(Evidence-Based Medicine) Donald A. Molony, Jonathan C. Craig, Giovanni Strippoli - Evidence-Based Nephrology, 2 Volume Set-Wiley-Blackwell (2022)

09-12-2022 15:09:44
Downloaded from https://onlinelibrary.wiley.com/doi/ by ibrahim ragab - Cochrane Germany , Wiley Online Library on [23/11/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Evidence-Based Nephrology
ffirs_Vol1.indd 1
Second Edition
VOLUME I
Edited by
Jonathan C. Craig, MBChB, DipCH, MMed(Clin Epi), PhD, FAHMS

Matthew Flinders Distinguished Professor
Vice President and Executive Dean, College of Medicine and Public Health
Flinders University
Adelaide, Australia
Donald A. Molony, MD
Professor of Medicine
Distinguished Teaching Professor of the University of Texas System
Division of Renal Diseases and Hypertension AND
Center for Clinical Research and Evidence-based Medicine
McGovern Medical School University of Texas, Houston, TX, USA
Giovanni F.M. Strippoli, MD, PhD, MPH, MM (Epi)

Professor of Nephrology, Department of Emergency and Organ Transplantation – University of Bari
Bari, Italy;
Adjunct Professor of Epidemiology, School of Public Health
University of Sydney
Sydney, NSW, Australia
With section editors
Aminu Bello Liz Lightstone

Mark Canney Paul Palevsky
Sara Davison Suetonia Palmer
Carmel Hawley Susan Samuel
David Johnson Allison Tong
Adeera Levin Germaine Wong
ffirs_Vol1.indd 3 09-12-2022 15:09:44

This edition first published 2023
© 2023 John Wiley & Sons Ltd
Edition History
Blackwell Publishing Ltd (1e, 2009)
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material
from this title is available at http://www.wiley.com/go/permissions.
The right of Jonathan C. Craig, Donald A. Molony, and Giovanni F.M. Strippoli to be identified as the authors of the editorial material in this
work has been asserted in accordance with law.
Registered Offices
John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
Editorial Office
9600 Garsington Road, Oxford, OX4 2DQ, UK
For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.
Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that appears in standard print versions
of this book may not be available in other formats.
Limit of Liability/Disclaimer of Warranty

The contents of this work are intended to further general scientific research, understanding, and discussion only and are not intended and
should not be relied upon as recommending or promoting scientific method, diagnosis, or treatment by physicians for any particular patient.
In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating
to the use of medicines, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert
or instructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indication of usage and
for added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no
representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties,
including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or
extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or
product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors
endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with
the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be
suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this
work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be
liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
Library of Congress Cataloging-in-Publication Data

Names: Molony, Donald A., editor. | Craig, Jonathan C., editor. |
Strippoli, Giovanni F.M., editor.
Title: Evidence-based nephrology / edited by Donald A. Molony, Jonathan C.
Craig, Giovanni F.M. Strippoli; with section editors, Aminu Bello [and 11 others].
Ýescription: Second edition. | Hoboken, NJ: Wiley-Blackwell, 2021. |
Includes bibliographical references and index.
Identifiers: LCCN 2020043376 (print) | LCCN 2020043377 (ebook) | ISBN
9781119105923 (cloth) | ISBN 9781119105930 (adobe pdf) | ISBN
9781119105947 (epub)
Subjects: MESH: Kidney Diseases | Evidence-Based Medicine–methods
Classification: LCC RC903 (print) | LCC RC903 (ebook) | NLM WJ 300 |
DDC 616.6/1–dc23
LC record available at https://lccn.loc.gov/2020043376
LC ebook record available at https://lccn.loc.gov/2020043377
Cover Design: Wiley

Cover Images: Jose Luis Calvo/Shutterstock, SEBASTIAN KAULITZKI/SCIENCE PHOTO LIBRARY
Set in 9.5/12.5pt STIXTwoText by Straive, Pondicherry, India
ffirs_Vol1.indd 4 09-12-2022 15:09:44

v
Contents
VOLUME I
Preface ix
List of Contributors xi
Part 1 Epidemiology 1
1 An Introduction to the Epidemiology of Chronic Kidney Disease 3
2 The Surveillance and Burden of Chronic Kidney Disease 14
3 Progression of Chronic Kidney Disease: An Evidence-based Approach to Risk Stratification 26
4 Screening for Chronic Kidney Disease 44
5 Risk Prediction in Chronic Kidney Disease 60
6 Chronic Kidney Disease in Disadvantaged Populations 72
Part 2 Acute Kidney Injury 85
7 Overview / Definition, Classification, and Epidemiology of Acute Kidney Disease 87
8 Pre-Renal Failure and Obstructive Disease 96
9 Hepatorenal Syndrome 107
10 Acute Tubular Necrosis 123
11 Iodinated Contrast and Acute Kidney Injury 145
12 Miscellaneous Etiologies of Acute Kidney Injury 163
13 Renal Replacement Therapy in Acute Kidney Injury 185
Part 3 Primary Diseases of the Kidney 203
14 Renal Biopsy 205
15 Minimal Change Disease and Focal Segmental Glomerulosclerosis in Adults 214
16 Membranous Nephropathy 235
ftoc_V1.indd 5 09-12-2022 15:12:29

vi Contents
17 IgA Nephropathy in Adults and Children 254
18 Membranoproliferative Glomerulonephritis 272
19 Autosomal Dominant Polycystic Kidney Disease 288
20 Urinary Tract Infections 305
21a Toxic Nephropathies: Environmental Agents and Metals 311
21b Toxic Nephropathies: Nonsteroidal Anti-Inflammatory Drugs 319
Part 4 Secondary Diseases of the Kidney 329
22 Hypertension 331
23 Renovascular Disease 354
24 Secondary Diseases of the Kidney: Diabetic Nephropathy 366
25 Lupus Nephritis 400
26 Hemolytic Uremic Syndrome 425
27 Pregnancy 444
28 ANCA-associated Vasculitis 461
29 Paraprotein-associated Kidney Disorders 478
30 Primary Immune Complex mediated Membranoproliferative Glomerulonephritis and

C3 Glomerulopathies: A Clinical Approach 494
Part 5 Chronic Kidney Disease and Complications 505
31 Cardiovascular Disease and CKD 507
32 Infection and CKD 526
33 Treatment of Anemia in Chronic Kidney Disease 542
34 Dyslipidemia in Chronic Kidney Disease 549
35 Chronic Kidney Disease and Hypertension 573
36 Chronic Kidney Disease-Mineral and Bone Disorder 589
37 Nutritional Disorders in Chronic Kidney Disease: Pathophysiology, Detection, and Treatment 617
38 Preparation for Dialysis 658
39 Choice of Hemodialysis or Peritoneal Dialysis for Kidney Replacement Therapy 671
Index 681
ftoc_V1.indd 6 09-12-2022 15:12:29

Contents vii
VOLUME II
Preface ix
Part 6 Hemodialysis 1
40 Modalities of Extracorporeal Therapy 3
41 Dialysis Dose and Adequacy for Hemodialysis 20
42 General Management of the Hemodialysis Patient 34
43 Infections in Hemodialysis Patients 48
44 Vascular Access for Hemodialysis 66
Part 7 Peritoneal Dialysis 91
45 Small Solute Clearance in Peritoneal Dialysis 93
46 Salt and Water Balance 113
47 Solutions 127
48 Peritoneal Dialysis: Infections 138
49 Urgent-start Peritoneal Dialysis 156
50 Peritoneal Dialysis Catheter Insertion 170
Part 8 Supportive Care 179
51 Overview of Kidney Supportive Care 181
52 Symptoms 194
53 Prognostication in Advanced Chronic Kidney Disease 208
54 Advance Care Planning 216
55 Conservative Kidney Management and Dialysis Withdrawal 227
Part 9 Transplantation 247
56 Evaluation of the Living Donor Kidney 249
57 The Impact of Deceased Donor Quality and Outcomes after Kidney Transplantation 257
58 Early Medical and Surgical Complications After Kidney Transplantation 271
59 Infections After Kidney Transplantation 294
ftoc_V1.indd 7 09-12-2022 15:12:29

viii Contents
Part 10 Electrolytes and Acid-Base Disorders 301
60 Electrolyte Disorders 303
61 Metabolic Evaluation and Prevention of Kidney Stone Disease 322
Part 11 Pediatrics 333
62 Growth, Nutrition, and Development 335
63 Bone Disease in Children with Chronic Kidney Disease 356
64 Anemia 379
65 Peritoneal Dialysis in Children 399
66 Hemodialysis 412
67 Urinary Tract Infections in Children 426
68 Henoch–Schonlein Purpura Glomerulonephritis and IgA Nephropathy in Children 439
69 Hereditary Nephritis in Children 451
70 Shigatoxin-related Hemolytic-uremic Syndrome 463
Part 12 Patient-centered Care and Outcomes 473
71 Shared Decision-making 475
72 Fatigue 488
73 Depression 499
74 Pain 517
75 Neurocognitive Disorders 539
76 Pruritus 551
77 Sexual Dysfunction 570
78 Family and Caregiver Support 584
Index 590
ftoc_V1.indd 8 09-12-2022 15:12:29

ix
Preface
The synthesis of the totality of evidence in kidney disease methods assist the summarizing and interpreting of evi-
was the original challenge we undertook when we con- dence in key areas of decision-making, which is useful to
ceived of the first edition of this evidence-based nephrol- patients, healthcare providers, and policy-makers.
ogy (EBN) textbook, thereby providing students of Using these methods, organizations including the
nephrology and practicing clinicians with a single conveni- Cochrane Kidney and Transplant group and several guide-
ent source of clinical evidence that had been passed line agencies have committed on a large scale to summa-
through an evidence-based filter. More than 12 years have rize the evidence and provide timely updates of established
passed since the first edition, and so it is timely to revisit and emerging new evidence, making research findings
the challenge in the form of a second edition. The reasons more manageable and reliable for end users. Much of this
for a second edition are principally three: the scale of new work has been included in this edition of the textbook. We
available data is substantial, newer methods of research hope this will allow the end-user to both access best evi-
synthesis have been developed, and there is a renewed dence from this diversity of resources as summarized by
focus on consumer engagement in general, including in chapter authors and to become more familiar with how the
the evaluation and management of kidney disease. newer techniques and resources contribute to a practice
that is informed by best evidence.
Scale of new data
Compared to the time of the first edition, the amount of Consumer engagement
new data is impressive. There are today more than 15 000 Recently, there has been increasing recognition of the criti-
randomized controlled trials in nephrology published in cal importance of engaging the end-users in research to
approximately 30 000 reports, 11 300 more than in 2009, ensure relevance and uptake. In particular there have been
resulting in more confidence and precision around the esti- large-scale efforts to engage patients in all stages of
mates of intervention effects and new evidence in areas research, including the evidence synthesis process and the
that were previously mostly evidence-free. These trials dissemination of findings. For example, the Cochrane
have also been summarized in systematic reviews, with Kidney and Transplant group now has a patient editor,
over 6500 additional reviews in Medline since the first edi- whose role includes translating scientific outputs into plain
tion. Much of this new evidence is incorporated in three language summaries to simplify the “medical jargon” and
new sections of this textbook and multiple either new or make evidence easily accessible to people without a medi-
entirely updated chapters. cal background, as well as prioritizing review topics. The
Standardized Outcomes in Nephrology (SONG) initiative
Novel methods was launched in 2014 to establish the core outcomes to be
Methods to assess the methodological quality of both sys- reported in all trials in chronic kidney disease, based on a
tematic reviews and other study designs, including nonran- consensus among patients, caregivers, health profession-
domized (case-control or cohort studies) and diagnostic test als, and policy-makers. This initiative has emphasized
studies, have developed since the first edition. In addition, patient-centered research and improved clinical outcomes
the number of syntheses of cohort studies has increased in nephrology, ensuring that the outcomes measured and
and the technique of network meta-analysis for comparing reported in clinical trials and other forms of research are
more than two interventions both directly (head to head) relevant and meaningful to patients and increase the
and indirectly (via a common comparator) has become acceptability, transparency, and generalizability of the
more developed and is commonly encountered. These results in this population. New studies assessing these
fpref_V1.indd 9 09-12-2022 15:12:55

x Preface
utcomes are now being developed, with advancement of

o and comprehensiveness has added complexity in the meth-
knowledge which could be incorporated into this edition of ods and may lead to residual uncertainty in the estimate of
the textbook. intervention effects. Finally, we focused on empiric evi-
In short, compared to the previous version published in dence for management decisions and so pathophysiologi-
2009, this updated textbook includes 20 new chapters and cal mechanisms of chronic kidney disease and its treatment
three new sections, covering supportive care and patient- were not considered in detail. We have attempted to include
centered care and outcomes. In addition, every chapter economic evaluations where particularly relevant for
that was in the first edition has been extensively updated. national and international management guidelines.
Specifically, the new textbook covers epidemiology, acute In the true spirit of evidence-based medicine, we hope
kidney injury, primary diseases of the kidney, secondary that this edition of Evidence-based Nephrology will con-
diseases of the kidney, chronic kidney disease and compli- tinue to push the specialty toward greater reliance on the
cations, hemodialysis, chronic kidney disease stage 5, peri- totality of evidence and the generation and utilization of
toneal dialysis, supportive care, transplantation, evidence that is the least biased and the most precise to
electrolytes and acid-base disorders, and patient-centered inform clinical decision making. To the material detailed in
care and outcomes for both adult and pediatric patients. all the chapters we would expect that our readers will add,
This effort was undertaken by existing and new authors through their own judicious application of evidence-based
and section editors, whom we would like to thank for their medicine principles, their local context, and incorporate
extensive work. their patients values and preferences, given that objective
Giovanni Strippoli has joined as the third co-editor of evidence is rightly only one consideration in the delivery of
Evidence-based Nephrology. Altogether, we hope that we true patient-centered care. Additionally, a textbook like
have produced an even better evidence-based nephrology this can at best be only one of several resources for the
tool, where each chapter provides a clear foundation of the evidence-based medicine practitioner, and should be sup-
topic that is supported by the best current evidence. plemented with current methodologically rigorous and
We believe that this updated textbook more broadly cov- transparent clinical practice guidelines. We hope this effort
ers available evidence and addresses crucial clinical ques- will provide a core resource for the evidence-based neph-
tions regarding the treatment and care of people with all rology practitioner who is otherwise limited by time con-
stages of chronic kidney disease, including people under- straints from researching every question that may arise
going any form of dialysis (hemodialysis or peritoneal dial- daily in the care of patients.
ysis), those requiring kidney transplantation, and pediatric We certainly wish that students and evidence-based
patients. nephrology practitioners will benefit from this updated
Inherently a textbook of this nature will always manifest edition in the search for answers to questions that arise in
potential limitations. We acknowledge that a textbook is daily care and their ambition to deliver the best care
unable to collect all new evidence in real time, and, like all possible.
of healthcare internationally, the COVID-19 pandemic did
impact this book, specifically the publication timelines as Jonathan C. Craig
contributors had to contend with major challenges. The Donald A. Molony
inclusion of nonrandomized studies to maximize relevance Giovanni F.M. Strippoli
fpref_V1.indd 10 09-12-2022 15:12:55

xi
List of Contributors
Diego Aguilar Bourne L. Auguste

PRS Population Health, Clinical Trial Service Division of Nephrology
Unit and Epidemiological Studies Unit, BHF Sunnybrook Health Sciences Centre
Centre of Research Excellence, Nuffield Department Toronto
of Population Health Canada
University of Oxford
UK Sevcan A. Bakkaloğlu
Department of Pediatric Nephrology
Gazi University
Kathleen E. Altemose School of Medicine, Ankara
Department of Pediatrics, Division of Pediatric Turkey
Nephrology & Hypertension
Penn State College of Medicine Joanne M. Bargman
Hershey Department of Medicine, Division of Nephrology
USA University of Toronto
Toronto, Canada
Oluwatoyin I. Ameh
Jonathan Barratt
Division of Nephrology
The John Walls Renal Unit, Leicester General Hospital,
Zenith Medical and Kidney Centre
and Department of Cardiovascular Sciences
Abuja
University of Leicester
Nigeria
UK
Sharon Phillips Andreoli
Nathan T. Beins
Department of Pediatrics
Division of Pediatric Nephrology
James Whitcomb Riley Hospital for Children
University of Missouri-Kansas City School of Medicine;
Indiana University School of Medicine
Children’s Mercy Hospital
Indianapolis, IN
Kansas City, Missouri, USA
USA
Aminu K. Bello
Chaisiri Angkurawaranon Division of Nephrology and Immunology, Department
Department of Family Medicine, Faculty of Medicine of Medicine
Chiang Mai University, Chiang Mai, Thailand University of Alberta
Thailand Edmonton, Alberta
Canada
Meredith A. Atkinson William M. Bennett

Department of Pediatrics, Division of Pediatric Legacy Transplant Services
Nephrology Legacy Good Samaritan Medical Center
Johns Hopkins University School of Medicine Portland, OR
Baltimore, MD, USA USA
flast_V1.indd 11 09-12-2022 15:13:20

xii List of Contributors
Meha Bhatt Ben Caplin

Division of Nephrology, Department of Medicine Department of Renal Medicine
Cumming School of Medicine University College London
University of Calgary UK
Alberta, Canada
Andres Cardenas
Bhadran Bose Institut de Malalties Digestives i Metabòliques,
Department of Nephrology Hospital Clínic
Nepean Hospital University of Barcelona
Kingswood, NSW, Australia; Spain and
University of Sydney, Sydney, NSW, Australia Institut d’Investigacions Biomèdiques August Pi-Sunyer y
Ciber de Enfermedades Hepáticas y Digestivas
Branko Braam
Barcelona
Division of Nephrology, Department of Medicine
Spain
University of Alberta
Canada
Juan J. Carrero
European Renal Nutrition Working Group of the
Frank Brennan
European Renal Association – European Dialysis
Departments of Nephrology and Palliative Care
Transplant Association
St George Hospital
Parma
Sydney
Italy and
Australia
Department of Medical Epidemiology and Biostatistics,
Karolinska Institutet
Frank Bridoux
Stockholm
Department of Nephrology, Centre Hospitalier
Sweden
Universitaire et Université de Poitiers
France and Department of Immunology CNRS UMR7276
Kerri Cavanaugh
Université de Limoges, Limoges
Division of Nephrology & Hypertension
France and
Department of Medicine
Centre de Référence Amylose AL et Autres Maladies par
Vanderbilt University Medical Center
Dépôt d’Immunoglobulines Monoclonales
Nashville, TN, USA
Université de Poitiers
Poitiers
France Christopher T. Chan
Victoria Briggs Toronto General Hospital
Department of Nephrology Toronto
Calderdale and Huddersfield NHS Foundation Trust Canada
Mark Brown Katharine L. Cheung

St. George Hospital Division of Nephrology, Department of Medicine
University of New South Wales The University of Vermont
Sydney Burlington
Australia USA
Neil Boudville Yeoungjee Cho

University of Western Australia, Medical Department of Nephrology
School, Sir Charles Gairdner Hospital, Nedlands, WA, University of Queensland at Princess Alexandra Hospital
Australia Woolloongabba, Brisbane, Australia
Mark Canney David Collister

Division of Nephrology Department of Medicine, Division of Nephrology
University of British Columbia McMaster University
Vancouver, BC, Hamilton
Canada Canada
flast_V1.indd 12 09-12-2022 15:13:20

List of Contributors xiii
Tess E Cooper Simon J Davies

Sydney School of Public Health Renal Research Group
The University of Sydney, Sydney Faculty of Medicine and Applied Clinical Sciences
Australia Keele University
and Newcastle-under-Lyme
Centre for Kidney Research, The Children’s Hospital at UK
Westmead, Westmead, Australia
Sara N. Davison
Sara A. Combs
Department of Medicine, Divisions of Nephrology and
Division of Nephrology and Immunology
Palliative Care
University of New Mexico School of Medicine
Edmonton, AB, Canada
Albuquerque, NM,
USA Mary Amanda Dew
Department of Psychiatry
Kelsey Connelly
University of Pittsburgh School of Medicine and
Max Rady College of Medicine,
Medical Center, Pittsburgh, PA, USA
University of Manitoba
Winnipeg, Manitoba Meghan J. Elliott
Canada Division of Nephrology, Department of Medicine,
Cumming School of Medicine
Michael J. Connor, Jr University of Calgary, Alberta
Division of Pulmonary, Allergy, Canada and
Critical Care, & Sleep Medicine Department of Community Health Sciences, Cumming
Department of Medicine School of Medicine
Emory University School of Medicine University of Calgary
Atlanta, GA, USA Alberta
and Canada
Division of Renal Medicine
Department of Medicine Fabrizio Fabrizi
Emory University School of Medicine Division of Nephrology
Atlanta, GA, USA Maggiore Hospital and IRCCS Foundation
Milano
Cecile Couchoud Italy
REIN Registry
Department Agence de la biomédecine Fadi Fakhouri
Saint Denis – La Plaine Department of Nephrology and Immunology
France Centre Hospitalier Universitaire de Nantes
France
Jonathan C Craig
College of Medicine and Public Health Kevin W. Finkel
Flinders University Division of Renal Diseases and Hypertension
Adelaide, Australia McGovern Medical School
Houston, TX
Neera K. Dahl USA
Section of Nephrology
Yale University School of Medicine Fred Finkelstein
New Haven, CT, Department of Medicine
USA Yale University
New Haven, CT, USA
Matthew J. Damasiewicz
Monash Health and Monash University James Fotheringham
Clayton, Victoria, School of Health and Related Research Regent Court,
Australia University of Sheffield
flast_V1.indd 13 09-12-2022 15:13:20

xiv List of Contributors
Martin Gallagher Camilla S. Hanson

Director of the Renal & Metabolic Division Sydney School of Public Health
The George Institute for Global Health University of Sydney
Newtown, NSW, Australia Sydney, NSW, Australia and
Centre for Kidney Research
Amit X. Garg The Children’s Hospital at Westmead
Western University Westmead, NSW, Australia
Institute for Clinical Evaluative Sciences Western Facility
Victoria Hospital Carmel M Hawley
London, ON, Canada Department of Nephrology, Princess Alexandra Hospital
University of Queensland
Michael J. Germain Brisbane, Australia
Department of Medicine, Division of Nephrology
Baystate Medical Center Swapnil Hiremath
Tufts University Department of Medicine
Springfield, MA, USA University of Ottawa
Canada
Pere Ginès
Institut de Malalties Digestives i Metabòliques, Hospital Htay Htay
Clínic Department of Renal Medicine
University of Barcelona Singapore General Hospital
Spain and Singapore
Institut d’Investigacions Biomèdiques August Pi-Sunyer y
Ciber de Enfermedades Hepáticas y Digestivas Emma Huarte
Barcelona Department of Nephrology
Spain Hospital San Pedro
La Rioja, Spain
David S. Goldfarb
Nephrology Section, New York Harbor Kwaifa S. Ibrahim
VA Medical Center, and Nephrology Department of Medicine
Division, NYU Grossman School of Wuse District General Hospital
Medicine, New York, Abuja, Nigeria
New York, USA
Georgina Irish
Silviu Grisaru Central and Northern Adelaide Renal and Transplantation
Alberta Children’s Hospital and University of Calgary Service
Calgary, Alberta, Canada Royal Adelaide Hospital Adelaide,
South Australia, Australia and
Vanessa Grubbs Department of Medicine
Department of Medicine, University of Adelaide
University of California Adelaide, South Australia, Australia
San Francisco, CA, USA
Masao Iwagami
Talia Gutman
Department of Health Services Research, University of
Sydney School of Public Health
Tuskuba, Ibaraki,
The University of Sydney
Japan
NSW, Australia and
and
The Centre for Kidney Research
Department of Non-Communicable
The Children’s Hospital at Westmead
Disease Epidemiology, London School of Hygiene &
NSW, Australia
Tropical Medicine, London, UK
flast_V1.indd 14 09-12-2022 15:13:20

List of Contributors xv
Matthew T. James David W Johnson

Division of Nephrology, Department of Medicine Department of Nephrology
Cumming School of Medicine University of Queensland at Princess
University of Calgary Alexandra Hospital
Alberta Woolloongabba, Brisbane, Australia
Canada and
Department of Community Health Sciences, Cumming Angela Ju
School of Medicine Sydney School of Public Health
University of Calgary University of Sydney
Alberta, Canada NSW, Australia and
Meg Jardine Children’s Hospital at Westmead
NHMRC Clinical Trials Centre Westmead, NSW, Australia
Camperdown Adrià Juanola
Australia and Institut de Malalties Digestives i Metabòliques,
Concord Repatriation General Hospital University of Barcelona
Australia Spain and
Sarbjit Vanita Jassal Ciber de Enfermedades Hepáticas y Digestivas
Division of Nephrology Barcelona
Department of Medicine Spain
University of Toronto, Toronto, ON
Canada
Peter G Kerr
Vincent Javaugue Professor and Director of Nephrology
Department of Nephrology Monash Health and Monash University
Centre Hospitalier Universitaire et Clayton, Victoria,
Université de Poitiers Australia
France and
Department of Immunology CNRS UMR7276 Myda Khalid
Université de Limoges Department of Pediatrics
Limoges Division of Pediatric Nephrology and Hypertension
France and James Whitcomb Riley Hospital for Children
Centre de Référence Amylose AL et Autres Maladies par Indiana University Medical Center
Dépôt d’Immunoglobulines Monoclonales Indianapolis, IN, USA
Poitiers Mubeen M. Khan
France Division of Renal Diseases and Hypertension
McGovern Medical School
Shilpanjali Jesudason Houston, TX
Central and Northern Adelaide Renal and Transplantation USA
Service
Royal Adelaide Hospital Amrit Kirpalani
Adelaide, South Australia, Australia and Division of Nephrology and Cell Biology Program
Department of Medicine Research Institute
University of Adelaide The Hospital for Sick Children
Adelaide, South Australia, Australia Toronto, ON, Canada
flast_V1.indd 15 09-12-2022 15:13:20

xvi List of Contributors
Jay L. Koyner Kelly Li

Section of Nephrology St. George Hospital
Department of Medicine University of New South Wales
University of Chicago, Chicago, IL Sydney
USA Australia
Raymond T. Krediet
Christoph Licht
Division of Nephrology and Cell Biology Program
Amsterdam UMC
Research Institute
University of Amsterdam
The Hospital for Sick Children
The Netherlands
Toronto, ON
Mark Lambie Canada and
Renal Research Group, Faculty of Medicine and Applied Department of Paediatrics
Clinical Sciences University of Toronto, Toronto, ON
Keele University Canada
Newcastle-under-Lyme
UK Wai H. Lim
Department of Renal Medicine
Nicholas G. Larkins Sir Charles Gairdner Hospital
Department of Nephrology, Perth Children’s Hospital Perth
Nedlands Australia and
Western Australia School of Medicine
Australia and Discipline of Internal Medicine
School of Paediatrics and Child Health University of Western Australia
University of Western Australia Perth
Nedlands, Western Australia, Australia
Australia
Hamidu M. Liman
Nelson Leung Division of Nephrology
Division of Nephrology, Hematology Usmanu Danfodiyo University Teaching Hospital
Department of Internal Medicine Sokoto
Mayo Clinic, Rochester, MN Nigeria
USA
Kathleen D. Liu
Asaf Lebel Division of Nephrology, Departments of Medicine and
Division of Nephrology and Cell Biology Program Anesthesia
Research Institute University of California
The Hospital for Sick Children San Francisco, CA
Toronto, ON, Canada USA
Adeera Levin Jayme E. Locke

Division of Nephrology Department of Surgery, Division of Transplantation
University of British Columbia University of Alabama at Birmingham, Birmingham, AL
Vancouver, BC USA
Canada
Charlotte Logeman
Guisen Li
Renal Department
Sydney, NSW, Australia and
Institute of Nephrology
Sichuan Provincial People’s Hospital
Medical School of UESTC
Westmead, NSW, Australia
China
flast_V1.indd 16 09-12-2022 15:13:20

List of Contributors xvii
Jicheng Lv Piergiorgio Messa

Peking University First Hospital Division of Nephrology
Beijing Maggiore Hospital and IRCCS Foundation
China Milano
Italy and
Bryan Ma University School of Medicine
Division of Nephrology, Department of Medicine Milano
Cumming School of Medicine Italy
University of Calgary
Alberta Pablo Molina
Canada Department of Nephrology, Hospital Universitario Dr Peset
Universitat de València
Robert MacGinley Spain and
Department of Renal Medicine European Renal Nutrition Working Group of the European
Eastern Health Renal Association – European Dialysis Transplant Association
Box Hill Parma
Australia Italy
and Renal Medicine Department
Eastern Health Clinical School Donald A. Molony
Monash University Division of Renal Diseases and Hypertension and Center
Melbourne for Clinical Research and Evidence-Based Medicine
Australia University of Texas Houston Medical School
Houston
Magdalena Madero
USA
Department of Nephrology,
National Institute of Cardiology David W. Mudge
Mexico Department of Nephrology
University of Queensland at Princess Alexandra Hospital
Muhammad A. Makusidi Woolloongabba, Brisbane,
Division of Nephrology Australia
Usmanu Danfodiyo University Teaching Hospital
Sokoto Elmi Muller
Nigeria Faculty of Medicine and Health Science & Dept of Surgery
tellenbosch University
Karine E. Manera
Sydney School of Public Health Alexandra Munt
University of Sydney Department of Renal Medicine
Australia and Westmead Hospital
Centre for Kidney Research Western Sydney Local Health District
The Children’s Hospital at Westmead Sydney
Sydney Australia and
Australia Michael Stern Laboratory for Polycystic Kidney Disease
Westmead Institute for Medical Research
Christina Mariyam Joy
Australia
Washington University in St. Louis
USA Margaux N. Mustian
Department of Surgery, Division of Transplantation
Stephen McAdoo
University of Alabama at Birmingham, Birmingham, AL
Imperial College London
USA
London, UK
Masaomi Nangaku
Helen McDonald
Division of Nephrology and Endocrinology
Department of Infectious Disease Epidemiology
The University of Tokyo Hospital
London School of Hygiene & Tropical Medicine
Bunkyo-ku, Tokyo
London, UK
Japan
flast_V1.indd 17 09-12-2022 15:13:20

xviii List of Contributors
Patrizia Natale Ali Olyaei

Department of Emergency and Organ Transplantation Department of Pharmacy
University of Bari Oregon State University/Oregon Health & Science University
Bari, Italy Portland, OR
and USA
The University of Sydney Mohamed A. Osman
Sydney, Australia Division of Nephrology and Immunology,
Sharon J. Nessim University of Alberta
Division of Nephrology Edmonton, Alberta
Jewish General Hospital Canada
McGill University
Montreal Marlies Ostermann
Canada Department of Critical Care
Guy’s & St Thomas’ Hospital
London
Javier A. Neyra
UK
Division of Nephrology, Bone and Mineral Metabolism
University of Kentucky Luis M. Pallardó
Lexington Department of Nephrology
USA Hospital Universitario Dr Peset
Spain
Thu T. Nguyen
Suetonia C. Palmer
Auckland City Hospital
Auckland
University of Otago Christchurch
New Zealand
Christchurch,
New Zealand
Dorothea Nitsch
Department of Non-Communicable Patrick S. Parfrey
Disease Epidemiology Faculty of Medicine
London School of Hygiene & Tropical Medicine Memorial University
London, UK St. John’s, NL,
Canada
Marlies Noordzij
Department of Medical Informatics Henry Pleass
Amsterdam UMC Specialty of Surgery
University of Amsterdam Faculty of Medicine and Health
The Netherlands University of Sydney and
Centre for Transplant and Renal Research
Ikechi G. Okpechi Westmead Hospital
Division of Nephrology and Hypertension Sydney
Groote Schuur Hospital Australia and
Cape Town, South Africa Department of Surgery
and Surgical Research and Education Centre
Kidney and Hypertension Research Unit Westmead Hospital
University of Cape Town Sydney
Cape Town, South Africa Australia
flast_V1.indd 18 09-12-2022 15:13:20

List of Contributors xix
Kevan R. Polkinghorne Gopala Rangan

Department of Nephrology Department of Renal Medicine
Monash Medical Centre Westmead Hospital
Melbourne, Victoria Western Sydney Local Health District
Australia Sydney, Australia and
Michael Stern Laboratory for Polycystic
Carol Pollock Kidney Disease
Sydney School of Public Health Westmead Institute for Medical Research
The University of Sydney University of Sydney
Sydney, Australia Australia
and
Kolling institute of Medical Research Amanda DeMauro Renaghan
St Leonards, Australia Division of Nephrology
University of Virginia Health System
Evgenia Preka Charlottesville
Southampton Children’s Hospital USA
University Hospital Southampton NHS Foundation Trust
Southampton Lesley Rees†
UK Paediatric Nephrology
Great Ormond Street Hospital NHS Foundation
Maria Prendecki Trust and UCL Great Ormond Street Institute of
Imperial College London Child Health
London, UK London, UK
Joseph B. Pryor Matthew A. Roberts

Department of Internal Medicine Eastern Health Clinical School
University of Washington, Seattle Monash University
WA, USA Melbourne, VIC
Australia
Lonnie Pyne
Department of Medicine Mitchell H. Rosner
Division of Nephrology Division of Nephrology
McMaster University University of Virginia Health System
Hamilton Charlottesville, VA
Canada USA
Qi Qian
Marinella Ruospo
Division of Nephrology and Hypertension
University of Sydney, NSW
Mayo Clinic School of Medicine
Australia
USA
Kannaiyan S. Rabindranath Fahad Saeed

Renal Unit Departments of Medicine and Public Health
Waikato Hospital Divisions of Nephrology and Palliative Care
Hamilton University of Rochester Medical Centre
New Zealand Rochester, NY, USA
Jörg Radermacher Valeria Saglimbene

Department of Nephrology Sydney School of Public Health
Johannes Wesling Klinikum Minden University of Sydney, NSW
UK- RUB, Minden, Germany Australia
flast_V1.indd 19 09-12-2022 15:13:20

xx List of Contributors
Aminu M. Sakajiki Emily J. See

Division of Nephrology Department of Intensive Care
Usmanu Danfodiyo University Teaching Hospital Austin Health
Sokoto, Nigeria Melbourne
Australia and
Joshua Samuels School of Medicine
The McGovern Medical School University of Melbourne
University of Texas HSC Melbourne
Houston, TX, USA Australia
Irene Sangadi Nikhil Shah

Department of Renal Medicine Division of Nephrology and Immunology, Department of
Westmead Hospital Medicine
Western Sydney Local Health District University of Alberta
Sydney Edmonton, Alberta
Australia and Canada
Michael Stern Laboratory for Polycystic Kidney Disease
Soroush Shojai
Division of Nephrology and Immunology, Department of
Australia
Medicine
Sayan Saravanabavan
Edmonton, Alberta
Canada
Westmead Hospital
Western Sydney Local Health District
Sydney, Australia and Badri Shrestha
Michael Stern Laboratory for Polycystic Kidney Disease Department of Renal Medicine
Westmead Institute for Medical Research Sheffield Teaching Hospitals NHS Foundation Trust
University of Sydney Sheffield
Australia S Yorkshire, UK
Judy Savige Rukshana Shroff

Department of Medicine (Melbourne Health) and Great Ormond Street Hospital for Children
Northern Health NHS Foundation Trust, London
The University of Melbourne UK
Australia
Brendan Smyth
Jane Schell Department of Renal Medicine
Section of Palliative Care and Medical Ethics, Division of St George Hospital, Kogarah
Renal-Electrolyte Australia
Department of Medicine and
University of Pittsburgh School of Medicine NHMRC Clinical Trials Centre
UPMC Health System University of Sydney
Pittsburgh, PA, Camperdown, Australia
USA
Belinda Stallard
Rebecca Schmidt Department of Nephrology
Section of Nephrology, Department of Medicine Princess Alexandra Hospital
West Virginia University School of Medicine Brisbane
Morgantown, WV Australia
USA
flast_V1.indd 20 09-12-2022 15:13:20

List of Contributors xxi
Dan A Streja Mathew Tabinor

David Geffen School of Medicine at UCLA Renal Research Group, Faculty of Medicine and Applied
Los Angeles, CA Clinical Sciences
USA and Keele University
Division of Endocrinology, Diabetes and Metabolism Newcastle-under-Lyme
VA Healthcare System UK
West Los Angeles VA Medical Center
Los Angeles, CA Tetsuhiro Tanaka
USA Division of Nephrology and Endocrinology
Elani Streja Bunkyo-ku, Tokyo
Division of Nephrology, Department of Medicine Japan
Harold Simmons Center for Kidney Disease Research and
Navdeep Tangri
Epidemiology
Max Rady College of Medicine
UC Irvine School of Medicine
Orange, CA, USA and
Winnipeg, Manitoba
Tibor Rubin VA Medical Center
Canada
Long Beach, CA
USA
J. Pedro Teixeira
Giovanni F.M. Strippoli Divisions of Nephrology and Pulmonary, Critical Care,
Department of Emergency and Organ Transplantation and Sleep Medicine
University of Bari University of New Mexico School of Medicine
Bari, Italy Albuquerque, NM, USA
and
School of Public Health Allison Tong
University of Sydney Sydney School of Public Health
Sydney, NSW, Australia University of Sydney
NSW, Australia and
Guobin Su Centre for Kidney Research
Department of Nephrology, Guangdong Provincial Children’s Hospital at Westmead
Hospital of Chinese Medicine Westmead, NSW,
The Second Affiliated Hospital Australia
Guangzhou University of Chinese Medicine
Hernán Trimachi
Guangzhou City
Nephrology Service and Kidney Transplant Unit
Guangdong Province
Hospital Británico de Buenos Aires
China
Buenos Aires
and
Argentina
National Clinical Research Center for Kidney Disease
State Key Laboratory of Organ Failure Research Katie Trinh
Guangdong Provincial Clinical Research Center for Department of Nephrology
Kidney Disease Nepean Hospital
and Kingswood, NSW, Australia;
Department of Nephrology University of Sydney, Sydney, NSW, Australia
Nanfang Hospital, Southern Medical University
Guangzhou city, Guangdong Province David J Tunnicliffe
China Global Health – Health Systems and Policy Sydney School of Public Health
Department of Global Public Health, Karolinska Institutet The University of Sydney, Sydney, NSW
Stockholm, Sweden Australia and
and Centre for Kidney Research
Department of Medical Epidemiology and Biostatistics The Children’s Hospital at Westmead
Karolinska Institutet Westmead, NSW
Stockholm, Sweden Australia
flast_V1.indd 21 09-12-2022 15:13:20

xxii List of Contributors
Mark Unruh Li Wang

Division of Nephrology Renal Department
University of New Mexico School of Medicine Institute of Nephrology
Albuquerque, NM Sichuan Provincial People’s Hospital
USA and Medical School of UESTC
New Mexico VA Health Care System China
Albuquerque, NM
USA Bradley A Warady
Tomoko Usui
Bunkyo-ku, Tokyo
Japan
Steven D. Weisbord
Renal Section and Center for Health Equity Research and
Anita van Zwieten
Promotion
VA Pittsburgh Healthcare System and Renal Electrolyte
Division
Australia
University of Pittsburgh School of Medicine
Pittsburgh
Andrea K. Viecelli USA
Department of Nephrology
Princess Alexandra Hospital Martin Wilkie
Brisbane, Queensland, Department of Renal Medicine, Sheffield Teaching
Australia Hospitals NHS Foundation Trust
Sheffield, S Yorkshire, UK
Anitha Vijayan
Division of Nephrology Annette Wong
Washington University in St. Louis Department of Renal Medicine
USA Westmead Hospital
Belén Vizcaíno Sydney Australia
Department of Nephrology and
Hospital Universitario Dr Peset Michael Stern Laboratory for Polycystic Kidney Disease
Universitat de València Westmead Institute for Medical Research
Spain University of Sydney
Australia
Michael Walsh
Department of Medicine, Division of Nephrology Muh Geot Wong
McMaster University, Hamilton The University of Sydney
Canada and Sydney, NSW, Australia
Department of Health Research Methods and
Evaluation and Impact Department of Renal Medicine
McMaster University Concord Repatriation General Hospital
Hamilton Concord, NSW, Australia
Canada and
Population Health Research Institute See Cheng Yeo
Hamilton Health Sciences/McMaster University Department of Renal Medicine
Hamilton Tan Tock Seng Hospital
Canada Singapore
flast_V1.indd 22 09-12-2022 15:13:20

List of Contributors xxiii
Jennifer Zhang Yuemiao Zhang

Department of Renal Medicine, Westmead Hospital Peking University First Hospital
Western Sydney Local Health District, Sydney Beijing
Australia and China
Australia
flast_V1.indd 23 09-12-2022 15:13:20

1
Epidemiology
Part 1
3
An Introduction to the Epidemiology of Chronic Kidney Disease

Mark Canney and Adeera Levin
Division of Nephrology, University of British Columbia, Vancouver, BC, Canada
I ntroduction well-studied outcomes. We also highlight some of the more

controversial aspects of the CKD paradigm: the definition
The purpose of this part of the book is to describe the of CKD in older individuals, and strengths and limitations
epidemiology of chronic kidney disease (CKD) across of different filtration markers and GFR estimating equa-
the spectrum of disease, from early detection through to tions. These important issues continue to be discussed and
major complications and end-stage kidney disease debated within the nephrology community. However, it
(ESKD). Fundamental to this is an understanding of what should be acknowledged that patients with kidney disease
defines CKD. We will therefore begin this chapter by are looked after by other healthcare providers, including
revisiting the 2002 National Kidney Foundation Kidney primary care physicians, cardiologists, and gerontologists.
Disease Outcomes Quality Initiative (NKF/DOQI) guide- Thus, the framework for diagnosis, staging, and risk strati-
line, which first proposed a harmonized definition of fication of patients with kidney disease must be clear and
CKD based on simple laboratory markers. A tremendous well-communicated to all stakeholders. This is especially
amount of data has been accrued since that seminal pub- pertinent in developing countries, where our understand-
lication, data which informed the updated CKD classifi- ing of the epidemiology of kidney disease and its risk fac-
cation system by Kidney Disease Improving Global tors is more limited.
Outcomes (KDIGO) in 2012. The subsequent chapters in this part expand on these
The definition and staging of CKD has generated debate concepts and provide the reader with an up-to-date review
and controversy among the nephrology community. of the state of knowledge of the epidemiology of kidney
There is confusion surrounding the definition and the disease. Chapter 2 describes the variety of surveillance
classification system. The former is an important, inclu- mechanisms currently utilized for CKD and ESKD, along
sive, and objective definition to describe the state of kid- with the challenges of obtaining accurate data from around
ney function and structure which is associated with the world. The importance of early detection of CKD and
disease. The latter describes the severity of disease and the evidence for population-level screening is discussed in
risk associated with two parameters: level of estimated Chapter 4. Chapters 3 and 5 provide an evidence-based
glomerular filtration rate (GFR) and degree of albuminu- approach to prognostication for progression of CKD and
ria. The classification of CKD is underpinned by the risk the risk of important clinical outcomes such as cardiovas-
of a series of clinical outcomes. The key outcomes in this cular disease and mortality. Finally, the emerging literature
context are progression to ESKD, cardiovascular, and all- regarding the burden of CKD in disadvantaged popula-
cause mortality. Other outcomes include acute kidney tions is described in Chapter 6. Throughout these chapters,
injury (AKI), hospitalization, cardiovascular events, and a standardized nomenclature for CKD is used to frame
infections. This chapter provides an overview of the the discussions. Hence, that is where we begin this
risk relationships between CKD and these first three introduction.
Evidence-Based Nephrology, Second Edition. Edited by Jonathan C. Craig, Donald A. Molony, and Giovanni F.M. Strippoli.
© 2023 John Wiley & Sons Ltd. Published 2023 by John Wiley & Sons Ltd.
4 An Introduction to the Epidemiology of Chronic Kidney Disease
Definition of CKD In the absence of any of these factors, an individual can be

diagnosed as having CKD if they have a structural abnor-
In 2002 NKF-KDOQI published a guideline proposing a mality identified on imaging tests, for example polycystic
new paradigm for the classification and evaluation of kid- kidney disease, or a histopathological abnormality such as
ney disease [1, 2]. This conceptual framework (Figure 1.1) tubulointerstitial disease.
addressed an unmet need in nephrology: the early identifi- KDIGO proposed a standardized approach to the classifi-
cation of patients at risk of either developing kidney dis- cation of CKD based on the cause of disease (C), level of
ease or progressing to more advanced stages of disease. The GFR (G), and amount of albuminuria (A), the CGA system.
NKF-KDOQI working group suggested a new classification The classification and staging of CKD can be done inde-
system that could categorize the severity of an individual’s pendent of cause, but the guidelines clearly recognized the
kidney disease, irrespective of etiology, using simple labo- importance of attempting to ascertain the etiology of CKD.
ratory parameters. One of the most significant contribu- Statements within the guidelines acknowledge the impor-
tions of this work was the harmonization of nomenclature tance of interpreting the GFR and albuminuria results in the
for kidney disease [3]. The term “chronic kidney disease” context of an underlying cause, as the specific cause may
and its stages would supplant ambiguous terminology such influence the risk or anticipated rate of disease progression.
as “azotemia,” “predialysis,” or “renal insufficiency” [4]. In many cases, however, the cause is unknown or can only
The original guideline has had a major impact over the last be speculated upon based on the presence or absence of key
15 years or so by facilitating collaborative research endeav- risk factors such as diabetes or hypertension. For many
ors, standardizing clinical and laboratory practices, and patients a kidney biopsy remains the only means of estab-
informing health policy. lishing a firm diagnosis, although recent advances in molec-
The definition of CKD was further refined by the KDIGO ular diagnostics have the potential to overcome this
working group in the 2012 guidelines, and is the one used barrier [6]. The clinical focus on GFR and albuminuria for
in contemporary clinical practice. KDIGO defines CKD as CKD staging is based on iterative studies demonstrating that
abnormalities of kidney structure or function, present for both parameters are powerful predictors of clinical out-
at least 3 months, with implications for health [5]. CKD comes. The cross-classification of GFR and albuminuria
can thus be defined by markers of kidney damage and/or stages provides a categorization of an individual’s risk status
markers of reduced kidney function. GFR is considered the (Figure 1.2). The synergistic effect of both risk markers is
best metric of kidney function and is estimated from the reflected in the fact that individuals with a combination of
serum concentration of a circulating filtration marker such reduced GFR and a high degree of albuminuria are at great-
as creatinine or cystatin C. CKD can be diagnosed by GFR est risk for ESKD. It is also evident from this risk classifica-
criteria alone if an individual has at least two estimates of tion that the presence of albuminuria is a strong risk factor
GFR below 60 ml/min/1.73 m2 at least 3 months apart. for ESKD even if the GFR is normal. The thresholds of GFR
Even in the presence of “preserved” kidney function and urinary albumin to creatinine ratio (ACR) that are used
(GFR 60 ml/min/1.73 m2), CKD can be diagnosed in the to establish the stage of CKD are based on large population-
presence of persistent urinary abnormalities such as albu- based studies examining the association between both mark-
minuria, hematuria, or evidence of tubular dysfunction. ers and hard clinical endpoints.
Potential antecedents of CKD

Stages of CKD
Consequences of CKD Complications
Increased ↓GFR Kidney

Normal Damage Death
risk failure
Screening CKD Diagnosis and Estimate Replacement

for CKD risk reduction; treatment; progression; by dialysis and
risk factors screening for treat comorbid treat complications; transplantation
CKD conditions; prepare for
slow progression replacement
Figure 1.1 Conceptual paradigm for chronic kidney disease. CKD, chronic kidney disease; GFR, glomerular filtration rate.
CKD and Outcome 5
No CKD Albuminuria stages, description, and range (mg/g)

Moderate-risk CKD
High-risk CKD A1 A2 A3
Very high-risk CKD
Optimum and High Very high and
high-normal nephrotic
<10 10–29 30–299 300–1999 ≥2000
>105
High and
G1
GFR stages, description, and range
optimum
90–104
(mL/min per 1.73 m2)
75–89
G2 Mild
60–74
G3a Mild-moderate 45–59
G3b Moderate-severe 30–44
G4 Severe 15–29
G5 Kidney failure <15
Figure 1.2 Cross-classification of glomerular filtration rate and albumin to creatinine ratio stages with respect to risk of clinical
outcomes. CKD, chronic kidney disease; GFR, glomerular filtration rate.
CKD and Outcomes ACR of 5 mg/mg [9]. CKD, whether defined by eGFR or

ACR criteria, is associated with substantial reductions in
Contemporary data pertaining to CKD and risk of clinical life expectancy [10, 11].
outcomes come predominantly from the CKD Prognosis
Consortium (CKD-PC), an international team of investiga-
Cardiovascular Events
tors from over 70 cohorts worldwide [7]. These cohorts rep-
resent individuals from the general population, those at CKD has been identified as a risk factor for several cardio-
high risk of vascular disease, as well as individuals with vascular outcomes, including coronary events, heart fail-
established kidney disease. CKD-PC investigators have ure, stroke, peripheral arterial disease, and cardiovascular
consistently demonstrated higher risk of adverse outcomes mortality [12–14]. The risk of myocardial infarction associ-
associated with reduced eGFR and/or increased ACR, ated with CKD may be at least as high as the risk associated
independent of demographic and clinical characteristics. with diabetes [15]. The presence of CKD is also a risk factor
for recurrent coronary events, with a magnitude of risk
that is higher than traditional cardiovascular risk factors
All-Cause Mortality
such as smoking [16]. These findings suggest that CKD
In general population cohorts, both a reduced eGFR and should be regarded as a coronary heart disease risk equiva-
an elevated ACR have been associated with higher risk of lent [17]. Data from CKD-PC quantified the risk of cardio-
all-cause mortality [8]. The risk relationship between vascular mortality for eGFR and ACR categories in
eGFR and mortality tends to be J-shaped, in that individ- 266 975 individuals with known cardiovascular risk fac-
uals with “high” eGFR (>95 ml/min/1.73 m2) also appear tors [18]. An increased risk of mortality was evident below
to have an increased mortality risk. This is most likely a threshold eGFR of 45 ml/min/1.73 m2. Even small
explained by malnutrition or low muscle mass (and there- increases in albuminuria were associated with higher car-
fore lower creatinine generation) in the setting of severe diovascular risk. Individuals with CKD often have other
illness. The association between albuminuria and mortal- comorbidities such as hypertension and diabetes, both of
ity tends to be broadly linear. In one large meta-analysis, which are also highly prevalent in the general population.
risk of death was increased by 20%, 63%, and 122% for The risk of cardiovascular disease associated with CKD
ACR values of 10, 30, and 300 mg/mg compared to an may be attributable to shared risk factors in people with
these conditions. This specific question was addressed in equation [24]. CKD-EPI has a number of advantages over
two large meta-analyses which demonstrated that the its predecessor, the Modification of Diet in Renal Disease
higher observed risk of cardiovascular mortality in indi- (MDRD) equation [25, 26]. These include a more diverse
viduals with reduced eGFR and/or increased ACR was derivation cohort which included people with and without
evident in those with and without diabetes and hyperten- kidney disease, less bias in the upper range of GFR ( 60 ml/
sion [19, 20]. The underlying mechanism driving poor min/1.73 m2), and superior risk discrimination for hard
cardiovascular outcomes in people with CKD is multifac- outcomes [27]. A significant limitation of the CKD-EPI
torial and continues to be an area of intensive research cohort was the under-representation of individuals over
efforts. Despite consistent evidence of higher cardiovas- the age of 70 years. An alternative GFR estimating equa-
cular risk, patients with CKD are often underinvestigated tion for this demographic was developed by the Berlin
and undertreated. One of the reasons for this is the under- Initiative Study (BIS) investigators [28].
representation of patients with kidney disease, particu- There is a trade-off in the use of any GFR estimating
larly advanced disease, in clinical trials. It is not known if equation: the widespread ease of their use versus the
interventions which have been shown to improve out- uncertainty in estimation. Some of this uncertainty is due
comes in people without kidney disease will have the to inherent variability in the regression models that gener-
same benefits and adverse effect profiles in people with ated the estimates, most notably at the higher range of
CKD. This knowledge gap needs to be urgently addressed. GFR where the models tend to perform less well. More
The strength of evidence for cardiovascular risk predic- importantly, creatinine generation can vary widely between
tion is expanded upon in Chapter 5. individuals due to differences in muscle mass and dietary
protein intake [29]. Several alternative filtration markers
have been studied, but none have managed to replace cre-
Progression to ESKD
atinine as the marker of choice in routine clinical practice.
The risk of ESKD is dramatically increased in the setting of One such alternative is cystatin C, a 13 kDa protein that is
reduced eGFR and/or high-grade albuminuria. In the produced by all nucleated cells and freely filtered at the
general population, this increased risk becomes apparent glomerulus. Although cystatin C is a better predictor of
below an eGFR of 75 ml/min/1.73 m2 and increases in measured GFR than creatinine [30], the adjustment of cre-
exponential fashion at the lower end of eGFR [21]. In CKD atinine for body composition using the variables age, sex,
cohorts, each 15 ml/min/1.73 m2 reduction in eGFR below and race largely overcomes this deficit [31]. Where the two
45 ml/min/1.73 m2 confers an approximate sixfold higher markers differ is with respect to risk stratification. GFR
risk of ESKD [22]. Similar to cardiovascular and mortality estimated from cystatin C is a stronger and more consistent
outcomes, the risk relationship between albuminuria and predictor of clinical outcomes compared to GFR estimated
ESKD appears to be more linear. The consistency of the from creatinine [32]. It should be noted that cystatin C
associations between eGFR, ACR, and ESKD risk facilitated also has “non-GFR” determinants, including adiposity
the creation of the Kidney Failure Risk Equation, which and inflammation, themselves recognized as risk factors
calculates the probability of progressing to ESKD at spe- for adverse clinical outcomes such as cardiovascular dis-
cific time points based on an individual’s age, sex, eGFR, ease [33, 34].
and ACR [23]. Risk prediction models and risk factors
for disease progression are comprehensively evaluated in
Chapter 5.
Risk Factors for CKD
GFR Estimating Equations and Risk Hypertension
Indirect quantification of GFR, via clearance of an exoge- Hypertension is estimated to affect 1.13 billion adults glob-
nous filtration marker, is the gold standard for measure- ally, with an age-standardized prevalence of 24.1% in men
ment of GFR but is rarely performed in practice outside of and 20.1% in women [35]. Hypertension is highly prevalent
specific clinical indications. The large-scale population among individuals with CKD, especially in more advanced
studies described above would not have been feasible with- stages of disease. The kidney is a vulnerable organ to the
out a method for calculating an estimate of GFR. A num- effects of sustained elevations in blood pressure (BP), and
ber of GFR estimating equations have been developed and hypertension is a risk factor for both incident CKD and
validated for use in clinical and research settings. In con- progression to ESKD [36, 37]. In the CKD population, the
temporary practice, the most widely used is the Chronic relationship between BP and cardiovascular outcomes and
Kidney Disease Epidemiology Collaboration (CKD-EPI) mortality tends to be J-shaped, suggesting higher risk of
Prevalence of CK 7
adverse outcomes with high and low BP [38–40]. This find- decrease in urine output over a specified time interval [55].
ing could be explained by underlying cardiac disease in AKI is associated with increased risk of morbidity and
individuals with low BP, which may be subclinical and not mortality, particularly from cardiovascular disease [56, 57].
optimally captured in studies [41]. Blood pressure is a Severe AKI necessitating temporary dialysis is a risk factor
dynamic measure of health, and what we consider as the for long-term development of ESKD [58]. On the other end
mean or “usual” BP may not be telling the whole story [42]. of the spectrum, apparently “mild” injury has been associ-
Variability in ambulatory BP recordings is a strong predic- ated with increased risk of developing CKD, especially if
tor of major cardiovascular events such as stroke, inde- recovery is slow and even if kidney function ultimately
pendent of the mean BP [43]. Similarly, large variability in returns to its baseline value after the event [59, 60]. In some
visit-to-visit BP has been identified as a risk factor for patients the relationship between AKI and CKD is bi-
adverse CKD outcomes [44]. Impaired stabilization of BP directional [61]. A diminished GFR and albuminuria are
during postural change has recently been highlighted as a cardinal risk factors for AKI [62–64], which in turn may
potential contributor to adverse outcomes [45, 46] and is contribute to progression of underlying CKD [65]. The
also associated with CKD [47, 48]. These studies illustrate importance of AKI as a risk factor for CKD progression,
the complexity of the relationship between changes in and methods to identify those at risk of progression after
blood pressure and their effects on the kidney. Our under- an AKI event, are discussed in detail in Chapter 3.
standing of hypertension in CKD is further hampered by
the under-representation of CKD patients in large clinical
trials, such that optimal management of hypertension in Prevalence of CKD
patients with kidney disease continues to be debated.
Several studies have provided estimates of CKD prevalence
from different jurisdictions around the world, with
Diabetes Mellitus
estimates between 8% and 16% [66]. In the United States,
Diabetes continues to be a major contributor to CKD and prevalence data come from a series of national surveys
associated outcomes. While the rates of diabetes-related called the National Health and Nutrition Examination
complications including ESKD have reduced over time in Survey (NHANES), a stratified cluster probability sample
relative terms, the prevalence of diabetes has continued to of US adults. The application of survey weights provides
increase such that the absolute rates of complications are representative estimates of disease prevalence. Data from
expected to continue to rise [49, 50]. The risk of kidney NHANES suggest an overall prevalence of CKD in the
disease attributable to diabetes in large population studies United States of 13%, with the majority of individuals
is subject to misclassification. Individuals with diabetes with CKD having mild to moderate reductions in eGFR
tend to have a higher prevalence of other CKD risk factors (30–59 ml/min/1.73 m2) and less than 1% having more
such as obesity and hypertension. As such, in the absence severe disease [67]. European prevalence data from various
of a tissue diagnosis the cause of their kidney disease can- cohorts were recently published from the European CKD
not be definitively ascribed to diabetes. This limitation Burden Consortium [68]. In keeping with US data, the
has historically hampered efforts to understand mecha- investigators documented an overall high prevalence of
nisms in diabetic kidney disease and has motivated initia- CKD. This group also found substantial heterogeneity in
tives like the Kidney Precision Medicine Project [51]. The the methodology of included studies, for example different
landscape of diabetic kidney disease could change dra- definitions of CKD using eGFR, ACR, or both, varying
matically with the advent of sodium-glucose cotrans- sample response rates, use or nonuse of a sampling frame,
porter type 2 (SGLT2) inhibition as a therapeutic target. and lack of standardization of assays for creatinine [69].
Clinical trials of SGLT2 inhibitors have demonstrated This heterogeneity poses a significant challenge when
improvements not only in glycemic control, but also in seeking to understand the true prevalence of CKD and is
cardiovascular and renal endpoints [52, 53]. These effects discussed further in Chapter 2.
may be mediated through beneficial changes in blood An important, but perhaps under-appreciated, source
pressure, glomerular hemodynamics, body weight, and of bias in prevalence studies is misclassification of CKD
uric acid [54]. stage based on a single measurement of eGFR or ACR.
While it is understandable from a feasibility perspective
that large population studies would not perform a second
Acute Kidney Injury
test of eGFR and ACR after 3 months to fulfill the chro-
AKI may be defined as a transient reduction in kidney nicity criterion of the KDIGO definition, recent studies
function, detected clinically by a change in creatinine or a have demonstrated the potentially large degree of
isclassification that can arise from a single test [70]. The

m Disease (GBD) studies. Since 1992, the GBD investigators
same is true of prognostic studies, yet the implication of have ranked individual diseases based on the number of
misclassification in that context is unclear. The amount of global deaths and disability-adjusted life years (DALY)
albumin excretion can vary significantly within an indi- attributable to each disease [80]. DALY is a time-based
vidual [71], while eGFR can be affected by a host of fac- measure that combines years of life lost due to morbidity
tors related to creatinine generation such as changes in and premature mortality, and was developed with the aim
diet and body composition [72]. of informing cost-effective analysis and public health pol-
icy. The scope of the GBD study continues to expand since
its first report in 1990. The 2015 iteration reported data
CKD in the Developing World for 300 diseases in 195 countries [81]. Those 25 years have
witnessed an ominous trend for CKD, which is now the
The bulk of our knowledge of CKD epidemiology comes 12th leading cause of death worldwide compared to a
from studies undertaken in high income countries with suf- ranking of 27th in 1990 [82]. Mortality attributable to
ficient data infrastructure to reliably estimate prevalence and CKD is an estimated 1.2 million deaths, representing a
incidence rates. The majority of the world’s population lives rate of 19.2 per 100 000 persons per year. Cardiovascular
in middle-income and low-income countries, where our disease is the biggest contributor to that mortality
understanding of the burden of kidney disease is sorely lack- rate [83]. Worryingly, this mortality rate may be an under-
ing. What is evident from the available data is that this bur- estimate, given that CKD is not always listed as a cause of
den is expected to increase dramatically in the coming death, for example in the setting of diabetes [84]. In keep-
decades due in large part to growing rates of diabetes, obesity, ing with mortality data, CKD is also a major contributor
and hypertension in the developing world [73]. Infection to DALYs. In some parts of the world, including Latin
continues to be a major driver of CKD in these countries, America and South-East Asia, CKD is currently ranked in
including the many and varied renal manifestations of infec- the top 10 for DALYs [81, 82]. It is evident, therefore, that
tion with human immunodeficiency virus [74, 75]. A current the impact of CKD on population health is considerable,
area of intensive research efforts is chronic kidney disease of and growing.
unknown etiology (CKDu), a condition highly prevalent
among young and middle-aged male agricultural and planta-
tion workers. Affected individuals lack traditional risk factors
CKD in Specific Subgroups
for kidney disease, and a number of pathogenic factors have
Older Adults
been hypothesized including heat stress, dehydration, and
exposure to heavy metals [76]. The final common pathway The diagnosis of CKD in older adults has been one of the
appears to be interstitial inflammation and fibrosis, as sug- more controversial aspects of the CKD classification sys-
gested in a biopsy series from Central America [77]. Similarly, tem. Measurement studies from decades ago demonstrated
in a large series from Sri Lanka, half of patients who under- an age-related decline in GFR. Similar findings have been
went kidney biopsy had evidence of a primary tubulointersti- shown in contemporary large population studies using
tial kidney disease [78]. Our knowledge of kidney disease estimated GFR. For example, in a cross-sectional study
epidemiology and appreciation of underlying mechanisms in from the Netherlands, the expected decline in eGFR was
developing countries is in its infancy. The multiple and 0.4 ml/min/1.73 m2 per year of age among apparently
diverse challenges that need to be overcome in order for us to healthy adults with no renal or cardiovascular risk fac-
enhance our understanding of, and effect meaningful change tors [85]. The current eGFR diagnostic threshold of 60 ml/
in, kidney disease burden among disadvantaged populations min/1.73 m2, without albuminuria or other evidence of
are discussed in detail in Chapter 6. kidney damage, captures a substantial proportion of older
community-dwelling adults as having CKD. Among indi-
viduals over 70 years of age, prevalence estimates have
idney Disease as a Global Public
K been as high as 40% [67, 68]. The majority of older adults
Health Priority with CKD have relatively modest reductions in eGFR (45–
59 ml/min/1.73 m2). It has been argued that the threshold
The importance of understanding the epidemiology of to define CKD in otherwise healthy adults over the age of
CKD is underscored by the impact of CKD on health at a 60 years should be lowered to an eGFR < 45 ml/min/1.73 m2
population level around the world [79]. The magnitude of to avoid overdiagnosis and disease-labeling in this subpop-
this impact, and how CKD compares to other chronic ulation [86]. Central to this viewpoint is the recognition of
conditions, has been quantified by the Global Burden of a mismatch between a seemingly high prevalence of
Knowledge Gaps and Future Direction 9
isease and low incidence of organ failure, partly explained

d of ESKD compared to other racial groups. Recent land-
by competing mortality from other age-related comorbidi- mark genetic studies have shed light on this phenome-
ties [87]. On the other side of the debate, investigators non. Mutations in the apolipoprotein L1 (APOL-1) gene,
point to the risks associated with CKD in older people [88]. found exclusively in individuals of recent African descent
A large meta-analysis leveraging data from over 1 million and protective against trypanosomal infection, are
participants from 33 general population and high-risk (of strongly associated with nondiabetic kidney disease,
vascular disease) cohorts specifically addressed this ques- including that attributed to idiopathic focal segmental
tion by examining the interaction of age and kidney glomerulosclerosis and hypertension [97, 98]. APOL-1
parameters for the risk of ESKD and mortality [89]. This variants have also been shown to confer a higher likeli-
study demonstrated that, while the relationship between hood of CKD progression [99]. The potential influence of
eGFR/ACR and outcomes varied by age, a reduced eGFR race and ethnicity on kidney outcomes around the world
or increased ACR was associated with increased risk of is not well understood due to a lack of data infrastructure
ESKD and mortality across all age categories. In addition and high-quality studies (see Chapter 2). While preva-
to hard outcomes, older adults with CKD are also at higher lence rates of CKD vary by race, data from the CKD-PC
risk of more proximal outcomes. Several studies have suggest that the risk relationships between reduced eGFR,
identified CKD as an independent risk factor for reduced albuminuria, and clinical outcomes are consistent across
physical performance, frailty, and cognitive impair- racial groups [100]. On a broader level, the use of race in
ment [90–92]. Thus, the notion of changing the threshold medicine has come under increasing scrutiny, particu-
for the definition of CKD based on age alone is not sub- larly the inclusion of race in diagnostic and treatment
stantiated with data; and would be analogous to changing algorithms [101]. From a nephrology perspective, there is
thresholds for other diseases such as diabetes or hyperten- a growing concern that the inclusion of a race coefficient
sion based on age. The implications of a diagnosis of CKD, in GFR estimation and in the evaluation of organ donors
potential interventions, and targets of therapy may differ is disadvantaging Black patients [102]. In 2021, a task
between older and younger individuals, but the actual force jointly created by the National Kidney Foundation
definition of CKD is not age-dependent. and American Society of Nephrology recommended the
implementation of a new race-free GFR estimating equa-
tion published by the CKD-EPI investigators [103, 104].
Sex Differences in Kidney Health
The implications of this for clinical decision-making and
and Outcomes
health policy are currently being evaluated, but it is hoped
The prevalence of CKD appears to be higher in women that this change will go some way towards achieving
compared to men, estimated at 16.5% and 13%, respec- equity for all patients with kidney disease.
tively, in the United States [93]. Possible reasons for this
discrepancy include a shorter life expectancy in men and
potential misclassification of CKD using GFR estimating nowledge Gaps and Future
K
equations in the general population [94]. On the other Directions
hand, proportionately more men start renal replacement
therapy owing, at least in part, to faster decline in kidney Our knowledge and understanding of kidney disease
function compared to women. The association between epidemiology has grown exponentially over the last
eGFR and albuminuria and hard outcomes is consistent in 15 years, which is testament to the intensive and rigor-
both sexes, although the slope of the relationship with ous research efforts of the international nephrology
mortality may be steeper in women [95]. While understud- community. Inevitably, as more data accrue more ques-
ied, women also have unique risk factors for kidney dis- tions will arise. The CKD classification system provided
ease, including pregnancy complications and higher a common platform from which to start from. There is
incidence of systemic autoimmune disease [96]. now a need for improved understanding of variability in
outcomes on the individual patient level to inform
shared decision-making based on individualized risk-
Race/Ethnicity
stratification. Integral to this is accurate definition of the
The impact of race on the incidence and severity of kid- disease phenotype, moving from assumptive to defini-
ney disease has been most comprehensively studied in tive diagnosis through molecular diagnostics or by low-
the United States. It has long been recognized that African ering the threshold for kidney biopsy in “common”
American patients with kidney disease have a higher-risk disorders. The data that currently inform clinical prac-
profile, evidenced by a substantially increased prevalence tice have been almost exclusively acquired from studies
in developed countries with an appropriate data infra- to care, and choice of therapies are fundamentally differ-
structure and access to nephrology care, including renal ent. The subsequent chapters in this part expand on
replacement therapy. This data cannot be extrapolated to these and other important issues, as we strive to meet
the majority of the world’s population in developing the current and future challenges of kidney disease
countries where susceptibility to kidney disease, access around the globe.
References
1 Levey, A.S., Coresh, J., Balk, E. et al. (2003). National 12 Matsushita, K., Coresh, J., Sang, Y. et al. (2015).
Kidney Foundation practice guidelines for chronic kidney Estimated glomerular filtration rate and albuminuria for
disease: evaluation, classification, and stratification. Ann. prediction of cardiovascular outcomes: a collaborative
Intern. Med. 139: 137–147. meta-analysis of individual participant data. Lancet
2 National Kidney F (2002). K/DOQI clinical practice Diabetes Endocrinol. 3: 514–525.
guidelines for chronic kidney disease: evaluation, 13 Matsushita, K., Ballew, S.H., Coresh, J. et al. (2017).
classification, and stratification. Am. J. Kidney Dis. 39: Measures of chronic kidney disease and risk of incident
S1–S266. peripheral artery disease: a collaborative meta-analysis of
3 Levin, A. (2003). The advantage of a uniform terminology individual participant data. Lancet Diabetes Endocrinol. 5:
and staging system for chronic kidney disease (CKD). 718–728.
Nephrol. Dial. Transplant. 18: 1446–1451. 14 Aguilar, M.I., O’Meara, E.S., Seliger, S. et al. (2010).
4 Hsu, C.Y. and Chertow, G.M. (2000). Chronic renal Albuminuria and the risk of incident stroke and stroke
confusion: insufficiency, failure, dysfunction, or disease. types in older adults. Neurology 75: 1343–1350.
Am. J. Kidney Dis. 36: 415–418. 15 Tonelli, M., Muntner, P., Lloyd, A. et al. (2012). Risk of
5 Stevens, P.E. and Levin, A., Kidney Disease: Improving coronary events in people with chronic kidney disease
Global Outcomes Chronic Kidney Disease Guideline compared with those with diabetes: a population-level
Development Work Group M. (2013). Evaluation and cohort study. Lancet 380: 807–814.
management of chronic kidney disease: synopsis of the 16 Baber, U., Gutierrez, O.M., Levitan, E.B. et al. (2013). Risk
kidney disease: improving global outcomes 2012 clinical for recurrent coronary heart disease and all-cause
practice guideline. Ann. Intern. Med. 158: 825–830. mortality among individuals with chronic kidney disease
6 Groopman, E.E., Marasa, M., Cameron-Christie, S. et al. compared with diabetes mellitus, metabolic syndrome,
(2019). Diagnostic utility of exome sequencing for kidney and cigarette smokers. Am. Heart J. 166: 373–380. e372.
disease. N. Engl. J. Med. 380: 142–151. 17 Gansevoort, R.T., Correa-Rotter, R., Hemmelgarn, B.R.
7 Matsushita, K., Ballew, S.H., Astor, B.C. et al. (2013). et al. (2013). Chronic kidney disease and cardiovascular
Cohort profile: the chronic kidney disease prognosis risk: epidemiology, mechanisms, and prevention. Lancet
consortium. Int. J. Epidemiol. 42: 1660–1668. 382: 339–352.
8 Wen, C.P., Cheng, T.Y., Tsai, M.K. et al. (2008). All-cause 18 van der Velde, M., Matsushita, K., Coresh, J. et al. (2011).
mortality attributable to chronic kidney disease: a Lower estimated glomerular filtration rate and higher
prospective cohort study based on 462 293 adults in albuminuria are associated with all-cause and
Taiwan. Lancet 371: 2173–2182. cardiovascular mortality. A collaborative meta-analysis of
9 Chronic Kidney Disease Prognosis Consortium, high-risk population cohorts. Kidney Int. 79: 1341–1352.
Matsushita, K., van der Velde, M. et al. (2010). 19 Mahmoodi, B.K., Matsushita, K., Woodward, M. et al.
Association of estimated glomerular filtration rate and (2012). Associations of kidney disease measures with
albuminuria with all-cause and cardiovascular mortality mortality and end-stage renal disease in individuals with
in general population cohorts: a collaborative meta- and without hypertension: a meta-analysis. Lancet 380:
analysis. Lancet 375: 2073–2081. 1649–1661.
10 Turin, T.C., Tonelli, M., Manns, B.J. et al. (2013). 20 Fox, C.S., Matsushita, K., Woodward, M. et al. (2012).
Proteinuria and life expectancy. Am. J. Kidney Dis. 61: Associations of kidney disease measures with mortality
646–648. and end-stage renal disease in individuals with and
11 Turin, T.C., Tonelli, M., Manns, B.J. et al. (2012). Chronic without diabetes: a meta-analysis. Lancet 380: 1662–1673.
kidney disease and life expectancy. Nephrol. Dial. 21 Gansevoort, R.T., Matsushita, K., van der Velde, M. et al.
Transplant. 27: 3182–3186. (2011). Lower estimated GFR and higher albuminuria are
Reference 11
associated with adverse kidney outcomes. A collaborative factors independently of measured GFR. J. Am. Soc.
meta-analysis of general and high-risk population Nephrol. 22: 927–937.
cohorts. Kidney Int. 80: 93–104. 35 NCD Risk Factor Collaboration (2017). Worldwide trends
22 Astor, B.C., Matsushita, K., Gansevoort, R.T. et al. (2011). in blood pressure from 1975 to 2015: a pooled analysis of
Lower estimated glomerular filtration rate and higher 1479 population-based measurement studies with
albuminuria are associated with mortality and end-stage 19.1 million participants. Lancet 389: 37–55.
renal disease. A collaborative meta-analysis of kidney 36 Fox, C.S., Larson, M.G., Leip, E.P. et al. (2004). Predictors
disease population cohorts. Kidney Int. 79: 1331–1340. of new-onset kidney disease in a community-based
23 Tangri, N., Stevens, L.A., Griffith, J. et al. (2011). A population. JAMA 291: 844–850.
predictive model for progression of chronic kidney 37 Anderson, A.H., Yang, W., Townsend, R.R. et al. (2015).
disease to kidney failure. JAMA 305: 1553–1559. Time-updated systolic blood pressure and the progression
24 Levey, A.S., Stevens, L.A., Schmid, C.H. et al. (2009). A of chronic kidney disease: a cohort study. Ann. Intern.
new equation to estimate glomerular filtration rate. Ann. Med. 162: 258–265.
Intern. Med. 150: 604–612. 38 Kovesdy, C.P., Bleyer, A.J., Molnar, M.Z. et al. (2013).
25 Levey, A.S., Bosch, J.P., Lewis, J.B. et al. (1999). A more Blood pressure and mortality in U.S. veterans with
accurate method to estimate glomerular filtration rate chronic kidney disease: a cohort study. Ann. Intern. Med.
from serum creatinine: a new prediction equation. 159: 233–242.
Modification of Diet in Renal Disease Study Group. Ann. 39 Kovesdy, C.P., Lu, J.L., Molnar, M.Z. et al. (2014).
Intern. Med. 130: 461–470. Observational modeling of strict vs conventional blood
26 Levey, A.S., Coresh, J., Greene, T. et al. (2006). Using pressure control in patients with chronic kidney disease.
standardized serum creatinine values in the JAMA Intern. Med. 174: 1442–1449.
modification of diet in renal disease study equation for 40 Weiss, J.W., Peters, D., Yang, X. et al. (2015). Systolic BP
estimating glomerular filtration rate. Ann. Intern. Med. and mortality in older adults with CKD. Clin. J. Am. Soc.
145: 247–254. Nephrol. 10: 1553–1559.
27 Matsushita, K., Mahmoodi, B.K., Woodward, M. et al. 41 Herrington, W., Staplin, N., Judge, P.K. et al. (2017).
(2012). Comparison of risk prediction using the CKD-EPI Evidence for reverse causality in the association
equation and the MDRD study equation for estimated between blood pressure and cardiovascular risk in
glomerular filtration rate. JAMA 307: 1941–1951. patients with chronic kidney disease. Hypertension 69:
28 Schaeffner, E.S., Ebert, N., Delanaye, P. et al. (2012). Two 314–322.
novel equations to estimate kidney function in persons 42 Rothwell, P.M. (2010). Limitations of the usual blood-
aged 70 years or older. Ann. Intern. Med. 157: 471–481. pressure hypothesis and importance of variability,
29 Stevens, L.A., Coresh, J., Greene, T. et al. (2006). instability, and episodic hypertension. Lancet 375:
Assessing kidney function – measured and estimated 938–948.
glomerular filtration rate. N. Engl. J. Med. 354: 43 Rothwell, P.M., Howard, S.C., Dolan, E. et al. (2010).
2473–2483. Prognostic significance of visit-to-visit variability,
30 Dharnidharka, V.R., Kwon, C., and Stevens, G. (2002). maximum systolic blood pressure, and episodic
Serum cystatin C is superior to serum creatinine as a hypertension. Lancet 375: 895–905.
marker of kidney function: a meta-analysis. Am. J. Kidney 44 Whittle, J., Lynch, A.I., Tanner, R.M. et al. (2016).
Dis. 40: 221–226. Visit-to-visit variability of BP and CKD outcomes:
31 Inker, L.A., Schmid, C.H., Tighiouart, H. et al. (2012). results from the ALLHAT. Clin. J. Am. Soc. Nephrol. 11:
Estimating glomerular filtration rate from serum 471–480.
creatinine and cystatin C. N. Engl. J. Med. 367: 20–29. 45 Xin, W., Lin, Z., and Mi, S. (2014). Orthostatic
32 Shlipak, M.G., Matsushita, K., Arnlov, J. et al. (2013). hypotension and mortality risk: a meta-analysis of cohort
Cystatin C versus creatinine in determining risk based on studies. Heart 100: 406–413.
kidney function. N. Engl. J. Med. 369: 932–943. 46 Fedorowski, A., Stavenow, L., Hedblad, B. et al. (2010).
33 Rule, A.D., Bailey, K.R., Lieske, J.C. et al. (2013). Orthostatic hypotension predicts all-cause mortality and
Estimating the glomerular filtration rate from serum coronary events in middle-aged individuals (The Malmo
creatinine is better than from cystatin C for evaluating Preventive Project). Eur. Heart J. 31: 85–91.
risk factors associated with chronic kidney disease. 47 Franceschini, N., Rose, K.M., Astor, B.C. et al. (2010).
Kidney Int. 83: 1169–1176. Orthostatic hypotension and incident chronic kidney
34 Mathisen, U.D., Melsom, T., Ingebretsen, O.C. et al. disease: the atherosclerosis risk in communities study.
(2011). Estimated GFR associates with cardiovascular risk Hypertension 56: 1054–1059.
8 Canney, M., O’Connell, M.D.L., Sexton, D.J. et al. (2017).

4 63 Grams, M.E., Sang, Y., Ballew, S.H. et al. (2015). A
Graded association between kidney function and meta-analysis of the association of estimated GFR,
impaired orthostatic blood pressure stabilization in older albuminuria, age, race, and sex with acute kidney injury.
adults. J. Am. Heart Assoc. 6. Am. J. Kidney Dis. 66: 591–601.
49 Gregg, E.W., Li, Y., Wang, J. et al. (2014). Changes in 64 James, M.T., Hemmelgarn, B.R., Wiebe, N. et al. (2010).
diabetes-related complications in the United States, Glomerular filtration rate, proteinuria, and the incidence
1990–2010. N. Engl. J. Med. 370: 1514–1523. and consequences of acute kidney injury: a cohort study.
50 Boyle, J.P., Thompson, T.J., Gregg, E.W. et al. (2010). Lancet 376: 2096–2103.
Projection of the year 2050 burden of diabetes in the US 65 Coca, S.G., Singanamala, S., and Parikh, C.R. (2012).
adult population: dynamic modeling of incidence, Chronic kidney disease after acute kidney injury: a
mortality, and prediabetes prevalence. Popul. Health systematic review and meta-analysis. Kidney Int. 81:
Metrics 8: 29. 442–448.
51 Saulnier, P.J. and Nelson, R.G. (2017). Burden of proof- 66 Jha, V., Garcia-Garcia, G., Iseki, K. et al. (2013). Chronic
when is kidney disease attributable to diabetes? Clin. J. kidney disease: global dimension and perspectives.
Am. Soc. Nephrol. 12: 1917–1918. Lancet 382: 260–272.
52 Wanner, C., Inzucchi, S.E., Lachin, J.M. et al. (2016). 67 Coresh, J., Selvin, E., Stevens, L.A. et al. (2007).
Empagliflozin and progression of kidney disease in type 2 Prevalence of chronic kidney disease in the United States.
diabetes. N. Engl. J. Med. 375: 323–334. JAMA 298: 2038–2047.
53 Zinman, B., Wanner, C., Lachin, J.M. et al. (2015). 68 Bruck, K., Stel, V.S., Gambaro, G. et al. (2016). CKD
Empagliflozin, cardiovascular outcomes, and mortality in prevalence varies across the European general
type 2 diabetes. N. Engl. J. Med. 373: 2117–2128. population. J. Am. Soc. Nephrol. 27: 2135–2047.
54 van Bommel, E.J., Muskiet, M.H., Tonneijck, L. et al. 69 Bruck, K., Jager, K.J., Dounousi, E. et al. (2015).
(2017). SGLT2 inhibition in the diabetic kidney-from Methodology used in studies reporting chronic kidney
mechanisms to clinical outcome. Clin. J. Am. Soc. disease prevalence: a systematic literature review.
Nephrol. 12: 700–710. Nephrol. Dial. Transplant. 30 (Suppl 4): iv6–iv16.
55 Mehta, R.L., Kellum, J.A., Shah, S.V. et al. (2007). Acute 70 Benghanem Gharbi, M., Elseviers, M., Zamd, M. et al.
kidney injury network: report of an initiative to improve (2016). Chronic kidney disease, hypertension, diabetes,
outcomes in acute kidney injury. Crit. Care 11: R31. and obesity in the adult population of Morocco: how to
56 Lafrance, J.P. and Miller, D.R. (2010). Acute kidney injury avoid “over”-and “under”-diagnosis of CKD. Kidney Int.
associates with increased long-term mortality. J. Am. Soc. 89: 1363–1371.
Nephrol. 21: 345–352. 71 Naresh, C.N., Hayen, A., Weening, A. et al. (2013).
57 Odutayo, A., Wong, C.X., Farkouh, M. et al. (2017). AKI Day-to-day variability in spot urine albumin-creatinine
and long-term risk for cardiovascular events and ratio. Am. J. Kidney Dis. 62: 1095–1101.
mortality. J. Am. Soc. Nephrol. 28: 377–387. 72 Fotheringham, J., Weatherley, N., Kawar, B. et al. (2014).
58 Wald, R., Quinn, R.R., Luo, J. et al. (2009). Chronic The body composition and excretory burden of lean,
dialysis and death among survivors of acute kidney injury obese, and severely obese individuals has implications for
requiring dialysis. JAMA 302: 1179–1185. the assessment of chronic kidney disease. Kidney Int. 86:
59 Jones, J., Holmen, J., De Graauw, J. et al. (2012). 1221–1228.
Association of complete recovery from acute kidney 73 Sharma, S. and Sarnak, M.J. (2017). Epidemiology: The
injury with incident CKD stage 3 and all-cause mortality. global burden of reduced GFR: ESRD, CVD and
Am. J. Kidney Dis. 60: 402–408. mortality. Nat. Rev. Nephrol. 13: 447–448.
60 Heung, M., Steffick, D.E., Zivin, K. et al. (2016). Acute 74 Cohen, S.D., Kopp, J.B., and Kimmel, P.L. (2018). Kidney
kidney injury recovery pattern and subsequent risk of diseases associated with human immunodeficiency virus
CKD: an analysis of veterans health administration data. infection. N. Engl. J. Med. 378: 1655–1656.
Am. J. Kidney Dis. 67: 742–752. 75 Swanepoel, C.R., Atta, M.G., D’Agati, V.D. et al. (2018).
61 Chawla, L.S., Eggers, P.W., Star, R.A. et al. (2014). Acute Kidney disease in the setting of HIV infection: conclusions
kidney injury and chronic kidney disease as from a kidney disease: improving global outcomes
interconnected syndromes. N. Engl. J. Med. 371: 58–66. (KDIGO) controversies conference. Kidney Int. 93:
62 Grams, M.E., Astor, B.C., Bash, L.D. et al. (2010). 545–559.
Albuminuria and estimated glomerular filtration rate 76 Lunyera, J., Mohottige, D., Von Isenburg, M. et al. (2016).
independently associate with acute kidney injury. J. Am. CKD of uncertain etiology: a systematic review. Clin. J.
Soc. Nephrol. 21: 1757–1764. Am. Soc. Nephrol. 11: 379–385.
Reference 13
7 Fischer, R.S.B., Vangala, C., Truong, L. et al. (2018). Early

7 92 Canney, M., Sexton, D.J., O’Connell, M.D.L. et al. (2017).
detection of acute tubulointerstitial nephritis in the Kidney function estimated from cystatin C, but not
genesis of Mesoamerican nephropathy. Kidney Int. 93: creatinine, is related to objective tests of physical
681–690. performance in community-dwelling older adults. J.
78 Anand, S., Montez-Rath, M.E., Adasooriya, D. et al. Gerontol. A Biol. Sci. Med. Sci. 72: 1554–1560.
(2019). Prospective biopsy-based study of chronic kidney 93 Saran, R., Robinson, B., Abbott, K.C. et al. (2017). US
disease of unknown etiology in Sri Lanka. Clin. J. Am. renal data system 2016 annual data report: epidemiology
Soc. Nephrol. 14: 224–232. of kidney disease in the United States. Am. J. Kidney Dis.
79 Eckardt, K.U., Coresh, J., Devuyst, O. et al. (2013). 69: A7–A8.
Evolving importance of kidney disease: from subspecialty 94 Carrero, J.J., Hecking, M., Chesnaye, N.C. et al. (2018).
to global health burden. Lancet 382: 158–169. Sex and gender disparities in the epidemiology and
80 Vos, T., Flaxman, A.D., Naghavi, M. et al. (2012). Years outcomes of chronic kidney disease. Nat. Rev. Nephrol.
lived with disability (YLDs) for 1160 sequelae of 289 14: 151–164.
diseases and injuries 1990-2010: a systematic analysis for 95 Nitsch, D., Grams, M., Sang, Y. et al. (2013).
the Global Burden of Disease Study 2010. Lancet 380: Associations of estimated glomerular filtration rate and
2163–2196. albuminuria with mortality and renal failure by sex: a
81 GBD 2015 DALYs and HALE Collaborators (2016). meta-analysis. BMJ 346: f324.
Global, regional, and national disability-adjusted 96 Piccoli, G.B., Liu, Z.H., and Levin, A. (2018). Women
life-years (DALYs) for 315 diseases and injuries and and kidney diseases: reflection on World Kidney Day
healthy life expectancy (HALE), 1990–2015: a systematic and International Woman’s Day. Arch. Argent. Pediatr.
analysis for the Global Burden of Disease Study 2015. 116: e273–e278.
Lancet 388: 1603–1658. 97 Freedman, B.I., Kopp, J.B., Langefeld, C.D. et al. (2010).
82 Jager, K.J. and Fraser, S.D.S. (2017). The ascending rank The apolipoprotein L1 (APOL1) gene and nondiabetic
of chronic kidney disease in the global burden of disease nephropathy in African Americans. J. Am. Soc. Nephrol.
study. Nephrol. Dial. Transplant. 32: ii121–ii128. 21: 1422–1426.
83 Thomas, B., Matsushita, K., Abate, K.H. et al. (2017). 98 Genovese, G., Friedman, D.J., Ross, M.D. et al. (2010).
Global cardiovascular and renal outcomes of reduced Association of trypanolytic ApoL1 variants with
GFR. J. Am. Soc. Nephrol. 28: 2167–2179. kidney disease in African Americans. Science 329:
84 Rao, C., Adair, T., Bain, C. et al. (2012). Mortality from 841–845.
diabetic renal disease: a hidden epidemic. Eur. J. Pub. 99 Parsa, A., Kao, W.H., Xie, D. et al. (2013). APOL1 risk
Health 22: 280–284. variants, race, and progression of chronic kidney
85 Wetzels, J.F., Kiemeney, L.A., Swinkels, D.W. et al. (2007). disease. N. Engl. J. Med. 369: 2183–2196.
Age-and gender-specific reference values of estimated 100 Wen, C.P., Matsushita, K., Coresh, J. et al. (2014).
GFR in Caucasians: the Nijmegen Biomedical Study. Relative risks of chronic kidney disease for mortality
Kidney Int. 72: 632–637. and end-stage renal disease across races are similar.
86 Glassock, R., Delanaye, P., and El Nahas, M. (2015). An Kidney Int. 86: 819–827.Epidemiology
age-calibrated classification of chronic kidney disease. Summaries and Visions
JAMA 314: 559–560. 101 Vyas, D.A., Eisenstein, L.G., Jones, D.S. (2020). Hidden
87 O’Hare, A.M., Choi, A.I., Bertenthal, D. et al. (2007). Age in plain sight – reconsidering the use of race correction
affects outcomes in chronic kidney disease. J. Am. Soc. in clinical algorithms. N. Engl. J. Med. 383: 874–882.
Nephrol. 18: 2758–2765. 102 Gopalakrishnan, C., Patorno, E. (2021). Time to end the
88 Levey, A.S., Inker, L.A., and Coresh, J. (2015). Chronic misuse of race in medicine: cases from nephrology. BMJ.
kidney disease in older people. JAMA 314: 557–558. 375:n2435.
89 Hallan, S.I., Matsushita, K., Sang, Y. et al. (2012). Age and 103 Delgado, C., Baweja, M., Crews, D.C. et al (2021). A
association of kidney measures with mortality and Unifying Approach for GFR Estimation:
end-stage renal disease. JAMA 308: 2349–2360. Recommendations of the NKF-ASN Task Force on
90 Dalrymple, L.S., Katz, R., Rifkin, D.E. et al. (2013). Reassessing the Inclusion of Race in Diagnosing Kidney
Kidney function and prevalent and incident frailty. Clin. Disease. J. Am. Soc. Nephrol. (published on-line ahead
J. Am. Soc. Nephrol. 8: 2091–2099. of print).
91 Roshanravan, B., Khatri, M., Robinson-Cohen, C. et al. 104 Inker, L.A., Eneanya, N.D., Coresh, J. et al (2021). New
(2012). A prospective study of frailty in nephrology-referred Creatinine- and Cystatin C-Based Equations to Estimate
patients with CKD. Am. J. Kidney Dis. 60: 912–921. GFR without Race. N. Engl. J. Med. 385: 1737–1749.
14
The Surveillance and Burden of Chronic Kidney Disease

Chaisiri Angkurawaranon1, Dorothea Nitsch2, and Ben Caplin3
1
Department of Family Medicine, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand
2
Department of Non-communicable Disease Epidemiology, London School of Hygiene and Tropical Medicine, London, United Kingdom
3
Department of Renal Medicine, University College London, London, United Kingdom
Surveillance of CKD benefit from early referral to specialist services [13].

However, the best approaches to identify this group remain
The World Health Organization defines surveillance as the the subject of ongoing research [14–16]. Combining test-
“continuous, systematic collection, analysis and interpre- ing for CKD with screening for other noncommunicable
tation of health-related data for the planning, implementa- diseases or risk factors means that surveillance for CKD in
tion and evaluation of public health practices.” In terms of high-risk populations may be the most efficient and effec-
chronic kidney disease (CKD), surveillance refers to activi- tive approach [1, 17]. These targeted populations include
ties aimed at providing key information on the burden of, those with hypertension, diabetes, and cardiovascular dis-
and risk factors for, CKD with a view to reducing variation ease [18–21]. Aside from benefits at the patient level, esti-
in, and improving the quality of, care processes and health mates of the projected incidence of ESRD and dialysis
outcomes associated with the condition [1]. Information needs in a local population are key for health services
from CKD surveillance can potentially be used to guide planning. Furthermore, understanding prevalence of
resource planning and implementation of medical and CKD in a population may have broader implications given
public health measures along with informing health policy. the relationship between CKD and hospitalizations from
Optimization in the management of CKD could prevent multiple causes such as cardiovascular events, acute kid-
progression to kidney failure and the need for dialysis and ney injury (AKI), and infections [3].
transplantation [2] as well as potentially reducing the For patients with ESRD (sometimes also including those
burden of hospitalization from nonrenal complications with the advanced stages of CKD), many countries have reg-
(e.g. cardiovascular disease) [3]. istries where treatments and outcomes are monitored [22].
As well as providing data on the incidence of RRT in a par-
ticular population, these registries aim to guide manage-
Potential Benefits of CKD Surveillance ment and improve care processes toward better health
While there is insufficient evidence to support the efficacy outcomes [23, 24].
of screening for CKD among the general population [4–6],
consistent evidence suggests that presence of proteinuria
Types of CKD Surveillance
or reduced estimated glomerular filtration rate (eGFR) is
strongly associated with end-stage renal disease (ESRD), Surveillance systems are generally divided into passive and
cardiovascular, and all-cause mortality in population- active approaches. A passive surveillance system relies on
based studies [7–10]. Equations that predict progression of data routinely collected in current practice. These include
CKD have been validated in high-income settings [11, 12]. death registries, hospital records, and primary care records
The absolute numbers reaching ESRD are small, but those (Table 2.1). These types of routine data sources are efficient
who do go onto need renal replacement therapy (RRT) and powerful tools in CKD surveillance, particularly where
Table 2.1 Types of passive surveillance for CKD.
Passive Potential Potential

surveillance indicators Numerator Denominator Generalizability strengths Potential limitations Potential biases
Death Mortality rate Deaths attributed to Population Local and national Resource Attributing deaths to Coverage and validity of death
registries Age-and CKD/ESRD covered by death level efficient CKD/ESRD certificates
sex-specific registration International Explore Disaggregation by area/
mortality rate comparisons (SMRs) geographical populations
SMR variations
Dialysis Incidence of Patients with ESRD Population with Patients with Resource Not representative of Coverage of RRT
registries and ESRD starting RRT because access to dialysis coverage for ESRD efficient burden of CKD Access to care
transplant Age-specific of ESRD (requiring and Explore (untreated ESRD) Competing mortality
registries incidence of dialysis for 90 days or transplantation geographical Obtaining patient-level
ESRD more) variations data
Standardized Limited data on
incidence of treatment and patient
ESRD outcomes
Hospital Prevalence of Patients diagnosed Patients who are Patients who seek or Availability of Not representative of Classification based on eGFR
records CKD and/or with CKD and/or in hospital care are referred to kidney (bio) true population and/or ICD 10
(admissions ESRD ESRD hospital care markers prevalence/incidence of Influence of AKI on results
data) Availability of CKD Creatinine calibration and
treatment and Completeness of method used
patient information Referral patterns from primary
outcome data care
Competing mortality
Primary care Incidence and Patients diagnosed Patients in People who are Detection of Not representation of selection bias due to self-
records prevalence of with CKD (from outpatient/ screened/attend early stages of true population presentation or volunteer of
CKD and/or kidney function primary care checkups or present CKD prevalence/incidence of community population
ESRD measures) for outpatient care CKD (rather among Classification based on eGFR
those who regularly get and ICD 10
renal function tests) Single measurement or
Completeness of multiple measure for diagnosis
information of CKD
Influence of AKI on results
Laboratory calibration issues
AKI, acute kidney injury; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; ICD 10, International Classification of Diseases-10; RRT,
renal replacement therapy; SMR, standardized mortality rate.
0005152390.INDD 15 09-12-2022 12:38:25

16 The Surveillance and Burden of Chronic Kidney Disease
laboratory test results are available in parallel to clinical the prevalence of reduced eGFR to monitor CKD at popu-
data. Outputs such as attributable deaths from CKD, and lation level, examples include the National Health and
prevalence or incidence of CKD can be derived. However, Nutrition Examination Survey (NHANES) [25] and the
there are important limitations to these datasets which Health Survey for England [26]. Although the single time
mostly surround data quality, coverage, and comprehen- point measurement in a survey participant does not meet
siveness. These limitations will vary substantially both criteria for CKD per se, as chronicity is not demonstrated,
between different types of electronic health records and conclusions can still be drawn about CKD prevalence,
between different health systems. Specifically, variation in assuming the methods used (at different points or in differ-
access to healthcare in some settings may lead to major ent regions if comparing populations) remain the same
issues with the representativeness of any estimates of dis- and survey nonresponse is appropriately dealt with.
ease burden, therefore the interpretation and generaliza-
bility of the conclusions derived from these data sets Registries of Patients Receiving RRT
requires a clear understanding of the context and ascer- Many health systems will use incidence of RRT as a meas-
tainment methods within each system. ure of CKD burden. However, there are two important
Active surveillance systems, such as national surveys or limitations in using this measure. First, it is based on the
outreach testing in high-risk populations, are often more assumption that all patients with ESRD can access dialysis
resource intensive than passive surveillance systems. With or transplantation. Across the world there are inequalities
an active system the accuracy of the diagnosis of CKD may in accessing care [27] and even in Europe substantial vari-
be improved and a wider range of patient factors, treat- ations exists [28]. Furthermore, health system incentives
ment, and health outcomes may potentially be explored. may influence the numbers starting RRT. Second, in an
However, with the exception of population-based surveys, attempt to capture those receiving RRT due to CKD rather
these data will not provide a measure of the population than AKI many registries only collect data on patients who
prevalence of CKD (Table 2.2). have received dialysis for a minimum duration, for exam-
In reality, most comprehensive health systems will ple 90 days. This means that all those patients who started
undertake a variety of surveillance approaches using dif- dialysis because of ESRD but died within a 90-day window
ferent data sources. Typically these might include death are not included in these estimates, potentially leading to
certification data, a dialysis and transplant registry, elec- an underestimate of disease burden.
tronic health record-based surveillance and targeted out-
reach in high-risk groups. Indeed unexpected differences CKD Outreach Screening
in estimates of disease burden or outcomes from these dif- Although the use of outreach approaches to detect CKD in
ferent sources may in turn provide key information about populations has the advantage that it can be combined
population coverage and patient care pathways within or with initiatives to improve the awareness of CKD, esti-
between regions. mates of disease burden from this type of work will be criti-
cally dependent on who presents for testing and hence
potentially misleading. This bias may drive estimates
Challenges of Surveillance for CKD
higher or lower, for example it may be that the more health-
Surveillance approaches have different challenges and lim- conscious members of the population are captured (“the
itations. Understanding these is key to the interpretation of worried well”), who are less likely to have kidney prob-
surveillance outputs as well as finding methods to mitigate lems. Alternatively, those who have risk factors for, or fam-
these problems. Any form of CKD surveillance should pro- ily members with, CKD, might seek the opportunity for
vide meaningful data as to whether burden of disease is kidney function testing that may otherwise be difficult to
increasing, decreasing, or stable in a population and many access. Hence, any estimate of disease burden that does not
outputs can also be used to detect differences between include a representative sample of the source population of
populations. interest (best ensured by random sampling) needs to be
treated with caution and may not provide useful insight
Cross-sectional Surveys of eGFR into the true prevalence of CKD or trends over time in dis-
Even an imperfect surveillance method, which misclassi- ease burden.
fies an individual as to whether they have CKD, might still
be useful to detect meaningful differences in CKD preva- Routine Health Records
lence between populations or changes of CKD prevalence Passive systems utilizing available data can have many of
over time within a population. This is the basis of a number similar problems to active approaches, in that the interpre-
of single time point, or snapshot, approaches quantifying tation of the data is dependent on the source population,
Table 2.2 Types of active surveillance for CKD.
Potential
Active surveillance indicators Numerator Denominator Generalizability Potential strengths Potential limitations Potential biases
National surveys Prevalence Participants Random/ Population targeted Representative of true Requires sampling Nonresponse bias
(population-based of CKD diagnosed with stratified sample by survey population incidence and frame, or
studies) and/or CKD (from of population prevalence of CKD in enumeration of
ESRD kidney function targeted by population targeted by population
measures) survey survey Obtaining resources
and Implementation
of fieldwork
Clinic/hospital- Prevalence Participants Hospital/clinic Self-presenting or Completeness of kidney Not representation of Selection bias due to
based surveys of CKD diagnosed with population population eligible (bio)markers true population self-presentation or
(testing of patients and/or CKD (from attending clinic for attending Completeness of treatment prevalence/incidence healthcare access
not routinely tested) ESRD kidney function hospital clinic and patient outcomes data of CKD Nonresponse bias
measures)
Outreach testing Prevalence Participants Selected Prevalence of CKD Completeness of kidney Not representation of Selection bias due to
of CKD diagnosed with reporting in selected (bio)markers true population self-presentation or
and/or CKD (from populations, populations Completeness of treatment prevalence/incidence volunteer of community
ESRD kidney function often volunteers and patient outcomes data of CKD population
measures)
Occupational or Prevalence Participants Random Generalization Less prone to nonresponse Identifying Selection bias restricts
workplace-based of CKD diagnosed with sample/stratifies within population bias appropriate generalizability, especially
surveys and/or CKD (from or whole of survey comparator if particular occupational
ESRD kidney function workforce population groups do not attend
measures)
CKD, chronic kidney disease; ESRD, end-stage renal disease.
0005152390.INDD 17 09-12-2022 12:38:25

i.e. in these cases the population that attended the respec- utputs from Estimates of CKD
O
tive clinical service. In the context of CKD there is also the Burden from Around the World
issue that diagnosis is often based on blood and urine tests
which the treating clinician must actively order, and there As discussed above, the population burden of CKD can be
may be reasons why certain groups of people are more/less estimated using a range of approaches. The corresponding
often tested. For example, a particular group of patients range of estimates will be of more or less valuable depend-
may have differing access to care, e.g. because of lack of ing on the purpose for which the data are sought (provision
insurance or because they are too unwell to attend routine of basic medical services, RRT planning, etiological insight,
appointments. Furthermore, a patient may receive a kid- or an understanding of secular or regional trends) as well
ney function test as the clinician worries about the poten- as the access the population has to specific healthcare
tial of AKI as the patient is acutely unwell, when a kidney resources.
function test may not otherwise be done, a situation that
would lead to an underestimate of kidney function in the
tested population. Therefore, surveillance for CKD using RRT Registries
hospital or primary care records should use results from
RRT incidence and prevalence are widely reported as part
steady state routine testing, i.e. when the patient was not
of nephrology society or health service-led registry initia-
acutely unwell. Furthermore, any changes in prevalence or
tives and the number of registries capturing these data
incidence over time may be driven by differences in the
worldwide is steadily increasing. Most of these registries
way people were tested and/or managed rather than
were initially developed to describe variation in provision
changes in disease burden. For example, changes in patient
and outcomes of dialysis but the same datasets can provide
pathway may explain apparent changes in prevalence, e.g.
a surrogate for the incidence of treated stage 5 CKD in
a particular group of patients may now be referred to sec-
some populations, specifically in those countries where
ondary care when they were previously managed else-
RRT is freely available.
where or vice versa.
Although RRT prevalence is a widely cited figure for
health service planning and resource calculations, this
Measurement Error figure reflects not only CKD burden but is also a func-
With all approaches based on the testing of renal bio- tion of the survival of those undergoing dialysis or
markers, it is also important to consider the issues sur- receiving kidney transplantation (a key metric, but one
rounding laboratory calibration and external quality which varies substantially between different regions of
assurance which can be the source of substantial meas- the world). Where available, RRT incidence is likely a
urement error. Compounding the problem with bio- more useful estimate of the burden of ESRD in a popula-
marker measurement error is the fact that the various tion. However, individuals receiving of conservative
estimating equations for the eGFR have generally been care for kidney failure, an increasing number of the
validated in US and European populations and whether elderly in high income countries, and of course those
the same relationship between serum creatinine (or cys- who have no prospect of receiving any kind of RRT, i.e.
tatin C) and kidney function exists in all populations and the majority of the ESRD population in lower-income
in other settings remains unclear. For example, although countries, will inevitably be excluded from such
there is a correction factor for African-Americans in both estimates.
the Modification of Diet in Renal Disease (MDRD) and Figure 2.1 summarizes reported RRT incidence from
CKD-Epi eGFR formulae, evidence suggests these correc- a number of international registries. The highest num-
tion factors cannot be applied with certainty to Sub- bers of patients starting RRT per million population are
Saharan African populations [29]. Finally, although again in the Americas (Mexico and the United States) and
less important if examining CKD within a region, when East Asia (Taiwan, Thailand, and Japan). Importantly
attempting to compare the burden of CKD between dif- although numbers appear to be stabilizing in a number
ferent areas it should be recognized that the distribution of high-income settings there continue to be dramatic
of healthy kidney function in many populations remains year-on-year increases in numbers starting RRT in
unknown. Therefore, whether the threshold GFR of middle-income settings such as Mexico and Thailand.
60 ml/min/1.73 m2 as an indicator of future adverse renal These numbers may represent improved ascertainment,
(and cardiovascular) outcomes is appropriate in non- true increases in incidence, or both, but of note,
European/North American settings is not established, i.e. Thailand recently introduced an insurance system by
is there a role for different GFR thresholds for CKD in which free peritoneal dialysis is made available to those
different populations? who require it.
Outputs from Estimates of CKD Burden from Around the Worl 19
Annual RRT incidence pmp

13.8 246.1 478.4
Figure 2.1 Global incidence of renal replacement therapy. Most recent data from publicly available renal registries [30–41], various
years. Crude rates per million population. See text for limitations.
Testing of High-Risk Populations the primary care population, for example when used as
part of audit and/or quality improvement programs. There
Evidence from high-income countries suggests that most
is evidence that these programs can in turn lead to improved
people with CKD can be identified by testing those with
care processes [44]. Although electronic diagnostic coding
risk factors, specifically diabetes, hypertension, those with
of CKD is generally poor in health systems where it is not
a family history and older age groups. For example, in the
incentivized (see reference [45] vs. reference [46]), esti-
Norwegian population almost 60% of those with an
mates of the burden of CKD can also be obtained by inter-
eGFR < 60 ml/min/1.73 m2 can be captured by testing
rogating laboratory records. Coverage is often high,
those with diabetes, hypertension, or a family history of
particularly among high-risk groups, because kidney func-
these conditions [17] and this proportion increases to
tion testing is required for appropriate treatment in those
around 90% if everyone in older age groups is also tested.
at risk (e.g. hypertension drug management, choice of dia-
A number of initiatives aimed at quantifying the burden
betes drug) and care (many older people receive multiple
of CKD have used a similar risk stratification approach to
renal function tests as part of other investigations), hence a
invite individuals for testing (Table 2.3).
high proportion of patients with reduced eGFR are
Although these approaches are efficient at identification
detected. Furthermore, this type of data is useful for dem-
of those with low eGFR and established risk factors such as
onstrating the chronicity of impaired eGFR or proteinuria.
diabetes, the utility of this approach has not been validated
Some examples are shown in Table 2.4.
in all populations. For example, those with nonproteinuric
CKD living in agricultural communities in low- and
middle-income countries would typically not be identified Population Representative Data
using this type of screening.
Population-based surveys provide the gold-standard for
assessing the prevalence of CKD. Although often challeng-
Health Systems Data
ing to conduct, studies of representative samples with high
Health systems that utilize electronic health records have response rates will provide the best estimates of the burden
access to a useful source of data on the burden of CKD in of disease.
Table 2.3 Examples of risk-stratification for CKD testing.
Study Countries Risk group included eGFR < 60 Albuminuria Notes Reference
KEEP USA 18 years + 17% 29% Over 90% of those with [25]
First-degree relative with (urinalysis) eGFR < 60 were unaware that they
diabetes, hypertension, or might have abnormal kidney
kidney disease function
Personal history of
hypertension or diabetes
KEEP Mexico 18 years + Mexico Mexico City 19% [42]
Mexico First-degree relative with City 7% (ACR > 30 mg/g)
diabetes, hypertension, or Jalisco Jalisco State 31%
kidney disease State 10% (urinalysis)
Personal history of
hypertension or diabetes
Hiperdia Brazil Diabetes and 24.6% 36.6% [43]
Programme hypertension
ACR, albumin creatinine ratio; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate.
Table 2.4 Studies using electronic health records to estimate As discussed above, there are a number of other chal-
CKD burden.
lenges that arise when comparing these data, including
laboratory biomarker quality assurance, validation of
Country Coverage Reference
GFR estimating equations in different groups as well as a
poor understanding of the population distribution of
Australia Melbourne [47]
healthy GFR across different regions and ethnicities.
Canada Alberta [48]
Despite these shortcomings this type of cross-sectional
Singapore Singhealth system [49]
survey probably provides the greatest insight into the bur-
UK National (~9% of English and [46] den of CKD, particularly in lower-income settings where
Welsh population)
access to health services, routine laboratory testing, and
Grampian, Scotland [50] especially RRT is incomplete. There have recently been
United States Ohio (Cleveland clinic) [45] international initiatives aimed at conducting simplified
National (Veterans Affairs [51] population-based studies using standardized protocols
Administration) with adequately refenced laboratory procedures [54]. It is
CKD, chronic kidney disease. only using this type of population-representative epide-
miological approach that the burden of impaired kidney
Population-based surveys have been conducted around function (as well as information on specific causes of
the world in attempts to establish the national or regional CKD) can be robustly captured and compared between
prevalence of eGFR < 60 ml/min/1.73 m2, with or without regions and internationally. Furthermore, these types of
also testing for urinary albumin. Often cross-sectional in study are likely to be key to understanding changes in the
nature and nested within epidemiological studies of other burden of kidney disease over time as well as providing
noncommunicable diseases, a key issue when interpreting the potential to gain insight into previously unappreci-
these studies is whether chronicity (i.e. kidney abnormali- ated risk-factors for disease.
ties present for more than 3 months) has been established. Examples of population-based studies examining the
This is important as evidence suggests that between a quar- prevalence of low eGFR (as a surrogate for CKD) are illus-
ter and one-third of those with a single measure of trated in Figure 2.2. The highest population prevalence of
eGFR < 60 ml/min/1.73 m2 (albeit those with values only an eGFR < 60 ml/min/1.73 m2 has been reported in stud-
just below the threshold) will have a value above this ies conducted in Cameroon, Iran, Japan, and the coun-
threshold on retesting [52, 53]. Similarly, variation in the tries of Central America. Whether these high rates reflect
underlying age structure of the population between studies the age structure of the population, the burden of known
will mean data are not easily comparable across regions CKD risk factors such as diabetes or HIV infection, or
without standardization to a reference population. reflect specific regional endemic nephropathies is often
Summary and Conclusion 21
Prevalence eGFR < 60 mL/min/1.7m2 (%)

1.6 8.25 14.9
Figure 2.2 Crude prevalence of eGFR < 60 ml/min/1.7 m2 reported from a range of studies around the world since the year 2000.
Includes representative population-based studies aimed at capturing prevalence of CKD across the age range (includes studies that
included minimum age as high as 40 years old). eGFR calculated from the CKD-Epi (or MDRD equation where CKD-Epi not provided,
and with local correction factors in some cases). Studies only included where external creatinine validation was reported (e.g. IDMS
referenced standards) and prevalence standardized to the local age structure where reported. Estimates based on a single measure of
eGFR only (see text). Source: [26, 52, 55–85].
not fully characterized. However, it is also important to approaches may be useful. The Global Burden of Disease
recognize that even country-level figures can often project combines data from a huge range of data sources.
obscure important regional/local differences in CKD There are a number of challenges when undertaking this
prevalence. For example, in Nicaragua age-standardized kind of quantification but nonetheless the results are
rates of eGFR < 60 ml/min/1.73 m2 reach 18% amongst informative. Estimates from this work report a global preva-
rural males but are less than half that in urban centers in lence of CKD of 3.7% (ranging from 3.4% in low-income
the same region [55]. regions to 5.3% in high-income regions) [86]. More concern-
As might be expected given the issues discussed in this ing perhaps is the increase in burden of kidney disease over
chapter prevalence of impaired kidney function at a popula- the last 25 years. Much of this increase originates from aging
tion level will not necessarily be reflected in high rates of but increasing rates of diabetes and other causes of CKD are
RRT. However, substantively different RRT rates are reported also a substantial contributor in a number of low- and
in settings with similar CKD prevalence even where there is middle-income settings. Globally, CKD now ranks 22nd
unrestricted access to RRT [56], suggesting there are impor- (from 30th in 1990) of all diseases and accounts for 1.5%
tant modifiable factors in the natural history of disease. (from 0.9% in 1990) of all disability adjusted life years lost.
Combined Estimates of CKD Burden: The

Global Burden of Disease Project
Summary and Conclusions
As none of the above approaches to determining CKD bur- No single approach to CKD surveillance will capture all
den can be easily applied to all regions of the world, sum- aspects of the burden of CKD comprehensively and data
mary approaches using a composite of all the above reported from most approaches need to be considered
alongside any limitations and the health systems context. ways or interventions by characterizing changes over
However, understanding the burden of CKD has an time. Furthermore, inter-regional comparisons may pro-
important role in health service planning as well as in vide insight into important health inequalities along with
quantifying benefits from the introduction of care path- the risk-factors and causes of some kidney diseases.
References
1 Levey, A.S., Atkins, R., Coresh, J. et al. (2007). Chronic associated with adverse kidney outcomes. A collaborative
kidney disease as a global public health problem: meta-analysis of general and high-risk population
approaches and initiatives – a position statement from cohorts. Kidney Int. 80 (1): 93–104.
kidney disease improving global outcomes. Kidney Int. 72 11 Tangri, N., Stevens, L.A., Griffith, J. et al. (2011). A
(3): 247–259. predictive model for progression of chronic kidney
2 Saran, R., Hedgeman, E., Plantinga, L. et al. (2010). disease to kidney failure. JAMA 305 (15): 1553–1559.
Establishing a national chronic kidney disease 12 Tangri, N., Grams, M.E., Levey, A.S. et al. (2016).
surveillance system for the United States. Clin. J. Am. Soc. Multinational assessment of accuracy of equations for
Nephrol. 5 (1): 152–161. predicting risk of kidney failure: a meta-analysis. JAMA
3 Iwagami, M., Caplin, B., Smeeth, L. et al. (2018). Chronic 315 (2): 164–174.
kidney disease and cause-specific hospitalisation: a 13 Black, C., Sharma, P., Scotland, G. et al. (2010). Early
matched cohort study using primary and secondary care referral strategies for management of people with
patient data. Br. J. Gen. Pract. 68 (673): e512–e523. markers of renal disease: a systematic review of the
4 Boulware, L., Jaar, B.G., Tarver-Carr, M.E. et al. (2003). evidence of clinical effectiveness, cost-effectiveness and
Screening for proteinuria in us adults: a cost-effectiveness economic analysis. Health Technol. Assess. 14 (21): 1–184.
analysis. JAMA 290 (23): 3101–3114. 14 Asselbergs, F.W., Hillege, H.L., and van Gilst, W.H.
5 Atthobari, J., Asselbergs, F.W., Boersma, C. et al. (2006). (2004). Framingham score and microalbuminuria:
Cost-effectiveness of screening for albuminuria with combined future targets for primary prevention? Kidney
subsequent fosinopril treatment to prevent cardiovascular Int. Suppl. 92: S111–S1114.
events: a pharmacoeconomic analysis linked to the 15 Ozyilmaz, A., de Jong, P.E., Bakker, S.J.L. et al. (2017).
prevention of renal and vascular endstage disease Screening for elevated albuminuria and subsequently
(PREVEND) study and the prevention of renal and hypertension identifies subjects in which treatment may
vascular endstage disease intervention trial (PREVEND be warranted to prevent renal function decline. Nephrol.
IT). Clin. Ther. 28 (3): 432–444. Dial Transplant. 32 (suppl_2): ii200–ii208.
6 Saunders, M.R., Cifu, A., and Vela, M. (2015). JAMA 16 Yarnoff, B.O., Hoerger, T.J., Simpson, S.K. et al. (2017).
guideline synopsis: chronic kidney disease screening. The cost-effectiveness of using chronic kidney disease
JAMA 314 (6): 615–616. risk scores to screen for early-stage chronic kidney
7 Mahmoodi, B.K., Matsushita, K., Woodward, M. et al. disease. BMC Nephrol. 18 (1): 85.
(2012). Associations of kidney disease measures with 17 Hallan, S.I., Dahl, K., Oien, C.M. et al. (2006). Screening
mortality and end-stage renal disease in individuals with strategies for chronic kidney disease in the general
and without hypertension: a meta-analysis. Lancet 380 population: follow-up of cross sectional health survey.
(9854): 1649–1661. BMJ 333 (7577): 1047.
8 Kwon, Y., Han, K., Kim, Y.H. et al. (2018). Dipstick 18 American Diabetes A (2010). Standards of medical Care
proteinuria predicts all-cause mortality in general in Diabetes – 2010. Diabetes Care 33 (Suppl 1): S11–S61.
population: a study of 17 million Korean adults. PLoS 19 Newman, D.J., Mattock, M.B., Dawnay, A.B. et al. (2005).
One 13 (6): e0199913. Systematic review on urine albumin testing for early
9 Chronic Kidney Disease Prognosis Consortium (2010). detection of diabetic complications. Health Technol.
Association of estimated glomerular filtration rate and Assess. 9 (30) iii-vi, xiii-163.
albuminuria with all-cause and cardiovascular mortality 20 Cifu, A.S. and Davis, A.M. (2017). Prevention, detection,
in general population cohorts: a collaborative meta- evaluation, and management of high blood pressure in
analysis. Lancet 375 (9731): 2073–2081. adults. JAMA 318 (21): 2132–2134.
10 Gansevoort, R.T., Matsushita, K., van der Velde, M. et al. 21 Howard, K., White, S., Salkeld, G. et al. (2010). Cost-
(2011). Lower estimated GFR and higher albuminuria are effectiveness of screening and optimal management for
Reference 23
diabetes, hypertension, and chronic kidney disease: a data focusing on elderly patients. Kidney Res. Clin. Pract.
modeled analysis. Value Health 13 (2): 196–208. 35 (4): 204–211.
22 Liu, F.X., Rutherford, P., Smoyer-Tomic, K. et al. (2015). A 35 Masakane, I., Taniguchi, M., Nakai, S. et al. (2018).
global overview of renal registries: a systematic review. Annual dialysis data report 2015, JSDT renal data
BMC Nephrol. 16: 31. registry. Ren. Replace.Ther. 4 (1): 19.
23 Liyanage, T., Ninomiya, T., Jha, V. et al. (2015). 36 National Registry of Diseases Office (2018). Singapore
Worldwide access to treatment for end-stage kidney renal registry annual report 2016. www.nrdo.gov.sg/
disease: a systematic review. Lancet 385 (9981): docs/librariesprovider3/default-document-library/
1975–1982. singapore-renal-registry-annual-report-2016_1999-
24 Jain, A.K., Blake, P., Cordy, P., and Garg, A.X. (2012). till-2016_v5_online_final.pdf?sfvrsn=0 (accessed 21 May
Global trends in rates of peritoneal dialysis. J. Am. Soc. 2021).
Nephrol. 23 (3): 533–544. 37 Wong, H.S. and Goh, B.L. (2018). 24th report of the
25 McCullough, P.A., Li, S., Jurkovitz, C.T. et al. (2008). Malaysian dialysis and transplant registry 2016. Kuala
CKD and cardiovascular disease in screened high-risk Lumpa. www.msn.org.my/msn/Doc/PublicDoc_PB/
volunteer and general populations: the Kidney Early Publication/mdtr2016/All%20Chapters.pdf (accessed
Evaluation Program (KEEP) and National Health and 21 May 2021).
Nutrition Examination Survey (NHANES) 1999–2004. 38 ERA-EDTA Registry (2018). ERA-EDTA registry annual
Am. J. Kidney Dis. 51 (4 Suppl 2): S38–S45. report 2016. Amsterdam. https://www.era-edta-reg.org/
26 Aitken, G.R., Roderick, P.J., Fraser, S. et al. (2014). files/annualreports/pdf/AnnRep2016.pdf.
Change in prevalence of chronic kidney disease in 39 Saran R, Li Y, Robinson B, et al. United States Renal Data
England over time: comparison of nationally System. 2015 USRDS Annual Data Report: Epidemiology
representative cross-sectional surveys from 2003 to 2010. of kidney disease in the United States. Am J Kidney Dis
BMJ Open 4 (9): e005480. 2016;67(3)(suppl 1):S1–S434.
27 Jha, V., Garcia-Garcia, G., Iseki, K. et al. (2013). Chronic 40 Canadian Institute for Health Information (2017).
kidney disease: global dimension and perspectives. Canadian organ replacement register, 2007 to 2016.
Lancet 382 (9888): 260–272. Ottawa, ON.
28 Chesnaye, N.C., Schaefer, F., Groothoff, J.W. et al. (2015). 41 ANZDATA Registry Australia and New Zealand Dialysis
Disparities in treatment rates of paediatric end-stage and Transplant Registry. 39th Annual Report. Adelaide.
renal disease across Europe: insights from the ESPN/ 2016. https://www.anzdata.org.au/report/
ERA-EDTA registry. Nephrol. Dial Transplant. 30 (8): anzdata-39th-annual-report-2016/
1377–1385. 42 Obrador, G.T., Garcia-Garcia, G., Villa, A.R. et al. (2010).
29 Stevens, L.A., Claybon, M.A., Schmid, C.H. et al. (2011). Prevalence of chronic kidney disease in the Kidney Early
Evaluation of the Chronic Kidney Disease Epidemiology Evaluation Program (KEEP) Mexico and comparison
Collaboration equation for estimating the glomerular with KEEP US. Kidney Int. Suppl. (116): S2–S8.
filtration rate in multiple ethnicities. Kidney Int. 79 (5): 43 Alves, L.F., Abreu, T.T., Neves, N.C.S. et al. (2017).
555–562. Prevalence of chronic kidney disease in a city of
30 Lin, Y.-C., Hsu, C.Y., Kao, C.C. et al. (2014). Incidence Southeast Brazil. J. Bras. Nefrol. 39 (2): 126–134.
and prevalence of ESRD in Taiwan Renal Registry Data 44 de Lusignana, S., Gallagher, H., Jones, S. et al. (2013).
System (TWRDS): 2005–2012. Acta Nephrologica 28 (2): Audit-based education lowers systolic blood pressure in
65–68. chronic kidney disease: the Quality Improvement in CKD
31 Gonzalez-Bedat, M., Rosa-Diez, G., Pecoits-Filho, R. et al. (QICKD) trial results. Kidney Int. 84 (3): 609–620.
(2015). Burden of disease: prevalence and incidence of 45 Navaneethan, S.D., Jolly, S.E., Schold, J.D. et al. (2011).
ESRD in Latin America. Clin. Nephrol. 83 (7 Suppl 1): 3–6. Development and validation of an electronic health
32 Leung, C.B., Cheung, W.L., and Li, P.K. (2015). Renal record-based chronic kidney disease registry. Clin. J. Am.
registry in Hong Kong-the first 20 years. Kidney Int. Soc. Nephrol. 6 (1): 40–49.
Suppl. (2011) 5 (1): 33–38. 46 Nitsch, D., Caplin, B., Hull, S., and Wheeler, D.C. (2017).
33 The Nephrology Society of Thailand (2015). Thailand On behalf of the National CKD audit and quality
renal replacement therapy year 2015. https://www. improvement programme in primary care. First national
nephrothai.org/wp-content/uploads/2020/08/Final_ CKD audit report. www.lshtm.ac.uk/files/ckd_audit_
TRT_report_2015_ฉบบแกไข.pdf. report.pdf (accessed 21 May 2021).
34 Jin, D.C., Yun, S.R., Lee, S.W. et al. (2016). Current 47 Pefanis, A., Botlero, R., Langham, R.G., and Nelson, C.L.
characteristics of dialysis therapy in Korea: 2015 registry (2018). eMAP: CKD: electronic diagnosis and
management assistance to primary care in chronic kidney 60 White, S.L., Polkinghorne, K.R., Atkins, R.C., and
disease. Nephrol. Dial Transplant. 33 (1): 121–128. Chadban, S.J. (2010). Comparison of the prevalence and
48 Ronksley, P.E., Tonelli, M., Quan, H. et al. (2012). mortality risk of CKD in Australia using the CKD
Validating a case definition for chronic kidney disease Epidemiology Collaboration (CKD-EPI) and Modification
using administrative data. Nephrol. Dial Transplant. 27 of Diet in Renal Disease (MDRD) Study GFR estimating
(5): 1826–1831. equations: the AusDiab (Australian Diabetes, Obesity and
49 Lew, Q.L.J., Allen, J.C., Nguyen, F. et al. (2018). Factors Lifestyle) Study. Am. J. Kidney Dis. 55 (4): 660–670.
associated with chronic kidney disease and their clinical 61 Anand, S., Shivashankar, R., Ali, M.K. et al. (2015).
utility in primary care clinics in a multi-ethnic Southeast Prevalence of chronic kidney disease in two major Indian
Asian population. Nephron 138 (3): 202–213. cities and projections for associated cardiovascular
50 Marks, A., Black, C., Fluck, N. et al. (2012). Translating disease. Kidney Int. 88 (1): 178–185.
chronic kidney disease epidemiology into patient care – 62 Hosseinpanah, F., Kasraei, F., Nassiri, A.A., and Azizi, F.
the individual/public health risk paradox. Nephrol. Dial (2009). High prevalence of chronic kidney disease in
Transplant. 27 (Suppl 3): iii65–iii72. Iran: a large population-based study. BMC Public Health
51 Wang, V., Maciejewski, M.L., Hammill, B.G. et al. (2014). 9: 44.
Recognition of CKD after the introduction of automated 63 Barreto, S.M., Ladeira, R.M., Duncan, B.B. et al. (2016).
reporting of estimated GFR in the Veterans Health Chronic kidney disease among adult participants of the
Administration. Clin. J. Am. Soc. Nephrol. 9 (1): 29–36. ELSA-Brasil cohort: association with race and
52 Benghanem Gharbi, M., Elseviers, M., Zamd, M. et al. socioeconomic position. J. Epidemiol. Community Health
(2016). Chronic kidney disease, hypertension, diabetes, 70 (4): 380–389.
and obesity in the adult population of Morocco: how to 64 Peraza, S., Wesseling, C., Aragon, A. et al. (2012).
avoid “over”-and “under”-diagnosis of CKD. Kidney Int. Decreased kidney function among agricultural workers in
89 (6): 1363–1371. El Salvador. Am. J. Kidney Dis. 59 (4): 531–540.
53 Hirst, J.A., Montes, M.D.V., Taylor, C.J. et al. (2018). 65 Eastwood, J.B., Kerry, S.M., Plange-Rhule, J. et al.
Impact of a single eGFR and eGFR-estimating equation (2010). Assessment of GFR by four methods in adults in
on chronic kidney disease reclassification: a cohort study Ashanti, Ghana: the need for an eGFR equation for lean
in primary care. Br. J. Gen. Pract. 68 (673): e524–e530. African populations. Nephrol. Dial Transplant. 25 (7):
54 Caplin, B., Jakobsson, K., Glaser, J. et al. (2017). 2178–2187.
International collaboration for the epidemiology of eGFR 66 Seck, S.M., Doupa, D., Gueye, L., and Ba, I. (2014).
in low and middle income populations – rationale and Chronic kidney disease epidemiology in northern
core protocol for the Disadvantaged Populations eGFR Senegal: a cross-sectional study. Iran. J. Kidney Dis. 8 (4):
Epidemiology Study (DEGREE). BMC Nephrol. 18 (1): 1. 286–291.
55 Lebov, J.F., Valladares, E., Pena, R. et al. (2015). A 67 Stanifer, J.W., Maro, V., Egger, J. et al. (2015). The
population-based study of prevalence and risk factors of epidemiology of chronic kidney disease in Northern
chronic kidney disease in Leon. Nicaragua. Can. J. Kidney Tanzania: a population-based survey. PLoS One 10 (4):
Health Dis. 2: 6. e0124506.
56 Hallan, S.I., Coresh, J., Astor, B.C. et al. (2006). 68 Kaze, F.F., Meto, D.T., Halle, M.P. et al. (2015). Prevalence
International comparison of the relationship of chronic and determinants of chronic kidney disease in rural and
kidney disease prevalence and ESRD risk. J. Am. Soc. urban Cameroonians: a cross-sectional study. BMC
Nephrol. 17 (8): 2275–2284. Nephrol. 16: 117.
57 Francis, E.R., Bernabe-Otiz, A., Gilman, M.D. et al. 69 Hooi, L.S., Ong, L.M., Ahmad, G. et al. (2013). A
(2015). Burden of chronic kidney disease in Peru: a population-based study measuring the prevalence of
population-based study. Lancet Glob. Health 3 chronic kidney disease among adults in West Malaysia.
(Supplement 1): S34. Kidney Int. 84 (5): 1034–1040.
58 Vinhas, J., Gardete-Correia, L., Boavida, J.M. et al. (2011). 70 Sabanayagam, C., Lim, S.C., Wong, T.Y. et al. (2010).
Prevalence of chronic kidney disease and associated risk Ethnic disparities in prevalence and impact of risk factors
factors, and risk of end-stage renal disease: data from the of chronic kidney disease. Nephrol. Dial Transplant. 25
PREVADIAB study. Nephron Clin. Pract. 119: c35–c40. (8): 2564–2570.
59 Jessani, S., Bux, R., and Jafar, T.H. (2014). Prevalence, 71 Ji, E. and Kim, Y.S. (2016). Prevalence of chronic kidney
determinants, and management of chronic kidney disease defined by using CKD-EPI equation and albumin-
disease in Karachi, Pakistan – a community based to-creatinine ratio in the Korean adult population. Korean
cross-sectional study. BMC Nephrol. 15: 90. J. Intern. Med. 31 (6): 1120–1130.
Reference 25
2 Zhang, L., Wang, F., Wang, L. et al. (2012). Prevalence of

7 and investigation of the related factors with chronic
chronic kidney disease in China: a cross-sectional survey. kidney disease. Ren. Fail. 31 (10): 920–927.
Lancet 379 (9818): 815–822. 81 Juutilainen, A., Kastarinen, H., Antikainen, R. et al.
73 Imai, E., Horio, M., Watanabe, T. et al. (2009). Prevalence (2012). Trends in estimated kidney function: the
of chronic kidney disease in the Japanese general FINRISK surveys. Eur. J. Epidemiol. 27 (4): 305–313.
population. Clin. Exp. Nephrol. 13 (6): 621–630. 82 Robles, N.R., Felix, F.J., Fernandez-Berges, D. et al.
74 Hsu, C.C., Hwang, S.J., Wen, C.P. et al. (2006). High (2013). Cross-sectional survey of the prevalence of
prevalence and low awareness of CKD in Taiwan: a study reduced estimated glomerular filtration rate, albuminuria
on the relationship between serum creatinine and and cardiovascular risk in a native Spanish population.
awareness from a nationally representative survey. Am. J. J. Nephrol. 26 (4): 675–682.
Kidney Dis. 48 (5): 727–738. 8 3 Ponte, B., Pruijm, M., Marques-V idal, P. et al. (2013).
75 Crews, D.C., Campbell, K.N., Liu, Y. et al. (2017). Chronic Determinants and burden of chronic kidney disease
kidney disease and risk factor prevalence in Saint Kitts in the population-b ased CoLaus study: a cross-
and Nevis: a cross-sectional study. BMC Nephrol. 18 (1): 7. sectional analysis. Nephrol. Dial Transplant. 28 (9):
76 Coresh, J., Selvin, E., Stevens, L.A. et al. (2007). 2329–2339.
Prevalence of chronic kidney disease in the United States. 84 Chudek, J., Wieczorowska-Tobis, K., Zejda, J. et al. (2014).
JAMA 298 (17): 2038–2047. The prevalence of chronic kidney disease and its relation
77 Arora, P., Vasa, P., Brenner, D. et al. (2013). Prevalence to socioeconomic conditions in an elderly polish
estimates of chronic kidney disease in Canada: results of a population: results from the national population-based
nationally representative survey. CMAJ 185 (9): E417–E423. study PolSenior. Nephrol. Dial Transplant. 29 (5):
78 Aumann, N., Baumeister, S.E., Rettig, R. et al. (2015). 1073–1082.
Regional variation of chronic kidney disease in Germany: 85 Ingsathit, A., Thakkinstian, A., Chaiprasert, A. et al.
results from two population-based surveys. Kidney Blood (2010). Prevalence and risk factors of chronic kidney
Press. Res. 40 (3): 231–243. disease in the Thai adult population: Thai SEEK study.
79 Gambaro, G., Yabarek, T., Graziani, M.S. et al. (2010). Nephrol. Dial Transplant. 25 (5): 1567–1575.
Prevalence of CKD in northeastern Italy: results of the 86 Xie, Y., Bowe, B., Mokdad, A.H. et al. (2018). Analysis of the
INCIPE study and comparison with NHANES. Clin. J. global burden of disease study highlights the global,
Am. Soc. Nephrol. 5 (11): 1946–1953. regional, and national trends of chronic kidney disease
80 Sahin, I., Yildirim, B., Cetin, I. et al. (2009). Prevalence of epidemiology from 1990 to 2016. Kidney Int. 94 (3):
chronic kidney disease in the Black Sea region, Turkey, 567–581.
26
Progression of Chronic Kidney Disease: An Evidence-based Approach to Risk Stratification

Meghan J. Elliott1,2, Meha Bhatt1, Bryan Ma1, and Matthew T. James1,2
1
Division of Nephrology, Department of Medicine, Cumming School of Medicine, University of Calgary, Alberta, Canada
2
Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Alberta, Canada
I ntroduction failure, which can be used to provide patients and their

care providers with individualized estimates of their
Chronic kidney disease (CKD) is associated with signifi- future risk of kidney failure.
cant risks of several adverse outcomes, including progres- This chapter provides an evidence-based approach to
sion to kidney failure, cardiovascular events, and risk stratification for the progression of CKD. A summary
mortality [1–4]. Assessment of CKD progression is a chief of evidence is provided for (i) individual prognostic factors
focus of CKD care and a central target of interventions to of CKD progression to kidney failure, (ii) validated multi-
prevent the development of kidney failure, an important variable clinical risk prediction models for progression to
patient-centered outcome. Kidney failure may be variably kidney failure, and (iii) the clinical impact of these existing
defined as initiation of kidney replacement therapy (KRT, risk prediction models. Where possible, evidence for each
including dialysis or kidney transplantation), death attrib- prognostic factor is provided from pooled meta-analyses of
uted to kidney disease, or sustained levels of estimated glo- multiple, high-quality cohort studies, many reported by the
merular filtration rate (eGFR) below a threshold of 15 ml/ CKD-PC. For multivariable risk prediction models, those
min/1.73 m2, which can influence estimates of disease inci- that have been derived, internally validated, and externally
dence and prevalence [5]. Regardless of definition, there validated in one or more cohorts have been included.
remains significant variability in the risk of progression to Because clinical impact analysis is part of the full evidence
kidney failure among patients with CKD. Reliable base for risk prediction models prior to widespread accept-
approaches to risk stratification are important in order to ance and adoption, studies on clinical application and
identify high-risk patients who may require specialized impact are also discussed.
care, need closer follow-up or derive greater benefit from The certainty of evidence was assessed using the Grading
certain interventions, and to provide timely psychological of Recommendations Assessment, Development and
or educational preparation and planning for initiation of Evaluation (GRADE) framework for prognosis studies [7].
KRT or a conservative approach to care. Because no GRADE criteria are currently available for pre-
Several prognostic factors for CKD progression have diction model studies, their quality has been described
been evaluated in large cohort studies that have charac- according to the extent of external validation and the con-
terized associations between clinical and laboratory sistency and generalizability of model performance.
measurements and risk of kidney failure. Much of this
work has been synthesized by the Chronic Kidney Disease
Prognosis Consortium (CKD-PC), in a series of patient- Epidemiology of CKD and Kidney Failure
level meta-analyses of prognostic factors, usually in
relation to eGFR and albuminuria [6]. More recently, pre- CKD is common, affecting up to 15% of the population
diction model studies have combined prognostic factors worldwide [8]. In the United States, the reported prevalence
to derive and validate multivariable models for kidney of CKD ranges from 7% to 15%, depending whether it is
Pathophysiology of CKD Progressio 27
defined using eGFR criteria alone or also by the presence of Kidney failure (ESRD)
albuminuria [9, 10]. The majority of patients with CKD have
ACR ACR ACR ACR
earlier stages of disease, with a much lower prevalence of <10 10–29 30–299 ≥300
kidney failure reported in the adult population. Among
older individuals, in whom multimorbidity can be common, eGFR
> 105 Ref Ref 7.8 18
the reported prevalence of CKD is as high as 40% worldwide
(approximately 25% in the United States), the majority of eGFR
Ref Ref 11 20
90–105
whom would be classified as having stage G3a CKD by eGFR
criteria (i.e. 45–59 ml/min/1.73 m2) [11]. The aging popula- eGFR
75–90 Ref Ref 3.8 48
tion and increased prevalence of cardiovascular disease, dia-
betes, and other noncommunicable diseases are anticipated eGFR
Ref Ref 7.4 67
to further increase the global burden of CKD and its associ- 60–75
ated morbidity and mortality [8]. Primary and secondary eGFR
prevention approaches are desirable so that modifiable risk 45–60 5.2 22 40 147
factors can be addressed to delay development or progres-
eGFR
sion to kidney failure, and to prevent associated complica- 30–45 56 74 294 763
tions such as cardiovascular events.
eGFR
There is substantial variability across the stages of CKD 433 1044 1056 2286
15–30
in risk of progression to kidney failure, and a significant
amount of research has aimed to characterize the factors Figure 3.1 Kidney Disease Improving Global Outcomes
that contribute to this risk. The current international (KDIGO) CKD Clinical Practice Guideline heat map for risk of
CKD staging system [5] categorizes patients according to kidney failure. eGFR is reported in units of ml/min/1.73 m2 and
both eGFR and albuminuria thresholds based on risk ACR in mg/g (multiply by 0.113 to convert to mg/mmol). Values
in the cells represent adjusted relative risks. ACR, albumin-to-
relationships, although some have suggested that age- creatinine ratio; ESRD, end-stage renal disease; eGFR, estimated
and gender-specific eGFR criteria may be more appropri- glomerular filtration rate. Source: Levey et al. [13]. © 2011
ate, particularly when compared to eGFR alone [12]. Elsevier.
Albuminuria was added to the previously accepted eGFR-
based classification system following studies demonstrat- cause [15]. These maladaptive changes are believed to con-
ing its association with increased risk of CKD progression tribute to a final common pathway of scarring, further
and mortality independent of eGFR [13]. Kidney Disease nephron loss, and progressive kidney damage that can ulti-
Improving Global Outcomes (KDIGO) has developed a mately lead to kidney failure [16]. Hemodynamic factors,
visual “heat map” to identify patients at progressively activation of the renin–angiotensin–aldosterone system,
increased risk for kidney failure at different levels of and inflammatory responses leading to tubulointerstitial
eGFR and albuminuria (Figure 3.1) [5]. Declining eGFR fibrosis are major pathophysiological mechanisms that
in combination with increasing albuminuria may better have been identified in this process.
predict progression to kidney failure than either one Both systemic and glomerular hypertension have been
alone [14]. However, while CKD staging can identify identified as common mediators of CKD progression. The
patients at increased risk of progressing to kidney failure, afferent and efferent arterioles can moderate glomerular
risk for progression on an individual level can vary greatly perfusion and pressure independently from systemic
even within these CKD stages. This suggests that other blood pressure. However, in the remnant kidney model in
factors beyond eGFR and albuminuria contribute to the rats, reduced nephron mass results in progressive hyper-
risk of kidney failure, and identification of variables that perfusion, glomerular hypertrophy, and glomerulosclero-
provide incremental prognostic value could lead to sis [17]. Micropuncture studies suggest that single nephron
improved risk stratification approaches. hyperfiltration in response to renal ablation leads to glo-
merulosclerosis [18], and animal and human studies sug-
gest that interventions such as renin-angiotensin system
Pathophysiology of CKD Progression inhibitors that decrease glomerular pressure can amelio-
rate sclerosis, reduce albuminuria, and slow progression
Although CKD may result from a wide variety of causes, of CKD [19].
once it is established, common pathological findings of Tubulointerstitial fibrosis is a second common pathophys-
vascular injury, glomerulosclerosis, and tubulointerstitial iological process associated with CKD progression [20, 21].
fibrosis have been described, regardless of the inciting While it has been traditionally viewed as a consequence of
28 Progression of Chronic Kidney Disease: An Evidence-based Approach to Risk Stratification
nephron ischemia in CKD [15], more recent mechanistic The CKD-EPI equation has been shown to more accu-
studies suggest that inflammatory stimuli in response to rately estimate GFR than the MDRD study equation, espe-
injury may play an important effect in fibrotic processes in cially at higher GFR. A meta-analysis of data from
the kidney, moderated via macrophage signaling, cytokines, 1.1 million adults from 21 general population, high-risk,
eicosanoids, and anti-angiogenic factors [22]. These factors and CKD cohorts demonstrated that although the general
may lead to decreased peritubular capillary density and pattern of the eGFR and kidney failure risk relationship
phenotypic transition of interstitial and parenchymal epi- was similar for both equations, eGFR computed by the
thelial cells into myofibroblasts, which in turn leads to CKD-EPI equation predicted risk of kidney failure more
interstitial fibrosis and glomerulosclerosis, and progressive accurately than the MDRD study equation when eGFR was
loss of kidney function [23]. used for classification into six categories (90, 60–89, 45–59,
30–44, 15–29, and 15 ml/min/1.73 m2) [24] by both equa-
tions. In the general population cohorts, 25.0% of patients
rognostic Factors for Progression
P were reclassified by the CKD-EPI equation, mostly to a
to Kidney Failure higher eGFR category, with overall reclassification leading
to lower risks of kidney failure in patients reclassified into
Several commonly ascertained measures have been identi- higher eGFR categories and higher risks of kidney failure
fied as risk factors or risk modifiers of progression of CKD in patients reclassified into lower eGFR categories. The
to kidney failure. The evidence for these laboratory (eGFR exception was the MDRD eGFR category of 15–19 ml/
and albuminuria), demographic (age, sex, and race), and min/1.73 m2, where no significant difference in risk was
clinical variables (acute kidney injury, blood pressure, dia- observed between patients reclassified into the higher
betes mellitus, and cardiovascular disease) is reviewed here eGFR category using CKD-EPI and those not reclassified.
and summarized in Table 3.1.
Change in eGFR
eGFR
A progressive decline in GFR can be viewed as an intermedi-
There is high certainty evidence that reduced eGFR is associ- ary on the pathway to kidney failure, and existing evidence
ated with kidney failure in people with an established diag- suggests with high certainty that a patient’s past trajectory of
nosis of CKD, as well as in the general population and those decline in eGFR provides additional prognostic information
with cardiovascular disease and a high risk of CKD. This about the risk of kidney failure, over and above the most
finding has been demonstrated when serum creatinine val- recent eGFR available. In a meta-analysis of 232 250 patients
ues, standardized to isotope dilution mass spectrometry- from 11 CKD cohorts, where the risk of kidney failure over
traceable methods, are used to estimate glomerular filtration 2 years of follow-up was 1.4%, a decline in eGFR of 6 ml/
rate (GFR) using either the Modification of Diet in Renal min/1.73 m2 per year over the previous 3 years (i.e. a decline
Disease (MDRD) study equation or the CKD Epidemiology of 18 ml/min/1.73 m2 versus no decline) was associated with
Collaboration (CKD-EPI) equation. The first phase of col- approximately double the adjusted relative risk of kidney
laborative meta-analysis published by the CKD-PC [4] exam- failure compared to no decline, whereas a current eGFR of
ined the risk of kidney failure in a total of 21 688 patients 30 versus 50 ml/min/1.73 m2 (i.e. a difference of 20 ml/
from 14 cohorts with CKD, in which the incidence rate of min/1.73 m2) was associated with 20-fold higher relative risk
kidney failure varied markedly from 13.6 to 115.3 per 1000 of kidney failure [33]. These findings thus demonstrate that
person-years. In adjusted analyses, each decrease in MDRD despite the prognostic value of a change in eGFR, current
eGFR by 15 ml/min/1.73 m2 below a threshold of 45 ml/ eGFR is associated more strongly with future risk of kidney
min/1.73 m2 was associated with a sixfold increase in hazard failure than the magnitude of past eGFR decline, especially
of kidney failure. In meta-analysis of cohorts drawn from the when eGFR falls below 30 ml/min/1.73 m2.
general population, with a higher mean eGFR (range
72–144 ml/min/1.73 m2) and the absolute risk for end-stage
Albuminuria
kidney disease (ESKD) of 0.31 events per 1000 person-years,
the risk of kidney failure was not related to MDRD eGFR at Proteinuria can be measured in several ways, including
values between 75 and 105 ml/min/1.73 m2, but was expo- urinary dipstick, albumin-to-creatinine ratio (ACR) or
nentially higher at lower levels of eGFR. At MDRD eGFRs protein-to-creatinine ratio (PCR) on a spot urine sample, or
averaging 60, 45, and 15 ml/min/1.73 m2, and adjusted haz- quantification from a 24-hour urine collection. Regardless
ard ratios for kidney failure were 4, 29, and 454, respectively, of the method of assessment, there is high certainty
compared with an eGFR of 95 ml/min/1.73 m2 [2]. evidence that increasing proteinuria confers a higher risk
Prognostic Factors for Progression to Kidney Failur 29
Table 3.1 Summary of findings: individual prognostic factors for kidney failure.
Definition of Certainty of
Prognostic factors prognostic factor Evidence source Summary of findingsa evidence (GRADE)b
eGFR [4] MDRD eGFR CKD prognosis Absolute risk: High

categorized as 30–44, consortium Incidence rate of kidney failure varied between ●●●●
15–29, and <15 ml/ meta-analysis cohorts from 13.6 to 115.3 per 1000 person-years.
min/1.73 m2, including 14 CKD Relative risk:
compared to a cohorts, with
reference group with 21 688 patients Each decrease in MDRD eGFR by 15 ml/min below
eGFR = 45–74 ml/ a threshold of 45 ml/min/1.73 m2 was associated
min/1.73 m2 with a HR 6.24 (95% CI 4.84–8.05), (I2 = 87.9%,
P < 0.001)
eGFR 30–44: HR 2.72 (95% CI 1.29–3.37)
eGFR 15–29: HR 10.21 (95% CI 8.36–12.46)
eGFR <15: HR 51.48 (95% 31.95–82.97)
CKD-EPI versus Estimated GFR CKD prognosis Absolute risk:
MDRD study classified into six consortium eGFR computed by the CKD-EPI equation predicted
eGFR [24] categories (90, 60–89, meta-analysis risk of ESKD more accurately than the MDRD Study
45–59, 30–44, 15–29, including equation; 35% of patients with eGFR 45–59 ml/
and 15 ml/ 1.1 million adults min/1.73 m2 by the MDRD study equation
min/1.73 m2) by from 21 general reclassified to eGFR of 60–89 ml/min/1.73 m2 by the
MDRD study and population, CKD-EPI equation and had a lower incidence rate of
CKD-EPI equations. high-risk, and kidney failure compared with those not reclassified
CKD cohorts (0.5 vs. 0.8 per 1000-person years).
Net reclassification improvement (NRI) significantly
favored the CKD-EPI equation for predicting kidney
failure; NRI 0.06 (95% CI, 0.02–0.10).
Change in Three-year eGFR 11 CKD cohorts Absolute risk: High
eGFR [25] slope evaluation (232 250 patients) Absolute risk of kidney failure over a mean ●●●●
follow-up of 2 years by eGFR slope; 4.7% for
eGFR slope <−5 ml/min/1.73 m2/year, 1% for
eGFR slope −5 to 5 ml/min/1.73 m2/year, 0.1% for
eGFR slope 5 ml/min/1.73 m2/year.

Relative risk:
Adjusted hazard ratios of kidney failure, compared
to eGFR slope of 0 ml/min/1.73 m2/year; eGFR slope
of −3 ml/min/1.73 m2/year over preceding 3 years
had a HR 1.73 (95% 1.50–2.00), I2 = 0, eGFR slope of
−6 ml/min/1.73 m2/year over 3 years had a HR 2.28
(95% CI 1.88–2.76), I2 = 34.3%. No evidence of effect
modification of the relative risk of kidney failure
associated with eGFR slope according to level of last
eGFR (P value for interaction 0.26).
Proteinuria [4] Proteinuria was Collaborative Absolute risk: High
categorically defined meta-analysis of Unadjusted absolute risk kidney failure by ACR ●●●●
as followsc: 13 CKD cohort category (range of incidence rates per 1000
studies, including person-years of follow-up)d:
ACR (mg/g; multiple 21 688 patients ACR <30 mg/g: 2.8–13.8
by 1.113 to convert to with follow-up of
1.9–7.5 years ACR 30–299 mg/g: 2.8–90.1
mg/mmol): <30, ACR 300–999 mg/g: 16.4–141.5
30–299, 300–999, and
1000 ACR 1000 mg/g: 50.2–247.4

Dipstick proteinuria: Unadjusted absolute risk of kidney failure by
−/±, +, ++, +++ dipstick category (incidence rate per 1000 person-
years of follow-up)d:
(Continued)
Table 3.1 (Continued)
−/±: 5.1
+: 1.2
++: 44.6
+++: 140.8

Relative risk:Adjusted relative risk of kidney
failure by ACR category (reference <30 mg/g):
ACR 30–299 mg/g: HR 2.87 (95% CI 1.91–4.34)
ACR 300–999 mg/g: HR 7.96 (95% CI 6.27–10.09)
ACR 1000 mg/g: HR 14.61 (95% CI 11.16–19.13)

Pooled relative risk for kidney in continuous
analyses with an eightfold increase in ACR: HR 2.92
(95% CI 1.96–4.35); I2 92.2%, P < 0.001

Adjusted relative risk of kidney failure by dipstick
category (reference −/±)e:
+: HR 2.92 (2.08–4.10)
++: HR 7.70 (4.52–13.10)
+++: HR 15.01 (8.36–26.95)
Acute kidney Variable definitions of Systematic review Absolute risk: Moderate

injury [26] AKI have been used and meta-analysis Absolute pooled rate of kidney failure of 8.6 (95% CI ●●●○
across studies of 13 cohort 0.63–28.1) events per 100 person-years of follow-up.
including studies, including
requirement for acute 1 472 743 patients
dialysis, diagnosis with follow-up of Relative risk:
codes, and creatinine- 6–75 months Adjusted relative risk of kidney failure associated
based staging with AKI: HR 3.1 (95% CI 1.9–5.0), I2 = 98%.
Relative risk of kidney failure increased in a graded
manner with increasing severity of AKI.
Age [27] Age was categorically Individual-level Absolute risk: Moderate

defined as follows (in meta-analysis of Absolute risk of kidney failure by age for eGFR of ●●●○
years): 18–54, 55–64, 46 cohorts, 50 ml/min/1.73 m2 in CKD cohorts (incidence rates
65–74, 75 including per 1000 person-years of follow-up) f:
2 051 244 patients 18–54 years: 38.2
with a mean
follow-up of 55–64 years: 20.7
5.8 years 65–74 years: 15.1
Separate analyses 75 years: 5.7
were conducted
on 13 CKD Relative risk:
cohorts (38 612 Adjusted relative risk of kidney failure associated
patients) with age at an eGFR of 30 ml/min/1.73 m2 in CKD
cohorts (reference eGFR 50 ml/min/1.73 m2 at age
55–64 years)d:
18–54 years: HR 2.85
55–64 years: HR 3.37
65–74 years: HR 3.22
75 years: HR 4.64
Sex [28] Male or female Meta-analysis of Absolute risk: Moderate

13 CKD cohorts Overall absolute risk of progression to kidney failure ●●●○
with 38 612 was 15.4% among the CKD cohorts.
patients, 47% of
whom were Relative risk:
women Men had higher risks of kidney failure than women,
but there was no difference by sex in the relative risk
of kidney failure associated with greater
albuminuria or reduced eGFR level.
Race [29] Race was categorized Meta-analysis of Absolute risk: Moderate
as Asian, White, or 45 cohorts, Unadjusted absolute risk of kidney failure by race ●●●○
Black including (incidence rates per 1 000 person-years of follow-up)
1 102 581 patients. in all cohorts f:
Among the 13 Asian 0.3
CKD cohorts
(36 295 patients), White 0.8
the majority were Black 2.8
White (83%), and Relative risk:
mean follow-up
Adjusted relative risk of kidney failure at eGFR
was 6.6–9.2 years
30 ml/min/1.73 m2 by race (reference eGFR 50 ml/
min/1.73 m2) in CKD cohortsd:
Asian 3.09 (95% CI 2.06–4.59)
White 2.98 (95% CI 2.48–3.77)
Black 2.57 (95% CI 1.3–5.08)
Hypertension [30] Systolic BP 13 cohorts of Absolute risk: Moderate
>140 mmHg or patients with Kidney failure occurred in 12.3% without ●●●○
diastolic CKD hypertension and 19.5% with hypertension.
BP > 90 mmHg (21 072 without
hypertension and Relative risk:
17 088 with Adjusted hazard ratio for kidney failure estimated at
hypertension) eGFR 30 ml/min/1.73 m2, by hypertensive status
(reference eGFR 50 ml/min/1.73 m2 in individuals
without hypertension)d:
No hypertension HR 3.20 (95% CI 2.14–4.78)
Hypertension HR 3.68 (95% CI 2.87–4.67)
Diabetes [31] Fasting glucose of 13 CKD cohorts Absolute risk: Low
7.0 mmol/l, or including 38, 612 Unadjusted absolute risk of kidney failure in all ●●○○
non-fasting glucose of patients cohorts f:
at least 11.1 mmol/l, Total of 5 960 kidney failure events during a mean
or glycated follow up time of 3.8 years. Absolute risk not
hemoglobin 6.5%, or stratified by those with versus without diabetes.
use of glucose-
lowering drugs, or Relative risk:
self-reported diabetes Relative HR = 0.79 (0.56–1.13) averaged across
range of eGFR, P value 0.19
Relative HR = 1.08 (0.95–1.23) averaged across
range of ACR, P value 0.22
Adjusted hazard ratio for kidney failure estimated at
eGFR 30 ml/min/1.73 m2, by diabetes status
(reference eGFR 50 ml/min/1.73 m2 in individuals
without diabetes)d:
No diabetes HR 3.26 (2.20–4.74)
Diabetes HR 4.93 (3.12–7.70)
(Continued)
Cardiovascular Previous diagnosis of 28 cohorts Absolute risk: Low

disease [32] myocardial infarction, including 185 024 Unadjusted absolute risk of kidney failure in all ●●○○
percutaneous participants from cohorts f:
coronary 30 different 22 301 (12.1%) kidney failure overall among the
intervention, bypass countries, greater 185 024 patients. Absolute risk not stratified by those
grafting, heart failure, than 18 years of with versus without history of CVD.
or stroke age, with eGFR
30 ml/ Relative risk:
min/1.73 m2 HR for history of CVD = 0.91 (0.82–1.02) in the 19
cohorts with outcome information on kidney failure,
CVD events, and death.
HR for time-varying CVD = 2.28 (2.02–2.57) in the
19 cohorts with outcome information on KRT, CVD
events, and death.
ACR, albumin-to-creatinine ratio; AKI, acute kidney injury; BP, blood pressure; CI, confidence interval; CKD, chronic kidney disease; CKD-EPI,
CKD Epidemiology Collaboration; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; HR, hazard ratio; KRT, kidney
replacement therapy; MDRD, Modification of Diet in Renal Disease.
a
The majority of the CKD-PC studies defined kidney failure as initiation of kidney replacement therapy (dialysis or kidney transplantation) or
death due to kidney disease other than acute kidney injury [4, 24, 25, 27–31], whereas two studies defined kidney failure as initiation of kidney
replacement therapy alone [26, 32].
b
GRADE assessment of the certainty of evidence: High: We are very confident that the true prognosis (probability of future events) lies close to
that of the estimate. Moderate: We are moderately confident that the true prognosis (probability of future events) is likely close to that of the
estimate, but there is a possibility that it is substantially different. Low: Our confidence in the estimate is limited: the true prognosis
(probability of future events) may be substantially different from the estimate. Very low: We have very little confidence in the estimate: the true
prognosis (probability of future events) is likely to be substantially different from the estimate.
c
PCR was not reported in this table as the results were similar to ACR.
d
Values that were only provided in figures and were approximated using graph digitizing software.
e
Results are from a single study, Kaiser Permanente Northwest.
f
CKD-PC includes three cohort types based on source population, general population cohorts, high-risk cohorts selecting subjects at high
cardiovascular risk (such as hypertension or diabetes), and CKD cohorts specifically enrolling subjects with CKD. Where not otherwise
specified results refer to overall estimates within all three of these cohorts.
of developing kidney failure independently of eGFR and increased risk of kidney failure and mortality in CKD
other clinical variables. In a collaborative meta-analysis that cohorts [4]. In one cohort study that stratified by eGFR,
included 12 CKD cohorts and 21 321 patients reporting on patients with heavier albuminuria within each eGFR cate-
ESKD outcomes, more severe proteinuria when measured gory had significantly increased adjusted rates of both kid-
by dipstick, ACR, or PCR was independently associated with ney failure and mortality compared to those with lesser or
an increased risk of kidney failure [4]. Among included no albuminuria [34]. In fact, those with heavy albuminuria
studies, the unadjusted incidence rate for kidney failure was but without overtly abnormal eGFR appeared to have
as high as 200 events per 1000 person-years in those with worse clinical outcomes than those with moderately
greatest degree of albuminuria, compared to rates of less reduced eGFR and no albuminuria. Increasing severity of
than 10 per 1000 person-years for those with no albuminu- albuminuria may predict a more rapid rate of decline in
ria [4]. Similarly, the pooled relative risk of developing kid- kidney function, regardless of baseline eGFR [35].
ney failure increased significantly with higher categories of These findings highlight the importance of considering
albuminuria, even after adjusting for age, comorbidities, the influence of albuminuria in addition to other clinical
and eGFR. Although there was significant heterogeneity factors, such as eGFR, when assessing risk of CKD progres-
between studies, the point estimates of association did not sion, and have contributed to interest in using albuminuria
vary significantly between studies with the exception of one as a surrogate for progression to kidney failure [36, 37]. The
outlier that demonstrated a more strongly positive associa- most recent international clinical practice guidelines for
tion between increasing ACR and kidney failure [33]. CKD management acknowledge albuminuria as a valid
In the meta-analysis by Astor et al., both lower eGFR and and important independent predictor of kidney, cardiovas-
higher albuminuria were independently associated with an cular, and mortality outcomes, resulting in a modification
of the previous eGFR-based CKD staging system to a risk- kidney function [42, 43]. A multivariable model developed
based one that considers both severity of eGFR decline and with 9973 participants with AKI and externally validated
albuminuria [5]. These guidelines refer specifically to use of in a cohort with 2761 participants from Canada included
albuminuria (i.e. ACR) rather than the proteinuria, as albu- age, sex, AKI stage, prehospitalization serum creatinine
min is the primary component of urinary protein in most level, albuminuria, and discharge serum creatinine in the
conditions, and the epidemiologic data supporting the asso- final risk index [44]. This model achieved a c-statistic of
ciation between proteinuria and outcomes is strongest for 0.81 and was well calibrated for predicting advanced CKD
urinary albumin quantification. after hospital discharge. However, the clinical impact of
using this model to guide care remains to be evaluated.
Whether interventions to prevent AKI or enhance recovery
Acute Kidney Injury
from AKI reduce the risk of progressive CKD and ESKD
Acute kidney injury (AKI) is a potentially avoidable compli- remains uncertain.
cation that may occur in hospital or the community. There is
moderate certainty evidence that patients who have survived
Age
AKI have an increased adjusted risk of subsequent progres-
sion of CKD and kidney failure compared to those without Age does not appear to uniformly influence the risk of
an episode of AKI [34]. In a large cohort study of nearly one developing kidney failure independent of the effect of
million individuals from Canada, an episode of AKI was baseline kidney function and albuminuria. In an individual-
associated with higher risk of long-term sustained doubling level meta-analysis of 46 cohorts and approximately two
of serum creatinine or ESKD, with the largest increases in million patients, the adjusted incidence rate of kidney fail-
risk seen in individuals with higher eGFR and lesser degrees ure at a given level of eGFR was lowest for persons over age
of proteinuria at baseline [38]. This illustrated that an epi- 75 years compared with those in younger age categories
sode of AKI provides extra prognostic information for pro- (e.g. 9.8 vs. 28.2 per 1000 person-years for ages >75 years
gression of CKD in addition to that provided by eGFR and and 55–64 years, respectively, at an eGFR of 45 ml/
proteinuria. Pathophysiologic processes that are believed to min/1.73 m2) [27]. However, once eGFR declined to 15 ml/
contribute to the transition from acute injury to chronic min/1.73 m2, the adjusted risk of kidney failure was similar
damage include a failure of differentiation and persistently across age categories. Among the 13 CKD cohorts in this
high signaling activity of regenerating cells in the tubuloint- meta-analysis, similar findings of lower adjusted incidence
erstitium, with inflammation, capillary rarefaction, and rates of kidney failure among the oldest age group were
fibroblast proliferation ultimately contributing to progres- reported, although the absolute risk of kidney failure
sion of CKD [39]. appeared to increase steeply for individuals’ 75 years of
In a systematic review and meta-analysis of 13 cohort age with an eGFR below 25 ml/min/1.73 m2 to reach risk
studies, including over one million patients with follow-up estimates comparable to other age groups. Overall, a trend
ranging from 6 to 75 months, the pooled incidence of kid- toward lower risk of kidney failure was identified among
ney failure was 8.6 events per 100 person-years, and was older individuals, although the absolute and relative risks
threefold higher for those with AKI than without AKI. The were not significantly different across age categories.
magnitude of the association between AKI and kidney fail- Kidney function generally declines with advancing age,
ure varied substantially between the studies included in this the significance of which (and the extent to which it is con-
systematic review; however, the definitions for AKI were sidered pathological) has been debated [45]. In the meta-
heterogeneous between studies and all studies reported analysis described above, low eGFR was independently
higher risks among patients with AKI. Furthermore, the associated with mortality regardless of age, although the
risk of kidney failure increased in a graded manner with relative mortality risk was attenuated with increasing
greater severity of AKI. More recently AKI has also been age [27]. Findings from previous studies suggest that
associated with a later risk of developing CKD even in elderly individuals are less likely to progress to kidney fail-
patients with recovery of kidney function to normal lev- ure than they are to die from other causes [46], although
els [40], and a longer duration of AKI has also been associ- these risks may vary depending on how kidney failure is
ated with a higher risk of developing CKD [41]. defined, an individuals’ baseline kidney function, and
Among patients who have had an episode of AKI, whether the kidney failure is treated with KRT. In the sys-
cohort studies have identified several factors that are asso- tematic review by Hallan et al., kidney failure was defined
ciated with high risks of developing advanced CKD (eGFR as initiation of dialysis, transplantation, or death as a result
<30 ml/min/1.73 m2), including older age, lower baseline of kidney disease. However, this definition may fail to iden-
eGFR, more severe AKI, and incomplete recovery of tify those whose who do not start dialysis for kidney failure
and those who die from a nonkidney-related cause. One kidney failure among the cohorts and racial groups. It is
large cohort study suggested that whereas the absolute and worth noting that in this study, most Black patients were
relative risks of “treated” kidney failure (i.e. receipt of dial- from US cohorts, most Asian patients were from East Asian
ysis or kidney transplantation) are higher in younger com- cohorts, and few cohorts included multiple racial groups.
pared to older people at a given eGFR, the rates of The absolute risk of kidney failure appeared to be high-
“untreated” kidney failure (progression to eGFR <15 ml/ est among Black patients in the general population cohorts
min/1.73 m2 without renal replacement therapy) were sig- of this meta-analysis (adjusted incidence rate 2.36 vs.
nificantly higher in older individuals [47]. This suggests 0.46 in White patients and 0.27 in Asians, per 1000 person-
that kidney disease does progress in older adults and that years) [29]. Unadjusted incidence rates in the CKD cohorts
the incidence of kidney failure in this population may be were similar. Genetic factors, including apolipoprotein L1
higher than what has been assumed when kidney failure is (APOL1) gene polymorphisms, contribute in large part to
defined only by receipt of KRT. These findings have impli- the significantly increased rates of kidney failure among
cations for clinical practice and decision making for adults Black patients; APOL1 mutations, found exclusively in
of all ages with CKD, including perception and communi- individuals of African descent, have been associated with
cation of risk, and evaluation of treatment options. earlier onset of CKD, increased albuminuria, and a more
rapid decline in eGFR [50–52]. Whereas findings from
other studies support differences in rates of kidney out-
Sex
comes across racial groups and nationalities [53–56], race
Although females have a higher prevalence of CKD than may also predict rate of eGFR decline and development of
males, observational studies from the United Kingdom and CKD in persons without known kidney disease. In one US
North America suggest that females have a lower risk of population-based study, Black Americans and Hispanics
kidney failure than males [48]. Moderate certainty evi- with normal kidney function experienced more rapid rates
dence from a meta-analysis of 13 CKD cohorts from of eGFR decline than White Americans and had higher
Europe, North and South America, Asia, and Australasia, rates of incident CKD [57]. Consistent findings of a rela-
including 38 612 patients, have confirmed this finding, and tionship between eGFR and renal outcomes regardless of
also found no differences by sex in the relative risk of kid- race support the use of internationally adopted practice
ney failure associated with a greater degree of albuminuria guidelines that stage CKD according to eGFR and albumi-
or reduced eGFR [28, 49]. It is not yet clear the extent to nuria without defining specific thresholds for racial
which biological differences related to sex versus societal groups [5]. However, to address modifiable risk factors and
and social norms related to gender influence the higher impact care delivery across populations with variable risk,
incidence of KRT in men than in women. it is important to consider the reasons for such baseline dif-
ferences, including contributions of socio-cultural, envi-
ronmental, and genetic influences [58].
Race
Moderate certainty evidence suggests that a patient’s race
Hypertension
does not independently influence his or her relative risk of
developing kidney failure for a given eGFR, despite varia- As a cause and consequence of CKD [59], hypertension is a
tions in baseline risk across racial groups. Pooled data from target of treatment in CKD, and there is moderate quality
a meta-analysis of 45 cohorts indicate that the relative risk of evidence that patients with CKD and hypertension have a
kidney failure and mortality increases similarly among the higher incidence of progression to kidney failure and
three studied races (i.e. Asian, Black, and White) across a death. A large cohort study of 300 000 adults within the
range of eGFR and/or albuminuria, with greatest risk at low general population found that there were graded increases
eGFR and high albuminuria [29]. The majority of cohorts in in the risk of kidney failure with increasing blood pressure
this meta-analysis were from the general population, and this relationship was independent of existing kidney
although a similar relationship was observed among the 13 disease [60]. These results are further supported by a large
CKD cohorts (n = 36 295, 3% of the study population). For US cohort study of 332 544 men that found increasing risk
example, compared to a reference eGFR of 50 ml/ of developing kidney failure with increasing severity of
min/1.73 m2, Asian, White, and Black patients at an eGFR of hypertension, with up to a 20 times greater risk among
30 ml/min/1.73 m2 had an approximately 2.5- to 3.5-times men with stage 4 hypertension in comparison to those with
increased risk of developing kidney failure after adjusting optimal blood pressure [60, 61]. Similarly, an increased risk
for measured confounders. This is despite differences in of kidney failure has also been associated with hyperten-
clinical and demographic characteristics and absolute risk of sion in women [62].
Among patients with CKD, the relationship between tory of cardiovascular disease. Findings indicated that dia-
hypertension and kidney failure is not uniform and varies betic patients had a higher relative risk of kidney failure
based on eGFR and ACR levels. A meta-analysis of 13 CKD when the references were set to an eGFR of 50 ml/
cohorts, comprising 21 072 patients with hypertension and min/1.73 m2 and an ACR of 20 mg/g in nondiabetic patients.
17 088 without hypertension, found an incidence of 19.5% However, this relationship did not extend across low eGFR
(3327 kidney failure events) among those with hypertension and high ACR levels, with no significant difference in the
and 12.3% (2597 kidney failure events) in the nonhyperten- relative risk of kidney failure by diabetes status across the
sive patients [30]. Hypertensive patients, characterized by full range of eGFR values (hazard ratio [HR] 0.79, 95% con-
systolic blood pressure 140 mmHg and diastolic blood pres- fidence interval [CI] 0.56–1.13, P value 0.19) and ACR val-
sure 90 mmHg, had a higher risk of kidney failure at an ues (HR 1.08, 95% CI 0.95–1.23, P value 0.22). There is low
eGFR of 50 ml/min/1.73 m2 and ACR of 100 mg/g. However, certainty in this evidence due to the variability observed
the effect of hypertensive status was attenuated at lower between studies.
eGFR and higher ACR levels, such that lower eGFR and While the aforementioned reviews did not specifically
increasing ACR levels corresponded with increasing risk of compare treated versus untreated diabetes, there is also
developing kidney failure [30] among patients both with and evidence that patients with poor glycemic control are at
without hypertension. No significant interaction was found greater risk of progression to kidney failure. A Canadian
when antihypertensive medication use was incorporated population-based cohort study found that among patients
into the analysis. with diabetes and an eGFR of 30–59 ml/min/1.73 m2, the
However, there is evidence for a protective effect on pro- risk of progression to kidney failure increased with increas-
gression to kidney failure based on the type of antihyper- ing HbA1c [66]. Such patients with an HbA1c between 7%
tensive treatment regimen. An individual patient-level to 9% and higher than 9% had increased relative risks of
meta-analysis of 11 randomized controlled trials (1946 22% and 152%, respectively, of developing kidney failure
patients) compared antihypertensive treatments with or compared to those with an HbA1c level <7%. This risk of
without angiotensin converting enzyme (ACE) inhibitors kidney failure with increasing HbA1C diminished at eGFR
and showed that patients with CKD who had proteinuria levels of 15–29 ml/min/1.73 m2.
and were treated with ACE inhibitors were at significantly Overall, individuals with diabetes and CKD have a higher
decreased risk of developing kidney failure (7.4%) in com- risk of kidney failure in comparison to those without dia-
parison to patients who received other antihypertensive betes at normal eGFR and ACR levels, and this risk is
regimens (11.6%) [63]. heightened among patients with poor glycemic control.
Taken together, this evidence suggests that hypertension However, diabetic status provides less additional prognos-
increases the risk of kidney failure in the general popula- tic value for kidney failure at lower eGFR and higher ACR
tion and CKD patients, but the relative prognostic impor- levels.
tance of hypertension for progression to kidney failure
tends to diminish at later stages of CKD.
Cardiovascular Disease
Many studies have shown that CKD is associated with the
Diabetes Mellitus
development of cardiovascular disease, and traditional car-
Diabetes is a common cause of CKD, and individuals with diovascular risk factors are also highly prevalent in patients
diabetes are at greater risk of kidney failure and mortal- with CKD [3, 67]. Some studies have shown that the risk of
ity [64, 65]. Evidence from a large meta-analysis of over cardiovascular events and mortality is a prominent com-
100 000 patients reported that patients with advanced (i.e. peting event in CKD and may exceed the risk of kidney fail-
stage G4) CKD and diabetes mellitus had a 30% higher ure in many patients. A meta-analysis of 28 cohorts across
event rate of kidney failure compared to those with 30 countries included 185 024 patients with advanced CKD
advanced (i.e. stage G4) CKD without diabetes [32]. (i.e. CKD G4+) and found an overall incidence of 12.1%
A second meta-analysis including 38 612 patients from 13 (22 301 events) for kidney failure requiring KRT [32].
CKD cohorts investigated the association of diabetes Adjusted meta-analyses suggested no increased risk of kid-
(defined as fasting glucose of at least 7.0 mmol/l, nonfasting ney failure requiring KRT among those with a history of
glucose of at least 11.1 mmol/l or HbA1c of at least 6.5%, use cardiovascular disease at cohort entry (defined by previous
of glucose-lowering drugs, or self-reported) with risks of diagnosis of myocardial infarction, percutaneous coronary
developing kidney failure [31]. These analyses were intervention, bypass grafting, heart failure, or stroke).
adjusted for other cardiovascular risk factors, including However, when time-varying cardiovascular events were
blood pressure, total cholesterol, body mass index, and his- added to the model, they were significantly associated with
increased risk of both progression to kidney failure (HR in these geographic regions. These findings provide strong
2.28, 95% CI 2.02–2.57) and mortality (HR 2.87, 95% CI evidence that estimates of the absolute risk of kidney fail-
2.57–3.20). The certainty of this evidence is low due to het- ure can be achieved with a level of accuracy considered
erogeneity between the study cohorts and imprecision of helpful for use in clinical decision-making, using demo-
pooled estimates. graphic and commonly ordered laboratory tests for patients
with a broad range of CKD stages.
Shroeder et al. have also developed and externally vali-
isk Prediction Models for Kidney
R dated a pragmatic prediction model for the 5-year risk of
Failure progression to KRT among patients with stage G3 or G4
CKD [69]. The development and validation cohorts were
The best ways to assess patient prognosis use prediction composed of patients with CKD from two geographically
models that combine multiple risk factors to provide esti- distinct cohorts of patients from a large health mainte-
mates of a patient’s absolute risk of an outcome. Risk pre- nance organization in the United States. The final model
diction models have been developed and tested in other included eight predictors: age, sex, eGFR, hemoglobin,
clinical settings to inform clinical decision making. To be proteinuria/albuminuria, systolic blood pressure, anti-
certain that a risk prediction model provides reliable prog- hypertensive medication use, and diabetes and its com-
nostic information, it should undergo validation, ideally in plications. This multivariable model had excellent
a separate group of patients from those in whom the mod- discrimination, with a c-statistic of 0.95, and maintained
els were derived, to demonstrate the model’s ability to dis- very good calibration in the external validation cohort,
criminate between patients who will have an event and with the exception of slightly lower observed than pre-
those who will not. Furthermore, the model should deliver dicted risk in the highest-risk quintile. These findings
predictions of absolute risk that are well calibrated to provide further evidence that a combination of demo-
observed risks [68]. Three multivariable risk prediction graphic, clinical, and laboratory variables that are rou-
models to predict progression to kidney failure have been tinely collected in primary care can provide accurate
developed and externally validated in cohorts with CKD estimates of the risk of kidney failure for patients across
(Table 3.2) [49, 69, 70]. a broad range of stages of CKD.
In 2011 Tangri et al. reported the derivation and perfor- Multivariable models have also been developed using
mance of kidney failure risk equations (KFREs) to estimate data from over 260 000 individuals with eGFR <30 ml/
2- and 5-year absolute risks of treated kidney failure, min/1.73 m2 from 30 countries included in the
defined by the need for dialysis or a pre-emptive kidney CKD-PC [70]. These models included nine prognostic fac-
transplant [49]. The models were derived in a cohort of tors (age, sex, race, eGFR, ACR, systolic blood pressure,
over 3000 patients with CKD (eGFR <60 ml/min/1.73 m2) smoking status, diabetes mellitus, and history of cardiovas-
cared for by nephrologists in Toronto, Canada, and exter- cular disease) and were developed to predict 2- and 4-year
nally validated in a separate cohort of almost 5000 patients risks of competing outcomes of CKD, including kidney
receiving similar care in nephrology clinics in British failure requiring KRT, a nonfatal cardiovascular event, and
Columbia, Canada. The most accurate model included death. Discrimination of the model for risk of kidney fail-
eight variables (age, sex, eGFR, albuminuria, serum cal- ure at 2 years was very good, and when compared to the
cium, serum phosphate, serum bicarbonate, and serum original four-variable KFRE developed by Tangri et al.
albumin) and achieved a c-statistic of 0.92 in the develop- showed similar discrimination (c-statistic 0.81 vs. 0.82),
ment cohort and 0.84 in the validation cohort. Prediction with good calibration when used for risk stratification into
model discrimination and calibration was also very good risk categories of <20%, 20–40%, and >40% predicted risk
for a simpler, four-variable model including age, sex, eGFR, of kidney failure. These findings provide evidence that
and albuminuria (Table 3.2). The generalizability of pre- multivariable risk models can also aid with risk stratifica-
dictions based on the KFREs have since been demonstrated tion for kidney failure in patients with advanced (G4) stage
through external validation of the original four-and eight- CKD and can be used to provide estimates of this risk in
variable equations in 31 cohorts from North America, relation to competing risks of other relevant clinical out-
Europe, Asia, and Oceania [71], including cohorts with comes of cardiovascular events and mortality before or
patients not under the care of a nephrologist. The four- after initiation of KRT.
variable risk equation maintained very good discrimina- Since publication of the original KFRE, several studies
tion and calibration in North American cohorts, although have examined strategies for model updating, with the
recalibration was required among non-North American addition of new or repeated measures in attempts to
cohorts to account for the lower incidence of kidney failure improve risk prediction for kidney failure. Tangri et al.
Table 3.2 Summary of findings: validated multivariable risk prediction models for kidney failure.
Model (author) Cohort descriptions Prognostic variables Predictive performance Tools for use
Kidney failure risk equations Derivation cohort of 3 449 Eight-variable model; age, sex, Eight-variable model; c-statistic A risk calculator is available
(Tangri et al.) [49] patients with CKD stages 3–5 eGFR, albuminuria, age, sex, 0.92; (95% CI 0.90–0.93) in the online at www.kidneyfailurerisk.
from 2001 to 2008 at Sunnybrook eGFR, albuminuria, serum derivation cohort and 0.84 (95% com, for use with an app for
Hospital, University of Toronto calcium, serum phosphate, CI, 0.82–0.86) in the validation handheld electronic devices at
Health network, Toronto Canada. serum bicarbonate, and serum cohort. https://qxmd.com/calculate/
11% developed kidney failure albumin calculator, and as excel risk
over 5 years Four-variable model; c-statistic calculator provided in the
Four-variable model; age, sex, 0.91 (95% CI 0.89–0.93) in supplementary material of the
Validation cohort of 4 942 eGFR, albuminuria derivation cohort and 0.85, 0.83, original publication
patients with CKD stages 3–5 and 0.83 at 1, 3, and 5 years,
from 2001 to 2008 from the respectively, in the external
British Columbia CKD Registry. validation cohort
24% developed kidney failure
over 5 years
An integer-based risk score is
Kaiser Permanente risk of renal Derivation cohort of 22 460 Age, sex, eGFR, proteinuria or Derivation cohort: c-statistic 0.96 available in the original
replacement therapy (RRT) patients in the Kaiser albuminuria, hemoglobin, (0.95–0.97), R2 = 79.7% (78.6– publication
model (Schroeder et al.) [69] Permanente Northwest health diabetes, systolic blood pressure, 80.8) at 5 years
maintenance organization with anti-hypertensive medication,
stage G3 or G4 CKD from 2002 to and diabetes with number of Validation cohort: c-statistic 0.95
2008. 4.7% developed kidney complications (0.94–0.97), R2 = 81.2%
failure over 5 years (79.6–82.6)

Validation cohort of 16 553
patients in Kaiser Permanent
Colorado from 2006 to 2008. 2.6%
developed kidney failure over
5 years.
A risk calculator is available
CKD Prognosis Consortium 264 296 patients with eGFR Age, sex, race, eGFR, systolic c-statistic 0.814 (within cohort online through KDIGO at https://
models (Grams et al.) [70] <30 ml/min/1.73 m2 from 29 blood pressure, history of range in c-static of 0.680–0.972). kdigo.org/equation
cohorts and 30 countries cardiovascular disease, diabetes,
included in the CKD Prognosis albuminuria, smoking history.
Consortium. 4-year risk of kidney
failure ranged from <0.5% to 28%
across cohorts.
CI, confidence interval; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate.
Note: In all risk prediction models, kidney failure is defined as initiation of dialysis or kidney transplantation.
0005152391.INDD 37 09-12-2022 12:40:33

assessed a dynamic risk prediction strategy using time- (the Diabetic Nephropathy Classification Score of the
varying laboratory data in a cohort study of patients Renal Pathology Society) to the KFRE did not significantly
with CKD [72]. A latest-available-measurement predictive improve risk prediction of kidney failure [74].
model with eGFR as a time-dependent predictor incremen-
tally improved risk prediction for kidney failure compared
to a static measurement model with only a single eGFR linical Application and Impact
C
(integrated discrimination improvement 0.73%, 95% CI of Kidney Failure Risk
0.56–0.90%), providing evidence that incorporating addi- Prediction Models
tional eGFR measurements over time can improve risk pre-
diction. There is no good evidence that addition of other Several potential clinical uses for prognostic information on
biomarkers can significantly improve upon the existing kidney failure risk have been proposed, which may include
KFREs for predicting risk of kidney failure in CKD popula- the use of this information by patients and family members,
tions. In one cohort study of patients with CKD who under- primary care clinicians, specialists, health system payers
went duplex ultrasonography testing, adding renal resistive and policy makers. However, there is little evidence guiding
indices did not improve risk prediction for kidney failure their application and assessing their impact in the clinical
beyond the four- and eight-variable KFRE [73] models. In setting. Three studies have reported on patient or provider
another cohort study of patients with CKD who had under- perspectives and experiences on the potential impact of
gone a kidney biopsy confirming diabetic nephropathy, the implementation of kidney failure risk prediction models in
addition of a score based on pathological biopsy features clinical care (Table 3.3) [75–77].
Table 3.3 Summary of findings: clinical implementation and impact of risk prediction models for kidney failure.
Objective and model

Study Design Population implementation Findings
Chiu Quantitative 111 nephrologists across Canada To determine 80% of respondents were not satisfied with
2015 [75] (surveys) perspectives on the their current ability to predict kidney
importance of predicting failure, cardiovascular outcomes, and death,
kidney failure, risk and most indicated that they would use
thresholds that would validated risk scores to predict clinical
influence CKD treatment outcomes. Risk prediction over a 1-to 5-year
plans, and experiences time frame was felt to be important for
using existing risk clinical decision making, such as guiding
prediction tools. referral for arteriovenous fistula creation.
Hingwala Quantitative Patients referred for nephrology To describe the 34% of referrals were classified as low risk
2017 [76] (quasi- assessment in Manitoba, Canada incorporation of the and referred back to their family physician.
experimental) from 1 January 2013 to 31 KFRE in a triage system The median wait time for nephrology
December 2013 for general nephrology consultation improved from 230 days to
referrals. Patients with a 58 days post triage.
5-year kidney failure risk
of 3% were booked with
nephrology either
urgently (risk 10%) or
nonurgently (risk
3–10%); those with a risk
of <3% were referred
back to their family
physician.
Smekal Qualitative Adult patients with non-dialysis To explore perceived Perceived challenges to implementing a
2018 [77] (focus groups, CKD and their family members benefits and challenges risk-based CKD care approach included lack
interviews, (n = 12 across two focus groups), of implementing a of capacity to manage low-risk patients
open-ended and healthcare providers (nurses, KFRE-based approach to outside of the CKD clinic setting and more
survey nephrologists [n = 16 interviews; determine eligibility for fragmented care. Perceived benefits
responses) n = 40 survey respondents]) from multidisciplinary CKD included targeted care for CKD patients at
a multidisciplinary CKD clinic in care. highest risk of progression and improved
Alberta, Canada program efficiency.
CKD, chronic kidney disease; KFRE, Kidney Failure Risk Equation.

Conclusion 39
Chiu et al. surveyed 111 Canadian nephrologists about the potential efficiency gains and focused allocation of
their views on and experiences with CKD risk prediction resources in providing targeted CKD care for those at high-
tools [75]. Approximately 80% were dissatisfied with their est risk of kidney failure.
current ability to predict risk of kidney failure, cardiovascu- These studies have primarily examined perceptions and
lar events, and death among CKD patients, and the majority experiences surrounding implementation of the KFRE to
indicated that they would use validated risk scores if such inform risk-informed nephrology care in the Canadian
tools accurately predicted risk and influenced their manage- context. However, the broader clinical impact of using risk
ment to improve clinical outcomes. Furthermore, this sur- prediction models in practice remains uncertain. Although
vey was able to characterize nephrologist consensus on risk the use of kidney failure risk estimates have been proposed
thresholds at which treatment decisions should be made, to inform several clinical decisions in CKD manage-
such as arteriovenous fistula planning when a patient’s ment [78], successful strategies for their implementation in
1-year predicted kidney failure risk exceeded 30–50%. clinical practice, utility in clinical decision making, and
Two studies have evaluated the impact of the KFRE in subsequent impact on patient-, provider-, and system-level
clinical nephrology practice, including in the triage of outcomes [79, 80] has yet to be comprehensively evaluated.
nephrology CKD referrals [76] and determining eligibility Such data will contribute to the generation of an important
for multidisciplinary CKD clinic care [77]. Hingwala et al. evidence base for the effectiveness of kidney failure risk
applied a KFRE-based threshold of 3% kidney failure risk prediction models in the care of people with CKD.
over 5 years to triage referrals to specialist nephrology
assessment. Approximately one-third of referrals were
considered low risk and consequently were referred back C
onclusions
to their family doctor with CKD care recommendations
provided in a standard “low risk letter.” In the post-triage High-quality evidence has established substantial variability
period, the median wait time for nephrology assessment in the risk of progression of CKD, and that demographic,
decreased from 230 to 58 days, suggesting that this approach clinical, and laboratory variables can provide independent
may help improve access to nephrology care for patients at prognostic value for the prediction of kidney failure. eGFR
highest risk of progression to kidney failure. Smekal et al. and albuminuria are well characterized, strong risk factors for
sought the perspectives of patients with CKD, their fami- progression of CKD to kidney failure, with high certainty evi-
lies, and nephrology healthcare providers on the use of the dence supporting the use of these variables in combination to
KFRE to determine patient eligibility for multidisciplinary assess CKD prognosis. Validated multivariable models that
CKD clinic care (2-year kidney failure risk 10%) or gen- incorporate these and other readily available demographic
eral nephrology care (2-year kidney failure risk <10%). In and clinical risk factors are now available for use as kidney
focus groups, interviews, and open-ended survey responses, failure risk stratification tools in clinical practice (Figure 3.2).
participants expressed concern about the lack of capacity Whether implementation of these risk prediction models in
to manage lower risk CKD patients in the community clinical care improves patient outcomes and experiences in
and loss of services for patients previously followed in the CKD care remains uncertain, and further research in the area
multidisciplinary setting. However, they also recognized of clinical application and impact is needed.
Prognostic
Factors Multivariable Prediction Models:
• Age30 Clinical Impact Implementation

• Sex35 Derivation Validation
Analysis
• Race37
• eGFR4,21
• Change in eGFR25 • Tangri’s KFRE36 Triaging general
• Albuminuria4 nephrology referrals67 Provider perceptions of
• Kaiser Permanente Model60
• Acute Kidney Injury26
• CKD-PC Prognostic Model61 risk estimation66
• Hypertension51 Determining eligibility
• Diabetes56 for multidisciplinary
• Cardiovascular disease55 CKD care68
Figure 3.2 Phases in evidence generation to establish risk prediction models for CKD progression in clinical practice. CKD, chronic
kidney disease; CKD-PC, Chronic Kidney Disease Prognosis Consortium; eGFR, estimated glomerular filtration rate; KFRE, kidney failure
risk equation.
List of Abbreviations
ACR albumin-to-creatinine ratio GFR glomerular filtration rate

AKI acute kidney injury GRADE Grading of Recommendations Assessment,
BP blood pressure Development and Evaluation
CI confidence interval HbA1C hemoglobin A1C
CKD chronic kidney disease HR hazard ratio
CKD-EPI CKD Epidemiology Collaboration KFRE kidney failure risk equation(s)
CKD-PC Chronic Kidney Disease Prognosis KRT kidney replacement therapy
Consortium MDRD Modification of Diet in Renal Disease
eGFR estimated glomerular filtration rate PCR protein-to-creatinine ratio
ESKD end-stage kidney disease U.S. United States
References
Matsushita, K., van der Velde, M., Astor, B.C. et al. (2010).
1 9 Coresh, J., Astor, B.C., Greene, T. et al. (2003). Prevalence
Association of estimated glomerular filtration rate and of chronic kidney disease and decreased kidney function
albuminuria with all-cause and cardiovascular mortality in in the adult US population: Third National Health and
general population cohorts: a collaborative meta-analysis. Nutrition Examination Survey. Am. J. Kidney Dis. 41 (1):
Lancet (London, England) 375 (9731): 2073–2081. 1–12.
2 Gansevoort, R.T., Matsushita, K., van der Velde, M. et al. 10 Murphy, D., McCulloch, C.E., Lin, F. et al. (2016). Trends
(2011). Lower estimated GFR and higher albuminuria are in prevalence of chronic kidney disease in the United
associated with adverse kidney outcomes. A collaborative States. Ann. Intern. Med. 165 (7): 473–481.
meta-analysis of general and high-risk population cohorts. 11 Stevens, L.A., Viswanathan, G., and Weiner, D.E. (2010).
Kidney Int. 80 (1): 93–104. Chronic kidney disease and end-stage renal disease in the
3 van der Velde, M., Matsushita, K., Coresh, J. et al. (2011). elderly population: current prevalence, future projections,
Lower estimated glomerular filtration rate and higher and clinical significance. Adv. Chronic Kidney Dis. 17 (4):
albuminuria are associated with all-cause and 293–301.
cardiovascular mortality. A collaborative meta-analysis of 12 Poggio, E.D. and Rule, A.D. (2007). Can we do better than
high-risk population cohorts. Kidney Int. 79 (12): a single estimated GFR threshold when screening for
1341–1352. chronic kidney disease? Kidney Int. 72 (5): 534–536.
4 Astor, B.C., Matsushita, K., Gansevoort, R.T. et al. (2011). 13 Levey, A.S., de Jong, P.E., Coresh, J. et al. (2011). The
Lower estimated glomerular filtration rate and higher definition, classification, and prognosis of chronic kidney
albuminuria are associated with mortality and end-stage disease: a KDIGO controversies conference report. Kidney
renal disease. A collaborative meta-analysis of kidney Int. 80 (1): 17–28.
disease population cohorts. Kidney Int. 79 (12): 1331–1340. 14 Hallan, S.I., Ritz, E., Lydersen, S. et al. (2009). Combining
5 Levin, A., Stevens, P.E., Bilous, R.W. et al. (2013). Kidney GFR and albuminuria to classify CKD improves
disease: improving global outcomes (KDIGO) CKD work prediction of ESRD. J. Am. Soc. Nephrol. 20 (5):
group. KDIGO 2012 clinical practice guideline for the 1069–1077.
evaluation and management of chronic kidney disease. 15 Fogo, A.B. (2007). Mechanisms of progression of chronic
Kidney Int. Suppl. 3 (1): 1–150. kidney disease. Pediatr. Nephrol. 22 (12): 2011–2022.
6 Matsushita, K., Ballew, S.H., Astor, B.C. et al. (2013). 16 Olson, J.L. and Heptinstall, R.H. (1988).
Cohort profile: the chronic kidney disease prognosis Nonimmunologic mechanisms of glomerular injury. Lab.
consortium. Int. J. Epidemiol. 42 (6): 1660–1668. Invest. 59 (5): 564–578.
7 Iorio, A., Spencer, F.A., Falavigna, M. et al. (2015). Use of 17 Shimamura, T. and Morrison, A.B. (1975). A progressive
GRADE for assessment of evidence about prognosis: rating glomerulosclerosis occurring in partial five-sixths
confidence in estimates of event rates in broad categories nephrectomized rats. Am. J. Pathol. 79 (1): 95–106.
of patients. BMJ (Clinical Research Ed.) 350: h870. 18 Hostetter, T.H., Olson, J.L., Rennke, H.G. et al. (1981).
8 Jha, V., Garcia-Garcia, G., Iseki, K. et al. (2013). Chronic Hyperfiltration in remnant nephrons: a potentially
kidney disease: global dimension and perspectives. Lancet adverse response to renal ablation. Am. J. Physiol. 241 (1):
(London, England) 382 (9888): 260–272. F85–F93.
Reference 41
19 Brenner, B.M., Meyer, T.W., and Hostetter, T.H. (1982). 32 Evans, M., Grams, M.E., Sang, Y. et al. (2018). Risk factors
Dietary protein intake and the progressive nature of for prognosis in patients with severely decreased GFR.
kidney disease: the role of hemodynamically mediated Kidney Int. Rep. 3 (3): 625–637.
glomerular injury in the pathogenesis of progressive 33 Brenner, B.M., Cooper, M.E., de Zeeuw, D. et al. (2001).
glomerular sclerosis in aging, renal ablation, and Effects of losartan on renal and cardiovascular outcomes
intrinsic renal disease. N. Engl. J. Med. 307 (11): in patients with type 2 diabetes and nephropathy. N. Engl.
652–659. J. Med. 345 (12): 861–869.
20 Lin, S.-L., Kisseleva, T., Brenner, D.A., and Duffield, J.S. 34 Hemmelgarn, B.R., Manns, B.J., Lloyd, A. et al. (2010).
(2008). Pericytes and perivascular fibroblasts are the Relation between kidney function, proteinuria, and
primary source of collagen-producing cells in obstructive adverse outcomes. JAMA 303 (5): 423–429.
fibrosis of the kidney. Am. J. Pathol. 173 (6): 1617–1627. 35 Turin, T.C., James, M., Ravani, P. et al. (2013). Proteinuria
21 Lan, R., Geng, H., Polichnowski, A.J. et al. (2012). PTEN and rate of change in kidney function in a community-
loss defines a TGF-β-induced tubule phenotype of failed based population. J. Am. Soc. Nephrol. 24 (10): 1661–1667.
differentiation and JNK signaling during renal fibrosis. 36 Jun, M., Turin, T.C., Woodward, M. et al. (2015).
Am. J. Physiol. Renal Physiol. 302 (9): F1210–F1223. Assessing the validity of surrogate outcomes for ESRD: a
22 Kipari, T. and Hughes, J. (2002). Macrophage-mediated meta-analysis. Am. Soc. Nephrol.
renal cell death. Kidney Int. 61 (2): 760–761. 37 Heerspink, H.J.L., Kröpelin, T.F., Hoekman, J., and De
23 Neilson, E.G. (2006). Mechanisms of disease: Zeeuw, D. (2015). Drug-induced reduction in albuminuria
fibroblasts – a new look at an old problem. Nat. Clin. is associated with subsequent renoprotection: a meta-
Pract. Nephrol. 2 (2): 101–108. analysis. Am. Soc. Nephrol. 26(8):2055–64.
24 Matsushita, K., Mahmoodi, B.K., Woodward, M. et al. 38 James, M.T., Hemmelgarn, B.R., Wiebe, N. et al. (2010).
(2012). Comparison of risk prediction using the CKD-EPI Glomerular filtration rate, proteinuria, and the incidence
equation and the MDRD study equation for estimated and consequences of acute kidney injury: a cohort study.
glomerular filtration rate. JAMA 307 (18): 1941–1951. Lancet 376 (9758): 2096–2103.
25 Kovesdy, C.P., Coresh, J., Ballew, S.H. et al. (2016). Past 39 Venkatachalam, M.A., Griffin, K.A., Lan, R. et al. (2010).
decline versus current eGFR and subsequent ESRD risk. Acute kidney injury: a springboard for progression in
J. Am. Soc. Nephrol. 27 (8): 2447–2455. chronic kidney disease. Am. J. Physiol. Renal Physiol. 298
26 Coca, S.G., Singanamala, S., and Parikh, C.R. (2012). (5): F1078–F1094.
Chronic kidney disease after acute kidney injury: a 40 Sawhney, S., Marks, A., Fluck, N. et al. (2017). Post-
systematic review and meta-analysis. Kidney Int. 81 (5): discharge kidney function is associated with subsequent
442–448. ten-year renal progression risk among survivors of acute
27 Hallan, S.I., Matsushita, K., Sang, Y. et al. (2012). Age and kidney injury. Kidney Int. 92 (2): 440–452.
association of kidney measures with mortality and 41 Heung, M., Steffick, D.E., Zivin, K. et al. (2016). Acute
end-stage renal disease. JAMA 308 (22): 2349–2360. kidney injury recovery pattern and subsequent risk of
28 Nitsch, D., Grams, M., Sang, Y. et al. (2013). Associations CKD: an analysis of veterans health administration data.
of estimated glomerular filtration rate and albuminuria Am. J. Kidney Dis. 67 (5): 742–752.
with mortality and renal failure by sex: a meta-analysis. 42 Pannu, N., James, M., Hemmelgarn, B., and Klarenbach,
BMJ (Clinical Research Ed.) 346: f324. S. (2013). Association between AKI, recovery of renal
29 Wen, C.P., Matsushita, K., Coresh, J. et al. (2014). Relative function, and long-term outcomes after hospital
risks of chronic kidney disease for mortality and end- discharge. Clin. J. Am. Soc. Nephrol. 8 (2): 194–202.
stage renal disease across races are similar. Kidney Int. 86 43 Chawla, L.S., Amdur, R.L., Amodeo, S. et al. (2011). The
(4): 819–827. severity of acute kidney injury predicts progression to
30 Mahmoodi, B.K., Matsushita, K., Woodward, M. et al. chronic kidney disease. Kidney Int. 79 (12): 1361–1369.
(2012). Associations of kidney disease measures with 44 James, M.T., Pannu, N., Hemmelgarn, B.R. et al. (2017).
mortality and end-stage renal disease in individuals with Derivation and external validation of prediction models
and without hypertension: a meta-analysis. Lancet 380 for advanced chronic kidney disease following acute
(9854): 1649–1661. kidney injury. JAMA 318 (18): 1787–1797.
31 Fox, C.S., Matsushita, K., Woodward, M. et al. (2012). 45 Stevens, R.J., Evans, J., Oke, J. et al. (2018). Kidney age, not
Associations of kidney disease measures with mortality kidney disease. Can. Med. Assoc. J. 190(13): E389–E393.
and end-stage renal disease in individuals with and 46 O’Hare, A.M., Bertenthal, D., Covinsky, K.E. et al. (2006).
without diabetes: a meta-analysis. Lancet 380 (9854): Mortality risk stratification in chronic kidney disease: one
1662–1673. size for all ages? J. Am. Soc. Nephrol. 17 (3): 846–853.
7 Hemmelgarn, B.R., James, M.T., Manns, B.J. et al. (2012).

4 62 Tozawa, M., Iseki, K., Iseki, C. et al. (2003). Blood
Rates of treated and untreated kidney failure in older vs pressure predicts risk of developing end-stage renal
younger adults. JAMA 307 (23): 2507–2515. disease in men and women. Hypertension 41 (6):
48 Van Dijk, P., Zwinderman, A., Dekker, F. et al. (2007). 1341–1345.
Effect of general population mortality on the north–south 63 Jafar, T.H., Stark, P.C., Schmid, C.H. et al. (2003).
mortality gradient in patients on replacement therapy in Progression of chronic kidney disease: the role of blood
Europe. Kidney Int. 71 (1): 53–59. pressure control, proteinuria, and angiotensin-converting
49 Tangri, N., Stevens, L.A., Griffith, J. et al. (2011). A enzyme inhibition: a patient-level meta-analysis. Ann.
predictive model for progression of chronic kidney Intern. Med. 139 (4): 244–252.
disease to kidney failure. JAMA 305 (15): 1553–1559. 64 Fox, C.S., Larson, M.G., Leip, E.P. et al. (2004). Predictors
50 Friedman, D.J., Kozlitina, J., Genovese, G. et al. (2011). of new-onset kidney disease in a community-based
Population-based risk assessment of APOL1 on renal population. JAMA 291 (7): 844–850.
disease. J. Am. Soc. Nephrol. 2011050519. 65 Collins, A.J., Foley, R.N., Herzog, C. et al. (2011). US
51 Parsa, A., Kao, W.L., Xie, D. et al. (2013). APOL1 risk renal data system 2010 annual data report. Am. J. Kidney
variants, race, and progression of chronic kidney disease. Dis. 57 (1): A8.
N. Engl. J. Med. 369 (23): 2183–2196. 66 Shurraw, S., Hemmelgarn, B., Lin, M. et al. (2011).
52 Kanji, Z., Powe, C.E., Wenger, J.B. et al. (2011). Genetic Association between glycemic control and adverse
variation in APOL1 associates with younger age at outcomes in people with diabetes mellitus and chronic
hemodialysis initiation. J. Am. Soc. Nephrol. 2010121234. kidney disease: a population-based cohort study. Arch.
53 Collins, A.J., Foley, R.N., Herzog, C. et al. (2013). US Intern. Med. 171 (21): 1920–1927.
Renal Data System 2012 annual data report. Am. J. Kidney 67 Go, A.S., Chertow, G.M., Fan, D. et al. (2004). Chronic
Dis. 61 (1): E1–E459. kidney disease and the risks of death, cardiovascular
54 McClellan, W., Warnock, D.G., McClure, L. et al. (2006). events, and hospitalization. N. Engl. J. Med. 351 (13):
Racial differences in the prevalence of chronic kidney 1296–1305.
disease among participants in the Reasons for 68 Alba, A.C., Agoritsas, T., Walsh, M. et al. (2017).
Geographic and Racial Differences in Stroke Discrimination and calibration of clinical prediction
(REGARDS) Cohort Study. J. Am. Soc. Nephrol. 17 (6): models: users’ guides to the medical literature. JAMA 318
1710–1715. (14): 1377–1384.
55 Mathur, R., Dreyer, G., Yaqoob, M.M., and Hull, S.A. 69 Schroeder, E.B., Yang, X., Thorp, M.L. et al. (2017).
(2018). Ethnic differences in the progression of chronic Predicting 5-year risk of RRT in stage 3 or 4 CKD:
kidney disease and risk of death in a UK diabetic development and external validation. Clin. J. Am. Soc.
population: an observational cohort study. BMJ Open 8 Nephrol. 12 (1): 87–94.
(3): e020145. 70 Grams, M.E., Sang, Y., Ballew, S.H. et al. (2018).
56 Iseki, K., Ikemiya, Y., Iseki, C., and Takishita, S. (2003). Predicting timing of clinical outcomes in patients with
Proteinuria and the risk of developing end-stage renal chronic kidney disease and severely decreased glomerular
disease. Kidney Int. 63 (4): 1468–1474. filtration rate. Kidney Int. 93 (6): 1442–1451.
57 Peralta, C.A., Katz, R., DeBoer, I. et al. (2011). Racial and 71 Tangri, N., Grams, M.E., Levey, A.S. et al. (2016).
ethnic differences in kidney function decline among Multinational assessment of accuracy of equations for
persons without chronic kidney disease. J. Am. Soc. predicting risk of kidney failure: a meta-analysis. JAMA
Nephrol. 2010090960. 315 (2): 164–174.
58 Norton, J.M., Moxey-Mims, M.M., Eggers, P.W. et al. 72 Tangri, N., Inker, L.A., Hiebert, B. et al. (2017). A
(2016). Social determinants of racial disparities in CKD. J. dynamic predictive model for progression of CKD. Am. J.
Am. Soc. Nephrol. 27 (9): 2576–2595. Kidney Dis. 69 (4): 514–520.
59 Ritz, E. (2010). Hypertension and kidney disease. Clin. 73 Lennartz, C.S., Pickering, J.W., Seiler-Mussler, S.
Nephrol. 74: S39–S543. et al. (2016). External validation of the kidney failure
60 C-y, H., McCulloch, C.E., Darbinian, J. et al. (2005). risk equation and re-calibration with addition of
Elevated blood pressure and risk of end-stage renal ultrasound parameters. Clin. J. Am. Soc. Nephrol. 11
disease in subjects without baseline kidney disease. Arch. (4): 609–615.
Intern. Med. 165 (8): 923–928. 74 Yamanouchi, M., Hoshino, J., Ubara, Y. et al. (2018).
61 Klag, M.J., Whelton, P.K., Randall, B.L. et al. (1996). Value of adding the renal pathological score to the kidney
Blood pressure and end-stage renal disease in men. N. failure risk equation in advanced diabetic nephropathy.
Engl. J. Med. 334 (1): 13–18. PLoS One 13 (1): e0190930.
Reference 43
5 Chiu, H.H., Tangri, N., Djurdjev, O. et al. (2015).

7 78 Eckardt, K.-U., Bansal, N., Coresh, J. et al. (2018).
Perceptions of prognostic risks in chronic kidney Improving the prognosis of patients with severely decreased
disease: a national survey. Can. J. Kidney Health glomerular filtration rate (CKD G4+): conclusions from a
Dis. 2: 53. Kidney Disease: Improving Global Outcomes (KDIGO)
76 Hingwala, J., Wojciechowski, P., Hiebert, B. et al. (2017). controversies conference. Kidney Int. 93 (6): 1281–1292.
Risk-based triage for nephrology referrals using the 79 Green, J.A., Ephraim, P.L., Hill-Briggs, F.F. et al. (2018).
kidney failure risk equation. Can. J. Kidney Health Dis. 4 Putting patients at the center of kidney care transitions:
2054358117722782. prepare now, a cluster randomized controlled trial.
77 Smekal, M.D., Tam-Tham, H., Finlay, J. et al. (2018). Contemp. Clin. Trials. 73:98–110.
Perceived benefits and challenges of a risk-based 80 Wojciechowski, P., Tangri, N., Rigatto, C., and Komenda,
approach to multidisciplinary chronic kidney disease P. (2016). Risk prediction in CKD: the rational alignment
care: a qualitative descriptive study. Can. J. Kidney Health of health care resources in CKD 4/5 care. Adv. Chronic
Dis. 5 2054358118763809. Kidney Dis. 23 (4): 227–230.
44
Screening for Chronic Kidney Disease

Mohamed A. Osman1, Kwaifa S. Ibrahim2, Nikhil Shah1, Soroush Shojai1, Ikechi G. Okpechi 3,4, and
Aminu K. Bello1
1
Division of Nephrology and Immunology, Department of Medicine, University of Alberta, Edmonton, Alberta, Canada
2
Department of Medicine, Wuse District General Hospital, Nigeria
3
Division of Nephrology and Hypertension, Groote Schuur Hospital, Cape Town, South Africa
4
Kidney and Hypertension Research Unit, University of Cape Town, Cape Town, South Africa
An ounce of prevention is worth a pound of cure. point-of-care screening and treatment for CKD among a
Benjamin Franklin rural adult Canadian indigenous population revealed an
incremental cost-effectiveness ratio (ICER) of $23 700 per
quality-adjusted life-year (QALY) compared to usual
I ntroduction care [7]. Findings from a systematic review show the cost-
effectiveness ratios for proteinuria screening to be $14 063–
Screening is performed to differentiate healthy people from $160 018 QALY for the general population, $5298–$54 943
those in the early stages of disease who are not experienc- QALY for diabetics, and $23 028–$73 939 QALY for hyper-
ing symptoms or exhibiting signs of illness. This might be tensives [8]. However, the cost-benefit ratios of screening
justified when effective early interventions are available have not been adequately studied in developing countries
that lead to better outcomes, for example screening for where such assessments are largely needed.
motor neuron disease would do more harm than good,
given the lack of effective interventions and risk of overdi-
Who Should be Screened for CKD?
agnosis, false-positive results, and associated anxiety [1].
This chapter aims to discuss the utility of a screening pro- Most countries do not have a coordinated approach to CKD
gram for chronic kidney disease (CKD) with regard to asso- screening, leading to so-called “opportunistic screening.”
ciated outcomes and costs. Several researchers have used World Kidney Day as a plat-
CKD is highly prevalent in many countries and access to form to increase awareness of the importance of CKD screen-
treatment is often limited [2, 3]. CKD tends to be an asymp- ing [9–12]. Because screening techniques are not
tomatic condition that goes undetected until its later stages standardized, data on CKD prevalence often are unreliable,
when most kidney function has been lost [4]. The asympto- even within a single country or region [10, 13]. Even in coun-
matic nature of kidney disease is one of the main reasons tries with national programs for early detection of CKD,
why several patients in developing countries often present treatment plans often are unclear and access to healthcare
late for kidney care when dialysis is urgently needed [5]. resources may be limited [14]. Large-scale, population-based
Late presentation, which is associated with increased mor- screening programs should include follow-up protocols and
tality, [6] may be due to late referral; however, a lack of successful care referrals for individuals who test positive to
access to healthcare plays a major role in many low-income ensure appropriate diagnosis, treatment, and counseling.
countries. Screening for CKD and early detection to slow A major problem with screening for CKD revolves
disease progression is thus important in these settings, around who to screen: the general population or only at-
given the high costs associated with treatment. The find- risk populations such as those with diabetes, hyperten-
ings of one study that assessed the cost utility of one-off, sion, family history of CKD, and other patient groups at
Existing CKD Early Detection and Prevention Program 45
increased risk for CKD, such as those with HIV. This harm will increase disproportionately. Therefore, it is
conundrum has generated much debate [15]. Those in important to choose the right frequency for any screening
favor of general population screening argue that early program [27].
identification is clinically and scientifically relevant to A successful detection method is a test with high valid-
reduce the burden of CKD [16, 17], while others use the ity (i.e. specificity and sensitivity). Such a test should also
same argument to support screening among at-risk popu- be highly reliable (i.e. consistently yield the same results
lations only [4, 8, 18–21]. In a population-based study of when repeated under the same conditions) [28].
dipstick screening, Boulware et al. [22] reported an unfa- Albuminuria is an early and sensitive marker of CKD,
vorable ICER for screening individuals with neither whether it is estimated via a 24-hour measurement (gold
hypertension nor diabetes ($282 818 QALY), but more standard) or dipstick assessment of spot urine. Screening
favorable ratios for screening individuals age 60 and older for CKD relies on quick and simple methods to assess kid-
($53 372 QALY) and those with hypertension ($18 621 ney function, such as testing urine for protein (albumin)
QALY). The authors concluded that routine proteinuria and/or measuring serum creatinine. Although a CKD
screening is only cost-effective for high-risk populations diagnosis is confirmed based on two abnormal results
or when performed very infrequently (i.e. every with a repeat test conducted after a 3-month period [29],
10 years) [22]. Those supporting targeted screening even for the purpose of screening and early disease detection,
suggest that harms may be associated with mass screen- one abnormal result is considered a positive test. The
ing due to false-positive results and unnecessary testing CKD prevalence rates reported from screening tests thus
and treatment. Overall, there is insufficient evidence that depend on the assessment methods utilized, with a poten-
providing early treatment for patients without risk factors tial to under-report or over-report identified cases of
who are identified via screening yields benefits [19]. The CKD [16, 30, 31]. For instance, in a large study in Japan,
American College of Physicians (ACP) recommends Uchida et al. [32] showed that a significant proportion of
against screening for asymptomatic adults without risk CKD might be under-reported if only dipstick proteinuria
factors for CKD, as well as adults who are receiving renin- is used to assess kidney function, and strongly recom-
angiotensin-aldosterone system (RAAS) blockade therapy mended that both urinalysis and serum creatinine meas-
due to low-quality evidence [23]. Due to insufficient evi- urement be used in CKD screening. Among 538 846
dence, Kidney Disease – Improving Global Outcomes people, 14.4% had an eGFR below 60 ml/min/1.73 m2,
(KDIGO) did not comment on the benefit of screening for 5.2% had proteinuria, and 18.1% had CKD, 71.4% of whom
early detection of disease versus the harm of “disease had nonproteinuric CKD (47.9% had diabetes mellitus
labeling” [24]. As one analysis suggested, screening for and 66.8% had hypertension). In contrast, a single assess-
CKD may be cost-effective for at-risk patients (i.e. indi- ment of serum creatinine level in the third National
viduals with diabetes and hypertension) and populations Health and Nutrition Examination Survey (NHANES III)
with higher incidences of CKD and rapid rates of progres- was reported as misleading due to age, gender, and ethnic
sion (e.g. HIV-positive patients and relatives of CKD variability [33].
patients) [8, 25].
xisting CKD Early Detection and

E
Methods of CKD Screening and Characteristics
Prevention Programs
of Screening Tests
CKD diagnosis requires detecting persistent structural
Ideally, we would like to have a test with very high sensi- (albuminuria) or functional (low eGFR) abnormalities of
tivity and specificity. However, there is usually a trade off the kidney through laboratory measurements [29].
between these two properties of the test. Imagine we want Collectively, the factors described above make CKD well-
to decide on a cut-off value for estimated glomerular fil- suited for population-based screening for early detection
tration rate (eGFR) to define CKD. The lower the cut-off and treatment. Wilson and Jungner’s first screening princi-
value, the lower the likelihood of mislabeling a healthy ple is the importance of the condition based on its preva-
person with CKD, but the higher the likelihood of miss- lence [34]. Therefore, in this section we examine CKD
ing some patients with CKD [26]. Any screening program prevalence data and summarize relevant population-based
is associated with costs and potential harms (i.e. false- studies of CKD epidemiology in various parts of the world,
positive results causing unnecessary anxiety, overdiagno- and report on recent findings that reveal national capaci-
sis, and overtreatment). As the frequency of screening ties for CKD screening and detection with the following
increases the marginal benefit will plateau, and cost and objectives:
46 Screening for Chronic Kidney Disease
●● To perform a review of the literature of the population- of the CKD staging system, has been instrumental in
based epidemiology of CKD. e stimating the scale of CKD prevalence in the United
●● To describe such studies in terms of design, scope, popu- States at 13.1% (95% CI 12.0–14.1%) [37]. If well executed,
lation, CKD outcomes, and relevant findings. surveys can provide precise data for estimating CKD prev-
●● To identify areas for further work. alence. However, they can be associated with limitations
related to (i) cross-sectional data collection, which typi-
We conducted an extensive literature review by per- cally entails a single measurement of kidney function
forming a broad and intensive search for population- instead of repeat testing to establish CKD diagnosis, (ii)
based studies of CKD published between 2000 and 2018. the vast financial and human resources required to per-
We searched several databases (i.e. PubMed, Cinahl, form large scale surveys [38], and/or (iii) the methods
Embase, and Cochrane) using the following keywords: used to estimate CKD, such as eGFR equations, calibra-
population-based, community, screening, chronic kidney tions, and standardized assays of serum creatinine and
disease, CKD, chronic kidney failure, CKF, end-stage albuminuria [39].
renal disease, ESRD, end-stage kidney disease, ESKD, Another approach to quantifying CKD burden is to ana-
early detection, prevention, management, risk factors, lyze data stored in administrative databases, which often
epidemiology, albuminuria, microalbuminuria, and pro- hold records on a large number of individuals that cannot
teinuria. Only studies that examined CKD and/or albumi- easily be surveyed. These databases often can be repur-
nuria in a community setting and were published in posed to provide information on CKD burden in the pop-
English were retrieved. To be included, a study had to: ulation. For instance, Go and colleagues used data from
●● include a measure of prevalence for CKD and/or albumi- the Kaiser Permanente renal registry, a sizeable health
nuria in a general population plan laboratory database in the United States, to estimate
●● be based on data collected from adults aged 18 years and the morbidity (cardiovascular events, hospitalization)
older who were recruited from a community-based pop- and mortality associated with CKD [40]. They reported an
ulation sample inverse and graded relationship between a decrease in
●● present sufficient detailed methodology and results, as kidney function and increased risk of death, cardiovascu-
well as ethical approval from the relevant authorities lar events, and hospitalization using large population
●● be the original and/or most recent publication if multi- data [40]. This technique has the advantage of obtaining
ple studies based on data from the same population exist data on a large segment of the population over long peri-
ods of time while requiring relatively fewer resources
To inform efforts for the prevention, detection, and than prospective studies. However, some known disad-
treatment of CKD, it is vital to have high-quality evidence vantages of this approach are (i) the accuracy of the data,
that yields accurate estimates of the disease burden in the since they typically are collected for purposes other than
population. The most recent systematic review and meta- research, (ii) the representativeness of the target popula-
analysis estimate the global prevalence (44 countries) of tion, because most data in administrative databases are
CKD to be 13.4% (95% confidence interval [CI] 11.7– from people with access to healthcare, making generaliz-
15.1%) [35]. Another systematic review and meta-analysis ability to the whole population challenging, (iii) the lack
estimates worldwide CKD prevalence to be 10.4% among of well-established health systems that collect data using
men (95% CI 9.3–11.9%) and 11.8% among women (95% rigorous methods in many parts of the world, and (iv) the
CI 11.2–12.6%) [36]. These estimates of global CKD bur- observational nature of these study designs and the poten-
den in the low double-digits were calculated by combin- tial for biases. Of note, data obtained from surveys can be
ing CKD prevalence data published in the literature. stored and later linked to other databases or analyzed
There are several methods to obtain population-level epi- in retrospective studies to answer different research
demiological data on CKD burden, and each has its ben- questions.
efits and drawbacks. The most commonly used are health The Global Burden of Disease (GBD) studies have
examination surveys [37]. When coupled with a repre- revealed additional insights about the burden of CKD. The
sentative random sample of the national population, GBD studies are systematic, scientific efforts using Bayesian
these cross-sectional studies can provide insights into the inferences and sophisticated statistical modeling to quan-
true prevalence of disease and associated risk factors. tify the burden of all major diseases, including CKD [41,
Moreover, periodic health examination surveys can be 42]. The GBD studies involve more than 3000 collaborators
used as surveillance tools to monitor disease trends over and cover more than 195 countries from 1990 to date. In
time. For example, the NHANES, one of the earliest the GBD studies, CKD’s ranking among the hierarchical
attempts to quantify CKD burden since the introduction causes of disability-adjusted life years (DALYs) rose from
Existing CKD Early Detection and Prevention Program 47
30 to 22 between 1990 and 2016 [43]. In addition, the GBD than 50% of countries tested for CKD in patients with
studies predicted an estimated 28.8% increase in CKD cardiovascular disease who were taking chronic nephro-
prevalence from 2006 to 2016; importantly, more than half toxic medications and with a family history of CKD. Elderly
of the increase was attributed to diabetes (+30.1%) and populations (age 65 years) also were tested for CKD in
hypertension (+34.7%) [44]. Although GBD studies are more than half of the countries in the high and upper-
currently at the forefront of global efforts to estimate dis- middle income groups, but not in the lower-middle (47%)
ease burden, they have several limitations, including cod- and low (41%) income groups. Overall, CKD testing rates
ing practices, sampling measurement, and case definition followed the economic development classification, with
errors involved in obtaining and merging this voluminous the highest rates in higher income countries and lowest
data from a wide array of sources [41, 42]. Finally, prospec- rates in low-income countries (Figure 4.2).
tive registries that collect data for surveillance and moni-
toring purposes are important sources for estimating CKD
CKD Detection Program
burden. Although end-stage kidney disease (ESKD) regis-
tries on disease incidence and prevalence are well estab- The GKHA also evaluated the presence of an active CKD
lished in many developing countries, predialysis CKD detection program based on national policies and/or guide-
registries are almost nonexistent. In the Global Kidney lines. Overall, less than a third of countries reported having
Health Atlas (GKHA) survey, only nine out of 125 coun- a CKD detection program. Most of the programs were in
tries surveyed reported viable CKD registries [45]. high-income countries (32%), followed by upper-middle
(27%) and lower-middle (22%) income countries. Among
the responding low-income countries, only one reported
National Capacity for CKD Screening
an active CKD detection program (Figure 4.3). The major-
The GKHA project evaluated national capacities to detect ity of countries with CKD detection programs implemented
CKD in both primary and secondary care settings them through active screening during routine health
(Figure 4.1), including several CKD screening-related encounters (68%) or through specific screening processes
anthropometric and laboratory measurements that are (57%), while 54% reported a reactive approach in their pro-
necessary to identify CKD risk factors or kidney damage. gram implementation (Figure 4.3). These findings indicate
Interestingly, less than half of low-income countries did that there is no single program implementation approach.
not have the capacity to routinely test for serum glucose, Moreover, many countries reported more than a single
serum creatinine urinalysis (qualitative and quantitative), method in their program implementation.
and urine portions in primary care settings. Moreover, From what has been presented so far, it can be inferred
none of the low-income countries reported the capacity to that many studies have been performed worldwide on
automatically report eGFR or test for urine proteins in pri- the prevalence of CKD at both the national and commu-
mary care settings. Similarly, lower-middle income coun- nity levels. Many of these studies are based on the initial
tries reported limited capacity to measure HbA1c and observation by the NHANES III that up to 11% of the US
serum creatinine with and without eGFR, and to perform population may have CKD. Prevalence of albuminuria/
urinalysis (qualitative and quantitative) and measure proteinuria in the results ranges from 0.6% to 29.7% in
urine proteins. While less than half of upper-middle general population studies, and from 2.7% to 45.4% in
income countries had these capacities, the majority of studies of high-risk individuals. Overall, there is a graded
high-income countries reported the availability of these effect in terms of the number and size of the studies
services (Figure 4.1). As expected, CKD-related health ser- observed, from high-income countries to low-income
vices were more available in secondary care settings; how- countries. These significant variations may undermine
ever, capacities for automated eGFR reporting and developing countries’ abilities to develop screening pro-
proteinuria assessment were limited compared to the grams tailored to their contexts. Recent findings from
other services, especially in low income countries. the GKHA that many countries have insufficient capac-
ity for detecting and monitoring CKD in both primary
and secondary care settings is concerning. Any discus-
CKD Testing for High-risk Populations
sion of CKD screening programs implies resource avail-
The GKHA evaluated routine testing for CKD among sev- ability and testing capabilities. Currently, the two pillars
eral high-risk populations (Figure 4.2). Across countries in of CKD detection are estimated GFR from serum creati-
all income groups, CKD testing was almost universal for nine and urine proteins, and capacities to perform these
patients with hypertension and diabetes (94–100%), but measurements are significantly limited in many parts of
minimal for high-risk ethnic populations (<30%). More the world.
Primary care
Blood Serum creatinine Serum creatinine Urinalysis Urinalysis

Serum glucose HbA1c UACR or UPCR
pressure without eGFR with eGFR (qualitative assays) (quantitative assays)
Low income 94 41 0 29 0 41 6 0
Lower-middle income 94 76 30 55 15 55 30 15
Upper-middle income 100 94 45 87 42 71 42 35
High income 97 97 76 71 68 87 71 58
Secondary care
Blood Serum creatinine Serum creatinine Urinalysis Urinalysis

Serum glucose HbA1c UACR or UPCR
pressure without eGFR with eGFR (qualitative assays) (quantitative assays)
Low income 100 100 44 81 19 88 57 6

Lower-middle income 100 97 79 94 47 85 58 53
Upper-middle income 100 100 94 100 74 90 74 77
High income 100 100 97 89 89 97 89 97
Figure 4.1 Capacity for identification and management of chronic kidney disease in primary and secondary care settings, by World Bank income group. HbA1C, glycated
hemoglobin; eGFR, estimated glomerular filtration rate; UACR, urine albumin-creatinine ratio; UPCR, urine protein-creatinine ratio. Source: Data sourced from [46].
Review of Current Guideline Recommendations for CKD Screenin 49
Low income
Lower-middle income
Upper-middle income
Hypertension
High income 100%
80%
Family history of CKD Diabetes mellitus
60%
40%
20%
0%
High-risk ethnic groups Cardiovascular diseases
Nephrotoxic medications Elderly (≥ 65 years)
Figure 4.2 Chronic kidney disease testing for high-risk populations, by World Bank income group. CKD, chronic kidney disease.
Source: Data sourced from [46].
Low income 6% Active screening

(specific screening processes) 57%
Lower-middle income 22%

Active screening
(routine health encounters) 68%
Upper-middle income 27%
Reactive approach 54%

High income 32%
Figure 4.3 Existence of CKD detection programs, by World Bank income group and implementation method. Source: Data sourced
from [46].
eview of Current Guideline

R screening, including disease labeling, adverse effects of
Recommendations for CKD Screening testing and treatment, and financial consequences. Thus,
they recommended against screening for CKD in asymp-
tomatic adults without risk factors for CKD (grade: weak
In 2012, the US Preventive Services Task Force (USPSTF) recommendation, low-quality evidence). They also rec-
reviewed evidence related to screening for CKD in ommended against testing for proteinuria in adults with
asymptomatic adults and concluded that it is insufficient or without diabetes who are currently taking an
to assess the balance of benefits and harms of routine angiotensin-converting enzyme inhibitor (ACEi) or an
testing. Thus, they did not provide any recommendations angiotensin II–receptor blocker (grade: weak recommen-
in their report, even for people with diabetes and hyper- dation, low-quality evidence). The ACP also found the
tension [47]. Likewise, the ACP did not find any rand- current evidence insufficient to evaluate the benefits and
omized control trials evaluating the benefits or harms of harms of screening for CKD in asymptomatic adults with
screening for CKD stages 1–3 among the general popula- CKD risk factors [23]. These recommendations were
tion. They pointed out potential harms of universal endorsed by the American Academy of Family Physicians
(AAFP) in a guideline statement published in 2014 [48]. ffectiveness of Screening for CKD in

E
In the UK, the National Institute of Health and Care Terms of Costs and Consequences
Excellence (NICE) made several recommendations
regarding CKD screening for at-risk populations [49].
NICE did not make any explicit recommendation for Given its silent nature and associated negative conse-
asymptomatic people without known risk factors for quences, early detection and treatment of CKD must be
CKD, and strongly recommended against using age, gen- emphasized. One way to achieve these goals is to perform
der, ethnicity, obesity in the absence of a metabolic syn- primary screening for CKD using blood or urine tests; how-
drome, diabetes, or hypertension as risk factors. For ever, it is important to understand the cost-effectiveness of
people with risk factors for CKD, they recommended doing so prior to adopting such an approach [8, 56–59]. In
offering CKD screening tests, and for patients with healthcare settings, decisions typically are made with con-
nephrotoxic drug prescriptions, they strongly suggested sideration to benefits, risks, and costs, yet the economic
annual screening for CKD. component of any program arguably is the key determinant.
Although the American Society of Nephrology does not Economic analyses are decision tools based on formal quan-
have a guideline, it issued a press release in response to the titative methods to investigate and compare program adop-
ACP guidelines in which they strongly suggested screening tion to the usual care and/or no care while taking into
for CKD among general populations, regardless of risk fac- account the costs of diverting resources from competing pri-
tors [50]. In 2013, KDIGO published one of the most com- orities, otherwise known as opportunity costs. The most
prehensive guideline recommendations for management widely used economic measurement to compare and evalu-
of CKD; however, these guidelines do not explicitly include ate programs with different costs and outcomes is ICER.
screening recommendations [29]. In 2002, the National ICER is the ratio of the difference in costs to the difference in
Kidney Foundation (NKF) and the Kidney Disease benefits of the program or intervention. Costs typically are
Outcome Quality Initiative (KDOQI) had recommended represented in monetary terms, while benefits are expressed
screening at-risk individuals for CKD [51]. In a 2014 comin clinical terms such as life-years gained (LYG) or lost, or
mentary on the 2012 KDIGO guidelines, they highlighted QALYs gained or lost. QALY is a bi-dimensional composite
KDIGO’s lack of specific recommendations for CKD measurement of length of life and quality of life expressed in
screening and endorsed their previous recommendation to a single number. ICER is a way to ascertain both the effec-
screen individuals at high risk for developing CKD [52]. tiveness and cost of an intervention compared to another,
They disagreed with the ACP’s recommendation not to hence the name cost-effectiveness. Additionally, interpret-
screen patients who take ACEi or angiotensin receptor ing ICER requires a value judgment to be placed on the
blocker (ARB) medications. opportunity cost of QALY or LYG. Since data on opportunity
Other renal organizations around the world have pub- costs are limited, determining this threshold can be chal-
lished their own commentaries or recommendations, with lenging and tends to be based either on historical funding
a common theme of screening individuals at high risk for decisions (e.g. $50 000 per QALY from dialysis) or consensus
CKD. In its commentary on the KDIGO guidelines, the agreements (e.g. $80 000 per QALY).
Canadian Society of Nephrology outlined consensus state- Several studies have looked into the cost-effectiveness of
ments. They did not recommend screening individuals for screening for CKD using eGFR and/or proteinuria [59–70].
CKD, but recommended that practitioners continue to use These studies (chapters 17, 22,59 and 62–67) are nicely
case findings as a means to identify patients with CKD in summarized in a recent systematic review on the cost-
high-risk populations [53]. The Kidney Health Australia – effectiveness of CKD screening studies from 1992 to 2012
Caring for Australasians with Renal Impairment (KHA- (Table 4.2) [8]. In this section, we use these systematic
CARI) guidelines strongly recommend screening people at review studies as a framework to evaluate the cost-
higher risk of CKD due to diabetes, hypertension or estab- effectiveness criteria for CKD screening.
lished cardiovascular disease (1B), or additional risk fac- None of the included studies were randomized control
tors such as obesity, a history of smoking, Aboriginal and trials, and all were based on Markov simulation models.
Torres Strait Islander heritage, family history of stage 5 Seven of the studies evaluated proteinuria, one study eval-
CKD or hereditary kidney disease in a first- or second- uated eGFR, and another study evaluated both proteinuria
degree relative, or severe socioeconomic disadvantage and eGFR as CKD screening methods. The comparator to
(1D) [54]. Similarly, the UK Renal Association strongly rec- screening was usual care in four studies, no screening in
ommends screening patients at higher risk of CKD three studies, and both in two studies. As cost-effectiveness
(1B) [55]. These recommendations are summarized in modeling assumptions depend on high-quality evidence,
Table 4.1. all of the included cost-effectiveness studies were based on
Table 4.1 Screening recommendations by the professional medical and nephrology societies.
Nature of
Organization organization Date Population Recommendation statement Strength and evidence of recommendation
US Preventive Government/ 2012 Asymptomatic adults without No recommendation Grade I (insufficient evidence): The
Services Task Force Generalist diagnosed CKD evidence is insufficient to assess the
balance of benefits and harms of routine
screening for CKD in asymptomatic adults.
Evidence is lacking, of poor quality, or
conflicting, and the balance of benefits and
harms cannot be determined.
American College Generalist 2013 Asymptomatic adults without risk ACP recommends against screening for chronic Grade: Weak recommendation, low-quality
of Physicians factors for CKD kidney disease in asymptomatic adults without evidence
risk factors for chronic kidney disease
American Academy Generalist 2014 Endorsed recommendation made by the
of Family American College of Physicians
Physicians
National Institute Generalist 2014; People prescribed nephrotoxic drugs 1.1.27 Monitor eGFR at least annually for Strong
of Health and Care updated people prescribed drugs known to be
Excellence (NICE), 2015 nephrotoxic, such as calcineurin inhibitors (for
UK example, cyclosporin, or tacrolimus), lithium
and non-steroidal anti-inflammatory drugs
(NSAIDs) (2008, amended 2014)
People with risk factors for CKD 1.1.28 Offer testing for CKD using eGFR Strong
creatinine and ACR to people with any of the
following risk factors: diabetes, hypertension,
acute kidney injury, cardiovascular disease
(ischemic heart disease, chronic heart failure,
peripheral vascular disease, or cerebral vascular
disease), structural renal tract disease, recurrent
renal calculi or prostatic hypertrophy, and
multisystem diseases with potential kidney
involvement, e.g. systemic lupus
erythematosus, family history of end-stage
kidney disease (GFR category G5), or
hereditary kidney disease, opportunistic
detection of hematuria
Asymptomatic adults 1.1.29 Do not use age, gender or ethnicity as Strong
risk markers to test people for CKD. In the
absence of metabolic syndrome, diabetes, or
hypertension, do not use obesity alone as a risk
marker to test people for CKD [2008, amended
2014].
(Continued)
0005152392.INDD 51 09-12-2022 12:42:25

Nature of
Organization organization Date Population Recommendation statement Strength and evidence of recommendation
American Society Renal 2013 Asymptomatic adults Strongly recommends regular screening for
of Nephrology organization kidney disease, regardless of an individual’s
risk factors
National Kidney Renal 2002 Individuals at risk for CKD Guideline 3: Individuals at increased risk of Opinion
Foundation/Kidney organization CKD:
Disease Outcomes Some individuals without kidney damage and
Quality Initiative with normal or elevated GFR are at increased
risk for development of CKD.
All individuals should be assessed, as part of
routine health encounters, to determine
whether they are at increased risk of developing
chronic kidney disease, based on clinical and
sociodemographic factors.
Individuals at increased risk of developing CKD
should undergo testing for markers of kidney
damage and to estimate the level of GFR.
Individuals found to have CKD should be
evaluated and treated as specified in Guideline
2.
Individuals at increased risk but found not to
have CKD should be advised to follow a
program of risk factor reduction, if appropriate,
and undergo repeat periodic evaluation.
2013 Individuals at risk for CKD The commentary work group endorses the Opinion
recommendations from the original guideline
for screening among individuals at high risk for
CKD despite the absence of this specific
recommendation in the KDIGO guideline, and
as such is in agreement with the
recommendation of the ACP to screen
asymptomatic adults only if they are at high
risk for CKD.
Kidney Disease Renal 2013 Screening of CKD was beyond the scope of the
Improving Global organization CKD Work Group and no specific
Outcomes recommendations were made.
Canadian Society Renal 2015 Asymptomatic individuals No screening Consensus

of Nephrology organization
0005152392.INDD 52 09-12-2022 12:42:25

At risk of CKD Practitioners should continue to use case Consensus
findings in keeping with usual clinical practice:
in people with new onset or long-standing
hypertension or diabetes, people with vascular
disease, people who are to undergo major
surgery or be exposed to other potential causes
of acute kidney injury, people with multisystem
or generalized symptoms, people who are being
considered for nephrotoxic medications or
medications that require dose adjustment for
renal function, people with a family history of
polycystic kidney disease or hereditary
nephritis, and people from First Nations
populations or other ethnic groups known to be
at increased risk.
KHA-CARI Renal 2013 Diabetes, hypertension, or We recommend that screening for CKD be 1B
Guidelines organization established cardiovascular disease. targeted and performed in individuals at
increased risk of developing CKD, including
those with diabetes mellitus, hypertension, and
established CVD
Additional risk factors such as: We recommend screening in those with 1D

obesity, cigarette smoking; additional CKD risk factors identified in
Aboriginal and Torres Strait Islander Guideline 2a (obesity, cigarette smoking,
peoples; family history of stage 5 Aboriginal and Torres Strait Islander peoples,
CKD or hereditary kidney disease in family history of stage 5 CKD, or hereditary
a first-or second-degree relative; and kidney disease in a first-or second-degree
severe socioeconomic disadvantage. relative and severe socioeconomic
disadvantage) (1D)
UK Renal Renal 2011 At-risk for CKD Guideline 1.4 – CKD: Detection and monitoring 1B
Association organization of CKD: We recommend that patients who are
at increased risk for developing CKD should be
offered screening tests to detect CKD.
KD, chronic kidney disease; ACP, American College of Physicians; NICE, National Institute of Health and Care Excellence; eGFR, estimated glomerular filtration rate; ACR, albumin to
C
creatinine ratio; KDIGO, Kidney Disease Improving Global Outcomes; KHA-CARI, Kidney Health Australia – Caring for Australasians with Renal Impairment; CVD, cardiovascular disease.
0005152392.INDD 53 09-12-2022 12:42:26

Table 4.2 CKD cost-effectiveness studies.
Summary Screening
Study Study population Screening method measure Comparator Perspective frequency
Proteinuria-based screening
Boersma General Dutch population aged Microalbuminuria (UAE 30–300 mg/day) Cost/LYG No screening Healthcare One-off
et al. [62] 28–75 years payer
Boulware US adults aged 50–75 years Proteinuria (dipstick) Cost/QALY Usual care Societal Annual
et al. [22]
Hoerger US adults aged 50–90 years Microalbuminuria (ACR 30–299 mg/g) Cost/QALY No screening and Healthcare 1-, 2-, 5-, 10-year
et al. [63] usual care payer intervals
Howard Hypertensive and diabetic Proteinuria (dipstick followed by spot Cost/QALY Usual care Healthcare Annual
et al. [64] Australian adults aged 50–69 years UPCR; 0.20 mg/mg confirmatory test) payer
Kessler et al. [65] Swiss adults aged 50–90 years Microalbuminuria (ACR 30–299 mg/g) Cost/QALY No screening and usual Healthcare 1-, 2-, 5-, 10-year
care payer intervals
Kondo et al. [17] Japanese adults aged 40–74 years Proteinuria (dipstick) Cost/QALY No screening Societal Annual
Palmer et al. [66] US hypertensive, type 2 diabetics Microalbuminuria (UAE 20–199 mg/min) Cost/QALY No screening Healthcare Annual
payer
Siegel et al. [67] US insulin-dependent diabetics Proteinuria (dipstick 300 mg/min) Cost/LYG Usual care Healthcare Biannual
(aged 15 years at diagnosis) payer
eGFR-based screening
Kondo et al. [17] Japanese adults aged 40–74 years eGFR (<50 ml/min/1.73 m2 and Cost/QALY No screening Societal Annual
hypertension, diabetes, or hyperlipidemia)
Manns et al. [59] Canadian healthcare system eGFR (<60 ml/min/1.73 m2) Cost/QALY Usual care Healthcare One-off
payer
ACR, albumin-creatinine ratio; eGFR, estimated glomerular filtration rate; LYG, life-year gained; QALY, quality-adjusted life-year; UAE, urinary albumin excretion; UPCR, urinary protein-
creatinine ratio. Source: Adapted with permission from Komenda et al. [8].
0005152392.INDD 54 09-12-2022 12:42:26

Effectiveness of Screening for CKD in Terms of Costs and Consequence 55
data from high-income countries in North America (the both, is cost-effective [23, 33]. Screening also is a cost-
United States and Canada), Europe (Netherlands, effective strategy for certain high-risk ethnic groups, such
Switzerland), Asia (Japan), and Australia. The majority of as African Americans and indigenous people of Canada,
studies took a healthcare perspective while two studies who are at increased risk of kidney failure and its numer-
adopted a broader social perspective. Screening frequency ous complications [31]. The cost-effectiveness of a screen-
varied from annual to biannual, to several year intervals, ing program for proteinuria among US adults with
and some evaluated on–off scenarios. The majority of hypertension and diabetes has been clearly demonstrated
included studies determined cost ratios by measuring cost in terms of reducing death and slowing the progression of
to QALYs; however, two studies used cost to LYG as the kidney disease by using an ACEi or ARB [59].
summary measure (Table 4.2). Proteinuria-based studies Overall, the cost-effectiveness ratio of screening versus
yielded a wide range of cost ratios ranging from $14 000 to the usual care (no screening) is unfavorable for the general
more than $160 000 per QALY among the general popula- population ($282 818 per QALY saved). Just 0.0022 QALYs
tion, from $5000 to more than $50 000 per QALY among per person are gained by preventing of one new case of
people with diabetes, and from $23 000 to more than ESKD and seven deaths per one million persons per year.
$70 000 among hypertensive populations (Figure 4.4). In For individuals with hypertension, the cost-effectiveness
eGFR-based studies, the cost ratio was around $100 000 per ratio of screening versus no screening is considered highly
QALY for the general population in two studies and $20 000 favorable ($18 621 per QALY-saved) [22], with 0.03 QALYs
per QALY for diabetes patients in a single study (Figure 4.4). per person gained through the prevention of 14 new cases
Although population-based screening offers potential of ESKD and 104 deaths per 1 million persons per year. For
benefits such as early identification and treatment of persons with diabetes, the cost-effectiveness ratio of screen-
affected patients, it also has drawbacks, such as identifica- ing versus no screening is overwhelmingly positive ($217
tion of patients with mild disease who may not need addi- per QALY saved), with a gain of 0.10 QALYS per person
tional treatment [59]. Findings show that screening for due to the prevention of approximately 84 new cases of
moderately increased albumin excretion (microalbuminu- ESKD and 541 deaths per 1 million persons per year [22].
ria), which is a marker of early-stage CKD, is cost-effective The results of a similar study carried out among members
for patients with diabetes or hypertension [22, 36, 68]. of the Canadian indigenous community demonstrate that
Early detection, appropriate risk stratification, and subse- screening and treatment for CKD is highly cost-effective,
quent treatment of CKD may delay or prevent many with a cost-effectiveness ratio of $23 700 per QALY saved.
associated complications, with concomitant increase in The cost-effectiveness ratio of CKD screening is even more
cost [7, 69]. profound in the most remote communities accessible only
Mass screening of the general population for CKD is not by air travel, at $7790 per QALY saved [7]. Integrating addi-
recommended because it is not cost-effective for low-risk tional benefits of screening and treatment into the existing
groups [23, 41]. In contrast, targeted screening of high-risk framework may improve the cost-effectiveness of the given
individuals, such as those with diabetes or hypertension or screening intervention [7, 22]. In previous studies,
ICER: $160,000 $140,000 $120,000 $100,000 $80,000 $80,000 $40,000 $20,000 $0

Not cost-effective Cost-effective
$160,018 [63]
General
population
$14,063 [17]
$54,973 [65]
Diabetic
population
$5,298 [64]
$73,939 [63]
Hypertensive
population
$23,028 [22]
Figure 4.4 Tornado plot comparing reported incremental cost-effectiveness ratios (ICERs) of CKD screening in selected
populations.$80 000 ICER selected based on 1–3 times gross domestic product per capita threshold for most G8 (Group of 8)
countries. Source: Adapted with permission from [8] © 2014 Elsevier.
researchers have considered the impact of albuminuria elucidated the key variables that might improve cost-
screening on the reduction of cardiovascular events and effectiveness, including adherence to ACEi or ARB ther-
found screening to be cost-effective, even for the general apy, the incidence of proteinuria, sensitivity, and specificity
population [62, 70]. Some potential benefits of screening of testing, the effectiveness of ACEi or ARB therapies in
include the reduction of cardiovascular events among slowing the progression toward ESKD or preventing death,
patients with diabetes through the use of blood pressure and decreased frequency of screening [22, 60, 68].
lowering medication, and the potential for increased utili-
zation of statins in newly identified CKD cases [71–73]. In
addition, newer therapies such as sodium-glucose cotrans- Summary
porter 2 inhibitors can provide benefits for patients with
type 2 diabetes and high cardiovascular risk [74]. Screening the general population for early CKD is not cost-
Multiple factors could influence the efficacy of these effective and therefore is unwarranted. Selective screening
treatment modalities, such as prevailing usage rates for should be directed toward high-risk groups (older persons
each pharmacotherapy, the underlying prevalence of and persons with hypertension and diabetes, and some eth-
untreated diabetes in screened communities, and ancillary nic groups such as African Americans and indigenous pop-
effects of screening programs such as community educa- ulations of Canada, the United States, and Australia).
tion, increased awareness, and lifestyle interventions [7]. Adherence to ACEi or ARB pharmacotherapy in prevent-
Overall, the effectiveness of CKD screening and treatment ing death or progression toward ESKD and the frequency
could be better measured in a cluster randomized trial that of screening play prominent roles in the cost-effectiveness
can better capture multiple concurrent effects and their of screening members of the general population.
influences on patient outcomes [7]. Direct and indirect cost Decision makers often are pressured to adopt policies
savings associated with screening are primarily achieved that are not necessarily economically sound. However, in a
through the prevention of new (and expensive) ESKD healthcare environment characterized by scarce resources
cases [64]. Years of life saved through the death prevention and competing demands, it is essential to consider how to
effects of ACEi and ARB therapies are augmented by the maximize value. Economic evaluations introduce an
prevention of ESKD cases and their associated high mor- evidence-based perspective to these decisions. In the case
tality [22, 60]. A lack of cost-effectiveness arises when the of CKD, the most recent evidence indicates that CKD
prevalence and incidence of proteinuria are very low (lead- screening is warranted only for high-risk patients, espe-
ing to few preventable cases of ESKD), and also when the cially those with diabetes and hypertension, to prevent
risk of death is very low [64]. Physicians must consider associated cardiovascular morbidity and mortality. More
many factors when making decisions regarding early research is needed to evaluate the armamentarium availa-
disease detection. Numerous studies [22, 60, 64] have ble for CKD screening and its cost-effectiveness.
References
Raffle A. (2018). Further resources to interactive learning

1 5 Adejumo, O.A., Akinbodewa, A.A., Okaka, E.I. et al.
module screening contents [Internet]. www. (2016). Chronic kidney disease in Nigeria: late presentation
healthknowledge.org.uk/interactive-learning/screening is still the norm. Niger. Med. J. 57 (3): 185–189.
(accessed 22 May 2021). 6 Udayaraj, U.P., Haynes, R., and Winearls, C.G. (2011). Late
2 Kaze, A.D., Ilori, T., Jaar, B.G., and Echouffo-Tcheugui, J.B. presentation of patients with end-stage renal disease for
(2018). Burden of chronic kidney disease on the African renal replacement therapy – is it always avoidable?
continent: a systematic review and meta-analysis. BMC Nephrol. Dial. Transplant. 26 (11): 3646–3651.
Nephrol. 19 (1): 125. 7 Ferguson, T.W., Tangri, N., Tan, Z. et al. (2017). Screening
3 Liyanage, T., Ninomiya, T., Jha, V. et al. (2015). Worldwide for chronic kidney disease in Canadian indigenous peoples
access to treatment for end-stage kidney disease: a is cost-effective. Kidney Int. 92 (1): 192–200.
systematic review. Lancet 385 (9981): 1975–1982. 8 Komenda, P., Ferguson, T.W., Macdonald, K. et al.
4 Berns, J.S. (2014). Routine screening for CKD should be (2014). Cost-effectiveness of primary screening for
done in asymptomatic adults. . . selectively. Clin. J. Am. CKD: a systematic review. Am. J. Kidney Dis. 63 (5):
Soc. Nephrol. 9 (11): 1988–1992. 789–797.
Reference 57
9 Perico, N., Plata, R., Anabaya, A. et al. (2005). Strategies 22 Boulware, L.E., Jaar, B.G., Tarver-Carr, M.E. et al. (2003).
for national health care systems in emerging countries: Screening for proteinuria in US adults: a cost-
the case of screening and prevention of renal disease effectiveness analysis. JAMA 290 (23): 3101–3114.
progression in Bolivia. Kidney Int. Suppl. 97: S87–S94. 23 Qaseem, A., Hopkins, R.H. Jr., Sweet, D.E. et al. (2013).
10 Makusidi, M.A., Liman, H.M., Yakubu, A. et al. (2013). Screening, monitoring, and treatment of stage 1 to 3
Prevalence of non-communicable diseases and its chronic kidney disease: a clinical practice guideline from
awareness among inhabitants of Sokoto metropolis: the American College of Physicians. Ann. Intern. Med.
outcome of a screening program for hypertension, 159 (12): 835–847.
obesity, diabetes mellitus and overt proteinuria. Arab. J. 24 Levey, A.S., de Jong, P.E., Coresh, J. et al. (2011). The
Nephrol. Transplant. 6 (3): 189–191. definition, classification, and prognosis of chronic kidney
11 Garcia-Garcia, G., Marquez-Magana, I., Renoirte-Lopez, disease: a KDIGO controversies conference report. Kidney
K. et al. (2010). Screening for kidney disease on World Int. 80 (1): 17–28.
Kidney Day in Jalisco. Mexico. J. Nephrol. 23 (2): 224–230.
25 Li, P.K., Ng, J.K., Cheng, Y.L. et al. (2017). Relatives in
12 Sumaili, E.K., Nseka, N.M., Lepira, F.B. et al. (2008).
silent kidney disease screening (RISKS) study: a Chinese
Screening for proteinuria and chronic kidney disease risk
cohort study. Nephrology 22 (Suppl 4): 35–42.
factors in Kinshasa: a world kidney day 2007 study.
Nephron Clin. Pract. 110 (4): c220–c228. 26 Pope, D. and Stanistreet, D. (2017). Quantitative Methods
13 Egbi, O.G., Okafor, U.H., Miebodei, K.E. et al. (2014). for Health Research: A Practical Interactive Guide to
Prevalence and correlates of chronic kidney disease Epidemiology and Statistics. Wiley.
among civil servants in Bayelsa state. Nigeria. J. Clin. 27 Bell, N.R., Grad, R., Dickinson, J.A. et al. (2017). Better
Pract. 17 (5): 602–607. decision making in preventive health screening:
14 Singh, A.K., Farag, Y.M., Mittal, B.V. et al. (2013). balancing benefits and harms. Can. Fam. Physician 63
Epidemiology and risk factors of chronic kidney disease (7): 521–524.
in India – results from the SEEK (Screening and Early 28 Jaar, B.G., Khatib, R., Plantinga, L. et al. (2008). Principles
Evaluation of Kidney Disease) study. BMC Nephrol. 14: of screening for chronic kidney disease. Clin. J. Am. Soc.
114. Nephrol. 3 (2): 601–609.
15 Kliger, A.S. (2014). Screening for CKD: a pro and con 29 Levin, A., Stevens, P.E., Bilous, R.W. and Kidney Disease:
debate. Clin. J. Am. Soc. Nephrol. 9 (11): 1987. Improving Global Outcomes (KDIGO) CKD Work Group
16 Benghanem Gharbi, M., Elseviers, M., Zamd, M. et al. et al. (2013). KDIGO 2012 clinical practice guideline for
(2016). Chronic kidney disease, hypertension, diabetes, the evaluation and management of chronic kidney
and obesity in the adult population of Morocco: how to disease. Kidney Int. Suppl. 3(1): 1–150.
avoid “over”-and “under”-diagnosis of CKD. Kidney Int. 30 Delanaye, P., Glassock, R.J., and De Broe, M.E. (2017).
89 (6): 1363–1371. Epidemiology of chronic kidney disease: think (at least)
17 Kondo, M., Yamagata, K., Hoshi, S.-L. et al. (2012). twice! Clin. Kidney J. 10 (3): 370–374.
Cost-effectiveness of chronic kidney disease mass
31 Komenda, P., Lavallee, B., Ferguson, T.W. et al. (2016).
screening test in Japan. J. Clin. Exp. Nephrol. 16 (2):
The prevalence of CKD in rural Canadian indigenous
279–291.
peoples: results from the first nations community based
18 Hallan, S.I. and Stevens, P. (2010). Screening for chronic
screening to improve kidney health and prevent dialysis
kidney disease: which strategy? J. Nephrol. 23 (2):
(FINISHED) screen, triage, and treat program. Am. J.
147–155.
Kidney Dis. 68 (4): 582–590.
19 Qaseem, A., Wilt, T.J., Cooke, M., and Denberg, T.D.
(2014). The paucity of evidence supporting screening for 32 Uchida, D., Kawarazaki, H., Shibagaki, Y. et al. (2015).
stages 1–3 CKD in asymptomatic patients with or without Underestimating chronic kidney disease by urine dipstick
risk factors. Clin. J. Am. Soc. Nephrol. 9 (11): 1993–1995. without serum creatinine as a screening tool in the general
20 de Lima, A.O., Kesrouani, S., Gomes, R.A. et al. (2012). Japanese population. J. Clin. Exp. Nephrol. 19 (3): 474–480.
Population screening for chronic kidney disease: a survey 33 Jones, C.A., McQuillan, G.M., Kusek, J.W. et al. (1998).
involving 38,721 Brazilians. Nephrol. Dial. Transplant. 27 Serum creatinine levels in the US population: third
(Suppl 3): iii135–iii138. National Health and Nutrition Examination Survey. Am.
21 Collins, A.J., Vassalotti, J.A., Wang, C. et al. (2009). Who J. Kidney Dis. 32 (6): 992–999.
should be targeted for CKD screening? Impact of 34 Wilson JMG, Jungner G, World Health Organization
diabetes, hypertension, and cardiovascular disease. Am. J. (1968). Principles and practice of screening for
Kidney Dis. 53 (3 Suppl 3): S71–S77. disease.
35 Hill, N.R., Fatoba, S.T., Oke, J.L. et al. (2016). Global 48 Lambert, M. (2014). ACP releases guideline on screening,
prevalence of chronic kidney disease – a systematic monitoring, and treatment of stage 1 to 3 chronic kidney
review and meta-analysis. PLoS One 11 (7): e0158765. disease. Am. Fam. Physician 90 (2): 121–122.
36 Mills, K.T., Xu, Y., Zhang, W. et al. (2015). A systematic 49 National Institute of Health and Care Excellence (NICE)
analysis of worldwide population-based data on the UK. Chronic kidney disease in adults: assessment and
global burden of chronic kidney disease in 2010. Kidney management clinical guideline [CG182]. (2014). www.
Int. 88 (5): 950–957. nice.org.uk/guidance/cg182/chapter/Introduction
37 Coresh, J., Selvin, E., Stevens, L.A. et al. (2007). (accessed 28 November 2018).
Prevalence of chronic kidney disease in the United States. 50 The American Society of Nephrology (ASN) (2013). ASN
JAMA 298 (17): 2038–2047. emphasizes need for early detection of kidney disease, a
38 Thygesen, L.C. and Ersbøll, A.K. (2014). When the entire silent killer [press release].
population is the sample: strengths and limitations in 51 Levey, A.S., Coresh, J., Bolton, K. et al. (2002). K/DOQI
register-based epidemiology. Eur. J. Epidemiol. 29 (8): clinical practice guidelines for chronic kidney disease:
551–558. evaluation, classification, and stratification. Am. J. Kidney
39 Stevens, L.A., Coresh, J., Greene, T., and Levey, A.S. Dis. 39 (2 suppl. 1) S1–S266.
(2006). Assessing kidney function – measured and 52 Inker, L.A., Astor, B.C., Fox, C.H. et al. (2014). KDOQI US
estimated glomerular filtration rate. N. Engl. J. Med. 354 commentary on the 2012 KDIGO clinical practice
(23): 2473–2483. guideline for the evaluation and management of CKD.
40 Go, A.S., Chertow, G.M., Fan, D. et al. (2004). Chronic Am. J. Kidney Dis. 63 (5): 713–735.
kidney disease and the risks of death, cardiovascular
53 Akbari, A., Clase, C.M., Acott, P. et al. (2015). Canadian
events, and hospitalization. N. Engl. J. Med. 351 (13):
Society of Nephrology commentary on the KDIGO
1296–1305.
clinical practice guideline for CKD evaluation and
41 Jager, K.J. and Fraser, S.D. (2017) The ascending rank of
management. Am. J. Kidney Dis. 65 (2): 177–205.
chronic kidney disease in the global burden of disease
54 Johnson, D.W., Atai, E., Chan, M. et al. (2013). KHA-
study. Nephrol. Dial. Transplant. 32 (suppl_2):
CARI guideline: early chronic kidney disease: detection,
ii121–ii128.
prevention and management. Nephrology 18 (5): 340–350.
42 Murray, C.J. and Lopez, A.D. (2017). Measuring global
55 MacGregor, M.S. and Taal, M.W. (2011). Renal
health: motivation and evolution of the Global Burden of
Association Clinical Practice Guideline on detection,
Disease Study. Lancet 390 (10100): 1460–1464.
monitoring and management of patients with CKD.
43 Hay, S.I., Abajobir, A.A., Abate, K.H. et al. (2017). Global,
Nephron Clin. Pract. 118 (Suppl. 1): c71–c100.
regional, and national disability-adjusted life-years
56 Saran, R., Robinson, B., Abbott, K.C. et al. (2017). US
(DALYs) for 333 diseases and injuries and healthy life
renal data system 2016 annual data report: epidemiology
expectancy (HALE) for 195 countries and territories,
of kidney disease in the United States. Am. J. Kidney Dis.
1990–2016: a systematic analysis for the Global Burden of
69 (3): A7–A8.
Disease Study 2016. Lancet 390 (10100): 1260–1344.
44 Naghavi, M., Abajobir, A.A., Abbafati, C. et al. (2017). 57 Jha, V., Wang, A.Y.-M., and Wang, H. (2012). The impact
Global, regional, and national age-sex specific mortality of CKD identification in large countries: the burden of
for 264 causes of death, 1980–2016: a systematic analysis illness. Nephrol. Dial. Transplant. 27 (suppl_3):
for the Global Burden of Disease Study 2016. Lancet 390 iii32–iii38.
(10100): 1151–1210. 58 Wang, H., Yang, L., Wang, F., and Zhang, L. (2017).
45 Bello, A.K., Levin, A., Tonelli, M. et al. (2017). Strategies and cost-effectiveness evaluation of persistent
Assessment of global kidney health care status. JAMA albuminuria screening among high-risk population of
317 (18): 1864–1881. chronic kidney disease. BMC Nephrol. 18 (1): 135.
46 Bello, A.K., Levin, A., Tonelli, M. et al. (2017). Global 59 Manns, B., Hemmelgarn, B., Tonelli, M. et al. (2010).
Kidney Health Atlas: A report by the International Population based screening for chronic kidney disease:
Society of Nephrology on the current state of cost effectiveness study. BMJ 341: c5869.
organization and structures for kidney care across the 60 Brown, W.W., Collins, A., Chen, S.-C. et al. (2003).
globe. Brussels, Belgium: International Society of. Identification of persons at high risk for kidney disease
Nephrology. via targeted screening: the NKF Kidney Early Evaluation
47 Moyer, V.A. (2012). Screening for chronic kidney disease: Program. Kidney Int. 63: S50–S55.
US Preventive Services Task Force recommendation 61 Vassalotti, J.A., Li, S., Chen, S.-C., and Collins, A.J.
statement. Ann. Intern. Med. 157 (8): 567–570. (2009). Screening populations at increased risk of CKD:
Reference 59
the Kidney Early Evaluation Program (KEEP) and the prevention and management of chronic kidney disease in
public health problem. Am. J. Kidney Dis. 53 (3): type 2 diabetes. Nephrology 15 (S1): S195–S203.
S107–S114. 69 Black C, Sharma P, Scotland G, McCullough K, McGurn
62 Boersma, C., Gansevoort, R.T., Pechlivanoglou, P. et al. D, Robertson L, et al. (2010). Early Referral Strategies for
(2010). Screen-and-treat strategies for albuminuria to Management of People with Markers of Renal Disease: A
prevent cardiovascular and renal disease: cost- Systematic Review of the Evidence of Clinical
effectiveness of nationwide and targeted interventions Effectiveness, Cost-Effectiveness and Economic Analysis.
based on analysis of cohort data from the Netherlands. Prepress Projects Ltd, Perth (www.prepress-projects.
Clin. Ther. 32 (6): 1103–1121. co.uk) on behalf of NETSCC (NIHR Evaluation, Trials
63 Hoerger, T.J., Wittenborn, J.S., Segel, J.E. et al. (2010). A and Studies Coordinating Centre (Great Britain).
health policy model of CKD: 2. The cost-effectiveness of (accessed 28 November 2018).
microalbuminuria screening. Am. J. Kidney Dis. 55 (3): 70 Hoerger, T.J., Wittenborn, J.S., Zhuo, X. et al. (2012).
463–473. Cost-effectiveness of screening for microalbuminuria
64 Howard, K., White, S., Salkeld, G. et al. (2010). among African Americans. J. Am. Soc. Nephrol. 23 (12):
Cost-effectiveness of screening and optimal management 2035–2041.
for diabetes, hypertension, and chronic kidney disease: a
71 Group SR (2015). A randomized trial of intensive versus
modeled analysis. Value Health 13 (2): 196–208.
standard blood-pressure control. N. Engl. J. Med. 373 (22):
65 Kessler, R., Keusch, G., Szucs, T.D. et al. (2012). Health
2103–2116.
economic modelling of the cost-effectiveness of
microalbuminuria screening in Switzerland. Swiss Med. 72 Tonelli, M., Isles, C., Curhan, G.C. et al. (2004). Effect
Wkly. 142: w13508. of pravastatin on cardiovascular events in people with
66 Palmer, A.J., Valentine, W.J., Chen, R. et al. (2008). A chronic kidney disease. Circulation 110 (12):
health economic analysis of screening and optimal 1557–1563.
treatment of nephropathy in patients with type 2 diabetes 73 Baigent, C., Landray, M.J., Reith, C. et al. (2011). The
and hypertension in the USA. Nephrol. Dial. Transplant. effects of lowering LDL cholesterol with simvastatin plus
23 (4): 1216–1223. ezetimibe in patients with chronic kidney disease (Study
67 Siegel, J.E., Krolewski, A.S., Warram, J.H., and Weinstein, of Heart and Renal Protection): a randomised placebo-
M.C. (1992). Cost-effectiveness of screening and early controlled trial. Lancet 377 (9784): 2181–2192.
treatment of nephropathy in patients with insulin- 74 Fitchett, D., Zinman, B., Wanner, C. et al. (2016). Heart
dependent diabetes mellitus. J. Am. Soc. Nephrol. 3 failure outcomes with empagliflozin in patients with type
(4 Suppl): S111–S119. 2 diabetes at high cardiovascular risk: results of the
68 Chadban, S., Howell, M., Twigg, S. et al. (2010). EMPA-REG OUTCOME® trial. Eur. Heart J. 37 (19):
Cost-effectiveness and socioeconomic implications of 1526–1534.
60
Risk Prediction in Chronic Kidney Disease

Joshua Samuels1, Kelsey Connelly 2, and Navdeep Tangri2
1
The McGovern Medical School, University of Texas HSC, Houston, TX, USA
2
Max Rady College of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada
I ntroduction accurately predict cardiovascular disease (CVD) in

patients with CKD may help identify patients whose pro-
Chronic kidney disease (CKD) is a major public health files warrant more intensive cardiovascular risk manage-
issue, with the global prevalence currently estimated to ment, such as antiplatelet therapy and lipid-lowering
range between 8.7% and 18.4% [1]. CKD is independently medications. The creation of such clinical decision tools is
associated with an increased risk for end-stage renal dis- based on the collection of clinical and demographic vari-
ease (ESRD), cardiovascular events, and all-cause mortal- ables and longitudinal follow-up. In addition to demo-
ity, but at the individual level these associations do not graphic risks, several biomarkers have been evaluated for
necessarily provide prognostic information for providers or their association with CKD disease severity and progres-
patients [2–4]. Accurate risk prediction integrated into sion. This chapter will review conventional biomarkers,
clinical practice, based on clinically relevant risk factors, novel biomarkers, and recently validated risk scores for
would allow physicians to determine patient-level risk of predicting CKD progression and outcomes (Tables 5.1
adverse events at point of care, and in turn stratify patients and 5.2).
based on their risk for adverse events [5].
Clinical methods of estimating risk provide assistance to Individual Predictors of CKD Progression
clinicians and patients in shared decision making, ulti-
mately ensuring that patient-centered care is provided Biomarkers are biological parameters used to differentiate
[6, 7]. While data-driven decision making is critical to between normal and abnormal physiological processes.
improve patient outcomes, there is no guarantee that For patients with CKD, biomarkers may reflect changes in
accurate estimation of risk translates into improved deci- renal function and thus may aid in predicting health out-
sion making nor to overall improved outcomes. With the comes, specifically progression of CKD or development of
effective use of these risk prediction models, most patients CVD [11]. There are conventional biomarkers that have
with stage 3 CKD (glomerular filtration rate [GFR] <60 ml/ been utilized for many years as well as newer, more novel
min/1.73 m2) can be stratified into the low-risk category markers to predict CKD progression.
and have the potential to be treated solely by their primary
care provider, allowing for observation and avoidance of
C
onventional Biomarkers
premature intensive treatment modalities. Conversely,
those at high risk can be referred to nephrology for urgent
Estimated Glomerular Filtration Rate
care [8]. On the same note, models predicting progression
toward ESRD can guide decisions about treatments aimed Perhaps the best indicator of future CKD progression is cur-
at slowing the progression of kidney disease, assist with rent burden of CKD. It has been appreciated for several dec-
dialysis planning and transplantation, and help triage ades that once glomerular filtration rate (GFR) has
highest-risk referrals to nephrologists [9]. Models that decreased to below a critical level, CKD tends to progress
Conventional Biomarker 61
Table 5.1 Prediction of chronic kidney disease [10].
Marker Recommendation Certainty of evidence Strength of recommendation
KDOQI Simple calculation of risk from Observational studies Strong

eGFR and proteinuria
KRFE Risk prediction using three, Validated in multiple Moderate/strong
four, or eight variables large cohorts
MCMC Risk prediction using nine Validated in multiple Moderate/strong
clinical variables large cohorts
Table 5.2 Prediction of cardiovascular disease.
Scoring Recommendation Certainty of evidence Strength of recommendation
eGFR Independently associated with Multiple observational studies Strong

risk of CVD
Proteinuria Independently associated with Multiple observational studies, several Strong
CVD risk and outcome RCT aimed a decreasing proteinuria
toward ESRD and is therefore the best indicator of global risk factor for CKD progression, as one study using dip-
kidney function [12–15]. This observation suggests that the stick urinalysis concluded that proteinuria was the most
loss of a critical number of nephrons provokes a vicious predictive biomarker of ESRD risk over a 10-year
cycle of further nephron loss. The natural history of CKD period [25]. In many glomerular diseases, changes in pro-
has been described with large variability depending on defi- teinuria have also been associated with changes in renal
nitions, the population being studied, as well as the health- prognosis, though these associations have not always
care environment [16, 17]. Using estimated GFR (eGFR) as been consistent. As with eGFR, the prognostic role of pro-
a marker of progression to ESRD, the African American teinuria/albuminuria is well established and it should be
Study of Kidney Disease and Hypertension (AASK) was considered as a potent biomarker of CKD progression. As
able to identify three distinct patterns of CKD progression: depicted in Figure 5.1, the 2012 Kidney Disease Improving
(i) stable/increasing eGFR (less than 2 ml/min/1.73 m2 Global Outcome (KDIGO) color-coded prognosis chart
decrease per year), (ii) rapid decrease in eGFR (a loss of includes only eGFR and degree of albuminuria in risk
more than 4 ml/min/1.73 m2 per year), and (iii) alternating assessment [26, 27].
combinations of the previously mentioned categories. The
most common pattern found in patients with CKD was a
Hypertension
stable/increasing eGFR, with 58% of patients falling under
this category [18]. The fact that only a minority of patients Hypertension is common in both adults and children with
will develop ESRD further supports a need for risk stratifi- CKD [28, 29]. Among patients with CKD, those with higher
cation development in order to appropriately allocate treat- blood pressure have significantly faster GFR decline and
ment decisions. Almost all models that attempt to predict the presence of hypertension has consistently been associ-
renal functional decline include an assessment of eGFR as ated with rapid progression [17, 29–32]. This finding has
a critical factor. led to recommendations regarding more intensive manage-
ment of hypertension in patients with CKD [33]. While
blood pressure appears to impart a deleterious effect on
Albuminuria/Proteinuria
CKD progression in a continuous fashion without a spe-
Proteinuria is another index of kidney disease severity in cific “threshold” value, for convenience most prediction
CKD patients, as it often results from glomerular models that include blood pressure as a factor simply
injury [11]. In several longitudinal studies of patients dichotomize blood pressure to hypertensive or not.
with CKD, higher baseline proteinuria was indicative of a
rapid decline in GFR [16–22] In a similar manner, base-
Serum Bicarbonate/Metabolic Acidosis
line urinary albumin/creatinine ratio (ACR) in patients
with diabetic nephropathy was found to be a strong indi- Metabolic acidosis is another common consequence of
cator of ESRD [23, 24]. Proteinuria itself may pose as a CKD and often causes impaired renal acid clearance [34].
62 Risk Prediction in Chronic Kidney Disease
Persistent albuminuria categories

Description and range
A1 A2 A3
Normal to
Guide to Frequency of Monitoring Moderately Severely
midly
(number of times per year) by increased increased
increased
GFR and Albuminuria Category
<30 mg/g 30–300 mg/g >300 mg/g
<3 mg/mmol 3–30 mg/mmol >30 mg/mmol
G1 Normal or high ≥90 1 if CKD 1 2

GFR categories (ml/min1.73 m2)
G2 Midly decreased 60–89 1 if CKD 1 2
Mildly to moderately
G3a decreased 45–59 1 2 3
Moderately to
G3b severely decreased 30–44 2 3 3
G4 Severely decreased 15–29 3 3 4+
G5 Kidney failure <15 4+ 4+ 4+
Figure 5.1 Prognosis of CKD dependent on eGFR and degree of albuminuria, with recommended frequency of follow-up. Green,
low risk (if no other markers of kidney disease, no CKD); yellow, moderately increased risk; orange, high risk; red, very high risk.
Source: KDIGO [26]. © 2013 KDIGO
The majority of relevant studies have indicated that lower Uric Acid
serum bicarbonate levels, even within normal range, are
Another biomarker which has been used to predict CKD
independently associated with CKD progression [7, 35–40].
progression is serum uric acid level [53–56]. An evalua-
Bicarbonate supplementation as a treatment for metabolic
tion of children and adolescents in the Chronic Kidney
acidosis, either as oral sodium bicarbonate or a diet rich in
Disease in Children Study (CKiD) cohort revealed that
fruits and vegetables, has been shown to decrease the rate of
hyperuricemia is associated with a more rapid renal
GFR decline in multiple small single-center studies [38–40].
decline [56]. Compared to subjects with the lowest uric
Recent studies have shown that metabolic acidosis has addi-
acid levels, those with intermediate and frankly elevated
tional end organ effects including impacts on cognition and
uric acid experienced a more rapid decline in eGFR. Uric
endothelial function; new nonalkali treatments for meta-
acid levels remained significantly associated with pro-
bolic acidosis are currently under development [35, 41–44].
gression even when controlling for other risk factors
such as hypertension, proteinuria, initial GFR, and glo-
Serum Phosphate and FGF23
merular etiology, all of which were also significant in
Various studies have indicated that elevated serum phos- multivariate analysis. A 2018 meta-analysis of adults
phate levels are independently associated with renal injury with CKD analyzed several randomized controlled trials
in CKD [7, 45, 46]. However, this has been contradicted by (RCTs) done to evaluate the effects of uric acid lowering
a large study including 10 672 CKD patients, which found therapy on the disease progression [53]. The analysis
no association between higher serum phosphate and risk included a total of 832 subjects with CKD randomized to
of progression toward ESRD [47]. Though not a conven- uric acid lowering treatments versus placebo. Active
tional biomarker in clinical practice, phosphatonin fibro- treatment conferred a 3.88 ml/min/1.73m2 (1.26–6.49 ml/
blast growth factor 23 (FGF23), a hormone involved in min/1.73m2) mean difference in eGFR. Additionally, risk
calcium and phosphate regulation, has also been suggested of worsened kidney function or ESRD or death was sig-
as an independent predictor of CKD progression [48, 49]. nificantly reduced with uricosuric pharmacotherapy.
Elevated levels of FGF23 in CKD patients were indepen- These data suggest that elevated uric acid represents a
dently associated with progression to ESRD, as well as car- risk prediction tool and potentially a therapeutic target to
diovascular events [50–52]. slow CKD progression.
CKD Risk Score 63
N
ovel Biomarkers of CKD and associated mortality [67]. To assess this asso-
ciation, one study looked at 343 AKI patients with a pread-
Recently, a number of novel biomarkers have emerged as mission eGFR greater than 45 ml/min/1.73 m2 who
potentially new clinically-relevant risk prediction tools, survived for at least 30 days after discharge without dialy-
but more research is needed to determine their full impact sis. The authors concluded that those who required dialysis
on CKD progression. For example, although levels of neu- for their AKI during hospital admission had a 28-fold
trophil gelatinase-associated lipocalin (NGAL) have been increase in the risk of developing CKD stage 4 or 5, and had
found to be overexpressed in patients with AKI [57] and more than double the risk of mortality when compared to
are highest in CKD patients [58], the Chronic Renal those who did not need dialysis [68].
Insufficiency Cohort (CRIC) study found that while uri- Due to increasing evidence indicating the high likeli-
nary NGAL levels were independently associated with hood of CKD development in post-AKI patients, there are
ESRD, they did not significantly improve prediction of ongoing efforts to develop a risk prediction model to iden-
CKD progression over predictions using only eGFR tify which of these patients are at the highest risk. Using a
decline and proteinuria [59]. The glycoprotein kidney cohort of 5351 patients diagnosed with an AKI, risk pre-
injury molecule-1 (KIM-1) is usually only present in the diction models were developed that utilized variables of
urine of patients with AKI or CKD [60]. One study found age, serum albumin, diabetes, baseline eGFR, mean
that serum KIM-1 levels increased with CKD progression serum creatinine, and severity of AKI as assessed by the
and were independently associated with ESRD [61]. RIFLE score (risk, injury, failure, loss, and end-stage kid-
Through the Multi-Ethnic Study of Atherosclerosis, it ney disease) or need for dialysis, non-African American
was also determined that doubling of urinary KIM-1 lev- ethnicity, and time spent at risk. External validation was
els was associated with an eGFR decrease greater than performed using 11 589 patients admitted for pneumonia
3 ml/min/1.73 m2 per year [62]. Elevated levels of soluble or myocardial infarction (MI), and yielded good predic-
urokinase-type plasminogen activator receptor (suPAR) tion accuracy (area under the receiver operating charac-
have also been observed in patients with renal disease, teristics [AUROC] value = 0.81–0.82) [69]. Further
specifically glomerulosclerosis [11]. One study investigat- investigations are required that specifically include study
ing suPAR determined that higher baseline levels were populations with a higher proportion of women before
independently associated with CKD. Despite this, 30% of clinical implementation.
patients with a normal eGFR (>90 ml/min/1.73 m2) were
also found to have increased baseline suPAR levels, indi-
cating that elevation may be a result of other biological CKD Risk Scores
processes, such as inflammation [63]. Lastly, the kidney-
specific glycoprotein uromoduliun (UMOD) is known to Clinical prediction rules or models (sometimes called clini-
have gene variants associated with eGFR decline and is cal decision tools) are tools that are created to assist clini-
associated with the development of CKD. One genome- cians in patient-level decisions [70–72]. Designed by
wide association study using data from 3203 Icelandic multivariate analysis, these scoring models complement
CKD patients discovered a variant adjacent to the UMOD clinical decision making by suggesting diagnostic or thera-
gene on chromosome 16p12 had an association with CKD peutic actions or providing prognostication of some clini-
and increasing creatinine levels [64, 65]. However, cal outcome. The goal of validated models is to improve
another study found that among 879 individuals of the individual patient outcomes and the cost-effectiveness of
Heart and Soul Study, UMOD single-nucleotide polymor- care [71]. Development of a clinical prediction model
phism variants influenced UMOD levels, but not CKD involves multiple steps, including (i) identification of
development [66]. important predictor variable (listed above in the case of
CKD), (ii) assigning relative weights to each factor to create
a combined score, (iii) assessing the score’s predictive abil-
he Role of AKI in Predicting
T ity, including calibration and discrimination, and then (iv)
Future CKD validation of the model using both internal and external
data.
Historically, clinicians assumed that patients who com- Calibration is the agreement between the predicted prob-
pletely recovered from acute kidney injury (AKI) would ability of the outcome of interest and the observed fre-
regain baseline renal function. However, recent studies quency. It is usually plotted graphically as the observed
have contradicted this by demonstrating that a previous outcome frequency against the mean predicted outcome
history of AKI is indeed associated with an increased risk probability (or risk). Goodness of fit for calibration can
then be statistically tested. Discrimination refers to a mod- Table 5.3 Hazard ratios and goodness of fit for sequential
el’s ability to distinguish individuals who have the model models in the development data set for three, four, and eight-
variable kidney failure risk equations (KFREs).
outcome from those who remain event free, usually
reported as the area under the receiver operating character-
Three- Four- Eight-
istic (ROC) curve (logistic regression c-statistic.) Models
variable variable variable
usually perform well within the data sample they are devel- Variable model model model
oped, as is often demonstrated using internal validation.
External validation, therefore, is advocated to ascertain Baseline GFR, per 0.54 0.57 0.61
utility of the model in clinical practice. Several techniques 5 ml/min/1.73 m2
of external validation are available [70, 73]. For a more Age, per 10 years 0.75 0.8 0.82
comprehensive review, Moons and colleagues provide an Male sex 1.46 1.26 1.16
in-depth analysis of clinical decision models in a two- Log spot urine ACR a
1.60 1.42
article series [70, 71]. Serum albumin, per 0.84
Investigation of risk factors/biomarkers used in predict- 0.5 g/dl
ing the development and/or progression of CKD has Serum phosphate, per 1.27
resulted in the discovery of a relatively small number of 1.0 mg/dl
risk factors common to different forms of CKD. These risk Serum bicarbonate, 0.92
factors are currently routinely being used to guide routine per 1.0 mEq/l
clinical practice, for example the revised KDIGO risk clas- Serum calcium, per 0.81
sification system uses GFR and albuminuria to categorize mg/dl
patients into CKD stages [26, 74]. This has led to the devel- c-statistics 0.89 0.91 0.92
opment of renal risk scores, which have the potential to be Akaike Information 4834 4520 4432
used in predicting the development and progression of Criterionb
CKD. Progress has already been made in developing a P value <0.001 <0.001 <0.001
variety of risk scores to facilitate more accurate risk pre- a
Hazard ratio for ACR represents a 1.0 higher ACR on the natural
diction, applying to those with and without CKD. We will
log scale. For the average patient with 20 mg/g of albuminuria, this
now review several models that have been developed to represents an increase to 55 mg/g.
predict risk. b
Null values for c-statistic and Akaike Information Criterion are 0.50
and 5569, respectively. Higher values for c-statistic and lower values
for Akaike Information Criterion indicate better models.
Source: Tangri et al. [7]. © 2011 American Medical Association.
The Kidney Failure Risk Equations
To predict the progression toward ESRD in patients with

also found that the simplified four-variable equation pro-
CKD, Tangri and colleagues developed the Kidney Failure
duced similar discrimination compared to the eight-
Risk Equations (KFREs) in 2011, using data from a cohort
variable equation (c-statistics were 0.89 vs. 0.90 at 2 years
of 3449 patients with late stage CKD [7]. Variables used in
and 0.86 vs. 0.88 at 5 years, respectively) [75].
the eight-variable equation were age, sex, eGFR, albumi-
nuria, serum calcium, serum phosphate, serum bicarbo-
nate, and serum albumin. Simpler three-variable (age, sex,
and eGFR) and four-variable (age, sex, eGFR, and albumi- Newer Prediction Models
nuria) models were also developed. While both of these
simplified models performed well (Table 5.3), the eight- Several newer prediction models have also been developed
variable model performed best with respect to discrimina- and some are now available online. In 2018 Grams et al.
tion, calibration, and reclassification in an external developed models using data from over 250 000 individu-
validation cohort. als across 30 countries who participated in the interna-
To assess the global functionality of the KFREs, a valida- tional CKD Prognosis Consortium with eGFR < 30 ml/
tion study of the equations was performed using 35 cohorts min/1.73m2 [76]. The Markov Chain Monte Carlo
consisting of 720 000 patients with late-stage CKD across (MCMC) model predicts 2-and 4-year risk of renal replace-
four continents. After a 4-year follow-up, the KFREs ment therapy, cardiovascular (CV) event, or any cause of
achieved accurate estimation of 23 829 cases of ESRD death using nine common predictor variables: age, sex,
(c-statistic 0.90, 95% confidence interval [CI] 0.89–0.92 at ethnicity, eGFR, systolic blood pressure, ACR, and any
2 years; c-statistic 0.88; 95% CI 0.86–0.90 at 5 years). It was history of smoking, diabetes, or prior CVD (available at
Newer Prediction Model 65
http://ckdpcrisk.org/lowgfrevents). This new model has Transition from Primary Care to Nephrology

significant overlap with KFRE (age, sex, eGFR, and albu-
minuria), though the additional variables provide an Patients with CKD will often develop hypertension and ane-
improved predictive ability in validation trials (within mia, which are often managed under the supervision of pri-
cohort R2 ranged from 0.89 to 0.97). A comparison of the mary care physicians [78]. However, these patients may be at
predicted versus observed outcomes in various trials using higher risk for developing ESRD, resulting in a need for
both KFRE and the Grams MCMC model is shown in referral to nephrology in order to delay CKD progression, in
Figure 5.2 [76]. turn preventing dialysis and reducing mortality risk [64, 79].
Successful clinical implementation of risk models Implementing risk equations in the primary care setting
requires both a validated risk prediction model and deter- may guide the appropriate timing of nephrology referrals.
mination of absolute risk thresholds. Using these to guide The 2012 KDIGO guidelines include recommendations for
therapy decisions will allow for a cost-effective treatment referral to specialty care, though the utility of this recom-
plan. Figure 5.3 demonstrates both the eGFR criteria (top) mendation has not been subject to critical review [26].
and the suggested KFRE risk-based thresholds (bottom) Some have argued that if patients have a moderately
that prompt transition between different levels of care [77]. reduced eGFR, they should be monitored at 3-or 6-month
1 AASK 1
CanPREDDICT
CRIB
CRIC
.8 CRISIS .8
GCKD
Geisinger
Observed risk
Observed risk
Gonryo
.6 Hongkong-CKD .6
MASTERPLAN
MDRD
Nanjing-CKD
NephroTest
.4 NRHP-URU .4
NZDCS
PSP-CKD
PSPA
.2 SCREAM .2
SMART
SRR-CKD
Sunnybrook
WestScot-CKD
0 0
0 .2 .4 .6 .8 1 0 .2 .4 .6 .8 1
Predicted risk Predicted risk
(a) (b)
Figure 5.2 Calibration of the predicted versus observed 2-year risk of ESRD for (a) the kidney failure risk equation (KFRE) and (b) the
MCMC model. Points represent groups of participants with <20% predicted KFRE risk, 20–40% predicted KFRE risk, and >40%
predicted KFRE risk. Source: From Grams et al. 2018 [76].
eGFR 30-60 eGFR <30 eGFR <20
eGFR-based Criteria
Transition from Primary Transition from Nephrology

Access and Transplant
Care to Nephrology Care to Interprofressional
Planning
Care Care
eGFR 90 60 30 20 <10 Kidney

Failure
KFRE ≥3% KFRE ≥10% KFRE ≥40%

Risk-based Criteria
5 years 2 years 2 years
Figure 5.3 Traditional paradigm whereby eGFR values are used to determine intensity of care and inclusion of specialists in
patients with CKD. Source: From Tangri N et al. 2017 [77].
intervals, whereas those who are at a lower risk can be healthcare system [81]. In the United States, the opposite
followed on an annual basis [80]. The KDIGO recom- may be happening with unacceptably late nephrology
mended frequency of follow-up is depicted in Figure 5.4. referral.
Dialysis Access and Modality Planning

Interdisciplinary Care
It is currently the standard of care to recommend dialysis
Interdisciplinary nephrology-led teams, which include modality planning to patients with an eGFR less than
nursing, pharmacy, and dietetic professionals, may 30 ml/min/1.73m2, despite the fact that the majority of
improve the course of treatment, in turn ensuring timely these patients will never progress to ESRD [82]. The weigh-
dialysis initiation and lowering mortality risk. Ideally ing of risks and benefits when it comes to modality plan-
risk stratification would ensure these teams are not over- ning (hemodialysis vs. peritoneal dialysis) is paramount.
burdened with low-risk CKD patients, thereby prevent- The process of dialysis planning, including the creation of
ing inappropriate allocation of limited healthcare an arteriovenous fistula (AVF), the standard method of
resources [18]. Following issues with excess demand for access for patients on hemodialysis, however crucial, may
interdisciplinary care in Ontario, Canada, the Ontario create anxiety for patients and caregivers. Furthermore,
Renal Network adopted a risk-based cutoff for clinical older adults are at a higher risk of AVF failure and subse-
resource eligibility. The cutoff for interdisciplinary care quent adverse events, including heart failure and steal syn-
was a 2-year risk of ESRD exceeding 10% or an eGFR less drome. Additionally, older adults who are at a low risk of
than 15 ml/min/1.73 m2. As a result, the number of progressing to kidney failure often have competing mor-
patients eligible was drastically lowered, as the previous bidities, further discouraging AVF formation [83]. By
threshold was solely based on an eGFR of less than 30 ml/ implementing risk prediction models such as the KFRE,
min/1.73 m2. Due to the implementation of this criterion, AVF insertion can be guided properly to avert some of
a proportion of patients with an eGFR between 30 and these risks while providing optimal treatment to low-risk
45 ml/min/1.73 m2 were also found to be eligible as they CKD patients. The efficacy of implementing KFREs was
had a high risk of ESRD development. As a result of this simulated in 2014 using the Sunnybrook hospital cohort.
change in cutoff, $6 million was saved annually for the The study concluded that a cutoff of 20% annual risk of
Persistent albuminuria categories

A1 A2 A3
Normal to
Moderately Severely
midly
increased increased
increased
<30 mg/g 30–300 mg/g >300 mg/g

<3 mg/mmol 3–30 mg/mmol >30 mg/mmol
G1 Normal or high ≥90 Monitor Refer*

GFR categories (ml/min1.73 m2)
G2 Midly decreased 60–89 Monitor Refer*
Mildly to moderately
G3a decreased 45–59 Monitor Monitor Refer
Moderately to
G3b severely decreased 30–44 Monitor Monitor Refer
G4 Severely decreased 15–29 Refer* Refer* Refer
G5 Kidney failure <15 Refer Refer Refer
Figure 5.4 KDIGO recommendations about referral to nephrology care in CKD. referral decision making by GFR and albuminuria.
aReferring clinicians may wish to discuss with their nephrology service depending on local arrangements regarding monitoring or
referring. Source: KDIGO [26] © 2013 KDIGO.
Reference 67
ESRD calculated by the KFREs was more effective than outcomes using data from 440 526 participants in the UK
eGFR-based cutoffs in predicting the need for AVF inser- Biobank [96]. Their simple prediction model, created in
tion and dialysis [84]. 2019, uses different eGFR to predict CVD. The most suc-
cessful model used cystatin C-based eGFR.
Predicting Cardiovascular Disease in CKD Patients
CKD substantially increases the risk of CVD in patients, P
roteinuria/Albuminuria
but it is also itself a consequence of CVD. One study found and Cardiovascular Risk
that amongst hospitalized patients with congestive heart
failure (CHF), the prevalence of late-stage CKD was 60.4% Proteinuria/albuminuria predicts not only renal progres-
and it was 51.7% for those with MI [85]. When accompany- sion but is an ominous marker of other outcomes as well,
ing heart disease, CKD also increased the risk of mortality most notably cardiovascular events. Multiple analyses
and ESRD in patients [86]. This may be a result of the have since established that ACR predicts all-cause mortal-
shared risk factors of CVD and CKD, which include ath- ity and cardiovascular death independent from baseline
erosclerosis, obesity, metabolic syndrome, hypertension, CKD and other covariates [97–101]. Early studies revealed
diabetes mellitus, dyslipidemia, and smoking. Renal ath- a link between degree of microalbuminuria and poor out-
erosclerosis has been found in 39% of patients undergoing comes in patients with type 2 diabetes [102]. Analysis of
elective coronary angiography [87]. Framingham participants revealed that proteinuria pre-
dicted poor outcome in the general population as
well [103]. Hillege analyzed a Dutch cohort of 40 548
eGFR and Cardiovascular Risk adults and found a dose–response relationship between
urinary ACR and both CV and non-CV mortality [98].
In the United States, individuals with CKD experience a
10-fold increased risk of vascular disease compared to
those without kidney disease [88, 89]. This finding may be F
uture Considerations
due to significant overlap of risk factors between CKD and
CVD [90–92]. The excess CV mortality appears to begin Although some disagreement on the validity of various risk
early in the advancing CKD process and increases with factors discussed in this chapter still remains, a relatively
worsened eGFR [76, 93]. Current CVD risk prediction small group of risk factors appear to be common to most cur-
models have proven to be unreliable when patients have rent studies. As a result, much progress has been made in
concurrent CKD [94]. Bansal et al. developed a prediction developing risk scores based on these select variables in their
tool that has shown promise in assessing 5-year mortality ability to aid in predicting the progression of CKD, and with
risk in patients with CKD but not ESRD (c-statistic 0.72, the development of risk prediction models. Future studies
95% CI 0.68–0.74). Their model incorporates age, sex, race, should focus on a more critical review of the aforementioned
eGFR, urine ACR, smoking status, diabetes mellitus, and biomarkers and genetic factors utilized in risk score develop-
history of CHF or stroke. While this model has been exter- ment, although the costs of measuring such markers is likely
nally validated in 789 participants (c-statistic 0.69, 95% CI to be much higher than simple risk factors used at the pre-
0.64–0.74), patients with CKD stages 4–5 were underrepre- sent time. Accurate risk scores can align risk and resources,
sented in the validation study and therefore further investi- while allowing for personalized care for patients with CKD.
gation of this demographic is required prior to successful Therefore, it is imperative to implement these risk scores
clinical implementation [95]. Researchers from Glasgow into routine clinical practice, in turn helping patients and
evaluated the effects of eGFR and proteinuria on CVD clinicians navigate the transition from CKD to dialysis.
R
eferences
Hill, N.R., Fatoba, S.T., Oke, J.L. et al. (2016). Global

1 3 Levey, A.S., Atkins, R., Coresh, J. et al. (2007). Chronic
prevalence of chronic kidney disease – a systematic review kidney disease as a global public health problem:
and meta-analysis. PLoS One 11 (7): e0158765. approaches and initiatives – a position statement from
2 Harrell, F.E. Jr., Lee, K.L., and Mark, D.B. (1996). Kidney Disease Improving Global Outcomes. Kidney Int.
Multivariable prognostic models: issues in developing 72 (3): 247–259.
models, evaluating assumptions and adequacy, and 4 Taal, M.W. and Brenner, B.M. (2008). Renal risk scores:
measuring and reducing errors. Stat. Med. 15 (4): 361–387. progress and prospects. Kidney Int. 73 (11): 1216–1219.
5 Tangri, N., Kitsios, G.D., Inker, L.A. et al. (2013). Risk adolescents: the chronic kidney disease in children
prediction models for patients with chronic kidney (CKiD) cohort. Am. J. Kidney Dis. 65 (6): 878–888.
disease. Ann. Intern. Med. 158 (8): 596–603. 20 The GISEN Group (Gruppo Italiano di Studi
6 Steyerberg, E.W. (2009). Clinical Prediction Models: Epidemiologici in Nefrologia) (1997). Randomised
A Practical Approach to Development, Validation placebo-controlled trial of effect of ramipril on decline in
and Updating. Rotterdam: Springer. glomerular filtration rate and risk of terminal renal
7 Tangri, N., Stevens, L.A., Griffith, J. et al. (2011). failure in proteinuric, non-diabetic nephropathy. Lancet
A predictive model for progression of chronic 349 (9069): 1857–1863.
kidney disease to kidney failure. JAMA 305 (15): 21 Lea, J., Greene, T., Hebert, L. et al. (2005). The
1553–1559. relationship between magnitude of proteinuria reduction
8 O’Hare, A.M., Choi, A.I., Bertenthal, D. et al. (2007). Age and risk of end-stage renal disease: results of the African
affects outcomes in chronic kidney disease. J. Am. Soc. American study of kidney disease and hypertension.
Nephrol. 18 (10): 2758–2765. Arch. Intern. Med. 165 (8): 947–953.
9 Oliver, M.J., Quinn, R.R., Garg, A.X. et al. (2012). 22 Peterson, J.C., Adler, S., Burkart, J.M. et al. (1995). Blood
Likelihood of starting dialysis after incident fistula pressure control, proteinuria, and the progression of
creation. Clin. J. Am. Soc. Nephrol. 7 (3): 466–471. renal disease. The modification of diet in renal disease
study. Ann. Intern. Med. 123 (10): 754–762.
10 Schünemann, H.J., Mustafa, R., Brozek, J. et al. (2016).
23 Atkins, R.C., Briganti, E.M., Lewis, J.B. et al. (2005).
GRADE Guidelines: 16. GRADE evidence to decision
Proteinuria reduction and progression to renal failure in
frameworks for tests in clinical practice and public
patients with type 2 diabetes mellitus and overt
health. J. Clin. Epidemiol. 76: 89–98.
nephropathy. Am. J. Kidney Dis. 45 (2): 281–287.
11 Collister, D., Ferguson, T., Komenda, P., and Tangri, N.
24 Keane, W.F., Zhang, Z., Lyle, P.A. et al. (2006). Risk scores
(2016). The patterns, risk factors, and prediction of
for predicting outcomes in patients with type 2 diabetes
progression in chronic kidney disease: a narrative review.
and nephropathy: the RENAAL study. Clin. J. Am. Soc.
Semin. Nephrol. 36 (4): 273–282.
Nephrol. 1 (4): 761–767.
12 Smith, H. (1951). Comparative physiology of the kidney.
25 Iseki, K., Iseki, C., Ikemiya, Y., and Fukiyama, K. (1996).
In: The Kidney: Structure and Function in Health and
Risk of developing end-stage renal disease in a cohort of
Disease (ed. H. Smith), 520–574. New York: Oxford
mass screening. Kidney Int. 49 (3): 800–805.
University Press.
26 KDIGO (2013). 2012 clinical practice guideline for the
13 Smith, H. (1957). Measurement of the rate of glomerular
evaluation and management of chronic kidney disease:
filtration. In: Principles of Renal Physiology (ed. H. Smith),
summary of recommendation statements. Kidney Int.
25–35. New York: Oxford University Press.
Suppl. 3 (1): 5–14.
14 Wesson, L. (1969). Renal hemodynamics in physiological 27 KDIGO (2013). 2012 clinical practice guideline for the
states. In: Physiology of the Human Kidney (ed. L. evaluation and management of chronic kidney disease;
Wesson), 96–108. New York: Grune & Stratton. Chapter 2: Definition, identification, and prediction of
15 Wesson, L. (1969). Renal hemodynamics in pathological CKD progression. Kidney Int. Suppl. 3 (1): 63–72.
states. In: Physiology of the Human Kidney (ed. L. 28 Samuels, J., Ng, D., Flynn, J.T. et al. (2012). Ambulatory
Wesson), 109–154. New York: Grune & Stratton. blood pressure patterns in children with chronic kidney
16 Hunsicker, L.G., Adler, S., Caggiula, A. et al. (1997). disease. Hypertension 60 (1): 43–50.
Predictors of the progression of renal disease in the 29 Griffin, K.A. (2017). Hypertensive kidney injury and the
Modification of Diet in Renal Disease Study. Kidney Int. progression of chronic kidney disease. Hypertension 70
51 (6): 1908–1919. (4): 687–694.
17 Atkinson, M.A., Ng, D.K., Warady, B.A. et al. (2020). The 30 Griffin, K.A., Polichnowski, A., Litbarg, N. et al. (2014).
CKiD study: overview and summary of findings related to Critical blood pressure threshold dependence of
kidney disease progression. Pediatr. Nephrol. hypertensive injury and repair in a malignant
36(3):527–538. nephrosclerosis model. Hypertension 64 (4): 801–807.
18 Li, L., Astor, B.C., Lewis, J. et al. (2012). Longitudinal 31 Bidani, A.K. and Griffin, K.A. (2004). Pathophysiology
progression trajectory of GFR among patients with CKD. of hypertensive renal damage. Hypertension 44 (5):
Am. J. Kidney Dis. 59 (4): 504–512. 595–601.
19 Warady, B.A., Abraham, A.G., Schwartz, G.J. et al. (2015). 32 (2004). K/DOQI clinical practice guidelines on
Predictors of rapid progression of glomerular and hypertension and antihypertensive agents in chronic
nonglomerular kidney disease in children and kidney disease. Am. J. Kidney Dis. 43: 11–13.
Reference 69
33 Appel, L.J., Wright, J.T. Jr., Greene, T. et al. (2010). the Chronic Renal Impairment in Birmingham (CRIB)
Intensive blood-pressure control in hypertensive chronic prospective cohort study. Am. J. Kidney Dis. 56 (6):
kidney disease. N. Engl. J. Med. 363 (10): 918–929. 1082–1094.
34 Loniewski, I. and Wesson, D.E. (2014). Bicarbonate 46 Levin, A., Djurdjev, O., Beaulieu, M., and Er, L. (2008).
therapy for prevention of chronic kidney disease Variability and risk factors for kidney disease progression
progression. Kidney Int. 85 (3): 529–535. and death following attainment of stage 4 CKD in a
35 Dobre, M., Yang, W., Chen, J. et al. (2013). Association of referred cohort. Am. J. Kidney Dis. 52 (4): 661–671.
serum bicarbonate with risk of renal and cardiovascular 47 Mehrotra, R., Peralta, C.A., Chen, S.C. et al. (2013). No
outcomes in CKD: a report from the Chronic Renal independent association of serum phosphorus with risk
Insufficiency Cohort (CRIC) study. Am. J. Kidney Dis. 62 for death or progression to end-stage renal disease in a
(4): 670–678. large screen for chronic kidney disease. Kidney Int. 84 (5):
36 Raphael, K.L., Wei, G., Baird, B.C. et al. (2011). Higher 989–997.
serum bicarbonate levels within the normal range are 48 Fliser, D., Kollerits, B., Neyer, U. et al. (2007). Fibroblast
associated with better survival and renal outcomes in growth factor 23 (FGF23) predicts progression of chronic
African Americans. Kidney Int. 79 (3): 356–362. kidney disease: the Mild to Moderate Kidney Disease
37 Shah, S.N., Abramowitz, M., Hostetter, T.H., and (MMKD) Study. J. Am. Soc. Nephrol. 18 (9): 2600–2608.
Melamed, M.L. (2009). Serum bicarbonate levels and the 49 Titan, S.M., Zatz, R., Graciolli, F.G. et al. (2011). FGF-23
progression of kidney disease: a cohort study. Am. J. as a predictor of renal outcome in diabetic nephropathy.
Kidney Dis. 54 (2): 270–277. Clin. J. Am. Soc. Nephrol. 6 (2): 241–247.
38 de Brito-Ashurst, I., Varagunam, M., Raftery, M.J., and 50 Isakova, T., Xie, H., Yang, W. et al. (2011). Fibroblast
Yaqoob, M.M. (2009). Bicarbonate supplementation slows growth factor 23 and risks of mortality and end-stage
progression of CKD and improves nutritional status. J. renal disease in patients with chronic kidney disease.
Am. Soc. Nephrol. 20 (9): 2075–2084. JAMA 305 (23): 2432–2439.
39 Goraya, N., Simoni, J., Jo, C.H., and Wesson, D.E. (2014). 51 Marthi, A., Donovan, K., Haynes, R. et al. (2018).
Treatment of metabolic acidosis in patients with stage 3 Fibroblast growth factor-23 and risks of cardiovascular
chronic kidney disease with fruits and vegetables or oral and noncardiovascular diseases: a meta-analysis. J. Am.
bicarbonate reduces urine angiotensinogen and preserves Soc. Nephrol. 29 (7): 2015–2027.
glomerular filtration rate. Kidney Int. 86 (5): 1031–1038. 52 Rebholz, C.M., Grams, M.E., Coresh, J. et al. (2015). Serum
40 Mahajan, A., Simoni, J., Sheather, S.J. et al. (2010). Daily fibroblast growth factor-23 is associated with incident
oral sodium bicarbonate preserves glomerular filtration kidney disease. J. Am. Soc. Nephrol. 26 (1): 192–200.
rate by slowing its decline in early hypertensive 53 Liu, X., Zhai, T., Ma, R. et al. (2018). Effects of uric
nephropathy. Kidney Int. 78 (3): 303–309. acid-lowering therapy on the progression of chronic
41 Afsar, B. and Elsurer, R. (2015). Association between kidney disease: a systematic review and meta-analysis.
serum bicarbonate and pH with depression, cognition Ren. Fail. 40 (1): 289–297.
and sleep quality in hemodialysis patients. Ren. Fail. 37 54 Madero, M., Sarnak, M.J., Wang, X. et al. (2009). Uric acid
(6): 957–960. and long-term outcomes in CKD. Am. J. Kidney Dis. 53
42 Bushinsky, D.A., Hostetter, T., Klaerner, G. et al. (2018). (5): 796–803.
Randomized, controlled trial of TRC101 to increase 55 Kim, K., Go, S., Son, H.E. et al. (2020). Association
serum bicarbonate in patients with CKD. Clin. J. Am. Soc. between serum uric acid level and ESRD or death in a
Nephrol. 13 (1): 26–35. Korean population. J. Korean Med. Sci. 35 (28): e254.
43 Wesson, D.E., Mathur, V., Tangri, N. et al. (2019). 56 Rodenbach, K.E., Schneider, M.F., Furth, S.L. et al. (2015).
Long-term safety and efficacy of veverimer in patients Hyperuricemia and progression of CKD in children and
with metabolic acidosis in chronic kidney disease: a adolescents: The Chronic Kidney Disease in Children
multicentre, randomised, blinded, placebo-controlled, (CKiD) Cohort Study. Am. J. Kidney Dis. 66 (6): 984–992.
40-week extension. Lancet 394 (10196): 396–406. 57 Haase, M., Bellomo, R., and Devarajan, P., and Group
44 Wesson, D.E., Mathur, V., Tangri, N. et al. (2019). NM-aI (2009). Accuracy of neutrophil gelatinase-
Veverimer versus placebo in patients with metabolic associated lipocalin (NGAL) in diagnosis and prognosis
acidosis associated with chronic kidney disease: a in acute kidney injury: a systematic review and meta-
multicentre, randomised, double-blind, controlled, phase analysis. Am. J. Kidney Dis. 54 (6): 1012–1024.
3 trial. Lancet 393 (10179): 1417–1427. 58 Bhavsar, N.A., Kottgen, A., Coresh, J., and Astor, B.C.
45 Landray, M.J., Emberson, J.R., Blackwell, L. et al. (2010). (2012). Neutrophil gelatinase-associated lipocalin
Prediction of ESRD and death among people with CKD: (NGAL) and kidney injury molecule 1 (KIM-1) as
predictors of incident CKD stage 3: the Atherosclerosis 72 Laupacis, A., Sekar, N., and Stiell, I.G. (1997). Clinical
Risk in Communities (ARIC) Study. Am. J. Kidney Dis. 60 prediction rules. A review and suggested modifications of
(2): 233–240. methodological standards. JAMA 277 (6): 488–494.
59 Liu, K.D., Yang, W., Anderson, A.H. et al. (2013). Urine 73 Toll, D.B., Janssen, K.J., Vergouwe, Y., and Moons, K.G.
neutrophil gelatinase-associated lipocalin levels do not (2008). Validation, updating and impact of clinical
improve risk prediction of progressive chronic kidney prediction rules: a review. J. Clin. Epidemiol. 61 (11):
disease. Kidney Int. 83 (5): 909–914. 1085–1094.
60 Haase, M. and Mertens, P.R. (2015). Biomarkers: more 74 Stevens, P.E. and Levin, A. (2013). Kidney Disease:
than just markers! Nephrol. Dial. Transpl. 30 (1): 33–38. Improving Global Outcomes Chronic Kidney Disease
61 Sabbisetti, V.S., Waikar, S.S., Antoine, D.J. et al. (2014). Guideline Development Work Group M. Evaluation
Blood kidney injury molecule-1 is a biomarker of acute and management of chronic kidney disease: synopsis
and chronic kidney injury and predicts progression to of the kidney disease: improving global outcomes 2012
ESRD in type I diabetes. J. Am. Soc. Nephrol. 25 (10): clinical practice guideline. Ann. Intern. Med. 158 (11):
2177–2186. 825–830.
62 Peralta, C.A., Katz, R., Bonventre, J.V. et al. (2012). 75 Tangri, N., Grams, M.E., Levey, A.S. et al. (2016).
Associations of urinary levels of kidney injury molecule 1 Multinational assessment of accuracy of equations for
(KIM-1) and neutrophil gelatinase-associated lipocalin predicting risk of kidney failure: a meta-analysis. JAMA
(NGAL) with kidney function decline in the Multi-Ethnic 315 (2): 164–174.
Study of Atherosclerosis (MESA). Am. J. Kidney Dis. 60 76 Grams, M.E., Sang, Y., Ballew, S.H. et al. (2018).
(6): 904–911. Predicting timing of clinical outcomes in patients with
63 Hayek, S.S., Sever, S., Ko, Y.A. et al. (2015). Soluble chronic kidney disease and severely decreased glomerular
urokinase receptor and chronic kidney disease. N. Engl. J. filtration rate. Kidney Int. 93 (6): 1442–1451.
Med. 373 (20): 1916–1925. 77 Tangri, N., Ferguson, T., and Komenda, P. (2017). Pro:
64 Kottgen, A., Glazer, N.L., Dehghan, A. et al. (2009). risk scores for chronic kidney disease progression are
Multiple loci associated with indices of renal function robust, powerful and ready for implementation. Nephrol.
and chronic kidney disease. Nat. Genet. 41 (6): 712–717. Dia. Transpl. 32 (5): 748–751.
65 Kottgen, A., Pattaro, C., Boger, C.A. et al. (2010). New loci 78 Kazmi, W.H., Obrador, G.T., Khan, S.S. et al. (2004). Late
associated with kidney function and chronic kidney nephrology referral and mortality among patients with
disease. Nat. Genet. 42 (5): 376–384. end-stage renal disease: a propensity score analysis.
66 Shlipak, M.G., Li, Y., Fox, C. et al. (2011). Uromodulin Nephrol. Dial. Transpl 19 (7): 1808–1814.
concentrations are not associated with incident CKD 79 Moynihan, R., Glassock, R., and Doust, J. (2013). Chronic
among persons with coronary artery disease. BMC kidney disease controversy: how expanding definitions
Nephrol. 12: 2. are unnecessarily labelling many people as diseased. BMJ
67 See, E.J., Jayasinghe, K., Glassford, N. et al. (2019). 347: f4298.
Long-term risk of adverse outcomes after acute kidney 80 Wojciechowski, P., Tangri, N., Rigatto, C., and Komenda,
injury: a systematic review and meta-analysis of cohort P. (2016). Risk prediction in CKD: the rational alignment
studies using consensus definitions of exposure. Kidney of health care resources in CKD 4/5 care. Adv. Chronic
Int. 95 (1): 160–172. Kidney Dis. 23 (4): 227–230.
68 Lo, L.J., Go, A.S., Chertow, G.M. et al. (2009). Dialysis- 81 Woodward, G.L., Iverson, A., Harvey, R., and Blake, P.G.
requiring acute renal failure increases the risk of (2015). Implementation of an agency to improve chronic
progressive chronic kidney disease. Kidney Int. 76 (8): kidney disease care in Ontario: lessons learned by the
893–899. Ontario renal network. Healthc. Q. 17 Spec No:44–47.
69 Chawla, L.S., Amdur, R.L., Amodeo, S. et al. (2011). The 82 Shirazian, S., Grant, C.D., Tangri, N., and Mattana, J.
severity of acute kidney injury predicts progression to (2014). Have we been overestimating the risk of end-stage
chronic kidney disease. Kidney Int. 79 (12): 1361–1369. renal disease in older adults with chronic kidney disease?
70 Moons, K.G., Kengne, A.P., Grobbee, D.E. et al. (2012). J. Am. Geriatr. Soc. 62 (12): 2455–2456.
Risk prediction models: II. External validation, model 83 Drew, D.A., Lok, C.E., Cohen, J.T. et al. (2015). Vascular
updating, and impact assessment. Heart 98 (9): 691–698. access choice in incident hemodialysis patients: a
71 Moons, K.G.M., Kengne, A.P., Woodward, M. et al. decision analysis. J. Am. Soc. Nephrol. 26 (1): 183–191.
(2012). Risk prediction models: I. Development, internal 84 Hsu, C., Yang, W., Appel, L. (2016). Optimizing Timing of
validation, and assessing the incremental value of a new Pre-Emptive AV Access Creation: Results from the CRIC
(bio)marker. Heart 98 (9): 683–690. Study (Abstracts of Kidney Week, Chicago, IL, 2016.
Reference 71
Abstract SA-PO779, p. 810A. American Society of 94 Lerner, B., Desrochers, S., and Tangri, N. (2017). Risk
Nephrology, Washington, DC, USA.) prediction models in CKD. Semin. Nephrol. 37 (2): 144–150.
85 Matsushita, K., van der Velde, M., Astor, B.C. et al. (2010). 95 Bansal, N., Katz, R., De Boer, I.H. et al. (2015).
Association of estimated glomerular filtration rate and Development and validation of a model to predict 5-year
albuminuria with all-cause and cardiovascular mortality risk of death without ESRD among older adults with
in general population cohorts: a collaborative meta- CKD. Clin. J. Am. Soc. Nephrol. 10 (3): 363–371.
analysis. Lancet 375 (9731): 2073–2081. 96 Lees, J.S., Welsh, C.E., Celis-Morales, C.A. et al. (2019).
86 McClellan, W.M., Langston, R.D., and Presley, R. (2004). Glomerular filtration rate by differing measures,
Medicare patients with cardiovascular disease have a albuminuria and prediction of cardiovascular disease,
high prevalence of chronic kidney disease and a high rate mortality and end-stage kidney disease. Nat. Med. 25
of progression to end-stage renal disease. J. Am. Soc. (11): 1753–1760.
Nephrol. 15 (7): 1912–1919. 97 de Zeeuw, D., Parving, H.H., and Henning, R.H. (2006).
87 Buller, C.E., Nogareda, J.G., Ramanathan, K. et al. (2004). Microalbuminuria as an early marker for cardiovascular
The profile of cardiac patients with renal artery stenosis. disease. J. Am. Soc. Nephrol. 17 (8): 2100–2105.
J. Am. Coll. Cardiol. 43 (9): 1606–1613. 98 Hillege, H.L., Fidler, V., Diercks, G.F. et al. (2002).
88 United States Renal Data System (2019). 2019 USRDS Urinary albumin excretion predicts cardiovascular and
annual data report: epidemiology of kidney disease in the noncardiovascular mortality in general population.
United States. In: National Institutes of Health NIDDKD Circulation 106 (14): 1777–1782.
(ed. M.D. Bethesda). 99 Schmieder, R.E., Mann, J.F., Schumacher, H. et al.
89 Saran, R., Robinson, B., Abbott, K.C. et al. (2020). US (2011). Changes in albuminuria predict mortality and
renal data system 2019 annual data report: epidemiology morbidity in patients with vascular disease. J. Am. Soc.
of kidney disease in the United States. Am. J. Kidney Dis. Nephrol. 22 (7): 1353–1364.
75 (1 Suppl 1): A6–a7. 100 Solomon, S.D., Lin, J., Solomon, C.G. et al. (2007).
90 Liu, M., Li, X.C., Lu, L. et al. (2014). Cardiovascular Influence of albuminuria on cardiovascular risk in
disease and its relationship with chronic kidney disease. patients with stable coronary artery disease. Circulation
Eur. Rev. Med. Pharmacol. Sci. 18 (19): 2918–2926. 116 (23): 2687–2693.
91 Fan, J. and Salameh, H. (2016). Impact of chronic kidney 101 Wagner, D.K., Harris, T., and Madans, J.H. (1994).
disease on risk for vascular events. Curr. Vasc. Pharmacol. Proteinuria as a biomarker: risk of subsequent
14 (5): 409–414. morbidity and mortality. Environ. Res. 66 (2): 160–172.
92 Gregg, L.P. and Hedayati, S.S. (2018). Management of 102 Mogensen, C.E. (1984). Microalbuminuria predicts
traditional cardiovascular risk factors in CKD: what are clinical proteinuria and early mortality in maturity-
the data? Am. J. Kidney Dis. 72 (5): 728–744. onset diabetes. N. Engl. J. Med. 310 (6): 356–360.
93 Go, A.S., Chertow, G.M., Fan, D. et al. (2004). Chronic 103 Kannel, W.B., Stampfer, M.J., Castelli, W.P., and Verter,
kidney disease and the risks of death, cardiovascular J. (1984). The prognostic significance of proteinuria:
events, and hospitalization. N. Engl. J. Med. 351 (13): the Framingham study. Am. Heart J. 108 (5):
1296–1305. 1347–1352.
72
Chronic Kidney Disease in Disadvantaged Populations

Magdalena Madero1 and Diego Aguilar 2
1
Department of Nephrology, National Institute of Cardiology, Mexico
2
PRS Population Health, Clinical Trial Service Unit and Epidemiological Studies Unit, BHF Centre of Research Excellence, Nuffield Department of Population Health,
University of Oxford, UK
I ntroduction other populations, frequently victims of illegal traffic prac-

tices, which contributes to the depletion of an already
Chronic kidney disease (CKD) is an established global scarce pool of grafts available to disadvantaged persons.
public health problem that represents a key determinant Given the complexity and systemic nature of this prob-
for poor health outcomes [1–4]. In recent decades it has lem, the required interventions are interdisciplinary and
been increasingly recognized that CKD is more frequent multidimensional. Therefore, it is not only a matter
and has poorer outcomes among populations or commu- incumbent to the medical community. Only global-scale
nities considered “disadvantaged” in both the developed efforts involving multiple stakeholders will be able to
and developing world [1, 5–8]. Such communities include close the observed gaps in kidney care that disadvantaged
“ethnic, socio-economically deprived, and other minority populations experience. Key partnerships between gov-
groups that are at greater risk of developing CKD,” accord- ernments, nongovernmental organizations, researchers,
ing to the International Society of Nephrology [9]. The and clinicians are paramount to achieve an improvement
causes of CKD in disadvantaged populations are varied in outcomes. The World Health Organization (WHO) has
and reflect exposures to environmental, socioeconomic, established sustainable health goals that are only possible
and other factors that contribute to a reduction in nephron with a concerted multidisciplinary approach. This chap-
mass over time. In any one individual, perinatal factors, ter presents the available evidence of the distribution of
nutritional factors, exposure to toxins, and infections may CKD in disadvantaged populations across the globe, with
all contribute to the development of CKD. The entity a particular focus on the challenges and opportunities
CKD of uncertain etiology (CKDu) is one condition that in the detection and treatment of CKD among such
is over-represented in disadvantaged populations. populations.
Apart from being more frequent, kidney disease among
disadvantaged populations tends to be underdiagnosed.
When detected, access to care is limited and, frequently, E
pidemiology
suboptimal [8]. Irrespective of the underlying cause of
CKD, the combination of those two factors – underdiagnosis According to a recent meta-analysis that pooled data from
and suboptimal treatment – contributes to poor clinical 100 studies of diverse quality comprising 6 908 440 patients,
outcomes. Moreover, the development of end-stage kidney an estimated 11% of people worldwide are affected with
disease (ESKD) in communities in which access to renal CKD stages 3–5 [10]. Table 6.1 shows the CKD prevalence
replacement therapy (RRT) is limited or nonexistent forces estimates for different regions in this study. The prevalence
households to incur substantial out-of-pocket expenses, of CKD varied substantially across regions, which was
perpetuating the poverty cycle. Additionally, disadvantaged partially attributable to substantial heterogeneity between
populations are a recurrent source of organ donors for the studies included as well as differences in quality.
Definitions and Theoretical Framewor 73
Table 6.1 Estimated prevalence of CKD in specific countries.
CKD prevalence (%, 95% CI)
Stages 1–5 Stages 3–5
Region N Prevalence (%) N Prevalence (%)
South Africa, Senegal, Congo 5497 8.6 (1.3–16.0) 1202 7.6 (6.1–9.1)
India, Bangladesh 1000 13.1 (1.0–15.2) 12 752 6.8 (3.7–9.9)
Iran 17 911 17.9 (7.4–28.5) 20 867 11.7 (4.5–18.9)
Chile 0 – 27 894 12.1 (11.7–12.5)
China, Taiwan, Mongolia 570 187 13.2 (12.1–14.3) 62 062 10.7 (6.6–13.5)
Japan, South Korea, Oceania 654 832 13.7 (10.8–16.7) 298 000 11.7 (5.3–18.1)
Australia 12 107 14.7 (11.7–17.7) 896 941 8.1 (4.5–11.8)
USA, Canada 20 352 15.5 (11.7–19.2) 1 319 003 14.4 (8.5–20.3)
Europe 821 902 18.4 (11.6–25.2) 2 169 183 11.9 (9.9–13.8)
Global 13.4 (11.7–15.1) 10.6 (9.2–12.2)
Source: Reproduced from Hill et al. [10]. © 2016 Hill et al. Licensed Under CCBY 4.0. This is an open access article distributed under the terms
of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the
original author and source are credited.
Nonetheless, the observed differences can also be explained Mexico City show that CKD is strikingly frequent among
by true variations in the prevalence of kidney disease population with diabetes [16]. A prospective analysis of a
among regions. Remarkably, the global prevalence of CKD large cohort in this population recently reported that
appears to be higher than that of diabetes, which has been adults (35–84 years) with diabetes had a 20-fold higher
reported to be about 8.8% [11], underscoring the relevance risk of dying from kidney disease compared with those
of CKD to global health. without diabetes [17]. This excess risk is most likely due
Another study that pooled data from 33 reports with to suboptimal management of diabetes: those with diabe-
gender-and age-specific prevalence of CKD in represent- tes had a mean glycated hemoglobin of 9.0% and poor gly-
ative population samples estimated that nearly 500 mil- cemic control was associated with increased mortality
lion individuals have CKD, of which almost 80% live in risk [18].
low-middle-income countries (LMICs) [12]. Importantly, The etiology and epidemiological determinants of kid-
the age-standardized prevalence of CKD was higher in ney diseases in areas like Africa, Asia, and Latin America
men and women from LMICs when compared to men and differ from what is observed in Europe and North
women from high-income countries. Despite the observed America and even within regions can vary substan-
disparities between developed and developing countries, tially [14]. As shown in Table 6.2, apart from diabetes
the epidemiological trends of CKD in LMICs, including and hypertension, communicable diseases such as HIV,
the causes of kidney disease and their distribution across malaria, and schistosomiasis may be important contribu-
the population, are either unknown or poorly character- tors to the burden of CKD in sub-Saharan Africa, even
ized [13, 14]. There are reasons to believe that the causes though evidence of their attributable fractions is limited
of CKD are widely heterogeneous in LMICs. As a result of or absent [19, 20].
an accelerated demographic transition and urbanization,
most of these countries have experienced an unprece-
dented upsurge in noncommunicable diseases [15]. A efinitions and Theoretical
D
sedentary lifestyle and unhealthy diet are intertwined in Framework
densely crowded cities with limited public health and
heathcare infrastructure, resulting in high rates of under- According to the WHO, social determinants of health
diagnosed and suboptimally treated hypertension and (SDH) are defined as “the conditions in which people
diabetes. These populations are enormously vulnerable are born, grow, live, work and age” [21]. SDH are
for adverse kidney outcomes. For instance, reports from mostly responsible for health inequities, “the unfair and
74 Chronic Kidney Disease in Disadvantaged Populations
Table 6.2 Environmental factors associated with CKD. groups” [9]. Even though the definition of minority
groups largely derives from sociodemographic phenom-
Factor Description ena (i.e. the definition of a minority is relative to a major-
ity in a population), [24] the concept of ethnicity is
Heavy metals Lead: particularly important in LMIC;
contained in paint, gasoline, adulterated closely related with genetic ancestry and, subsequently,
gasoline, contaminated food and water, soil, with a complex cluster of social and biological factors.
mines, garages The definition of “disadvantaged populations” can there-
Cadmium: sources include tobacco, batteries, fore be ambiguous, as a myriad of factors can predispose
fuel, industrial waste, certain South-Eastern
Asian medicinal herbs or foods
individuals and populations to kidney disease. Three
Arsenic: fish contamination, fuel, cereals main factors, however, can be identified as underlying
treated with ethyl mercury as pesticide and sometimes overlapping causes of disadvantage: (i)
Uranium: contaminated food or water socioeconomic factors that affect health behavior and
access to care, (ii) biological factors that predispose to
Other toxins Aristolochic acid associated with chronic
tubulointerstitial nephritis and urothelial adverse kidney outcomes, and (iii) environmental fac-
cancer; the causal agent of Balkan endemic tors, e.g. labor conditions and exposures to nephrotoxic
nephropathy agents. These factors are not necessarily a direct cause of
Exposure to other chemicals used for kidney disease but may act as mediators on the causal
consumer activities or medical treatment; pathway (Figure 6.1).
industrial chemicals; occupational chemicals
apart from heavy metals (e.g. methylene
chloride, trichloroethylene, toluene); most
associated with acute disease and/or
Socioeconomic Factors
accentuated progression of underlying Socioeconomic status (SES) usually encompasses educa-
disease
tion, occupation, and income as proxies for measurement
Occupational Mesoamerican nephropathy affecting of social and economic wellbeing. Lower SES has con-
exposures agricultural workers; most likely mechanism sistently been associated with increased morbidity and
consists in repeated subclinical episodes of
AKI due to dehydration during work shifts; all-cause mortality [25]. Frequently, access to health –
uric acid crystalluria might be involved in frequently measured as insurance status – and risk
kidney damage; several other potential behaviors have been suggested as the mechanisms
synergistic factors (e.g. NSAIDs, fructose- explaining the link between SES and health. However, it
high beverages, alcohol consumption,
pesticides) has become clear that those factors do not fully account
Sri Lanka, India, and Pakistan: for the associations observed, and other more complex
contamination of water/food by unknown factors such as social isolation and adversity may be
agent (industrial chemicals, agrochemicals, involved [22]. This is not different for CKD, where the
heavy metals suggested) incidence and prevalence of CKD and the progression of
Medications NSAIDs (linked with AKI, but not yet with disease varies by SES, being higher at lower SES lev-
CKD); mis-regulated alternative or novel els [26, 27]. For instance, low income or education below
therapies or medications (usually high school level have been associated with the presence
counterfeit); herbal remedies
of albuminuria in the United States [28]. In the Chronic
Infections Related to poor hygiene conditions in LMICs; Renal Insufficiency Cohort, lower income and education
parasites (e.g. schistosomiasis, malaria, were associated with lower estimated glomerular filtra-
leptospirosis, tuberculosis, etc.); enteric and
diarrheal diseases; HIV-associated
tion rate (eGFR) [29]. Progression of CKD into ESKD is
nephropathy; hepatitis B and C viruses more likely among individuals with low income and edu-
cation [30–32]. Additionally, those with ESKD who have
less education or are unemployed are less likely to receive
avoidable differences in health status seen within and a transplant [33, 34] and more likely to experience dialy-
between countries” [21]. Therefore SDH, and the result- sis complications such as peritonitis [35, 36] or graft
ing disparities in health, are key components of what is rejection if they receive a transplant [37]. Irrespective of
considered a “disadvantaged population” [22, 23]. the mediating factors in the association between SES and
However, the definition most widely accepted for “disad- CKD, it is clear that a low SES confers a greater risk of
vantaged population” is not restricted to those at a greater developing CKD and having poor outcomes, and there-
risk of kidney disease due to SDH, and further includes fore such segments of the population can be considered
“ethnic, socio-economically deprived, and other minority disadvantaged.
Definitions and Theoretical Framewor 75
Environmental
exposures:
- Occupation
- Heavy metals and
chemicals
- Herbal remedies
Poor
Socioeconomic Risk factors: Outcomes
factors Access to care - CVD
Health behaviour: - DM & HTN
- Education - Diagnosis - ESKD
- Lifestyle - Chronic
- Income - Management - Catastrophic
- Psychosocial infections; viral,
- Cultural - Referral expenditure
- Health literacy bacterial,
factors - Access to RRT - QoL
parasitary
- Geography - Premature
death
Biological factors:
- Age & sex
- Maternal-fetal
health
- Genetics
- Others
Figure 6.1 Theoretical model: interaction of socioeconomic, environmental, and biological factors and their impact on health
behavior, risk factors, and direct causes of CKD, access to care, and outcomes. CKD, chronic kidney disease; CVD, DM, ESKD, end-stage
kidney disease; HTN, QoL,
Environmental Exposures professionals. This limited capacity for systematic research

has hindered efforts to accurately identify hotspots for
Environmental exposures associated with the development
disease and reliably estimate incidence rates [13, 44, 61].
of kidney disease are well known [38–40]. They include
heavy metals such as lead, arsenic, mercury, and cadmium,
Biological Factors
other toxins such as the aristolochic acid (the cause of the
Balkan endemic nephropathy), industrial and agricultural Apart from aging, there are two main determinants that
chemicals, occupational exposures, which are not fully predispose individuals to developing kidney disease and
delineated but appear to correlate with specific occupations can be considered inherently biological: genetics and
(e.g. Mesoamerican nephropathy and agricultural workers), developmental programming. Genome-wide association
medications and herbal remedies, and infections (Table 6.3). studies and other novel sequencing techniques have
Some of these factors have been recently linked with the identified at least 50 genetic regions or variants associated
emerging CKDu hotspots around the globe [38, 60]. Defined with adverse kidney outcomes and the development of
as “countries, regions, communities, or ethnicities with CKD [62]. As genetic variants are inherently linked with
higher than an average incidence of CKD” [61], these hot- ancestry and vary across populations, population-based
spots are currently of major research interest, even though genetic studies have given key insights into kidney function
there is a paucity of evidence to establish a cause–effect rela- traits and have been helpful toward understanding the
tionship in most cases [13]. Importantly, most of these hot- distribution of kidney disease in specific populations.
spots are located in LMICs with limited health resources, In the United States, a major disparity in CKD has
and thus are closely linked to socioeconomic factors. The long been documented between black and nonblack popu-
cases arise mainly from socioeconomically deprived com- lations [56]. African Americans have historically had
munities, mostly consisting of working-class individuals higher progression of diabetic and nondiabetic kidney
whose income directly depends on their ability to continue disease. Even though African Americans represent about
working in the very conditions that potentially expose them 13% of the US population, they comprise 33% of incident
to kidney injury. Frequently, affected communities lack cases of ESKD. Moreover, the lifetime risk of ESKD in
social security and have limited availability of public health African Americans is an estimated 8.2% (compared
protection programs. Associated with this deprivation comes with only 2.7% in European Americans) [57], with a
a lack of health resources, both infrastructure and trained disproportionate number of patients requiring RRT.
Table 6.3 Description of identified CKD hotspots or disadvantaged populations with CKD.
Certainty of evidence
Region Cause or type of CKD Comments (GRADE)
Sub-Saharan Africa HIV-related HIV-related GN in countries with high High

prevalence of the infection [20]; certain genetic ●●●●
variants in the APOL1 gene may predispose to
HIV-related nephropathy in African population
[41]; the prevalence of CKD among populations
with HIV is estimated to be around 12% in
sub-Saharan Africa [19]
Malaria Mostly caused by Plasmodium falciparum [20];
immune-mediated glomerular and tubular
disease [42]; 250 million cases of malaria per year of
which 90% in sub-Saharan Africa; 445 000 deaths in
2015, of which 91% in sub-Saharan Africa; renal
involvement (i.e. AKI) in at least half of malaria
deaths [42, 43]
Limited evidence of a direct association of malaria
and CKD
Central America and Mesoamerican nephropathy Working conditions; unknown etiology; repeated Low
Mexico episodes of acute kidney injury which evolve into ●●○○
overt CKD; possibly related with heat-stress,
exposure to agrochemicals, use of NSAID and other
over-the-counter drugs [13, 44, 45]
Diabetic nephropathy Unhealthy diet and sedentary lifestyle; high High
rates of overweight and obesity; high prevalence ●●●●
of diabetes with suboptimal management [17,
18]; high prevalence of CKD among those with
diabetes and or hypertension, with most of them
being undiagnosed [16, 46]; higher risks of
kidney-associated death [17, 18]
Southern Asia Sri Lanka nephropathy Chronic tubule-interstitial nephritis in agricultural Low
and construction workers; unknown cause; exposure ●●○○
to cadmium through food chain and pesticides
exacerbated by genetic factors [47] or selenium
deficiency [48] suggested as possible causes;
clinically similar to MeN [38]
Andhra Pradesh, India Coastal regions known as Udannam area; unknown
cause; contamination of drinking water suspected,
but no evidence found of chemicals above
permissible limits [49]
Oceania Australian Northern Territory Higher incidence of CKD among rural aboriginal Low
and Maori and Pacific communities in central Australia; probably ●●○○
communities in New Zealand associated with higher prevalence of overweight
and obesity as early as from childhood; a potentially
strong role of developmental origins of disease [27,
50–52]
Europe Balkan endemic nephropathy Endemic along the Danube river Moderate
Most likely cause is consumption of aristolochic ●●●○
acid obtained from wheat frown in fields
contaminated with Aristolochia clematitis [53–55]
North America African-American communities Genetic predisposition for CKD (APOL1 high-risk High
variants) in combination with health ●●●●
disparities [56, 57]
Hispanic agricultural workers AKI among agricultural workers after workshifts; Low
potentially similar to MeN [58, 59]. ●●○○
AKI, acute kidney injury; CKD, chronic kidney disease; MeN, NSAID,
Challenges in the Diagnosis of CKD Among Disadvantaged Population 77
It was recently discovered that two coding variants of gain after birth is associated with childhood and adult-
the Apolipoprotein L1 (APOL1) gene on chromosome hood obesity [77–79], a well-known risk factor for diabe-
22q13 were associated with higher rates of focal segmental tes, hypertension, and kidney disease. Likewise, low
glomerulosclerosis (odds ratio 10.5) and hypertension- birth weight, prematurity, and small size for gestational
attributed ESKD (odds ratio 7.3) [63, 64]. Interestingly, age have been associated with reduced nephron mass
these two variants are common in individuals of recent and an increased risk of developing CKD and hyperten-
African ancestry, mostly absent in European populations, sion in adult life [80]. The global impact of these risk fac-
and appear to be protective against African sleeping sick- tors is substantial. In LMICs, a total of 43.3 million
ness as Apolipoprotein L-1 has an active lytic role in trypa- deliveries were either premature or small for gestational
nosomes. Moreover, it has been proposed that these age in 2010 [81], thus giving them a higher risk of devel-
variants may be involved in the development of HIV- oping noncommunicable diseases including kidney dis-
associated nephropathy [41], sickle cell kidney disease [65], ease [67]. Albeit a reduced nephron number may not be
and severe lupus nephritis [66]. sufficient to develop CKD in adult life, the interaction of
Understanding the role of APOL1 is paramount for such predisposition with high-risk genetic variants, envi-
assessing the burden of CKD among African Americans ronmental factors, and other stressors (e.g. episodes of
and, globally, in populations with African ancestry. It is acute kidney injury) may become relevant in a fraction of
estimated that about five million African Americans cases. A life course approach that assesses the interaction
possess at least two APOL1 nephropathy variants, which between genetics, perinatal, childhood, and adulthood
puts these populations at a higher risk of developing exposures is likely to better inform our understanding of
CKD. APOL1-associated kidney disease could poten- the development of CKD in disadvantaged popula-
tially account for up to 40% of the total burden observed tions [14, 44].
in African Americans [57]. The interaction of APOL1 There are important gaps to be addressed in the under-
with environmental – potentially modifiable – risk fac- standing of biological factors and kidney disease [38].
tors and other genes represents a major area of interest First, education and awareness of clinicians, researchers,
for research that will allow a more detailed phenotyping and patients is still limited. This is particularly relevant in
of kidney disease, particularly in those individuals with terms of translation of research findings into therapeu-
poor response to therapies for blood pressure control tics, as well understanding the attributable fraction of
and proteinuria reduction. The example of APOL1 genetics to the overall burden of kidney disease. Second,
underscores the important role that genetic factors may the majority of genetic research is carried out in popula-
have in the burden of kidney disease and its clustering tions of European ancestry, even though indigenous and
in specific populations. Research on interactions other disadvantaged populations have shown higher rates
between genes and environment is still at an early stage, of kidney disease. This lack of diversity substantially hin-
although it is an area of great interest for CKD hotspots. ders the generalizability of genetic findings, raises the
Some examples include Sri Lanka nephropathy and a issue of potentially misclassifying benign variants as
single-nucleotide polymorphism in the SCL13A3 gene, pathogenic [82], and limits the understanding of kidney
which appears to predispose to CKD [47]. Likewise, a disease among disadvantaged populations. An improved
gene in 3q25 could explain the familial clustering of the knowledge of genetic and developmental determinants of
Balkan endemic nephropathy [53]. Further studies are, kidney disease and their interaction with environmental
however, still required to replicate and confirm such factors may provide important insights into renal patho-
associations. physiology, which in turn could facilitate the develop-
In addition to genetic and lifestyle factors as determi- ment of more effective prevention strategies and pave the
nants of noncommunicable diseases, the “developmental way for novel therapeutics.
origins of disease” has emerged as a comprehensive
pathophysiological model of disease risk, which may be
of particular relevance to the kidney [67]. This model hallenges in the Diagnosis of CKD
C
proposes that individuals at greatest risk of kidney dis- Among Disadvantaged Populations
ease [68, 69] – as well as diabetes, [70, 71] hyperten-
sion, [72] and cardiovascular disease [73, 74] – are those There are several challenges which need to be systemati-
who experienced adverse physiological conditions such cally addressed in the context of CKD in disadvantaged
as maternal malnutrition [75, 76] during gestation, populations. These include lack of awareness, strategies for
resulting in low birth weight and underdevelopment of screening, workforce capacity, and availability of labora-
fetal kidneys. Furthermore, consequent rapid weight tory tests.
Awareness hypertension, maintenance of good glycemic control, and

the use of lipid-lowering therapy [89, 90]. There is, how-
Lack of awareness is a major barrier to tackling the public
ever, uncertainty surrounding CKDu hotspots since little
health issue of CKD. This is particularly problematic in
is known about the natural history of disease and there is
LMICs, where the majority of individuals with CKD go
no evidence of the effectiveness of therapy. Nevertheless,
undetected [23]. It is estimated that only one-third of
the global consensus is that an early diagnosis is preferred
individuals with CKD are diagnosed worldwide [4]. A
with the objective of identifying the cause of kidney
population-wide study in the United States in high-risk
injury and preventing progression and the requirement of
individuals reported that the prevalence and awareness of
RRT [2].
CKD were, respectively, 23.4% and 11.9% for Asians and
Pacific Islanders respondents, 29.5% and 8.6% for European
Americans, 20.3% and 11.1% for Hispanics, 22.8% and 6.3% Diagnostic Tools
for African Americans, and 29.2% and 6.8% for Native
A related challenge is the selection of tests for screening
Americans [83]. A study in China reported that the
and diagnosis. There is very limited evidence to guide
prevalence and awareness of CKD were 5.7% and 1.0% [84],
choices, and decisions should be based on local and
whereas in another study from this country, only 8% of
regional resources as well as the cultural acceptability of
rural Chinese individuals with CKD were aware of having
each mode of testing. An initial approach with risk scores
the disease [85]. Finally, a study from Mexico reporting
and questionnaires to identify high-risk individuals is
prevalence of CKD in high-risk populations showed that
potentially useful for large-scale screening. The models
only 1% of the participants were aware of having CKD
used for predicting the risk of having CKD and its pro-
despite 71% of them having had a doctor’s appointment in
gression are largely based on European or North American
the last year [16]. Further reports show that the proportion
populations and often require measuring biomarkers.
of the CKD population that is aware of their condition is
This is a major inconvenience in resource-poor settings
low in many other LMICs, including Tanzania (10%) [14],
where laboratory testing is not readily available [87, 91].
Senegal (23%), and Taiwan (3.5%) [86]. These data suggest
In addition, the diagnosis and staging of CKD require the
that the gap in awareness is important in all populations,
measurement of both glomerular filtration rate and albu-
but may be substantially larger in LMICs.
minuria [89], with multiple tests available. The quality of
such tests in LMICs varies widely, with limited use of vali-
dated tools [19, 92]. These differences could potentially
Screening
explain the heterogeneity observed in the estimates of
A major challenge in the diagnosis of CKD is deciding if a CKD prevalence in these populations. In a 2014 system-
screening program – either in the general population or atic review and meta-analysis on the epidemiology of
restricted to high-risk individuals – is appropriate [87]. CKD in sub-Saharan Africa that included 90 studies from
Wilson and Junger proposed 10 criteria in 1968 to help 96 sites, the measurement of urine protein was the most
decision-makers in the implementation of screening pro- common method for determining the presence of kidney
grams. In disadvantaged populations, special considera- disease (62 out of 90 studies [69%]) [19]. Only 22 studies
tion of the availability of diagnostic tools and treatment (24%) used the Cockcroft–Gault formula and 17 (19%)
options that are affordable and sustainable must be prior- used the Modification of Diet in Renal Disease (MDRD)
itized. The arguments in favor of screening for CKD in formula to eGFR. This study reported an overall preva-
disadvantaged populations are, first, that the burden is lence of CKD of 13.9% (95% CI 12.2–15.7). Substantial
substantial, as shown by the observed high prevalence, variability was found between countries, with estimates
the low rate of awareness, and unduly high risks of pre- ranging from 2% in Cote d’Ivoire to 30% in Zimbabwe.
mature death and morbidity reported among such popu- Apart from differences in sampling procedures, this het-
lations. Second, that the economic burden of CKD is erogeneity was also attributed to differences in the meas-
substantial and appears to be higher in LMIC, for both urement methods. One plausible solution to facilitate
households and health systems [4]. Third, the natural his- screening and diagnosis of CKD in resource-limited set-
tory of CKD is relatively well understood. The progres- tings is the use of point-of-care tests for measuring creati-
sion of CKD from its early stages to kidney failure ranges nine [93] and albuminuria [94]. Although these methods
between 12 and 15 years, although the range can be vari- are not as reliable as laboratory-based measurements,
able [88]. This window allows for the implementation of they offer advantages including prompt decision-making
effective strategies that can slow the progression of dis- by the clinician, avoidance of time-consuming logistics
ease and/or prevent complications, such as treatment of for transportation of specimens (of major relevance for
Challenges and Opportunities in Treatmen 79
isolated tropical communities in which samples can be were offered hemodialysis (73%), with only a small propor-
damaged by heat), and cost savings for patients through tion having access to peritoneal dialysis (1%) or transplan-
reductions in clinic visits. tation (<1%). The remaining 25% received conservative
management due to unaffordability of dialysis or ongoing
viral infection which precluded their access to hemodialy-
Other Challenges
sis. Moreover, only 5% of those who initiated hemodialysis
There are several additional challenges associated with were able to afford treatment for longer than 3 months [96].
diagnosing CKD in disadvantaged populations, including A similar scenario has been described in India, where less
diversity in the underlying causes of kidney disease and a than 10% of patients requiring RRT receive it and of those
lack of consensus on diagnostic criteria for specific >70% die within the first 12 weeks due to inability to afford
diseases, i.e. mesoamerican nephropathy and other cases sustained treatment [97, 98]. Likewise, in Asian countries
of unknown etiology. The limited availability of reliable without funded renal programs, around 90% of patients
and validated measures of kidney function is compounded with ESKD die within months of diagnosis due to a lack of
by a paucity of adequately trained health professionals, access [60]. As the costs of dialysis rise to US$20000–
particularly in under-resourced communities. 100 000 per year [99], for a great proportion of individuals
from disadvantaged communities RRT becomes prohibi-
tively expensive.
hallenges and Opportunities
C A systematic review and meta-analysis that pooled data
in Treatment from 123 countries and included 93% of the worldwide
population estimated that, in 2010, 2.618 million individu-
Given the high costs associated with provision of RRT, the als received RRT, of which 2.050 million (78%) received
main focus of therapeutic interventions should be on CKD dialysis and the remainder a transplant [8]. It was conserv-
prevention, slowing disease progression, and avoiding atively estimated that the number of patients needing RRT
complications. was 4.902 million, but it could be closer 9.701 million.
Thus, between half (2.284 million, 47%) and up to three-
quarters (7.083 million, 73%) of individuals requiring RRT
Early Prevention Programs
worldwide did not receive it, which translated to at least
These programs are centered on early and aggressive man- 2.284 million premature deaths. Notably, an alarming dis-
agement of risk factors, and have been shown to be cost- parity between the need for and the receipt of RRT was
effective for cardinal risk factors such as diabetes and found in terms of income level. Of all those receiving RRT
hypertension [95]. Interventions may include early detec- in 2010, 92.8% resided in high-income or high-middle
tion using low-cost tools such as albuminuria testing, and income countries. This represents a striking 70-fold preva-
the use of renoprotective agents such as renin angiotensin lence gradient [6]. By contrast, only 7.2% resided in low-
aldosterone system blockers. income countries where 96% of the patients requiring RRT
did not receive it. This proportion decreases to 88% in
LMICs and 58% in high-middle income countries, which
Access to Optimal Care for CKD
suggests that access to RRT is socially graded. The projec-
Limited access to primary care and specialist nephrology tions from this meta-analysis estimated that by 2030 the
care is a common problem in LMICs. There is scope for number of individuals requiring RRT will more than dou-
international training programs to enhance capacity in ble to a total of 5.439 million, with the greatest increase in
CKD care in these countries, for example the fellowship Asia and Africa, the regions where the population is
opportunities provided by the International Society of expected to grow the most.
Nephrology. In addition, efforts should be made to widen Unhealthy aging has been largely documented as the
the availability and affordability of medications with result of gains in economic development and increases in
proven benefit for CKD patients, such as antihypertensives life expectancy in the developing world and is arguably the
and lipid-lowering medications. demographic reason pushing the expected increase in the
demand of RRT. This is of relevance in those disadvan-
taged populations in which unhealthy behaviors have
Availability of RRT
increased the prevalence of risk factors for CKD such as
ESKD in disadvantaged populations has an unduly poor hypertension and diabetes. Countries must prepare their
outcome. A retrospective study in Nigeria that included health systems to face the challenge ahead and meet the
760 patients with incident ESKD showed that the majority demands of RRT in the present and foreseeable future.
RRT rationing resulting from limited facilities and unique genetic, environmental, and socioeconomic fac-
resources in low and middle-income countries is complex. tors. Implementation of effective prevention programs
Key areas of opportunity in regard to RRT include estab- and improvements in quality of care for the diverse causes
lishing registries of dialysis and transplantation in all of kidney disease in each population is required. Such
countries. Likewise, identifying patients who need RRT programs must be tailored to local requirements so that
but do not receive it is important. Open sharing of method- they can meet the epidemiological, cultural, and societal
ology, sampling frameworks, and innovative strategies needs of the communities they serve, and ultimately
between countries could potentially facilitate this work in translate into better clinical outcomes. Identifying the
countries without a registry. major risk factors in each community is paramount to
maximize the efficacy of any intervention. Similarly, pri-
oritizing low-cost interventions, such as albuminuria test-
Health Literacy
ing followed by treatment with blockers of the
Limited health literacy affects 25% of the population with renin-angiotensin system or bicarbonate supplementa-
CKD and this percentage is likely higher in disadvantaged tion for metabolic acidosis, may have a significant impact
populations and has been associated with adverse out- on CKD prevention and disease progression. Only by
comes in CKD populations [100]. A recent meta-analysis tracking these outcomes can the implementation of prac-
that examined 13 studies found that limited health literacy tical interventions be appropriately evaluated. Inequity in
was significantly associated with hospitalizations, emer- health is at the very core of the observed disparities in
gency department use, missed dialysis sessions, cardiovas- CKD burden among disadvantaged populations. It is a
cular events, and mortality [101]. moral imperative to solve the existing disparities, espe-
cially given the evidence that, in some cases, the lack of
access to RRT is due to inefficient allocation of available
C
onclusion resources rather than a lack of them [102]. The WHO has
established sustainable development goals for universal
CKD in disadvantaged populations is increasingly recog- health coverage. It is paramount that efforts from global
nized in low-, middle- and high-income countries and is health organizations are championed by researchers and
due in part to rising incidence and prevalence of hyper- clinicians at every level to achieve an improvement in
tension and diabetes in those populations, as well as CKD care worldwide.
R
eferences
Bello, A.K., Levin, A., Tonelli, M. et al. (2017). Global

1 7 Ene-Iordache, B., Perico, N., Bikbov, B. et al. (2016).
Kidney Health Atlas: A Report by the International Society of Chronic kidney disease and cardiovascular risk in six
Nephrology on the Current State of Organization and regions of the world (ISN-KDDC): a cross-sectional study.
Structures for Kidney Care Across the Globe. Brussels, Lancet Glob. Health 4 (5): e307–e319.
Belgium: International Society of Nephrology. 8 Liyanage, T., Ninomiya, T., Jha, V. et al. (2015).
2 Levin, A., Tonelli, M., Bonventre, J. et al. (2017). Global Worldwide access to treatment for end-stage kidney
kidney health 2017 and beyond: a roadmap for closing gaps disease: a systematic review. Lancet 385 (9981):
in care, research, and policy. Lancet. 1975–1982.
21;390(10105):1888–1917. 9 Disadvantaged Populations. (2018). Available from:
3 Eckardt, K.-U., Coresh, J., Devuyst, O. et al. (2013). https://www.theisn.org/ (accessed 22 May 2021).
Evolving importance of kidney disease: from subspecialty 10 Hill, N.R., Fatoba, S.T., Oke, J.L. et al. (2016). Global
to global health burden. Lancet 382 (9887): 158–169. prevalence of chronic kidney disease – a systematic
4 Jha, V., Garcia-Garcia, G., Iseki, K. et al. (2013). Chronic review and meta-analysis (ed. G. Remuzzi). PLoS One 11
kidney disease: global dimension and perspectives. Lancet (7): e0158765.
382 (9888): 260–272. 11 International Diabetes Federation (2017). IDF
5 Barsoum, R.S. (2006). Chronic kidney disease in the Diabetes Atlas, 8e. Brussels: International Diabetes
developing World. N. Engl. J. Med. 354 (10): 997–999. Federation. ISBN: 978-2 -9 30229-8 7-4 www.
6 Coresh, J. and Jafar, T.H. (2015). Disparities in worldwide diabetesatlas.org.
treatment of kidney failure. Lancet 385 (9981): 12 Mills, K.T., Xu, Y., Zhang, W. et al. (2015). A systematic
1926–1928. analysis of worldwide population-based data on the
Reference 81
global burden of chronic kidney disease in 2010. Kidney 27 White, S.L., McGeechan, K., Jones, M. et al. (2008).
Int. 88 (5): 950–957. Socioeconomic disadvantage and kidney disease in the
13 Lunyera, J., Mohottige, D., Isenburg, M.V. et al. (2016). United States, Australia, and Thailand. Am. J. Public
CKD of uncertain etiology: a systematic review. Clin. J. Health 98 (7): 1306–1313.
Am. Soc. Nephrol. 11 (3): 379–385. 28 Martins, D., Tareen, N., Zadshir, A. et al. (2006). The
14 Stanifer, J.W., Muiru, A., Jafar, T.H., and Patel, U.D. association of poverty with the prevalence of
(2016). Chronic kidney disease in low-and middle- albuminuria: data from the Third National Health and
income countries. Nephrol. Dial. Transplant. 31 (6): Nutrition Examination Survey (NHANES III). Am. J.
868–874. Kidney Dis. 47 (6): 965–971.
15 McKeown, R.E. (2009). The epidemiologic transition: 29 Lash, J.P., Go, A.S., Appel, L.J. et al. (2009). Chronic
changing patterns of mortality and population dynamics. Renal Insufficiency Cohort (CRIC) Study: baseline
Am. J. Lifestyle Med. 3 (1_suppl): 19S–26S. characteristics and associations with kidney function.
16 Obrador, G.T., García-García, G., Villa, A.R. et al. (2010). Clin. J. Am. Soc. Nephrol. 4 (8): 1302–1311.
Prevalence of chronic kidney disease in the Kidney Early 30 Lipworth, L., Mumma, M.T., Cavanaugh, K.L. et al. (2012).
Evaluation Program (KEEP) México and comparison Incidence and predictors of end stage renal disease among
with KEEP US. Kidney Int. Suppl. 116: S2–S8. low-income blacks and whites. PLoS One 7 (10): e48407.
17 Alegre-Díaz, J., Herrington, W., López-Cervantes, M. 31 Hsu, C., Iribarren, C., McCulloch, C.E. et al. (2009). Risk
et al. (2016). Diabetes and cause-specific mortality in factors for end-stage renal disease: 25-year follow-up.
Mexico City. N. Engl. J. Med. 375 (20): 1961–1971. Arch. Intern. Med. 169 (4): 342–350.
18 Herrington, W.G., Alegre-Díaz, J., Wade, R. et al. (2018). 32 Crews, D.C., Gutiérrez, O.M., Fedewa, S.A. et al. (2014).
Effect of diabetes duration and glycaemic control on Low income, community poverty and risk of end stage
14-year cause-specific mortality in Mexican adults: a renal disease. BMC Nephrol. 15: 192.
blood-based prospective cohort study. Lancet Diabetes 33 Sandhu, G.S., Khattak, M., Pavlakis, M. et al. (2013).
Endocrinol. 19 (6):455–463. Recipient’s unemployment restricts access to renal
19 Stanifer, J.W., Jing, B., Tolan, S. et al. (2014). The transplantation. Clin. Transpl. 27 (4): 598–606.
epidemiology of chronic kidney disease in sub-Saharan 34 Goldfarb-Rumyantzev, A.S., Sandhu, G.S., Baird, B.
Africa: a systematic review and meta-analysis. Lancet et al. (2012). Effect of education on racial disparities in
Glob. Health 2 (3): e174–e181. access to kidney transplantation. Clin. Transpl. 26 (1):
20 Arogundade, F.A., Hassan, M.O., Omotoso, B.A. et al. 74–81.
(2016). Spectrum of kidney diseases in Africa: malaria, 35 Chow, K.M., Szeto, C.C., Leung, C.B. et al. (2005). Impact
schistosomiasis, sickle cell disease, and toxins. Clin. of social factors on patients on peritoneal dialysis.
Nephrol. 86 (13): 53–60. Nephrol. Dial. Transpl. 20 (11): 2504–2510.
21 WHO (2018). About social determinants of health 36 Martin, L.C., Caramori, J.C.T., Fernandes, N. et al. (2011).
[Internet]. WHO. http://www.who.int/social_determinants/ Geographic and educational factors and risk of the first
sdh_definition/en (accessed on 22 May 2021). peritonitis episode in Brazilian Peritoneal Dialysis study
22 Norton, J.M., Moxey-Mims, M.M., Eggers, P.W. et al. (BRAZPD) patients. Clin. J. Am. Soc. Nephrol. 6 (8):
(2016). Social determinants of racial disparities in CKD. J. 1944–1951.
Am. Soc. Nephrol. 27 (9): 2576–2595. 37 Taber, D.J., Hamedi, M., Rodrigue, J.R. et al. (2016).
23 Garcia-Garcia, G., Jha, V., and World Kidney Day Steering Quantifying the race stratified impact of socioeconomics
Committee (2015). Chronic kidney disease in on graft outcomes in kidney transplant recipients.
disadvantaged populations. Curr. Opin. Nephrol. Transplantation 100 (7): 1550–1557.
Hypertens. 24 (3): 203–207. 38 Obrador, G.T., Schultheiss, U.T., Kretzler, M. et al. (2017).
24 Williams, D.R. and Sternthal, M. (2010). Genetic and environmental risk factors for chronic
Understanding racial-ethnic disparities in health: kidney disease. Kidney Int. Suppl. 7 (2): 88–106.
sociological contributions. J. Health Soc. Behav. 51 39 Weaver, V.M., Fadrowski, J.J., and Jaar, B.G. (2015).
(Suppl): S15–S27. Global dimensions of chronic kidney disease of unknown
25 Adler, N.E., Boyce, W.T., Chesney, M.A. et al. (1993). etiology (CKDu): a modern era environmental and/or
Socioeconomic inequalities in health. No easy solution. occupational nephropathy? BMC Nephrol. 16 (1) 16:145.
JAMA 269 (24): 3140–3145. 40 Soderland, P., Lovekar, S., Weiner, D.E. et al. (2010).
26 Patzer, R.E. and McClellan, W.M. (2012). Influence of Chronic kidney disease associated with environmental
race, ethnicity and socioeconomic status on kidney toxins and exposures. Adv. Chronic Kidney Dis. 17 (3):
disease. Nat. Rev. Nephrol. 8 (9): 533–541. 254–264.
41 Kopp, J.B., Nelson, G.W., Sampath, K. et al. (2011). 54 Debelle, F.D., Vanherweghem, J.-L., and Nortier, J.L.
APOL1 genetic variants in focal segmental (2008). Aristolochic acid nephropathy: a worldwide
glomerulosclerosis and HIV-associated nephropathy. J. problem. Kidney Int. 74 (2): 158–169.
Am. Soc. Nephrol. 22 (11): 2129–2137. 55 Yang, L., Su, T., Li, X.-M. et al. (2012). Aristolochic acid
42 Das, B.S. (2008). Renal failure in malaria. J. Vector Borne nephropathy: variation in presentation and prognosis.
Dis. 45 (2): 83–97. Nephrol. Dial. Transpl. 27 (1): 292–298.
43 Tran, T.H., Day, N.P., Nguyen, H.P. et al. (1996). A 56 Muntner, P., Newsome, B., Kramer, H. et al. (2012). Racial
controlled trial of artemether or quinine in Vietnamese differences in the incidence of chronic kidney disease.
adults with severe falciparum malaria. N. Engl. J. Med. Clin. J. Am. Soc. Nephrol. 7 (1): 101–107.
335 (2): 76–83. 57 Freedman, B.I. and Skorecki, K. (2014). Gene-gene and
44 Wesseling, C., Crowe, J., Hogstedt, C. et al. (2014). gene-environment interactions in Apolipoprotein L1
Resolving the enigma of the Mesoamerican nephropathy: gene-associated nephropathy. Clin. J. Am. Soc. Nephrol. 9
a research workshop summary. Am. J. Kidney Dis. 63 (3): (11): 2006–2013.
396–404. 58 Mix, J., Elon, L., Mac, V. et al. (2017). Hydration status,
45 Correa-Rotter, R., Wesseling, C., and Johnson, R.J. (2014). kidney function and kidney injury in Florida agricultural
CKD of unknown origin in Central America: the case for workers. J. Occup. Environ. Med. 60(5):e253–e260.
a Mesoamerican nephropathy. Am. J. Kidney Dis. 63 (3): 59 Moyce, S., Joseph, J., Tancredi, D. et al. (2016).
506–520. Cumulative incidence of acute kidney injury in
46 Obrador, G.T., Villa, A.R., Olvera, N. et al. (2013). California’s agricultural workers. J. Occup. Environ. Med.
Longitudinal analysis of participants in the KEEP 58 (4): 391–397.
Mexico’s chronic kidney disease screening program. Arch.
60 Abraham, G., Varughese, S., Thandavan, T. et al. (2016).
Med. Res. 44 (8): 650–654.
Chronic kidney disease hotspots in developing countries
47 Nanayakkara, S., Senevirathna, S., Abeysekera, T. et al.
in South Asia. Clin. Kidney J. 9 (1): 135–141.
(2014). An integrative study of the genetic, social and
61 Martin-Cleary, C. and Ortiz, A. (2014). CKD hotspots
environmental determinants of chronic kidney disease
around the world: where, why and what the lessons are.
characterized by tubulointerstitial damages in the north
A CKJ review series. Clin. Kidney J. 7 (6): 519–523.
central region of Sri Lanka. J. Occup. Health 56 (1):
62 Wuttke, M. and Köttgen, A. (2016). Insights into kidney
28–38.
diseases from genome-wide association studies. Nat. Rev.
48 Jayatilake, N., Mendis, S., Maheepala, P., and Mehta, F.R.
Nephrol. 12 (9): 549–562.
(2013). Chronic kidney disease of uncertain aetiology:
63 Genovese, G., Friedman, D.J., Ross, M.D. et al. (2010).
prevalence and causative factors in a developing country.
Association of trypanolytic ApoL1 variants with kidney
BMC Nephrol. 14 (1) http://bmcnephrol.biomedcentral.
disease in African Americans. Science 329 (5993):
com/articles/10.1186/1471-2369-14-180 (2013).
841–845.
49 Reddy, D.V. and Gunasekar, A. (2013). Chronic kidney
disease in two coastal districts of Andhra Pradesh, India: 64 Freedman, B.I., Kopp, J.B., Langefeld, C.D. et al. (2010).
role of drinking water. Environ. Geochem. Health 35 (4): The Apolipoprotein L1 (APOL1) gene and nondiabetic
439–454. nephropathy in African Americans. J. Am. Soc. Nephrol.
50 Hoy, W.E. (2014). Kidney disease in aboriginal 21 (9): 1422–1426.
Australians: a perspective from the Northern Territory. 65 Ashley-Koch, A.E., Okocha, E.C., Garrett, M.E. et al.
Clin. Kidney J. 7 (6): 524–530. (2011). MYH9 and APOL1 are both associated with sickle
51 Kim, S., Macaskill, P., Hodson, E.M. et al. (2017). cell disease nephropathy: MYH9, APOL1 and SCD
Beginning the trajectory to ESKD in adult life: nephropathy. Br. J. Haematol. 155 (3): 386–394.
albuminuria in Australian aboriginal children and 66 Larsen, C.P., Beggs, M.L., Saeed, M., and Walker, P.D.
adolescents. Pediatr. Nephrol. 32 (1): 119–129. (2013). Apolipoprotein L1 risk variants associate with
52 McDonald, S.P., Maguire, G.P., and Hoy, W.E. (2003). systemic lupus erythematosus-associated collapsing
Renal function and cardiovascular risk markers in a Glomerulopathy. J. Am. Soc. Nephrol. 24 (5): 722–725.
remote Australian aboriginal community. Nephrol. Dial. 67 Luyckx, V.A. and Brenner, B.M. (2015). Birth weight,
Transpl. 18 (8): 1555–1561. malnutrition and kidney-associated outcomes – a global
53 Bui-Klimke, T. and Wu, F. (2014). Evaluating weight of concern. Nat. Rev. Nephrol. 11 (3): 135–149.
evidence in the mystery of Balkan endemic nephropathy. 68 Lackland, D.T., Bendall, H.E., Osmond, C. et al. (2000).
Risk Anal. 34 (9): 1688–1705. Low birth weights contribute to high rates of early-onset
Reference 83
chronic renal failure in the Southeastern United States. income countries in 2010. Lancet Glob. Health 1 (1):
Arch. Intern. Med. 160 (10): 1472–1476. e26–e36.
69 White, S.L., Perkovic, V., Cass, A. et al. (2009). Is low 82 Manrai, A.K., Funke, B.H., Rehm, H.L. et al. (2016).
birth weight an antecedent of CKD in later life? A Genetic misdiagnoses and the potential for health
systematic review of observational studies. Am. J. Kidney disparities. N. Engl. J. Med. 375 (7): 655–665.
Dis. 54 (2): 248–261. 83 Vassalotti, J.A., Li, S., McCullough, P.A., and Bakris, G.L.
70 Eriksson, J.G., Osmond, C., Kajantie, E. et al. (2006). (2010). Kidney early evaluation program: a community-
Patterns of growth among children who later develop based screening approach to address disparities in
type 2 diabetes or its risk factors. Diabetologia 49 (12): chronic kidney disease. Semin. Nephrol. 30 (1): 66–73.
2853–2858. 84 Lu, C., Zhao, H., Xu, G. et al. (2010). Prevalence and risk
71 Phillips, D.I., Barker, D.J., Hales, C.N. et al. (1994). factors associated with chronic kidney disease in a Uygur
Thinness at birth and insulin resistance in adult life. adult population from Urumqi. J. Huazhong Univ. Sci.
Diabetologia 37 (2): 150–154. Technolog. Med. Sci. 30 (5): 604–610.
72 Kwong WY, Wild AE, Roberts P, Willis AC, Fleming TP. 85 Liu, Q., Li, Z., Wang, H. et al. (2012). High prevalence and
(2000). Maternal undernutrition during the associated risk factors for impaired renal function and
preimplantation period of rat development causes urinary abnormalities in a rural adult population from
blastocyst abnormalities and programming of postnatal southern China. PLoS One 7 (10): e47100.
hypertension. Development. 27(19):4195–202. 86 Hwang, S.-J., Tsai, J.-C., and Chen, H.-C. (2010).
73 Andersen, L.G., Angquist, L., Eriksson, J.G. et al. (2010). Epidemiology, impact and preventive care of chronic
Birth weight, childhood body mass index and risk of kidney disease in Taiwan. Nephrology 15 (Suppl 2): 3–9.
coronary heart disease in adults: combined historical 87 George, C., Mogueo, A., Okpechi, I. et al. (2017). Chronic
cohort studies. PLoS One 5 (11): e14126. kidney disease in low-income to middle-income
74 Osmond, C., Kajantie, E., Forsén, T.J. et al. (2007). Infant countries: the case for increased screening. BMJ Glob.
growth and stroke in adult life: the Helsinki birth cohort Health 2 (2): e000256.
study. Stroke 38 (2): 264–270. 88 Levey, A.S. and Coresh, J. (2012). Chronic kidney disease.
75 Duggleby, S.L. and Jackson, A.A. (2002). Higher weight at Lancet 379 (9811): 165–180.
birth is related to decreased maternal amino acid 89 Kidney Disease Improving Global Outcomes (KDIGO)
oxidation during pregnancy. Am. J. Clin. Nutr. 76 (4): CKD Work Group (2013). KDIGO 2012 clinical practice
852–857. guideline for the evaluation and management of chronic
76 Duggleby, S.L. and Jackson, A.A. (2001). Relationship of kidney disease. Kidney Int. Suppl. 3: 1–150.
maternal protein turnover and lean body mass during 90 Cueto-Manzano, A.M., Martínez-Ramírez, H.R., and
pregnancy and birth length. Clin. Sci. (Lond.) 101 (1): Cortés-Sanabria, L. (2010). Management of chronic
65–72. kidney disease: primary health-care setting, self-care and
77 Uauy, R., Kain, J., and Corvalan, C. (2011). How can the multidisciplinary approach. Clin. Nephrol. 74 (Suppl 1):
developmental origins of health and disease (DOHaD) S99–S104.
hypothesis contribute to improving health in developing 91 Echouffo-Tcheugui, J.B. and Kengne, A.P. (2012). Risk
countries? Am. J. Clin. Nutr. 94 (6 Suppl): 1759S–1764S. models to predict chronic kidney disease and its
78 Yajnik, C.S. (2014). Transmission of obesity-adiposity and progression: a systematic review. PLoS Med. 9 (11):
related disorders from the mother to the baby. Ann. Nutr. e1001344.
Metab. 64 (Suppl 1): 8–17. 92 Eastwood, J.B., Kerry, S.M., Plange-Rhule, J. et al.
79 Adair, L.S., Fall, C.H.D., Osmond, C. et al. (2013). (2010). Assessment of GFR by four methods in adults in
Associations of linear growth and relative weight gain Ashanti, Ghana: the need for an eGFR equation for lean
during early life with adult health and human capital in African populations. Nephrol. Dial. Transplant. 25 (7):
countries of low and middle income: findings from five 2178–2187.
birth cohort studies. Lancet 382 (9891): 525–534. 93 Shephard, M.D. (2011). Point-of-care testing and
80 Hsu, C.W., Yamamoto, K.T., Henry, R.K. et al. (2014). creatinine measurement. Clin. Biochem. Rev. 32 (2):
Prenatal risk factors for childhood CKD. J. Am. Soc. 109–114.
Nephrol. 25 (9): 2105–2111. 94 McTaggart, M.P., Newall, R.G., Hirst, J.A. et al. (2014).
81 Lee, A.C., Katz, J., Blencowe, H. et al. (2013). National Diagnostic accuracy of point-of-care tests for detecting
and regional estimates of term and preterm babies born albuminuria: a systematic review and meta-analysis. Ann.
small for gestational age in 138 low-income and middle- Intern. Med. 160 (8): 550–557.
95 Komenda, P., Ferguson, T.W., Macdonald, K. et al. (2014). 99 Dirks, J.H. and Levin, N.W. (2006). Dialysis rationing in
Cost-effectiveness of primary screening for CKD: a South Africa: a global message. Kidney Int. 70 (6):
systematic review. Am. J. Kidney Dis. 63 (5): 789–797. 982–984.
96 Arogundade, F.A., Sanusi, A.A., Hassan, M.O., and 100 Young, B.A. (2013). Health literacy in nephrology: why
Akinsola, A. (2011). The pattern, clinical characteristics is it important? Am. J. Kidney Dis. 62 (1): 3–6.
and outcome of ESRD in Ile-Ife, Nigeria: is there a
change in trend? Afr. Health Sci. 11 (4): 594–601. 101 Taylor, D.M., Fraser, S., Dudley, C. et al. (2017). Health
97 Sakhuja, V. and Sud, K. (2003). End-stage renal disease in literacy and patient outcomes in chronic kidney disease:
India and Pakistan: burden of disease and management a systematic review. Nephrol. Dial. Transplant. Vol 33,
issues. Kidney Int. Suppl. 83: S115–S118. Issue 9, 545–1558.
98 Rajapurkar, M.M., John, G.T., Kirpalani, A.L. et al. 102 Obrador, G.T., Rubilar, X., Agazzi, E., and Estefan, J.
(2012). What do we know about chronic kidney disease in (2016). The challenge of providing renal replacement
India: first report of the Indian CKD registry. therapy in developing countries: the Latin American
BMC Nephrol 13, 10 http://bmcnephrol.biomedcentral. perspective. Am. J. Kidney Dis. 67 (3): 499–506.
com/articles/10.1186/1471-2369-13-10.
85
Acute Kidney Injury

Part 2
87
Overview/Definition, Classification, and Epidemiology of Acute Kidney Disease

Kathleen D. Liu1 and Jay L. Koyner2
1
Division of Nephrology, Departments of Medicine and Anesthesia, University of California, San Francisco, CA, USA
2
Section of Nephrology, Department of Medicine, University of Chicago, Chicago, IL, USA
I ntroduction the Risk, Injury, Failure, Loss, and End-stage (RIFLE) clas-
sification system for AKI (Table 7.1) [1]. A major goal was
Acute kidney injury (AKI), previously known as acute to ensure that studies in the literature were comparable:
renal failure, refers to a sudden reduction in kidney func- because of differences in the definition of acute renal fail-
tion. This acute reduction in kidney function typically ure used in prior studies (e.g. ranging from a 0.5 mg/dl rise
manifests as a fall in urine output and/or a rise in blood in creatinine to the need for renal replacement therapy
urea nitrogen and creatinine, two blood biomarkers that [RRT]), the incidence and outcomes of AKI varied widely.
are filtered and excreted by the kidney and are readily That is, AKI defined as a 0.5 mg/dl increase in creatinine
measured in the clinical laboratory. While initial descrip- was much more common and associated with better out-
tions of AKI focused on the most severe forms of disease, comes than AKI defined as the need for RRT.
over the past 15 years there has been a growing recognition Importantly, the RIFLE criteria defined three levels of
that small changes in serum creatinine are associated with decreased kidney function (risk, injury, failure; Table 7.1)
adverse outcomes, including longer hospital length of stay along with loss of kidney function (persistent AKI, defined
and death. Thus, a major focus of the field for the past as the need for RRT for 4 weeks) and end-stage renal dis-
15 years has been the development of consensus defini- ease. The rationale for defining different levels of disease
tions and approaches to defining AKI. These have been severity was that earlier AKI recognition might allow for
important to moving the field forward because they have earlier treatment (including supportive care and nephro-
allowed comparison of studies across settings and disease toxin avoidance, for example), recognizing that there is an
subtypes. inherent tradeoff between sensitivity and specificity that
In this chapter, we will review the consensus definitions may occur when small changes in serum creatinine are
for AKI, along with recent work to develop consensus used to define AKI. In addition, the RIFLE criteria defined
regarding acute kidney disease (AKD) that does not meet AKI using changes in either serum creatinine or urine out-
the technical definition of AKI and recovery from AKI. We put, with the rationale that these are both markers of
will also review common settings in which AKI may occur decreased kidney function. However, it should be noted
and the epidemiology of the disease, along with the poten- that studies prior to the development of the RIFLE criteria
tial role of novel biomarkers in disease recognition and almost universally used serum creatinine, not urine out-
classification. put, to define acute renal failure. In addition, at the time
the RIFLE criteria were published, the literature on the
predictive value of graded reductions in urine output for
O
verview/Definition adverse outcomes was limited, with no data to support the
concept that the urine output criteria proposed were in any
In 2004, the Acute Dialysis Quality Initiative (ADQI), a conway “equivalent” to the creatinine-based AKI criteria.
sensus group of intensivists and nephrologists, proposed Subsequent studies have suggested that AKI defined by
88 Overview/Definition, Classification, and Epidemiology of Acute Kidney Disease
Table 7.1 Comparison of consensus AKI definitions.
Urine output
(common to all
Stage three) Creatinine (KDIGO) Creatinine (AKIN) RIFLE category Creatinine (RIFLE)
1 <0.5 ml/kg/h for 1.5–1.9 times baseline 1.5–2 times baseline or Risk 1.5 increase in SCr within
6–12 h over 7 days or 0.3 mg/ 0.3 mg/dl absolute increase 7 days, sustained for 24 h
dl absolute increase within 48 h
over 48 h
2 <0.5 ml/kg/h for 2.0–2.9 times baseline >2–3 times baseline Injury 2 increase in SCr
12 h
3 <0.3 ml/kg/h for 3.0 times baseline, or >3.0 times baseline, or Failure 3.0 increase in SCr or
24 h or anuria increase to 4.0 mg/dl increase to 4.0 mg/dl (with increase to 4.0 mg/dl (with
for 12 h or initiation of RRT an increase of 0.5 mg/dl) or an increase of 0.5 mg/dl) or
initiation of RRT initiation of RRT
Loss Complete loss of kidney

function >4 weeks
End-stage End-stage renal disease

kidney disease >3 months
Urine output criteria are common to all three. The ADQI RIFLE definition incorporated three categories of injury and two outcomes that varied
by severity. Outcomes were eliminated from the subsequent Acute Kidney Injury Network (AKIN) and KDIGO definitions. The AKIN definition
incorporated smaller changes in SCr and the KDIGO definition added more definitive time frames to the definition.
urine output criteria is much more common than AKI Network (AKIN) [6]. These studies included a large retro-
defined by creatinine criteria. Isolated AKI by urine output spective study by Chertow et al. [7], which demonstrated
criteria is generally associated with a significant increase in that a 0.3–0.4 mg/dl increase in serum creatinine was asso-
the risk of death compared to those with no AKI, but gen- ciated with a 1.7-fold higher odds of death in multivariable
erally lower risk compared to those with isolated AKI by analysis adjusting for age, gender, chronic kidney disease,
creatinine criteria; those with AKI by both urine output and diagnosis codes. Thus, a 0.3 mg/dl increase in serum
and creatinine criteria have the greatest risk of death [2–5]. creatinine over 48 hours was added to the definition of
However, despite these caveats, creating consensus criteria mild (Stage 1) AKI (see Table 7.1). Importantly, both “loss”
for both changes in serum creatinine and urine output to and “end-stage kidney disease” were recognized to be out-
define AKI was a critical step to moving the field forward. comes of AKI and dropped from the definition.
Finally, the RIFLE criteria recognized that relative changes In 2012, Kidney Disease Improving Global Outcomes
in serum creatinine are important, that is, in someone with (KDIGO) comprehensively reviewed the field of AKI to
normal kidney function (e.g. baseline creatinine of 0.8 mg/ develop a series of consensus guidelines focused on AKI
dl), an increase in serum creatinine to 1.6 mg/dl represents diagnosis and treatment [8]. As part of these guidelines,
a larger change in renal function than a rise of the same the RIFLE and AKIN criteria were combined into the
absolute magnitude in an individual with underlying KDIGO definition (Table 7.1). In addition, the concept of
chronic kidney disease (e.g. a rise from 4.0 to 4.8 mg/dl). AKD was introduced. This comprises AKI, along with a
Therefore, inherent in the RIFLE criteria is knowledge of a decrease in eGFR <60 ml/min/1.73 m2, decrease in glo-
baseline creatinine. In the absence of a baseline creatinine, merular filtration rate (GFR) by 35% or increase in
the RIFLE criteria proposed the use of a creatinine that serum creatinine (SCr) by >50% for less than 3 months.
represents an estimated glomerular filtration rate (eGFR) Subsequently, a staging system for AKD analogous to that
of 75 ml/min/1.73 m2 by the Modification of Diet in Renal for AKI has been proposed, though this definition has not
Disease (MDRD) estimating equation (also referred to as been as well validated [9]. Along the same lines, criteria
MDRD-75). to define and stage kidney recovery have yet to be well
Following a number of studies suggesting that small standardized and are, when creatinine is used as the
absolute changes in serum creatinine are also associated measure of kidney function, potentially confounded by
with an increased risk of adverse outcomes, the RIFLE cri- changes in nutritional status and muscle mass (see below
teria were modified slightly by the Acute Kidney Injury for additional discussion).
Classification of AK 89
Classification of AKI Broadly speaking, AKI can be subdivided into prerenal,

intrarenal, and postrenal etiologies (Table 7.2). Prerenal
Since at present the majority of care for AKI is supportive physiology refers to any cause of decreased kidney perfu-
and focused on the underlying etiology (e.g. volume resus- sion, which may be due to volume depletion itself.
citation for the patient with volume depletion and “prere- Alternatively, decreased kidney perfusion may occur due
nal” physiology), it is imperative to search for and identify to other pathologic conditions, including heart failure,
the primary cause of AKI [10]. Careful history-taking, liver disease, nephrotic syndrome, and impaired venous
chart review, and physical examination remain the funda- return in the setting of abdominal compartment syn-
mental tenets of the workup. Specific causes of AKI that drome. Intrarenal failure may be due to damage to the vas-
may warrant cause-specific therapy include acute intersti- culature, glomerulus, tubules, or interstitium. Finally,
tial nephritis (withdrawal of offending medication, ster- postrenal AKI is due to obstruction anywhere along the
oids), vasculitis/acute glomerulonephritis (consideration collecting system. In particular in those who have had a
of immunosuppression), thrombotic microangiopathy kidney transplant or who have a solitary kidney, it is essen-
(consideration of pheresis or monoclonal antibody therapy tial to assess for and rule out obstruction. In hospitalized
targeting the complement pathway), and urinary tract patients, the most common form of AKI is acute tubular
obstruction (relief/decompression of obstruction). necrosis (intrarenal) due to ischemia, nephrotoxins, or
Table 7.2 Classification of AKI by etiology.
Decreased renal perfusion, e.g. “prerenal” states

Hypovolemia Increased losses; poor oral intake
Reduced cardiac output Heart failure, cardiac tamponade, massive pulmonary embolism
Renal vasomodulation/shunting Medications (NSAIDs, ACE/ARB, cyclosporine, iodinated contrast), hypercalcemia,
hepatorenal syndrome, abdominal compartment syndrome
Systemic vasodilation Sepsis, hepatorenal syndrome
Intrarenal causes
Vascular Renal artery stenosis
Microvascular Thrombotic microangiopathies, cholesterol emboli
Glomerular Rapidly progressive (crescentic) glomerulonephritis
Immune complex diseases: IgA nephropathy, post-infectious, lupus, mixed
cryoglobuminemia with MPGN
Pauci-immune glomerulonephritis: ANCA-associated and ANCA-negative
Some causes of nephrotic range proteinuria may associate with AKI: in particular light-chain
cast nephropathy
Tubulointerstitium Acute interstitial nephritis: medications, infection, lymphoproliferative disease
Pigment nephropathy: rhabdomyolysis (myoglobin), massive hemolysis (hemoglobin)
Crystal nephropathy: uric acid (tumor lysis), acyclovir, sulfonamides, protease inhibitors
(indinavir, atazanavir), methotrexate, ethylene glycol, acute phosphate nephropathy, oxalate
nephropathy
Myeloma-associated AKI (cast nephropathy)
Acute tubular necrosis: ischemia (shock), medications
Sepsis-associated AKI
Postrenal
Bladder outlet Benign prostatic hypertrophy, strictures, blood clots
Ureteral Bilateral obstruction (or unilateral with one kidney): stones, malignancy, retroperitoneal
fibrosis
Renal pelvis Papillary necrosis (NSAIDs), stones
Etiologies of AKI can be broadly divided into prerenal, intrarenal, and postrenal causes and then further subdivided.
sepsis. At present, the classification of AKI is based pre- Limitations of SCr to Define AKI
dominantly on history/clinical context and may be sup-
ported by laboratory testing. The laboratory studies will With regards to defining AKI itself, in individuals with lim-
depend on the suspected etiology of AKI and may include ited muscle mass (e.g. older adults, those with end-stage
a serologic workup for vasculitis or a peripheral blood liver disease), SCr may overestimate eGFR because it is a
smear to assess for hemolysis if thrombotic microangiopa- breakdown product of creatine derived from muscle.
thy is suspected. Urinalysis and urine microscopy may be Across the world, SCr measurements are now harmonized
helpful. In particular, the presence of casts in the urine to IDMS standard materials, so there is not much variation
may provide insight into the cause of AKI (Table 7.3). in the laboratory measurement of serum creatinine.
However, casts are not required for these diagnoses. However, at high levels, chromogens, both endogenous
Urinalysis scores which quantify the number of renal tubu- and exogenous (e.g. bilirubin, uric acid, ascorbic acid, some
lar epithelial cells or cast elements per microscopic fields cephalosporins, and ketones), may interfere with the cre-
have been repeatedly shown to be specific for AKI and cor- atinine assay [8]. Thus, there has been tremendous interest
relate with AKI severity (in this case, with acute tubular in identifying other endogenous filtration markers (e.g.
necrosis as the etiology of AKI) [11–13]. Widespread functional markers) that are not associated with muscle
acceptance has been hampered by the lack of large-scale mass to define chronic kidney disease (e.g. cystatin C [17]).
validation and the relatively modest sensitivity for urine However, changes in markers including cystatin C have
microscopy for detecting AKI [11, 14–16]. Table 7.4 sum- not been as well-studied in the context of AKI, so there is
marizes several of the published urinalysis risk scores. no consensus regarding absolute biomarker levels or
changes in biomarker levels that could be used to define
Table 7.3 Overview of urine cast types with typical AKI. Finally, there is growing interest in other filtration
associations. markers, including proenkephalin [18], but more studies
are needed.
Waxy Chronic kidney disease
Hyaline Prerenal azotemia
Granular Acute tubular necrosis Biomarkers of AKI/AKD
Dark/muddy brown Acute tubular necrosis
While there have been numerous publications on the role
White blood cells Interstitial nephritis
of novel biomarkers in the risk stratification, diagnosis,
Red blood cells Glomerulonephritis
prognosis, and long-term outcomes of AKI, to date there
has not been the formal inclusion of these novel tests into
Table 7.4 Review of urine microscopy scoring systems for the diagnostic criteria and none have gained wide clinical
diagnosis of AKI due to acute tubular necrosis. acceptance. Regardless, these investigations have helped to
focus the ideal characteristics of a novel biomarker specific
Study Scoring system for AKI/AKD. Such biomarkers should (i) be organ specific
and allow differentiation between the traditional causes of
Chawla Grade 1: No casts or RTE AKI (intrarenal, prerenal, and postrenal), (ii) detect AKI
et al. [11] Grade 2: At least 1 cast or RTE but <10% of LPF early in the course and predict the course/future implica-
Grade 3: Many casts or RTEs (between 10% and tions of AKD, (iii) identify the source/nature of AKI, (iv) be
90% of LPF)
site-specific and able to inform pathologic changes in vari-
Grade 4: Sheet of muddy brown casts and RTEs
in >90% of LPF ous segments of renal tubules and correlate with the histol-
Perazella 0 points: No casts or RTE seen
ogy on kidney biopsies, and (v) be easily and reliably
et al. [12] 1 point each: 1–5 casts per LPF or 1–5 RTEs per measured in a noninvasive, rapid manner, which implies
HPF that the marker is stable in its matrix and relatively inex-
2 points each: 6 casts per LPF or 6 RTEs per pensive to measure.
HPF The concept of functional versus damage markers of kid-
Bagshaw 0 points: No casts or RTE seen ney injury is similar to diagnostic models from other
et al. [14] 1 point each: 1 casts or 1 RTEs per HPF aspects of medical care. It is common to see functional bio-
2 points each: 2–4 casts or RTEs per HPF markers (e.g. electrocardiograms or imaged-based stress
3 points each: 5 casts or 5 RTEs per HPF tests) paired with damage biomarkers (creatine kinase, tro-
RTE, renal tubule epithelial cells; LPF, low power field; HPF, high ponins). An ADQI consensus statement recommended
power field. that the combination of functional and damage biomarkers
Biomarkers of AKI/AK 91
would be helpful to define AKI. (Figure 7.1) [19]. By strat- The incidence and duration of these outcomes were simi-
ifying patients across these two domains, a novel cohort lar in the NGAL-positive/creatinine-negative and NGAL-
with “subclinical AKI” can be identified: those with a negative/creatinine-positive cohorts. This concept of
change in damage biomarkers in the absence of change in damage biomarker positive/functional biomarker nega-
functional biomarkers [20, 21]. This group of patients, tive patients being at increased risk for adverse outcomes
who would not be detected by creatinine or urine output- has been validated in other cohorts [22, 23].
based KDIGO criteria, appears to have the same risk for In 2014, the combination of urine tissue inhibitor of
severe AKI and adverse outcomes (e.g. inpatient mortal- metalloproteinase-2 and insulin-like growth factor bind-
ity) as patients who have a change in functional markers ing protein-7 (TIMP2*IGFBP7) became the first novel AKI
in the absence of change in damage biomarkers (more biomarker approved by the Food and Drug Administration
traditionally thought of as having prerenal azotemia). In a (FDA). The intended use of TIMP-2*IGFBP7 is to assess
pooled, prospective observational cohort of 2322 subjects the risk of developing KDIGO Stage 2 AKI over the next
(1452 with cardiac surgery and 870 with critical illness), 12 hours in high-risk critically ill patients [24, 25]. Some
Haase et al. designated subjects as neutrophil gelatinase studies have shown improved outcomes when supportive
associated lipocalin (NGAL)-positive or NGAL-negative care bundles are applied to patients identified by high lev-
and creatinine-positive or creatinine-negative. They dem- els of TIMP-2*IGFBP7 [26, 27], but TIMP-2*IGFBP7 has
onstrated that individuals who were NGAL-positive/ not yet gained widespread acceptance. In other parts of
creatinine-negative needed RRT more than 16 times more the world (e.g. Europe and Japan, respectively), urinary
frequently than those who were NGAL-negative/ NGAL and liver-fatty acid binding protein (L-FABP)
creatinine-negative (odds ratio 16.4, 95% confidence can be measured in clinical practice as measures of
interval 3.6–76.9) [20]. NGAL-positive/creatinine- AKI, though clear thresholds for AKI have not been
negative patients were also in ICU and hospital longer developed.
and more likely to experience inpatient mortality com- As previously mentioned, there has been extensive inves-
pared to NGAL-negative/creatinine-negative subjects. tigation into several biomarkers of kidney injury in several
No Damage / Damage Present /

Biomarker Negative Biomarker Positive
“Sub-
No Functional
clinical
Change /
AKI”
Creatinine No Functional Damage
Negative Changes or without loss of
damage function
Functional
Change / Loss of function
Creatinine Damage with
without
Positive loss of function
damage
“Pre-
“Intrinsic
Renal
AKI”
Azotemia”
Figure 7.1 The Acute Dialysis Quality Initiative recommended criteria for defining acute kidney injury (AKI) in terms of changes in
biomarkers of kidney function (SCr) and biomarkers of kidney damage/injury. This paradigm allows for the combination of injury
biomarkers with SCr and has proven useful in the discrimination of patients with AKI and resulted in the concept of “subclinical AKI.”
Source: Adapted from [19].
clinical settings, including after cardiac surgery, critical ill- Stage 3 and the need for RRT [33–35]. As a functional bio-
ness/ICU admission, or admission to the emergency marker of proximal tubule and loop of Henle function, the
department. In each of these settings biomarkers have FST has outperformed several damage biomarkers when
been examined to determine their association with a vari- compared directly [33, 36]. With the renal angina index,
ety of short and long-term outcomes, including but not patient demographics and other risk factors (comorbidi-
limited to the early detection of AKI, the temporal nature ties, procedure/medication exposure) are combined with
of the AKI (transient versus persistent), the detection of small changes in serum creatinine and/or urine output and
severe (e.g. dialysis requiring) AKI, and inpatient mortal- the presence of volume overload to identify patients where
ity. It should not be surprising that a single biomarker has damage biomarkers should be measured and may identify
not been associated with all of these outcomes across mul- those with early AKI or at high risk of AKI [37]. Multiple
tiple settings, given the biological heterogeneity that under- novel risk scores (some of which include AKI biomarkers)
lies the diverse etiologies of AKI. However, this does not have also been proposed [38, 39]. However, to date no
diminish the importance of their performance for individ- single risk score (or biomarker) has been universally
ual endpoints in specific clinical settings. Table 7.5 sum- validated.
marizes the current state of the ability of biomarkers to In addition to this explosion of more traditional risk
detect clinical endpoints related to AKI in a variety of set- scores, there is increasing interest in developing complex
tings. We anticipate continued investigation of these bio- risk assessment models which can utilize all of the infor-
markers as well as novel tests over the course of the next mation that is available in the medical record to identify
decade. patients at risk for AKI [40–43]. The appeal of such risk
assessment models is that they can incorporate the most
up-to-date data to evaluate changes in risk. While such
pidemiology of AKI: Identification
E electronic biomarkers of AKI represent a potential future
of High-risk Patients option, they will require wide-scale validation prior to
becoming part of the standard of care. In the near future,
In terms of identification of high-risk patients, both electronic risk scores and biochemical biomarkers (in par-
chronic kidney disease and the presence of proteinuria/ ticular, damage biomarkers) are likely to be incorporated in
albuminuria (as dip-stick, protein to creatinine ratio AKI/AKD definitions.
or albumin to creatinine ratio) have been strongly
implicated as risk factors for AKI [29–31]. However, in
particular in the perioperative setting, preoperative risk
stratification, while helpful, does not necessarily robustly pidemiology of AKI: Incidence and
E
predict postoperative AKI because AKI typically is due to Outcomes
a complicated operative course in a high-risk patient. For
example, in a classic analysis, using data from more than The exact incidence and outcomes of AKI will vary
33 000 patients at the Cleveland Clinic, a scoring system depending on the definition used and clinical setting. For
to identify those at high risk of dialysis requiring AKI was example, more severe AKI requiring dialysis in sicker
developed, with a robust area under the curve of 0.81 and patients (e.g. ICU patients) is associated with a greater
0.82 in the test and validation datasets, respectively [32]. risk of death than stage 1 KDIGO AKI in all hospitalized
However, given the very large size of the low risk group, patients. Although AKI has long been recognized to asso-
more dialysis events occurred in the lower risk groups ciate with a high risk of short-term adverse outcomes
than the highest risk group. Thus, while such scores may such as death, more recently it has become clear that AKI
be useful for preoperative risk assessment and for enrich- is associated with an increased risk of a number of long-
ment of populations for clinical trials, they are of some- term adverse outcomes [44, 45], including de novo
what limited utility because they typically do not chronic kidney disease/chronic kidney disease progres-
incorporate risk based on the ongoing clinical course/cur- sion, cardiovascular disease, and death. Again, these
rent situation. associations are likely graded, with greater risk associated
Thus, there has been significant interest in risk stratifica- with more severe AKI. That said, even fully recovered
tion to identify patients who may benefit from novel/dam- AKI (based on serum creatinine) appears to be associated
age biomarker assessment. For example, the urine output with an increased risk of adverse outcomes, including
following a protocolized intravenous dose of furosemide in chronic kidney disease and death [46]. Thus, in AKI sur-
the setting of stage 1 or 2 AKI (furosemide stress test, FST) vivors, clinicians should be aware of the increased risk of
has been shown to reliably predict AKI progression to long-term adverse outcomes.
Table 7.5 Biomarker performance in detecting AKI from multicenter studies at a variety of clinical timepoints.
Perioperative AKI Critically ill Emergency room
Early Type of AKI Type of AKI

Preoperative postoperative Long-term Early diagnosis (transient vs. Early diagnosis (transient vs.
AKI risk AKI AKI progression mortality of AKI intrinsic) Need for RRT of AKI intrinsic)
Urine NGAL N/A + − + + + + + +
Blood NGAL − + + + − ? − ? ?
Blood CysC + + − ? + + + ? ?
Urine CysC N/A − − − + + + + +
Proenkephalin ? ? ? ? + + + ? ?
Urine IL-18 N/A + + + + + + + +

Urine KIM-1 N/A + − + + − − + +
Urine L-FABP N/A − − + ? ? − + +
TIMP-2 IGFBP-7 N/A + + + + ? + + ?

Urine protein/ + + + + ? ? ? ? ?
albumin
NGAL, neutrophil gelatinase associated lipocalin; CysC, cystatin C; IL-18, interleukin-18; KIM-1, kidney injury molecule-1; L-FABP, liver fatty acid binding protein; TIMP-2 IGFBP-7, tissue
injury metalloprotease 2 insulin-like growth factor binding protein-7.
+, data published displays the ability to detect this aspect of AKI.
−, data published does not display the ability to detect this aspect of AKI.
?, no large multicenter data published on this biomarker/aspect of AKI.
N/A, not applicable either because (a) biomarkers of tubular injury have no role in preoperative risk screening or (b) serum creatinine is intrinsic to the definitions of AKI being tested.
Source: Adapted and expanded from Koyner and Parikh et al. [28]
0005152395.INDD 93 09-12-2022 12:52:52

R
eferences
1 Bellomo, R., Ronco, C., Kellum, J. et al. (2004). Acute 14 Bagshaw, S.M., Haase, M., Haase-Fielitz, A. et al. (2012).
renal failure -definition, outcome measures, animal A prospective evaluation of urine microscopy in septic
models, fluid therapy and information technology needs: and non-septic acute kidney injury. Nephrol. Dial.
the second international consensus conference of the Transplant. 27 (2): 582–588.
acute dialysis quality initiative group. Crit. Care 8 (4): 15 Hall, I.E., Coca, S.G., Perazella, M.A. et al. (2011). Risk of
R204–R212. poor outcomes with novel and traditional biomarkers at
2 Kellum, J.A., Sileanu, F.E., Murugan, R. et al. (2015). clinical AKI diagnosis. Clin. J. Am. Soc. Nephrol. 6 (12):
Classifying AKI by urine output versus serum creatinine 2740–2749.
level. J. Am. Soc. Nephrol. 26 (9): 2231–2238. 16 Perazella, M.A., Coca, S.G., Kanbay, M. et al. (2008).
3 Quan, S., Pannu, N., Wilson, T. et al. (2016). Prognostic Diagnostic value of urine microscopy for differential
implications of adding urine output to serum creatinine diagnosis of acute kidney injury in hospitalized patients.
measurements for staging of acute kidney injury after Clin. J. Am. Soc. Nephrol. 3 (6): 1615–1619.
major surgery: a cohort study. Nephrol. Dial. Transplant. 17 Shlipak, M.G., Mattes, M.D., and Peralta, C.A. (2013).
31 (12): 2049–2056. Update on cystatin C: incorporation into clinical practice.
4 Jin, K., Murugan, R., Sileanu, F.E. et al. (2017). Intensive Am. J. Kidney Dis. 62 (3): 595–603.
monitoring of urine output is associated with increased 18 Hollinger, A., Wittebole, X., Francois, B. et al. (2018).
detection of acute kidney injury and improved outcomes. Proenkephalin A 119–159 (Penkid) is an early biomarker
Chest 152 (5): 972–979. of septic acute kidney injury: the kidney in sepsis and
5 Mizota, T., Yamamoto, Y., Hamada, M. et al. (2017). septic shock (kid-SSS) study. Kidney Int. Rep. 3 (6):
Intraoperative oliguria predicts acute kidney injury after 1424–1433.
major abdominal surgery. Br. J. Anaesth. 119 (6): 1127–1134. 19 Endre, Z.H., Kellum, J.A., Di Somma, S. et al. (2013).
6 Mehta, R.L., Kellum, J.A., Shah, S.V. et al. (2007). Acute Differential diagnosis of AKI in clinical practice by
kidney injury network: report of an initiative to improve functional and damage biomarkers: workgroup
outcomes in acute kidney injury. Crit. Care 11 (2): R31. statements from the tenth acute dialysis quality initiative
7 Chertow, G.M., Burdick, E., Honour, M. et al. (2005). consensus conference. Contrib. Nephrol. 182: 30–44.
Acute kidney injury, mortality, length of stay, and costs in 20 Haase, M., Devarajan, P., Haase-Fielitz, A. et al. (2011).
hospitalized patients. J. Am. Soc. Nephrol. 16 (11): The outcome of neutrophil gelatinase-associated
3365–3370. lipocalin-positive subclinical acute kidney injury a
8 KDIGO AKI Work Group (2012). KDIGO clinical practice multicenter pooled analysis of prospective studies. J. Am.
guideline for acute kidney injury. Kidney Int. Suppl. 2: Coll. Cardiol. 57 (17): 1752–1761.
1–138. 21 Ronco, C., Kellum, J.A., and Haase, M. (2012). Subclinical
9 Chawla, L.S., Bellomo, R., Bihorac, A. et al. (2017). Acute AKI is still AKI. Crit. Care 16 (3): 313.
kidney disease and renal recovery: consensus report of 22 Nickolas, T.L., Schmidt-Ott, K.M., Canetta, P. et al.
the acute disease quality initiative (ADQI) 16 workgroup. (2012). Diagnostic and prognostic stratification in the
Nat. Rev. Nephrol. 13 (4): 241–257. emergency department using urinary biomarkers of
10 Moore, P.K., Hsu, R.K., and Liu, K.D. (2018). nephron damage: a multicenter prospective cohort study.
Management of acute kidney Injury: core curriculum J. Am. Coll. Cardiol. 59 (3): 246–255.
2018. Am. J. Kidney Dis. 72 (1): 136–148. 23 Coca, S.G., Garg, A.X., Thiessen-Philbrook, H. et al.
11 Chawla, L.S., Dommu, A., Berger, A. et al. (2008). Urinary (2014). Urinary biomarkers of AKI and mortality 3 years
sediment cast scoring index for acute kidney injury: a after cardiac surgery. J. Am. Soc. Nephrol. 25 (5):
pilot study. Nephron Clin. Pract. 110 (3): c145–c150. 1063–1071.
12 Perazella, M.A., Coca, S.G., Hall, I.E. et al. (2010). Urine 24 Kashani, K., Al-Khafaji, A., Ardiles, T. et al. (2013).
microscopy is associated with severity and worsening of Discovery and validation of cell cycle arrest biomarkers
acute kidney injury in hospitalized patients. Clin. J. Am. in human acute kidney injury. Crit. Care 17 (1): R25.
Soc. Nephrol. 5 (3): 402–408. 25 Bihorac, A., Chawla, L.S., Shaw, A.D. et al. (2014).
13 Schinstock, C.A., Semret, M.H., Wagner, S.J. et al. (2013). Validation of cell-cycle arrest biomarkers for acute kidney
Urinalysis is more specific and urinary neutrophil injury using clinical adjudication. Am. J. Respir. Crit. Care
gelatinase-associated lipocalin is more sensitive for early Med. 189 (8): 932–939.
detection of acute kidney injury. Nephrol. Dial. 26 Meersch, M., Schmidt, C., Hoffmeier, A. et al. (2017).
Transplant. 28 (5): 1175–1185. Prevention of cardiac surgery-associated AKI by
Reference 95
implementing the KDIGO guidelines in high risk patients prediction of AKI severity. J. Am. Soc. Nephrol. 26 (8):
identified by biomarkers: the PrevAKI randomized 2023–2031.
controlled trial. Intensive Care Med. 43 (11): 1151–1161. 37 Goldstein, S.L. and Chawla, L.S. (2010). Renal angina.
27 Gocze, I., Jauch, D., Gotz, M. et al. (2018). Biomarker- Clin. J. Am. Soc. Nephrol. 5 (5): 943–949.
guided intervention to prevent acute kidney injury after 38 Hodgson, L.E., Sarnowski, A., Roderick, P.J. et al. (2017).
major surgery: the prospective randomized BigpAK study. Systematic review of prognostic prediction models for
Ann. Surg. 267 (6): 1013–1020. acute kidney injury (AKI) in general hospital
28 Koyner, J.L. and Parikh, C.R. (2013). Clinical utility of populations. BMJ Open 7 (9): e016591.
biomarkers of AKI in cardiac surgery and critical illness. 39 Flechet, M., Guiza, F., Schetz, M. et al. (2017).
Clin. J. Am. Soc. Nephrol. 8 (6): 1034–1042. AKIpredictor, an online prognostic calculator for acute
29 Grams, M.E., Sang, Y., Ballew, S.H. et al. (2015). A kidney injury in adult critically ill patients: development,
meta-analysis of the association of estimated GFR, validation and comparison to serum neutrophil
albuminuria, age, race, and sex with acute kidney injury. gelatinase-associated lipocalin. Intensive Care Med. 43 (6):
Am. J. Kidney Dis. 66 (4): 591–601. 764–773.
30 James, M.T., Grams, M.E., Woodward, M. et al. (2015). A 40 Bihorac, A., Ozrazgat-Baslanti, T., Ebadi, A. et al. (2019).
meta-analysis of the association of estimated GFR, MySurgeryRisk: development and validation of a
albuminuria, diabetes mellitus, and hypertension with machine-learning risk algorithm for major complications
acute kidney injury. Am. J. Kidney Dis. 66 (4): 602–612. and death after surgery. Ann. Surg. 269 (4): 652–662.
31 James, M.T., Hemmelgarn, B.R., Wiebe, N. et al. (2010). 41 Koyner, J.L., Adhikari, R., Edelson, D.P., and Churpek,
Glomerular filtration rate, proteinuria, and the incidence M.M. (2016). Development of a multicenter ward-based
and consequences of acute kidney injury: a cohort study. AKI prediction model. Clin. J. Am. Soc. Nephrol. 11 (11):
Lancet 376 (9758): 2096–2103. 1935–1943.
32 Thakar, C.V., Arrigain, S., Worley, S. et al. (2005). A 42 Koyner, J.L., Carey, K.A., Edelson, D.P., and Churpek,
clinical score to predict acute renal failure after cardiac M.M. (2018). The development of a machine learning
surgery. J. Am. Soc. Nephrol.: JASN. 16 (1): 162–168. inpatient acute kidney injury prediction model. Crit. Care
33 Matsuura, R., Komaru, Y., Miyamoto, Y. et al. (2018). Med. 46 (7): 1070–1077.
Response to different furosemide doses predicts AKI 43 Wilson, F.P. and Greenberg, J.H. (2018). Acute kidney
progression in ICU patients with elevated plasma NGAL injury in real time: prediction, alerts, and clinical
levels. Ann. Intensive Care 8 (1): 8. decision support. Nephron: 140(2):116–119.
34 Lumlertgul, N., Peerapornratana, S., Trakarnvanich, T. 44 Chawla, L.S., Eggers, P.W., Star, R.A., and Kimmel, P.L.
et al. (2018). Early versus standard initiation of renal (2014). Acute kidney injury and chronic kidney disease as
replacement therapy in furosemide stress test non- interconnected syndromes. N. Engl. J. Med. 371 (1):
responsive acute kidney injury patients (the FST trial). 58–66.
Crit. Care 22 (1): 101. 45 Chawla, L.S. and Kimmel, P.L. (2012). Acute kidney
35 Chawla, L.S., Davison, D.L., Brasha-Mitchell, E. et al. injury and chronic kidney disease: an integrated clinical
(2013). Development and standardization of a furosemide syndrome. Kidney Int. 82 (5): 516–524.
stress test to predict the severity of acute kidney injury. 46 Bucaloiu, I.D., Kirchner, H.L., Norfolk, E.R. et al. (2012).
Crit. Care 17 (5): R207. Increased risk of death and de novo chronic kidney
36 Koyner, J.L., Davison, D.L., Brasha-Mitchell, E. et al. disease following reversible acute kidney injury. Kidney
(2015). Furosemide stress test and biomarkers for the Int. 81 (5): 477–485.
96
Prerenal Failure and Obstructive Disease

Mubeen M. Khan and Kevin W. Finkel
Division of Renal Diseases and Hypertension, McGovern Medical School, Houston, TX, USA
I ntroduction angiotensin II antagonists, such as angiotensin converting

enzyme (ACE) inhibitors and angiotensin receptor blockers
Prerenal azotemia is the most common form of acute (ARBs), can precipitate ATN. The vasoconstrictor effects of
kidney injury (AKI) in hospitalized patients and, if uncor- angiotensin II are opposed by endogenously produced prosta-
rected, can lead to ischemic parenchymal damage and glandins and nitric oxide. When used acutely in prerenal fail-
acute tubular necrosis (ATN) [1, 2]. The clinical syndrome ure, nonsteroidal anti-inflammatory drugs lead to unopposed
of prerenal failure is characterized by intact tubular func- vasoconstriction and worsening of renal function.
tion with impaired filtration function because of renal The differential vasoconstrictor activity of AII between
hypoperfusion and usually manifests as low urine volume, the afferent and efferent arteriole may be best explained by
reduced urinary concentration of sodium (UNa) and frac- the interaction of AII-mediated vasoconstriction with vari-
tional excretion of sodium (FeNa%), and an elevated ratio ous local mediators of vasodilation stimulated by AII. This
of blood urea nitrogen to serum creatinine (Cr). is illustrated by contrasting results between in vivo and
Prerenal failure is the result of either true volume deple- in vitro vasoconstrictor studies. Microperfusion studies
tion, as occurs with hemorrhage or excessive diuresis, or have demonstrated a 100-fold greater sensitivity for AII to
“effective” volume depletion in circumstances such as con- elicit a vasoconstrictor response in the efferent versus affer-
gestive heart failure (cardiorenal syndrome), liver failure ent arteriole. In vitro studies, however, failed to show a dif-
with ascites (hepatorenal syndrome), and sepsis. The etiolo- ference in vasoconstrictor sensitivity in these vascular
gies for prerenal failure are outlined in Table 8.1. In both beds, suggesting that the production of other local factors
instances decreased intravascular volume causes barorecep- stimulated by AII may contribute to the differential
tor stimulation and enhanced adrenergic output, leading to response. These include prostaglandins secreted by the glo-
systemic and local regulatory mechanisms aimed at restor- merulus in response to AII. Experimental inhibition of
ing normal perfusion pressure. prostaglandin synthesis with indomethacin enhances
At the level of the kidney, a complex interplay between neu- afferent and efferent arteriole vasoconstriction by AII, with
roendocrine and vasoactive agents initially compensates for afferent arteriolar response to AII increasing tenfold [3].
decreased perfusion pressure. Specifically, angiotensin II In response to decreased renal perfusion, activation of
causes preferential vasoconstriction of the efferent arteriole, the renin–aldosterone axis and release of antidiuretic hor-
increased proximal tubular reabsorption, elevation of glomer- mone also occur, increasing sodium and water retention in
ular hydrostatic pressure, and a reduced glomerular ultrafil- an attempt to restore intravascular volume. These tempo-
tration coefficient. Because of angiotensin II preferential rizing physiologic measures are necessary to maintain per-
constriction of the efferent arteriole, with mild to moderate fusion pressures in the face of frank volume depletion but
volume depletion, the glomerular filtration rate (GFR) is are ineffective in sepsis and clearly deleterious in condi-
maintained. With severe volume depletion, frank ischemic tions such as heart failure with pulmonary edema.
ATN may ensue. Since the GFR in such situations is depend- Given this background, prerenal failure is not a simple
ent on relative efferent arteriolar vasoconstriction, the use of case of hypovolemia that can be corrected by administration
Introductio 97
Table 8.1 Causes of prerenal failure.
Decreased cardiac output Redistribution/vasodilation
Myocardial infarction Cirrhosis

Congestive heart failure Nephrotic syndrome
Pericardial tamponade Pancreatitis
Positive pressure ventilation Sepsis
Pulmonary embolism Crush injuries
Intestinal obstruction
Hypovolemia Vascular disease
Hemorrhage Renal artery stenosis
Diuresis Atheroembolism
Diarrhea Vasculitis
Excessive sweating Aortic aneurysm dissection
of intravascular crystalloid or colloid. Treatment must, by leg raise (PLR) has been shown to be predictive of a response
necessity, be directed at the underlying pathophysiology to a fluid challenge. In this maneuver, the patient is passively
that has triggered the compensatory physiologic response moved from a 45-degree semi-recumbent position to supine
common to all etiologies. Therefore, an evidence-based with legs raised and held to 45 degrees for one minute.
approach to the treatment of prerenal azotemia requires a Baseline measurement (e.g., cardiac output) is obtained
review of the available data on several subgroups of patients before the maneuver and repeated afterward. In a prospec-
and cannot take a “one size fits all” approach. tive, observational study, Duus, et al investigated the repro-
ducibility of PLR vs fluid bolus using non-invasive cardiac
output monitor (NICOM) for assessment of fluid respon-
Diagnosis
siveness in spontaneously breathing ED patients. They
Many causes of prerenal failure are clinically obvious. With enrolled 109 patients and each patient received two PLR and
reliance on history, physical examination, and certain labora- two boluses. PLR was strongly correlated with fluid respon-
tory findings, patients with true volume depletion are often siveness as measured by NICOM, whereas the boluses were
easily identified and can be treated with aggressive volume less strongly correlated. In addition, PLR technique was
repletion. Such patients demonstrate orthostatic hypoten- more reproducible than the fluid bolus technique [5].
sion, tachycardia, dry mucous membranes, low urinary out- POCUS has also been used increasingly in recent years. One
put, and poor skin turgor. Other patients require a more such use to assess volume status is IVC measurements. In a
careful evaluation for fluid responsiveness. Accurate predic- systematic review and meta-analysis, Orso, et al reviewed 26
tors of fluid responsiveness are essential in the management studies that investigated the role of IVC collapsibility or dis-
of hemodynamic instability. For those who are fluid respon- tensibility or IVC diameter in critically ill patients (both ven-
sive, they guide fluid replacement. For those who are not, tilated and non-ventilated). Ultimately, they found that
they guide vasopressor and inotrope use and help avoid the ultrasound evaluation of the diameter of the IVC and its res-
detrimental effects of excess fluid. Pulse pressure variation piratory variations was not a reliable method to predict fluid
(PPV) may be measured in the mechanically ventilated responsiveness [6]. Although much has been made of the
patient and is derived from the arterial waveform. It is utility of urinary indices in AKI (vide infra), they are unneces-
defined as the maximum pulse pressure in a respiratory sary in such situations, and patients promptly respond to vol-
cycle minus the minimum pulse pressure divided by the ume resuscitation.
mean of these two values [PPV = 100 x (PPmax – PPmin)/
PPmean] In a systematic review to determine the ability of
CVP and PAOP
dynamic changes in arterial waveform-derived variables,
such as PPV, to predict fluid responsiveness and compare Although physicians often rely on central venous pressure
these with static indices of fluid responsiveness, dynamic (CVP) and pulmonary artery occlusion pressure (PAOP) to
changes of arterial waveform-derived variables during guide fluid administration to determine whether or not a
mechanical ventilation were highly accurate in predicting patient has true volume depletion, there is little evidence that
volume responsiveness in critically ill patients when com- such measurements are clinically useful. There are no trials
pared with traditional static indices of volume responsive- that have assessed the utility of these measurements specifi-
ness, such as CVP [4]. In non-ventilated patients, the passive cally for determining the reversibility of prerenal failure.
98 Prerenal Failure and Obstructive Disease
The CVP and PAOP are frequently used as means of deter- Urinalysis
mining cardiac preload and volume status. However, their use
Urinalysis is often used to differentiate various causes of AKI.
has been criticized because of poor ability to predict which
In prerenal failure, the urine is concentrated and the urinary
patients will have improvement in their hemodynamic status
sediment is bland. Bland urinary sediment, however, does
when given fluid resuscitation [7]. As reviewed by Michard
not allow for discrimination between true and effective vol-
and Teboul, in three of five trials assessing the ability of CVP
ume depletion. Furthermore, patients with prerenal failure
to predict fluid responsiveness in critically ill patients, defined
can have abnormal sediment if they have a pre-existing renal
as an improvement in either stroke volume or cardiac output,
disorder, bilirubin in the urine (causing granular casts), or
there was no significant difference in the baseline CVP meas-
microscopic hematuria and bacteruria from an indwelling
urements between responders and nonresponders. The two
bladder catheter. The utility of the urinalysis in differentiat-
remaining studies reported lower baseline values of CVP in
ing prerenal failure from AKI has been called further into
responders; however, the marked overlap of individual CVP
question by the results of two recent systematic reviews on
levels did not allow for the identification of a discriminatory
the urinary findings in sepsis-related AKI (human and exper-
cutoff value. Likewise, baseline PAOP was not significantly
imental animal studies). These systematic reviews demon-
lower in responders versus nonresponders in seven of nine
strated substantial heterogeneity between studies and
studies. Three studies reported a significant difference in the
suggested that the scientific basis for the use of urinary
baseline PAOP between the groups, being higher in respond-
microscopy in septic AKI and, by extension, to differentiating
ers in a single study and lower in responders in two studies.
ATN from prerenal failure is weak [14, 15].
However, in none of the studies could a cutoff PAOP value be
On the other hand, Varghese, et al, showed in a recent
found that predicted the hemodynamic response to volume
prospective study that serial microscopic examinations of
expansion. In normal subjects, Kumar and associates demon-
the urinary sediment revealed diagnostic findings of ATI
strated a lack of correlation between initial CVP and PAOP
otherwise not identified in a single examination [16]. This
values, and both end diastolic volume and stroke volume [8].
study used two different urinary cast scores developed by
Furthermore, changes in both parameters following 3 l of
Chawla and Parazella, each of which has been previously
saline infusion did not correlate with changes in end diastolic
validated to predict renal outcomes in AKI [17] and [18].
volume and cardiac performance.
With Chawla’s cast scoring index, there was a 99.8% inter-
In addition, numerous randomized controlled trials (RCTs)
observer agreement among three nephrologists who inde-
and meta-analyses assessing the benefit of pulmonary artery
pendently scored 30 patients with AKI due to ATI. In
catheters (PAC) in a variety of clinical circumstances have been
addition, the ROC AUC curve for the scoring index to pre-
reported [9–11]. In a meta-analysis of 13 RCTs, the use of PAC
dict renal recovery was 0.79, indicating that the scoring
neither increased overall mortality and hospital days nor con-
index may be useful in predicting renal outcomes.
ferred any benefit [12]. Likewise, in an RCT of 676 patients with
Parazella’s scoring system was significantly associated with
shock mainly from sepsis, ARDS, or both, early use of PAC did
increased risk of worsening AKI (adjusted relative risk: 7.3;
not significantly affect mortality or morbidity, including the
95% confidence interval: 4.5 to 9.7 for worsening with score
need for dialysis [13]. Therefore, in critically ill patients with
of > or =3 versus score of 0) and was more predictive than
multiple reasons for prerenal failure (sepsis, capillary leak with
AKI Network stage at the time of consultation.
third spacing, positive pressure ventilation, hypoalbuminemia,
cardiac dysfunction, and liver disease), measurement of CVP
and/or PAOP does not improve outcome and has poor predic-
Urinary Indices
tive value for improvement in renal function.
The measurement of urinary sodium (UNa) and calculation
of the fractional excretion of sodium (FeNa%) has been rou-
Urine Output
tinely recommended as a means to differentiate oliguric pre-
The development of oliguria (urine output of less than 400 ml/ renal failure from oliguric ATN, as shown in Table 8.2. In the
day) does not ensure the presence of prerenal failure. It occurs sentinel study reported by Miller et al., a UNa concentration
in both prerenal failure states and numerous other causes of of less than 20 mEq/l had an 80% sensitivity and specificity
AKI, including ATN, obstructive nephropathy, glomerulone- for differentiating prerenal failure from ATN in the face of
phritis, and atheroembolic disease. Despite the compensatory oliguria [19]. The FeNa% performed even better, with sensi-
mechanisms invoked in response to decreased renal perfusion tivity and specificity of 98% and 95%, respectively. The utility
that increase salt and water retention by the kidneys in prere- of either measurement was significantly less in the absence
nal failure states, higher urine outputs can ensue in the face of of oliguria. However, there are numerous exceptions to the
underlying chronic kidney disease, salt wasting states such as general rule [20–22]. Both UNa and FeNa% can be low, sug-
adrenal insufficiency and cerebral salt wasting, concomitant gestive of prerenal failure with AKI from rhabdomyolysis,
use of diuretics, osmotic diuresis, and the presence of nonreab- radiocontrast nephropathy, acute glomerulonephritis,
sorbed anions, such as ketones and bicarbonate in the urine. multiple myeloma, amphotericin B toxicity, and early
Introductio 99
Table 8.2 Urinary indices in ARF.
Number of patients with condition/total number of patients
Index and range Prerenal failure Oliguric ATN Nonoliguric ATN
Urine Na (mEq/l)
<20 18/30 (60%) 0/24 (0%) 2/13 (6%)
20–40 12/30 (40%) 14/24 (59%) 11/31 (35%)
>40 0/30 (0%) 10/24 (41%) 18/31 (59%)
FeNa%
<1 27/30 (90%) 1/24 (4%) 4/31 (12%)
Source: Adapted from Miller et al. [19].
obstructive nephropathy. Thus, the finding of renal sodium prerenal failure is due to true or effective volume depletion,
avidity does not allow one to discriminate between true and while concern is great for the potential harm of indiscrimi-
effective volume depletion nor reliably predict reversibility nant fluid administration in the face of acute lung injury or
with volume administration. frank ARDS. There is also debate on what type of fluid, col-
Despite these limitations, a recent study demonstrates a loid versus crystalloid, is best for volume resuscitation in
potential role for FENa% in patients with cirrhosis. In a pro- such patients. Although not specifically done to address
spective analysis of 200 patients with decompensated cirrho- prerenal failure, review of recent clinical trials in these
sis with AKI, Gowda, et al evaluated the diagnostic utility of areas can provide some type of general guideline.
FeNa% and fractional excretion of urea nitrogen (FEUN%) for
differentiating AKI phenotypes (i.e., prerenal, ATN, and Goal-directed Hemodynamic Management
HRS). After determining the optimal cutoffs, they validated and Fluid Therapy
their findings in an independent cohort of 50 patients. The
AUC for FENa% was 86.6% and for FEUN% was 60.3% for Previous observational nonrandomized studies have shown
ATN vs non-ATN. The AUC for FENa% to differentiate that patients who survived critical illness had higher values
between HRS and non-HRS was 74.5%. FEUN% could not dif- for cardiac index and oxygen delivery than those who died
ferentiate HRS vs non-HRS satisfactorily (60.4%). FENa% and and that those values were higher than normal physiologic
FEUN% were unable to differentiate between prerenal and levels [26, 27]. Two studies of surgical patients showed sig-
HRS (AUC < 70). Thus, among cirrhotics, FENa% at admis- nificant decreases in mortality associated with therapy
sion can be used to differentiate ATN from prerenal and HRS. directed at increasing hemodynamic parameters to supra-
The use of diuretics can increase urinary sodium loss even physiologic values, whereas no benefit was seen in patients
in the face of prerenal failure, thus negating the utility of the with sepsis and mixed groups of critically ill patients [28–
FeNa%. Since the (FeUN%) is primarily dependent on pas- 31]. Subsequently, two RCTs assessed the benefit of goal-
sive forces, it is less influenced by the administration of diu- directed hemodynamic therapy in critically ill patients. In
retics and may be useful in the evaluation of AKI. A group of one trial of 100 patients randomized to dobutamine or pla-
102 patients with AKI was divided into three subgroups: pre- cebo there were no differences in mean arterial pressure
renal failure, prerenal failure treated with diuretics, and (MAP) or oxygen consumption despite higher cardiac index
ATN. The FeNa% was low in 92% of the prerenal failure and oxygen delivery in the treatment group [32]. In fact, the
group and 48% of prerenal failure patients receiving diuret- treatment group had a significantly higher mortality rate
ics. Both groups had similar and significantly lower FeUN% than the placebo group. In another trial of 762 patients ran-
values than the ATN patients [23]. domized into one of three intervention groups, hemody-
namic therapy aimed at achieving supranormal values for
cardiac index or normal values for mixed venous oxygena-
Treatment
tion saturation did not reduce morbidity or mortality [33].
In the patient with clinically obvious volume depletion, ade- In 2001, Rivers et al reported the results of a nRCT on
quate fluid resuscitation can be achieved by the administra- early goal-directed therapy (EGDT) in the treatment of
tion of crystalloids, such as normal saline or lactated Ringers severe sepsis and septic shock [34]. Patients with severe
solution. Such patients often show remarkable improve- sepsis or septic shock were randomly assigned upon emer-
ment in their renal function over a short period of time. gency center arrival to either 6 hours of EGDT or standard
In the complex or critically ill patient, the situation is therapy prior to admission to the ICU. Active intervention
more challenging. In this population, it is difficult to accu- aimed at achieving a central venous oxygen saturation of
rately determine intravascular volume and whether the 70% included saline boluses targeted to a CVP level of
8–12 mmHg, vasopressor agents to maintain a MAP greater significantly lower for crystalloids compared with HES
than 65 mmHg, transfusion to a hematocrit of 30%, and (relative risk 0.91; p = .009 and 0.9; p = .005, respectively).
dobutamine. Although the patients assigned to active These findings suggest that crystalloids are less effective than
intervention received more fluid and blood products in the colloids at hemodynamic stabilization in this population
initial 24-hour period, they experienced a significantly [41]. In a recent Cochrane review assessing the effect of using
reduced rate of organ failure and death. colloids vs crystalloids in critically ill patients on mortality
Subsequently, the ARDS Clinical Trials Network assessed and need for blood transfusion or renal replacement therapy
optimal fluid management in patients in the ICU with acute (RRT), using various colloids vs crystalloids appeared to
lung injury [4]. In this RCT, 1000 patients were assigned to make little or no difference to mortality. Starches probably
either a conservative or liberal strategy of fluid manage- slightly increase the need for blood transfusion and RRT [42].
ment using explicit protocols applied for 7 days. The mean
cumulative fluid balances in the conservative versus liberal High- chloride Versus Balanced Fluids
groups were −136 and 6992 ml, respectively. Although there
was no difference in mortality, the conservative group had Recent attention has shifted toward the potential negative
significantly reduced ventilator and ICU days, and improved effects of chloride in intravenous fluids on patient out-
oxygenation. Importantly, conservative fluid management comes. The concentration of chloride in human plasma is
did not increase the need for renal replacement therapy approximately 100 meq/l. In 0.9% saline, it is much higher
(RRT). Three independent multicenter RCTs evaluated at 154 meq/l. In Ringer’s lactate (LR), it is still somewhat
EGDT [24]. All three trials confirmed that there was no sur- elevated at 109 meq/l, and in Plasma-Lyte, it is more physi-
vival benefit compared to usual resuscitation. In a patient ologic at 98 meq/l.
level meta-analysis from these trial, EGDT did not result in Two recent studies from a single center investigated
better outcomes than usual care and was associated with whether fluid resuscitation of noncritically ill (SALT-ED)
higher hospitalization costs across a broad range of patient and critically ill (SMART) patients with balanced crystalloid
and hospital characteristics [25]. vs isotonic crystalloid in the emergency department resulted
in earlier hospital discharge. The primary outcome was
hospital-free days (days alive after discharge before day 28)
Crystalloid Versus Colloid Fluids
and the secondary outcome included a composite of major
The putative superiority of colloid over crystalloid fluids in adverse kidney events within 30 days. The type of crystalloid
the resuscitation of the critically ill patient has been a source (balanced vs isotonic) administered in the emergency depart-
of considerable controversy. Aggressive hydration with crys- ment was assigned to each patient on the basis of calendar
talloid solutions such as normal saline can worsen intersti- month, with the entire emergency department crossing over
tial edema and pulmonary function. Colloidal solutions, between balanced crystalloids (mostly lactated ringers) and
such as various starches and human albumin, might appear saline monthly during the 16-month trial. Results showed
to be attractive alternatives, but there is little solid evidence that the number of hospital-free days did not differ between
of their superiority in clinical trials. Systematic reviews of the balanced crystalloids and saline groups (median, 25 days
RCTs comparing crystalloids with colloids have yielded con- in each group; adjusted odds ratio with balanced crystalloids,
flicting results. Some trials have found an increased mortal- 0.98; 95% confidence interval [CI], 0.92 to 1.04; P = 0.41).
ity rate associated with the administration of human However, balanced crystalloids resulted in a lower incidence
albumin and hydroxyethylstarch, while others have not [35– of major adverse kidney events within 30 days than saline
38]. More recently, a large, randomized, controlled prospec- (4.7% vs. 5.6%; adjusted odds ratio, 0.82; 95% CI, 0.70 to 0.95;
tive trial of albumin versus saline in almost 7000 critically ill P = 0.01) [43]. In the SMART trial, critically patients were
patients found no benefit of one over the other [39]. assigned to receive saline or balanced crystalloids in a rand-
Specifically, there was no demonstrable effect on mortality, omized fashion. The primary outcome was a major adverse
renal function, or the frequency of RRT. Of note, patients kidney event within 30 days—a composite of death from any
with cirrhosis were excluded from this trial, and limited data cause, new renal replacement therapy, or persistent renal
suggest that albumin is useful to prevent AKI in cirrhotic dysfunction—or 30 days, whichever occurred first. Among
patients with spontaneous bacterial peritonitis [40]. the 7942 patients in the balanced-crystalloids group, 1139
In a systematic review and meta-analysis that evaluated (14.3%) had a major adverse kidney event, as compared with
hemodynamic response to crystalloids/colloids in critically 1211 of 7860 patients (15.4%) in the saline group (marginal
ill adults, central venous pressure was significantly lower odds ratio, 0.91; 95% confidence interval [CI], 0.84 to 0.99;
with crystalloids vs. albumin (mean difference [MD]: −3.5 conditional odds ratio, 0.90; 95% CI, 0.82 to 0.99; P = 0.04).
mm Hg; p = .03) or gelatin (MD: −9.2 mm Hg; p = .02). In-hospital mortality at 30 days was 10.3% in the balanced-
Compared with the albumin group, cardiac index was crystalloids group and 11.1% in the saline group (P = 0.06).
significantly lower in the crystalloid group (MD: −0.6 L/ The incidence of new renal-replacement therapy was 2.5%
min/m2, p < .001). All mortality and 90-day mortality were and 2.9%, respectively (P = 0.08), and the incidence of
Summar 101
ersistent renal dysfunction was 6.4% and 6.6%, respectively

p water retention and systemic vasoconstriction, which in
(P = 0.60). The authors concluded that in critically ill adults, turn decrease renal perfusion. Activation of the previously
the use of balanced crystalloids for IV resuscitation resulted mentioned systems overrides the vasodilatory effects of
in a lower rate of the composite outcome of death from any natriuretic peptides, nitric oxide, prostaglandins, and brad-
cause, new renal-replacement therapy, or persistent renal ykinin. Chronic and acute exacerbations of CHF are asso-
dysfunction than the use of saline. [44]. ciated with widespread inflammation and high levels of
proinflammatory cytokines, which may further boost car-
diac, kidney, and other tissue dysfunction [46].
S
ummary Renal perfusion is regulated by adenosine receptors
involved with tubuloglomerular feedback (TGF) signaling,
In the complex medical patient, prerenal failure can occur and maintain intrarenal vascular tone via a number of
from either true or effective volume depletion. Since both cir- pathways which, in CRS, result in afferent arteriolar vaso-
cumstances share the same physiologic derangements, clini- constriction and efferent arteriolar dilation. Actions of
cal parameters such as urine output and urinary electrolytes adenosine at the macula densa and mesangium depend on
do not have any discriminatory power between the two. It is angiotensin, renin, nitric oxide, and prostaglandin levels,
also difficult to predict which patients will benefit from vol- which are altered in CRS [46].
ume resuscitation based on measurement of the CVP or Right ventricular dysfunction increases central venous
PAOP. Often, the only way to determine if the renal failure is pressure, which in turn increases renal venous pressure.
reversible is to provide an empiric “fluid challenge” and Right ventricular dilation and dysfunction also reduce left
assess the clinical response. With regard to approaching the ventricular filling and decrease renal arterial pressure. The
critically ill patient who may be functionally prerenal failure, combination of increased renal venous and decreased
fluid management is dependent on timing. Early interven- renal arterial pressures diminishes the pressure gradient
tion with goal-directed therapy, resulting in a positive fluid across the glomerulus and decreases GFR [46].
balance in the first 24 hours, is associated with reduced organ
failure. In patients already in the ICU with lung injury, a con- Treatment of Cardiorenal Syndrome
servative fluid approach improves pulmonary function with-
out adverse renal effects. Loop diuretics remain the cornerstone of treatment of car-
diorenal syndrome for effecting natriuresis, net negative salt
and water balance, and extracellular volume reduction.
Cardiorenal Syndrome
Extracellular fluid volume reduction in patients with CHF
A recent classification of CRS proposed by the 7th Acute returns hemodynamics to a more optimal position on the
Dialysis Quality Initiate consensus conference has divided Frank-Starling curve. Reducing volume overload improves
the syndromes into those that are “cardiorenal” referring to renal perfusion pressure by decreasing central venous pres-
when cardiac dysfunction leads to kidney dysfunction and sure, renal venous pressure, and right ventricular dilation,
those that are “renocardiac” referring to when primary kid- which improves right and left ventricular performance.
ney dysfunction leads to cardiac dysfunction. These syn- Diuretic resistance remains a challenge and is manifested as
dromes are further classified based on their acuity and the a failure to to resolve congestion and a low urine Na despite
presence of a systemic (non-cardiac, non-renal) illness that maximal dose of diuretic. This can occur as a result of inad-
may play a role in the pathophysiology [45]. There are 4 equate dose, nonadherence, pharmacokinetic factors, CKD,
major mechanisms implicated in CRS: increased renal hypotension, hypoalbuminemia, antin-atriuretic drugs,
venous pressure (ie, venous congestion), neurohumoral nephron remodeling, and neurohormonal braking. This last
imbalances, reduced renal perfusion, and right ventricular phenomenon occurs when diuretics reduce extracellular
dilation and dysfunction [46]. fluid volume to such an extent that the sympathetic nervous
Venous hypertension is thought to result in decreased system and angiotensin II production are activated, which
renal perfusion, raised kidney interstitial pressure, nar- then stimulate proximal tubular sodium reabsorption, effec-
rowed arterial-to-venous renal pressure gradient, decreased tively neutralizing the therapeutic effect of the diuretic [46].
glomerular filtration rate (GFR), maladaptive autoregula- Despite the centrality of diuretics in the management of car-
tory responses, and other characteristic neurohumoral diorenal syndrome, there is a paucity of data to guide their
imbalances. This higher renal venous pressure attenuates use. The Diuretic Optimization Strategies Evaluation
glomerular filtration, causes tubule collapse, and may trig- (DOSE) trial was a prospective, double-blind, randomized
ger tubulointerstitial fibrosis [46]. (The neurohumoral trial of patients with acute decompensated heart failure that
mechanisms underlying CRS consist of activation of the compared intravenous furosemide delivered by either a
renin-angiotensin-aldosterone and sympathetic nervous bolus every 12 hours or by continuous infusion and at either
systems as well as increased production of antidiuretic hor- a low dose (equivalent to the patient’s previous oral dose) or
mone and endothelin 1. These processes increase salt and a high dose (2.5 times the previous oral dose). Co-primary
endpoints were patients’ global assessment of symptoms n = 573), OVERTURE (Omapatrilat Versus Enalapril
and the change in serum creatinine level from baseline to 72 Randomized Trial of Utility in Reducing Events trial;
hours. At the conclusion, the were no significant differences n = 5770), and PARADIGM-HF (Prospective Comparison of
in patients’ global assessment of symptoms or in the change ARNI With ACEI to Determine Impact on Global Mortality
in renal function when diuretic therapy was administered by and Morbidity in Heart Failure; n = 8399). The composite
bolus as compared with continuous infusion or at a high outcome of death or HHF was reduced numerically in
dose as compared with a low dose. However, the high-dose patients receiving combined neprilysin/RAAS inhibition in
strategy was associated with a greater diuresis [47]. With all 3 trials, with a pooled HR of 0.86 (95% CI, 0.76–0.97;
regard to diuretic dosing regimen, Grodin, et al compared P = 0.013). Combined neprilysin/RAAS inhibition compared
the diuretic effect of a urine output-guided diuretic adjust- with ACE inhibitor alone was associated with more hypoten-
ment and standard therapy for the management of cardiore- sion but less renal dysfunction and hyperkalemia in all 3 trials.
nal syndrome in acute decompensated heart failure. They [54]. In a subset analysis of PARADIGM-HF, treatment with
showed that compared with a standard diuretic regimen, sacubitril/valsartan resulted a slower rate of decrease in
patients who received stepwise intensification of diuretic eGFR compared with enalapril, including in patients with
therapy had greater net fluid and weight loss without being CKD, despite a modest increased in albuminuria [55]. The
associated with renal compromise [48]. Another strategy to recently published HARP-III trial, investigated the effects of
guide dosing is to follow urine sodium concentrations. An sacubitril/valsartan on kidney function and cardiac biomark-
intravenous loop diuretic dose that produces a spot urine ers in patients with moderate to severe chronic kidney dis-
sodium <50 mmol/l at 2 hours is inadequate and should ease (GFR 20-60 ml/min/1.73 m2). Over 12 months,
generally be doubled until a maximal dose is reached (not sacubitril/valsartan had similar effects on kidney function
clearly defined but often up to a bolus dose of 200 to 300 mg and albuminuria when compared with irbesartan, but had
of furosemide equivalents) and urine sodium remeasured. the additional effect of lowering blood pressure and cardiac
Once an adequate natriuretic response is achieved, the dose biomarkers. Addition of sacubitril to a patient on RAAS inhi-
can be repeated every 6 to 12 h to achieve the goal net nega- bition could further reduce cardiovascular risk in patients
tive sodium balance [49]. Addition of low-dose dopamine or with CKD, without further worsening the CKD [56]. Data on
low-dose nesiritide to diuretic therapy in patients hospital- the safety and efficacy of Mineralocorticoid receptor antago-
ized with acute decompensated heart failure and renal dys- nists (MRAs in heart failure with advanced CKD are limited.
function were studied in the Renal Optimization Strategies In the EMPHASIS-HF trial (Eplerenone in Mild Patients
Evaluation (ROSE) trial, which showed that neither had a Hospitalization and Survival Study in Heart Failure), the
significant effect on 72-hour cumulative urine volume or effect of eplerenone on the primary composite of end point
change in cystatin C level when compared to placebo [50]. on HHF or cardiovascular death was similar in patients
dichotomized at an eGFR <60 mL/min per 1.73 m2 [57a].
ACE Inhibitors and ARBs
Natriuretic Peptides
The strength of evidence of ACE inhibitors in HF with pre-
dialytic CKD is not established given the lack of inclusion of B-type natriuretic peptide (BNP) is produced by the myocardial
these patients in RCTs for HF. Both CONSENSUS ventricles in response to increased stress. BNP results in arterial
(Cooperative North Scandinavian Enalapril Survival Study) and venous vasodilation, increased sodium excretion, and sup-
and SOLVD (Study of Left Ventricular Dysfunction) trial pression of the renin–angiotensin–aldosterone system.
demonstrated the benefits of enalapril for HF symptoms and Nesiritide, a synthetic BNP, was assessed in heart failure
hospitalization reduction. The enalapril group in SOLVD patients in the Vasodilation in the Management of Acute
showed a 33% higher likelihood of a serum creatinine rise of Congestive Heart Failure trial [57]. Compared to intrave-
>0.5 mg/dL, but no data on progression of CKD, ESKD, or nous nitroglycerine, nesiritide significantly lowered PAOP
doubling of creatinine were reported [51] A post hoc analysis at 1 hour, although there was no difference in the degree of
of SOLVD with HF and CKD demonstrated a mortality ben- dyspnea. Nesiritide was equally effective in patients with
efit even in subjects with higher degrees of CKD [52]. In a or without renal dysfunction (Cr > 2.0 mg/dl) [58].
propensity score analysis of 1665 patients with HF (EF <45%) In another trial nesiritide was given to patients with
and CKD (eGFR <60 mL/min per 1.73 m2), treatment with recent increases in baseline Cr associated with congestive
an ACE inhibitor or ARB was associated with significant heart failure exacerbation (mean Cr increase from 1.5 to
reductions in all-cause mortality (HR, 0.68 [95% CI, 0.74– 1.8 mg/dl) [59]. Compared to placebo, nesiritide resulted in
0.996]; P = 0.04) [53]. A recent meta-analysis of data from 3 no changes in urine output, sodium excretion, GFR, or
trials in HFrEF that compared combined neprilysin/RAAS effective renal plasma flow. This finding suggests that nesi-
inhibition with RAAS inhibition alone. The trials were the ritide is ineffective in cardiorenal syndrome.
IMPRESS (Inhibition of Metallo Protease by Omapatrilat in a Additionally, the use of nesiritide in congestive heart
Randomized Exercise and Symptoms Study of Heart Failure; failure may be associated with an increased risk of acute
Summar 103
renal dysfunction and mortality. In a meta-analysis of five depending on the etiology of the obstruction. Patients may
randomized trials including 1269 patients, the use of nesir- have very dilute urine due to the presence of an acquired
itide was significantly associated with the development of form of nephrogenic diabetes insipidus [65].
acute renal dysfunction (an increase of Cr of 0.5 mg/dl) Ultrasonography is the most useful test for the presence
[60]. Furthermore, in three RCTs that reported mortality of obstruction. Although hydronephrosis is usually dem-
rates, the use of nesiritide was associated with a hazard onstrated, there are circumstances when hydronephrosis is
ratio for death of 1.8 (P = 0.057) [61]. not seen despite urinary tract obstruction: (i) early in the
course of obstruction (12–24 hours) when the collecting
Ultrafiltration system is relatively noncompliant; (ii) in the face of severe
volume depletion when glomerular filtration is severely
The CARRESS-HF (Cardiorenal Rescue Study in Acute
depressed; and (iii) when the collecting system is encased
Decompensated Heart Failure) trial did not show a differ-
by retroperitoneal lymphadenopathy or fibrosis [66, 67].
ence in weight loss with ultrafiltration vs a stepped pharma-
Conversely, the finding of hydronephrosis on ultrasound
cologic therapy algorithm in patients hospitalized for acute
does not prove the presence of obstruction, since it is also seen
decompensated heart failure and worsened renal function.
in high urinary flow states such as diuretic use and diabetes
However, ultrafiltration was associated with increased cre-
insipidus, pregnancy, previous obstruction, and congenital
atinine level at 96 hours [62].
megaureter. In a series of 192 patients with AKI, hydronephro-
A recent Cochrane review looking at the benefits and harms
sis unrelated to obstruction was noted in 11% of patients [68].
of pharmacological interventions for heart failure in patients
Although ultrasound is the preferred imagining modality for
with heart failure and CKD concluded that it is uncertain
initial assessment of obstruction, its use has been supplanted
whether these therapies, including nesiritide, make any differ-
by noncontrast helical computed tomography (CT) scan in the
ence to death, hospitalizations, or adverse effects including
evaluation of flank pain and nephrolithiasis. In a single-center
worsening kidney function, hyperkalemia, or hypotension [63].
study of 864 patients evaluated for flank pain, 34 underwent
both helical CT scan and ultrasound [69]. Compared to helical
Obstructive Nephropathy
CT, ultrasound was found to have sensitivity and specificity for
Urinary tract obstruction should always be considered in renal stones of 81% and 100%, respectively, although the sensi-
the differential diagnosis of AKI. Regardless of cause, tivity for ureteric stones was only 46%. In the same study, the
obstruction of urinary flow leads to renal impairment, sensitivity and specificity for the detection of hydronephrosis
which early in the course of the condition is reversible if were 93% and 100%, respectively.
the obstruction is alleviated. Tubular function is initially In another study, 181 consecutive patients with acute
affected; however, prolonged obstruction leads to tubular flank pain underwent a combination of helical CT, ultra-
damage and parenchymal atrophy. sound, and unenhanced radiography [70]. When compared
Clinical manifestations of urinary tract obstruction vary with the diagnostic accuracy for ureterolithiasis of com-
depending on the location, duration, and degree of obstruc- bined ultrasound and unenhanced radiography, helical CT
tion. In patients with complete bilateral obstruction or had greater sensitivity (92% vs. 77%).
with an obstructed solitary kidney, anuria (<50 ml urine Treatment of obstruction and its timing is dependent on the
output in 24 hours) can be the presenting feature, whereas etiology. Patients with small stones (>5 mm) can be managed
in patients with partial obstruction, the urinary output can conservatively with aggressive hydration and pain control in
vary from oliguria to polyuria. Although pain is more likely the absence of infection. Bladder catheterization should be
to be associated with acute blockage, obstruction may be performed if there is reason to suspect that bladder outlet
totally asymptomatic and occur without overt clinical obstruction is present; possible clues to this diagnosis include
manifestations or suggestive laboratory findings. Therefore, suprapubic pain, a palpable bladder, or an older man with
obstructive nephropathy should always be considered as a unexplained renal failure. Measurement of a urinary postvoid
cause of renal failure when another obvious prerenal fail- residual is often advocated in the detection of outlet obstruc-
ure or intrinsic renal cause is not identified. The location of tion but has not been subjected to any rigorous evaluation.
obstruction is anatomically divided into upper and lower Nor is there a standard method of measurement or definition.
urinary tract at or below the level of the bladder. Common Up to one-third of patients with bladder outlet obstruction
causes of lower urinary tract obstruction include urethral demonstrated by urodynamic testing do not have an elevated
strictures, prostatic hypertrophy, and neurogenic bladder. postvoid residual [71]. Upper tract obstruction is relieved with
Diagnosis of urinary tract obstruction can be difficult. either percutaneous nephrostomy tubes or ureteral stenting,
Anuria, flank pain with a palpable mass, or a palpable depending on cause, availability, and local expertise.
bladder are obvious clues. Hyperkalemia with a nonanion The duration and severity of obstruction are the major
gap metabolic acidosis is suggestive of a renal tubular aci- determinants for the recovery of renal function after its
dosis associated with obstruction [64]. The urinary sedi- correction. The longer the duration of obstruction, the less
ment may be bland or demonstrate crystals or hematuria, likely are the chances for complete renal recovery. There are
few published data on the relationship between duration and treatment of postobstructive diuresis, there is very little in
obstruction and renal recovery rates. The variability is likely the medical literature that describes its clinical course or out-
related to the severity and cause of obstruction, the presence come other than case reports and case series. Patients with
of underlying chronic kidney disease, and concurrent infec- relief of complete obstruction are considered at risk for devel-
tion. Case reports, however, have documented renal recov- oping postobstructive diuresis. It has been recommended for
ery after correction of obstruction lasting 12 months [72]. patients without pulmonary edema, congestive heart failure,
or altered consciousness that urine losses be replaced by oral
intake. Intravenous fluids are given only if the patients
Postobstructive Diuresis
develop orthostatic hypotension, tachycardia, hyponatremia,
Postobstructive diuresis occurs when there is correction of or a urine output of more than 200 ml/h [74]. In high-risk
complete bilateral obstruction or complete obstruction of a patients with altered sensorium, congestive heart failure, or
solitary kidney. It involves the production of a large volume pulmonary edema, replacement of half the hourly urine
of urine that results from a defect in urine-concentrating output with half-normal saline has been recommended. If
ability, impaired reabsorption of urinary sodium, and solute the patient is hyponatremic, normal saline should be used
diuresis from the retained urea and intravenous administra- instead. None of these recommendations has been evaluated
tion of sodium-containing solutions [73]. Although numer- with rigorous clinical trial. They are considered to be prudent
ous textbooks discuss the proposed pathogenesis and and reasonable based on clinical experience.
R
eferences
1 Liano, F. and Pascual, J. (1996). Epidemiology of acute artery catheters in high-risk surgical patients. N. Engl. J.
renal failure: a prospective, multicenter, community- Med. 348: 5–14.
based study. Madrid Acute Renal Failure Study Group. 10 Harvey, S., Harrison, D.A., Singer, M. et al. (2005). Assessment
Kidney Int. 50: 811–818. of the clinical effectiveness of pulmonary artery catheters in
2 Thadhani, R., Pascual, M., and Bonventre, J.V. (1996). management of patients in intensive care (PAC-man): a
Acute renal failure. N. Engl. J. Med. 334: 1448–1460. randomised controlled trial. Lancet 366: 472–477.
3 Ko, Benjamin and Bakris, George. “The Renin- 11 Binanay, C., Califf, R.M., Hasselblad, V. et al. (2005).
Angiotensin-Aldosterone System and the Kidney.” Evaluation study of congestive heart failure and
Textbook of Nephro-Endocrinology, 2nd Ed., edited by pulmonary artery catheterization effectiveness: the
Williams, Gordon, Elsevier, 2017;29–30. ESCAPE trial. JAMA 294: 1625–1633.
4 Marik PE, Cavallazzi R, Vasu T, Hirani A. Dynamic changes in 12 Shah, M.R., Hasselblad, V., Stevenson, L.W. et al. (2005).
arterial waveform derived variables and fluid responsiveness Impact of the pulmonary artery catheter in critically ill
in mechanically ventilated patients: a systematic review of patients: meta-analysis of randomized clinical trials.
the literature. Crit Care Med. 2009; 37 (9): 2642–2647. JAMA 294: 1664–1670.
5 Duus N, Shogilev DJ, Skibsted S, et al. The reliability and 13 Richard, C., Warszawski, J., Anguel, N. et al. (2003). Early
validity of passive leg raise and fluid bolus to assess fluid use of the pulmonary artery catheter and outcomes in
responsiveness in spontaneously breathing emergency patients with shock and acute respiratory distress syndrome:
department patients. J Crit Care. 2015; 30 (1): 217. a randomized controlled trial. JAMA 290: 2713–2720.
e1–217.e2175. 14 Bagshaw, S.M., Langenberg, C., and Bellomo, R. (2006).
6 Orso D, Paoli I, Piani T, Cilenti FL, Cristiani L, Guglielmo Urinary biochemistry and microscopy in septic acute
N. Accuracy of Ultrasonographic Measurements of renal failure: a systematic review. Am. J. Kidney Dis. 48:
Inferior Vena Cava to Determine Fluid Responsiveness: A 695–705.
Systematic Review and Meta-Analysis. J Intensive Care 15 Bagshaw, S.M., Langenberg, C., Wan, L. et al. (2007). A
Med. 2020; 35 (4): 354–3630. systematic review of urinary findings in experimental
7 Michard, F. and Teboul, J.L. (2002). Predicting fluid septic acute renal failure. Crit. Care Med. 35: 1592–1598.
responsiveness in ICU patients: a critical analysis of the 16 Varghese V, Rivera MS, Alalwan AA, Alghamdi AM,
evidence. Chest 121: 2000–2008. Gonzalez ME, Velez JCQ. Diagnostic Utility of Serial
8 Kumar, A., Anel, R., Bunnell, E. et al. (2004). Pulmonary Microscopic Examination of the Urinary Sediment in
artery occlusion pressure and central venous pressure fail Acute Kidney Injury. Kidney 360. 2020; 2 (2): 182–191.
to predict ventricular filling volume, cardiac Published 2020 Dec 11. doi:10.34067/KID.0004022020.
performance, or the response to volume infusion in 17 Chawla LS, Dommu A, Berger A, Shih S, Patel SS.
normal subjects. Crit. Care Med. 32: 691–699. Urinary sediment cast scoring index for acute kidney
9 Sandham, J.D., Hull, R.D., Brant, R.F. et al. (2003). A injury: a pilot study. Nephron Clin Pract. 2008; 110 (3):
randomized, controlled trial of the use of pulmonary- c145–c150. doi:10.1159/000166605
Reference 105
18 Perazella MA, Coca SG, Hall IE, Iyanam U, Koraishy M, 33 Gattinoni, L., Brazzi, L., Pelosi, P. et al. (1995). A trial of
Parikh CR. Urine microscopy is associated with severity goal-oriented hemodynamic therapy in critically ill patients.
and worsening of acute kidney injury in hospitalized SvO2 collaborative group. N. Engl. J. Med. 333: 1025–1032.
patients. Clin J Am Soc Nephrol. 2010; 5 (3): 402–408. 34 Rivers, E., Nguyen, B., Havstad, S. et al. (2001). Early
doi:10.2215/CJN.06960909 goal-directed therapy in the treatment of severe sepsis
19 Miller, T.R., Anderson, R.J., Linas, S.L. et al. (1978). and septic shock. N. Engl. J. Med. 345: 1368–1377.
Urinary diagnostic indices in acute renal failure: a 35 Schierhout, G. and Roberts, I. (1998). Fluid resuscitation
prospective study. Ann. Intern. Med. 89: 47–50. with colloid or crystalloid solutions in critically ill patients:
20 Corwin, H.L., Schreiber, M.J., and Fang, L.S. (1984). Low a systematic review of randomised trials. BMJ 316: 961–964.
fractional excretion of sodium. Occurrence with 36 Choi, P.T., Yip, G., Quinonez, L.G., and Cook, D.J. (1999).
hemoglobinuric-and myoglobinuric-induced acute renal Crystalloids vs. colloids in fluid resuscitation: a
failure. Arch. Intern. Med. 144: 981–982. systematic review. Crit. Care Med. 27: 200–210.
21 Vaz, A.J. (1983). Low fractional excretion of urine sodium 37 Ragaller, M.J., Theilen, H., and Koch, T. (2001). Volume
in acute renal failure due to sepsis. Arch. Intern. Med. 143: replacement in critically ill patients with acute renal
738–739. failure. J. Am. Soc. Nephrol. 12 (Suppl 17): S33–S39.
22 Zarich, S., Fang, L.S., and Diamond, J.R. (1985). Fractional 38 Cochrane Injuries Group Albumin Reviewers (1998).
excretion of sodium. Exceptions to its diagnostic value. Human albumin administration in critically ill patients:
Arch. Intern. Med. 145: 108–112. systematic review of randomised controlled trials. BMJ
23 Carvounis, C.P., Nisar, S., and Guro-Razuman, S. (2002). 317: 235–240.
Significance of the fractional excretion of urea in the 39 Finfer, S., Bellomo, R., Boyce, N. et al. (2004). A
differential diagnosis of acute renal failure. Kidney Int. comparison of albumin and saline for fluid resuscitation
62: 2223–2229. in the intensive care unit. N. Engl. J. Med. 350: 2247–2256.
24 Osborn TM. Severe Sepsis and Septic Shock Trials 40 Sort, P., Navasa, M., Arroyo, V. et al. (1999). Effect of
(ProCESS, ARISE, ProMISe): What is Optimal intravenous albumin on renal impairment and mortality
Resuscitation?. Crit Care Clin. 2017; 33 (2): 323–344. in patients with cirrhosis and spontaneous bacterial
doi:10.1016/j.ccc.2016.12.004 peritonitis. N. Engl. J. Med. 341: 403–409.
25 PRISM Investigators, Rowan KM, Angus DC, et al. Early, 41 Martin GS, Bassett P. Crystalloids vs. colloids for fluid
Goal-Directed Therapy for Septic Shock - A Patient-Level resuscitation in the Intensive Care Unit: A systematic
Meta-Analysis. N Engl J Med. 2017; 376 (23): 2223–2234. review and meta-analysis. J Crit Care. 2019; 50: 144–154.
doi:10.1056/NEJMoa1701380 doi:10.1016/j.jcrc.2018.11.031
26 Bland, R.D., Shoemaker, W.C., Abraham, E., and Cobo, 42 Lewis SR, Pritchard MW, Evans DJ, et al. Colloids versus
J.C. (1985). Hemodynamic and oxygen transport patterns crystalloids for fluid resuscitation in critically ill people.
in surviving and nonsurviving postoperative patients. Cochrane Database Syst Rev. 2018; 8 (8): CD000567. Published
Crit. Care Med. 13: 85–90. 2018 Aug 3. doi:10.1002/14651858.CD000567.pub7)
27 Shoemaker, W.C., Montgomery, E.S., Kaplan, E., and Elwyn, 43 Self WH, Semler MW, Wanderer JP, et al. Balanced
D.H. (1973). Physiologic patterns in surviving and Crystalloids versus Saline in Noncritically Ill Adults. N Engl
nonsurviving shock patients. Use of sequential J Med. 2018; 378 (9): 819–828. doi:10.1056/NEJMoa1711586.
cardiorespiratory variables in defining criteria for therapeutic 44 Semler MW, Self WH, Wanderer JP, et al. Balanced
goals and early warning of death. Arch. Surg. 106: 630–636. Crystalloids versus Saline in Critically Ill Adults. N Engl J
28 Shoemaker, W.C., Appel, P.L., Kram, H.B. et al. (1988). Med. 2018; 378 (9): 829–839. doi:10.1056/NEJMoa1711584
Prospective trial of supranormal values of survivors as 45 Kumar U, Wettersten N, Garimella PS. Cardiorenal
therapeutic goals in high-risk surgical patients. Chest 94: Syndrome: Pathophysiology. Cardiol Clin. 2019; 37 (3):
1176–1186. 251–265. doi:10.1016/j.ccl.2019.04.001.
29 Boyd, O., Grounds, R.M., and Bennett, E.D. (1993). A 46 Chitturi C, Novak JE. Diuretics in the Management of
randomized clinical trial of the effect of deliberate Cardiorenal Syndrome. Adv Chronic Kidney Dis. 2018;
perioperative increase of oxygen delivery on mortality in 25 (5): 425–433. doi:10.1053/j.ackd.2018.08.008
high-risk surgical patients. JAMA 270: 2699–2707. 47 Felker GM, Lee KL, Bull DA, et al. Diuretic strategies in
30 Tuchschmidt, J., Fried, J., Astiz, M., and Rackow, E. patients with acute decompensated heart failure. N Engl J
(1992). Elevation of cardiac output and oxygen delivery Med. 2011; 364 (9): 797–805. doi:10.1056/NEJMoa1005419
improves outcome in septic shock. Chest 102: 216–220. 48 Grodin JL, Stevens SR, de Las Fuentes L, et al.
31 Yu, M., Levy, M.M., Smith, P. et al. (1993). Effect of Intensification of Medication Therapy for Cardiorenal
maximizing oxygen delivery on morbidity and mortality Syndrome in Acute Decompensated Heart Failure. J Card
rates in critically ill patients: a prospective, randomized, Fail. 2016; 22 (1): 26–32. doi:10.1016/j.cardfail.2015.07.007
controlled study. Crit. Care Med. 21: 830–838. 49 Felker GM, Ellison DH, Mullens W, Cox ZL, Testani JM.
32 Hayes, M.A., Timmins, A.C., Yau, E.H. et al. (1994). Diuretic Therapy for Patients With Heart Failure: JACC
Elevation of systemic oxygen delivery in the treatment of State-of-the-Art Review. J Am Coll Cardiol. 2020; 75 (10):
critically ill patients. N. Engl. J. Med. 330: 1717–1722. 1178–1195. doi:10.1016/j.jacc.2019.12.059)
50 Chen HH, Anstrom KJ, Givertz MM, et al. Low-Dose 60 Sackner-Bernstein, J.D., Skopicki, H.A., and Aaronson,
Dopamine or Low-Dose Nesiritide in Acute Heart K.D. (2005). Risk of worsening renal function with
Failure With Renal Dysfunction: The ROSE Acute Heart nesiritide in patients with acutely decompensated heart
Failure Randomized Trial. JAMA. 2013; 310 (23): failure. Circulation 111: 1487–1491.
2533–2543. doi:10.1001/jama.2013.282190 61 Sackner-Bernstein, J.D., Kowalski, M., Fox, M., and
51 Rangaswami J, Bhalla V, Blair JEA, et al. Cardiorenal Aaronson, K. (2005). Short-term risk of death after
Syndrome: Classification, Pathophysiology, Diagnosis, treatment with nesiritide for decompensated heart
and Treatment Strategies: A Scientific Statement From failure: a pooled analysis of randomized controlled trials.
the American Heart Association. Circulation. 2019; JAMA 293: 1900–1905.
139 (16): e840–e878. doi:10.1161/CIR.0000000000000664 62 Bart BA, Goldsmith SR, Lee KL, et al. Ultrafiltration in
52 Khan NA, Ma I, Thompson CR, Humphries K, Salem decompensated heart failure with cardiorenal syndrome.
DN, Sarnak MJ, Levin A. Kidney function and mortality N Engl J Med. 2012; 367 (24): 2296–2304. doi:10.1056/
among patients with left ventricular systolic dysfunction. NEJMoa1210357
J Am Soc Nephrol. 2006; 17: 244–253. doi: 10.1681/ 63 Lunney M, Ruospo M, Natale P, Quinn RR, Ronksley PE,
ASN.2005030270 Konstantinidis I, Palmer SC, Tonelli M, Strippoli GFM,
53 Ahmed A, Fonarow GC, Zhang Y, Sanders PW, Allman Ravani P. Pharmacological interventions for heart failure
RM, Arnett DK, Feller MA, Love TE, Aban IB, Levesque in people with chronic kidney disease. Cochrane
R, Ekundayo OJ, Dell’Italia LJ, Bakris GL, Rich MW. Database of Systematic Reviews 2020, Issue 2. Art. No.:
Renin-angiotensin inhibition in systolic heart failure CD012466. DOI: 10.1002/14651858.CD012466.pub2.
and chronic kidney disease. Am J Med. 2012; 125: 64 Batlle, D.C., Arruda, J.A., and Kurtzman, N.A. (1981).
399–410. doi:10.1016/j.amjmed.2011.10.013 Hyperkalemic distal renal tubular acidosis associated
54 Solomon SD, Claggett B, McMurray JJ, Hernandez AF, with obstructive uropathy. N. Engl. J. Med. 304: 373–380.
Fonarow GC. Combined neprilysin and renin- 65 Kato, A., Hishida, A., Ishibashi, R. et al. (1994).
angiotensin system inhibition in heart failure with Nephrogenic diabetes insipidus associated with bilateral
reduced ejection fraction: a meta-analysis. Eur J Heart ureteral obstruction. Intern. Med. 33: 231–233.
Fail. 2016; 18: 1238–1243. doi:10.1002/ejhf.603 66 Webb, J.A. (1990). Ultrasonography in the diagnosis of
55 Damman K, Gori M, Claggett B, Jhund PS, Senni M, renal obstruction. BMJ 301: 944–946.
Lefkowitz MP, Prescott MF, Shi VC, Rouleau JL, Swedberg 67 Rascoff, J.H., Golden, R.A., Spinowitz, B.S., and
K, Zile MR, Packer M, Desai AS, Solomon SD, McMurray Charytan, C. (1983). Nondilated obstructive nephropathy.
JV. Renal effects and associated outcomes during Arch. Intern. Med. 143: 696–698.
angiotensin-neprilysin inhibition in heart failure. JACC 68 Amis, E.S. Jr., Cronan, J.J., P fister, R.C., and Yoder, I.C.
Heart Fail. 2018; 6: 489–498. doi:10.1016/j.jchf.2018.02.004 (1982). Ultrasonic inaccuracies in diagnosing renal
56 UK HARP-III Collaborative Group. Randomized obstruction. Urology 19: 101–105.
multicentre pilot study of sacubitril/valsartan versus 69 Ather, M.H., Jafri, A.H., and Sulaiman, M.N. (2004).
irbesartan in patients with chronic kidney disease: Diagnostic accuracy of ultrasonography compared to
United Kingdom Heart and Renal Protection (HARP), unenhanced CT for stone and obstruction in patients
III: rationale, trial design and baseline data. Nephrol Dial with renal failure. BMC Med. Imag. 4: 2.
Transplant. 2017; 32: 2043–2051. doi:10.1093/ndt/gfw321 70 Catalano, O., Nunziata, A., Altei, F., and Siani, A. (2002).
57 Publication Committee for the VMAC Investigators Suspected ureteral colic: primary helical CT versus
(2002). Intravenous nesiritide vs nitroglycerin for selective helical CT after unenhanced radiography and
treatment of decompensated congestive heart failure: a sonography. Am. J. Roentgenol. 178: 379–387.
randomized controlled trial. JAMA 287: 1531–1540. 71 Turner-Warwick, R., Whiteside, C.G., Worth, P.H. et al.
57a Zannad, F., McMurray, J.J.V., Krum, H, et.al. (2011). (1973). A urodynamic view of the clinical problems
Eplerenone in patients with systolic heart failure and associated with bladder neck dysfunction and its
mild symptoms. N Engl J Med 364: 11–21. treatment by endoscopic incision and trans-trigonal
58 Butler, J., Emerman, C., Peacock, W.F. et al. (2004). The posterior prostatectomy. Br. J. Urol. 45: 44–59.
efficacy and safety of B-type natriuretic peptide 72 Cohen, E.P., Sobrero, M., Roxe, D.M., and Levin, M.L. (1992).
(nesiritide) in patients with renal insufficiency and Reversibility of long-standing urinary tract obstruction
acutely decompensated congestive heart failure. requiring long-term dialysis. Arch. Intern. Med. 152: 177–179.
Nephrol. Dial. Transplant. 19: 391–399. 73 Schlossberg, S.M. and Vaughan, E.D. Jr. (1984). The
59 Wang, D.J., Dowling, T.C., Meadows, D. et al. (2004). mechanism of unilateral postobstructive diuresis. J. Urol.
Nesiritide does not improve renal function in patients 131: 534–536.
with chronic heart failure and worsening serum 74 Narins, R.G. (1970). Post-obstructive diuresis: a review.
creatinine. Circulation 110: 1620–1625. J. Am. Geriatr. Soc. 18: 925–936.
107
Hepatorenal Syndrome
Adrià Juanola1, Andres Cardenas1, and Pere Ginès1,2
1
Institut de Malalties Digestives i Metabòliques, Hospital Clínic, University of Barcelona, Spain
2
Institut d’Investigacions Biomèdiques August Pi-Sunyer (IDIBAPS) y Ciber de Enfermedades Hepáticas y Digestivas (CIBEREHD), Barcelona, Spain
Renal failure is a common complication in patients with in patients with advanced cirrhosis and ascites; however, it
cirrhosis that accounts for major morbidity and mortal- occasionally occurs in patients with alcoholic hepatitis or
ity [1]. The definition of acute kidney injury (AKI) is based acute liver failure [1–4]. Although advances in medical
on small changes or a percentage increase serum creati- therapy of HRS are highly promising, liver transplantation
nine from baseline (Table 9.1). A frequent cause of AKI is remains the treatment of choice in suitable candidates.
hepatorenal syndrome (HRS), which is characterized by Several controlled and uncontrolled studies demonstrate
renal vasoconstriction and a low glomerular filtration rate that the use of splanchnic vasoconstrictors, such as terli-
(GFR) with minimal renal histologic abnormalities. pressin, midodrine, octreotide, and/or norepinephrine, in
However, establishing the diagnosis of HRS may be diffi- combination with albumin as a plasma expander are effec-
cult because there are no specific diagnostic tests or mark- tive in reducing serum creatinine and increasing GFR to
ers for this syndrome. In addition, patients with liver within normal limits. HRS can be effectively prevented in
disease may develop other types of renal impairment such the setting of spontaneous bacterial peritonitis with intra-
as those related to volume depletion, acute tubular necro- venous albumin [5, 6]. This chapter will review the patho-
sis, drug toxicity and/or glomerular diseases. HRS develops genesis, clinical features, therapy, and prevention of HRS.
Table 9.1 Definition and staging of AKI.
Definition:
P
athogenesis
●● Increase in serum creatinine 0.3 mg/dl (26.5 μmol/l) within
48 h; or HRS represents the end stage of a circulatory dysfunction
●● A percentage increase in serum creatinine of 50% which is
that occurs late in the natural history of patients with cir-
known, or presumed, to have occurred within the prior 7 days rhosis. The main characteristic of HRS is the presence of
Staging: renal vasoconstriction, which leads to a functional prere-
●● Stage 1A: An increase in serum creatinine 0.3 mg/dl nal AKI. This vasoconstriction is a consequence of a con-
(26.5 μmol/l) or an increase in serum creatinine of 1.5-fold to tinuous process where several underlying mechanisms
2-fold from baseline to a value lower than 1.5 mg/dl
(133 μmol/l) from baseline to diagnosis of AKI. interact, including changes in systemic arterial circulation,
●● Stage 1B: An increase in serum creatinine 0.3 mg/dl
increased portal pressure, impaired cardiac function, and
(26.5 μmol/l) or an increase in serum creatinine of 1.5-fold to activation of systemic and renal vasoconstrictor factors
2-fold from baseline to a value 1.5 mg/dl (133 μmol/l) from (Figure 9.1) [7].
baseline to diagnosis of AKI.
●● Stage 2: An increase in serum creatinine greater than twofold
to threefold from baseline.
●● Stage 3: An increase of serum creatinine greater than threefold
A
rterial Vasodilation
from baseline or serum creatinine 4.0 mg/dl (353.6 μmol/l)
with an acute increase 0.3 mg/dl (26.5 μmol/l) or initiation of Portal hypertension, due to an abnormal liver architecture
renal replacement therapy. and increased intrahepatic resistance of blood flow, causes
Source: Based on Angeli et al. [2]. the release of nitric oxide and other vasodilator substances
108 Hepatorenal Syndrome
Cirrhosis
Sinusoidal portal hypertension
Vasodilator
Bacterial substances (NO, CO,
DAMPs translocation endocannabinoids)
(HMGB1, HSPs and (PAMPs)
hyaluronic acid)
Cardiac
Activation of Splanchnic function
immune cells arterial impairment
Proinflammatory cytokines
(IL-6, TNFα and VCAM) vasodilatation
Reduced effective
arterial volume
Activation of endogenous
vasoconstrictor systems
(RAAS, AVP and SNS)
Inflammatory
mediators
Sodium and water retention
Renal vasoconstriction
Figure 9.1 Patients with advanced cirrhosis have a marked splanchnic arterial vasodilation triggered by portal hypertension.
Splanchnic vasodilation leads to a decreased systemic vascular resistance with the development of effective arterial hypovolemia. The
activation of vasoconstrictor systems leads to a marked renal vasoconstriction, low glomerular filtration rate, and development of
hepatorenal syndrome. In this advanced stage there is a reduced cardiac output and decreased effective arterial blood volume.
Systemic inflammation is altered as well. Pathogen-associated molecular patterns (PAMPs) and damage-associated molecular patterns
(DAMPs) of bacterial translocation may lead to a marked inflammatory response. AVP, arginine vasopressin; CO, carbon monoxide;
HMGB1, high-mobility group box 1; HSPs, heat shock proteins; NO, nitric oxide; RAAS, renin–angiotensin–aldosterone system; SNS,
sympathetic nervous system; TNF, tumor necrosis factor; VCAM, vascular cell adhesion protein.
(i.e. carbon monoxide and endogenous cannabinoids), may trigger renal vasoconstriction [6, 14]. At this stage, the
responsible for splanchnic vasodilatation [8]. Plasma vol- stimulation of the RAAS, SNS, and AVP is so intense that
ume accumulates in the splanchnic bed, causing a com- their vasoconstrictor effects cannot be overcome, and HRS
pensatory response due to a fall in central (effective) blood develops.
volume. This activates the renin–angiotensin–aldosterone
system (RAAS), the sympathetic nervous system (SNS), Reduced Cardiac Output
and arginine vasopressin (AVP), accounting for sodium
and water retention and, in advanced stages, renal vaso- Circulatory dysfunction in cirrhosis has been the classic
constriction [9–11]. The ongoing action of vasoconstrictor defining pathophysiological alteration that leads to HRS;
factors acting on the kidney eventually leads to HRS. This however, evidence suggests that a reduction in cardiac
occurs because although in the early stages of cirrhosis output may play a role as well [15]. As discussed above, a
renal blood flow may be kept within normal limits due to complex interplay between the splanchnic, systemic, and
the effect of local renal vasodilators, with time circulating renal circulation comes into play once portal hyperten-
vasoconstrictors overcome the effect of renal vasodilators, sion is established. In the initial stages of cirrhosis in the
leading to severe renal vasoconstriction and reduction in process of progressive vasodilation, cardiac output (CO)
GFR [12, 13]. In some patients precipitating factors such as increases to maintain systemic hemodynamic homeostasis.
bacterial infections worsen circulatory dysfunction and However, with progression to decompensated cirrhosis the
Diagnostic Approac 109
progressive effects of splanchnic vasodilatation cannot be area, leading to a further vasodilation of the splanchnic
overcome by increasing CO. Moreover, in patients with circulation (Figure 9.1) [26, 27].
ascites or HRS, CO may be normal or even reduced, which Bacterial infections, particularly SBP, are one of the lead-
is the extreme expression of the circulatory dysfunction [1, ing triggers of HRS. Patients who develop HRS associated
16, 17]. A consequent decrease in effective arterial blood with SBP have higher levels of IL-6 and TNFα compared
volume may occur, thereby reducing renal perfusion and with those patients with SBP who do not develop HRS [24].
possibly contributing to the development of HRS. In fact, Moreover, a recent study suggested that HRS
the development of HRS occurs in the setting of a reduc- is characterized by an elevated systemic inflammatory
tion in CO and the progression of circulatory and renal dys- response with higher serum levels of proinflammatory
function in cirrhosis is not only due to splanchnic cytokines, particularly IL-6, TNF-α, and VCAM, compared
vasodilation but also due to a drop in CO. [18–20] In a to pre-renal AKI [28].
study of 23 patients with SBP, eight patients who developed
renal failure had a significantly reduced cardiac output
compared to those that did not develop renal failure [18]. R
enal Vasoconstriction
In a second study, 66 patients with cirrhosis and ascites
were longitudinally followed. Twenty-seven (41%) devel- Marked renal vasoconstriction in individuals with HRS
oped HRS, and in all cases CO at the time of HRS was sig- causes reduced perfusion of the renal cortex and this is evi-
nificantly decreased compared to the CO in the same denced by methods such renal arteriography, duplex
patients measured before the occurrence of HRS [19]. The Doppler ultrasonography, and paraaminohippuric acid
third study, which included 24 patients with decompen- excretion among others [29–35]. Renal vasoconstriction is
sated cirrhosis, showed that in patients with a cardiac mainly due to the activation of systemic and local vasocon-
index (CI) below 1.5 l/min/m2, GFR and renal blood flow strictor factors and the suppression of renal vasodilators.
were lower and serum creatinine was higher. Moreover, the Increased angiotensin II as a result of activation of the
number of patients that developed HRS was higher in the RAAS causes renal vasoconstriction [19, 36, 37]. Infections
group with low CI [20]. These studies support the concept such as SBP which can trigger HRS are associated with sig-
that HRS occurs in the setting of worsening circulatory nificant activation of the RAAS. The SNS also plays a role
function, with a significant decrease in CO in the setting of and plasma levels of noradrenaline are greatly increased in
marked splanchnic vasodilation [18–20]. The exact patho- patients with HRS compared to those of patients with cir-
genic mechanisms that cause this decrease in cardiac out- rhosis without HRS. In addition, noradrenaline levels cor-
put are not completely understood. relate inversely with GFR [38]. Other vasoconstrictor
factors such as endothelin and cysteinyl leukotrienes may
also play a pathogenic role [39, 40]. A reduced production
S
ystemic Inflammation of endogenous renal vasodilators such prostaglandins has
been suggested as playing a pathogenic role in many
Recent evidence indicates that cirrhosis is associated with patients with cirrhosis and ascites. This is supported by
increased systemic inflammation and that the grade of the observation that patients with cirrhosis who receive
inflammation is higher with more severe liver, circulatory, NSAIDs, which inhibit prostaglandin synthesis, can
and renal dysfunction [21]. Patients with cirrhosis may develop AKI resembling HRS [41].
have elevated white blood cell count, activated circulating
neutrophils and monocytes, plasma C-reactive protein (C-
RP), pro-inflammatory cytokines, and systemic oxidative D
iagnostic Approach
stress. The highest levels of inflammatory cytokines in
patients with decompensated cirrhosis are seen in those Patients with cirrhosis may develop AKI for a variety of
with acute on chronic liver failure (ACLF), in which AKI reasons aside from HRS. Other forms of AKI include those
frequently occurs [22–25]. related to hypovolemia and shock, the use of nephrotoxic
The main mechanism of systemic inflammation is the drugs, and intrinsic renal diseases. AKI occurs in 20–50%
translocation of bacteria and/or pathogen-associated molec- of patients with cirrhosis admitted to hospital for compli-
ular patterns (PAMPs) from intestinal lumen to systemic cations of the disease. However, HRS appears in about
circulation. Bacterial translocation induces an inflammatory 10–20% of hospitalized patients with cirrhosis and
response, with increased production of proinflammatory ascites [3]. The probability of developing HRS in patient
cytokines (particularly interleukin 6 and tumor necrosis with cirrhosis and ascites is 18% at 1 year and 39% at 5 years
factor-α) and vasoactive factors (i.e. NO) in the splanchnic of follow-up [42].
The initial management of patients with a rising serum nine production because of reduced muscle mass a low
creatinine depends on the cause. Therefore, the most serum creatinine level may result in an overestimation of the
important step in treating renal failure in a patient with GFR [50–52]. A drawback of this cutoff was that it identified
cirrhosis is to identify its etiology. However, even in the patients with a severely reduced GFR (<30 ml per minute)
absence of a definitive recognized cause of AKI, the man- and underestimated the prevalence of this clinical problem.
agement should be immediately started according to the Accordingly, the definition of renal failure in patients with
initial stage (Table 9.1). Irrespective of the stage, diuretics cirrhosis was modified [2]. AKI is now defined as either an
should be discontinued. Similarly, beta-blockers should be absolute increase in serum creatinine of more than or equal
temporarily stopped [43]. Other precipitating factors of to 0.3 mg/dl ( 26.4 lmol/l) in less than 48 hours, or by a per-
AKI should be identified and treated, including screening centage increase in serum creatinine of more or equal to 50%
and treatment of infection, volume expansion with albu- (1.5-fold from baseline) in less than 7 days. A new staging
min when appropriate, Box 9.1 and discontinuation of all system (1A, 1B, 2, and 3) was also introduced, mainly based
nephrotoxic drugs [2]. Volume replacement should be used on the percentage increase of serum creatinine from base-
in accordance with the cause and the severity of fluid loss. line at the time of the first fulfillment of the criteria together
Patients with diarrhea or excessive diuresis should be with the value of [53, 54]. This definition has the advantage
treated with crystalloids, while those with acute gastroin- of detecting earlier phases of kidney dysfunction with the
testinal bleeding should be given packed red blood cells to goal of implementing early therapy. Minor increases in
maintain hemoglobin level between 7 and 9 g/dl [44]. In serum creatinine detected with the AKI definition are inde-
case of no obvious cause, 20% or 25% albumin solution at pendently associated with increased mortality for hospital-
the dose of 1 g of albumin/kg of body weight (with a maxi- ized patients with cirrhosis [55].
mum of 100 g of albumin) for two consecutive days should Several studies have shown that the new AKI criteria and
be given. An algorithm for diagnosis and management of its stages are prognostic as AKI is associated with increased
AKI in cirrhosis is depicted in Figure 9.2. length of hospital stay and mortality [54, 56, 57]. The defi-
nition of the new ICA-AKI criteria also led to changes in
the diagnostic criteria of HRS-AKI (Table 9.2). Traditionally
C
linical Features HRS was classified into two entities according to severity
and progression of kidney failure: type-1 HRS and type-2
Individuals with HRS for the most part exhibit clinical fea- HRS [49]. In type 1, patients had a rapid increase of serum
tures of advanced cirrhosis, and most have high Model for creatinine to a level >2.5 mg/dl in <2 weeks and type 2 was
End-stage Liver Disease (MELD) and Child–Pugh scores, characterized with lower serum creatinine levels usually
along with low arterial blood pressure and marked urinary ranging from 1.5 to 2.5 mg/dl. The main change that was
sodium retention (urine sodium, 10 mEq/l) [42, 46]. The made to the classical diagnostic criteria of HRS was the
MELD score is a mathematically derived score calculated removal of the cutoff value of serum creatinine [2, 45] to
from serum bilirubin, serum creatinine, and the interna- allow an earlier diagnosis of HRS-AKI. Therefore, the cur-
tional normalized ratio for prothrombin time used for organ rently HRS-AKI, includes patients with classical type-1
allocation by liver transplant centers [47]. An updated HRS criteria (serum creatinine 2.5 mg/dl), but also those
MELD score including the serum sodium as a variable patients presenting with lower values of serum creatinine
(MELD-Na) has been implemented in the United States and that fulfill the remaining HRS-AKI criteria and that would
other countries. This is mainly due to the fact that serum not have been included in the classical definition of type-1
sodium is a strong prognostic factor in patients with HRS. On the other hand, the term type-2 HRS is not
advanced liver disease and ascites. In fact, hypervolemic included in the current concept of HRS-AKI because is not
hyponatremia (serum sodium, 130 mEq/l) is commonly an acute impairment but rather a chronic impairment of
present in patients with HRS due to increased solute-free kidney function. Therefore, type-2 HRS is currently consid-
water retention because of elevated levels of AVP [48]. ered a form of chronic kidney disease (HRS-CKD) that is
characteristic of cirrhosis [2, 45, 58].
A
KI and HRS-AKI Criteria U
rinary Biomarkers
Renal impairment in patients with cirrhosis was previously Differentiating HRS and ATN is quite challenging given that
defined by a serum creatinine value 1.5 mg/dl [49]. both conditions are common in hospitalized or critically ill
However, because patients with cirrhosis have a low creati- patients with cirrhosis. The routine urine biomarkers for
Urinary Biomarker 111
Assessment of AKI stage according to modified ICA-AKI criteria
Stage 1A AKI Stage 1B, stage 2 or stage 3 AKI
• Close monitoring
• Remove risk factors (nephrotoxic • Remove risk factors (nephrotoxic
drugs, vasodilators, NSAIDs and drugs, vasodilators, NSAIDs and
diuretics) diuretics)
• Treatment of infections, if present • Volume expansion with albumin

(1 g/kg) for 2 days
• Plasma volum expansion, if fluid
loses
Resolution Stable Progression Resolution Stable or progression
Close follow-up Criteria for HRS-AKI
No Yes
Further treatment of AKI decided on a Renal replacement therapy, if Vasoconstrictors

case-by-case basis required (terlipressin) plus
albumin
Figure 9.2 An algorithm for diagnosis and management of acute kidney injury (AKI) in cirrhosis. AKI-HRS, AKI-hepatorenal syndrome;
ICA-AKI, International Club of Ascites-AKI. aReturn of serum creatinine levels to <0.3 mg/dl from baseline. bIntravenous noradrenaline
can be used if terlipressin is not available. The combination of midodrine, octreotide, and albumin is much less effective than
terlipressin or noradrenaline plus albumin. Source: Modified with permission from Angeli et al. [45].
ATN (urine sodium, fractional excretion of sodium or urine Table 9.2 New diagnostic criteria of acute kidney injury
osmolality) are not very useful in patients with cirrhosis (AKI)-HRS.
because these can be modified by diuretics and also urinary
sodium in advanced cirrhosis may be markedly low owing to ●● Cirrhosis with ascites
increased sodium retention [51]. Currently, there are no ●● Diagnosis of AKI according to International Club of Ascites-
Acute Kidney Injury (ICA-AKI) criteria
objective tools with high accuracy for differential diagnosis
●● No response after two consecutive days of diuretic withdrawal
of AKI, particularly between ATN and HRS-AKI. In this set- and plasma volume expansion with albumin
ting, urine biomarkers seem to be very useful in the differen- ●● Absence of shock
tial diagnosis and distinguishing structural from functional ●● No current or recent use of nephrotoxic drugs (NSAIDs,
causes of AKI. The most promising biomarkers are urinary aminoglycosides or iodinated contrast media)
neutrophil gelatinase-associated lipocalin (uNGAL), fol- ●● No signs of structural kidney injury. Structural kidney injury is
lowed by interleukin-18 (IL-18), albumin, and Liver fatty indicated by proteinuria (>500 mg/d), microhaematuria (>50
red blood cells per high-power field), and/or abnormal renal
acid-binding protein (L-FABP) [59–63]. ultrasonography
The utility of biomarkers in the field of cirrhosis may be
particularly relevant in the differential diagnosis of the Source: Based on Angeli et al. [2].
type of AKI, especially in distinguishing ATN from prere-
nal AKI or HRS. The urinary concentration of biomarkers prerenal-AKI. uNGAL may help identifying the cause of
which are upregulated in the setting of tubular injury are AKI in patients with cirrhosis, especially in differentiating
increased in patients with ATN compared to patients with ATN from HRS [63, 64]. Patients with hypovolemia and
AKI have low uNGAL levels. The values in patients with receive plasma expansion when ruling out other causes
HRS are in between those of patients with hypovolemia of renal failure, but the routine use of central pressure
and AKI and the markedly elevated levels observed among monitoring has yet to be established. Given that patients
patients with ATN. Moreover, studies suggest that elevated are frequently malnourished and require a sodium-
uNGAL is predictive of early mortality in patients with restricted diet, a nutritionist should be a part of the team
AKI [61]. IL-18 has also been evaluated in patients with taking care of the patient. Finally, probably the most
cirrhosis, with higher urinary IL-18 levels in patients with important aspect of providing care to patients with HRS
ATN as compared to non-ATN AKI [62]. Moreover, urinary is assessment of candidacy for liver transplantation. In
IL-18 has also been shown to have a relatively good ability suitable candidates for liver transplantation, all efforts
to predict need for renal replacement therapy (RRT) and should be made to improve renal function to obtain a
mortality during hospitalization [61]. Plasma levels of better outcome after transplantation. Current available
cystatin C seem to be also predictive of AKI in patients therapies for HRS-AKI include the use of splanchnic
with cirrhosis but there are scarce studies. Other urine bio- vasoconstrictors and albumin and transjugular intrahe-
markers, including albumin and liver fatty acid binding patic portosystemic shunt (TIPS).
protein have also demonstrated that they have the ability to
distinguish ATN from non-ATN in patients with cirrhosis
and AKI [59, 62]. P
harmacological Therapy
Although in the past when considering treatment the type

M
anagement of HRS was taken into account, currently it is considered
that there is no specific cutoff value of serum creatinine
The first step of management of patient with HRS is levels for the diagnosis and treatment of HRS-AKI. That
optimization of clinical status with adequate manage- said, vasoconstrictor therapy is recommended for those
ment of fluid balance and blood pressure. Moreover, all individuals that meet HRS-AKI stage 1B criteria or greater.
patients should have blood, urine, and ascites cultured There is very limited data on the efficacy and side effects of
to rule out bacterial infections. Diuretics should be vasoconstrictor therapy in HRS-AKI stage 1a. Systemic
stopped and low volume therapeutic paracentesis (<5 l) vasoconstrictors given with plasma expansion are cur-
with albumin performed if needed [2, 43, 44] for man- rently the most effective method form of pharmacologic
agement of symptoms. Patients need to be hospitalized therapy for HRS [45]. Splanchnic vasoconstrictors are used
in a closely monitored acute care unit. A central line in conjunction with albumin because they counteract
may be helpful in assessing volume status after patients the intense vasodilation of the splanchnic circulation
Box 9.1 Treatment recommendations box, based on GRADE, for AKI-HRS
Comment Certainty of evidence Strength of Recommendation
Patients meeting the current definition of HRS-AKI stage > 1A should be Moderate Strong
treated with vasoconstrictors and albumin
Terlipressin can be administered by IV boluses (1 mg every 4–6 h) or by High Strong

continuous IV infusion (2 mg/day)†
●● In case of nonresponse (decrease in SCr <25% from the peak value)
after 2 days, the dose of terlipressin should be increased in a stepwise
fashion to a maximum of 12 mg/day
Albumin solution (20% or 25%) should be used at 20–40 g/day Moderate Strong

●●Serial assessment of central blood volume aids in titrating the dose
Noradrenaline can be an alternative to terlipressin‡ High Weak

●● Requires a central venous line often in an ICU Moderate Weak
idodrine + octreotide can be an option when terlipressin or

M
noradrenaline are unavailable (but efficacy is lower)
Source: Modified with permission from EASL CPG decompensated cirrhosis [45].

Pharmacological Therap 113
and improve effective arterial blood volume, which in turn than nonresponders. In two meta-analyses, the use of terli-
suppresses the endogenous vasoconstrictor factors respon- pressin was associated with a significant improvement in
sible for HRS. Vasoconstrictor drugs used in HRS include short-term survival [80, 83].
terlipressin, noradrenaline, and the combination of Albumin infusion along with terlipressin is always rec-
midodrine and octreotide. Albumin is used concomitantly ommended. The combination of albumin and terlipressin
with these vasoconstrictors as a means of improving effec- is more effective than terlipressin alone [78]. The dose of
tive arterial blood volume and therefore further suppress- albumin in HRS treatment has not been well established,
ing endogenous vasoconstrictor factors (RAAS and SNS) but in most studies 20–40 g/d are given. However, the
responsible for the intense renal vasoconstriction and dose of albumin administered is determined by the level
reduced cardiac contractility in HRS [65–76]. of central venous pressure. Aside from an increase in sys-
Pharmacological treatment recommentdations are sum- temic vascular resistance and cardiac output, albumin
marized in Box 9.1. has nononcotic effects including antioxidative and anti-
inflammatory properties [84–87].
Diarrhea and abdominal cramps are common adverse
Vasoconstrictors (Table 9.3)
effects of terlipressin (up to 10%). Other more serious
Terlipressin adverse events are uncommon (<5%), including ischemic
Terlipressin is the most commonly used drug for HRS in and cardiovascular events. Although rare, patients can
countries in which it is available. It is a synthetic analog of develop angina, intestinal, digital or glossal ischemia, car-
vasopressin and several studies and at least six randomized diac arrhythmias, or severe hypertension. Patients with
controlled trials have proven the efficacy of terlipressin established ischemic heart disease and peripheral vascular
plus albumin in the treatment of HRS [65–68, 73, 76, 78]. disease should not be treated with terlipressin.
According to a recent meta-analysis, reversal of HRS occurs
in approximately 35–50% of patients with terlipressin plus Norepinephrine, Midodrine, and Octreotide
albumin vs. 11–14% of patients treated with placebo [79]. The use of vasoconstrictors other than terlipressin is an
Also, an improvement in renal function was observed in alternative in countries where terlipressin is not available.
48% of patients with terlipressin plus albumin against 24% Noradrenaline is an (alfa)-adrenergic and (beta)-adrenergic
of patients receiving placebo [79, 80]. Terlipressin plus receptor agonist with vasoconstrictor effect activity on
albumin seems to confer a survival benefit against placebo, splanchnic circulation that can also improve renal perfu-
but there are some conflicting results [80, 81]. sion. It has been evaluated in four randomized controlled
In the initial studies, terlipressin was administered by trials compared to terlipressin [69–72]. A meta-analysis
repeated intravenous boluses (starting dose of 0.5–1 mg compared terlipressin plus albumin with noradrenaline
every 4–6 hours, increasing to a maximum of 2 mg every plus albumin and found no difference in the reversal of
4–6 hours in case of a reduction of baseline serum creati- HRS, 30-day mortality and recurrence of HRS [80].
nine <25%) but a randomized trial compared the efficacy Although data is limited due to the low number of patients
and safety of terlipressin given by continuous intravenous included in the studies (n = 78), noradrenaline seems to be
infusion (dose 2 mg/d up to 12 mg/d) against intravenous as effective as terlipressin in the management of HRS and
boluses [82]. The study showed similar response rates represents a good alternative treatment [80], however the
between both groups. Moreover, mean effective dose of ter- quality of evidence supporting the use of noradrenaline is
lipressin was lower in the continuous infusion group, and low according to a meta-analysis [80]. Noradrenaline must
that was associated with a lower rate of adverse events. The be administered under continuous monitoring and in some
dose of terlipressin should be increased in a stepwise man- countries it cannot be used outside intensive care units.
ner if serum creatinine does not decrease by at least 25% Midodrine, a selective (alfa)-1-adrenergic receptor agonist,
with respect to the pretreatment value after 3 days of treat- and octreotide, a somatostatin analog, have been evaluated in
ment. Treatment with terlipressin plus albumin should be the management of HRS; there are no placebo-controlled tri-
continued until serum creatinine is <1.5 mg/dl or close to als and only nonrandomized observational cohort studies in
the baseline value before HRS-AKI. In patients with no very small series of patients demonstrate a mild improve-
response or only partial response the treatment should be ment in renal function and GFR with suppression of renin,
discontinued within 2 weeks. Recent studies showed that aldosterone, norepinephrine, and vasopressin to normal or
55.5% of patients with HRS had a complete response to ter- near-normal levels [88, 89]. However, this data was not con-
lipressin plus albumin, defined as a reduction of serum cre- firmed in a recent randomized control trial, where midodrine
atinine by >50% to a final value of <1.5 mg/dl [76, 82]. and octreotide plus albumin were compared to terlipressin
Patients who respond to treatment have a better survival plus albumin, and significant higher response rates were
Table 9.3 Studies evaluating vasoconstrictors in patients with HRS.
Reversal HRS Reversal HRS in Treatment related severe Median survival, control
Study (reference) Study design (%)a Recurrenceb control group side effectsc (%) Median survival (days) group (days)
Terlipressin vs. placebo or control

Prashant et al. [65] Single-center, single blind, 5/12 NA 0/12 3/12 NA NA
placebo-controlled,
randomized trial
Neri et al. [66] Single-center, open-label, 21/26 NA 5/26 2/52 87% at day 15, 72% at day 53% at day 15, 42% at day
randomized controlled trial 30, 67% at day 60, 54% at 30, 21% at day 60, 18% at
day 90, 42% at day 180 day 90, 16% at day 180
Sanyal et al. [67] Multicenter, double-blind, 19/56 1/9 7/56 5/56 24/56 at day 180 21/56 at day 180
placebo-controlled trial
Martín-Llahí Multicenter, open-label, 9/23 1/10d 1/23 10/23 6/23 at day 90 4/23 at day 90
et al. [73] randomized controlled trial
Zafar et al. [77] Single-center, randomized 10/25 NA 2/25 NA NA NA
controlled trial
Boyer et al. [68] Multicenter, double-blind, 19/97 3/19 13/99 14/93e 23.8 20.7 days
placebo-controlled, 56/97 at day 90 54/99 at day 90
randomized trial
Noradrenaline vs. terlipressin
Alessandria Single-center, open-label, 7/10 2/7 10/12 0/10 NA NA
et al. [72] randomized controlled trial
Indrabi et al. [69] Single-centre, randomized 16/30 1/16 17/30 NA 37.7 ± 49.9 45.6 ± 44.1
controlled trial
Sharma et al. [70] Single-center, open-label, 10/20 NA 10/20 5/40 15 13
randomized controlled trial
Singh et al. [71] Single-center, open-label, 9/23 0/9 10/23 0/46 9/23 at day 15, 11/23 at day 15, 8/23 at
randomized controlled trial 7/23 at day 30 day 30
Midodrine plus octreotide vs. terlipressin
Cavallin et al. [76] Multicenter, open-label, 1/21 15/27 13/48 14/21 at day 30, 9/21 at 19/27 at day 30, 16/27 at
randomized controlled trial day 90 day 90
Midodrine plus octreotide vs. noradrenaline
Tavakkoli et al. [75] Single-center, open-label, 9/12 3/9 8/11 0/23 NA NA
randomized controlled trial
a
The definition of response varied between studies, but in all studies it was defined as a marked decrease in serum creatinine.
b
Recurrence of HRS after treatment withdrawal in responder patients; the definition of recurrence also varied between studies.
c
Many patients presented with self-limited abdominal cramps and/or diarrhea during the administration of the first doses of terlipressin, which were not counted as severe side effects.
d
10 out of 23 improve renal function, nine with complete response.
e
Eight patients (four in each treatment group) did not receive study medication primarily because of declining health and SCr levels that did not meet inclusion criteria after randomization.
0005152397.INDD 114 09-12-2022 14:19:08

Nonpharmacological Therap 115
found in the terlipressin group (70.4% vs. 28.6%) [76]. serum creatinine is a strong variable in the mathematical
Therefore, terlipressin is clearly more effective than the com- model [47]. The presence of HRS-AKI at the time of liver
bination of midodrine and octreotide. transplantation has a negative impact on survival after the
operation. Patients with HRS who undergo transplantation
have more complications, spend more days in the intensive
Aim of Pharmacological Therapy care unit, and have a higher in-hospital mortality rate than
transplanted patients without prior HRS [94].
The most important objective when treating HRS-AKI is Although liver transplantation is the treatment of
to reverse renal failure to provide a successful bridge to choice, renal function does not completely recover in all
liver transplant so that suitable candidates can undergo patients [94, 95]. The reason for this is not known.
transplantation with less morbidity and side effects to Simultaneous liver-kidney (SLK) has been considered
post-transplant medications and have similar outcomes for patients with HRS. Recent recommendations for SLK
as patients without HRS-AKI. As mentioned above, in the United States include patients with AKI with an
response to therapy is considered when there is marked estimated GFR of 25 ml/min/1.73 m2 for 6 weeks or a
reduction of the high serum creatinine levels, at least period of dialysis 6 weeks, patients with an eGFR
below 1.5 mg/dl, which is usually associated with of < 45 ml/min/1.73 m2 for >90 days (stage 3B, 4, or 5
increased urine output and improvement of hypona- CKD), or patients with comorbidities and presence of
tremia [45]. Studies suggest that response to treatment metabolic diseases [96]. European guidelines suggest
with vasoconstrictors and albumin correlates with an that patients with end-stage liver disease who have a
increase in mean arterial pressure (MAP). One study GFR of <30 ml/min/1.73 m2 or type 1 HRS requiring
showed that patients who experienced a significant RRT of >8–12 weeks and patients with renal biopsy sam-
increase in MAP during terlipressin treatment had a ples revealing >30% fibrosis and glomerulosclerosis
higher probability of recovering kidney function [90]. should receive simultaneous liver and kidney trans-
Another study also reported that responders to terlipres- plant [45]. SLK is also indicated in patients with cirrho-
sin had a higher increase in MAP at the end of treatment sis and sustained AKI irrespective of its type, including
compared with nonresponders. The improvement of kid- HRS-AKI when refractory to drug therapy if requiring
ney function in patients with HRS treated with vasocon- RRT for 4 weeks, or if the estimated GFR is 35 ml/
strictors tightly correlates with the rise in MAP. Therefore, min/1.73 m2 or measured GFR 25 ml/min/1.73 m2 for
a goal-directed approach to the treatment of HRS based 4 weeks [45]. Beyond these two conditions, in a candi-
on targeting a rise in MAP during vasoconstrictor therapy date with high priority for liver transplantation due to a
may lead to better outcomes [90–92]. Nevertheless, pro- high MELD score, the option of SLK may be considered
spective studies evaluating this approach are needed in the presence of risk factors for underlying undiag-
before integrating it into practice. Recurrence of HRS in nosed CKD (diabetes, hypertension, abnormal renal
responders, after the end of the treatment, has been imaging, and proteinuria >2 g/d) [45].
reported in up to 20% of cases. Re-treatment is usually
effective, but in some cases continuous recurrence occurs.
Transjugular Intrahepatic Portosystemic Shunt
Patients who respond to treatment with terlipressin plus
albumin show a better survival than nonresponders to Transjugular intrahepatic portosystemic shunt (TIPS) is
treatment. a nonsurgical intervention performed percutaneously
that creates a connection (by means of a prosthetic
shunt) within the liver between the portal and hepatic
vein, thereby reducing portal pressure. This method has
N
onpharmacological Therapy been used as an alternative therapy to surgical shunts for
cirrhotic patients bleeding from esophageal or gastric
Liver Transplantation
varices who are refractory to endoscopic and medical
Liver transplantation is the best available effective therapy treatment. The rationale for using TIPS in HRS is based
for patients with HRS-AKI since it cures liver failure and on the fact that reduction of portal pressure improves cir-
portal hypertension. Since patients with HRS-AKI have a culatory function and suppresses RAAS and SNS activity.
poor prognosis, this group of patients should be given higher Four uncontrolled studies indicate that TIPS may improve
priority for transplantation [93]. In fact, the implementation renal function and GFR as well as reduce the activity of
of the MELD score in many countries has placed these RAAS and SNS in patients with cirrhosis and HRS.
patients at a higher priority level for liver transplant because Improvement in renal function observed in these uncontrolled
studies after TIPS placement alone was slow and successful P

revention
in approximately 60–70% of patients. However, studies
assessing TIPS for type 1 HRS have only included patients In SBP, intravenous albumin (1.5 g/kg at diagnosis of infec-
with preserved liver function and excluded those with a tion and 1 g/kg 48 hours later) mitigates the risk of develop-
history of hepatic encephalopathy, Child–Pugh scores of ment of HRS because it counteracts an enhanced arterial
12, or serum bilirubin of >5 mg/dl. As existing data are splanchnic vasodilation triggered by the infection, causing an
limited and the applicability of TIPS in these patients is even more reduced effective arterial blood volume with addi-
very low, TIPS placement should not be recommended in tional activation of vasoconstrictor systems. The incidence
the setting of HRS-AKI [89, 97, 98]. of HRS in patients with SBP not administered albumin
was 33% and only 10% in those who received albumin.
In addition, there was less in-hospital mortality in those
Renal Replacement Therapy receiving albumin (10%) than in those not receiving
albumin (29%) [6]. Albumin is not effective in the preven-
RRT is not recommended as a first-line treatment in
tion of HRS associated with infections other than
patients with HRS-AKI, but it should be considered in
SBP [103, 104]. Prophylaxis with norfloxacin in patients
nonresponders to vasoconstrictors. Indications for RRT
with cirrhosis and ascites reduces the incidence of SBP
are the same in patients with cirrhosis as in the general
and delays the development of HRS, but can increase the
population, including severe and/or refractory electro-
risk of infections caused by multidrug-resistant organ-
lyte or acid-base imbalance, severe or refractory volume
isms [105]. Previous studies suggested that the use of pen-
overload, and/or symptomatic azotemia. However, pub-
toxyfilline decreased the incidence of renal failure in
lished data on RRT in patients with cirrhosis are scant,
patients with alcoholic hepatitis. However, recent data do
with controversial effects on survival [99, 100].
not confirm these results [106–109].
Alternative dialysis methods such as the molecular
adsorbent recirculating system (MARS) and albumin
dialysis have been evaluated. These systems remove S
ummary
protein-bound solutes from plasma, such as bilirubin,
bile acids, and cytokines, that are not well removed by HRS-AKI is a devastating complication of patients with cir-
conventional hemodialysis or hemofiltration. Although rhosis and ascites. Triggers of AKI such the use diuretics
these treatments may have some potential beneficial and nephrotoxic drugs, infections, hypovolemia, and
effect, they should still be considered experimental and parenchymal kidney disease should be sought for and
further controlled studies are clearly needed before promptly managed. Differential diagnosis of HRS-AKI
advocating their use in the management of patients with from other forms of AKI, such as ATN, is often difficult.
HRS [101, 102]. Treatment guidelines do not recom- Specific biomarkers such as NGAL or IL-18 may aid in the
mend MARS for the management of HRS in clinical correct diagnosis of AKI in cirrhosis but have not yet been
practice. introduced into clinical routine. Vasoconstrictors such as
Hepatorenal Syndrome
Evaluate for liver transplantation
Candidate Not a candidate
High priority for liver transplantation

Vasoconstrictors + Albumin
Vasoconstrictors + Albumin * case by case basis
Response Consider renal replacement

no
therapy
yes No Response
Stop after complete response Consider renal replacement

or until a maximum of 14 days therapy in selected cases
Figure 9.3 Treatment algorithm of AKI-HRS depending on whether the patient can undergo liver transplantation.
Reference 117
terlipressin and noradrenaline in combination with intra- A

cknowledgment
venous albumin are the first-line therapy for HRS-AKI.
Liver transplantation is considered a definitive and effec- The authors would like to thank Miss Nicola van Berckel
tive treatment for HRS and patients with cirrhosis and for administrative support.
HRS-AKI should be prioritized for liver transplantation.
An algorithm for management of AKI in cirrhosis is sum-
marized in Figure 9.3.
R
eferences
1 Ginès, P. and Schrier, R.W. (2009). Renal failure in Dysfunction in Liver Disease. Wiley Online Books. Available
cirrhosis. N. Engl. J. Med. 361 (13): 1279–1290. Available from: https://doi.org/10.1002/9780470987476.ch5.
from: https://doi.org/10.1056/NEJMra0809139. 11 Schrier, R.W. (2006). Water and sodium retention in
2 Angeli, P., Ginès, P., Wong, F. et al. (2015). Diagnosis and edematous disorders: role of vasopressin and aldosterone.
management of acute kidney injury in patients with Am. J. Med. 119: S47–S53.
cirrhosis: revised consensus recommendations of the 12 Schrier, R.W., Arroyo, V., Bernardi, M. et al. (2005).
International Club of Ascites. J. Hepatol. 62 (4): 968–974. Peripheral arterial vasodilation hypothesis: a proposal for
Available from: http://dx.doi.org/10.1016/j. the initiation of renal sodium and water retention in
jhep.2014.12.029. cirrhosis. Hepatol. Int. 8 (5): 1151–1157. Available from:
3 Garcia-Tsao, G., Parikh, C.R., and Viola, A. (2008). Acute https://doi.org/10.1002/hep.1840080532.
kidney injury in cirrhosis. Hepatol. Int. 48 (6): 2064–2077. 13 Arroyo, V., Terra, C., and Ginès, P. (2006). New treatments
Available from: https://aasldpubs.onlinelibrary.wiley. of Hepatorenal syndrome. Semin. Liver Dis. 26 (03):
com/doi/abs/10.1002/hep.22605. 254–264.
4 Ginès, P., Guevara, M., Arroyo, V., and Rodés, J. (2003). 14 Terra, C., Guevara, M., Torre, A. et al. (2005). Renal
Hepatorenal syndrome. Lancet 362 (9398): 1819–1827. failure in patients with cirrhosis and sepsis unrelated to
Available from: http://www.sciencedirect.com/science/ spontaneous bacterial peritonitis: value of MELD score.
article/pii/S0140673603149033. Gastroenterol. Int. 129 (6): 1944–1953. Available from:
5 Solé, C., Pose, E., Solà, E., and Ginès, P. (2018). http://www.sciencedirect.com/science/article/pii/
Hepatorenal syndrome in the era of acute kidney injury. S0016508505018500.
Liver Int.. Available from: https://doi.org/10.1111/liv.13893. 15 Ginès, P., Guevara, M., and Perez-Villa, F. (2004).
6 Sort, P., Navasa, M., Arroyo, V. et al. (1999). Effect of Management of hepatorenal syndrome: another piece of
intravenous albumin on renal impairment and mortality the puzzle. Hepatol. Int. 40 (1): 16–18. Available from:
in patients with cirrhosis and spontaneous bacterial https://doi.org/10.1002/hep.20313.
peritonitis. N. Engl. J. Med. 341 (6): 403–940. Available 16 Angeli, P. and Merkel, C. (2008). Pathogenesis and
from: https://doi.org/10.1056/NEJM199908053410603. management of hepatorenal syndrome in patients with
7 Møller, S. and Bendtsen, F. (2017). The pathophysiology cirrhosis. J. Hepatol. 48: S93–S103. Available from: http://
of arterial vasodilatation and hyperdynamic circulation in www.sciencedirect.com/science/article/pii/
cirrhosis. Liver Int. 38 (4): 570–580. Available from: S0168827808000561.
https://doi.org/10.1111/liv.13589. 17 Arroyo, V., Fernandez, J., and Ginès, P. (2008).
8 Iwakiri, Y. and Groszmann, R.J. (2006). The hyperdynamic Pathogenesis and treatment of hepatorenal syndrome.
circulation of chronic liver diseases: from the patient to the Semin. Liver Dis. 28 (01): 81–95.
molecule. Hepatol. Int. 43 (S1): S121–S131. Available from: 18 Ruiz-del-Arbol, L., Urman, J., Fernández, J. et al. (2003).
https://doi.org/10.1002/hep.20993. Systemic, renal, and hepatic hemodynamic derangement
9 Bernardi, M. and Domenicali, M. (2007). The renin- in cirrhotic patients with spontaneous bacterial
angiotensin-aldosterone system in cirrhosis [internet]. In: peritonitis. Hepatol. Int. 38 (5): 1210–1218. Available
Ascites and Renal Dysfunction in Liver Disease. Wiley from: https://doi.org/10.1053/jhep.2003.50447.
Online Books. Available from: https://doi. 19 Ruiz-del-Arbol, L., Monescillo, A., Arocena, C. et al.
org/10.1002/9780470987476.ch4. (2005). Circulatory function and hepatorenal syndrome
10 Dudley, F.J. and Esler, M.D. (2007). The sympathetic in cirrhosis. Hepatol. Int. 42 (2): 439–447. Available from:
nervous system in cirrhosis [internet]. In: Ascites and Renal https://doi.org/10.1002/hep.20766.
20 Krag, A., Bendtsen, F., Henriksen, J.H., and Møller, S. 30 Epstein, M., Berk, D.P., Hollenberg, N.K. et al. (1970).
(2010). Low cardiac output predicts development of Renal failure in the patient with cirrhosis. Am. J. Med. 49
hepatorenal syndrome and survival in patients with (2): 175–185. Available from: https://doi.org/10.1016/
cirrhosis and ascites. Gut 59 (01): 105–110. Available S0002-9343(70)80073-0.
from: http://gut.bmj.com/content/59/01/105.abstract. 31 Gonwa, T., Klintmalm, G., Levy, M. et al. (1995). Impact
21 Bernardi, M., Moreau, R., Angeli, P. et al. (2015). of pretransplant renal function on survival after liver
Mechanisms of decompensation and organ failure in transplantation. Transplantation 59 (3): 361–365.
cirrhosis: from peripheral arterial vasodilation to systemic 32 Schroeder, E.T., Shear, L., Sancetta, S.M., and Gabuzda,
inflammation hypothesis. J. Hepatol. 63 (5): 1272–1284. G.J. (1967). Renal failure in patients with cirrhosis of
Available from: http://www.sciencedirect.com/science/ the liver: III. Evaluation of intrarenal blood flow by
article/pii/S0168827815004663. Para-aminohippurate extraction and response to
22 Waidmann, O., Brunner, F., Herrmann, E. et al. (2013). angiotensin. Am. J. Med. 43 (6): 887–896. Available from:
Macrophage activation is a prognostic parameter for variceal http://www.sciencedirect.com/science/article/
bleeding and overall survival in patients with liver cirrhosis. pii/0002934367902471.
J. Hepatol. 58 (5): 956–961. Available from: http://www. 33 Kew, M.C., Varma, R.R., Williams, H.S. et al. (1971).
sciencedirect.com/science/article/pii/S0168827813000196. Renal and intrarenal blood-flow in cirrhosis of the liver.
23 Oettl, K., Birner-Gruenberger, R., Spindelboeck, W. et al. Lancet 298 (7723): 504–510. Available from: http://www.
(2013). Oxidative albumin damage in chronic liver sciencedirect.com/science/article/pii/
failure: relation to albumin binding capacity, liver S0140673671904351.
dysfunction and survival. J. Hepatol. 59 (5): 978–983. 34 Sacerdoti, D., Merlo, A., Merkel, C. et al. (1986).
Available from: http://www.sciencedirect.com/science/ Redistribution of renal blood flow in patients with liver
article/pii/S0168827813004297. cirrhosis. J. Hepatol. 2 (2): 253–261. Available from:
24 Navasa, M., Follo, A., Filella, X. et al. (2003). Tumor https://doi.org/10.1016/S0168-8278(86)80084-8.
necrosis factor and interleukin-6 in spontaneous bacterial 35 Maroto, A., Ginès, A., Saló, J. et al. (2018). Diagnosis of
peritonitis in cirrhosis: relationship with the development functional kidney failure of cirrhosis with Doppler
of renal impairment and mortality. Hepatol. Int. 27 (5): sonography: prognostic value of resistive index. Hepatol.
1227–1232. Available from: https://doi.org/10.1002/ Int. 20 (4): 839–844. Available from: https://doi.
hep.510270507. org/10.1002/hep.1840200411.
25 Albillos, A., Lario, M., and Álvarez-Mon, M. (2014). 36 Bernardi, M., Trevisani, F., Gasbarrini, A., and
Cirrhosis-associated immune dysfunction: distinctive Gasbarrini, G. (1994). Hepatorenal disorders: role of the
features and clinical relevance. J. Hepatol. 61 (6): renin-angiotensin-aldosterone system. Semin. Liver Dis.
1385–1396. Available from: http://www.sciencedirect. 14 (01): 23–34.
com/science/article/pii/S0168827814005492. 37 Arroyo, V., Ginés, P., Rimola, A., and Gaya, J. (1986). Renal
26 Girón-González, J.A., Martínez-Sierra, C., Rodriguez- function abnormalities, prostaglandins, and effects of
Ramos, C. et al. (2004). Implication of inflammation- nonsteroidal anti-inflammatory drugs in cirrhosis with
related cytokines in the natural history of liver cirrhosis. ascites: an overview with emphasis on pathogenesis. Am. J.
Liver Int. 24 (5): 437–445. Available from: https://doi. Med. 81 (Suppl. 2): 104–122. Available from: http://www.
org/10.1111/j.1478-3231.2004.0951.x. sciencedirect.com/science/article/pii/0002934386909125.
27 Solé, C., Solà, E., Morales-Ruiz, M. et al. (2016). 38 Henriksen, J.H., Møller, S., Ring-Larsen, H., and
Characterization of inflammatory response in acute-on- Christensen, N.J. (1998). The sympathetic nervous system
chronic liver failure and relationship with prognosis. Sci. in liver disease. J. Hepatol. 29 (2): 328–341. Available
Rep. 31 (6): 32341. Available from: http://dx.doi. from: http://www.sciencedirect.com/science/article/pii/
org/10.1038/srep32341. S0168827898800226.
28 Sole, C., Solà, E., Huelin, P. et al. (2018). Characterization 39 Moore, K., Wendon, J., Frazer, M. et al. (1992). Plasma
of inflammatory response in hepatorenal syndrome. Endothelin Immunoreactivity in liver disease and the
Relationship with kidney outcome and survival. Liver Int. Hepatorenal syndrome. N. Engl. J. Med. 327 (25):
https://doi.org/10.1111/liv.14037. [Epub ahead of print]. 1774–1778. Available from: https://doi.org/10.1056/
29 Gines, P., Cardenas, A., Solà, E., and Schrier, R.W. (2012). NEJM199212173272502.
Schrier’s diseases of the kidney. In: Schrier’s Diseases of 40 Moore, K.P., Taylor, G.W., Maltby, N.H. et al. (1990).
the Kidney (eds. R.W. Schrier, T.M. Coffman, R.J. Falk, Increased production of cysteinyl leukotrienes in
et al.), 1965–1996. Lippincott Williams & Wilkins. hepatorenal syndrome. J. Hepatol. 11 (2): 263–271.
Reference 119
Available from: http://www.sciencedirect.com/science/ http://www.sciencedirect.com/science/article/pii/

article/pii/0168827890901239. S0168827809008204.
41 Elia, C., Graupera, I., Barreto, R. et al. (2015). Severe 52 Bernardi, M., Gitto, S., and Biselli, M. (2011). The MELD
acute kidney injury associated with non-steroidal score in patients awaiting liver transplant: strengths and
anti-inflammatory drugs in cirrhosis: a case-control weaknesses. J. Hepatol. 54 (6): 1297–1306. Available from:
study. J. Hepatol. 63 (3): 593–600. Available from: http:// http://www.sciencedirect.com/science/article/pii/
www.sciencedirect.com/science/article/pii/ S0168827810010706.
S0168827815002500. 53 Khwaja, A. (2012). KDIGO clinical practice guidelines for
42 Ginès, A., Escorsell, A., Ginès, P. et al. (1993). Incidence, acute kidney injury. Nephron. Clin. Pract. 120 (4):
predictive factors, and prognosis of the hepatorenal c179–c184. Available from: https://www.karger.com/
syndrome in cirrhosis with ascites. Gastroenterol. Int. 105 DOI/10.1159/000339789.
(1): 229–236. Available from: http://www.sciencedirect. 54 Huelin, P., Piano, S., Solà, E. et al. (2017). Validation of a
com/science/article/pii/0016508593900317. staging system for acute kidney injury in patients with
43 de Franchis, R. (2015). Expanding consensus in portal cirrhosis and association with acute-on-chronic liver
hypertension: report of the Baveno VI consensus failure. Clin. Gastroenterol. Hepatol. 15 (3): 438–445.e5.
workshop: stratifying risk and individualizing care for Available from: http://dx.doi.org/10.1016/j.
portal hypertension. J. Hepatol. 63 (3): 743–752. Available cgh.2016.09.156.
from: http://www.sciencedirect.com/science/article/pii/ 55 Arroyo, V., Terra, C., and Ginès, P. (2007). Advances in
S0168827815003499. the pathogenesis and treatment of type-1 and type-2
44 Nadim, M.K., Durand, F., Kellum, J.A. et al. (2016). hepatorenal syndrome. J. Hepatol. 46 (5): 935–946.
Management of the critically ill patient with cirrhosis: a Available from: http://www.sciencedirect.com/science/
multidisciplinary perspective. J. Hepatol. 64 (3): 717–735. article/pii/S0168827807001067.
Available from: http://www.sciencedirect.com/science/ 56 Fagundes, C., Barreto, R., Guevara, M. et al. (2013). A
article/pii/S0168827815007187. modified acute kidney injury classification for diagnosis
45 Angeli, P., Bernardi, M., Villanueva, C. et al. (2018). EASL and risk stratification of impairment of kidney function
clinical practice guidelines for the management of in cirrhosis. J. Hepatol. 59 (3): 474–481. Available from:
patients with decompensated cirrhosis. J. Hepatol. http://www.sciencedirect.com/science/article/pii/
46 Alessandria, C., Ozdogan, O., Guevara, M. et al. (2005). S0168827813002936.
MELD score and clinical type predict prognosis in 57 Piano, S., Rosi, S., Maresio, G. et al. (2013). Evaluation of
hepatorenal syndrome: relevance to liver transplantation. the acute kidney injury network criteria in hospitalized
Hepatol. Int. 41 (6): 1282–1289. Available from: https:// patients with cirrhosis and ascites. J. Hepatol. 59 (3):
doi.org/10.1002/hep.20687. 482–489. Available from: http://www.sciencedirect.com/
47 Kamath, P.S., Wiesner, R.H., Malinchoc, M. et al. (2003). science/article/pii/S0168827813002766.
A model to predict survival in patients with end-stage 58 Wong, F., Nadim, M.K., Kellum, J.A. et al. (2011).
liver disease. Hepatol. Int. 33 (2): 464–470. Available from: Working party proposal for a revised classification system
https://doi.org/10.1053/jhep.2001.22172. of renal dysfunction in patients with cirrhosis. Gut 60 (5):
48 Ginès, P. and Cárdenas, A. (2008). The Management of 702–709.
ascites and hyponatremia in cirrhosis. Semin. Liver Dis. 59 Belcher, J.M., Sanyal, A.J., Peixoto, A.J. et al. (2013).
28 (01): 43–58. Kidney biomarkers and differential diagnosis of patients
49 Arroyo, V., Ginès, P., Gerbes, A.L. et al. (1996). Definition with cirrhosis and acute kidney injury. Hepatol. Int. 60
and diagnostic criteria of refractory ascites and hepatorenal (2): 622–632. Available from: https://doi.org/10.1002/
syndrome in cirrhosis. Hepatology 23 (1): 164–176. hep.26980.
50 Caregaro, L., Menon, F., Angeli, P. et al. (1994). 60 Francoz, C., Nadim, M.K., and Durand, F. (2016). Kidney
Limitations of serum creatinine level and creatinine biomarkers in cirrhosis. J. Hepatol. 65 (4): 809–824.
clearance as filtration markers in cirrhosis. Arch. Intern. Available from: http://dx.doi.org/10.1016/j.
Med. 154 (2): 201–205. Available from: http://www. jhep.2016.05.025.
jamanetwork.com/journals/jamainternalmedicine/ 61 Puthumana, J., Ariza, X., Belcher, J.M. et al. (2017). Urine
article-abstract/618345. interleukin 18 and Lipocalin 2 are biomarkers of acute
51 Francoz, C., Glotz, D., Moreau, R., and Durand, F. (2010). tubular necrosis in patients with cirrhosis: a systematic
The evaluation of renal function and disease in patients review and meta-analysis. Clin. Gastroenterol. Hepatol. 15
with cirrhosis. J. Hepatol. 52 (4): 605–613. Available from: (7): 1003–1013.e3. Available from: http://www.
sciencedirect.com/science/article/pii/ Am. J. Gastroenterol. 103: 1689. Available from: http://

S1542356516312319. dx.doi.org/10.1111/j.1572-0241.2008.01828.x.
62 Ariza, X., Solà, E., Elia, C. et al. (2015). Analysis of a 71 Singh, V., Ghosh, S., Singh, B. et al. (2012). Noradrenaline
urinary biomarker panel for clinical outcomes assessment vs. terlipressin in the treatment of hepatorenal syndrome:
in cirrhosis. PLoS One 10 (6): e0128145. Available from: a randomized study. J. Hepatol. 56 (6): 1293–1298.
https://doi.org/10.1371/journal.pone.0128145. Available from: http://www.sciencedirect.com/science/
63 Fagundes, C., Pépin, M.-N., Guevara, M. et al. (2012). article/pii/S0168827812001031.
Urinary neutrophil gelatinase-associated lipocalin as 72 Alessandria, C., Ottobrelli, A., Debernardi-Venon, W.
biomarker in the differential diagnosis of impairment of et al. (2007). Noradrenalin vs terlipressin in patients with
kidney function in cirrhosis. J. Hepatol. 57 (2): 267–273. hepatorenal syndrome: a prospective, randomized,
Available from: http://www.sciencedirect.com/science/ unblinded, pilot study. J. Hepatol. 47 (4): 499–505.
article/pii/S016882781200270X. Available from: http://www.sciencedirect.com/science/
64 Barreto, R., Elia, C., Solà, E. et al. (2014). Urinary article/pii/S0168827807002875.
neutrophil gelatinase-associated lipocalin predicts kidney 73 Martín–Llahí, M., Pépin, M., Guevara, M. et al. (2008).
outcome and death in patients with cirrhosis and Terlipressin and albumin vs albumin in patients with
bacterial infections. J. Hepatol. 61 (1): 35–42. Available cirrhosis and Hepatorenal syndrome: a randomized
from: http://www.sciencedirect.com/science/article/pii/ study. Gastroenterol. Int. 134 (5): 1352–1359. Available
S0168827814001317. from: http://www.sciencedirect.com/science/article/pii/
65 Prashant, S., Chawla, A., Ramesh, G. et al. (2003). S0016508508002527.
Beneficial effects of terlipressin in hepatorenal 74 Srivastava, S., Shalimar, V.S., Prakash, S. et al. (2015).
syndrome: a prospective, randomized placebo-controlled Randomized controlled trial comparing the efficacy of
clinical trial. J. Gastroenterol. Hepatol. 18 (2): 152–156. Terlipressin and albumin with a combination of
Available from: https://doi. concurrent dopamine, furosemide, and albumin in
org/10.1046/j.1440-1746.2003.02934.x. Hepatorenal syndrome. J. Clin. Exp. Hepatol. 5 (4):
66 Neri, S., Pulvirenti, D., Malaguarnera, M. et al. (2008). 276–285. Available from: http://www.sciencedirect.com/
Terlipressin and albumin in patients with cirrhosis and science/article/pii/S0973688315004181.
type I Hepatorenal syndrome. Dig. Dis. Sci. 53 (3): 75 Tavakkoli, H., Yazdanpanah, K., and Mansourian, M.
830–835. Available from: https://doi.org/10.1007/ (2012). Noradrenalin versus the combination of
s10620-007-9919-9. Midodrine and Octreotide in patients with Hepatorenal
67 Sanyal, A.J., Boyer, T., Garcia–Tsao, G. et al. (2008). A syndrome: randomized clinical trial. Int. J. Prev. Med. 3
randomized, prospective, double-blind, placebo- (11): 764–769. Available from: http://www.ncbi.nlm.nih.
controlled trial of Terlipressin for type 1 Hepatorenal gov/pmc/articles/PMC3506087.
syndrome. Gastroenterol. Int. 134 (5, 1368): 1360. 76 Cavallin, M., Kamath, P.S., Merli, M. et al. (2015).
Available from: http://www.sciencedirect.com/science/ Terlipressin plus albumin versus midodrine and
article/pii/S001650850800245X. octreotide plus albumin in the treatment of hepatorenal
68 Boyer, T.D., Sanyal, A.J., Wong, F. et al. (2016). syndrome: a randomized trial. Hepatol. Int. 62 (2):
Terlipressin plus albumin is more effective than albumin 567–574. Available from: https://doi.org/10.1002/
alone in improving renal function in patients with hep.27709.
cirrhosis and Hepatorenal syndrome type 1. 77 Zafar, S. Haque, I and Tayyab, G. U. et al. (2012). Role of
Gastroenterol. Int. 150 (7): 1579–1589.e2. Available from: Terlipressin and Albumin Combination versus Albumin
http://www.sciencedirect.com/science/article/pii/ Alone in Hepatorenal Syndrome. Am. J. Gastroenterol.
S0016508516002171. 107: S175–S176.
69 Indrabi, R.A., Javid, G., Zargar, S.A. et al. (2013). 78 Ortega, R., Ginès, P., Uriz, J. et al. (2003). Terlipressin
Noradrenaline is equally effective as terlipressin in therapy with and without albumin for patients with
reversal of type 1 hepatorenal syndrome: a randomized hepatorenal syndrome: results of a prospective,
prospective study. J. Clin. Exp. Hepatol. 3 (Suppl. 1): S97. nonrandomized study. Hepatol. Int. 36 (4): 941–948.
Available from: http://www.sciencedirect.com/science/ Available from: https://doi.org/10.1053/
article/pii/S0973688313002491. jhep.2002.35819.
70 Sharma, P., Kumar, A., Shrama, B.C., and Sarin, S.K. 79 Nanda, A., Reddy, R., Safraz, H. et al. (2018). J. Clin.
(2008). An open label, pilot, randomized controlled trial Gastroenterol. 52 (4). Available from: https://journals.
of noradrenaline versus Terlipressin in the treatment of lww.com/jcge/Fulltext/2018/04000/Pharmacological_
type 1 Hepatorenal syndrome and predictors of response. Therapies_for_Hepatorenal.15.aspx.
Reference 121
0 Facciorusso, A., Chandar, A.K., Murad, M.H. et al. (2017).

8 90 Velez, J.C.Q. and Nietert, P.J. (2011). Therapeutic
Comparative efficacy of pharmacological strategies for response to vasoconstrictors in Hepatorenal syndrome
management of type 1 hepatorenal syndrome: a parallels increase in mean arterial pressure: a pooled
systematic review and network meta-analysis. Lancet analysis of clinical trials. Am. J. Kidney Dis. 58 (6):
Gastroenterol. Hepatol. 2 (2): 94–102. Available from: 928–938. Available from: http://www.ncbi.nlm.nih.gov/
http://dx.doi.org/10.1016/S2468-1253(16)30157-1. pmc/articles/PMC3251915.
81 Israelsen, M., Krag, A., Allegretti, A.S. et al. Terlipressin 91 Velez, J.C.Q., Kadian, M., Taburyanskaya, M. et al.
versus other vasoactive drugs for hepatorenal syndrome. (2015). Hepatorenal acute kidney injury and the
Cochrane Database Syst. Rev. 2017 (9). Available from: importance of raising mean arterial pressure. Nephron
http://dx.doi.org/10.1002/14651858.CD011532.pub2. 131 (3): 191–201. Available from: http://www.ncbi.nlm.
82 Marta, C., Salvatore, P., Antonietta, R. et al. (2015). nih.gov/pmc/articles/PMC4655825.
Terlipressin given by continuous intravenous infusion 92 Nazar, A., Pereira, G.H., Guevara, M. et al. (2009).
versus intravenous boluses in the treatment of Predictors of response to therapy with terlipressin and
hepatorenal syndrome: a randomized controlled study. albumin in patients with cirrhosis and type 1
Hepatol. Int. 63 (3): 983–992. Available from: https://doi. hepatorenal syndrome. Hepatol. Int. 51 (1): 219–126.
org/10.1002/hep.28396. Available from: https://doi.org/10.1002/hep.23283.
83 Gluud, L.L., Christensen, K., Christensen, E., and Krag, 93 Cárdenas, A. and Ginès, P. (2011). Management of
A. (2010). Systematic review of randomized trials on patients with cirrhosis awaiting liver transplantation.
vasoconstrictor drugs for hepatorenal syndrome. Hepatol. Gut 60 (3): 412 LP–421. Available from: http://gut.bmj.
Int. 51 (2): 576–584. Available from: https://doi. com/content/60/3/412.abstract.
org/10.1002/hep.23286. 94 Wong, F., Leung, W., Al Beshir, M. et al. (2014). Outcomes
84 Jalan, R., Schnurr, K., Mookerjee, R.P. et al. (2009). of patients with cirrhosis and hepatorenal syndrome type
Alterations in the functional capacity of albumin in 1 treated with liver transplantation. Liver Transplant. 21
patients with decompensated cirrhosis is associated with (3): 300–307. Available from: https://doi.org/10.1002/
increased mortality. Hepatol. Int. 50 (2): 555–564. lt.24049.
Available from: https://doi.org/10.1002/hep.22913. 95 Marik, P.E., Wood, K., and Starzl, T.E. (2006). The course
85 Fernández, J., Monteagudo, J., Bargallo, X. et al. (2005). of type 1 hepato-renal syndrome post liver
A randomized unblinded pilot study comparing albumin transplantation. Nephrol. Dial. Transplant. 21 (2): 478–482.
versus hydroxyethyl starch in spontaneous bacterial Available from: http://dx.doi.org/10.1093/ndt/gfi212.
peritonitis. Hepatol. Int. 42 (3): 627–634. Available from: 96 Formica, R.N.J. (2016). Simultaneous liver kidney
https://doi.org/10.1002/hep.20829. transplantation. Curr. Opin. Nephrol. Hypertens. 25 (6).
86 Bortoluzzi, A., Ceolotto, G., Gola, E. et al. (2012). Positive Available from: https://journals.lww.com/co-
cardiac inotropic effect of albumin infusion in rodents nephrolhypertens/Fulltext/2016/11000/Simultaneous_
with cirrhosis and ascites: molecular mechanisms. liver_kidney_transplantation.18.aspx.
Hepatol. Int. 57 (1): 266–276. Available from: https://doi. 97 Guevara, M., Ginès, P., Bandi, J.C. et al. (2003).
org/10.1002/hep.26021. Transjugular intrahepatic portosystemic shunt in
87 Domenicali, M., Baldassarre, M., Giannone, F.A. et al. hepatorenal syndrome: effects on renal function and
(2014). Posttranserum creatinineiptional changes of vasoactive systems. Hepatol. Int. 28 (2): 416–422.
serum albumin: clinical and prognostic significance in Available from: https://doi.org/10.1002/hep.510280219.
hospitalized patients with cirrhosis. Hepatol. Int. 60 (6): 98 Testino, G., Ferro, C., Sumberaz, A. et al. (2003). Type-2
1851–1860. Available from: https://doi.org/10.1002/ hepatorenal syndrome and refractory ascites: role of
hep.27322. transjugular intrahepatic portosystemic stent-shunt in
88 Angeli, P., Volpin, R., Gerunda, G. et al. (2003). Reversal eighteen patients with advanced cirrhosis awaiting
of type 1 hepatorenal syndrome with the administration orthotopic liver transplantation. Hepato-
of midodrine and octreotide. Hepatol. Int. 29 (6): Gastroenterology 50 (54): 1753–1755.
1690–1697. Available from: https://doi.org/10.1002/ 99 Sourianarayanane, A., Raina, R., Garg, G. et al. (2014).
hep.510290629. Management and outcome in hepatorenal syndrome:
89 Florence, W., Lavinia, P., and Kenneth, S. (2004). need for renal replacement therapy in non-transplanted
Midodrine, octreotide, albumin, and TIPS in selected patients. Int. Urol. Nephrol. 46 (4): 793–800. Available
patients with cirrhosis and type 1 hepatorenal syndrome. from: https://doi.org/10.1007/s11255-013-0527-7.
Hepatol. Int. 40 (1): 55–64. Available from: https://doi. 100 Staufer, K., Roedl, K., Kivaranovic, D. et al. (2017).
org/10.1002/hep.20262. Renal replacement therapy in critically ill liver cirrhotic
patients—outcome and clinical implications. Liver Int. cirrhosis. Gastroenterol. Int. 133 (3): 818–824. Available
37 (6): 843–850. Available from: https://doi.org/10.1111/ from: http://www.sciencedirect.com/science/article/pii/
liv.13389. S001650850701298X.
101 Bañares, R., Nevens, F., Larsen, F.S. et al. (2012). 106 Park, S.H., Kim, D.J., Kim, Y.S. et al. (2014).
Extracorporeal albumin dialysis with the molecular Pentoxifylline vs. corticosteroid to treat severe alcoholic
adsorbent recirculating system in acute-on-chronic hepatitis: a randomised, non-inferiority, open trial. J.
liver failure: the RELIEF trial. Hepatol. Int. 57 (3): Hepatol. 61 (4): 792–798. Available from: http://dx.doi.
1153–1162. Available from: https://doi.org/10.1002/ org/10.1016/j.jhep.2014.05.014.
hep.26185. 107 Mathurin, P., Louvet, A., Duhamel, A. et al. (2013).
102 Kribben, A., Gerken, G., Haag, S. et al. (2012). Effects of Prednisolone with vs without pentoxifylline and
fractionated plasma separation and adsorption on survival of patients with severe alcoholic hepatitis: a
survival in patients with acute-on-chronic liver failure. randomized clinical trial. JAMA 310 (10): 1033–1041.
Gastroenterol. Int. 142 (4): 782–789.e3. Available from: Available from: http://dx.doi.org/10.1001/
http://www.sciencedirect.com/science/article/pii/ jama.2013.276300.
S0016508512000273. 108 Akriviadis, E., Botla, R., Briggs, W. et al. (2000).
103 Guevara, M., Terra, C., Nazar, A. et al. (2012). Albumin Pentoxifylline improves short-term survival in severe
for bacterial infections other than spontaneous bacterial acute alcoholic hepatitis: a double-blind, placebo-
peritonitis in cirrhosis. A randomized, controlled study. controlled trial. Gastroenterol. Int. 119 (6): 1637–1648.
J. Hepatol. 57 (4): 759–765. Available from: http://www. Available from: http://www.sciencedirect.com/science/
sciencedirect.com/science/article/pii/ article/pii/S0016508500511834.
S0168827812004394. 109 Parker, R., Armstrong, M., Corbett, C. et al. (2013).
104 Thévenot, T., Bureau, C., Oberti, F. et al. (2015). Effect Systematic review: pentoxifylline for the treatment of
of albumin in cirrhotic patients with infection other severe alcoholic hepatitis. Aliment. Pharmacol. Ther. 37
than spontaneous bacterial peritonitis. A randomized (9): 845–854. Available from: https://doi.org/10.1111/
trial. J. Hepatol. 62 (4): 822–830. Available from: http:// apt.12279.
www.sciencedirect.com/science/article/pii/ 110 EASL Clinical Practice Guidelines for the management
S0168827814008629. of patients with decompensated cirrhosis. J. Hepatol.
105 Fernández, J., Navasa, M., Planas, R. et al. (2007). 2018;doi: https://doi.org/10.1016/j.jhep.2018.03.024.
Primary prophylaxis of spontaneous bacterial peritonitis
delays Hepatorenal syndrome and improves survival in
123
10
Acute Tubular Necrosis

Amanda DeMauro Renaghan and Mitchell H. Rosner
Division of Nephrology, University of Virginia Health System, Charlottesville, VA, USA
I ntroduction centers in the United States. Ischemic ATN was the pre-
sumed etiology for 50% of all patients with renal failure,
Acute kidney injury (AKI) is defined as a rapid deteriora- while an additional ~12% were attributed to unresolved
tion in kidney function resulting in a reduction in glomeru- prerenal factors and ~25% to nephrotoxic ATN [5]. It is
lar filtration rate (GFR) and a failure of the kidneys to important to note that the diagnosis of ATN is often
excrete nitrogenous waste products and to maintain fluid, inferred based upon clinical characteristics, since kidney
electrolyte, and acid-base balance [1]. The two most com- biopsies are not routinely performed in patients with AKI.
mon etiologies of AKI are prerenal azotemia, caused by While several classes of agents have been investigated,
reversible renal hypoperfusion, and acute tubular necrosis thus far no specific pharmacologic strategies have proven
(ATN) [2–6]. ATN results from intrinsic parenchymal dam- effective for preventing or treating ATN. Therefore, it is
age mediated by endothelial and epithelial cell injury, intra- essential that clinicians focus on preventative measures
tubular obstruction, and immunologic and inflammatory such as maintaining adequate systemic perfusion and lim-
processes. ATN is a histopathologic and pathophysiologic iting nephrotoxin exposure, especially in those deemed to
entity that can stem from a variety of distinct renal insults, be at highest risk.
often seen in combination, and these can be divided broadly
into ischemic (prolonged volume depletion, hypotension),
septic, and nephrotoxic (medications, radiocontrast dye,
rhabdomyolysis) categories [2]. Ischemic ATN is seen most
commonly and occurs on a continuum beginning with pre-
renal azotemia [1, 5]. When prerenal insults are severe or
prolonged they may give rise to tubular damage [1].
Histology reveals dilation of proximal tubular lumens, flat-
tening of the tubular epithelium, simplification and loss of
brush border, and vacuolization (Figure 10.1) [7]. These his-
tological changes are reflected in the urinalysis, which typi-
cally reveals the presence of cellular debris and “muddy
brown casts” which consist of sloughed and necrotic epi-
thelial cells (Figure 10.2) [8].
ATN is common, however the incidence and outcomes
vary significantly depending on the definition used and
the patient population studied [2]. The landmark Program Figure 10.1 High-power image demonstrating flattened
epithelium with loss of the brush border and cells becoming
to Improve Care in Acute Renal Disease (PICARD) study disconnected from the basement membrane as seen in acute
was an observational study of a cohort of 618 patients with tubular necrosis. Source: Image courtesy of Helen P. Cathro,
AKI in the intensive care units (ICUs) of five academic University of Virginia Health System, Division of Renal Pathology.
124 Acute Tubular Necrosis
Failure (ORs: Risk 2.5, Injury 5.4, Failure 10.1) [13]. An

even higher incidence of AKI has been demonstrated in
the intensive care setting, with one retrospective cohort
study of 5383 patients admitted to the ICUs of a single ter-
tiary care academic center revealing an AKI incidence of
67%. Again, increased hospital mortality was seen with
progressive RIFLE stage [12].
The vast majority (two thirds or more) of cases of AKI in
hospitalized patients can be attributed to ATN or prerenal
injury [3–6]. This attribution is often based upon compati-
ble history and diagnostic findings and not based upon his-
tology, and to our knowledge there are no studies that have
systematically evaluated the causes of AKI in hospitalized
Figure 10.2 Muddy brown casts as seen in a patient with acute patients based on biopsy findings. In a study by the Madrid
kidney injury and presumed acute tubular necrosis due to sepsis.
Acute Renal Failure Study Group, ATN was identified in
45% of patients, while prerenal azotemia was determined
Definition to be the cause of AKI in 21% of patients [4]. In the PICARD
study, an observational cohort study of 618 patients with
ATN is a histopathologic and pathophysiologic entity that
AKI in the intensive care setting, ischemic ATN occurred
is the end result of numerous ischemic, nephrotoxic, and
in >50% of cases, while nephrotoxic ATN played a role in
inflammatory insults. In many cases, the decline in kid-
25% of cases [5].
ney function may be more prominent than the severity of
Certain patients and populations are at greater risk for the
histologic change, and necrotic cell death may not be
development of ATN based on age, comorbidities, severity
readily apparent [2, 9–11]. Clinical manifestations of
of illness, and exposures to procedures and medications. The
ATN vary by patient and the type of insult. While some
elderly, diabetic patients, those with preexisting chronic kid-
patients maintain normal urine flow, others may develop
ney disease (CKD) or hypertension, those with renal artery
a decline in urine output (oliguria) and subsequent signs
stenosis, and those taking nonsteroidal anti-inflammatory
and symptoms of volume overload. Similar to the meta-
drugs (NSAIDs), angiotensin-converting-enzyme (ACE)
bolic derangements seen in other forms of AKI, reduced
inhibitors, or angiotensin-receptor blockers (ARBs) may be
GFR and impaired tubular function in ATN may lead to
at increased risk for ischemic ATN due to impaired renal
hyperkalemia, metabolic acidosis, and azotemia. These
autoregulation [9]. The combination of an ACE inhibitor
may range from mild to severe and life-threatening. In
or an ARB along with an NSAID is a particularly high-risk
some cases, other kidney manifestations of a toxic insult
scenario especially in states of volume depletion, where
may be present such as glomerular disease (nonsteroidal
renal autoregulation and perfusion can be severely
anti-inflammatory agents) or tubular dysfunction such as
impaired [14]. Other populations at high risk for the devel-
the Fanconi syndrome (ifosfamide) or sodium and mag-
opment of ATN include (i) patients in ICUs, where 35–56%
nesium wasting (cisplatin).
of cases of AKI are attributed to sepsis; (ii) postoperative
patients, in particular those underdoing abdominal aortic
aneurysm surgery, surgery to correct obstructive jaundice,
E
pidemiology or combined coronary artery bypass graft (CABG) surgery
with valve surgery; and (iii) patients receiving iodinated
While it is clear that AKI is commonly encountered in radiocontrast dye (especially when delivered intra-
clinical practice, the reported incidence varies depending arterially and at high doses) [2, 15–22].
on the definition used and the patient population evalu-
ated [2, 12, 13]. In one study of 20 126 patients with a
mean age of 64 years admitted to an Australian tertiary
P
athophysiology
care hospital, almost 20% of patients met criteria for AKI
as defined by the RIFLE (Risk, Injury, Failure, Loss, and
Histology
End-stage) criteria. Importantly, multivariate logistic
regression analysis showed that all RIFLE criteria were The histologic lesions of ATN predominantly affect the
significantly predictive factors for hospital mortality, with proximal tubules (PTs) and vary with the severity of kidney
an almost linear increase in odds ratios (ORs) from Risk to injury and the stage of evolution of the lesion. On light
Pathophysiolog 125
microscopy, ATN typically manifests as dilation of proxi- response, driven by a combination of improper tubule
mal tubular lumens, flattening of the tubular epithelium, repair, vascular dropout, hypoxia, and persistent activation
simplification and loss of brush border, and vacuolization of the immune system may lead to fibrosis and CKD [24].
(see Figure 10.1) [7]. Downstream, distal tubular casts and
obstruction may be present and result from the binding of Tubular Factors
desquamated tubular cells and debris to Tamm-Horsfall The abrupt fall in GFR seen in ATN has traditionally been
protein [23]. Interstitial edema with mild to moderate leu- attributed to an interplay between tubular and hemody-
kocyte infiltration can be seen. Morphologic changes may namic abnormalities [26]. Ischemic insults to the kidney
be subtle, and frankly necrotic cells may not be present preferentially cause cell damage and death of tubular epi-
even in those with marked functional impairment [2, thelial cells, leading to decreased kidney function. These
9–11]. For this reason, many clinicians and pathologists cells are especially susceptible to ischemic injury because
prefer the term “acute tubular injury” (ATI) to “ATN” [11]. of their high demand for oxygen and ATP to drive transport
functions, along with their location in an environment of
very low oxygen tension [27]. Cells in the S3 segment of the
Ischemic ATN
PT and the thick ascending limb of Henle are most prone
Phases of Ischemic ATN to ischemic injury due to a combination of low medullary
Much remains to be discovered about the complex biology blood flow and countercurrent diffusion of oxygen causing
of ATN. Current research suggests that the course of lower partial pressure of oxygen (pO2) [28].
ischemic ATN progresses across five phases: prerenal, ini- Renal ischemia leads to a loss of cytoskeletal integrity
tiation, extension, maintenance, and repair. During the and cell polarity, with shedding of both necrotic and viable
prerenal phase, alterations in renal blood flow (RBF) lead PT brush border cells into the tubule lumens [2]. This is
to a decrease in intraglomerular pressure, however a vari- thought to be driven, in part, by redistribution of sodium/
ety of cellular and vascular adaptations are able to main- potassium ATPase and integrins from the basolateral to the
tain renal epithelial integrity [24]. These alterations may be apical membrane [29–31]. Additional biochemical path-
the result of volume depletion (vomiting, diarrhea, use of ways that have been implicated in severe cell injury and
diuretics), hypotension (shock), cirrhosis (splanchnic vaso- ATN include (i) depletion of ATP; (ii) increased tissue con-
dilation), heart failure (primarily through venous conges- centrations of reactive oxygen species; (iii) intracellular
tion and increased intra-abdominal pressure), selective acidosis; (iv) increased concentrations of cytosolic calcium;
renal ischemia (renal artery stenosis, cross-clamping of the (v) increased activity of phospholipases; (vi) release of pro-
renal artery during surgery for renal cell carcinoma), or teases from the tubular cell brush border; (vii) mitochon-
drugs that cause afferent arteriolar constriction (NSAIDs, drial injury; and (viii) cell apoptosis [2, 26, 32, 33]. As cells
calcineurin inhibitors) or efferent arteriolar dilation (ACE desquamate and enter the lumen, backleak of filtrate may
inhibitors, ARBs) [1, 25]. The initiation phase of ischemic occur into the vascular space across the damaged tubular
ATN occurs when hypoperfusion is sufficiently severe or epithelium, potentially contributing to further decline in
prolonged to deplete adenosine triphosphate (ATP), lead- GFR, especially when intratubular pressure is increased
ing to tubular cell injury and a continued decline in due to intratubular obstruction [26, 34].
GFR [24]. The subsequent extension phase is characterized
by (i) vascular endothelial damage, which contributes to Renal Blood Supply
severely reduced blood flow, persistent hypoxia, microvas- In ATN, RBF can decrease by as much as 30–50%, with a
cular obstruction, coagulopathy, and further injury to disproportionate decrease in medullary blood flow relative
tubular epithelial cells; and (ii) a pronounced inflamma- to total RBF contributing to pronounced medullary
tory response mediated by an influx of leukocytes and the ischemia [2, 35–38]. A variety of mechanisms have been
release of chemokines and cytokines that enhance the proposed to explain this and include (i) increased vasocon-
inflammatory cascade [24]. During the maintenance phase, striction in response to endothelin, adenosine, angiotensin
GFR nadirs as cells undergo migration, apoptosis, and pro- II, thromboxane A2, leukotrienes, and sympathetic nerve
liferation in an attempt to reestablish cellular and tubular activity; (ii) decreased vasodilation in response to nitric
integrity. Blood flow returns toward normal. Finally, dur- oxide, prostaglandin E2, acetylcholine, and bradykinin; (iii)
ing the recovery phase, GFR begins to improve as cellular structural damage to endothelial and vascular smooth
differentiation and repair occur. It is in this phase that epi- muscle cells; and (iv) increased leukocyte-endothelial
thelial polarity is reestablished and cellular and organ adhesion, vascular obstruction, leukocyte activation, and
function is restored [1, 24]. Incomplete recovery and/or inflammation from release of chemokines and
conversion from a physiological to a pathological repair cytokines [32]. In addition, the combination of continued
filtration and impaired tubular reabsorptive function leads Table 10.1 Mechanisms of nephrotoxin-induced acute tubular
to increased sodium chloride delivery to the macular densa, necrosis.
ultimately activating tubuloglomerular feedback (TGF)
and causing afferent arteriolar constriction with resultant Mechanism of injury Toxin
further decline in GFR [32, 39].
Exogenous toxins
Altered Afferent vasoconstriction:
Septic ATN intraglomerular NSAIDs
hemodynamics Calcineurin inhibitors
Sepsis is a frequent cause of AKI in hospitalized patients [15, (prolonged prerenal
injury predisposing to Iodinated radiocontrast agents
16, 20, 21]. While previously thought to induce ATN through
ATN)
decreased renal perfusion and systemic hypotension, more
Efferent vasodilation:
recent data suggest that septic ATN may be associated with ACE inhibitors
systemic vasodilation, high-output cardiac failure, and ARBs
increased RBF (“hyperdynamic sepsis”). In one study of
Direct tubular toxicity Antimicrobials:
ewes treated with Escherichia coli infusion to induce sepsis,
Aminoglycosides, outdated
RBF was found to be significantly increased. Importantly, tetracyclines, vancomycin
recovery from sepsis was associated with relative renal vaso- Amphotericin B
constriction and return in RBF back to control levels [40]. In Cidofovir, tenofovir
a follow-up study, sepsis induced by E. coli was again associ- Pentamidine
ated with increased RBF, along with oliguria and reduced
GFR. Infusion of angiotensin II restored arterial pressure, Antineoplastic agents:
decreased RBF, and led to a significant increase in urine out- Cisplatin
put and normalization of creatinine [41]. One possible Ifosfamide
explanation for these findings is that kidney injury in this
state may occur due to preferential dilation of the efferent Recreational drugs:
arteriole, leading to a decrease in intraglomerular pressure Synthetic cannabinoids
and drop in GFR [42].
Miscellaneous:
Acetaminophen
Nephrotoxic ATN Calcineurin inhibitors
Nephrotoxic ATN may be associated with exposure to a Iodinated radiocontrast agents
Mannitol
variety of exogenous and endogenous toxins, some of
Sucrose-containing IVIG formulations
which are shown in Table 10.1. Exogenous toxins include
Zoledronate
medications that impair renal autoregulation and predis-
pose patients to prolonged prerenal ischemia and subse- Intratubular Acyclovir, indinavir, methotrexate,
obstruction sulfonamides
quent ATN (NSAIDs, ACE inhibitors, ARBs), as well as
drugs that are directly toxic to renal tubules, including Endogenous toxins
many antimicrobial, iodinated radiocontrast, and antineo- Intratubular pigments Rhabdomyolysis (may be precipitated
plastic agents. Other exogenous toxins, such as acyclovir by medications [statins, haloperidol]
and methotrexate, may cause ATN through intratubular and drugs of abuse [cocaine, alcohol])
crystal precipitation and obstruction. ATN secondary to
Hemolysis (may be precipitated by
the production of endogenous toxins may occur in patients
quinine, quinidine, sulfonamides)
with rhabdomyolysis or, less commonly, hemolysis. In
Intratubular Multiple myeloma (cast nephropathy)
these cases, AKI is caused by the release of heme pigment
obstruction by proteins
from myoglobin and hemoglobin, respectively, ultimately
Intratubular Uric acid (tumor lysis syndrome)
leading to (i) tubular obstruction; (ii) direct proximal precipitation
tubular cell injury; and (iii) vasoconstriction with reduced
medullary blood flow [43, 44]. Additionally, endogenous ACE, angiotensin-converting-enzyme; ARB, angiotensin-receptor
production of abnormal paraproteins (as in multiple mye- blocker; ATN, acute tubular necrosis; IVIG, intravenous
immunoglobulin; NSAID, nonsteroidal anti-inflammatory drug.
loma) and uric acid (as in tumor lysis syndrome) may Source: Adapted by permission from Rosner MH, Okusa MD (2008).
injure renal tubules through precipitation and intratubu- Drug-associated acute kidney injury in the intensive care unit.
lar obstruction [45]. Clinical Nephrotoxins, 29–41. © 2008, Springer Nature.
Diagnosi 127
D
iagnosis Urine Microscopy
The histologic changes of ATN are reflected in the urinaly-
History and Physical Examination sis, which classically reveals the presence of cellular debris
Determining the etiology of AKI begins with a careful eval- and “muddy brown casts” consisting of sloughed and
uation of the patient’s history and physical examination. necrotic epithelial cells. Perazella and colleagues demon-
For example, a history of vomiting, diarrhea, or hypoten- strated the high value of urine microscopy for the diagnosis
sion, combined with physical examination findings of vol- of ATN using a urinary sediment score. In patients with a
ume depletion, may point toward prerenal AKI or ischemic high pretest probability of ATN, the presence of any casts
ATN. Recent exposure to nephrotoxic medications or iodi- or renal tubular epithelial (RTE) cells (score 2) resulted in
nated radiocontrast dye may suggest nephrotoxic ATN. A very high positive predictive value (100%) and low negative
careful and thorough history of drug and other potential predictive value (44%) for a final diagnosis of ATN [8]. In a
nephrotoxin exposure is critical. subsequent study, the same group determined that a uri-
nary sediment score may be useful to predict worsening of
kidney injury due to ATN during hospitalization [53]. A
Urine Sodium Concentration and Fractional
subsequent systematic review assessing the utility of urine
Excretion of Sodium
microscopy for differentiating prerenal AKI and ATN in
Fractional excretion of sodium (FENa) is a measure of the hospitalized patients determined that the presence and
extraction of sodium and water from the glomerular fil- number of RTE cells and RTE cell casts and/or granular
trate; it is the ratio of the sodium filtration rate to the over- casts in the urine sediment may be useful in predicting
all GFR as estimated by the renal filtration of creatinine: more severe kidney damage as reflected by nonrecovery of
AKI or need for dialysis [54].
urine sodium plasma creatinine 100
plasma sodium urine creatinine 46

The Role of Novel Biomarkers
FENa has traditionally been used to distinguish between
prerenal azotemia and intrinsic forms of AKI such as ATN In more recent years, a number of investigational biomark-
in patients with oliguria. In cases of prerenal AKI, and in ers have been studied with the goals of (i) detecting AKI at
early ischemic ATN where tubular function remains intact, an earlier stage (“subclinical AKI” occurring prior to a rise
urinary sodium concentration (UNa) should be low in serum creatinine [SCr]); (ii) determining the location
(<20 mmol/l) and FENa should be <1%, while in estab- and severity of kidney injury; and (iii) predicting outcomes
lished ATN with impaired tubular function, UNa will be such as the development of CKD following AKI [55].
higher (usually >40 mmol/l) and FENa will be >1% [2, 46]. Urinary kidney injury molecule-1 (KIM-1) is a biomarker
While these urine indices may help differentiate prerenal that has shown promise for both the early diagnosis of AKI
azotemia from ATN in some cases, there are limitations to and the differentiation of ATN from other etiologies of kid-
the use of UNa and FENa. For example, in the presence of ney disease. In one case-control study of 20 children who
CKD or diuretic use, the diagnostic accuracy of urine underwent cardiopulmonary bypass surgery with or with-
chemistries to distinguish these conditions decreases. A out the subsequent development of AKI (defined as a >50%
low fractional excretion of urea (FEUrea) (<35%) is more increase in SCr from baseline within 48 hours), urinary
sensitive than FENa for distinguishing prerenal from KIM-1 increased postoperatively only in those children
intrinsic renal failure when diuretics have been recently who developed AKI [56]. In another study, normalized uri-
administered [47]. For instance, in a prospective study of nary KIM-1 levels were significantly higher in patients
AKI patients, a FEUrea 35% was less sensitive than a with biopsy-proven ATN compared to those with other
FENa 1% in differentiating between prerenal disease and forms of kidney disease.
ATN in patients not treated with diuretics (48% vs. 78%) The Translational Research Investigating Biomarkers
and was more sensitive (79% vs. 58%) but less specific (33% and End Points for Acute Kidney Injury (TRIBE-AKI) con-
vs. 81%) in patients who were treated with diuretics [47]. sortium conducted a prospective, multicenter cohort study
Additionally, ATN in the setting of rhabdomyolysis, hemol- in adults undergoing cardiac surgery to assess whether
ysis, sepsis, or recent contrast administration has been postoperative measures of urine interleukin-18 (IL-18),
shown to be associated with low UNa (<10 mmol/l) and urine neutrophil gelatinase-associated lipocalin (NGAL),
FENa <1% [48–52]. Given these limitations, measurements or plasma NGAL could identify patients destined to develop
of UNa and FENa must be interpreted in the context of AKI and other adverse outcomes. Results revealed that
patient’s overall clinical course and exposure history. urine IL-18 and urine and plasma NGAL levels peaked
within 6 hours after surgery. After multivariable adjust- P

rognosis
ment, the highest quintiles of urine IL-18 and plasma
NGAL were associated with 6.8-fold and 5-fold higher odds Renal Recovery After ATN
of AKI, respectively, compared with the lowest quintiles.
Elevated urine IL-18 and urine and plasma NGAL levels The duration and outcome of ATN vary depending on the
associated with longer length of hospital stay, longer ICU length and severity of the initial insult and the presence or
stay, and higher risk for dialysis or death [57]. absence of recurrent ischemia and/or repeated exposure to
More recently, the multicenter Sapphire study discovered nephrotoxins [60, 61]. While the majority of patients return
and validated two novel cell cycle arrest urine biomarkers for to or near their previous level of kidney function, recovery
risk stratification of AKI, tissue inhibitor of metalloprotein- does not occur in all cases [62]. Progression from AKI to
ases-2 (TIMP-2) and insulin-like growth factor-binding pro- CKD and end-stage renal disease (ESRD) is an area of
tein 7 (IGFBP7) [58]. The primary endpoint was Kidney increasing interest and active research, largely focusing on
Disease Improving Global Outcomes (KDIGO) stages 2 and 3 understanding the mechanisms of injury and maladaptive
AKI within 12 hours of sample collection. The combination repair after ATN with the goal of slowing or halting
of TIMP-2 and IGFBP7 produced an area under the receiver progression [63].
operating characteristics curve (AUC) of 0.80, which was sig- Schiffl and colleagues prospectively evaluated critically
nificantly superior to all previously described biomarkers for ill patients with normal baseline kidney function who
AKI. Additionally, risk for major adverse kidney events developed clinically diagnosed severe ATN necessitating
(death, need for renal replacement therapy [RRT], or persis- RRT. At discharge, 57% of the 226 surviving patients had
tent renal dysfunction) within 30 days elevated sharply for normal kidney function, while 33% had mild to moderate
[TIMP-2]·[IGFBP7] above 0.3 (ng/ml)2/1000 and doubled renal failure (SCr 1.3–3 mg/dl), and 10% had severe renal
when values were >2.0 (ng/ml)2/1000. Based in part on these failure (SCr 3–6 mg/dl). Multivariate analysis showed that
observations, a point of care device measuring these two bio- neither patient characteristics (age, gender, comorbid
markers was approved by the Food and Drug Administration conditions), severity of illness (APACHE III, number of
(FDA) in 2014 to aid in the rapid and early identification of failed organs), nor mode and duration of RRT were
patients at high risk for developing AKI. related to recovery of renal function. The authors con-
The above data clearly suggest potential value for next- cluded that, when critically ill patients with normal base-
generation biomarkers for both early diagnosis of and line kidney function survive the precipitating cause of
prognostication in AKI, however many questions remain ATN, the vast majority will recover sufficient renal
about the best ways to apply these markers in clinical prac- function [62].
tice, and it remains to be seen whether early knowledge of Not surprisingly, pre-existing CKD represents a major
risk of AKI will translate into improved clinical out- risk factor for the development of more severe AKI, greater
comes [55]. In addition, the sensitivity and specificity of need for RRT, and lower probability of renal recovery [64].
these biomarkers for the specific diagnosis of ATN versus Additional risk factors for irreversible decline in kidney
other forms of AKI is not yet determined. In this regard, a function include (i) older age; (ii) hypoalbuminemia; (iii)
single-center trial examined the effect of a “KDIGO bun- comorbidities including diabetes and heart failure; and (iv)
dle” consisting of optimization of volume status and hemo- more severe AKI requiring RRT [65, 66].
dynamics, avoidance of nephrotoxic drugs, and preventing
hyperglycemia in high-risk patients defined by urinary
Mortality After ATN
[TIMP-2]·[IGFBP7] >0.3 (ng/ml)2/1000 undergoing car-
diac surgery. The primary endpoint was the rate of AKI Several studies have demonstrated an association between
defined by KDIGO criteria within the first 72 hours after AKI during hospitalization and in-hospital and long-term
surgery. AKI was significantly reduced with the interven- mortality.
tion compared to controls (55.1 vs. 71.7%; ARR, 16.6%). The In the Acute Kidney Injury-Epidemiologic Prospective
implementation of the bundle resulted in significantly Investigation (AKI-EPI) study, a total of 1032 ICU patients
improved hemodynamic parameters at different time out of 1802 total patients (57%) developed AKI by KDIGO
points, less hyperglycemia, and use of ACE inhibitors/ criteria. Increasing AKI severity was associated with hospi-
ARBs compared to controls. Rates of moderate to severe tal mortality when adjusted for other variables (ORs: stage
AKI were also significantly reduced by the intervention 1 = 1.679, stage 2 = 2.945, stage 3 = 6.884) [67].
compared to controls. This promising study suggests that Importantly, AKI has been shown to represent a risk fac-
biomarker-guided interventions may be of benefit in tor for long-term mortality even in those patients who
decreasing the rate and progression of AKI [59]. experience normalization of SCr after AKI. Bucaloiu and
Preventio 129
colleagues used a propensity score-matched cohort method inform our evidence-based recommendations for the pre-
to retrospectively evaluate the risk of death after reversible, vention of ATN. An exhaustive list of the various com-
hospital-associated AKI in patients with normal pre- pounds and strategies that have been reported is beyond
hospital kidney function. Of patients alive at 90 days, 1610 the scope of this chapter. Thus, representative examples
patients with reversible AKI were matched with 3652 con- are discussed and other compounds and strategies are
trols. At a median follow-up of just over 3 years, the risk of listed in Tables 10.2 and 10.3.
death associated with reversible AKI was significant (haz-
ard ratio, 1.50) in Cox proportional hazard models.
Avoiding Nephrotoxic Medications and
Adjustment for the development of CKD during follow-up
Mitigating Risk of ATN
attenuated but did eliminate this increased risk (hazard
ratio 1.18). Based on this data, investigators concluded that Aminoglycosides
a resolved episode of hospital-associated AKI has impor- Aminoglycosides remain important therapies for the man-
tant implications for the longitudinal surveillance of agement of life-threatening gram-negative infections [48].
patients [68]. Nephrotoxicity represents the major dose-limiting toxicity
of these drugs, occurring in approximately 10–25% of
patients [119–123]. Toxicity is mediated primarily by dam-
Loss of Renal Reserve as a Predictor and
age to PT cells, impairing tubular reabsorption and com-
Consequence of ATN
promising cell viability [124].
Renal functional reserve (RFR) can be used to measure the More than 60 trials have been conducted to evaluate the
capacity of the kidney to increase GFR under conditions of benefits of altering the dosing schedule of aminoglyco-
physiologic stress and may serve as a functional marker sides, with the goal of limiting uptake by PT cells. Blaser
that assesses susceptibility to injury. One method of meas- and König evaluated the results of 24 randomized clinical
uring RFR is to measure changes in GFR in response to a trials including 3181 patients comparing once-daily versus
high-protein meal. In a recent study, preoperative RFR was multiple-daily dosing regimens for amikacin, netilmicin,
lower in patients who experienced AKI and predicted AKI and gentamicin. Analysis revealed superior results for
with an AUC of 0.83. Patients with preoperative RFRs not once-daily regimens with respect to clinical and bacterio-
greater than 15 ml/min/1.73m2, despite normal resting logical efficacy. While nephrotoxicity occurred less fre-
GFRs, were 11.8 times more likely to experience AKI [69]. quently with once-daily dosing (4.5% vs. 5.5%), results did
Thus, measurement of RFR preoperatively may identify not reach statistical significance [72]. In another meta-
patients who are likely to benefit from preventive measures analysis of similar size, Barza et al. demonstrated a reduc-
or to select for use of biomarkers for early detection. In a tion in nephrotoxicity with once-daily dosing, but no
follow-up to this study, AKI patients displayed a significant difference in patient mortality [73]. Hatala et al. evaluated
decrease in RFR (from 14.4 to 9.1 ml/min/1.73 m2) and 13 studies and revealed a trend toward reduced nephrotox-
patients without AKI but with positive postoperative cell- icity with once-daily dosing that did not reach statistical
cycle arrest biomarkers also experienced a similar decrease significance, while Ferriols-Lisart and Alós-Almiñana ana-
of RFR (from 26.7 to 19.7 ml/min/1.73m2). In contrast, lyzed 18 studies including 2317 patients and demonstrated
patients with neither clinical AKI nor positive biomarkers a reduction in summary odds of nephrotoxicity with once-
had no such decrease of RFR. Thus, AKI, and even AKI daily aminoglycoside dosing [74, 75]. Two additional meta-
defined by biomarkers (but not elevations in SCr), is associ- analyses by Munckhof et al. and Ali and Goetz failed to
ated with persistent falls in RFR and highlights the long- show a significant difference in the rates of kidney
term deleterious effects of this occurrence [70]. Further injury [71, 76].
studies are needed in this area. The cumulative conclusions of these analyses indicate
that once-daily dosing strategies generally lead to less AKI
when compared with multiple-dose strategies, though the
Prevention benefits of this approach are modest [125].
Many studies have been performed to determine if any Amphotericin B

approaches can minimize the risk of ATN when patients Amphotericin B remains an important antifungal therapy.
are faced with a high-risk clinical scenario, such as sepsis, Unfortunately, nephrotoxicity is common and may occur in
major surgery, or the need for certain antibiotics or iodi- close to 25% of patients receiving conventional (nonlipid-
nated radiocontrast dye. This section seeks to review the based) formulations [126, 127]. Renal toxicity may take the
studies currently available that guide best practice and form of renal tubular acidosis, impaired renal concentrating
Table 10.2 Prevention of acute tubular necrosis in select high-risk settings.
Certainty Strength of
Risk setting Select key studies Results of studies Recommendation of evidence recommendation
Aminoglycosides Ali and Goetz [71] (M) No significant difference in the incidence of nephrotoxicity We suggest that, in patients with normal Moderate Strong
with SDD vs. MDD kidney function in steady state,
Blaser and König [72] Trend toward decreased nephrotoxicity with SDD vs. MDD aminoglycosides be administered in a single
(M) (4.5% vs. 5.5%); did not reach statistical significance daily dose rather than multiple daily doses.
We also recommend that, for patients with
Barza et al. [73] (M) Reduced risk of nephrotoxicity with SDD vs. MDD CKD stage 3b and later, aminoglycosides be
Ferriols-Lisart and Reduced summary odds of nephrotoxicity with SDD vs. avoided unless there are no other
Alós-Almiñana [74] (M) MDD alternatives.
Hatala et al. [75] (M) Trend toward decreased nephrotoxicity with SDD vs.
MDD; did not reach statistical significance
Munckhof et al. [76] No significant difference in the incidence of nephrotoxicity
(M) with SDD vs. MDD
Amphotericin B Anderson [77] (M) Supplementing dietary sodium chloride intake with We recommend the administration of a Moderate Strong
150 mEq intravenously or orally before or at time of drug saline load (1 l of 0.9% sodium chloride IV)
administration “will likely prevent much of the observed prior to amphotericin B administration.
nephrotoxicity” associated with amphotericin B
Llanos et al. [78] (RCT) Creatinine clearance decreased in patients pretreated with
dextrose in water, but remained unchanged in patients
pretreated with 0.9% saline
Botero Aguirre and Significantly lower risk of nephrotoxicity with liposomal We suggest using lipid formulations of Moderate Strong
Restrepo Hamid [79] amphotericin B vs. conventional amphotericin B amphotericin B rather than conventional
(Cochrane systematic deoxycholate formulations of amphotericin B except in
review) circumstances were the efficacy of
conventional formulations may be greater.
Vancomycin + Chen et al. [80] (M) Increased risk of AKI with concurrent use of We recommend using piperacillin- Moderate Strong
piperacillin- vancomycin + piperacillin-tazobactam vs. tazobactam cautiously in those patients
tazobactam vancomycin + other β-lactam receiving vancomycin. Use should be
Giuliano et al. [81] (M) Increased risk of AKI with vancomycin + piperacillin- limited to those who have appropriate
tazobactam vs. vancomycin ± other β-lactam indications. Alternative antibiotics should
be considered whenever possible.
Hammond et al. [82] Increased risk of AKI with concomitant
(M) vancomycin + piperacillin-tazobactam vs. vancomycin
without piperacillin-tazobactam
Luther et al. [83] (M) Increased odds of AKI with vancomycin + piperacillin-
tazobactam vs. vancomycin monotherapy,
vancomycin +cefepime or carbapenem, and piperacillin-
monotherapy; NNH = 11
Mellen et al. [84] (M) Increase in the incidence of nephrotoxicity with
vancomycin + piperacillin-tazobactam vs. vancomycin
alone or in combination with cefepime or meropenem
0005152398.INDD 130 09-12-2022 15:14:54

Iodinated Brar et al. [85] (RCT) Decrease in CIN with LVEDP-guided 0.9% saline volume We recommend IV volume expansion for Moderate Strong
radiocontrast expansion vs. standard hydration group (6.7% vs. 16.3%) in patients at risk for CIN without
dyea patients undergoing cardiac catheterization contraindications to volume expansion.
Jurado-Román [86] Decreased incidence of CIN with 0.9% saline vs. no hydration
et al. (RCT) in the setting of emergent PCI for STEMI (11% vs. 21%)
Luo et al. [87] (RCT) Lower rate of CIN with 0.9% saline vs. no hydration after
PCI for STEMI (20.4% vs. 35.2%)
Nijssen EC et al. [88] No difference in rate of CIN between 0.9% saline and no
(RCT) prophylaxis (2.6% vs. 2.7%); not an especially high-risk
group (low rate of CIN, 65% had only mild CKD with eGFR
45–59 ml/min/1.73m2, only 48% received intra-arterial
contrast)
Trivedi et al. [89] (RCT) Decreased risk of CIN 0.9% saline vs. unrestricted oral
intake (1/27 vs. 9/26 patients)
Weisbord et al. [90] No significant difference in primary endpoint (composite We do not recommend the routine use of Strong Strong
(RCT) of death, need for RRT, or persistent decline in kidney sodium bicarbonate solution over NS for
function at 90 days) or CIN with IV 1.26% BIC vs. 0.9% the prevention of CIN.
saline
Zoungas et al. [91] (M) Lower pooled risk of CIN with BIC (RR, 0.62); significant
heterogeneity; small, poor-quality studies more likely to
suggest benefit; no clear effects on risk of RRT, heart
failure, or mortality
Weisbord et al. [90] No significant difference in primary endpoint (composite We do not recommend the routine use of Strong Strong
(RCT) of death, need for RRT, or persistent decline in kidney N-acetylcysteine for the prevention of CIN.
function at 90 days) or CIN with PO NAC vs. placebo
Rhabdomyolysis Gunal et al. [92] (case 16 crush injury victims; time between rescue and initiation We recommend early and aggressive Moderate Strong
series) of IVF significantly longer in victims requiring RRT hydration in all patients with
Ron et al. [93] (case 7 patients with crush injury treated immediately with rhabdomyolysis with CK values >5000 U/l
series) alkaline solute diuresis on extraction; no patient developed or increasing CK values in the absence of
ARF (attributed to prompt administration of IVF) volume overload.
Brown et al. [94] No difference in rates of ARF, dialysis, or mortality with We do not recommend the routine use of Low Weak
(retrospective chart BIC/MAN vs. without BIC/MAN among patients with CK mannitol to prevent or treat rhabdomyolysis
review) >5000 U/l; trend toward improved outcome in subgroup given the risks of volume depletion and
with CK >30 000 U/l hypernatremia, and conflicting evidence
Kim et al. [95] (RCT) Significant difference in incidence of rhabdomyolysis (CK regarding potential benefit.
>5000 U/l) with early urine alkalization group vs. Use of bicarbonate-based fluids to prevent
hydration alone in patients admitted with doxylamine or treat rhabdomyolysis may be considered
overdose in patients in whom a diuresis has been
Scharman and 27 studies; 3 studies concluded no significant difference in established with 0.9% saline. Bicarbonate
Troutman [96] rates of ARF with BIC vs. no BIC; no evidence supported a infusion should be discontinued and
(systematic literature preferred fluid type; interpretation limited by lack of isotonic saline resumed for arterial pH 7.5,
review) controlled trials and heterogeneity serum bicarbonate >30 mEq/l, or if
symptomatic hypocalcemia develops.
(Continued)
0005152398.INDD 131 09-12-2022 15:14:54

Certainty Strength of
Risk setting Select key studies Results of studies Recommendation of evidence recommendation
Surgery (all Zacharias et al. [97] Investigated preventative strategies including dopamine We do not recommend the routine use of Very low Weak
types) (Cochrane systematic and its analogues, diuretics, CCBs, ACE inhibitors, NAC, these interventions for the prevention of
review) ANP, and erythropoietin; no group showed a significant AKI/ATN in the perioperative setting.
change in mortality or kidney injury in treatment groups
vs. controls
Cardiac surgery Lamy et al. [98] (RCT) Lower rate of AKI at 30 d with off-pump vs. on-pump We suggest that off-pump coronary artery Low Weak
CABG; no difference in renal failure requiring RRT or bypass graft surgery not be selected solely
death; increased risk of early revascularization for the purpose of reducing perioperative
AKI.
Nigwekar et al. [99] (M) Reduction in overall AKI (OR, 0.57) and AKI requiring
RRT (OR, 0.55) with off-pump vs. on-pump CABG; limited
by lack of uniform definition for AKI and heterogeneity
Seabra et al. [100] (M) Significant 40% lower odds of postoperative AKI and a
nonsignificant 33% lower odds of need for RRT compared
with off-pump vs. on-pump CABG
a
For a more detailed discussion of the prevention and treatment of CIN, please see Chapter 11 on contrast-induced nephropathy.
ACE, angiotensin-converting-enzyme; AKI, acute kidney injury; ANP, atrial natriuretic peptide; ARF, acute renal failure; ATN, acute tubular necrosis; BIC, sodium bicarbonate; BIC/MAN,
sodium bicarbonate + mannitol; CABG, coronary artery bypass graft; CCB, calcium channel blocker; CIN, contrast-induced nephropathy; CK, creatine kinase; CKD, chronic kidney disease;
eGFR, estimated glomerular filtration rate, IV, intravenous, IVF, intravenous fluid; LVEDP, left ventricular end-diastolic pressure; M, meta-analysis; MDD, multiple daily dosing; NAC,
N-acetylcysteine; NNH, number needed to harm; NS, normal saline; OR, odds ratio; PCI, percutaneous coronary intervention; RCT, randomized controlled trial; RR, relative risk; RRT, renal
replacement therapy; SCr, serum creatinine; SDD, single daily dosing; STEMI, ST elevation myocardial infarction.
0005152398.INDD 132 09-12-2022 15:14:54

Table 10.3 Select agents investigated for the prevention of acute tubular necrosis.
Agent or procedure Certainty of Strength of

for prevention of ATN Select key studies Results of studies Recommendation evidence recommendation
Dopamine Bellomo et al. [101] No significant difference in renal function, need for RRT, ICU In the absence of convincing High Strong
(RCT) or hospital length of stay, or mortality in critically ill patients evidence of benefit, and in
with early renal dysfunction receiving low-dose dopamine light of potential risks, we do
(2 mcg/kg/min) vs. placebo not recommend the routine
Friedrich et al. [ 102] Increased urine output and improvement in SCr on day 1 with use of dopamine for
(M) low-dose dopamine ( 5 mcg/kg/min); no sustained prevention of AKI.
improvement in renal function, decrease in need for RRT, or
decrease in mortality (did not separate prophylactic from
therapeutic trials)
Kellum and No significant benefit of low-dose dopamine (<5 mcg/kg/min)
Decker [103] (M) for prevention of AKI
Fenoldopam Bove et al. [104] Similar rates of AKI with the use of fenoldopam vs. dopamine In the absence of high-quality Low Weak
(RCT) in high-risk patients following induction of anesthesia for adequately powered trials and
cardiac surgery consistent data demonstrating
Gillies et al. [105] (M) Reduction in risk of AKI in patients receiving fenoldopam vs. improvement in patient-
placebo; no difference in mortality or need for RRT; high risk of centered outcomes, and in
bias in 5/6 studies light of the risk of
hypotension, we suggest not
Landoni et al. [106] Lower risk of AKI in fenoldopam-treated patients vs. placebo or using fenoldopam to prevent
(M) other therapy (primarily low-dose dopamine); no significant ischemic ATN.
increase in complications; heterogeneous populations; 6 studies
lacked placebo controls
Morelli et al. [107] Lower rate of AKI (increase in SCr above 1.7 mg/dl) in patients
(RCT) treated with prophylactic fenoldopam vs. placebo; no significant
difference in rate of severe acute renal failure (SCr >3.4 mg/dl)
Loop diuretics Bove et al. [108] (M) Significant improvement in survival in patients treated with We do not recommend the Moderate Strong
preventative furosemide bolus(es) vs. any comparator (standard routine use of loop diuretics
care, 0.9% saline, 0.9% saline ± mannitol, or furosemide for the prevention of ATN.
infusion)
Hager et al. [109] No significant difference in post-operative decline in creatinine
(RCT) clearance between low-dose furosemide (1 mg/h) and placebo
groups; increased incidence of hypokalemia with use of
furosemide
Ho et al. [110] (M) No significant difference in in-hospital mortality or need for
Ho et al. [111] (M) RRT with the use of furosemide to prevent acute renal failure
Lassnigg et al. [112] Increased rate of AKI in patients randomized to receive

(RCT) furosemide (0.5 mcg/kg/min) vs. 0.9% saline or low-dose
dopamine
(Continued)
0005152398.INDD 133 09-12-2022 15:14:54


for prevention of ATN Select key studies Results of studies Recommendation evidence recommendation
Natriuretic peptides Nigwekar et al. [113] Decreased need for RRT but no difference in mortality in In the absence of high-quality Low Weak
(Cochrane systematic patients receiving prophylactic low-dose ANP compared to adequately powered trials, and
review) high-dose ANP or controls; authors acknowledged paucity of in light of the risk of
high-quality, adequately powered data hypotension, we suggest not
Nigwekar et al. [114] Natriuretic peptide administration (ANP, BMP, urodilantin) was using natriuretic peptides to
(M) associated with a reduction in need for RRT and a prevent ischemic ATN.
nonsignificant trend toward reduction in 30 days or in-hospital
mortality; most studies were small and lacked adequate power
to reach statistical significance on their own
Remote ischemic Menting et al. [115] RIPC by cuff inflation made little or no difference in SCr In the absence of clear benefit, High Strong
preconditioning (Cochrane systematic (postoperative days 1–3), need for dialysis, length of hospital we do not recommend the
review) stay, or all-cause mortality compared with no RIPC prior to routine use of RIPC for the
surgery (renal transplant excluded) prevention of ischemic ATN.
Statinsa Billings et al. [116] No reduction in the risk of AKI (0.3 mg/dl increase in SCr In the absence of clear benefit, High Strong
(RCT) within 48 hours) in patients undergoing cardiac surgery with we do not recommend the
use of high-dose atorvastatin vs. placebo; increased AKI among routine use of statins for the
statin-naïve CKD patients prevention of ischemic ATN.
Park et al. [117] Acute perioperative atorvastatin use not associated with lower
(RCT) incidence of AKI (AKIN criteria) or improved clinical outcomes
after valvular heart surgery compared with placebo
Prowle et al. Short-term perioperative atorvastatin use not associated with
(RCT) [118] reduction in AKI (RIFLE R or greater) or smaller increases in
urinary NGAL
a
For a discussion of statin administration as it pertains to the use of iodinated radiocontrast dye, please see Chapter 11 on contrast-induced nephropathy.
AKI, acute kidney injury; AKIN, Acute Kidney Injury Network; ANP, atrial natriuretic peptide; ATN, acute tubular necrosis; BNP, brain natriuretic peptide; CKD, chronic kidney disease; ICU,
intensive care unit; NGAL, neutrophil gelatinase-associated lipocalin; M, meta-analysis; RCT, randomized controlled trial; RIFLE, Risk, Injury, Failure, Loss of kidney function, and End-stage
kidney disease criteria; RIPC, remote ischemic preconditioning; RRT, renal replacement therapy; SCr, serum creatinine concentration.
0005152398.INDD 134 09-12-2022 15:14:54

Preventio 135
ability, electrolyte abnormalities (hypokalemia, sodium, and Vancomycin and Piperacillin-Tazobactam

magnesium wasting), and ATN [128]. In attempts to mini- Vancomycin is a known cause of nephrotoxic ATN, with
mize kidney injury, several preventative strategies have been incidence of AKI ranging from 5% to 43% depending on the
investigated, including salt loading (volume expansion) and population studied [130–132]. Risk factors include vanco-
the use of lipid-based formulations. mycin exposure (trough levels 15 mg/l or higher, larger
AUC, longer duration of therapy), host susceptibility to
vancomycin (increased body weight, pre-existing kidney
Volume Expansion (Salt Loading) In response to animal
dysfunction, critical illness), and concurrent nephrotoxin
data suggesting that enhanced TGF contributes to the
therapy [130].
nephrotoxicity of amphotericin B, a number of studies
Since 2011, multiple studies have demonstrated an
have evaluated the utility of salt loading through saline
increased risk of AKI with the use of vancomycin in combi-
infusion with the goal of decreasing sensitivity to TGF
nation with piperacillin-tazobactam, and five meta-analyses
and preserving GFR [129].
have been conducted to investigate this proposed increased
In 1995, Anderson published a meta-analysis of the
risk [80–84]. Giuliano et al. evaluated 15 observational
five available human studies investigating the role of salt
cohort studies including a total of 3258 patients and docu-
loading. He concluded that 150 mEq of sodium chloride
mented an association with the development of AKI with
daily intravenously or orally will likely prevent much of
vancomycin + piperacillin-tazobactam compared with van-
the observed nephrotoxicity and should be carried out
comycin ± beta-lactam (OR, 3.649) [81]. Additional studies
routinely [77]. Anderson’s conclusions were based
have demonstrated similar findings [80, 82, 84]. In the larg-
largely on one placebo-controlled trial, in which patients
est meta-analysis, the overall occurrence rate of AKI was
treated with thrice-weekly amphotericin for leishmania-
22.2% for the vancomycin + piperacillin-tazobactam group
sis were randomized to receive either 1 l of 0.9% saline or
compared with 12.9% for the comparator groups (vancomy-
1 l of 5% dextrose in water immediately prior to each
cin monotherapy, vancomycin + cefepime or carbapenem,
amphotericin B dose [48]. Results showed a reduction in
piperacillin monotherapy), yielding an overall number
creatinine clearance in the control group compared with
needed to harm of 11 [83].
patients in the saline group (P < 0.05). None of the
Taken together, these studies provide compelling evi-
patients sustained an increase in SCr to values greater
dence supporting an increased risk for AKI in those patients
than 1.7 mg/dl [78].
treated with vancomycin and piperacillin-tazobactam. This
There has been minimal recent research in this area, and
combination should be initiated cautiously and only in
it is unlikely that any randomized trials will be conducted
patients with appropriate indications for use.
due to practical and ethical considerations. While the data
above is clearly limited, there appears to be minimal harm
in giving patients a saline load (1 l of 0.9% sodium chloride
Iodinated Radiocontrast Dye
intravenously) prior to amphotericin B administration, and
Iodinated radiocontrast dye is a common cause of ATN,
this should be considered the standard of care.
with incidence of contrast-induced nephropathy (CIN)
reported to be as high as 15–25% (depending on the exact
Use of Lipid-Based Amphotericin B Formulations Several definition and agent used) in certain high-risk popula-
animal and human studies suggest that lipid-based tions receiving intra-arterial injections [133–135]. These
formulations of amphotericin B (either lipid complex or include patients with pre-existing renal impairment or
liposomal) may decrease the incidence and severity of certain risk factors such as diabetes, congestive heart fail-
kidney injury. ure, or concurrent administration of nephrotoxic medica-
In a recent systematic Cochrane review, 10 randomized tion [125, 136, 137]. A number of measures (hydration,
controlled trials (RCTs) comparing amphotericin B types of contrast, and medications) have been investi-
sodium deoxycholate with liposomal amphotericin B gated in attempts to prevent CIN [137]. Of these, the most
were meta-analyzed. The liposomal formulation was critical appears to be optimization of volume status, and
found to be significantly safer in terms of nephrotoxicity, we recommend the administration of isotonic fluids prior
defined as a 2-fold increase in SCr from baseline (RR, to and for several hours after the administration of iodi-
0.49). The sensitivity analysis performed did not change nated contrast in at-risk patients without contraindica-
the significance of difference in favor of the liposomal tions to volume expansion (see Table 10.2) [125]. For a
formulation, and the overall certainty of evidence was detailed discussion of the pathophysiology, prevention,
determined to be moderate based on GRADE Working and treatment of contrast-induced AKI, see Chapter 11
Group criteria [79]. [Weisbord and Gallagher].
T
reatment An association between positive fluid balance and mor-
tality has been demonstrated in patients with established
Many drug classes and strategies have been studied as AKI [142, 145]. In a secondary analysis of AKI patients
potential treatments for established ATN, and several rep- from a multicenter, prospective cohort study in 10 Italian
resentative examples are outlined in Table 10.4. ICUs, higher mean fluid balance was associated with
Unfortunately, the results of these investigations have been greater mortality, and diuretic use was associated with
largely disappointing. Because no single effective treat- better survival in this population (adjusted HR,
ment for ATN has been identified, the mainstays of therapy 0.25) [145]. Using data from the FACTT, 306 AKI patients
remain appreciation of AKI risk, early detection of injury, were identified and included in a post hoc analysis. Grams
initiation of appropriate supportive care, and avoidance of et al. showed that post-AKI fluid balance was significantly
further harm [154]. Treatment priorities must include (i) associated with 60-day mortality in both crude and
discontinuation of nephrotoxic medications whenever pos- adjusted analysis, and that higher post-AKI furosemide
sible; (ii) optimization of hemodynamics and RBF; (iii) doses had a protective effect on mortality but no signifi-
attention to volume management with caution to avoid cant effect after adjustment for post-AKI fluid bal-
both hypovolemia and fluid overload; (iv) correction of ance [142]. While the data does not support the routine
electrolyte and acid-base derangements; and (v) timely ini- use of diuretics in euvolemic patients with established
tiation of RRT in appropriately selected patients. ATN (who may develop volume depletion and subsequent
worsening of ischemic kidney injury), diuretics and/or
RRT may be considered in those with evidence of volume
Discontinuation of Nephrotoxic Medications overload. For a review of relevant studies investigating
All patients with established ATN require a thorough the role of loop diuretics in the treatment of ATN, see
review of their medications. Nephrotoxic drugs (NSAIDs, Table 10.4.
ACE inhibitors, ARBs, aminoglycosides) and agents such
as iodinated radiocontrast dye should be avoided or Correction of Electrolyte and Acid-Base
stopped, if possible. Additionally, medication doses should Disturbances and Timely Initiation of RRT
be adjusted to account for reduced renal clearance.
Metabolic derangements seen in established ATN include
hyperkalemia, hyponatremia, hyperphosphatemia, and
Optimization of Hemodynamics and RBF and
acidemia, and these may range in severity from mild to life-
Volume Management
threatening. Rapid rises in serum potassium concentra-
Early measures to treat established ATN must include opti- tion, as may occur in the presence of oliguria and tissue
mization of cardiac output and maintenance of hemody- breakdown (rhabdomyolysis, tumor lysis syndrome), are
namics to ensure adequate systemic perfusion and RBF; especially worrisome. In these cases, emergency treatment
the specific approach will vary based on patient factors and may be required.
the clinical scenario (hypovolemic shock, tumor lysis syn- Despite extensive research, debate continues regarding
drome, heart failure) [2]. the optimal timing, modality, and dosing of RRT in ATN.
Close attention must be paid to volume status. While vol- The KDIGO Clinical Practice Guideline for AKI suggests
ume depletion should be corrected when present, there is a initiation of RRT emergently when life-threatening
growing body of evidence documenting poorer patient out- changes in fluid, electrolyte, and acid-base balance exist. In
comes with excessive fluid administration. In the Fluid and the absence of emergent indications, KDIGO suggests that
Catheter Treatment Trial (FACTT), a conservative fluid clinicians consider the broader clinical context, the pres-
strategy (achieved through the protocolized use of furo- ence of conditions that can be modified with RRT, and
semide to target low central venous pressure or pulmonary trends in laboratory results when making the decision to
capillary wedge pressure) was associated with decreased start RRT (recommendations not graded) [125].
ICU length of stay and decreased length of mechanical
ventilation in patients with acute lung injury without an
increase in nonpulmonary-organ failures or need for dialy- C
onclusions
sis [155]. Additionally, a positive fluid balance at 72 hours
post-admission has been shown to be associated with ATN is a common clinical problem associated with
increased mortality, and some data suggest that a positive increased risk of CKD and mortality. Unfortunately, while
fluid balance as early as 12 hours may be associated with several classes of agents have been investigated, thus far no
increased risk of death [156–159]. specific pharmacologic strategies have proven effective for
Table 10.4 Select agents investigated for the treatment of established acute tubular necrosis.

for treatment of AKI Select key studies Results of studies Recommendation evidence recommendation
Alkaline Pickkers et al. [138] (RCT) No significant improvement in short-term (7 days) creatinine clearance In the absence of data Moderate Strong
phosphatase with human recombinant AP (vs. placebo) in critically ill patients with demonstrating
sepsis-associated AKI consistent
improvement in
Pickkers et al. [139] (RCT) Significant (P = 0.02) difference in favor of AP relative to controls for the
patient-centered
primary endpoint (progress in renal outcome variables as measured by
outcomes, we suggest
ECC and requirement and duration of RRT) at 28 days; ECC 50 ± 27 to
not using alkaline
108 ± 73 ml/min (mean ± SEM) in AP group vs. 40 ± 37 to 65 ± 30 ml/min
phosphatase for the
for placebo (P = 0.01); reductions in RRT requirement and duration did
treatment of AKI.
not reach statistical significance
Dopamine Friedrich et al. [102] (M) Increased urine output and improvement in SCr on day 1 with low-dose In the absence of High Strong
dopamine ( 5 mcg/kg/min); no sustained improvement in renal convincing evidence
function, decrease in need for RRT, or decrease in mortality (did not of benefit, and in
separate prophylactic from therapeutic trials) light of potential
risks, we do not
Kellum and Decker [103] No significant benefit of low-dose dopamine (<5 mcg/kg/min) for
recommend the use
(M) treatment of AKI
of dopamine for the
Lauschke et al. [140] (RCT) Increase in renal vascular resistance with low-dose dopamine (2 mcg/kg/ treatment of AKI.
min) in patients with ARF (vs. decrease in renal vascular resistance
without ARF); concern for worsening of renal perfusion in dopamine-
treated patients
Loop diuretics Cantarovich et al. [141] No difference in survival or rate of renal recovery with high dose We suggest not using Moderate Weak
(RCT) furosemide (25 mg/kg/d IV or 35 mg/kg/d PO) vs. placebo; increased diuretics to treat AKI,
UOP in furosemide group without reduction in number of treatments or except in the
time on dialysis management of
volume overload.
Grams et al. [142](RTC, Higher post-AKI fluid balance associated with increased 60-day
post hoc analysis) mortality; higher post-AKI furosemide associated with improved survival
(no significant effect after adjustment for post-AKI fluid balance)
Mehta et al. [143] Increased risk of death and nonrecovery renal function with use of
(retrospective cohort study) furosemide, bumetanide, metolazone, and/or hydrodiuril vs. no diuretics
Shilliday et al. [144] (RCT) Randomized to receive torsemide, furosemide, or placebo in addition to
dopamine and mannitol; no decrease in rate of renal recovery, need for
RRT, or death with diuretics despite increased UOP
Teixeira et al. [145] Higher MFB in nonsurviving AKI patients vs. surviving AKI patients;
(prospective cohort study, higher MFB independently associated with 28-day mortality; diuretic use
secondary analysis) associated with better survival in patients with higher MFB and lower MUV
Uchino et al. [146] No significant difference in patient mortality associated with the use of
(prospective cohort study) diuretics (almost exclusively furosemide) vs. no diuretics in critically ill
ICU patients
(Continued)
0005152398.INDD 137 09-12-2022 15:14:54


for treatment of AKI Select key studies Results of studies Recommendation evidence recommendation
Fenoldopam Brienza et al. [147] (RCT) Randomized to receive fenoldopam (0.1 mcg/kg/min) or dopamine; In the absence of data Low Weak
lower SCr in fenoldopam group after 4-day infusion (change from demonstrating
baseline, −0.041 vs. -0.09 mg/dl, P = 0.02); no decrease in need for RRT consistent improvement
in patient-centered
Bove et al. [148] (RCT) No difference in need for RRT or 30-day mortality with fenoldopam vs. outcomes, and in light
placebo; significantly more hypotension in fenoldopam group (26% vs. of the risk of
15%, P = 0.001); stopped early for futility hypotension, we
Tumlin et al. [149] (RCT) Trend toward lower 21-day mortality and decreased need for RRT with suggest not using
fenoldopam vs. placebo, but not statistically significant (underpowered, fenoldopam for the
11% absolute risk reduction) treatment of established
ischemic ATN.
Insulin-like Hirschberg et al. [150] In patients with sepsis and hypotension or hemodynamic shock, no In the absence of data Moderate Strong
growth factor-1 (RCT) differences noted in changes in GFR, urine output, need for RRT, or from high-quality
mortality with IGF-1 vs. placebo (mean treatment period 10.6 days) trials demonstrating
improvement in
patient-centered
outcomes, we do not
recommend the
routine use of IGF-1
for the treatment of
established AKI.
Natriuretic Allgren et al. [151] (RCT) No significant difference in overall dialysis-free survival at 21 days with In the absence of data Moderate Weak
peptides anaritide (0.2 mcg/kg/min) vs. placebo; improved dialysis-free survival in demonstrating
subgroup with oliguria receiving anaritide; increased incidence of consistent benefit in
hypotension in anaritide group (46% vs. 18%, P = 0.001) patient-centered
outcomes, and in
Lewis et al. [152] (RCT) No statistically significant difference in dialysis-free survival or mortality with light of the risk of
anaritide (0.2 mcg/kg/min) vs. placebo; greater incidence of hypotension hypotension, we
(SBP <90) in the anaritide-treated group (95% vs. 55%, P < 0.001) suggest not using
Nigwekar et al. [110] No difference in mortality with either low-or high-dose ANP; low-dose ANP to treat ATN.
(Cochrane systematic ANP associated with reduction in need for RRT; high-dose ANP
review) associated with more adverse events (hypotension, arrhythmias); authors
acknowledge paucity of high-quality, adequately powered data
Swärd et al. [153] (RCT) 61 postoperative patients with heart failure and AKI (not ischemic ATN)
randomized to h-ANP (50 ng/kg/min) or placebo; decreased need for
RRT and improved dialysis-free survival in h-ANP group vs. placebo; no
significant difference in incidence of hypotension
AKI, acute kidney injury; ANP, atrial natriuretic peptide; AP, alkaline phosphatase; ARF, acute renal failure; ATN, acute tubular necrosis; ECC, endogenous creatinine clearance; GFR,
glomerular filtration rate; h-ANP, human atrial natriuretic peptide; ICU, intensive care unit; IGF-1, insulin-like growth factor-1; IV, intravenous; M, meta-analysis; MFB, mean fluid balance;
MUV, mean urine volume; PO by mouth, RCT, randomized controlled trial; RRT, renal replacement therapy; SEM; standard error of the mean; SBP, systolic blood pressure; UOP, urine output.
0005152398.INDD 138 09-12-2022 15:14:54

Reference 139
preventing or treating this condition. Thus, the approach to kidney injury, and aggressive supportive care, including
ATN centers on identifying patients at increased risk, mini- the maintenance of adequate hemodynamics and manage-
mizing ischemic and toxic exposures, early detection of ment of electrolyte and acid-base derangements.
R
eferences
1 Thadhani, R., Pascual, M., and Bonventre, J.V. (1996). 15 Brivet, F.G., Kleinknecht, D.J., Loirat, P., and Landais, P.J.
Acute renal failure. N. Engl. J. Med. 334 (22): 1448–1460. (1996). Acute renal failure in intensive care units – causes,
2 Lameire, N., Van Biesen, W., and Vanholder, R. (2005). outcome, and prognostic factors of hospital mortality; a
Acute renal failure. Lancet 365 (9457): 417–430. prospective, multicenter study. French study group on
3 Hou, S.H., Bushinsky, D.A., Wish, J.B. et al. (1983). acute renal failure. Crit. Care Med. 24 (2): 192–198.
Hospital-acquired renal insufficiency: a prospective study. 16 Cole, L., Bellomo, R., Silvester, W., and Reeves, J.H.
Am. J. Med. 74 (2): 243–248. (2000). A prospective, multicenter study of the
4 Liano, F. and Pascual, J. (1996). Epidemiology of acute epidemiology, management, and outcome of severe acute
renal failure: a prospective, multicenter, community- renal failure in a “closed” ICU system. Am. J. Respir. Crit.
based study. Madrid acute renal failure study group. Care Med. 162 (1): 191–196.
Kidney Int. 50 (3): 811–818. 17 Dawson, J.L. (1965). Post-operative renal function in
5 Mehta, R.L., Pascual, M.T., Soroko, S. et al. (2004). obstructive jaundice: effect of a mannitol diuresis. Br.
Spectrum of acute renal failure in the intensive care unit: Med. J. 1 (5427): 82–86.
the PICARD experience. Kidney Int. 66 (4): 1613–1621. 18 Gornick, C.C. Jr. and Kjellstrand, C.M. (1983). Acute
6 Nash, K., Hafeez, A., and Hou, S. (2002). Hospital- renal failure complicating aortic aneurysm surgery.
acquired renal insufficiency. Am. J. Kidney Dis. 39 (5): Nephron 35 (3): 145–157.
930–936. 19 Grayson, A.D., Khater, M., Jackson, M., and Fox, M.A.
7 Fogo, A.B., Lusco, M.A., Najafian, B., and Alpers, C.E. (2003). Valvular heart operation is an independent risk
(2016). AJKD atlas of renal pathology: ischemic acute factor for acute renal failure. Ann. Thorac. Surg. 75 (6):
tubular injury. Am. J. Kidney Dis. 67 (5): e25. 1829–1835.
8 Perazella, M.A., Coca, S.G., Kanbay, M. et al. (2008). 20 Liano, F., Junco, E., Pascual, J. et al. (1998). The spectrum
Diagnostic value of urine microscopy for differential of acute renal failure in the intensive care unit compared
diagnosis of acute kidney injury in hospitalized patients. with that seen in other settings. The Madrid acute renal
Clin. J. Am. Soc. Nephrol. 3 (6): 1615–1619. failure study group. Kidney Int. Suppl. 66: S16–S24.
9 Abuelo, J.G. (2007). Normotensive ischemic acute renal 21 Neveu, H., Kleinknecht, D., Brivet, F. et al. (1996).
failure. N. Engl. J. Med. 357 (8): 797–805. Prognostic factors in acute renal failure due to sepsis.
10 Bohle, A., Christensen, J., Kokot, F. et al. (1990). Acute Results of a prospective multicentre study. The French
renal failure in man: new aspects concerning study group on acute renal failure. Nephrol. Dial.
pathogenesis. A morphometric study. Am. J. Nephrol. 10 Transplant. 11 (2): 293–299.
(5): 374–388. 22 Rosner, M.H. and Okusa, M.D. (2006). Acute kidney
11 Moeckel, G.W. (2018). Pathologic perspectives on acute injury associated with cardiac surgery. Clin. J. Am. Soc.
tubular injury assessment in the kidney biopsy. Semin. Nephrol. 1 (1): 19–32.
Nephrol. 38 (1): 21–30. 23 Wangsiripaisan, A., Gengaro, P.E., Edelstein, C.L., and
12 Hoste, E.A., Clermont, G., Kersten, A. et al. (2006). Schrier, R.W. (2001). Role of polymeric tamm-horsfall
RIFLE criteria for acute kidney injury are associated with protein in cast formation: oligosaccharide and tubular
hospital mortality in critically ill patients: a cohort fluid ions. Kidney Int. 59 (3): 932–940.
analysis. Crit. Care 10 (3): R73. 24 Basile, D.P., Anderson, M.D., and Sutton, T.A. (2012).
13 Uchino, S., Bellomo, R., Goldsmith, D. et al. (2006). An Pathophysiology of acute kidney injury. Compr. Physiol. 2
assessment of the RIFLE criteria for acute renal failure in (2): 1303–1353.
hospitalized patients. Crit. Care Med. 34 (7): 1913–1917. 25 Mullens, W., Abrahams, Z., Skouri, H.N. et al. (2008).
14 Dixit, M., Doan, T., Kirschner, R., and Dixit, N. (2010). Elevated intra-abdominal pressure in acute
Significant acute kidney injury due to non-steroidal decompensated heart failure: a potential contributor to
anti-inflammatory drugs: inpatient setting. worsening renal function? J. Am. Coll. Cardiol. 51 (3):
Pharmaceuticals (Basel) 3 (4): 1279–1285. 300–306.
6 Lameire, N.H. and Vanholder, R. (2004). Pathophysiology

2 42 Bellomo, R., Wan, L., Langenberg, C. et al. (2011). Septic
of ischaemic acute renal failure. Best Pract. Res. Clin. acute kidney injury: the glomerular arterioles. Contrib.
Anaesthesiol. 18 (1): 21–36. Nephrol. 174: 98–107.
27 Kinsey, G.R. and Okusa, M.D. (2011). Pathogenesis of 43 Heyman, S.N., Rosen, S., Fuchs, S. et al. (1996).
acute kidney injury: foundation for clinical practice. Am. Myoglobinuric acute renal failure in the rat: a role for
J. Kidney Dis. 58 (2): 291–301. medullary hypoperfusion, hypoxia, and tubular
28 Brezis, M., Rosen, S.N., and Epstein, F.H. (1989). The obstruction. J. Am. Soc. Nephrol. 7 (7): 1066–1074.
pathophysiological implications of medullary hypoxia. 44 Zager, R.A. (1996). Rhabdomyolysis and
Am. J. Kidney Dis. 13 (3): 253–258. myohemoglobinuric acute renal failure. Kidney Int. 49
29 Fish, E.M. and Molitoris, B.A. (1994). Alterations in (2): 314–326.
epithelial polarity and the pathogenesis of disease states. 45 Rosner, M.H. and Perazella, M.A. (2017). Acute kidney injury
N. Engl. J. Med. 330 (22): 1580–1588. in patients with cancer. N. Engl. J. Med. 377 (5): 500–501.
30 Gailit, J., Colflesh, D., Rabiner, I. et al. (1993). 46 Makris, K. and Spanou, L. (2016). Acute kidney injury:
Redistribution and dysfunction of integrins in cultured diagnostic approaches and controversies. Clin. Biochem.
renal epithelial cells exposed to oxidative stress. Am. J. Rev. 37 (4): 153–175.
Phys. 264 (1 Pt 2): F149–F157. 47 Carvounis, C.P., Nisar, S., and Guro-Razuman, S. (2002).
31 Molitoris, B.A., Dahl, R., and Geerdes, A. (1992). Significance of the fractional excretion of urea in the
Cytoskeleton disruption and apical redistribution of differential diagnosis of acute renal failure. Kidney Int. 62
proximal tubule Na(+)-K(+)-ATPase during ischemia. (6): 2223–2229.
Am. J. Phys. 263 (3 Pt 2): F488–F495. 48 Molony, D.A. and Craig, J.C. (2009). Evidence-based
32 Bonventre, J.V. and Weinberg, J.M. (2003). Recent nephrology. Wiley Online Library UBCM All Obooks.
advances in the pathophysiology of ischemic acute renal http://RE5QY4SB7X.search.serialssolutions.com/?V=1.0
failure. J. Am. Soc. Nephrol. 14 (8): 2199–2210. &L=RE5QY4SB7X&S=JCs&C=TC0000151296&T=marc
33 Kaushal, G.P., Basnakian, A.G., and Shah, S.V. (2004). (accessed 22 May 2020).
Apoptotic pathways in ischemic acute renal failure. 49 Corwin, H.L., Schreiber, M.J., and Fang, L.S. (1984). Low
Kidney Int. 66 (2): 500–506. fractional excretion of sodium. Occurrence with
34 Donohoe, J.F., Venkatachalam, M.A., Bernard, D.B., and hemoglobinuric-and myoglobinuric-induced acute renal
Levinsky, N.G. (1978). Tubular leakage and obstruction failure. Arch. Intern. Med. 144 (5): 981–982.
after renal ischemia: structural-functional correlations. 50 Fang, L.S., Sirota, R.A., Ebert, T.H., and Lichtenstein, N.S.
Kidney Int. 13 (3): 208–222. (1980). Low fractional excretion of sodium with contrast
35 Hollenberg, N.K., Adams, D.F., Oken, D.E. et al. (1970). media-induced acute renal failure. Arch. Intern. Med. 140
Acute renal failure due to nephrotoxins. N. Engl. J. Med. (4): 531–533.
282 (24): 1329–1334. 51 Vaz, A.J. (1983). Low fractional excretion of urine sodium
36 Hollenberg, N.K., Epstein, M., Rosen, S.M. et al. (1968). in acute renal failure due to sepsis. Arch. Intern. Med. 143
Acute oliguric renal failure in man: evidence for (4): 738–739.
preferential renal cortical ischemia. Medicine (Baltimore) 52 Zarich, S., Fang, L.S., and Diamond, J.R. (1985).
47 (6): 455–474. Fractional excretion of sodium. Exceptions to its
37 Karlberg, L., Norlen, B.J., Ojteg, G., and Wolgast, M. diagnostic value. Arch. Intern. Med. 145 (1): 108–112.
(1983). Impaired medullary circulation in postischemic 53 Perazella, M.A., Coca, S.G., Hall, I.E. et al. (2010). Urine
acute renal failure. Acta Physiol. Scand. 118 (1): 11–17. microscopy is associated with severity and worsening of
38 Reubi, F.C. (1974). The pathogenesis of anuria following acute kidney injury in hospitalized patients. Clin. J. Am.
shock. Kidney Int. 5 (2): 106–110. Soc. Nephrol. 5 (3): 402–408.
39 Mason, J., Takabatake, T., Olbricht, C., and Thurau, K. 54 Kanbay, M., Kasapoglu, B., and Perazella, M.A. (2010).
(1978). The early phase of experimental acute renal failure. Acute tubular necrosis and pre-renal acute kidney injury:
III. Tubologlomerular feedback. Pflugers Arch. 373 (1): 69–76. utility of urine microscopy in their evaluation-a
40 Langenberg, C., Wan, L., Egi, M. et al. (2007). Renal blood systematic review. Int. Urol. Nephrol. 42 (2): 425–433.
flow and function during recovery from experimental 55 Malhotra, R. and Siew, E.D. (2017). Biomarkers for the
septic acute kidney injury. Intensive Care Med. 33 (9): early detection and prognosis of acute kidney injury. Clin.
1614–1618. J. Am. Soc. Nephrol. 12 (1): 149–173.
41 Wan, L., Langenberg, C., Bellomo, R., and May, C.N. 56 Han, W.K., Waikar, S.S., Johnson, A. et al. (2008). Urinary
(2009). Angiotensin II in experimental hyperdynamic biomarkers in the early diagnosis of acute kidney injury.
sepsis. Crit. Care 13 (6): R190. Kidney Int. 73 (7): 863–869.
Reference 141
7 Parikh, C.R., Coca, S.G., Thiessen-Philbrook, H. et al.

5 after cardiac surgery-associated acute kidney injury
(2011). Postoperative biomarkers predict acute kidney despite clinical recovery. Nephrol. Dial. Transplant.
injury and poor outcomes after adult cardiac surgery. J. 71 Ali, M.Z. and Goetz, M.B. (1997). A meta-analysis of the
Am. Soc. Nephrol. 22 (9): 1748–1757. relative efficacy and toxicity of single daily dosing versus
58 Kashani, K., Al-Khafaji, A., Ardiles, T. et al. (2013). multiple daily dosing of aminoglycosides. Clin. Infect. Dis.
Discovery and validation of cell cycle arrest biomarkers 24 (5): 796–809.
in human acute kidney injury. Crit. Care 17 (1): R25. 72 Blaser, J. and Konig, C. (1995). Once-daily dosing of
59 Meersch, M., Schmidt, C., Hoffmeier, A. et al. (2017). aminoglycosides. Eur. J. Clin. Microbiol. Infect. Dis. 14
Prevention of cardiac surgery-associated AKI by (12): 1029–1038.
implementing the KDIGO guidelines in high risk patients 73 Barza, M., Ioannidis, J.P., Cappelleri, J.C., and Lau, J.
identified by biomarkers: the PrevAKI randomized (1996). Single or multiple daily doses of aminoglycosides:
controlled trial. Intensive Care Med. 43 (11): 1551–1561. a meta-analysis. BMJ 312 (7027): 338–345.
60 Myers, B.D. and Moran, S.M. (1986). Hemodynamically 74 Ferriols-Lisart, R. and Alos-Alminana, M. (1996).
mediated acute renal failure. N. Engl. J. Med. 314 (2): Effectiveness and safety of once-daily aminoglycosides: a
97–105. meta-analysis. Am. J. Health Syst. Pharm. 53 (10):
61 Spurney, R.F., Fulkerson, W.J., and Schwab, S.J. (1991). 1141–1150.
Acute renal failure in critically ill patients: prognosis for 75 Hatala, R., Dinh, T., and Cook, D.J. (1996). Once-daily
recovery of kidney function after prolonged dialysis aminoglycoside dosing in immunocompetent adults: a
support. Crit. Care Med. 19 (1): 8–11. meta-analysis. Ann. Intern. Med. 124 (8): 717–725.
62 Schiffl, H. (2006). Renal recovery from acute tubular 76 Munckhof, W.J., Grayson, M.L., and Turnidge, J.D. (1996).
necrosis requiring renal replacement therapy: a A meta-analysis of studies on the safety and efficacy of
prospective study in critically ill patients. Nephrol. Dial. aminoglycosides given either once daily or as divided
Transplant. 21 (5): 1248–1252. doses. J. Antimicrob. Chemother. 37 (4): 645–663.
63 Basile, D.P., Bonventre, J.V., Mehta, R. et al. (2016). 77 Anderson, C.M. (1995). Sodium chloride treatment of
Progression after AKI: understanding maladaptive repair amphotericin B nephrotoxicity. Standard of care? West. J.
processes to predict and identify therapeutic treatments. Med. 162 (4): 313–317.
J. Am. Soc. Nephrol. 27 (3): 687–697. 78 Llanos, A., Cieza, J., Bernardo, J. et al. (1991). Effect of
64 Hsu, C.Y., Chertow, G.M., McCulloch, C.E. et al. (2009). salt supplementation on amphotericin B nephrotoxicity.
Nonrecovery of kidney function and death after acute on Kidney Int. 40 (2): 302–308.
chronic renal failure. Clin. J. Am. Soc. Nephrol. 4 (5): 79 Botero Aguirre, J.P. and Restrepo Hamid, A.M. (2015).
891–898. Amphotericin B deoxycholate versus liposomal
65 Chawla, L.S., Amdur, R.L., Amodeo, S. et al. (2011). The amphotericin B: effects on kidney function. Cochrane
severity of acute kidney injury predicts progression to Database Syst. Rev. (11): CD010481.
chronic kidney disease. Kidney Int. 79 (12): 1361–1369. 80 Chen, X.Y., Xu, R.X., Zhou, X. et al. (2018). Acute kidney
66 Schmitt, R., Coca, S., Kanbay, M. et al. (2008). Recovery injury associated with concomitant vancomycin and
of kidney function after acute kidney injury in the piperacillin/tazobactam administration: a systematic
elderly: a systematic review and meta-analysis. Am. J. review and meta-analysis. Int. Urol. Nephrol.
Kidney Dis. 52 (2): 262–271. 81 Giuliano, C.A., Patel, C.R., and Kale-Pradhan, P.B. (2016).
67 Hoste, E.A., Bagshaw, S.M., Bellomo, R. et al. (2015). Is the combination of piperacillin-tazobactam and
Epidemiology of acute kidney injury in critically ill vancomycin associated with development of acute kidney
patients: the multinational AKI-EPI study. Intensive Care injury? A meta-analysis. Pharmacotherapy 36 (12):
Med. 41 (8): 1411–1423. 1217–1228.
68 Bucaloiu, I.D., Kirchner, H.L., Norfolk, E.R. et al. (2012). 82 Hammond, D.A., Smith, M.N., Li, C. et al. (2017).
Increased risk of death and de novo chronic kidney Systematic review and meta-analysis of acute kidney
disease following reversible acute kidney injury. Kidney injury associated with concomitant vancomycin and
Int. 81 (5): 477–485. piperacillin/tazobactam. Clin. Infect. Dis. 64 (5): 666–674.
69 Husain-Syed, F., Ferrari, F., Sharma, A. et al. (2018). 83 Luther, M.K., Timbrook, T.T., Caffrey, A.R. et al. (2018).
Preoperative renal functional reserve predicts risk of Vancomycin plus piperacillin-tazobactam and acute
acute kidney injury after cardiac operation. Ann. Thorac. kidney injury in adults: a systematic review and meta-
Surg. 105 (4): 1094–1101. analysis. Crit. Care Med. 46 (1): 12–20.
70 Husain-Syed, F., Ferrari, F., Sharma, A. et al. (2018). 84 Mellen, C.K., Ryba, J.E., and Rindone, J.P. (2017). Does
Persistent decrease of renal functional reserve in patients piperacillin-tazobactam increase the risk of
nephrotoxicity when used with vancomycin: a meta- 96 Scharman, E.J. and Troutman, W.G. (2013). Prevention
analysis of observational trials. Curr. Drug Saf. 12 (1): of kidney injury following rhabdomyolysis: a systematic
62–66. review. Ann. Pharmacother. 47 (1): 90–105.
85 Brar, S.S., Aharonian, V., Mansukhani, P. et al. (2014). 97 Zacharias, M., Mugawar, M., Herbison, G.P. et al. (2013).
Haemodynamic-guided fluid administration for the Interventions for protecting renal function in the
prevention of contrast-induced acute kidney injury: the perioperative period. Cochrane Database Syst. Rev. (9):
POSEIDON randomised controlled trial. Lancet 383 CD003590.
(9931): 1814–1823. 98 Lamy, A., Devereaux, P.J., Prabhakaran, D. et al. (2012).
86 Jurado-Roman, A., Hernandez-Hernandez, F., Garcia- Off-pump or on-pump coronary-artery bypass grafting
Tejada, J. et al. (2015). Role of hydration in contrast- at 30 days. N. Engl. J. Med. 366 (16): 1489–1497.
induced nephropathy in patients who underwent primary 99 Nigwekar, S.U., Kandula, P., Hix, J.K., and Thakar, C.V.
percutaneous coronary intervention. Am. J. Cardiol. 115 (2009). Off-pump coronary artery bypass surgery and
(9): 1174–1178. acute kidney injury: a meta-analysis of randomized and
87 Luo, Y., Wang, X., Ye, Z. et al. (2014). Remedial observational studies. Am. J. Kidney Dis. 54 (3): 413–423.
hydration reduces the incidence of contrast-induced 100 Seabra, V.F., Alobaidi, S., Balk, E.M. et al. (2010).
nephropathy and short-term adverse events in patients Off-pump coronary artery bypass surgery and acute
with ST-segment elevation myocardial infarction: a kidney injury: a meta-analysis of randomized controlled
single-center, randomized trial. Intern. Med. 53 (20): trials. Clin. J. Am. Soc. Nephrol. 5 (10): 1734–1744.
2265–2272. 101 Bellomo, R., Chapman, M., Finfer, S. et al. (2000).
88 Nijssen, E.C., Rennenberg, R.J., Nelemans, P.J. et al. Low-dose dopamine in patients with early renal
(2017). Prophylactic hydration to protect renal function dysfunction: a placebo-controlled randomised trial.
from intravascular iodinated contrast material in patients Australian and New Zealand intensive care society
at high risk of contrast-induced nephropathy (ANZICS) clinical trials group. Lancet 356 (9248):
(AMACING): a prospective, randomised, phase 3, 2139–2143.
controlled, open-label, non-inferiority trial. Lancet 389 102 Friedrich, J.O., Adhikari, N., Herridge, M.S., and
(10076): 1312–1322. Beyene, J. (2005). Meta-analysis: low-dose dopamine
89 Trivedi, H.S., Moore, H., Nasr, S. et al. (2003). A increases urine output but does not prevent renal
randomized prospective trial to assess the role of saline dysfunction or death. Ann. Intern. Med. 142 (7):
hydration on the development of contrast nephrotoxicity. 510–524.
Nephron Clin. Pract. 93 (1): C29–C34. 103 Kellum, J.A. and M Decker J. (2001). Use of dopamine
90 Weisbord, S.D., Gallagher, M., Jneid, H. et al. (2018). in acute renal failure: a meta-analysis. Crit. Care Med.
Outcomes after angiography with sodium bicarbonate 29 (8): 1526–1531.
and acetylcysteine. N. Engl. J. Med. 378 (7): 603–614. 104 Bove, T., Landoni, G., Calabro, M.G. et al. (2005).
91 Zoungas, S., Ninomiya, T., Huxley, R. et al. (2009). Renoprotective action of fenoldopam in high-risk
Systematic review: sodium bicarbonate treatment patients undergoing cardiac surgery: a prospective,
regimens for the prevention of contrast-induced double-blind, randomized clinical trial. Circulation 111
nephropathy. Ann. Intern. Med. 151 (9): 631–638. (24): 3230–3235.
92 Gunal, A.I., Celiker, H., Dogukan, A. et al. (2004). Early 105 Gillies, M.A., Kakar, V., Parker, R.J. et al. (2015).
and vigorous fluid resuscitation prevents acute renal Fenoldopam to prevent acute kidney injury after major
failure in the crush victims of catastrophic earthquakes. J. surgery-a systematic review and meta-analysis. Crit.
Am. Soc. Nephrol. 15 (7): 1862–1867. Care 19:449:1–9.
93 Ron, D., Taitelman, U., Michaelson, M. et al. (1984). 106 Landoni, G., Biondi-Zoccai, G.G., Tumlin, J.A. et al.
Prevention of acute renal failure in traumatic (2007). Beneficial impact of fenoldopam in critically ill
rhabdomyolysis. Arch. Intern. Med. 144 (2): 277–280. patients with or at risk for acute renal failure: a
94 Brown, C.V., Rhee, P., Chan, L. et al. (2004). Preventing meta-analysis of randomized clinical trials. Am. J.
renal failure in patients with rhabdomyolysis: do Kidney Dis. 49 (1): 56–68.
bicarbonate and mannitol make a difference? J. Trauma 107 Morelli, A., Ricci, Z., Bellomo, R. et al. (2005).
56 (6): 1191–1196. Prophylactic fenoldopam for renal protection in sepsis: a
95 Kim, E., Choi, Y.H., Lim, J.Y. et al. (2018). The effect of randomized, double-blind, placebo-controlled pilot trial.
early urine alkalinization on occurrence rhabdomyolysis Crit. Care Med. 33 (11): 2451–2456.
and hospital stay in high dose doxylamine ingestion. Am. 108 Bove, T., Belletti, A., Putzu, A. et al. (2018). Intermittent
J. Emerg. Med. 36 (7): 1170–1173. furosemide administration in patients with or at risk for
Reference 143
acute kidney injury: meta-analysis of randomized trials. 122 Leehey, D.J., Braun, B.I., Tholl, D.A. et al. (1993). Can
PLoS ONE 13 (4): e0196088. pharmacokinetic dosing decrease nephrotoxicity
109 Hager, B., Betschart, M., and Krapf, R. (1996). Effect of associated with aminoglycoside therapy. J. Am. Soc.
postoperative intravenous loop diuretic on renal Nephrol. 4 (1): 81–90.
function after major surgery. Schweiz. Med. Wochenschr. 123 Lopez-Novoa, J.M., Quiros, Y., Vicente, L. et al. (2011).
126 (16): 666–673. New insights into the mechanism of aminoglycoside
110 Ho, K.M. and Sheridan, D.J. (2006). Meta-analysis of nephrotoxicity: an integrative point of view. Kidney Int.
frusemide to prevent or treat acute renal failure. BMJ 79 (1): 33–45.
333 (7565): 420. 124 Wargo, K.A. and Edwards, J.D. (2014). Aminoglycoside-
111 Ho, K.M. and Power, B.M. (2010). Benefits and risks of induced nephrotoxicity. J. Pharm. Pract. 27 (6): 573–577.
furosemide in acute kidney injury. Anaesthesia 65 (3): 125 Kidney Disease: Improving Global Outcomes (KDIGO)
283–293. Acute Kidney Injury Work Group (2012). Section 3:
112 Lassnigg, A., Donner, E., Grubhofer, G. et al. (2000). Prevention and treatment of AKI. Kidney Int. Suppl. 2
Lack of renoprotective effects of dopamine and (1): 37–68.
furosemide during cardiac surgery. J. Am. Soc. Nephrol. 126 Bates, D.W., Su, L., Yu, D.T. et al. (2001). Correlates of
11 (1): 97–104. acute renal failure in patients receiving parenteral
113 Nigwekar, S.U., Navaneethan, S.D., Parikh, C.R., and amphotericin B. Kidney Int. 60 (4): 1452–1459.
Hix, J.K. (2009). Atrial natriuretic peptide for preventing 127 Harbarth, S., Pestotnik, S.L., Lloyd, J.F. et al. (2001). The
and treating acute kidney injury. Cochrane Database epidemiology of nephrotoxicity associated with
Syst. Rev. (4): CD006028. conventional amphotericin B therapy. Am. J. Med. 111
114 Nigwekar, S.U. and Hix, J.K. (2009). The role of (7): 528–534.
natriuretic peptide administration in cardiovascular 128 Sabra, R. and Branch, R.A. (1990). Amphotericin B
surgery-associated renal dysfunction: a systematic nephrotoxicity. Drug Saf. 5 (2): 94–108.
review and meta-analysis of randomized controlled 129 Gerkens, J.F. and Branch, R.A. (1980). The influence of
trials. J. Cardiothorac. Vasc. Anesth. 23 (2): 151–160. sodium status and furosemide on canine acute
115 Menting, T.P., Wever, K.E., Ozdemir-van Brunschot, amphotericin B nephrotoxicity. J. Pharmacol. Exp. Ther.
D.M. et al. (2017). Ischaemic preconditioning for the 214 (2): 306–311.
reduction of renal ischaemia reperfusion injury. 130 Carreno, J.J., Kenney, R.M., and Lomaestro, B. (2014).
Cochrane Database Syst. Rev. 3: CD010777. Vancomycin-associated renal dysfunction: where are we
116 Billings, F.T. 4th, Hendricks, P.A., Schildcrout, J.S. et al. now? Pharmacotherapy 34 (12): 1259–1268. doi:
(2016). High-dose perioperative atorvastatin and acute 10.1002/phar.1488 [doi].
kidney injury following cardiac surgery: a randomized 131 Farber, B.F. and Moellering, R.C. Jr. (1983).
clinical trial. JAMA 315 (9): 877–888. Retrospective study of the toxicity of preparations of
117 Park, J.H., Shim, J.K., Song, J.W. et al. (2016). Effect of vancomycin from 1974 to 1981. Antimicrob. Agents
atorvastatin on the incidence of acute kidney injury Chemother. 23 (1): 138–141.
following valvular heart surgery: a randomized, placebo- 132 Jeffres, M.N., Isakow, W., Doherty, J.A. et al. (2007).
controlled trial. Intensive Care Med. 42 (9): 1398–1407. A retrospective analysis of possible renal toxicity
118 Prowle, J.R., Calzavacca, P., Licari, E. et al. (2012). Pilot associated with vancomycin in patients with health
double-blind, randomized controlled trial of short-term care-associated methicillin-resistant staphylococcus
atorvastatin for prevention of acute kidney injury after aureus pneumonia. Clin. Ther. 29 (6): 1107–1115.
cardiac surgery. Nephrology (Carlton) 17 (3): 215–224. 133 Aspelin, P., Aubry, P., Fransson, S.G. et al. (2003).
119 Bertino, J.S. Jr., Booker, L.A., Franck, P.A. et al. (1993). Nephrotoxic effects in high-risk patients undergoing
Incidence of and significant risk factors for angiography. N. Engl. J. Med. 348 (6): 491–499.
aminoglycoside-associated nephrotoxicity in patients 134 Brar, S.S., Shen, A.Y., Jorgensen, M.B. et al. (2008).
dosed by using individualized pharmacokinetic Sodium bicarbonate vs sodium chloride for the
monitoring. J. Infect. Dis. 167 (1): 173–179. prevention of contrast medium-induced
120 Kacew, S. and Bergeron, M.G. (1990). Pathogenic factors nephropathy in patients undergoing coronary
in aminoglycoside-induced nephrotoxicity. Toxicol. Lett. angiography: a randomized trial. JAMA 300 (9):
51 (3): 241–259. discussion 237-9. 1038–1046.
121 Laurent, G., Kishore, B.K., and Tulkens, P.M. (1990). 135 Rudnick, M.R., Davidson, C., Laskey, W. et al. (2008).
Aminoglycoside-induced renal phospholipidosis and VALOR trial investigators. Nephrotoxicity of iodixanol
nephrotoxicity. Biochem. Pharmacol. 40 (11): 2383–2392. versus ioversol in patients with chronic kidney disease:
the visipaque angiography/interventions with laboratory 148 Bove, T., Zangrillo, A., Guarracino, F. et al. (2014). Effect
outcomes in renal insufficiency (VALOR) trial. Am. of fenoldopam on use of renal replacement therapy
Heart J. 156 (4): 776–782. among patients with acute kidney injury after cardiac
136 Barrett, B.J. (1994). Contrast nephrotoxicity. J. Am. Soc. surgery: a randomized clinical trial. JAMA 312 (21):
Nephrol. 5 (2): 125–137. 2244–2253.
137 Kwok, C.S., Pang, C.L., Yeong, J.K., and Loke, Y.K. 149 Tumlin, J.A., Finkel, K.W., Murray, P.T. et al. (2005).
(2013). Measures used to treat contrast-induced Fenoldopam mesylate in early acute tubular necrosis: a
nephropathy: overview of reviews. Br. J. Radiol. 86 randomized, double-blind, placebo-controlled clinical
(1021): 20120272. trial. Am. J. Kidney Dis. 46 (1): 26–34.
138 Pickkers, P., Mehta, R.L., Murray, P.T. et al. (2018). 150 Hirschberg, R., Kopple, J., Lipsett, P. et al. (1999).
Effect of human recombinant alkaline phosphatase on Multicenter clinical trial of recombinant human
7-day creatinine clearance in patients with sepsis- insulin-like growth factor I in patients with acute renal
associated acute kidney injury: a randomized clinical failure. Kidney Int. 55 (6): 2423–2432.
trial. JAMA 320 (19): 1998–2009. 151 Allgren, R.L., Marbury, T.C., Rahman, S.N. et al. (1997).
139 Pickkers, P., Heemskerk, S., Schouten, J. et al. (2012). Anaritide in acute tubular necrosis. auriculin anaritide
Alkaline phosphatase for treatment of sepsis-induced acute renal failure study group. N. Engl. J. Med. 336 (12):
acute kidney injury: a prospective randomized double- 828–834.
blind placebo-controlled trial. Crit. Care 16 (1): R14. 152 Lewis, J., Salem, M.M., Chertow, G.M. et al. (2000).
140 Lauschke, A., Teichgraber, U.K., Frei, U., and Eckardt, Atrial natriuretic factor in oliguric acute renal failure.
K.U. (2006). ’Low-dose’ dopamine worsens renal anaritide acute renal failure study group. Am. J. Kidney
perfusion in patients with acute renal failure. Kidney Dis. 36 (4): 767–774.
Int. 69 (9): 1669–1674. 153 Sward, K., Valsson, F., Odencrants, P. et al. (2004).
141 Cantarovich, F., Rangoonwala, B., Lorenz, H. et al. (2004). Recombinant human atrial natriuretic peptide in
High-dose flurosemide in acute renal failure study group. ischemic acute renal failure: a randomized placebo-
High-dose furosemide for established ARF: a prospective, controlled trial. Crit. Care Med. 32 (6): 1310–1315.
randomized, double-blind, placebo-controlled, 154 Alobaidi, R., Basu, R.K., Goldstein, S.L., and Bagshaw,
multicenter trial. Am. J. Kidney Dis. 44 (3): 402–409. S.M. (2015). Sepsis-associated acute kidney injury.
142 Grams, M.E., Estrella, M.M., Coresh, J. et al. (2011). Semin. Nephrol. 35 (1): 2–11.
National Heart, Lung, and Blood Institute acute 155 National Heart, Lung, and Blood Institute Acute
respiratory distress syndrome network. Fluid balance, Respiratory Distress Syndrome (ARDS) Clinical Trials
diuretic use, and mortality in acute kidney injury. Clin. Network, Wiedemann HP, Wheeler APet al (2006).
J. Am. Soc. Nephrol. 6 (5): 966–973. Comparison of two fluid-management strategies in
143 Mehta, R.L., Pascual, M.T., Soroko, S., and Chertow, acute lung injury. N. Engl. J. Med. 354 (24):
G.M., PICARD Study Group. (2002). Diuretics, 2564–2575.
mortality, and nonrecovery of renal function in acute 156 Alsous, F., Khamiees, M., DeGirolamo, A. et al. (2000).
renal failure. JAMA 288 (20): 2547–2553. Negative fluid balance predicts survival in patients with
144 Shilliday, I.R., Quinn, K.J., and Allison, M.E. (1997). Loop septic shock: a retrospective pilot study. Chest 117 (6):
diuretics in the management of acute renal failure: a 1749–1754.
prospective, double-blind, placebo-controlled, randomized 157 Boyd, J.H., Forbes, J., Nakada, T.A. et al. (2011). Fluid
study. Nephrol. Dial. Transplant. 12 (12): 2592–2596. resuscitation in septic shock: a positive fluid balance
145 Teixeira, C., Garzotto, F., Piccinni, P. et al. (2013). Fluid and elevated central venous pressure are associated with
balance and urine volume are independent predictors of increased mortality. Crit. Care Med. 39 (2): 259–265.
mortality in acute kidney injury. Crit. Care 17 (1): R14. 158 Jaffee, W., Hodgins, S., and McGee, W.T. (2018). Tissue
146 Uchino, S., Doig, G.S., Bellomo, R. et al. (2004). edema, fluid balance, and patient outcomes in severe
Diuretics and mortality in acute renal failure. Crit. Care sepsis: an organ systems review. J. Intensive Care Med.
Med. 32 (8): 1669–1677. 33 (9): 502–509.
147 Brienza, N., Malcangi, V., Dalfino, L. et al. (2006). 159 Sakr, Y., Rubatto Birri, P.N., Kotfis, K. et al. (2017).
A comparison between fenoldopam and low-dose Higher fluid balance increases the risk of death from
dopamine in early renal dysfunction of critically ill sepsis: results from a large international audit. Crit. Care
patients. Crit. Care Med. 34 (3): 707–714. Med. 45 (3): 386–394.
145
11
Iodinated Contrast and Acute Kidney Injury

Steven D. Weisbord1 and Martin Gallagher2
1
Renal Section and Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System and Renal Electrolyte Division, University of Pittsburgh School of
Medicine, Pittsburgh USA
2
Renal & Metabolic Division The George Institute for Global Health Newtown, NSW, Australia
Acute kidney injury (AKI) is a widely recognized complica- P

athophysiology
tion of intravascular administration of iodinated contrast
media. The pathophysiology of contrast-associated acute The potential harms of the intravascular administration of
kidney injury (CA-AKI) is thought to be related to hemody- radiological, iodine-based contrast media have been appre-
namic perturbations in the kidney from contrast, direct toxic ciated for over 60 years, with the recognized risk of AKI well
effects of contrast on tubular epithelial cells, and the genera- described since the 1970s [1]. A series of pathophysiological
tion of reactive oxygen species following contrast adminis- processes likely underlie the development of contrast-
tration. While typically defined by small increments in induced CKI (Figure 11.1). After iodinated contrast is
serum creatinine within several days following contrast administered into the systemic circulation, blood flow to
administration, the incidence of CA-AKI varies based on the the outer renal medulla decreases while active transport
precise definition employed, clinical characteristics of the and oxygen requirements in the distal nephron increase.
patient population, and nature of the procedure being per-
formed. The primary patient-related risk factor for CA-AKI
Intravascular
is underlying kidney impairment, while absolute and effec- administration
tive circulating volume also increases risk. A multitude of of iodinated contrast
studies have demonstrated that CA-AKI is associated with media
serious adverse short- and long-term outcomes, including

death. However, it is currently unknown whether these
Mismatch of
associations are causal. This is important as a growing num- oxygen supply Direct
Generation of
reactive oxygen
ber of studies suggest that patients with chronic kidney dis- demand in outer cytotoxicity
species
medulla
ease (CKD), who are elevated risk for CA-AKI, are less likely
to undergo indicated contrast-enhanced procedures such as
coronary angiography compared with patients without
CKD, likely out of fear by providers of precipitating CA-AKI. Renal tubular
epithelial cell
As CA-AKI is a potentially preventable form of renal injury, injury/acute
there has been substantial effort to identify interventions to tubular necrosis
reduce risk. Studies over the past several decades inform the
current evidence basis for the prevention of this condition.
Understanding the data that inform the current evidence
Reduced GFR/CA-AKI
basis for prevention is essential to minimize the burden of
AKI related to contrast and to enhance the appropriate
implementation of clinically indicated contrast-enhanced Figure 11.1 Potential processes underlying the
procedures in patients at risk for this iatrogenic condition. pathophysiology of CA-AKI.
146 Iodinated Contrast and Acute Kidney Injury
The resulting mismatch in oxygen supply and demand in an variables, as well as the criteria employed to define the
anatomical portion of the kidney with low baseline oxygen development of AKI. Weisbord et al. demonstrated that
reserve leads to ischemia of tubular epithelial cells [2, 3]. CA-AKI, defined by an increase in serum creatinine of
Further epithelial cell injury results from both the direct 25%, developed in 13.2%, 8.5%, and 6.5% of patients fol-
toxicity of the filtered iodinated contrast and the generation lowing nonemergent noncoronary angiography, coronary
of injurious reactive oxygen species [4–9]. After being fil- angiography, and contrast-enhanced computed tomogra-
tered at the glomerulus, iodinated contrast enters the tubu- phy, respectively [20]. In a study of 1111 hospitalized
lar lumen where it has been shown to have cytotoxic effects patients by Shema et al., CA-AKI, defined as an increase in
that result in tubular epithelial cell necrosis [6–8, 10]. serum creatinine of 0.5 mg/dl within 1–5 days, was identi-
Contrast administration also leads to increased production fied in 44% of diabetic patients with underlying kidney dis-
of oxygen free radicals in the kidney that further contribute ease [21]. D’Elia et al. demonstrated that 33% of patients
to epithelial and endothelial cell injury [11–17]. The cumu- with CKD developed CA-AKI, defined based on an increase
lative result of these effects is acute tubular necrosis and an in serum creatinine of 1.0 mg/dl following angiogra-
abrupt decline in kidney function. phy [22]. Hence, the incidence is highly dependent upon
the definition used, the characteristics of the patient popu-
lation, and the nature of the procedure for which radiocon-
Nomenclature and Definition trast is administered.
Over the past decade, a series of studies have questioned
Historically, AKI following the administration of iodinated the relationship between intravascular radiocontrast
contrast media was referred to as contrast nephropathy or administration and CA-AKI, demonstrating comparable
contrast-induced nephropathy. More recently, the Kidney rates of AKI between patients who underwent radiographic
Disease Improving Global Outcomes (KDIGO) organiza- procedures without contrast and patients who underwent
tion proposed the term contrast-induced AKI [18, 19]. contrast-enhanced procedures [23–31] (Table 11.1). Bruce
However, in many patients, other factors along with con- et al. investigated the incidence of AKI, defined by an
trast administration likely contribute to the observed decline increase in serum creatinine 0.5 mg/dl or decrease in esti-
in kidney function. Consequently, contrast-associated AKI mated glomerular filtration rate (eGFR) 25%, among
(CA-AKI) may be a more appropriate term. CA-AKI has 11 588 patients who underwent computed tomography
commonly been defined by an increment in serum creati- either with (n = 5790) or without (n = 7484) intravenous
nine of 0.5 mg/dl or 25% within 48–72 hours of contrast (IV) contrast [23]. Among those with CKD, the incidence of
administration. While this definition is highly sensitive, it AKI following computed tomography without contrast
has poor specificity for contrast-induced damage as multi- (8.8%) was comparable to the incidence following com-
ple other factors, including fluid shifts, medications, and puted tomography with IV contrast (9.7% with iso-osmolal
atheroembolic disease can cause an increase in serum cre- iodixanol and 9.9% with low-osmolal iohexol). Several sub-
atinine, with or without underlying kidney injury, inde- sequent studies demonstrated similar findings [26, 28,
pendent of the effect of contrast. Nonetheless, this 32–34]. McDonald and colleagues conducted a meta-
continues to be among the most commonly used defini- analysis comprising 13 studies with nearly 26 000 patients
tions of this condition. In keeping with efforts to establish a and reported a nonstatistically significant lower risk for
unifying definition of AKI that is independent of etiology, AKI following IV contrast-enhanced procedures compared
CA-AKI has also been defined based on an increase in to unenhanced procedures (relative risk 0.79, 95% confi-
serum creatinine of 0.3 mg/dl within 48 hours [18]. dence interval [CI] 0.62–1.02) [35]. While the majority of
However, this definition also has poor specificity. In light of such studies examined patients undergoing computed
the issues associated with defining a condition based on tomography, others have broadened the study population.
nominal increments in a biological parameter (i.e. serum In a study of hospitalized patients, Wilhelm-Leen et al.
creatinine) that is not specific for kidney injury, efforts to demonstrated a lower risk for AKI among hospitalized
identify other biomarkers that better define and more accu- patients who received intravascular contrast as part of any
rately predict the prognosis of CA-AKI are needed. type of procedure during their hospitalization compared
with patients who did not receive contrast (5.1% vs. 5.6%,
adjusted odds ratio [OR] 0.93, 95% CI 0.88–0.97) [26]. Caspi
Incidence et al. demonstrated a similar incidence of AKI in a cohort of
2025 patients with ST segment elevation myocardial infarc-
The reported incidence of CA-AKI has varied across studies tion treated with percutaneous intervention compared with
due to differences in patient characteristics and procedural 1025 patients treated with fibrinolysis (10.3% vs. 12.1%,
Risk Factor 147
Table 11.1 Studies comparing incidence of CA-AKI based on receipt of contrast.
Authors Type of procedure AKI with contrast (%) AKI without contrast (%) OR/RR (95% CI)
Bruce et al. [23] CT 9.9/9.7a 8.8 Not reported

Ehrmann et al. [31] Not specified 7.5 5.3 1.57 (0.69–3.53)
McDonald et al. [30] CT 4.8 5.1 0.94 (0.83–1.07)
McDonald et al. [28]b CT 9.9 16 0.68 (0.55–0.84)
McDonald et al. [29] CT 31 34 0.88 (0.75–1.05)
Hinson et al. [27] CT 10.6 10.2 1.05 (0.94–1.18)
Caspi et al. [25] Coronary angiographyc 10.3 12.1 0.79 (0.60–1.05)
Zealley et al. [24] Not specified 10.9 10.7 0.91 (0.82–1.00)
Wilhelm-Leen et al. [26] Not specified 5.1 5.6 0.93 (0.88–0.97)
a
Data denote AKI rates with iohexol and iodixanol, respectively.
b
Comparison is for patients with stage 3 CKD.
c
With percutaneous intervention.
AKI, acute kidney injury; CA-AKI, contrast-associated acute kidney injury; OR, odds ratio; RR, relative risk.
P = 0.38) [25]. In analyses that included propensity score conditions that augment the risk for CA-AKI are those
matching, AKI rates remained comparable in patients who that impair the capacity of the kidneys to effectively
were and were not treated with percutaneous intervention counteract the pathophysiological processes that lead to
(8.6% vs. 10.9%, P = 0.12). the development of CA-AKI. Pre-procedural kidney
While collectively these studies suggest that the adminis- impairment is the most potent patient-related risk factor
tration of intravascular contrast does not increase the risk for for CA-AKI, with lower levels of kidney function corre-
AKI, most are retrospective in design and have limited capac- sponding with higher degrees of risk [43]. In a study of
ity to adjust for the effects of unmeasured confounding. 378 hospitalized patients undergoing nonrenal angiogra-
Specifically, patients at higher baseline risk for CA-AKI were phy, D’Elia et al. demonstrated a markedly higher inci-
likely to have selectively undergone radiological tests without dence of CA-AKI among those defined as azotemic
intravascular contrast. Without understanding and account- (blood urea nitrogen 30 mg/dl and serum creatinine
ing for the factors that might influence provider decision 1.5 mg/dl) compared to nonazotemic patients (33% vs.
making regarding the use of contrast, the reported relation- 2%). Pre-existing azotemia was the only significant risk
ships are likely to be confounded by differences in baseline factor for CA-AKI in this study [22]. McCullough et al.
characteristics between those who did and those who did not documented a strong association of baseline kidney dis-
receive contrast. This is best illustrated by the fact that certain ease with risk of CA-AKI in a large population of patients
of these studies demonstrated lower rates of AKI among undergoing angiography [38]. The findings from a trial
patients who were administered intravascular contrast com- by Rudnick et al. that compared different contrast agents
pared to those who did not receive contrast [26]. As contrast suggest that diabetes, while not an independent risk fac-
media are not nephroprotective, this observation strongly tor for CA-AKI, substantially amplifies the risk in those
implies differential use of contrast based on clinicians’ per- patients with existing CKD [44]. Intravascular volume
ceptions of patients’ baseline risk for AKI. Accordingly, it is depletion, either absolute (e.g. GI losses) or effective (e.g.
inappropriate to conclude based on these studies that the risk heart failure, liver failure), also increases the risk for CA-
for AKI is independent of contrast administration in patients AKI [45, 46]. Similarly, the consumption of selective or
undergoing radiographic/angiographic procedures. nonselective nonsteroidal anti-inflammatory medica-
tions, which inhibit vasodilatory prostaglandins in the
kidney, may also increase the risk for CA-AKI [47]. Other
patient-related factors that have been linked with CA-
Risk Factors AKI include increasing age, hypertension, anemia, and
proteinuria; however, the independent associations of
Patient-related
these factors with CA-AKI remain unproven as each
Risk factors for CA-AKI are categorized as patient-related is strongly linked with the presence and severity of
and procedure-related (Table 11.2). Patient-related CKD [48–51].
Table 11.2 Association of CA-AKI with short-term mortality risk.
Study authors n CA-AKI definition OR 95% CI
Clec’h et al. [36] 143 ↑ Serum creatinine 25% or 0.3 mg/dl 3.48 (1.10–11.46)

Levy et al. [37] 357 ↑ Serum creatinine 25% to 2.0 mg/dl 5.5 (2.9–13.2)
Shema et al. [21] 1 111 ↑ Serum creatinine 50% or ↓ eGFR 25% 3.9 (1.2–12.0)
McCullough et al. [38] 1 826 ↑ Serum creatinine >25% 6.6 (3.3–12.9)
From et al. [39] 3 236 ↑ Serum creatinine 25% or 0.5 mg/dl 3.4 (2.6–4.4)
Rihal et al. [40] 7 586 ↑ Serum creatinine >0.5 mg/dl 10.8 (6.9–17.0)
Bartholomew et al. [41] 20 479 ↑ Serum creatinine 1.0 mg/dl 22 (16–31)

Weisbord et al. [42] 27 608 ↑ Serum creatinine 0.25–0.5 mg/dl 1.8 (1.4–2.5)
CA-AKI, contrast-associated acute kidney injury; CI, confidence interval; eGFR, estimated glomerular filtration rate; OR, odds ratio.
Procedure-related Associated Outcomes

Factors associated with the radiological procedure itself
also affect the risk for CA-AKI. The administration of Association with Short-term Adverse
larger volumes of radiocontrast increases the risk. Outcomes
Although the precise volume above which risk increases CA-AKI has been associated with an increased risk for
significantly not been definitively determined, various for- short-term mortality (Table 11.2) [36–42]. McCullough
mulas to calculate the upper threshold dose in milliliters et al. demonstrated that patients who developed CA-AKI
of contrast have been proposed. These include multiplying following angiography with percutaneous coronary
the body weight (in kg) by 5 and dividing by the patient’s intervention were more likely to die during the index
serum creatinine (mg/dl), or alternatively multiplying the hospitalization compared to patients who did not develop
patient’s eGFR by 2 [52–55]. The performance of sequen- CA-AKI (7.1% vs. 1.1%, I < 0.0001) [38]. Several other
tial radiocontrast procedures over a short period of time observational studies reported an increased incidence of
also increases the risk for CA-AKI. High-osmolal contrast short-term mortality risk associated with the development
media are associated with an increased risk compared of CA-AKI [36, 37, 39–42] (Table 11.2). Data from clinical
with low- and iso-osmolal contrast [44, 56]. More wide- trials also document this association. In a clinical trial of
spread use of low and iso-osmolal contrast likely explains, acetylcysteine, Marenzi et al. reported a higher incidence
at least in part, the trend toward lower overall rates of of in-hospital death among patients who developed CA-
CA-AKI [57]. AKI compared with patients without CA-AKI (26% vs.
The intra-arterial administration of contrast appears 1.4%, P < 0.001) [58]. Similarly, in a trial comparing IV
to confer a higher risk compared with IV administration, fluids, Maioli et al. demonstrated that patients who
although it remains unclear if the differential risk relates experienced CA-AKI had a higher incidence of in-hospital
to the route of contrast administration or underlying death than patients who did not develop CA-AKI (11.1% vs.
characteristics that differ across patients undergoing dif- 0.2%, P = 0.001) [59]. CA-AKI has also been associated
ferent types of procedures. For example, Weisbord et al. with increased duration of hospitalization, as noted in a
demonstrated lower rates of CA-AKI following IV study by Adolph et al., who found that the development of
administration of contrast in the setting of computed CA-AKI was associated with an average prolongation in
tomography (6.5%) as compared with the intra-arterial hospitalization of 2 days [60].
administration for coronary (8.5%) and noncoronary
angiography (13.2%) in patients with CKD [20]. However,
Association with Longer-term Adverse
differences in clinical characteristics such as diabetes
Outcomes
mellitus, heart failure, or severity of CKD, along with
differential utilization of preventive interventions such Several studies demonstrate that CA-AKI is associated with
as IV fluids may have accounted for the differing inci- increased longer-term risk of mortality and more rapid pro-
dent rates across procedure types. gression of underlying CKD [40, 59, 61–68] (Table 11.3).
Associated Outcome 149
Table 11.3 Association of CA-AKI with longer-term risk of death.
Authors n CA-AKI definition Follow-up (years) Adjusted HR 95% CI
Goldenberg et al. [61] 78 ↑ Serum creatinine 0.5 mg/dl or 25% 5 2.7 1.7–4.5

Solomon et al. [64] 294 ↑ Serum creatinine 0.3 mg/dl 1 3.2a 1.1–8.7
Harjai et al. [62] 985 ↑ Serum creatinine 0.5 mg/dl 2 2.6 1.5–4.4
Zlatanovic et al. [68] 1017 ↑ Serum creatinine 0.5 mg/dl or 25% 4.8 2.6 1.3–3.8
Roghi et al. [63] 2860 ↑ Serum creatinine 0.5 mg/dl 2 1.8 1.0–3.4
b b
Rihal et al. [40] 7075 ↑ Serum creatinine >0.5 mg/dl 0.5
Brown et al. [65] 7856 ↑ Serum creatinine 0.5 mg/dl 7.5 3.1 2.4–4.0
a
Denotes incident rate ratio of death, cerebrovascular accident, myocardial infarction, end-stage renal disease.
b
Six-month mortality 9.8% vs. 2.3%, p < 0.0001.
CA-AKI, contrast-associated acute kidney injury; CI, confidence interval; HR, hazard ratio.
Solomon and colleagues reported that the development of the treatment of acute myocardial infarction in 15 093
CA-AKI was associated with a more than threefold increased patients with CKD and 42 191 without CKD [73]. In analy-
risk of death, stroke, myocardial infarction, and/or end- ses accounting for the appropriateness of coronary angiog-
stage kidney disease at 1 year following angiography [64]. raphy, those with CKD were around half as likely to
Goldenberg et al. reported that patients with transient CA- undergo this procedure than patients without CKD (OR
AKI manifested a larger decrement in kidney function 0.47, 95% CI 0.40–0.52), and patients with CKD who did
2 years following angiography compared with patients who undergo coronary angiography were also less likely to pro-
did not experience CA-AKI (∆eGFR −20 ± 11 ml/ ceed to revascularization (55% vs. 62%, P < 0.0001). Concern
min/1.73 m2 vs. −6 ± 16 ml/min/1.73 m2, P = 0.02), a finding for the risk of CA-AKI was a hypothesized explanation for
consistent with other types of AKI [61]. James et al. demon- the underutilization of angiography and revascularization
strated more rapid deterioration of kidney function over in subjects with CKD.
3 months following angiography among patients who had Renalism has also been documented among patients
developed CA-AKI compared to those without this condi- with CKD presenting with ST elevation myocardial infarc-
tion (decrement in eGFR 0.8 ml/min/1.73 m2/yr vs. 0.2 ml/ tion (STEMI). Nauta et al. found that rates of percutaneous
min/1.73 m2/yr) [66]. coronary revascularization were lower in patients with
STEMI and stage 4 CKD compared to those with STEMI
but without CKD [72]. The findings of these studies docu-
Potential Clinical Implications of Data
menting renalism are particularly relevant considering
Associating CA-AKI with Adverse Outcomes
data from multiple studies demonstrating improved sur-
While consistent in demonstrating associations of CA- vival among patients with CKD who undergo coronary
AKI with serious, adverse short-and long-term outcomes, angiography and revascularization in the setting of acute
these and similar past studies have not established coronary syndrome [74]. A study by Wright et al. of 1786
whether the associations are causal. CA-AKI, defined by patients with CKD demonstrated that initial invasive coro-
small increments in serum creatinine, may simply be a nary care was associated with 35% and 50% relative reduc-
marker of a population of patients with more significant tions in in-hospital mortality compared with conservative
vascular disease and propensity for hemodynamic insta- medical therapy in patients with ST elevation myocardial
bility that places them at heightened risk for serious infarction and non-ST elevation myocardial infarction,
events. This fact is particularly pertinent given research respectively [75]. In a more recent study of patients with
documenting the underutilization of clinically indicated, non-ST elevation acute coronary syndrome, James et al.
potentially life-saving contrast-enhanced procedures in found that among those with stage 3 CKD, early perfor-
patients with CKD [69–72]. This practice is referred to as mance of coronary angiography was associated with
“renalism” and is likely driven by providers’ fear of pre- lower long-term mortality (risk ratio 0.66, 95% CI 0.52–
cipitating CA-AKI. 0.85) compared with conservative management, without
Chertow et al. compared the use of invasive coronary a corresponding increase in risk for the development
care, including coronary angiography, percutaneous coro- of end-stage kidney disease (risk ratio 0.88, 95% CI
nary intervention, and coronary artery bypass surgery, for 0.42–1.86) [76].
Collectively, these studies suggest that patients with were referred to as “low-osmolal” contrast. Compared with
CKD are less likely to undergo contrast-enhanced proce- high-osmolal contrast, the low-osmolal agents were associ-
dures than those without CKD and that this practice may ated with lower rates of CA-AKI. The Iohexol Cooperative
be resulting in an increase in serious adverse outcomes. It Study was a clinical trial of 1196 patients undergoing coro-
is hypothesized that the underutilization of contrast- nary angiography that demonstrated a lower incidence of
enhanced procedures in those with CKD is due to provider CA-AKI with low-osmolal iohexol compared with high-
fear of causing CA-AKI. The importance of these observa- osmolal diatrizoate (3.2% vs. 7.1%, P = 0.002) [44]. A meta-
tions is underscored by the absence of data establishing a analysis by Barrett and Carlisle that included results from
causal link between CA-AKI and adverse downstream this trial documented a lower risk of CA-AKI with low-
events. Efforts to examine whether CA-AKI causes adverse osmolal contrast compared to high-osmolal contrast (OR
outcomes or simply serves as a marker of patients more 0.61, 95% CI 0.48–0.77) [56]. Among patients with impaired
likely to experience such events are essential. Until research kidney function, low-osmolal contrast was associated with
answers this question, an important focus is ensuring that an even lower risk of CA-AKI compared with high-osmolal
patients at risk for CA-AKI undergo indicated procedures, agents (OR 0.5, 95% CI 0.36–0.68).
albeit with the implementation of evidence-based care to The development of a third-generation contrast agent,
prevent this iatrogenic condition. iodixanol, a nonionic dimer that was iso-osmolal to
plasma, led to comparisons with low-osmolal media.
Initial trials suggested lower rates of CA-AKI with iodix-
Prevention anol compared with certain low-osmolal agents (i.e.
iohexol, ioxaglate), while other studies demonstrated
Overview similar risk for CA-AKI [77–84]. Clinical practice guide-
lines from the American College of Cardiology/American
CA-AKI is potentially preventable as the principal risk fac- Heart Association cite a lack of sound data to support the
tors are well known and the timing of contrast administra- preferential use of iodixanol compared with low-osmolal
tion is frequently known in advance, which facilitates the contrast agents [85]. Similarly, the European Society of
timely implementation of preventive care in patients most Urogenital Radiology supports use of iso-or low-osmolal
likely to benefit. Identifying patients at heightened risk for contrast in patients at elevated risk of CA-AKI [86].
CA-AKI allows for the consideration of alternative imaging
procedures that do not necessitate intravascular contrast,
but that provide comparable diagnostic information. For Renal Replacement Therapies
at-risk patients who require procedures with intravascular Iodinated contrast media remain largely in the extracellular
contrast, evidence-based prevention should be imple- space, have low protein binding, and are highly water solu-
mented to reduce the risk for kidney damage. ble. Hence, they can be efficiently removed from the circula-
Research on preventive strategies for CA-AKI has tion by renal replacement therapies. Lee et al. randomized
focused on comparing different contrast agents, using 82 patients with advanced CKD undergoing coronary angi-
renal replacement therapy to filter iodinated contrast from ography to receive prophylactic hemodialysis or not, in addi-
the vasculature prior to glomerular filtration, administer- tion to IV saline and demonstrated a smaller decrement in
ing pharmacologic agents to oppose the nephrotoxic creatinine clearance and lower likelihood of requiring
actions of contrast, and expanding the intravascular space chronic hemodialysis in those who received prophylactic
and enhancing urine volume with IV fluids to mitigate the dialysis [87]. Unlike this study, most other trials of prophy-
adverse hemodynamic and tubular effects of contrast in lactic hemodialysis either failed to demonstrate a benefit or
the kidney. reported a higher risk for CA-AKI associated with prophy-
lactic hemodialysis [88–93]. Continuous renal replacement
therapy (CRRT) also removes iodinated contrast from the
Selection of Contrast Agent
vascular space, and there have been two randomized clinical
Iodinated contrast media are commonly categorized based trials of continuous veno-venous hemofiltration that have
on their osmolality. First-generation contrast agents were suggested benefit [94, 95]. Both of these studies utilized
ionic derivatives of tri-iodobenzoic acid with osmolalities change in serum creatinine relative to baseline as their pri-
of approximately 1500 to over 2000 mOsm/kg; accordingly, mary outcome, an endpoint confounded by the lowering of
they were termed “high-osmolal” contrast media. Second- serum creatinine by the intervention itself. As such, prophy-
generation agents were nonionic monomers or ionic lactic renal replacement therapy as either intermittent
dimers with lower osmolalities (600–1000 mOsm/kg) and hemodialysis or CRRT is not indicated to prevent CA-AKI.
Preventio 151
Pharmacological Agents A study by Kurnik et al. compared atrial natriuretic

peptide (ANP), which has renal vasodilatory properties,
A multitude of clinical trials have investigated various
with placebo in 247 patients with CKD and found no ben-
pharmacological agents for the prevention of CA-AKI.
efit for the prevention of CA-AKI in the ANP group [110].
Disappointingly, much of this literature has been suscepti-
Conversely, in a trial of 254 patients with CKD undergo-
ble to type 1 error and bias, and no prophylactic pharmaco-
ing angiography, Morikawa et al. reported a lower inci-
logical treatments have been definitively demonstrated to
dence of CA-AKI among patients who received ANP and
be effective (Table 11.4).
IV fluids compared to patients who received just IV flu-
By blocking active sodium transport in the ascending limb
ids [111]. Notwithstanding the discordant findings on the
of the loop of Henle and reducing oxygen utilization, loop
benefit of ANP, this treatment has potentially deleterious
diuretics have been the subject of several studies. A clinical
side effects, such as reductions in systolic blood pressure,
trial by Solomon et al. that compared IV hypotonic saline
so its use for the prevention of CA-AKI is currently not
alone to IV hypotonic saline with either furosemide or man-
recommended.
nitol reported a lower incidence of CA-AKI among patients
Adenosine has been implicated in the pathogenesis of
who received peri-procedural IV hypotonic saline alone
CA-AKI. Accordingly, studies have evaluated antagonists
compared with IV hypotonic saline with furosemide (11% vs.
of adenosine, including theophylline, for the prevention of
40%, P = <0.02) [96]. The lack of benefit with furosemide in
this condition [112, 113]. In a trial by Early et al. that
this study mirrored the findings of an earlier study that
enrolled 80 patients, theophylline was not found to be
found a higher risk for CA-AKI with the use of this agent [97].
effective for the prevention of CA-AKI [112]. Conversely, a
A subsequent study by Majumdar et al. that attempted to
study by Huber et al. of 100 patients with CKD undergoing
ensure preservation of euvolemia despite diuretic adminis-
coronary angiography demonstrated that pre-procedure
tration again found increased risk of CA-AKI associated with
theophylline was associated with a lower incidence of CA-
administration of furosemide and mannitol [98].
AKI compared with placebo. Consistent with the findings
On the basis of studies demonstrating renal vasodilata-
of clinical trials, meta-analyses of theophylline have
tion when used in low doses, dopamine has been examined
reported discordant findings [114–116]. Recognizing the
as a potential preventive therapy [99–101]. Initial studies
potential adverse side effects of theophylline, particularly
suggested a beneficial effect on reducing CA-AKI, yet sub-
in patients with underlying cardiovascular disease, its use
sequent clinical trials with larger numbers of patients
to prevent CA-AKI is not current recommended.
found no benefit [99, 101, 102]. Weisberg et al. compared
A multitude of clinical trials have investigated statins for
dopamine with IV hypotonic saline with IV hypotonic
the prevention of CA-AKI, many of which suggested a ben-
saline and found no difference in the incidence of CA-AKI
efit, particularly with high intensity doses. The results of
overall, yet reported that diabetics who received dopamine
subsequent meta-analyses reflected the findings of the
experienced higher rates of CA-AKI [101]. Fenoldopam is a
clinical trials upon which their results were based [117,
selective dopamine receptor agonist that was found in
118]. However, a number of these studies enrolled lower
early, mostly small nonrandomized studies, to prevent the
risk patients and many used small increments in serum
development of CA-AKI [103–108]. As with dopamine,
creatinine as the primary outcome, which may be problem-
larger, methodologically stronger studies failed to demon-
atic in light of the observation that the administration of
strate a benefit of fenoldopam [109].
Table 11.4 Pharmacological interventions investigated for contrast-associated

acute kidney injury prevention.
Ineffective Effectiveness not determined
Mannitola Atrial natriuretic peptide

Loop diureticsa Theophylline/aminophylline
Dopaminea Statins
Fenoldopama Prostaglandin analogs
Calcium channel blockers Allopurinol
N-Acetylcysteine Acetazolamide
Sodium bicarbonate
a
Potentially deleterious.
rosuvastatin may result in an increase in eGFR [119]. Intravenous Crystalloid

Consequently, there is insufficient evidence at present to
Administration of intravenous crystalloid is believed to
conclude that statins are protective against CA-AKI.
reduce the risk of CA-AKI by counteracting contrast-
N-acetylcysteine (NAC) is an antioxidant with vasodila-
induced vasoconstriction in the kidneys and by reducing
tory effects that has been investigated in numerous trials
the concentration and viscosity of contrast media in the
over nearly 20 years as a preventive intervention for CA-
tubular lumen [156]. Initial observational studies suggested
AKI. The initial clinical trial of NAC by Tepel et al. rand-
a benefit of IV fluid, but were limited by the lack of
omized 83 patients undergoing computed tomography
randomization. A clinical trial by Trivedi et al. randomized
to receive oral NAC or placebo and demonstrated a
patients undergoing coronary angiography to receive IV
lower incidence of CA-AKI with NAC (2% vs. 21%,
isotonic saline for 12 hours prior to and 12 hours following
P = 0.01) [120]. Since the publication of this study, dozens
the procedure or unrestricted fluid by mouth [157]. The
of clinical trials have examined NAC for the prevention of
trial was stopped early after demonstrating a substantially
CA-AKI, but the literature has again been susceptible to
lower incidence of CA-AKI among patients who received
type 1 error and variable study quality, yielding divergent
IV saline compared with oral fluid (3.7% vs. 34.6%,
results [58, 120–144]. Numerous meta-analyses sought to
P = 0.005). Mueller et al. randomized over 1600 patients
reconcile the incongruent findings of the clinical trials,
undergoing coronary angiography to receive peri-
yet similarly reported discordant results [116, 145–154].
procedural IV 0.45% saline or 0.9% saline [158]. The
The two largest studies, the Acetylcysteine for Contrast
administration of isotonic saline was associated with a
Nephropathy (ACT) trial and the Prevention of Serious
lower incidence of CA-AKI than hypotonic saline (0.7% vs.
Adverse Events Following Angiography (PRESERVE)
2.0%, P = 0.04), although the study population was at
trial, have provided more definitive evidence that NAC is
generally low risk for CA-AKI as evidence by the low event
not efficacious for the prevention of CA-AKI. The ACT
rates in both study arms.
trial randomized more than 2300 patients undergoing
Notwithstanding the limitations of these and other
angiography to receive NAC or placebo and found
studies, IV isotonic saline was considered the standard
no reduction in the risk for CA-AKI with NAC.
treatment to prevent CA-AKI. It should be noted that a
Notwithstanding this finding, this study was limited by
recent clinical trial questioned the need for IV fluid to
the inclusion of many patients at low risk for CA-AKI.
Using a 2 × 2 factorial design, the PRESERVE trial rand- prevent CA-AKI. The A MAstricht Contrast-Associated
omized 2495 patients with CKD undergoing angiography Nephropathy Guideline (AMACING) trial randomized
to receive 1200 mg of oral NAC prior to and for 4 days fol- 660 patients who were undergoing procedures with IV or
lowing the procedure, and 2498 patients with CKD to intra-arterial contrast to receive IV isotonic saline or no
receive matching oral placebo [155]. The results demon- IV fluids. Designed as a noninferiority trial, the investiga-
strated no difference in the incidence of the primary trial tors documented that a strategy of no IV fluids was not
endpoint comprised of 90-day death, need for dialysis, or inferior to a strategy of IV saline for the prevention of CA-
persistent impairment in kidney function between AKI (2.6% vs. 2.7%, absolute difference −0.10%, 95% CI
patients randomized to receive NAC and those rand- −2.25 to 2.06) [159]. Despite this surprising result, key
omized to placebo (OR 1.02, 95% CI 0.78–1.33). Similarly, methodological issues, including under-enrollment (only
there was no difference in the incidence of CA-AKI, 660 of 1300 planned patients were randomized), low rate
which was a secondary trial endpoint (OR 1.06, 95% CI of enrollment of patients undergoing procedures with
0.87–1.28). These two large trials provide sufficiently intra-arterial contrast (48%) and interventional proce-
sound evidence that NAC has no role for the prevention dures (16%), and inclusion of a relatively large population
of CA-AKI or associated serious, adverse outcomes. of patients with less advanced CKD (i.e. eGFR 45–60 ml/
Clinical trials of other agents including calcium channel min/1.73 m2) substantially diminished the validity of the
blockers, prostaglandins, acetazolamide, and mannitol results.
have been inconclusive, yet like the clinical trials of many Over the past 15 years, investigators have focused on
pharmacological therapies were limited by small numbers comparing IV isotonic sodium bicarbonate with IV isotonic
of patients, reliance on a surrogate primary study endpoint sodium chloride based on the hypothesis that enhanced
defined by small increments in serum creatinine, and, in urinary alkalinization with HCO3− decreases the genera-
some cases, inclusion of patients at very low risk for kidney tion of reactive oxygen species and may attenuate tubular
injury. Consequently, current recommendations for the epithelial cell injury compared with Cl−. Merten et al.
prevention of CA-AKI do not incorporate the administra- examined this hypothesis in a clinical trial that randomized
tion of any of these pharmacological therapies. 119 patients and documented that IV isotonic sodium
Preventio 153
bicarbonate was associated with a lower incidence of CA- to the standard treatment group, who received IV isotonic
AKI compared with IV isotonic saline (1.6% vs. 13.6%, saline at 1.5 ml/kg/h during and for 4 hours after the
P = 0.02) [160]. A proliferation of clinical trials followed procedure, the experimental group received IV isotonic
with discordant results [59, 60, 161–168]. Multiple system- saline based on the measured left ventricular end-diastolic
atic reviews and meta-analyses reported results that were pressure (i.e. 5, 3, or 1.5 ml/kg/h for left ventricular
as disparate as the component clinical trials [162, 168–177]. end-diastolic pressure measurements of <13, 13–18, and
More recently, Solomon et al. compared IV isotonic sodium >18 mmHg, respectively). CA-AKI occurred less com-
bicarbonate to IV isotonic saline in 391 patients with eGFR monly in the left ventricular end-diastolic pressure-guided
<45 ml/min/1.73 m2 undergoing angiography and found group (6.7% vs. 16.3%, relative risk 0.41, 95% CI 0.22–0.79,
no differences in the incidence of a primary composite out- P = 0.005) compared with the standard IV saline group.
come of 6-month death, need for dialysis, or sustained 20% The overall rate of pulmonary compromise related to IV
decrease in eGFR (14.9% vs. 16.3%, P = 0.78) [178]. This fluid administration was very low [181]. A similar study by
study also documented no difference in the incidence of Qian et al. demonstrated a lower incidence of CA-AKI in
CA-AKI between the study arms. The largest trial compar- patients who received IV saline administered based on cen-
ing the efficacy of sodium bicarbonate to saline was the tral venous pressure measurements compared with patients
PRESERVE trial, which randomized 4993 high risk patients who received standard IV saline treatment [182]. The over-
to receive IV isotonic sodium bicarbonate (n = 2511) or IV all rate of acute heart failure during IV fluid administration
isotonic sodium chloride (n = 2482) prior to, during, and was very low and comparable across patient groups (3.8%
following angiography [155]. Compared with IV saline, IV vs. 3.0%, P = 0.5).
sodium bicarbonate use was not associated with a decreased
risk for 90-day death, need for dialysis, or persistent impair-
Limitations of Clinical Trials on Preventive
ment in kidney function (OR 0.93, 95% CI 0.72–1.22) or in
Interventions
the risk for CA-AKI (OR 1.16, 95% CI 0.96–1.41) [155]. A
subsequent subgroup analysis of the PRESERVE trial Many of the clinical trials investigating preventive inter-
found no benefit of IV sodium bicarbonate compared with ventions for CA-AKI had important methodological limi-
IV sodium chloride among study participants who under- tations. Most importantly, the vast majority of trials used
went percutaneous coronary intervention [179]. small increments in serum creatinine as their primary
A series of small studies examined the efficacy of induc- outcome, based on observational data associating such
ing a very high urine flow (>300 ml/h) with saline and biochemical perturbations with serious, adverse events.
furosemide using a device that permits accurate matching That such correlations exist is not in dispute, but it is
of urine flow and IV saline administration for the preven- essential to recognize that for these serum creatinine
tion of CA-AKI. A recent meta-analysis by Shah et al. that changes to truly be considered a “surrogate” for serious
comprised three clinical trials and 586 patients demon- adverse clinical events, they “must track with the
strated a reduction in CA-AKI, major adverse clinical frequency of the end point (i.e. serious clinical events)
events, and need for renal replacement therapy with this both as an epidemiologic marker and as a therapeutic
device compared with standard preventive care comprised responder” [183]. Definitive evidence that the changes in
principally of conventional isotonic IV fluid administra- serum creatinine that define CA-AKI track with changes
tion [180]. However, as pointed out by the authors, these in subsequent clinical outcomes is lacking. Thus, although
findings were based on small, underpowered studies and CA-AKI remains defined by small changes in serum
require confirmation in larger, adequately powered clinical creatinine, definitive clinical trials require the use of
trials. more meaningful outcomes.
While heart failure is recognized as a risk factor for The use of small changes in serum creatinine as the sur-
CA-AKI, there are data that the administration of IV fluids rogate primary endpoint has resulted in multiple clinical
to patients with heart failure is both effective and safe. The trials that are susceptible to type 1 error with inflated event
Prevention of Contrast Renal Injury with Different rates, small sample sizes, and inadequate statistical power
Hydration Strategies (POSEIDON) trial randomized 396 to discern effects of the interventions on key patient-
patients who were undergoing coronary angiography and centered outcomes. Clinical trials focused upon far more
compared a standard regimen of IV isotonic saline (3 ml/kg serious patient-centered outcomes (e.g. death, need for
for 1 hour prior to and 1.5 ml/kg/h during and for 4 hours dialysis) as the primary study outcome require markedly
following the procedure) with a strategy of IV isotonic larger samples sizes, as was demonstrated with the
saline based on the left ventricular end-diastolic pressure PRESERVE trial [155]. As such, all trials of treatments to
value at the beginning of angiography [181]. As compared prevent CA-AKI need to use clinically significant events as
their primary outcome until there is sound evidence that contrast administration. While a meta-analysis by Wu
prevention of small changes in serum creatinine also et al. that comprised 14 studies demonstrated a higher risk
diminishes these “hard” outcomes. for CA-AKI among patients receiving angiotensin con-
verting enzyme inhibitors or angiotensin receptor block-
ers at the time of angiography compared with patients not
Current Recommendations for Prevention
receiving these medications, that authors acknowledged
Identifying patients at high risk for CA-AKI and determin- this finding was based on low-level evidence from small
ing if alternative imaging procedures that can avoid intra- observational studies [184]. Furthermore, the initiation or
vascular iodinated contrast with comparable diagnostic discontinuation of these agents at the time of contrast
yield are available should be the initial step in efforts to administration may confound the interpretation of small
prevent this condition. For most patients who require pro- changes in serum creatinine. Assessing for the develop-
cedures with the use of intravascular contrast, all modifi- ment of CA-AKI by monitoring serum creatinine
able risk factors should be addressed. Nephrotoxic 48–96 hours following contrast administration is impor-
medications such as nonsteroidal anti-inflammatory med- tant to identify the development of this condition and to
ications should be discontinued prior to contrast adminis- provide supportive care to attenuate the risk for further
tration and held until CA-AKI has been ruled out. The renal parenchymal damage.
lowest required dose of either low or iso-osmolal contrast
should be used. Isotonic IV sodium chloride should be
administered prior to, during, and after the procedure, Conclusion
particularly for patients with lower eGFR values and those
undergoing angiography [155]. While IV isotonic sodium AKI remains a common iatrogenic event following proce-
bicarbonate is an acceptable alternative, it offers no advan- dures that involve the intravascular administration of iodi-
tage over IV isotonic saline and poses a small risk for com- nated contrast. The pathophysiology of and risk factors for
pounding errors as there is currently no commercially CA-AKI have been demonstrated in past studies, as have
available “off-the-shelf” form. For hospitalized patients, the associations of this condition with serious, adverse
isotonic IV saline can be administered at a rate of 1 ml/ short- and long-term outcomes. Such associations, which
kg/h for 12 hours preceding, during, and for 12 hours fol- have not been confirmed to be causal, have likely led to the
lowing the procedure if the timing of the procedure per- underutilization of clinically indicated diagnostic and ther-
mits. An alternative regimen, particularly for outpatients apeutic contrast-enhanced procedures in patients with
and those undergoing more urgent procedures, includes CKD. Patients with clear clinical indications for procedures
IV isotonic saline at a rate of 3 ml/kg/h over 1–2 hours that require intravascular contrast should undergo such
prior to and 6 ml/kg infused over 2–6 hours following con- procedures, albeit with the implementation of evidence-
trast administration. IV isotonic saline should be adminis- based preventive care. The cornerstone of prevention pres-
tered to patients with nondecompensated heart failure ently is the peri-procedural IV administration of isotonic
with careful monitoring of patients’ respiratory status. crystalloid, preferably sodium chloride. Well-designed,
There is no current role for the use of pharmacological adequately powered studies that utilize clinically relevant
agents for the prevention of CA-AKI. Similarly, studies to primary outcomes and that enroll high-risk patients are
date do not support the discontinuation of diuretics or needed to expand our understanding of the clinical impli-
blockers of the renin-angiotensin aldosterone axis prior to cations and prevention of this iatrogenic condition.
References
Alexander, R.D., Berkes, S.L., and Abuelo, G. (1978).

1 4 Nicot, G.S., Merle, L.J., Charmes, J.P. et al. (1984). Transient
Contrast media-induced oliguric renal failure. Arch. Intern. glomerular proteinuria, enzymuria, and nephrotoxic reaction
Med. 138: 381–384. induced by radiocontrast media. JAMA 252: 2432–2434.
2 Heyman, S.N., Brezis, M., Epstein, F.H. et al. (1991). Early 5 Haller, C. and Hizoh, I. (2004). The cytotoxicity of
renal medullary hypoxic injury from radiocontrast and iodinated radiocontrast agents on renal cells in vitro.
indomethacin. Kidney Int. 40: 632–642. Investig. Radiol. 39: 149–154.
3 Heyman, S.N., Reichman, J., and Brezis, M. (1999). 6 Hizoh, I. and Haller, C. (2002). Radiocontrast-induced
Pathophysiology of radiocontrast nephropathy: a role for renal tubular cell apoptosis: hypertonic versus oxidative
medullary hypoxia. Investig. Radiol. 34: 685–691. stress. Investig. Radiol. 37: 428–434.
Reference 155
7 Hizoh, I., Strater, J., Schick, C.S. et al. (1998). 21 Shema, L., Ore, L., Geron, R., and Kristal, B. (2009).
Radiocontrast-induced DNA fragmentation of renal Contrast-induced nephropathy among Israeli hospitalized
tubular cells in vitro: role of hypertonicity. Nephrol. Dial. patients: incidence, risk factors, length of stay and
Transplant. 13: 911–918. mortality. Isr. Med. Assoc. J. 11: 460–464.
8 Hardiek, K., Katholi, R.E., Ramkumar, V., and Deitrick, 22 D’Elia, J.A., Gleason, R.E., Alday, M. et al. (1982).
C. (2001). Proximal tubule cell response to radiographic Nephrotoxicity from angiographic contrast material. A
contrast media. Am. J. Physiol. Renal Physiol. 280: prospective study. Am. J. Med. 72: 719–725.
F61–F70. 23 Bruce, R.J., Djamali, A., Shinki, K. et al. (2009).
9 Heyman, S.N., Rosen, S., Khamaisi, M. et al. (2010). Background fluctuation of kidney function versus
Reactive oxygen species and the pathogenesis of contrast-induced nephrotoxicity. Am. J. Roentgenol. 192:
radiocontrast-induced nephropathy. Investig. Radiol. 45: 711–718.
188–195. 24 Zealley, I., Wang, H., Donnan, P.T., and Bell, S. (2018).
10 Moreau, J.F., Droz, D., Noel, L.H. et al. (1980). Tubular Exposure to contrast media in the perioperative period
nephrotoxicity of water-soluble iodinated contrast media. confers no additional risk of acute kidney injury in
Investig. Radiol. 15: S54–S60. surgical patients. Nephrol. Dial. Transplant. 33:
11 Humes, H.D., Hunt, D.A., and White, M.D. (1987). Direct 1751–1756.
toxic effect of the radiocontrast agent diatrizoate on renal 25 Caspi, O., Habib, M., Cohen, Y. et al. (2017). Acute kidney
proximal tubule cells. Am. J. Phys. 252: F246–F255. injury after primary angioplasty: is contrast-induced
12 Bakris, G.L., Lass, N., Gaber, A.O. et al. (1990). nephropathy the culprit? J. Am. Heart Assoc. 6: 1–10.
Radiocontrast medium-induced declines in renal 26 Wilhelm-Leen, E., Montez-Rath, M.E., and Chertow, G.
function: a role for oxygen free radicals. Am. J. Phys. 258: (2017). Estimating the risk of radiocontrast-associated
F115–F120. nephropathy. J. Am. Soc. Nephrol. 28: 653–659.
13 Bakris, G.L., Gaber, A.O., and Jones, J.D. (1990). Oxygen 27 Hinson, J.S., Ehmann, M.R., Fine, D.M. et al. (2017). Risk
free radical involvement in urinary Tamm-Horsfall of acute kidney injury after intravenous contrast media
protein excretion after intrarenal injection of contrast administration. Ann. Emerg. Med. 69: 577–586. e4.
medium. Radiology 175: 57–60. 28 McDonald, J.S., McDonald, R.J., Lieske, J.C. et al. (2015).
14 Yoshioka, T., Fogo, A., and Beckman, J.K. (1992). Risk of acute kidney injury, dialysis, and mortality in
Reduced activity of antioxidant enzymes underlies patients with chronic kidney disease after intravenous
contrast media-induced renal injury in volume depletion. contrast material exposure. Mayo Clin. Proc. 90:
Kidney Int. 41: 1008–1015. 1046–1053.
15 Erley, C.M., Heyne, N., Burgert, K. et al. (1997). 29 McDonald, J.S., McDonald, R.J., Williamson, E.E. et al.
Prevention of radiocontrast-induced nephropathy by (2017). Post-contrast acute kidney injury in intensive care
adenosine antagonists in rats with chronic nitric oxide unit patients: a propensity score-adjusted study. Intensive
deficiency. J. Am. Soc. Nephrol. 8: 1125–1132. Care Med. 43: 774–784.
16 Schnackenberg, C.G. (2002). Physiological and 30 McDonald, R.J., McDonald, J.S., Carter, R.E. et al. (2014).
pathophysiological roles of oxygen radicals in the renal Intravenous contrast material exposure is not an
microvasculature. Am. J. Physiol. Regul. Integr. Comp. independent risk factor for dialysis or mortality.
Physiol. 282: R335–R342. Radiology 273: 714–725.
17 Szabo, G., Bahrle, S., Stumpf, N. et al. (2002). Poly(ADP- 31 Ehrmann, S., Badin, J., Savath, L. et al. (2013). Acute
Ribose) polymerase inhibition reduces reperfusion injury kidney injury in the critically ill: is iodinated contrast
after heart transplantation. Circ. Res. 90: 100–106. medium really harmful? Crit. Care Med. 41: 1017–1026.
18 Kidney Disease: Improving Global Outcomes (KDIGO) 32 McDonald, J.S., McDonald, R.J., Carter, R.E. et al. (2014).
Acute Kidney Injury Work Group (2012). KDIGO clinical Risk of intravenous contrast material-mediated acute
practice guideline for acute kidney injury. Kidney Int. kidney injury: a propensity score-matched study stratified
Suppl. 2: 1–126. by baseline-estimated glomerular filtration rate.
19 Palevsky, P.M., Liu, K.D., Brophy, P.D. et al. (2013). Radiology 271: 65–73.
KDOQI US commentary on the 2012 KDIGO clinical 33 McDonald, R.J., McDonald, J.S., Bida, J.P. et al. (2013).
practice guideline for acute kidney injury. Am. J. Kidney Intravenous contrast material-induced nephropathy:
Dis. 61: 649–672. causal or coincident phenomenon? Radiology 267:
20 Weisbord, S.D., Mor, M.K., Resnick, A.L. et al. (2008). 106–118.
Prevention, incidence, and outcomes of contrast-induced 34 McDonald, R.J., McDonald, J.S., Newhouse, J.H., and
acute kidney injury. Arch. Intern. Med. 168: 1325–1332. Davenport, M.S. (2015). Controversies in contrast
material-induced acute kidney injury: closing in on the 49 Mehran, R., Aymong, E.D., Nikolsky, E. et al. (2004).
truth? Radiology 277: 627–632. A simple risk score for prediction of contrast-induced
35 McDonald, J.S., McDonald, R.J., Comin, J. et al. (2013). nephropathy after percutaneous coronary intervention:
Frequency of acute kidney injury following intravenous development and initial validation. J. Am. Coll. Cardiol.
contrast medium administration: a systematic review and 44: 1393–1399.
meta-analysis. Radiology 267: 119–128. 50 Hsu, R.K. and Hsu, C.Y. (2011). Proteinuria and
36 Clec’h, C., Razafimandimby, D., Laouisset, M. et al. reduced glomerular filtration rate as risk factors for
(2013). Incidence and outcome of contrast-associated acute kidney injury. Curr. Opin. Nephrol. Hypertens.
acute kidney injury in a mixed medical-surgical ICU 20: 211–217.
population: a retrospective study. BMC Nephrol. 14: 31. 51 He, F., Zhang, J., Lu, Z.Q. et al. (2012). Risk factors and
37 Levy, E.M., Viscoli, C.M., and Horwitz, R.I. (1996). The outcomes of acute kidney injury after intracoronary stent
effect of acute renal failure on mortality. A cohort implantation. World J Emerg Med 3: 197–201.
analysis. JAMA 275: 1489–1494. 52 Cigarroa, R.G., Lange, R.A., Williams, R.H., and Hillis,
38 McCullough, P.A., Wolyn, R., Rocher, L.L. et al. (1997). L.D. (1989). Dosing of contrast material to prevent
Acute renal failure after coronary intervention: incidence, contrast nephropathy in patients with renal disease. Am.
risk factors, and relationship to mortality. Am. J. Med. J. Med. 86: 649–652.
103: 368–375. 53 Brown, J.R., Robb, J.F., Block, C.A. et al. (2010). Does
39 From, A.M., Bartholmai, B.J., Williams, A.W. et al. (2008). safe dosing of iodinated contrast prevent contrast-
Mortality associated with nephropathy after radiographic induced acute kidney injury? Circ. Cardiovasc. Interv.
contrast exposure. Mayo Clin. Proc. 83: 1095–1100. 3: 346–350.
40 Rihal, C.S., Textor, S.C., Grill, D.E. et al. (2002). Incidence 54 Ando, G., Cortese, B., Frigoli, E. et al. (2015). Acute
and prognostic importance of acute renal failure after kidney injury after percutaneous coronary intervention:
percutaneous coronary intervention. Circulation 105: Rationale of the AKI-MATRIX (acute kidney injury-
2259–2264. minimizing adverse hemorrhagic events by TRansradial
41 Bartholomew, B.A., Harjai, K.J., Dukkipati, S. et al. access site and systemic implementation of angioX)
(2004). Impact of nephropathy after percutaneous sub-study. Catheter. Cardiovasc. Interv. 86: 950–957.
coronary intervention and a method for risk stratification. 55 Gurm, H.S., Dixon, S.R., Smith, D.E. et al. (2011). Renal
Am. J. Cardiol. 93: 1515–1519. function-based contrast dosing to define safe limits of
42 Weisbord, S.D., Chen, H., Stone, R.A. et al. (2006). radiographic contrast media in patients undergoing
Associations of increases in serum creatinine with percutaneous coronary interventions. J. Am. Coll. Cardiol.
mortality and length of hospital stay after coronary 58: 907–914.
angiography. J. Am. Soc. Nephrol. 17: 2871–2877. 56 Barrett, B.J. and Carlisle, E.J. (1993). Metaanalysis of the
43 McCullough, P.A., Adam, A., Becker, C.R. et al. (2006). relative nephrotoxicity of high-and low-osmolality
Risk prediction of contrast-induced nephropathy. Am. J. iodinated contrast media. Radiology 188: 171–178.
Cardiol. 98: 27K–36K. 57 Amin, A.P., Salisbury, A.C., McCullough, P.A. et al.
44 Rudnick, M.R., Goldfarb, S., Wexler, L. et al. (1995). (2012). Trends in the incidence of acute kidney injury in
Nephrotoxicity of ionic and nonionic contrast media in patients hospitalized with acute myocardial infarction.
1196 patients: a randomized trial. The iohexol cooperative Arch. Intern. Med. 172: 246–253.
study. Kidney Int. 47: 254–261. 58 Marenzi, G., Assanelli, E., Marana, I. et al. (2006).
45 Taliercio, C.P., Vlietstra, R.E., Fisher, L.D., and Burnett, N-acetylcysteine and contrast-induced nephropathy in
J.C. (1986). Risks for renal dysfunction with cardiac primary angioplasty. N. Engl. J. Med. 354: 2773–2782.
angiography. Ann. Intern. Med. 104: 501–504. 59 Maioli, M., Toso, A., Leoncini, M. et al. (2008). Sodium
46 Gomes, A.S., Baker, J.D., Martin-Paredero, V. et al. (1985). bicarbonate versus saline for the prevention of contrast-
Acute renal dysfunction after major arteriography. AJR induced nephropathy in patients with renal dysfunction
Am. J. Roentgenol. 145: 1249–1253. undergoing coronary angiography or intervention. J. Am.
47 Ahmad, S.R., Kortepeter, C., Brinker, A. et al. (2002). Coll. Cardiol. 52: 599–604.
Renal failure associated with the use of celecoxib and 60 Adolph, E., Holdt-Lehmann, B., Chatterjee, T. et al.
rofecoxib. Drug Saf. 25: 537–544. (2008). Renal Insufficiency Following Radiocontrast
48 Nikolsky, E., Mehran, R., Lasic, Z. et al. (2005). Low Exposure Trial (REINFORCE): a randomized comparison
hematocrit predicts contrast-induced nephropathy after of sodium bicarbonate versus sodium chloride hydration
percutaneous coronary interventions. Kidney Int. 67: for the prevention of contrast-induced nephropathy.
706–713. Coron. Artery Dis. 19: 413–419.
Reference 157
1 Goldenberg, I., Chonchol, M., and Guetta, V. (2009).

6 in patients with chronic kidney disease: evolution from
Reversible acute kidney injury following contrast the prethrombolysis to the percutaneous coronary
exposure and the risk of long-term mortality. Am. J. intervention era. Kidney Int. 84: 353–358.
Nephrol. 29: 136–144. 73 Chertow, G.M., Normand, S.L., and McNeil, B.J.
62 Harjai, K.J., Raizada, A., Shenoy, C. et al. (2008). A (2004). “Renalism”: inappropriately low rates of
comparison of contemporary definitions of contrast coronary angiography in elderly individuals with
nephropathy in patients undergoing percutaneous renal insufficiency. J. Am. Soc. Nephrol.
coronary intervention and a proposal for a novel 15: 2462–2468.
nephropathy grading system. Am. J. Cardiol. 101: 74 Shaw, C., Nitsch, D., Lee, J. et al. (2016). Impact of an
812–819. early invasive strategy versus conservative strategy for
63 Roghi, A., Savonitto, S., Cavallini, C. et al. (2008). Impact unstable angina and non-st elevation acute coronary
of acute renal failure following percutaneous coronary syndrome in patients with chronic kidney disease: a
intervention on long-term mortality. J. Cardiovasc. Med. systematic review. PLoS One 11: e0153478.
(Hagerstown) 9: 375–381. 75 Wright, R.S., Reeder, G.S., Herzog, C.A. et al. (2002).
64 Solomon, R.J., Mehran, R., Natarajan, M.K. et al. (2009). Acute myocardial infarction and renal dysfunction: a
Contrast-induced nephropathy and long-term adverse high-risk combination. Ann. Intern. Med. 137: 563–570.
events: cause and effect? Clin. J. Am. Soc. Nephrol. 4: 76 James, M.T., Tonelli, M., Ghali, W.A. et al. (2013). Renal
1162–1169. outcomes associated with invasive versus conservative
65 Brown, J.R., Malenka, D.J., DeVries, J.T. et al. (2008). management of acute coronary syndrome: propensity
Transient and persistent renal dysfunction are predictors matched cohort study. BMJ 347: f4151.
of survival after percutaneous coronary intervention: 77 Aspelin, P., Aubry, P., Fransson, S.G. et al. (2003).
insights from the Dartmouth Dynamic Registry. Catheter. Nephrotoxic effects in high-risk patients undergoing
Cardiovasc. Interv. 72: 347–354. angiography. N. Engl. J. Med. 348: 491–499.
66 James, M.T., Ghali, W.A., Tonelli, M. et al. (2010). Acute 78 Jo, S.H., Youn, T.J., Koo, B.K. et al. (2006). Renal toxicity
kidney injury following coronary angiography is evaluation and comparison between visipaque (iodixanol)
associated with a long-term decline in kidney function. and hexabrix (ioxaglate) in patients with renal
Kidney Int. 78: 803–809. insufficiency undergoing coronary angiography: the
67 Bandeali, S.J., Kayani, W.T., Lee, V.V. et al. (2013). RECOVER study: a randomized controlled trial. J. Am.
Association between preoperative diuretic use and Coll. Cardiol. 48: 924–930.
in-hospital outcomes after cardiac surgery. Cardiovasc. 79 Carraro, M., Malalan, F., Antonione, R. et al. (1998).
Ther. 31: 291–297. Effects of a dimeric vs a monomeric nonionic contrast
68 Zlatanovic, P., Koncar, I., Dragas, M. et al. (2018). medium on renal function in patients with mild to
Combined impact of chronic kidney disease and contrast moderate renal insufficiency: a double-blind, randomized
induced acute kidney injury on long-term outcomes in clinical trial. Eur. Radiol. 8: 144–147.
patients with acute lower limb ischaemia. Eur. J. Vasc. 80 Chalmers, N. and Jackson, R.W. (1999). Comparison of
Endovasc. Surg. 56: 78–86. iodixanol and iohexol in renal impairment. Br. J. Radiol.
69 Han, J.H., Chandra, A., Mulgund, J. et al. (2006). Chronic 72: 701–703.
kidney disease in patients with non-ST-segment elevation 81 Juergens, C.P., Winter, J.P., Nguyen-Do, P. et al. (2009).
acute coronary syndromes. Am. J. Med. 119: 248–254. Nephrotoxic effects of iodixanol and iopromide in
70 Szummer, K., Lundman, P., Jacobson, S.H. et al. (2010). patients with abnormal renal function receiving
Relation between renal function, presentation, use of N-acetylcysteine and hydration before coronary
therapies and in-hospital complications in acute coronary angiography and intervention: a randomized trial. Intern.
syndrome: data from the SWEDEHEART register. J. Med. J. 39: 25–31.
Intern. Med. 268: 40–49. 82 Laskey, W., Aspelin, P., Davidson, C. et al. (2009).
71 Goldenberg, I., Subirana, I., Boyko, V. et al. (2010). Nephrotoxicity of iodixanol versus iopamidol in patients
Relation between renal function and outcomes in with chronic kidney disease and diabetes mellitus
patients with non-ST-segment elevation acute coronary undergoing coronary angiographic procedures. Am. Heart
syndrome: real-world data from the European Public J. 158: 822–882. e3.
Health Outcome Research and Indicators Collection 83 Nguyen, S.A., Suranyi, P., Ravenel, J.G. et al. (2008).
Project. Arch. Intern. Med. 170: 888–895. Iso-osmolality versus low-osmolality iodinated contrast
72 Nauta, S.T., van Domburg, R.T., Nuis, R.J. et al. (2013). medium at intravenous contrast-enhanced CT: effect on
Decline in 20-year mortality after myocardial infarction kidney function. Radiology 248: 97–105.
4 Solomon, R.J., Natarajan, M.K., Doucet, S. et al. (2007).

8 95 Marenzi, G., Marana, I., Lauri, G. et al. (2003). The
Cardiac Angiography in Renally Impaired Patients prevention of radiocontrast-agent-induced nephropathy
(CARE) study: a randomized double-blind trial of by hemofiltration. N. Engl. J. Med. 349: 1333–1340.
contrast-induced nephropathy in patients with chronic 96 Solomon, R., Werner, C., Mann, D., and D’Elia, J. (1994).
kidney disease. Circulation 115: 3189–3196. Silva P. Effects of saline, mannitol, and furosemide to
85 Anderson, J.L., Adams, C.D., Antman, E.M. et al. (2013). prevent acute decreases in renal function induced by
2012 ACCF/AHA focused update incorporated into the radiocontrast agents. [see comments.]. N. Engl. J. Med.
ACCF/AHA 2007 guidelines for the management of 331: 1416–1420.
patients with unstable angina/non-ST-elevation 97 Weinstein, J.M., Heyman, S., and Brezis, M. (1992).
myocardial infarction: a report of the American College Potential deleterious effect of furosemide in
of Cardiology Foundation/American Heart Association radiocontrast nephropathy. Nephron 62: 413–415.
Task Force on Practice Guidelines. Circulation 127: 98 Majumdar, S.R., Kjellstrand, C.M., Tymchak, W.J. et al.
e663–e828. (2009). Forced euvolemic diuresis with mannitol and
86 ESUR (2008). Guidelines on contrast media. ESUR.org. furosemide for prevention of contrast-induced
87 Lee, P.T., Chou, K.J., Liu, C.P. et al. (2007). Renal nephropathy in patients with CKD undergoing coronary
protection for coronary angiography in advanced renal angiography: a randomized controlled trial. Am. J.
failure patients by prophylactic hemodialysis. A Kidney Dis. 54: 602–609.
randomized controlled trial. J. Am. Coll. Cardiol. 50: 99 Hall, K.A., Wong, R.W., Hunter, G.C. et al. (1992).
1015–1020. Contrast-induced nephrotoxicity: the effects of
88 Reinecke, H., Fobker, M., Wellmann, J. et al. (2007). A vasodilator therapy. J. Surg. Res. 53: 317–320.
randomized controlled trial comparing hydration therapy 100 Kellum, J.A. (1997). The use of diuretics and dopamine
to additional hemodialysis or N-acetylcysteine for the in acute renal failure: a systematic review of the
prevention of contrast medium-induced nephropathy: the evidence. Crit. Care (Lond) 1: 53–59.
Dialysis-versus-Diuresis (DVD) Trial. Clin. Res. Cardiol. 101 Weisberg, L.S., Kurnik, P.B., and Kurnik, B.R. (1993).
96: 130–139. Dopamine and renal blood flow in radiocontrast-
89 Holscher, B., Heitmeyer, C., Fobker, M. et al. (2008). induced nephropathy in humans. Ren. Fail. 15: 61–68.
Predictors for contrast media-induced nephropathy and 102 Kapoor, A., Sinha, N., Sharma, R.K. et al. (1996). Use of
long-term survival: prospectively assessed data from the dopamine in prevention of contrast induced acute renal
randomized controlled dialysis-versus-diuresis (DVD) failure--a randomised study. Int. J. Cardiol. 53: 233–236.
trial. Can. J. Cardiol. 24: 845–850. 103 Bakris, G.L., Lass, N.A., and Glock, D. (1999). Renal
90 Hsieh, Y.C., Ting, C.T., Liu, T.J. et al. (2005). Short-and hemodynamics in radiocontrast medium-induced renal
long-term renal outcomes of immediate prophylactic dysfunction: A role for dopamine-1 receptors. Kidney
hemodialysis after cardiovascular catheterizations in Int. 56: 206–210.
patients with severe renal insufficiency. Int. J. Cardiol. 104 Madyoon, H. and Croushore, L. (2001). Use of
101: 407–413. fenoldopam for prevention of radiocontrast
91 Berger, E.D., Bader, B.D., Bosker, J. et al. (2001). Contrast nephropathy in the cardiac catheterization laboratory: a
media-induced kidney failure cannot be prevented by case series. J. Interv. Cardiol. 14: 179–185.
hemodialysis. Dtsch. Med. Wochenschr. 126: 162–166. 105 Madyoon, H., Croushore, L., Weaver, D., and Mathur, V.
92 Frank, H., Werner, D., Lorusso, V. et al. (2003). (2001). Use of fenoldopam to prevent radiocontrast
Simultaneous hemodialysis during coronary nephropathy in high-risk patients. Catheter. Cardiovasc.
angiography fails to prevent radiocontrast-induced Interv. 53: 341–345.
nephropathy in chronic renal failure. Clin. Nephrol. 106 Singer, I. and Epstein, M. (1998). Potential of dopamine
60: 176–182. A-1 agonists in the management of acute renal failure.
93 Huber, W., Jeschke, B., Kreymann, B. et al. (2002). Am. J. Kidney Dis. 31: 743–755.
Haemodialysis for the prevention of contrast-induced 107 Madyoon, H. (2001). Clinical experience with the use of
nephropathy: outcome of 31 patients with severely fenoldopam for prevention of radiocontrast
impaired renal function, comparison with patients at nephropathy in high-risk patients. Rev. Cardiovasc. Med.
similar risk and review. Investig. Radiol. 37: 471–481. 2 (Suppl. 1): S26–S30.
94 Marenzi, G., Lauri, G., Campodonico, J. et al. (2006). 108 Mathur, V.S. (2003). The role of the DA1 receptor
Comparison of two hemofiltration protocols for agonist fenoldopam in the management of critically ill,
prevention of contrast-induced nephropathy in high-risk transplant, and hypertensive patients. Rev. Cardiovasc.
patients. Am. J. Med. 119: 155–162. Med. 4 (Suppl. 1): S35–S40.
Reference 159
109 Stone, G.W., McCullough, P.A., Tumlin, J.A. et al. 121 Baker, C.S. and Baker, L.R. (2001). Prevention of
(2003). Fenoldopam mesylate for the prevention of contrast nephropathy after cardiac catheterisation.
contrast-induced nephropathy: a randomized controlled Heart 4: 361–362.
trial. JAMA 290: 2284–2291. 122 Briguori, C., Manganelli, F., Scarpato, P. et al. (2002).
110 Kurnik, B.R., Allgren, R.L., Genter, F.C. et al. (1998). Acetylcysteine and contrast agent-associated
Prospective study of atrial natriuretic peptide for the nephrotoxicity. J. Am. Coll. Cardiol. 40: 298–303.
prevention of radiocontrast-induced nephropathy. Am. 123 Coyle, L.C., Rodriguez, A., Jeschke, R.E. et al. (2006).
J. Kidney Dis. 31: 674–680. Acetylcysteine In Diabetes (AID): a randomized study
111 Morikawa, S., Sone, T., Tsuboi, H. et al. (2009). Renal of acetylcysteine for the prevention of contrast
protective effects and the prevention of contrast-induced nephropathy in diabetics. Am. Heart J.
nephropathy by atrial natriuretic peptide. J. Am. Coll. 151: 1032. e9–e12.
Cardiol. 53: 1040–1046. 124 Kay, J., Chow, W.H., Chan, T.M. et al. (2003).
112 Erley, C.M., Duda, S.H., Rehfuss, D. et al. (1999). Acetylcysteine for prevention of acute deterioration of
Prevention of radiocontrast-media-induced renal function following elective coronary angiography
nephropathy in patients with pre-existing renal and intervention: a randomized controlled trial. JAMA
insufficiency by hydration in combination with the 289: 553–558.
adenosine antagonist theophylline. Nephrol. Dial. 125 Gomes, V.O., Poli de Figueredo, C.E., Caramori, P. et al.
Transplant. 14: 1146–1149. (2005). N-acetylcysteine does not prevent contrast
113 Katholi, R.E., Taylor, G.J., McCann, W.P. et al. (1995). induced nephropathy after cardiac catheterisation with
Nephrotoxicity from contrast media: attenuation with an ionic low osmolality contrast medium: a multicentre
theophylline. Radiology 195: 17–22. clinical trial. Heart 91: 774–778.
114 Bagshaw, S.M. and Ghali, W.A. (2005). Theophylline for 126 Fung, J.W., Szeto, C.C., Chan, W.W. et al. (2004). Effect
prevention of contrast-induced nephropathy: a of N-acetylcysteine for prevention of contrast
systematic review and meta-analysis. Arch. Intern. Med. nephropathy in patients with moderate to severe renal
165: 1087–1093. insufficiency: a randomized trial. Am. J. Kidney Dis. 43:
115 Ix, J.H., McCulloch, C.E., and Chertow, G.M. (2004). 801–808.
Theophylline for the prevention of radiocontrast 127 Durham, J.D., Caputo, C., Dokko, J. et al. (2002). A
nephropathy: a meta-analysis. Nephrol. Dial. Transplant. randomized controlled trial of N-acetylcysteine to
19: 2747–2753. prevent contrast nephropathy in cardiac angiography.
116 Kelly, A.M., Dwamena, B., Cronin, P. et al. (2008). Kidney Int. 62: 2202–2207.
Meta-analysis: effectiveness of drugs for preventing 128 Allaqaband, S., Tumuluri, R., Malik, A.M. et al. (2002).
contrast-induced nephropathy. Ann. Intern. Med. 148: Prospective randomized study of N-acetylcysteine,
284–294. fenoldopam, and saline for prevention of radiocontrast-
117 Han, Y., Zhu, G., Han, L. et al. (2014). Short-term induced nephropathy. Catheter. Cardiovasc. Interv. 57:
rosuvastatin therapy for prevention of contrast-induced 279–283.
acute kidney injury in patients with diabetes and 129 Shyu, K.G., Cheng, J.J., and Kuan, P. (2002).
chronic kidney disease. J. Am. Coll. Cardiol. 63: 62–70. Acetylcysteine protects against acute renal damage in
118 Leoncini, M., Toso, A., Maioli, M. et al. (2014). Early patients with abnormal renal function undergoing a
high-dose rosuvastatin for contrast-induced coronary procedure. J. Am. Coll. Cardiol. 40: 1383–1388.
nephropathy prevention in acute coronary syndrome: 130 Sandhu, C., Belli, A.M., and Oliveira, D.B. (2006). The
Results from the PRATO-ACS Study (protective effect of role of N-acetylcysteine in the prevention of contrast-
rosuvastatin and antiplatelet therapy on contrast- induced nephrotoxicity. Cardiovasc. Intervent. Radiol.
induced acute kidney injury and myocardial damage in 29: 344–347.
patients with Acute Coronary Syndrome). J. Am. Coll. 131 Rashid, S.T., Salman, M., Myint, F. et al. (2004).
Cardiol. 63: 71–79. Prevention of contrast-induced nephropathy in vascular
119 Vidt, D.G., Harris, S., McTaggart, F. et al. (2006). Effect patients undergoing angiography: a randomized
of short-term rosuvastatin treatment on estimated controlled trial of intravenous N-acetylcysteine. J. Vasc.
glomerular filtration rate. Am. J. Cardiol. 97: 1602–1606. Surg. 40: 1136–1141.
120 Tepel, M., van der Giet, M., Schwarzfeld, C. et al. (2000). 132 Oldemeyer, J.B., Biddle, W.P., Wurdeman, R.L. et al.
Prevention of radiographic-contrast-agent-induced (2003). Acetylcysteine in the prevention of contrast-
reductions in renal function by acetylcysteine. N. Engl. J. induced nephropathy after coronary angiography. Am.
Med. 343: 180–184. Heart J. 146: E23.
133 Ochoa, A., Pellizzon, G., Addala, S. et al. (2004). nephropathy but not clinical events during long-term
Abbreviated dosing of N-acetylcysteine prevents follow-up. Am. Heart J. 148: 690–695.
contrast-induced nephropathy after elective and urgent 145 Alonso, A., Lau, J., Jaber, B.L. et al. (2004).
coronary angiography and intervention. J. Interv. Prevention of radiocontrast nephropathy with
Cardiol. 17: 159–165. N-acetylcysteine in patients with chronic kidney
134 MacNeill, B.D., Harding, S.A., Bazari, H. et al. (2003). disease: a meta-analysis of randomized, controlled
Prophylaxis of contrast-induced nephropathy in patients trials. Am. J. Kidney Dis. 43: 1–9.
undergoing coronary angiography. Catheter. Cardiovasc. 146 Bagshaw, S.M. and Ghali, W.A. (2004). Acetylcysteine
Interv. 60: 458–461. for prevention of contrast-induced nephropathy after
135 Kefer, J.M., Hanet, C.E., Boitte, S. et al. (2003). intravascular angiography: a systematic review and
Acetylcysteine, coronary procedure and prevention of meta-analysis. BMC Med. 2: 38.
contrast-induced worsening of renal function: which 147 Birck, R., Krzossok, S., Markowetz, F. et al. (2003).
benefit for which patient? Acta Cardiol. 58: 555–560. Acetylcysteine for prevention of contrast nephropathy:
136 Goldenberg, I., Shechter, M., Matetzky, S. et al. (2004). meta-analysis. Lancet 362: 598–603.
Oral acetylcysteine as an adjunct to saline hydration for 148 Duong, M.H., MacKenzie, T.A., and Malenka, D.J.
the prevention of contrast-induced nephropathy (2005). N-acetylcysteine prophylaxis significantly
following coronary angiography. A randomized reduces the risk of radiocontrast-induced nephropathy:
controlled trial and review of the current literature. Eur. comprehensive meta-analysis. Catheter. Cardiovasc.
Heart J. 25: 212–218. Interv. 64: 471–479.
137 Drager, L.F., Andrade, L., Barros de Toledo, J.F. et al. 149 Gonzales, D.A., Norsworthy, K.J., Kern, S.J. et al. (2007).
(2004). Renal effects of N-acetylcysteine in patients at A meta-analysis of N-acetylcysteine in contrast-induced
risk for contrast nephropathy: decrease in oxidant nephrotoxicity: unsupervised clustering to resolve
stress-mediated renal tubular injury. Nephrol. Dial. heterogeneity. BMC Med. 5: 32.
Transplant. 19: 1803–1807. 150 Isenbarger, D.W., Kent, S.M., and O’Malley, P.G. (2003).
138 Diaz-Sandoval, L.J., Kosowsky, B.D., and Losordo, D.W. Meta-analysis of randomized clinical trials on the
(2002). Acetylcysteine to prevent angiography-related usefulness of acetylcysteine for prevention of contrast
renal tissue injury (the APART trial). Am. J. Cardiol. 89: nephropathy. Am. J. Cardiol. 92: 1454–1458.
356–358. 151 Kshirsagar, A.V., Poole, C., Mottl, A. et al. (2004).
139 Azmus, A.D., Gottschall, C., Manica, A. et al. (2005). N-acetylcysteine for the prevention of radiocontrast
Effectiveness of acetylcysteine in prevention of contrast induced nephropathy: a meta-analysis of prospective
nephropathy. J. Invasive Cardiol. 17: 80–84. controlled trials. J. Am. Soc. Nephrol. 15: 761–769.
140 Webb, J.G., Pate, G.E., Humphries, K.H. et al. (2004). A 152 Misra, D., Leibowitz, K., Gowda, R.M. et al. (2004). Role
randomized controlled trial of intravenous of N-acetylcysteine in prevention of contrast-induced
N-acetylcysteine for the prevention of contrast-induced nephropathy after cardiovascular procedures: a
nephropathy after cardiac catheterization: lack of effect. meta-analysis. Clin. Cardiol. 27: 607–610.
Am. Heart J. 148: 422–429. 153 Nallamothu, B.K., Shojania, K.G., Saint, S. et al. (2004).
141 Balderramo, D.C., Verdu, M.B., Ramacciotti, C.F. et al. Is acetylcysteine effective in preventing contrast-related
(2004). Renoprotective effect of high periprocedural nephropathy? A meta-analysis. Am. J. Med. 117:
doses of oral N-acetylcysteine in patients scheduled to 938–947.
undergo a same-day angiography. Rev. Fac. Cien. Med. 154 Pannu, N., Manns, B., Lee, H., and Tonelli, M. (2004).
Univ. Nac. Cordoba 61: 13–19. Systematic review of the impact of N-acetylcysteine on
142 Carbonell, N., Blasco, M., Sanjuan, R. et al. (2007). contrast nephropathy. Kidney Int. 65: 1366–1374.
Intravenous N-acetylcysteine for preventing contrast- 155 Weisbord, S.D., Gallagher, M., Jneid, H. et al. (2017).
induced nephropathy: a randomised trial. Int. J. Cardiol. Outcomes after angiography with sodium bicarbonate
115: 57–62. and acetylcysteine. N. Engl. J. Med. 378 (7): 603–614.
143 Amini, M., Salarifar, M., Amirbaigloo, A. et al. (2009). 156 Weisbord, S.D. and Palevsky, P.M. (2008). Prevention of
N-acetylcysteine does not prevent contrast-induced contrast-induced nephropathy with volume expansion.
nephropathy after cardiac catheterization in patients Clin. J. Am. Soc. Nephrol. 3: 273–280.
with diabetes mellitus and chronic kidney disease: a 157 Trivedi, H.S., Moore, H., Nasr, S. et al. (2003). A
randomized clinical trial. Trials 10: 45. randomized prospective trial to assess the role of saline
144 Miner, S.E., Dzavik, V., Nguyen-Ho, P. et al. (2004). hydration on the development of contrast
N-acetylcysteine reduces contrast-associated nephrotoxicity. Nephron 93: C29–C34.
Reference 161
58 Mueller, C., Buerkle, G., Buettner, H.J. et al. (2002).

1 nephropathy in patients undergoing coronary
Prevention of contrast media-associated nephropathy: angiography: a randomized controlled trial. Am. J.
randomized comparison of 2 hydration regimens in Kidney Dis. 54: 610–618.
1620 patients undergoing coronary angioplasty. [see 168 Dong, Y., Zhang, B., Liang, L. et al. (2016). How strong
comments.]. Arch. Intern. Med. 162: 329–336. is the evidence for sodium bicarbonate to prevent
159 Nijssen, E.C., Rennenberg, R.J., Nelemans, P.J. et al. contrast-induced acute kidney injury after coronary
(2017). Prophylactic hydration to protect renal function angiography and percutaneous coronary intervention?
from intravascular iodinated contrast material in Medicine (Baltimore) 95: e2715.
patients at high risk of contrast-induced nephropathy 169 Zoungas, S., Ninomiya, T., Huxley, R. et al. (2009).
(AMACING): a prospective, randomised, phase 3, Systematic review: sodium bicarbonate treatment
controlled, open-label, non-inferiority trial. Lancet 389: regimens for the prevention of contrast-induced
1312–1322. nephropathy. Ann. Intern. Med. 151: 631–638.
160 Merten, G.J., Burgess, W.P., Gray, L.V. et al. (2004). 170 Navaneethan, S.D., Singh, S., Appasamy, S. et al. (2009).
Prevention of contrast-induced nephropathy with Sodium bicarbonate therapy for prevention of contrast-
sodium bicarbonate: a randomized controlled trial. induced nephropathy: a systematic review and meta-
JAMA 291: 2328–2334. analysis. Am. J. Kidney Dis. 53: 617–627.
161 Brar, S.S., Shen, A.Y., Jorgensen, M.B. et al. (2008). 171 Meier, P., Ko, D.T., Tamura, A. et al. (2009). Sodium
Sodium bicarbonate vs sodium chloride for the bicarbonate-based hydration prevents contrast-induced
prevention of contrast medium-induced nephropathy in nephropathy: a meta-analysis. BMC Med. 7: 23.
patients undergoing coronary angiography: a 172 Hoste, E.A., De Waele, J.J., Gevaert, S.A. et al. (2009).
randomized trial. JAMA 300: 1038–1046. Sodium bicarbonate for prevention of contrast-induced
162 Kanbay, M., Covic, A., Coca, S.G. et al. (2009). Sodium acute kidney injury: a systematic review and meta-
bicarbonate for the prevention of contrast-induced analysis. Nephrol. Dial. Transplant. 25 (3): 747–758.
nephropathy: a meta-analysis of 17 randomized trials. 173 Brown, J.R., Block, C.A., Malenka, D.J. et al. (2009).
Int. Urol. Nephrol. 41: 617–627. Sodium bicarbonate plus N-acetylcysteine prophylaxis:
163 Masuda, M., Yamada, T., Mine, T. et al. (2007). a meta-analysis. JACC Cardiovasc. Interv. 2: 1116–1124.
Comparison of usefulness of sodium bicarbonate versus 174 Joannidis, M., Schmid, M., and Wiedermann, C.J.
sodium chloride to prevent contrast-induced (2008). Prevention of contrast media-induced
nephropathy in patients undergoing an emergent nephropathy by isotonic sodium bicarbonate: a meta-
coronary procedure. Am. J. Cardiol. 100: 781–786. analysis. Wien. Klin. Wochenschr. 120: 742–748.
164 Ozcan, E.E., Guneri, S., Akdeniz, B. et al. (2007). 175 Hogan, S.E., L’Allier, P., Chetcuti, S. et al. (2008).
Sodium bicarbonate, N-acetylcysteine, and saline for Current role of sodium bicarbonate-based preprocedural
prevention of radiocontrast-induced nephropathy. A hydration for the prevention of contrast-induced acute
comparison of 3 regimens for protecting contrast- kidney injury: a meta-analysis. Am. Heart J. 156:
induced nephropathy in patients undergoing coronary 414–421.
procedures. A single-center prospective controlled trial. 176 Ho, K.M. and Morgan, D.J. (2008). Use of isotonic
Am. Heart J. 154: 539–544. sodium bicarbonate to prevent radiocontrast
165 Pakfetrat, M., Nikoo, M.H., Malekmakan, L. et al. nephropathy in patients with mild pre-existing renal
(2009). A comparison of sodium bicarbonate infusion impairment: a meta-analysis. Anaesth. Intensive Care 36:
versus normal saline infusion and its combination with 646–653.
oral acetazolamide for prevention of contrast-induced 177 Kunadian, V., Zaman, A., Spyridopoulos, I., and Qiu, W.
nephropathy: a randomized, double-blind trial. Int. Urol. (2011). Sodium bicarbonate for the prevention of
Nephrol. 41: 629–634. contrast induced nephropathy: a meta-analysis of
166 Recio-Mayoral, A., Chaparro, M., Prado, B. et al. (2007). published clinical trials. Eur. J. Radiol. 79 (1): 48–55.
The reno-protective effect of hydration with sodium 178 Solomon, R., Gordon, P., Manoukian, S.V. et al. (2015).
bicarbonate plus N-acetylcysteine in patients Randomized trial of bicarbonate or saline study for the
undergoing emergency percutaneous coronary prevention of contrast-induced nephropathy in patients
intervention: the RENO Study. J. Am. Coll. Cardiol. 49: with CKD. Clin. J. Am. Soc. Nephrol. 10: 1519–1524.
1283–1288. 179 ACT Trial Investigators (2009). Rationale, design, and
167 Vasheghani-Farahani, A., Sadigh, G., Kassaian, S.E. baseline characteristics of the Acetylcystein for
et al. (2009). Sodium bicarbonate plus isotonic saline Contrast-Induced nephropaThy (ACT) Trial: a
versus saline for prevention of contrast-induced pragmatic randomized controlled trial to evaluate the
efficacy of acetylcysteine for the prevention of contrast- 182 Qian, G., Fu, Z., Guo, J. et al. (2016). Prevention of
induced nephropathy. Trials 10: 38. contrast-induced nephropathy by central venous
180 Shah, R., Wood, S.J., Khan, S.A. et al. (2017). High- pressure-guided fluid administration in chronic kidney
volume forced diuresis with matched hydration using disease and congestive heart failure patients. JACC
the RenalGuard System to prevent contrast-induced Cardiovasc. Interv. 9: 89–96.
nephropathy: a meta-analysis of randomized trials. Clin. 183 Cohn, J.N. (2004). Introduction to surrogate markers.
Cardiol. 40: 1242–1246. Circulation 109: IV20–IV21.
81 Brar, S.S., Aharonian, V., Mansukhani, P. et al. (2014).
1 184 Wu, Z., Zhang, H., Jin, W. et al. (2015). The effect of
Haemodynamic-guided fluid administration for the renin-angiotensin-aldosterone system blockade
prevention of contrast-induced acute kidney injury: the medications on contrast-induced nephropathy in
POSEIDON randomised controlled trial. Lancet 383: patients undergoing coronary angiography: a meta-
1814–1823. analysis. PLoS One 10: e0129747.
163
12
Miscellaneous Etiologies of Acute Kidney Injury

Christina Mariyam Joy and Anitha Vijayan
Division of Nephrology, Washington University in St. Louis St. Louis USA
I ntroduction in the kidney interstitium, either alone or as a hapten [8].

This is usually not dose-dependent, but idiosyncratic with
This chapter will review some of the less frequent causes of recrudescence in disease activity upon re-exposure to
acute kidney injury (AKI): acute interstitial nephritis compounds with similar biochemical structure, often
(AIN), atheroembolic renal disease, cast nephropathy, with multiorgan involvement. Newer chemotherapeutic
crystalline nephropathies, and renovascular syndromes. agents, including immune checkpoint regulators and pro-
Due to the relative rarity of these conditions, there are few grammed cell death proteins, cause dose-dependent tox-
randomized controlled trials to guide management. icity by immune upregulation [9]. On histology, there are
However, several observational and retrospective studies focal to diffuse interstitial infiltrates composed of T-cell
provide some insight into the clinical features, natural predominant lymphocytes. Immune complexes or antitu-
histories, and therapeutic responses. bular basement membrane antibodies are infrequently
noted, suggesting that humoral immunity may also be
involved [10, 11]. Granulomatous interstitial nephritis
Acute Interstitial Nephritis (GIN) has been identified in approximately 0.5–0.9% of all
Definition, Etiology, and Pathogenesis kidney biopsy specimens. Approximately 17.5–45% of
AIN is a hypersensitivity reaction that causes AKI by GIN is attributed to drug toxicity [12]. The most common
immune-mediated tubular injury localized to the kidney etiologies of AIN are drugs (70%), autoimmune diseases
interstitium with sparing of the glomeruli. The first (20%), infections (4%), and idiopathic (5%) [5] (Table 12.1).
description of AIN was by Biermer in 1860 [1]. In 1898, Antibiotics (49%), proton pump inhibitors (14%), and
Councilman used the term acute interstitial lesions to nonsteroidal anti-inflammatory drugs (NSAIDs; 11%)
describe the histopathological changes noted in autopsy accounted for the majority of cases, although any drug
specimens of children who had succumbed to scarlet fever, can potentially cause AIN [5]. The offending drug may be
diphtheria, and other infectious diseases [2]. In 1946, sul- difficult to identify due to polypharmacy [13].
fonamides became one of the first drugs noted to cause
intestinal nephritis [3] followed by penicillin and methicil-
Clinical Features
lin in 1968 [4]. The true incidence of AIN is difficult to
ascertain but data from various literature searches provide The classical signs and symptoms of drug-induced AIN
estimates of 0.5–3% of all kidney biopsies and 5–27% of include a triad of rash, fever, and eosinophilia [14], which
biopsies done for AKI [5–7]. The etiology of AIN is esti- suggests an allergic type reaction. However, it should be
mated to be drug-induced in as many as 78% of cases in noted that the constellation of the three findings is an
developed countries while infectious etiologies are noted exception and not the rule. These allergic symptoms may
in 50% of cases from developing countries. be absent, especially in those cases associated with
AIN is a delayed hypersensitivity reaction mediated by NSAID, where nonspecific symptoms of acute kidney fail-
T-cells, which may occur through diverse mechanisms, ure including nausea, vomiting, malaise, and fatigue
including molecular mimicry with tubular epitopes or maybe seen. Oliguric or nonoliguric kidney dysfunction
deposition of immunogenic portions of the inciting agent can occur. The latent period for disease manifestation is
164 Miscellaneous Etiologies of Acute Kidney Injury
Table 12.1 Etiology of acute interstitial nephritis.
Drugs Antibiotics Penicillina (penicillin, ampicillin, and methicillin), cephalosporins, fluoroquinolonesa

(ciprofloxacin, levofloxacin), rifampina and sulfonamidesa (including cotrimoxazole)
Abacavir, acyclovir, indinavir
NSAIDs Celecoxib, diclofenac, fenoprofen, indomethacin, meloxicam, sulindac
Mesalamine, sulfasalazine
Antacid/antisecretory Esomeprazole, omeprazole, pantoprazole
Cimetidine, ranitidine, famotidine
Chemotherapeutic Atezolizumab, iplilimumab, nivolumab, pembrolizumab
agents Sorafenib, sunitinib
Adriamycin, BCG, bevacizumab, bortezomib, gemcitabine, lenalidomide, pemetrexed
Diuretics Chlorthalidone, hydrochlorothiazide, metolazone, furosemide, amiloride, triamterene
Antihypertensive Captopril, lisinopril, losartan, amlodipine, nifedipine
Anticonvulsants Carbamazepine, levetiracetam, phenobarbital, phenytoin, valproic acid
Miscellaneous Allopurinol, atorvastatin, carbimazole, feboxustat, gemfibrozil, leflunomide, sildenafil
Infectious agents Bacteria Brucella spp.,a Campylobacter jejuni, Corynebacterium diphtheriae, Chlamydia spp.,
Escherichia coli, Legionella spp., L. interrogans, Mycobacterium tuberculosis,a
Mycoplasma pneumoniae, Rickettsia spp., Salmonella spp.,a Staphylococcus spp.,
Streptococcus spp., Yersinia pseudotuberculosis
Virus Cytomegalovirus, Epstein–Barr virus,a hantaviruses, hepatitis B virus, human immune
deficiency virus, herpes simplex virus, measles virus, polyomaviruses
Parasites Leishmania donovani, Toxoplasma gondiia
Systemic diseases Sarcoidosis,a Sjogren’s syndrome,a systemic lupus erythematosus, tubulointerstitial
nephritis and uveitis syndrome,a Antineutrophil cytoplasmic antibody-associated
vasculitis, other vasculitis
a
Denotes association with granulomatous interstitial nephritis.
usually 3 weeks after the initiation of the offending drug can be seen, usually detected more reliably with Hansel’s
in 80% of individuals, with an average of 10 days (although stain than with Wright’s stain [23]. However, the diagnos-
can range from 1 day to more than a year). NSAIDs have a tic utility of testing for urine eosinophils has recently been
slightly longer latency of 2–3 months. The temporal rela- invalidated and is no longer recommended. In a retrospec-
tionship between the initiation of a new drug and the tive study of 566 patients who had urine eosinophil testing
development of kidney injury may also aid in making the and a kidney biopsy within a week of each other, 91 patients
diagnosis. Disease manifestations develop within 3 weeks had evidence of AIN on pathology. Urine eosinophils at 1%
of initiation of the offending drug in about 80% of patients, and 5% cutoff had sensitivities of 30.8% and 19.8% and spe-
with an average latency of onset of 10 days (range 1 day to cificities of 68.2% and 91.2%, respectively [24]. Urine chem-
>1 year) [8]. The duration of onset may be longer when istries are not useful in distinguishing AIN from other
associated with NSAIDs, with a mean latent period of etiologies of AKI. Fractional excretion of sodium can either
2–3 months [15]. When AIN is associated with a systemic be above or below 1% [25].
disease, the clinical features of the respective diseases Other laboratory abnormalities that may be present
may be present (Table 12.2). include metabolic acidosis secondary to renal tubular aci-
dosis, low specific gravity due to a concentrating defect,
glycosuria, or phosphaturia secondary to Fanconi syn-
Laboratory Diagnosis
drome. Mild proteinuria is common, but sometimes it may
Urinalysis in patients suspected to have AIN often show be in the nephrotic range. Though heavy proteinuria has
hematuria and/or sterile pyuria, but these are nonspecific been reported with AIN from various drugs, it is classically
findings. The presence of white blood cell (WBC) casts is associated with NSAIDs. Concomitant minimal change
more specific, although they can also be seen in pyelone- glomerulopathy may be seen in NSAID-associated AIN,
phritis and certain proliferative glomerulonephritis. although the exact incidence is unclear. One paper noted
Eosinophiluria (urine eosinophils >1% of the urine WBCs) that 70% of patients with NSAID-associated AIN had
Table 12.2 Causes of AIN and characteristic clinical and laboratory features.
Inciting agent
[reference(s)] Clinical features Laboratory findings Clinical course
All drugs other than Fever (45%) Hematuria (53%) Mean 10-day exposure prior to onset
methicillin [2, 8, 9] Rash (42%) Pyuria (50%) Temporary dialysis required in 32–50%
Arthralgia (12%) Mild proteinuria (58%) CKD remains in 36–40%
Flank pain (45%) Eosinophilia (40%)
Oliguria (40%)
Macroscopic hematuria (17%)
New or worsened hypertension (20%)
Methicillin [2, 8] Hypersensitivity symptoms common: Hematuria (90%) Mean duration of impaired kidney function,
Fever (85%) Pyuria (95%, often w/WBC casts) 1.5 months
Rash (25%) Mild proteinuria (80%) Temporary dialysis required in 17%
Arthralgias (10%) • Eosinophilia (80%) CKD remains in only 10%
Oliguria (25%) Eosinophiluria (almost all patients)
Macroscopic hematuria (80%)
NSAIDs [2, 8–10] Hypersensitivity symptoms uncommon: Nephrotic-range proteinuria (38%) Mean exposure 2–3 months before
Fever, rash, or arthralgias (10%) Hematuria (38%) presentation
Macroscopic hematuria (7%) Pyuria (40%) Temporary dialysis required in 20–38%
New or worsened hypertension (17%) Eosinophilia (40%) CKD remains in 56%
Kidney biopsy may also show minimal
change disease
Allopurinol [10] Hypersensitivity symptoms very common and Eosinophilia fairly common Mortality may be as high as 25%
robust Hepatitis common Rate of kidney recovery unknown
Signs of vasculitis possible Kidney biopsy may reveal immune complex
Often occurs in setting of renal insufficiency deposition at TBM
from accumulation of the inciting metabolite,
oxypurinol
Rifampin [16] Hypersensitivity symptoms common and Eosinophilia (7%) Usually occurs 24 hours after dose with
robust Coombs-positive hemolysis (25%) current intermittent dosing or after previous
Fever (45%) Thrombocytopenia (50%) continuous exposure (up to 1 year prior)
Nausea or vomiting (72%) Hemoglobinuria (17%) Dialysis required in almost all cases
Abdominal pain (40%) Hepatitis (25%) CKD remains in only 3%
Rash uncommon Antirifampin antibodies (almost all
Flank pain (17%) patients)
Oligo-anuria (96%) Kidney biopsy rarely shows immune complex
deposition at TBM
(Continued)
0005152400.INDD 165 09-12-2022 15:22:11

Inciting agent
[reference(s)] Clinical features Laboratory findings Clinical course
PPI [17] Fever (<50%) Pyuria Median time from exposure to diagnosis exceeds
Rash (<10%) Hematuria 6 months
Eosinophilia (33%) White and red cell cast Absolute risk for CKD is increased (3.3%)
Fatigue and nausea (39%) Biopsy: interstitial infiltrates with or without
Weakness (22%) tubulitis
Checkpoint Systemic manifestations (pneumonitis, vasculitis, Eosinophilia Usually occur 3–12 months after starting
inhibitors [9] colitis, hepatitis, pancreatitis, myocarditis, Kidney biopsy may show AIN and/or treatment.
enteritis, neuropathy) occur with varying glomerulonephritis and membranous GN Treatment of choice is prednisone and stopping
frequency CPI
CPI is restarted after improvement in SCr
Leptospirosis [18] Preceding exposure to animal excrement Cholestatic hepatitis (93%) Nephropathy occurs in 40% of cases of
Fever (93%) Hemolytic anemia (72%) leptospirosis
Jaundice (93%) Thrombocytopenia (81%) Mortality of 26%
Hepatomegaly (76%) Hypokalemia (renal wasting) (38%) Temporary dialysis required in 74%
Gingival/GI bleeding, purpura (79%) Hyponatremia (79%) Persistent tubular transport defects may
Macroscopic hematuria (26%) Rhabdomyolysis (62%) remain in 29%
Conjunctival suffusion (12%) Positive blood/urine cultures or serology CKD remains in only 10.3%
Altered mental status (50%) Kidney biopsy reveals inflammation
Hypotension (62%) predominating at proximal tubules early;
Oligoanuria (95%) interstitial hemorrhage possible
Rhabdomyolysis (62%)
BK nephropathy [14] Usually occurs in kidney allografts 1 year after Decoy cells (tubular cells with enlarged Acute or gradual deterioration in kidney
transplant in setting of aggressive nucleus, intranuclear inclusions) in urine function
immunosuppression sediment, 100% sensitivity/71% specificity Decreases 5-year kidney allograft survival from
May occur in other immunosuppressed states also Viremia by PCR 100% sensitive/88% specific 76% to 46%
(e.g. HIV) Kidney biopsy reveals SV40 stain positive Often resolves with decrease in
Fever uncommon intranuclear inclusion bodies immunosuppression
Macroscopic hematuria rare
0005152400.INDD 166 09-12-2022 15:22:11

Sarcoidosis [19] Extrarenal symptoms of sarcoidosis predominate Eosinophilia (25%) Often a remitting/relapsing course responsive to
with pulmonary, ocular, and skin symptoms Hypercalcemia or normocalcemia despite pulse increase in steroids
most common advanced kidney failure CKD remains in 90%
Renal limited disease is very rare Chest radiography with hilar adenopathy
Most often occurs in young adults and/or infiltrates (90%)
Higher incidence in black population Elevated ACE level not reliable with kidney
involvement
Renal biopsy reveals noncaseating
granulomas and giant cells
TINU [20] 3 : 1 female predominance Eosinophilia (17%) Uveitis precedes renal disease in 21%, is
Median age of onset, 15 years Elevated serum IgG (83%) concurrent in 15%, and follows it in 65%
Eye pain or redness (32%) Kidney biopsy shows granulomas in 13%; Complete renal recovery often occurs
Fever (53%) unlike sarcoidosis, uveitis is not spontaneously within 1 year
Weight loss (47%) granulomatous Uveitis recurs in 54%, but recurrence of AIN is
rare
Abdominal or flank pain (28%)
CKD is rare
Arthralgias or myalgias (17%)
Rash (1%)
IgG4 disease [21] Pancreatitis Elevated IgG4 levels Good response to corticosteroids but recurrence
Sclerosing cholangitis Biopsy shows lymphoplasmocytic infiltrates is common.
Autoimmune hypophysitis of IgG4-plasma cells, storiform fibrosis, tissue
Sialadenitis eosinophilia and obliterative phlebitis
Interstitial lung disease
Retro-peritoneal and mediastinal fibrosis
DRESS syndrome [22] Fever Leukocytosis with eosinophilia (90%) or High mortality rates up to 10% from fulminant
Lymphadenopathy mononucleosis (40%) hepatitis
Rash that can progress to Exfoliative dermatitis Anemia Supportive treatment including glucocorticoids
Hepatitis Thrombocytopenia
Pneumonitis Elevated liver enzymes
Myocarditis
Nephritis
Colitis
CKD, chronic kidney disease; CPI, checkpoint inhibitor; DRESS, drug reaction with eosinophilia and systemic symptoms; GI, gastrointestinal; IgG, immunoglobulin G; PPI, proton pump
inhibitors; SCr, serum creatinine; SV40, simian virus 40; TBM, tubular basement membrane.
Particularly distinctive features are noted in bold. The percentages are estimates of the prevalence of the associated finding for each etiology.
0005152400.INDD 167 09-12-2022 15:22:11

nephrotic range proteinuria [8]. Signs of multiorgan dys-

function, such as elevated transaminases and hemolysis,
are occasionally seen.
Kidney imaging may reveal normal to large kidneys
with increased echogenicity, although this is a nonspe-
cific finding [25]. Gallium scintigraphy and 2-[18F]-fluor
o-2-deoxy-d -glucose-positron emission tomography
(FDG-PET) scanning have shown some promise in diag-
nosis of AIN, but lacks sufficient sensitivity and specific-
ity to be recommended as diagnostic tools. Kidney biopsy
is the gold standard for the definitive diagnosis of AIN. A
cellular infiltrate consisting mostly of T-cells and mac-
rophages with edema increases the separation of tubules
from one another, and there may be tubulitis or frank
tubular necrosis in severe AIN (Figure 12.1a,b).
(a)
Neutrophils, eosinophils, and plasma cells can often also
be found in the cellular infiltrate. The presence of neutro-
phils may be associated with better response to steroid
therapy [26]. Occasionally there is granulomatous inflam-
mation, which might provide clues as to the offending
agent, since this is associated with a limited list of causes
(Table 12.1). Immunofluorescence and electron micros-
copy usually do not reveal the presence of immune com-
plexes, but tubular basement membrane deposition can
rarely be seen. Glomeruli are usually normal, but electron
microscopy may reveal foot process effacement in NSAID-
associated AIN. A summary of the relevant laboratory
findings is given in Table 12.2.
Course and Treatment (b)
Given the polymorphic nature of the disease and the mul- Figure 12.1 Acute interstitial nephritis (AIN), drug-induced,
with hematoxylin and eosin stain. Note the presence of
tiple potential causative agents, AIN does not have a uni- interstitial edema and predominantly mononuclear
form course or response to treatment. In the prototype, inflammatory cells separating the normally adjacent tubules (a).
methicillin-induced AIN, the prognosis was excellent, A few eosinophils (arrows) and a plasma cell (arrowhead) can be
with complete recovery of kidney function noted in 90% seen on higher power (b). Source: Courtesy of Rosa Davila MD,
Washington University School of Medicine.
of patients after a mean period of 1.5 months after cessa-
tion of the drug exposure [2]. In nonmethicillin, drug-
induced AIN, chronic kidney disease (CKD) is an expected Due to the heterogeneity of AIN and severity of the ill-
sequela in 36–40% of cases [2]. The prevalence of CKD is ness, most studies that guide therapy in AIN are retrospec-
even higher (56%) with NSAID-induced AIN [27]. In one tive and there is a dearth of randomized controlled trials to
series consisting of seven cases of infection-associated define therapeutic strategies. Clearly, the most important
AIN, all had complete renal recovery. However, in this therapeutic maneuver is prompt removal of the inciting
case series the inciting infections were readily treatable or agent. Corticosteroids have been used in the treatment of
self-limited [13, 27]. Indeed, the prognosis for AIN may hypersensitivity, although the benefits of its use remain
depend on the promptness of elimination of the inciting controversial. Clarkson et al. studied 60 patients with AIN
agent, with those etiologies associated with milder symp- (92% drug-induced), 60% of whom were treated with corti-
toms and therefore delayed diagnosis (e.g. NSAIDs, costeroids, and found no difference in serum creatinine at
chronic infections, sarcoidosis) having worse prognosis 1 year between those who received and did not receive cor-
than those with more acute and dramatic presentations ticosteroids [28]. In contrast, Gonzales et al. reported that
(e.g. methicillin, rifampin, acute bacterial, or viral infec- use of steroids was beneficial, especially when started
tions) (Table 12.2). early, being associated with a lower final serum creatinine
Introductio 169
and less need for renal replacement therapy (RRT) [29]. In preexisting CKD, and histology with intense and diffuse
a single center, retrospective analysis from 1993 to interstitial infiltrate, granuloma formation, or significant
2011 which studied 133 patients with biopsy proven AIN fibrosis and tubular atrophy. If corticosteroids are consid-
(70% drug-induced AIN), 49% of patients with drug- ered, then they should be given early, as this is associated
induced AIN achieved complete recovery after initiation of with improvement in kidney function [33]. A frequently
corticosteroids. An additional 39% of patients had partial used regimen is oral prednisone, 1 mg/kg, with the dura-
recovery and 12% had no recovery. Correlates for poor tion of therapy guided by the improvement in kidney func-
recovery included longer duration of exposure to the drug tion, usually 3 weeks. The presence of conditions that can
and time to initiation of steroids [5]. A more recent, large, be exacerbated by corticosteroid therapy (e.g. slow-healing
single-center retrospective study from London included wounds, brittle diabetes, or active infection) should dis-
187 eligible patients with AIN out of 3983 native kidney suade the clinician from this option. Furthermore, studies
biopsies, and 158 patients were treated with steroids. It was have suggested that corticosteroids do not alter the course
concluded that steroid-treated patients had higher eGFR at of NSAID-induced AIN [28], but if poor prognostic factors
all follow-up intervals up to 24 months. Fewer patients are present, treatment can still be considered. Other immu-
were dependent on RRT at the end of 6 months and nosuppressants should be reserved for cases that are not
24 months (3.2% and 5.1%) when compared to patients not responsive to withdrawal of the inciting agent alone when
treated with steroids (20.6% and 24.1%, respectively) [30]. corticosteroid resistance or dependence occurs.
An attempt to match steroid-treated with untreated
patients showed greater improvement in renal outcomes,
Atheroembolic Renal Disease
although this was not statistically significant. Based on
their results, the authors suggested a regimen of predniso- Definition, Prevalence, and Pathogenesis
lone at 1 mg/kg with a maximum dose of 60 mg daily, with Atheroembolic renal disease causes progressive AKI and
rapid weaning over 8–12 weeks to prevent worsening of acute kidney disease (AKD) resulting from inflammation
kidney function. However, one must be careful in extrapo- in the renal microvasculature caused by cholesterol crys-
lating this data to all patients with AIN, since this study tals and lipid debris. The dislodged atheromatous plaques
only looked at patients with biopsy proven AIN and may can travel through the bloodstream and affect any organ,
not have included those who had improvement in renal including the kidneys, skin, brain, and gastrointestinal sys-
function after prompt withdrawal of the offending agent at tem. Atheroembolism often occurs after an inciting event
the onset of kidney injury. (like cardiac catheterization, vascular surgeries, etc.) or use
One retrospective series described the use of mycopheno- of systemic anticoagulation but may occur spontaneously.
late mofetil in eight cases of steroid-dependent AIN, defined The first description of atheroembolic disease was by
as worsening kidney function with steroid withdrawal, Panum in 1883, but the nature of cholesterol embolization
even after a 6-month course of corticosteroids. Only two of was described by Flory in 1945 [34].
the eight cases were thought to be drug-induced, with the The disease is often under recognized. From various
remaining cases being idiopathic or associated with systemic autopsy studies, the incidence was about 1% but increased
diseases such as sarcoidosis or collagen vascular disease. to 4.0–6.5% in people 60 years or older [35]. There is usually
Treatment with mycophenolate mofetil resulted in a stabili- an inciting event leading to plaque destabilization and dis-
zation or slight improvement in kidney function (mean cre- tal showering of lipid debris. Plaque destabilization may
atinine decreased from 2.3 to 1.6 mg/dl); however, an average occur from vessel wall trauma from vascular surgery or
of 2 years of therapy was required [31]. More recently a case endovascular procedures of which coronary angiography is
report demonstrated benefit of antithymocyte globulin in a the most common iatrogenic procedure causing atheroem-
patient on RRT as the result of steroid refractory AIN who bolism. Although less common as an inciting factor, sys-
achieved independence from hemodialysis following temic anticoagulation and thrombolytics are thought to
treatment [32]. prevent protective thrombus formation over ulcerated
Given these data, a reasonable treatment strategy would plaques, thereby increasing risk of atheroembolism.
be to reserve corticosteroids for patients with idiopathic In atheroembolism, the cholesterol-containing debris
AIN, systemic diseases for which corticosteroids have a lodge in small arterioles, inciting thrombus formation, and
proven role (e.g. sarcoidosis, Sjögren’s syndrome, vascu- causing distal ischemia and infarction. Within 5 days there
litides), or cases with characteristics associated with worse may be recanalization of the thrombus, but an inflamma-
renal prognosis. The latter includes delayed onset of kid- tory foreign body arteritis ensues. Because the cholesterol
ney recovery after removal of the inciting agent (>1 week), is not soluble and it is never successfully cleared by phago-
prolonged exposure to the offending agent (>2–3 weeks), cytes, inflammation persists and leads to progressive
fibrosis and later obliteration of the lumen of the small ves- Table 12.3 Clinical findings in patients with renal
sel. With involvement of multiple arterioles, the result is atheroembolic disease and frequency of occurrence.
persistent, patchy ischemia of several nephrons, instigating
progressive renal insufficiency. Frequency of
Clinical finding occurrence (%)
Clinical Features and Laboratory Findings Signs and symptoms

New-onset, accelerated, or labile hypertension 78
Atheroembolic renal disease is preeminently a disease of
Skin findings (cyanotic or ulcerated digits or 80
the elderly (mean age 66–70 years), with a male predomi- scrotum, livedo reticularis on back or lower
nance (4 : 1). It has a higher incidence in whites than extremities, nodules, less commonly purpura)
African Americans (30 : 1) [36]. It is possible that this dis- GI symptoms (nausea, abdominal pain, GI 24
crepancy represents under-recognition of the disease in the bleeding)
African American population, as the characteristic skin CNS symptom (focal neurologic deficits, 11
changes may by missed in dark-skinned individuals. The progressive dementia)
major risk factors include age more than 60 years, male Retinal emboli (Hollenhorst plaque on 19
gender, comorbidities like diabetes mellitus and hyperten- fundoscopy)
sion, and tobacco use. Also, patients with significant aortic Fevers Uncommon
atherosclerosis have a higher incidence of cholesterol Laboratory findings [reference]
embolic disease [34]. The onset of disease is quite variable, Microscopic hematuria [38, 39] 53
ranging from 3 days to 3 months (as the embolic shower is Mild proteinuria (rarely nephrotic associated 9
often unpredictable). As mentioned earlier, in addition to with a secondary collapsing FSGS or
the kidneys, any organ system may be involved, including accelerated hypertension) [39, 40]
skin, gastrointestinal, and brain. Transient eosinophilia Eosinophiluria (by Hansel’s stain) 88
(seen in 48%) and elevated erythrocyte sedimentation rate Peripheral blood eosinophilia [41] 48–88%
and occasionally hypocomplementemia may be pre- Hypocomplementemia [41] <15–12
sent [37]. The characteristic skin changes include livedo
Various markers of ischemic organ injury Not
reticularis or “blue toe” syndrome. These constellations of (elevated creatinine kinase, amylase/lipase, uncommon
clinical and laboratory features often mimic a systemic vas- transaminases)
culitis (Table 12.3). Elevated erythrocyte sedimentation rate and/ Very common
Renal histology reveals empty clefts in arcuate and inter- or C-reactive protein
lobular arteries and, less commonly, afferent arterioles. Renal artery stenosis by Doppler or 44
These result from the dissolution of lipid from these sites angiography [38, 42]
by the fixation process leaving voids at the sites of choles-
CK, creatine kinase; CNS, central nervous system; FSGS, focal and
terol crystals. Depending on the age of the lesion, varying segmental glomerulosclerosis; GI, gastrointestinal.
degrees of arteritis and intimal fibrosis can be seen Unless otherwise indicated, frequency data come from combining
(Figure 12.2). In severe, acute cases there may be signs of three case series of clinically significant kidney disease totaling 171
acute tubular necrosis (ATN). Late in evolution, patchy glo- patients.
Source: Belenfant et al. [38], Thadhani et al. [39], and Scolari et al. [42].
merular sclerosis and tubular atrophy may be visualized in
areas supplied by the affected vessels. Similar arteriolar
inflammation or fibrosis can be found in other tissues, which represents repeated smaller embolic showers and
especially the muscle, gastrointestinal tract, and skin. their subsequent inflammatory endothelial response. Thus,
Biopsy of skin lesions may have the highest diagnostic there is a stepwise deterioration in kidney function averag-
yield, with a reported sensitivity of 92% [43]. ing 3–8 weeks. The third entity involves a slowly progressing
renal insufficiency secondary to chronic ischemia mimick-
ing hypertensive nephrosclerosis or ischemic renovascular
Course and Management
disease. This is often underdiagnosed because of the paucity
Three patterns of disease evolution are apparent in athero- of systemic organ involvement and the failure to obtain kid-
embolic renal disease [44]. An acute course (20–30% of ney biopsy.
cases) resulting from a massive embolic load is characterized Progression to dialysis dependence occurs in one-third
by an abrupt deterioration in kidney function and multior- of the patients who survive the initial insult. Preexisting
gan involvement within a week after the inciting event. CKD and a history of significant claudication confers
More common, however, is the subacute course (56%), greater risk for this outcome. Some mild improvement in
AKI Associated with Malignancie 171
embolic disease. Other studies have used high doses of

steroids with variable benefits. Prospective studies have
shown beneficial effects of statins in the progression to end
stage renal disease (ESRD). Observations have shown that
statins help to decrease lipid level, stabilize plaques, and
are anti-inflammatory. Lipoprotein aphesis is known to sig-
nificantly reduced cholesterol levels and has been recom-
mended for use in patients with high risk [47, 48]. Surgical
procedures such as endoluminal stenting could provide a
definitive treatment, although this is often not feasible
because the disrupted plaque is difficult to identify or the
patient may be too fragile to undergo any surgical interven-
tion. If endovascular procedures are required in high-risk
patients, strategies such as “no-touch” techniques to mini-
Figure 12.2 Renal atheroembolic disease, hematoxylin and
eosin stain. Arrows represent cholesterol clefts remaining in an mize plaque disruption, and embolic protection devices
intralobular artery where lipid had been present prior to the may help to prevent the development of atheroembolic
fixation process. The early inflammatory arteritis composed of disease [49–51].
eosinophils, neutrophils, and macrophages is later replaced by a Atheroembolic disease generally has poor prognosis as it
giant cell foreign body reaction (arrowhead), intimal
proliferation, and fibrosis. Source: Courtesy of Helen Liapis MD, is associated with multiorgan system involvement, often
Washington University School of Medicine. with irreversible damage. About 30–55% of patients with
acute/subacute onset of kidney failure need dialysis, of
kidney function can occur with time, from progressive whom 21–28% have renal recovery and 23–32% progress to
hyperfiltration by the remaining nephrons. As a result, ESRD. Preexisting CKD and hypertension are determi-
21–32% of patients on chronic dialysis from atheroembolic nants for progression to ESRD [39, 42].
renal disease can recover enough kidney function to stop
dialysis [38, 41, 45]. Early studies revealed a grim outcome
for patients with acute, multivisceral atheroembolic dis-
ease, with 1-year survival rates of 13–36% [39, 46]. Death
AKI Associated with Malignancies
usually occurs from cardiac ischemia, heart failure, stroke,
Cast Nephropathy
or gastrointestinal ischemia with malnutrition [39, 44, 45].
An aggressive supportive regimen specifically targeting Definitions and Etiology
these mechanisms of mortality was prospectively evalu- Multiple myeloma (MM) is a hematological malignancy
ated in a population with acute multivisceral atheroem- characterized by abnormal proliferation of monoclonal
bolic disease [38]. plasma cells in the bone marrow. It is the second most
The principles of treatment for atheroembolic disease are common hematological malignancy, diagnosed in approx-
intended to decrease the ischemic damage and prevent imately 13% of all hematological malignancies, and the
recurrent embolization. The mainstay of treatment is often American Cancer Society estimates a lifetime risk of
preventative and supportive. The major preventative meas- 0.76% in the United States. The definition of renal impair-
ures include avoiding anticoagulation (if feasible after care- ment in MM, as defined by the International Myeloma
fully considering risks and benefits) and avoiding any further Working Group (IMWG) criteria, includes a serum creati-
invasive procedures that might trigger further plaque dis- nine level of 2 mg/dl or creatinine clearance of <40 ml/
ruption. To delay progression, blood pressure control and min [52]. Based on these criteria, kidney involvement can
judicious use of statins and renin-angiotensin-aldosterone range from 20% to 30% of patients with MM. AKI associ-
system (RAAS) blockers should be considered. Nutritional ated with plasma cell dyscrasias can result from (i) prere-
replacement, including parenteral replacements, may be nal AKI secondary to hypercalcemia and/or volume
needed in settings of intestinal ischemia. With advanced depletion; (ii) glomerular disease from light-chain deposi-
kidney failure, dialysis should be considered. tion disease or amyloidosis; and (iii) tubulointerstitial dis-
Several adjunctive therapies to improve renal outcome ease from cast nephropathy and drug-induced AKI (from
have been attempted in cases of atheroembolic renal dis- intravenous bisphosphonates). Other etiologies of AKI in
ease. Belenfast et al. [38] determined that low-dose corti- MM include glomerular diseases (e.g. membranoprolifer-
costeroids (0.3 mg/kg of prednisone) improved nutritional ative glomerulonephritis, light-chain deposition disease,
intake and symptoms in 18 patients with recurrent athero- amyloidosis) associated with paraproteinemia and, to a
lesser extent, tumor lysis syndrome (TLS), which will be

discussed separately. In the updated Diagnostic Criteria
and Staging System for MM only light-chain cast nephrop-
athy is defined as myeloma defining illness [53].
Myeloma cells synthesize free light chains (FLCs) which
are freely filtered by the glomerulus. These proteins are
then reabsorbed by the megalin and cubulin receptors of
the proximal convoluted tubule [54]. When the filtered
load exceeds the maximal reabsorption of the receptors,
they are excreted in the urine and referred to as Bence–
Jones proteins. The FLCs may interact with Tamm–Horsfall
protein (THP) causing tubular obstruction inducing
inflammation and resultant interstitial nephritis and
fibrosis [55].
Clinical Features and Laboratory Findings Figure 12.3 Myeloma cast nephropathy, with hematoxylin and
eosin stain. There is a characteristic eosinophilic, “corrugated
Cast nephropathy is suspected when patients present with paper” appearance of this cast that is adherent to the wall of the
rapidly declining kidney function in the setting of FLCs tubule. There is resulting tubular cell toxicity and inflammation,
with an interstitial infiltrate and cells engulfing the cast at its
detected by plasma or urine immunofixation or by serum point of adherence (arrow). Source: Courtesy of Helen Liapis, MD,
FLC measurement. IMWG recommends a 24-hour urine Washington University School of Medicine.
protein electrophoresis and a serum FLC assay. In the set-
ting of reduced kidney function, the serum FLC may be
Newer chemotherapeutic agents have increased overall
elevated, but not in a clonal pattern. If a clonal serum FLC
survival in MM. Renal protective chemotherapy with tha-
is >500 mg/l, quantitative urinary albumin excretion
lidomide or bortezomib aims at rapidly decreasing FLC
should be assessed. In glomerular causes of AKI, the pre-
concentrations. Bortezomib-based regimens have been
dominant urinary protein is albumin; but when albumin
associated with improved kidney function and are the
10% of total urinary protein, the diagnosis is likely to be
treatment of choice [54, 56, 57]. A Cochrane meta-analysis
cast nephropathy and kidney biopsy can be avoided. If
with 16 studies and 5626 patients suggested better survival
quantitative urinary light chains are <1 g/day, an alternate
rates and longer time without progression with bortezomib.
diagnosis should be considered and a kidney biopsy may be
Metabolized by liver, it can be safely administered to
warranted.
patients with kidney failure. Thalidomide and lenalido-
The definite diagnosis is obtained from kidney biopsy. On
mide are also targeted chemotherapy for MM. In one study
light microscopy, cast nephropathy is manifested by the
of 133 patients, renal recovery was noted in 77%, 55%, and
presence of large lamellated eosinophilic casts in the distal
43% of patients treated with bortezomib, thalidomide, and
convoluted and collecting tubules. Features of interstitial
lenalidomide, respectively [57]. In a multivariate analysis,
inflammation and nephritis are also seen (Figure 12.3). The
bortezomib had superior results with respect to probability
casts are comprised primarily of monoclonal light chains
of response as well as the median response time compared
but also contain albumin, THP, fibrin, and polyclonal light
to the other two agents. A retrospective cohort study ana-
chains.
lyzing autologous stem cell transplant and renal recovery
showed that 14 out of 15 patients with advanced kidney
failure, who received transplant remained dialysis depend-
Course and Treatment
ent despite hematological recovery; however, it is impor-
Treatment of cast nephropathy includes general measures tant to note that having kidney failure did not exclude them
and disease-specific treatments. General measures include from this curative therapy [58].
adequate volume expansion, correction of hypercalcemia The idea that therapeutic plasma exchange (TPE)
and hyperuricemia, discontinuation of potential nephro- would be effective in removal of FLC stems from the fact
toxic medications, such as NSAIDs and angiotensin con- that FLC have small molecular masses. Trials evaluating
verting enzyme inhibitors (ACEI), and angiotensin receptor the role of TPE in cast nephropathy were performed
blockers (ARBs), loop diuretics, and avoidance of iodinated before the advent of newer bortezomib-based regimens.
intravascular contrast. Burnette et al. analyzed 14 patients from 2004 to 2010 who
Pathophysiolog 173
received TPE and bortezomib, and showed full renal causing hyperuricemia, hyperkalemia, hyperphos-
recovery in 43% of patients in 6 months [59]. Given the phatemia, and hypocalcemia. Multiorgan failure, includ-
lack of sufficient data from current literature, the role of ing AKI, life-threatening cardiac arrhythmias, seizures,
TPE in cast nephropathy remains nebulous and cannot be and death, is known to occur. The Cairo–Bishop diagnos-
recommended. Two recent trials – EuLite and tic criteria for laboratory and clinical TLS are shown in
MYRE – evaluated the role of high-cut-off hemodialysis Table 12.4.
(HCO-HD) along with concomitant chemotherapy in the TLS is most commonly described in malignancies with
treatment of biopsy confirmed cast nephropathy [60, 61]. large tumor burden: non-Hodgkin lymphoma, especially
In both trials, the primary outcome of dialysis independ- Burkitt’s lymphoma, acute and chronic leukemias, and
ence at 30 days was not significantly different between the MM [67–69]. Certain solid organ tumors like germ cell
HCO-HD arm versus the control groups. In the MYRE tumors, neuroblastoma, breast, and small cell lung cancer,
study, dialysis independence at 6 months was superior to especially when associated with widespread metastasis,
conventional treatment alone (35% vs. 57%, P = 0.04) [61, also predispose to TLS. Although previously described with
62]. It should be noted that there was a higher incidence chemotherapy, other forms of treatment, including corti-
of death due to infections in the HCO-HD arm in the costeroids, tamoxifen, radiotherapy, and newer targeted
MYRE study, but 11% of patients in the HCO-HD arm therapy with biologicals, can also cause TLS. Elevated
compared to 4% in the conventional HD arm had infec- serum lactate dehydrogenase (LDH) is an additional risk
tion at baseline. There was no difference in infectious factor predisposing to TLS [70].
complications in the EuLite trial. HCO-HD is not FDA
approved for treatment of cast nephropathy in the United
States. The IMWG recommends that HCO-HD can be Pathophysiology
considered as an adjunct to bortezomib-based chemo-
therapy in patients with MM and severe kidney impair- Uric acid is the end product of the metabolism of purine
ment (grade B evidence) [63]. nucleotides in humans and higher primates. Uric acid is
freely filtered by the glomerulus, but in contrast to previous
studies, recent data suggest that our understanding of the
AKI Secondary to Bisphosphonate Therapy
tubular handling of uric acid is incomplete [71]. Uric acid
Intravenous bisphosphonates are widely used in the treat- is actively reabsorbed and secreted at the proximal tubule
ment of hypercalcemia of malignancy. Both zoledronate via complicated interplay among various transport path-
and pamidronate have been associated with AKI. Acute ways involving urate transporter 1 (URAT1), glucose trans-
tubular injury and collapsing focal segmental glomerulo- porter 9 (GLUT9), and organic anion transporters (OATs),
sclerosis have been described [64]. The 2015 European among others [71]. Uric acid crystallization occurs when
Myeloma network guidelines [65] recommend measuring concentration in the tubules exceeds the solubility, and this
creatinine clearance, serum electrolytes, and urine albu- usually occurs in setting of low urine volume and acidic
min in all patients receiving bisphosphonates (grade 1A).
These drugs should be avoided if CrCl <30 mg/dl (grade Table 12.4 Cairo–Bishop definition of tumor lysis
1A). Dose reduction is necessary for zoledronate with no syndrome [66].
change in infusion time (grade1A) whereas pamidronate
should be given over a 4-hour infusion (grade 1C). Laboratory tumor Two or more occurring within 3 days
lysis syndrome before or 7 days after chemotherapy
Clodronate can be used as alternative in patients with CrCl
(LTLS) ●● Uric acid 8 mg/dl or 25% increase
>12 ml/min (Grade 2C). To prevent hypocalcemia, calcium
from baseline
and vitamin D supplementation should be administered. ●● Potassium 6 mEq/dl or 25% increase
(Grade 1A). from baseline
●● Phosphorus 4.5 mg/dl or 25%
increase from baseline
Tumor Lysis Syndrome ●● Calcium 7mg/dl or 25% decrease
TLS is one of the most common hematological emergen- from baseline

cies. TLS describes an array of biochemical abnormalities Clinical tumor lysis LTLS with one or more of the following:
syndrome (CTLS) serum creatinine 1.5 upper limit of
secondary to cell lysis that occur spontaneously or after
normal cardiac arrhythmia or death
initiation of cytotoxic chemotherapy. The cell damage seizures
releases intracytoplasmic molecules, including proteins,
electrolytes, and nucleic acids, into systemic circulation, Source: Cairo and Bishop [66].
pH. Hyperphosphatemia results when tumor cells with recombinant uricase enzyme that converts uric acid into
high metabolic rates and turnover die, releasing high levels allantoin, which is 5–10 times more soluble in urine than
of intracellular phosphorus into bloodstream, overwhelm- uric acid. The dosing approved by FDA is 0.2 mg/kg in a
ing the kidney’s capacity to eliminate it. In spontaneous 30-minute intravenous infusion once daily for up to
TLS, severe hyperphosphatemia is less likely, since the sur- 5 days [74, 75]. Rasburicase does not require renal dosing.
viving tumor cells, with their high metabolic rate, rapidly It can cause methemoglobinemia and hemolysis in patients
recycle released phosphate. Phosphorus binds with cal- with glucose-6-phosphate dehydrogenase deficiency hence
cium cations, causing hypocalcemia, and calcium phos- all patients should be checked for enzyme activity prior to
phate complexes can precipitate in the kidney and other starting rasburicase [76].
tissues. Patients can develop signs and symptoms of hypoc- Most patients with AKI will not require dialysis and
alcemia, including tetany and seizures. AKI in TLS is thus will recover kidney function, although patient and renal
thought to be secondary to uric acid and calcium phos- prognosis may be worse with spontaneous TLS. Before
phate crystal nephropathy. In addition to the crystalline the availability of rasburicase, hemodialysis was often ini-
pathway, vasoconstriction and inflammation also play a tiated early in order to rapidly reduce uric acid levels,
role in the pathogenesis of AKI. with a 50% reduction occurring after 6 hours of hemodi-
alysis (clearance of 70–145 ml/min) [77]. Rasburicase
results in more prompt reduction of uric acid and elimi-
Management nates this indication for RRT. However, when significant
hyperphosphatemia is present, early RRT may still be
The management of TLS can be divided into preventative required along with noncalcium-based phosphate binders
and therapeutic measures. Patients who are at high risk to prevent further phosphate precipitation and/or hypoc-
for TLS should be closely monitored for electrolyte abnor- alcemia. Treatment of hypocalcemia with intravenous
malities and metabolic acidosis. General measures such calcium without first lowering the phosphorus should
as avoidance of nephrotoxic drugs including NSAIDs and be reserved for symptomatic patients or those with
radiocontrast is necessary. Volume expansion with the EKG changes as calcium administration in the setting of
use of crystalloids to maintain a urine output of 2 ml/ hyperphosphatemia can cause calcium phosphate precip-
kg/h [72] helps ensure adequate glomerular filtration and itation. Even though phosphorus clearance is high
high tubular flow rates, thereby enhancing excretion of (approximately 50–90 ml/min) with intermittent hemodi-
uric acid, phosphorus, and potassium, and minimizing alysis (IHD), the short duration of IHD (usually 4 hours/
the risk of uric acid and calcium phosphate precipitation. day) is not sufficient to achieve negative phosphorus bal-
Even though uric acid is soluble in alkaline urine, urine ance. The longer duration of treatment offered by con-
alkalization can decrease the solubility of calcium phos- tinuous or prolonged intermittent RRT is more effective
phate and increase the risk for calcium phosphate for phosphorus clearance and should be considered for
nephropathy. Ideally this should be initiated 2 days prior patients with TLS [78].
to the initiation of chemotherapy. Also, alkalosis leads to
increased binding of calcium to albumin, reducing ion-
Crystalline Nephropathies
ized calcium, and hence routine alkalization of urine is
not recommended [69, 72]. Definition, Etiology, and Pathogenesis
Allopurinol (which gets converted to oxypurinol in vivo) Crystalline nephropathies describe the kidney injury that
is a competitive inhibitor of the enzyme xanthine oxidase. results from the intratubular precipitation of substances.
Oxypurinol competes with purines and decreases their Uric acid nephropathy was discussed earlier and this
conversion to uric acid. Hence it is most effective in the section will focus on other crystalline nephropathies.
prevention of hyperuricemia [73]. However, the drug also Intratubular crystal formation and deposition of a sub-
inhibits the conversion of xanthine to uric acid, hence stance is promoted by three mechanisms: (i) high concen-
there is a risk of formation of xanthine crystal in the renal trations of the offending agent in the tubular fluid; (ii)
tubules [70]. A large study including 1172 adults and chil- prolonged intratubular transit time; and (iii) decreased
dren, treated with intravenous allopurinol, was found to solubility [79]. The first two mechanisms occur in the set-
decrease uric acid level in 57% of adults when used for ting of decreased effective circulating volume, which
treatment and by 93% when used prophylactically [73]. results in proximal tubular sodium and water reabsorption,
Febuxostat is a newer xanthine oxidase inhibitor that is resulting in high concentration of the offending agent in
metabolized in the liver and hence can be used safely in the distal tubule along with decreased distal flow rates.
patients with kidney insufficiency [70]. Rasburicase is a Underlying CKD is an important risk factor for the crystalline
Managemen 175
nephropathies. In CKD, the amount of compound excreted and hence does not rule out crystal nephropathy. Conversely,
per remaining functioning nephron is increased, raising the presence of crystals does not prove their pathogenic role
the likelihood of intratubular precipitation. The third because calcium oxalate, calcium phosphate, and uric acid
mechanism, decreased solubility, is often dependent on the crystalluria may be seen in normal individuals. Furthermore,
distal tubular fluid pH. Compounds with pKa < 7, such as crystalluria has been found in patients without AKI in fre-
uric acid, calcium oxalate, sulfonamides, methotrexate, quencies varying from 100% of healthy individuals receiv-
and triamterene, tend to precipitate in acidic urine, ing a single 100-mg dose of triamterene and ascorbic acid
whereas compounds with pKa > 7, such as indinavir and (for urinary acidification) to 17% of all patients taking
calcium phosphate, tend to precipitate in alkaline urine. indinavir [90].
Once crystals are formed in the tubular lumen, the tubular Ultrasound may reveal bilaterally enlarged and echo-
epithelial cells attempt to phagocytose and digest the crys- genic kidneys and can identify concomitant macroscopic
tals with proteolytic enzymes, resulting in direct cellular lithiasis. Kidney biopsy is sometimes required to make a
toxicity and necrosis. Crystal-induced toxicity can also lead definitive diagnosis. Light microscopy may reveal crystal-
to significant peritubular inflammation, sometimes mim- line deposits (usually in the distal tubules), with a sur-
icking AIN [80, 81]. rounding interstitial infiltrate that may contain giant cells
Inadequate dose reduction for the level of kidney func- as part of a foreign body reaction. Evidence of ATN can
tion is frequent, leading to increased drug exposure. The also be present, secondary to direct tubular cell toxicity.
clinical contexts in which the more common crystalline Polarized microscopy may demonstrate birefringence,
nephropathies occur are summarized in Table 12.5. Besides depending on the offending agent (Figure 12.5a,b). A sum-
TLS, other causes of crystalline nephropathy include cal- mary of the characteristic laboratory findings is outlined in
cium oxalate, acyclovir, sulfonamides, methotrexate, indi- Table 12.5.
navir, and triamterene. Ethylene glycol intoxication is
discussed separately due to its unique pathogenic mecha-
Course and Management
nisms and treatment strategies.
In most cases of crystalline nephropathy, the renal progno-
sis is excellent if further exposure to the precipitating sub-
Clinical Features
stance can be avoided. In drug-related crystalline
Extensive crystal deposition results in pain from distention nephropathies, recovery of kidney function is expected to
of the renal capsule and is similar in nature to the pain due occur within days to weeks after cessation or dose reduc-
to nephrolithiasis. Oliguria is uncommon unless there is tion of the offending drug. An exception is the use of
bilateral obstruction. In certain cases (especially with indi- phosphate-containing colonoscopy preparation, causing
navir and sulfonamides), macroscopic lithiasis and its phosphate nephropathy. Although there was likely selec-
characteristic presenting symptoms can exist alone or con- tion bias for more severe cases, the largest series character-
comitantly with intratubular crystal nephropathy [88, 90]. izing this disease found that none of 21 affected patients
Precipitation of calcium with phosphates and oxalates can returned to their previous level of kidney function and four
lead to symptomatic hypocalcemia, leading to paresthesia, progressed to require chronic dialysis [82].
tetany, and lethargy. High levels of acyclovir seen in AKI The mainstay in preventing crystalline nephropathy is
can lead to hallucinations, delirium, and myoclonus. avoidance of the two most frequent predisposing factors:
Methotrexate can cause stroke-like symptoms (not dose- volume depletion and drug overdosing relative to the
related) as well as nausea, rash, and mucositis (usually level of kidney function. Establishing a brisk urine output
dose-related). (e.g. 100–150 ml/h or greater) in high-risk patients is
extremely important. For substances with pKa < 7 (e.g.
uric acid, calcium oxalate, sulfonamides, methotrexate,
Laboratory Findings
triamterene) urinary alkalization by administering intra-
Urinalysis and microscopic examination of the urine sedi- venous isotonic bicarbonate solutions or oral citrate can
ment are often nonspecific, generally revealing hematuria, be considered. Urine pH should be periodically followed
pyuria, and mild proteinuria. Although the offending sub- to ensure an appropriate level of alkalization to achieve
stances have unique crystal morphologies on urine micros- pH > 7.15. Excess alkalization of urine can lead to precipi-
copy, finding urine crystals may not be independently tation of calcium phosphate crystals and should be
diagnostic. Urine microscopic images of some crystals are avoided. Acetazolamide may be added if a metabolic alka-
shown in Figure 12.4a–e. Absence of crystals on urine losis ensues. Attempting to acidify the urine to increase
microscopy can occur when there is tubular obstruction the solubility of weakly basic compounds is dangerous
Table 12.5 Causes of crystalline nephropathy, their distinctive clinical and laboratory features, and strategies for prevention and treatment.
Inciting agent
[reference] Clinical context Laboratory findings Prevention and treatment Disease course
Phosphate [82] TLS, especially post treatment form Crystalluria with weakly birefringent, long Prevention Incidence of CKD
Phosphasoda bowel prep with risk prisms, often in rosettes Forced diuresisa higher in cases caused
increased by ACEi/diuretic use, Hyperphosphatemia out of proportion to Noncalcium-based phosphate binders by phosphasoda
advanced age, CKD, and/or female kidney insufficiency Treatment
gender Hypocalcemia Forced diuresisa
●● Rhabdomyolysis (rare) Hyperkalemia out of proportion to kidney Noncalcium-based phosphate binders
insufficiency
Avoid treatment of hypocalcemia unless
High LDH in TLS, rhabdomyolysis, and symptomatic or ECG changes present
hemolysis
Consider early initiation of RRT,
Kidney biopsy: von Kossa-positive crystals especially CRRT
Uric acid [83, 84] TLS, especially spontaneous form Crystalluria with brownish, strongly Prevention Full renal recovery
Rhabdomyolysis (rare) birefringent, rhomboid plates, rosettes, or Forced diuresisa expected in the absence
HGPRT deficiency (rare) needles Allopurinol or rasburicase of other insults; higher
UA >15 mg/dl (peaking 2–3 days after mortality in
Treatment spontaneous TLS
initiation of treatment in TLS)
Rasburicase
Urine UA/urine Cr >1 (<0.75 suggests
alternative diagnosis) RRT (if rasburicase not available)
Hyperkalemia out of proportion to kidney
insufficiency
Elevated LDH in TLS and rhabdomyolysis
Oxalate [85, 86] EG poisoning Crystalluria with birefringent monohydrate Prevention Full renal recovery
High-dose i.v. ascorbic acid–xylitol– needles or dihydrate envelope shapes Forced diuresisa occurred in 6/9
sorbitol infusions (rare) appearing >4 h after EG ingestion Consider urine alkalinization patients; duration and
Primary hyperoxaluria (rare) Hypocalcemia severity of AKI is
Fomepizole for high-risk EG ingestionb predicted by severity of
In EG poisoning, osmolal gap >10 mOsm/l (ethanol is second line)
and detectable serum and urine EG early with acidosis at presentation
i.v. thiamine, magnesium, and pyridoxine
subsequent development of severe anion gap in alcoholic patients
acidosis
Treatment
Kidney biopsy: silver nitrate/rubeanic acid
stain-positive crystals Forced diuresisa
Consider urine alkalinization
Avoid treatment of hypocalcemia unless
symptomatic or ECG changes present
Fomepizole (ethanol is second line) if EG
level >20 mg/dl
Early initiation of RRT, especially if EG
>50 mg/dl and renal insufficiency or
acidosis present
0005152400.INDD 176 09-12-2022 15:22:16

Acyclovir [87] Associated with intravenous dosing Crystalluria with birefringent needles with Prevention Full renal recovery
11% develop AKI with high-dose occasional engulfment by WBC Forced diuresisa expected
rapid i.v. bolus Increase time of i.v. infusion to 1 hour
Treatment
Forced diuresisa
Decreasing the dose is sufficient in
40–50% of patients
Indinavir [88] 18.6% develop AKI, especially with Crystalluria with birefringent plates, fans, or Treatment AKI is easily reversible
longer treatment, smaller body size, starbursts Forced diuresisa as opposed to the
and concurrent TMP-SMX Isosthenuria common Urologic consultation if concomitant slowly progressive
CT with i.v. contrast shows wedge-shaped macroscopic obstruction present (not injury seen in
perfusion defects in up to 50% uncommon) indinavir-induced AIN
Methotrexate [89] 2–4% develop AKI with high-dose i.v. Crystalluria with amorphous yellow casts Prevention Median time to renal
therapy; increased risk in older High serum methotrexate level Forced diuresisa recovery to 2–3 weeks
patients and with concomitant use of Anemia, leukopenia, or thrombocytopenia Urinary alkalinization to pH 8
NSAID or other highly protein- possible
bound drugs which increase free Treatment
methotrexate levels Forced diuresisa
Urinary alkalinization to pH 8
Leucovorin rescue ± thymidine for
extrarenal toxicity until methotrexate
level <0.05 μmol/l
Consider carboxypeptidase G2 (>98% fall
in levels by 15 minutes) vs. daily 6-hour
high-flux hemodialysis (clears drug in
5.5 days)
Sulfonamides [90] Usually with high-dose therapy Crystalluria with variable shapes from shocks Treatment Full renal recovery is
More common with sulfadiazine of wheat to spheres Forced diuresisa prompt, sometimes
(1.9–7.5% incidence in AIDS Positive lignin test (orange urine upon mixing Urine alkalinization to pH > 7.1 within hours, but
patients) with 10% hydrochloric acid) median 6 days
Dose reduction usually sufficient
Low serum albumin may increase Densities in parenchyma and in collection Urologic consultation if macroscopic
risk from higher free drug levels system on ultrasound are common obstruction present (not uncommon)
Triamterene [79] Rare; must distinguish from the Crystalluria with birefringent orange casts Treatment Full renal recovery is
much more common AIN, or AKI and spheres Forced diuresisa expected
with concomitant NSAID use Hyperkalemia out of proportion to renal Urine alkalinization to pH > 7.5
insufficiency May require urologic consultation for the
more common triamterene stone if
obstructed
CRRT, continuous renal replacement therapy; CT, computerized tomography; ECG, electrocardiogram; EG, ethylene glycol; HGPRT, hypoxanthine-guanine phosphoribosyl transferase; i.v.,
intravenous; LDH, lactate dehydrogenase; TLS, tumor lysis syndrome; TMP-SMX, trimethoprim-sulfamethoxazole; UA, uric acid.
a
Forced diuresis refers to aggressive hydration with judicious use of diuretics to achieve and maintain brisk urine flow rate.
b
Patients at high risk for organ dysfunction in EG poisoning are those with serum EG of >20 mg/dl, OR with known recent ingestion of EG with osmolal gap >10 mOsm/l, OR strong
suspicion of recent ingestion and two of the following: pH <7.3, HCO3 < 20 mEq/l, osmolal gap >10 mOsm/l, and urinary oxalate crystals [85].
0005152400.INDD 177 09-12-2022 15:22:16

(b1)
(a)
(b2)
(c)
(d) (e)
Figure 12.4 (a)–(e) Crystals on urine microscopy. (a) Calcium oxalate crystals may appear in the monohydrate (dumbbell shaped) and
dihydrate (envelope shaped) form as shown here (brightfield illumination, original magnification 400×). (b) Uric acid crystals can be
seen in various forms. The rhombic and rosette forms are shown in 4B:1 and the barrel-shaped forms are shown in 4B:2 (brightfield
illumination, magnification 400×). Source: Images (a), 4B:1 and 4B:2: courtesy of Jay Seltzer MD, St. Louis Kidney Consultants, St. Louis,
MO. (c) Acyclovir crystals are typically needle shaped, as shown here (brightfield illumination, magnification 400×). (D) Sulfadiazine
crystals can appear as striated shells or “shocks of wheat” (brightfield illumination, magnification 400×). Source: Images (c) and (d):
courtesy of José Antonio Tesser Poloni, Urinalysis Section – Carlos Franco Voegeli Clinical Analysis Laboratory, Santa Casa de
Misericórdia de Porto Alegre, Brazil. (e) Indinavir crystals are shown here as rectangular plates with irregular borders and occasional
tapering (brightfield illumination, magnification 400×).
Ethylene Glycol Poisonin 179
High-dose methotrexate causing AKI can in turn

increase its serum levels, causing acute systemic metho-
trexate toxicity from decreased urinary clearance. In
addition to large volume hydration and forced alkaline
diuresis, leucovorin can be administered to decrease the
cytotoxicity of the drug. Glucarpidase (carboxypeptidase)
was approved in 2012 for MTX concentrations >1 umol/l
to decrease its systemic toxicity if the above measures are
unsuccessful in bringing down the levels [91]. Studies
have shown daily hemodialysis with high flux dialyzers
and sessions of 4–6 hours in duration are effective in
removing methotrexate [89]. Again, evidence for
improvement in renal outcome with these maneuvers is
(a)
lacking. See Table 12.5 for details on specific manage-
ment strategies.
Ethylene Glycol Poisoning
Ethylene glycol (or antifreeze) is metabolized by the

hepatic alcohol dehydrogenase into four toxic metabolites:
glycoaldehyde, glycolic acid, glyoxylic acid, and oxalic acid.
Accumulation of organic anions (glycolate, glyoxylate, and
oxalate) leads to a severe anion gap metabolic acidosis.
These compounds, especially glycolic acid, are direct cel-
lular toxins and cause multiorgan dysfunction, including
(b) cardiotoxicity, ATN, and nervous system depression. In
Figure 12.5 (a) and (b) Oxalate nephropathy, hematoxylin and addition, oxalate precipitates with calcium in several
eosin stain. This patient developed renal failure in the setting of tissues, including the renal tubules, causing crystalline
primary hyperoxaluria. (a) Note the distortion of the tubular nephropathy (Figure 12.4a,b).
architecture associated with intratubular deposits of calcium The clinical manifestations of ethylene glycol intoxica-
oxalate (arrow heads). There is an interstitial infiltrate along
interstitial fibrosis consistent with disease chronicity. (b) Under tion evolve over time as the alcohol is metabolized.
polarized light the deposits exhibit birefringence (black arrows). During the first 30 minutes to 12 hours, ethylene glycol
Source: Courtesy of Dr. Rosa Davila, Washington University causes inebriation, with progression to seizures or coma.
School of Medicine.
Twelve to 36 hours post ingestion, profound acidosis
ensues as the concentration of the organic acid increases,
and not recommended. These preventive strategies are leading to Kussumals breathing and cardiorespiratory
based on underlying pathophysiologic mechanisms, and failure. Twenty-four to 72 hours later, the oxalate end
evidence proving reduced renal complications with these product accumulates in tissues, including the kidneys,
measures is lacking, except perhaps with high-dose meth- resulting in AKI. The timeline is extended when there is
otrexate administration. co-ingestion of ethanol, which acts as a competitive
Treatment of established AKI consists of discontinua- inhibitor of alcohol dehydrogenase. The focus of treat-
tion of the drug and, if nonoliguric and in the absence of ment in ethylene glycol toxicity involves decreasing the
volume overload, applying many of the same principles concentration of its toxic metabolites. This can be
used in prevention: forced diuresis (aggressive volume achieved by (i) reducing organic acid formation by use of
expansion with judicious use of diuretics) and, for weak competitive alcohol dehydrogenase inhibitors, such as
acids, urinary alkalization. In the setting of AKI, moderate fomepizole or ethanol; (ii) increasing metabolite clear-
to large doses of diuretics may be required to establish ade- ance through early initiation of RRT, and (iii) conversion
quate urine flow, and care must be taken with bicarbonate to less toxic metabolites by cofactor supplementation like
loading to avoid profound alkalosis. Additionally, early ini- thiamine and pyridoxine (Table 12.5). Other causes of
tiation of RRT can rapidly decrease the concentration of oxalate nephropathy include enteric hyperoxaluria after
some inciting agents (e.g. phosphate, oxalate). bariatric surgeries and primary oxaluria.
enovascular Etiologies of Acute

R Renal Vein Thrombosis
Kidney Injury Renal vein thrombosis (RVT), either unilateral or bilateral,
is an extremely rare cause of AKI. The common etiologies
Renal Artery Thrombosis and Renal Infarction are NS and other hypercoagulable states, such as antiphos-
Acute renal artery thrombosis (RAT) is a rare clinical pholipid antibody syndrome. Of all the causes of NS,
entity characterized by acute onset of abdominal or flank patients with membranous glomerulonephritis have the
pain, nausea, vomiting, hematuria, hypertension, and highest prevalence of RVT (37%) [98]. In a case series, low
even more rarely AKI. Most common causes of renal serum albumin of less than 2.8 g/dl, rather than proteinu-
infarction are cardioembolic, although the percentage ria, was found to have increased risk for thrombotic epi-
involvement varies with different studies, ranging from sodes [99]. Selective renal venogram is considered the gold
18% to 55.7% [92–95]. Renal artery injury following blunt standard for diagnosis, but it is not routinely performed
abdominal trauma and hypercoagulable states like due to a high risk for contrast-induced AKI as well as vas-
nephrotic syndrome (NS), antiphospholipid antibody cular complications. In one series, computerized tomogra-
syndrome, paroxysmal nocturnal hemoglobinuria, as phy had a sensitivity of 92% and specificity of 100%. A
well as systemic malignancies can predispose to RAT as series investigating Doppler ultrasonography revealed a
well. In one large case series, AKI developed in 21% of sensitivity of 85% and specificity of 56%, but this modality
patients with renal infarction [95]. AKI is unusual with is heavily operator dependent [98]. Magnetic resonance
unilateral involvement unless there is baseline CKD. CT imaging can also be used, although adequate studies have
scan with intravenous contrast is the imaging modality not evaluated this modality.
of choice in suspected RAT, however intravenous con- Treatment of RVT involves conventional anticoagulation
trast should be used with caution in those with kidney with intravenous heparin and warfarin sodium. Use of
injury. Magnetic resonance angiography is another thrombolytics in RVT has been reported, but given the risk
option, but exposure to gadolinium is associated with for bleeding their use should be considered only in patients
nephrogenic systemic fibrosis in patients with CKD. For with severe bilateral RVT with rapidly declining kidney
patients with GFR <30 ml/min, including those under- function [100]. There are insufficient data to support the
going dialysis, it is generally recommended that gadolin- use of primary prophylactic anticoagulation in patients
ium exposure be avoided if possible and alternative with NS and the decision to pre-emptively anticoagulate a
imaging modalities be pursued [96]. Management of patient should be made on a case-by-case basis. In 2014, a
RAT involves systemic anticoagulation with heparin and risk calculator based on the Markov decision model was
a bridge to warfarin to maintain therapeutic INR between published to calculate the risk vs. benefits for prophylactic
2 and 3. There are limited data regarding use of direct anticoagulation for patients with primary membranous
oral anticoagulants in this setting. Endovascular proce- nephropathy [101]. Once a patient has developed a throm-
dures like stenting and thrombolysis have favorable out- botic event, anticoagulation should be continued until the
comes when performed early [97]. Prognosis depends on resolution of the NS, since the risk of recurrence is
severity of kidney injury and underlying comorbidities. extremely high. There are few case reports of using direct
In the largest series, mortality was 5% and 2% progressed oral anticoagulants for treatment of thrombosis when war-
to ESRD [95]. farin is contraindicated, but there is no consensus or evi-
dence for their use at this time.
References
Raghavan, R. and Eknoyan, G. (2014). Acute interstitial

1 4 Baldwin, D.S. et al. (1968). Renal failure and interstitial
nephritis -a reappraisal and update. Clin. Nephrol. 82 (3): nephritis due to penicillin and methicillin. N. Engl. J. Med.
149–162. 279 (23): 1245–1252.
2 Councilman, W.T. (1898). Acute interstitial nephritis. J. 5 Muriithi, A.K. et al. (2014). Biopsy-proven acute interstitial
Exp. Med. 3 (4–5): 393–420. nephritis, 1993-2011: a case series. Am. J. Kidney Dis. 64
3 More, R.H., Mc, M.G., and Duff, G.L. (1946). The (4): 558–566.
pathology of sulfonamide allergy in man. Am. J. Pathol. 6 Nast, C.C. (2017). Medication-induced interstitial nephritis
22: 703–735. in the 21st century. Adv. Chronic Kidney Dis. 24 (2): 72–79.
Reference 181
7 Praga, M. et al. (2015). Changes in the aetiology, 23 Nolan, C.R. 3rd, Anger, M.S., and Kelleher, S.P. (1986).
clinical presentation and management of acute Eosinophiluria--a new method of detection and
interstitial nephritis, an increasingly common cause definition of the clinical spectrum. N. Engl. J. Med. 315
of acute kidney injury. Nephrol. Dial. Transplant. (24): 1516–1519.
30: 1472–1479, 1479. 24 Muriithi, A.K., Nasr, S.H., and Leung, N. (2013). Utility
8 Rossert, J. (2001). Drug-induced acute interstitial of urine eosinophils in the diagnosis of acute
nephritis. Kidney Int. 60 (2): 804–817. interstitial nephritis. Clin. J. Am. Soc. Nephrol. 8 (11):
9 Wanchoo, R. et al. (2017). Adverse renal effects of 1857–1862.
immune checkpoint inhibitors: a narrative review. Am. J. 25 Perazella, M.A. and Markowitz, G.S. (2010). Drug-
Nephrol. 45 (2): 160–169. induced acute interstitial nephritis. Nat. Rev. Nephrol. 6
10 Arellano, F. and Sacristan, J.A. (1993). Allopurinol (8): 461–470.
hypersensitivity syndrome: a review. Ann. Pharmacother. 26 Ramachandran, R. et al. (2015). Drug-induced acute
27 (3): 337–343. interstitial nephritis: a clinicopathological study and
11 Mori, Y. et al. (2005). Predominant tubulointerstitial comparative trial of steroid regimens. Indian J. Nephrol.
nephritis in a patient with systemic lupus nephritis. Clin. 25 (5): 281–286.
Exp. Nephrol. 9 (1): 79–84. 27 Schwarz, A. et al. (2000). The outcome of acute
12 Shah, S., Carter-Monroe, N., and Atta, M.G. (2015). interstitial nephritis: risk factors for the transition from
Granulomatous interstitial nephritis. Clin. Kidney J. 8 (5): acute to chronic interstitial nephritis. Clin. Nephrol. 54
516–523. (3): 179–190.
13 Fernandez-Juarez, G. et al. (2018). Duration of treatment 28 Clarkson, M.R. et al. (2004). Acute interstitial
with corticosteroids and recovery of kidney function in nephritis: clinical features and response to
acute interstitial nephritis. Clin. J. Am. Soc. Nephrol. 13 corticosteroid therapy. Nephrol. Dial. Transplant. 19
(12): 1851–1858. (11): 2778–2783.
14 Baker, R.J. and Pusey, C.D. (2004). The changing profile 29 Gonzalez, E. et al. (2008). Early steroid treatment
of acute tubulointerstitial nephritis. Nephrol. Dial. improves the recovery of renal function in patients with
Transplant. 19 (1): 8–11. drug-induced acute interstitial nephritis. Kidney Int. 73
15 Kleinknecht, D., Landais, P., and Goldfarb, B. (1986). (8): 940–946.
Analgesic and non-steroidal anti-inflammatory drug- 30 Prendecki, M. et al. (2017). Long-term outcome in
associated acute renal failure: a prospective collaborative biopsy-proven acute interstitial nephritis treated with
study. Clin. Nephrol. 25 (6): 275–281. steroids. Clin. Kidney J. 10 (2): 233–239.
16 Covic, A. et al. (2004). A clinical description of 31 Preddie, D.C. et al. (2006). Mycophenolate mofetil for the
rifampicin-induced acute renal failure in 170 consecutive treatment of interstitial nephritis. Clin. J. Am. Soc.
cases. J. Indian Med. Assoc. 102 (1): 20, 22–25. Nephrol. 1 (4): 718–722.
17 Moledina, D.G. and Perazella, M.A. (2016). Proton pump 32 Shao, T. et al. (2017). Severe acute interstitial nephritis:
inhibitors and CKD. J. Am. Soc. Nephrol. 27 (10): response to therapy with antithymocyte globulin. Kidney
2926–2928. Int. Rep. 2 (2): 138–141.
18 Covic, A. et al. (2003). A retrospective 5-year study in 33 Moledina, D.G. and Perazella, M.A. (2018). Treatment of
Moldova of acute renal failure due to leptospirosis: 58 drug-induced acute Tubulointerstitial nephritis: the
cases and a review of the literature. Nephrol. Dial. search for better evidence. Clin. J. Am. Soc. Nephrol. 13
Transplant. 18 (6): 1128–1134. (12): 1785–1787.
19 Hannedouche, T. et al. (1990). Renal granulomatous 34 Flory, C.M. (1945). Arterial occlusions produced by
sarcoidosis: report of six cases. Nephrol. Dial. Transplant. emboli from eroded aortic Atheromatous plaques. Am. J.
5 (1): 18–24. Pathol. 21 (3): 549–565.
20 Mandeville, J.T., Levinson, R.D., and Holland, G.N. 35 Preston, R.A. et al. (1990). Renal biopsy in patients 65
(2001). The tubulointerstitial nephritis and uveitis years of age or older -an analysis of the results of 334
syndrome. Surv. Ophthalmol. 46 (3): 195–208. biopsies. J. Am. Geriatr. Soc. 38 (6): 669–674.
21 Masaki, Y. et al. (2009). Proposal for a new clinical entity, 36 Saklayen, M.G. (1989). Atheroembolic renal disease:
IgG4-positive multiorgan lymphoproliferative syndrome: preferential occurrence in whites only. Am. J. Nephrol. 9
analysis of 64 cases of IgG4-related disorders. Ann. (1): 87–88.
Rheum. Dis. 68 (8): 1310–1315. 37 Cosio, F.G., Zager, R.A., and Sharma, H.M. (1985).
22 Cacoub, P. et al. (2011). The DRESS syndrome: a Atheroembolic renal-disease causes
literature review. Am. J. Med. 124 (7): 588–597. Hypocomplementemia. Lancet 2 (8447): 118–121.
8 Belenfant, X., Meyrier, A., and Jacquot, C. (1999).

3 54 Finkel, K.W. et al. (2016). Paraprotein-related kidney
Supportive treatment improves survival in multivisceral disease: evaluation and treatment of myeloma cast
cholesterol crystal embolism. Am. J. Kidney Dis. 33 (5): nephropathy. Clin. J. Am. Soc. Nephrol. 11 (12):
840–850. 2273–2279.
39 Thadhani, R.I. et al. (1995). Atheroembolic renal failure 55 Sanders, P.W. (2012). Mechanisms of light chain injury
after invasive procedures. Natural history based on 52 along the tubular nephron. J. Am. Soc. Nephrol. 23 (11):
histologically proven cases. Medicine (Baltimore) 74 (6): 1777–1781.
350–358. 56 Bringhen, S. et al. (2018). Once-weekly carfilzomib,
40 Greenberg, A. et al. (1997). Focal segmental pomalidomide, and low-dose dexamethasone for
glomerulosclerosis associated with nephrotic syndrome relapsed/refractory myeloma: a phase I/II study.
in cholesterol atheroembolism: clinicopathological Leukemia 32 (8): 1803–1807.
correlations. Am. J. Kidney Dis. 29 (3): 334–344. 57 Dimopoulos, M.A. et al. (2013). The role of novel agents
41 Theriault, J. et al. (2003). Atheroembolic renal failure on the reversibility of renal impairment in newly
requiring dialysis: potential for renal recovery? diagnosed symptomatic patients with multiple myeloma.
A review of 43 cases. Nephron Clin. Pract. 94 (1): Leukemia 27 (2): 423–429.
c11–c18. 58 Glavey, S.V. et al. (2013). Long-term outcome of patients
42 Scolari, F. et al. (2003). Predictors of renal and patient with multiple [corrected] myeloma-related advanced
outcomes in atheroembolic renal disease: a prospective renal failure following auto-SCT. Bone Marrow
study. J. Am. Soc. Nephrol. 14 (6): 1584–1590. Transplant. 48 (12): 1543–1547.
43 Falanga, V., Fine, M.J., and Kapoor, W.N. (1986). The
59 Burnette, B.L., Leung, N., and Rajkumar, S.V. (2011).
cutaneous manifestations of cholesterol crystal
Renal improvement in myeloma with bortezomib
embolization. Arch. Dermatol. 122 (10): 1194–1198.
plus plasma exchange. N. Engl. J. Med. 364 (24):
44 Scolari, F. et al. (2000). Cholesterol crystal embolism: a
2365–2366.
recognizable cause of renal disease. Am. J. Kidney Dis. 36
60 Hutchison, C.A. et al. (2019). High cutoff versus high-flux
(6): 1089–1109.
haemodialysis for myeloma cast nephropathy in patients
45 Lye, W.C., Cheah, J.S., and Sinniah, R. (1993). Renal
receiving bortezomib-based chemotherapy (EuLITE): a
cholesterol embolic disease. Case report and review of the
phase 2 randomised controlled trial. Lancet. Haematol. 6
literature. Am. J. Nephrol. 13 (6): 489–493.
(24): e217–e228.
46 Dahlberg, P.J., Frecentese, D.F., and Cogbill, T.H. (1989).
61 Bridoux, F. et al. (2017). Effect of high-cutoff
Cholesterol embolism: experience with 22 histologically
hemodialysis vs conventional hemodialysis on
proven cases. Surgery 105 (6): 737–746.
hemodialysis Independence among patients with
47 Ishiyama, K., Sato, T., and Taguma, Y. (2015). Low-
myeloma cast nephropathy: a randomized clinical trial.
density lipoprotein apheresis ameliorates renal prognosis
JAMA 318 (21): 2099–2110.
of cholesterol crystal embolism. Ther. Apher. Dial. 19 (4):
355–360. 62 Finkel, K. and Fabbrini, P. (2017). High Cut-Off
48 Vogt, A. (2017). Hyperlipoproteinaemia(a) -apheresis Hemodialysis for Myeloma Cast Nephropathy – Do
and emerging therapies. Clin. Res. Cardiol. Suppl. 12 We Finally Have An Answer? J. Onco-Nephrol.
(Suppl 1): 12–17. 1 (2): 67–70.
49 White, C.J. (2010). Optimizing outcomes for renal 63 Dimopoulos, M.A. et al. (2016). International Myeloma
artery intervention. Circ. Cardiovasc. Interv. 3 (2): Working Group recommendations for the diagnosis and
184–192. Management of Myeloma-Related Renal Impairment.
50 Scolari, F. and Ravani, P. (2010). Atheroembolic renal J. Clin. Oncol. 34 (13): 1544–1557.
disease. Lancet 375 (9726): 1650–1660. 64 Perazella, M.A. and Markowitz, G.S. (2008).
51 Henry, M. et al. (2003). Renal angioplasty and stenting Bisphosphonate nephrotoxicity. Kidney Int. 74 (11):
under protection: the way for the future? Catheter. 1385–1393.
Cardiovasc. Interv. 60 (3): 299–312. 65 Terpos, E. et al. (2015). European myeloma network
52 Rajkumar, S.V. et al. (2014). International Myeloma guidelines for the management of multiple myeloma-
Working Group updated criteria for the diagnosis of related complications. Haematologica 100 (10):
multiple myeloma. Lancet Oncol. 15 (12): e538–e548. 1254–1266.
53 Rajkumar, S.V. (2016). Updated diagnostic criteria and 66 Cairo, M.S. and Bishop, M. (2004). Tumour lysis
staging system for multiple myeloma. Am. Soc. Clin. syndrome: new therapeutic strategies and classification.
Oncol. Educ. Book 35: e418–e423. Br. J. Haematol. 127 (1): 3–11.
Reference 183
7 Baeksgaard, L. and Sorensen, J.B. (2003). Acute tumor

6 purgative: an underrecognized cause of chronic renal
lysis syndrome in solid tumors--a case report and review failure. J. Am. Soc. Nephrol. 16 (11): 3389–3396.
of the literature. Cancer Chemother. Pharmacol. 51 (3): 83 Kelton, J., Kelley, W.N., and Holmes, E.W. (1978). A rapid
187–192. method for the diagnosis of acute uric acid nephropathy.
68 Cairo, M.S. et al. (2010). Recommendations for the Arch. Intern. Med. 138 (4): 612–615.
evaluation of risk and prophylaxis of tumour lysis 84 Oldfield, V. and Perry, C.M. (2006). Rasburicase: a
syndrome (TLS) in adults and children with malignant review of its use in the management of anticancer
diseases: an expert TLS panel consensus. Br. J. Haematol. therapy-induced hyperuricaemia. Drugs 66 (4):
149 (4): 578–586. 529–545.
69 Coiffier, B. et al. (2008). Guidelines for the management 85 Barceloux, D.G. et al. (1999). American Academy of
of pediatric and adult tumor lysis syndrome: an evidence- clinical toxicology practice guidelines on the treatment of
based review. J. Clin. Oncol. 26 (16): 2767–2778. ethylene glycol poisoning. Ad hoc committee. J. Toxicol.
70 Wossmann, W. et al. (2003). Incidence of tumor lysis Clin. Toxicol. 37 (5): 537–560.
syndrome in children with advanced stage Burkitt’s 86 Megarbane, B. (2010). Treatment of patients with
lymphoma/leukemia before and after introduction of ethylene glycol or methanol poisoning: focus on
prophylactic use of urate oxidase. Ann. Hematol. 82 (3): fomepizole. Open Access Emerg. Med. 2: 67–75.
160–165. 87 Keeney, R.E., Kirk, L.E., and Bridgen, D. (1982).
71 Bobulescu, I.A. and Moe, O.W. (2012). Renal transport of Acyclovir tolerance in humans. Am. J. Med. 73 (1A):
uric acid: evolving concepts and uncertainties. Adv. 176–181.
Chronic Kidney Dis. 19 (6): 358–371.
88 Boubaker, K. et al. (1998). Changes in renal function
72 Howard, S.C., Jones, D.P., and Pui, C.H. (2011). The
associated with indinavir. AIDS 12 (18): F249–F254.
tumor lysis syndrome. N. Engl. J. Med. 364 (19):
89 Wall, S.M. et al. (1996). Effective clearance of
1844–1854.
methotrexate using high-flux hemodialysis membranes.
73 Smalley, R.V. et al. (2000). Allopurinol: intravenous use
Am. J. Kidney Dis. 28 (6): 846–854.
for prevention and treatment of hyperuricemia. J. Clin.
90 Becker, K., Jablonowski, H., and Haussinger, D. (1996).
Oncol. 18 (8): 1758–1763.
Sulfadiazine-associated nephrotoxicity in patients with
74 Annemans, L. et al. (2003). Pan-European multicentre
the acquired immunodeficiency syndrome. Medicine
economic evaluation of recombinant urate oxidase
(Baltimore) 75 (4): 185–194.
(rasburicase) in prevention and treatment of
91 Ramsey, L.B. et al. (2018). Consensus guideline for use of
hyperuricaemia and tumour lysis syndrome in
Glucarpidase in patients with high-dose methotrexate
haematological cancer patients. Support Care Cancer 11
induced acute kidney injury and delayed methotrexate
(4): 249–257.
clearance. Oncologist 23 (1): 52–61.
75 Wilson, F.P. and Berns, J.S. (2014). Tumor lysis syndrome:
new challenges and recent advances. Adv. Chronic Kidney 92 Antopolsky, M. et al. (2012). Renal infarction in the ED:
Dis. 21 (1): 18–26. 10-year experience and review of the literature. Am. J.
76 Sonbol, M.B. et al. (2013). Methemoglobinemia and Emerg. Med. 30 (7): 1055–1060.
hemolysis in a patient with G6PD deficiency treated with 93 Bae, E.J. et al. (2014). A retrospective study of short-and
rasburicase. Am. J. Hematol. 88 (2): 152–154. long-term effects on renal function after acute renal
77 Kjellstrand, C.M. et al. (1974). Hyperuricemic acute renal infarction. Ren. Fail. 36: 1385–1389, 1360.
failure. Arch. Intern. Med. 133 (3): 349–359. 94 Bourgault, M. et al. (2013). Acute renal infarction: a case
78 Ratanarat, R. et al. (2005). Phosphate kinetics during series. Clin. J. Am. Soc. Nephrol. 8 (3): 392–398.
different dialysis modalities. Blood Purif. 95 Oh, Y.K. et al. (2016). Clinical characteristics and
23 (1): 83–90. outcomes of renal infarction. Am. J. Kidney Dis. 67 (2):
79 Perazella, M.A. (1999). Crystal-induced acute renal 243–250.
failure. Am. J. Med. 106 (4): 459–465. 96 Schieda, N. et al. (2018). Gadolinium-based contrast
80 Mulay, S.R. and Anders, H.J. (2016). Crystallopathies. N. agents in kidney disease: a comprehensive review and
Engl. J. Med. 375 (13): e29. clinical practice guideline issued by the Canadian
81 Mulay, S.R. and Anders, H.J. (2017). Crystal Association of Radiologists. Can. J. Kidney Health Dis. 5:
nephropathies: mechanisms of crystal-induced kidney 1–17.
injury. Nat. Rev. Nephrol. 13 (4): 226–240. 97 Blum, U. et al. (1993). Effect of local low-dose
82 Markowitz, G.S. et al. (2005). Acute phosphate thrombolysis on clinical outcome in acute embolic renal
nephropathy following oral sodium phosphate bowel artery occlusion. Radiology 189 (2): 549–554.
8 Singhal, R. and Brimble, K.S. (2006). Thromboembolic

9 100 Markowitz, G.S. et al. (1995). Renal vein thrombosis
complications in the nephrotic syndrome: treated with thrombolytic therapy: case report and brief
pathophysiology and clinical management. Thromb. Res. review. Am. J. Kidney Dis. 25 (5): 801–806.
118 (3): 397–407. 101 Lee, T. et al. (2014). Personalized prophylactic
99 Lionaki, S. et al. (2012). Venous thromboembolism in anticoagulation decision analysis in patients with
patients with membranous nephropathy. Clin. J. Am. Soc. membranous nephropathy. Kidney Int. 85 (6):
Nephrol. 7 (1): 43–51. 1412–1420.
185
13
Renal Replacement Therapy in Acute Kidney Injury

Michael J. Connor, Jr1,2, Javier A. Neyra3, and Marlies Ostermann4
1
Division of Pulmonary, Allergy, Critical Care, & Sleep Medicine, Department of Medicine, Emory University School of Medicine, Atlanta, GA, USA
2
Division of Renal Medicine, Department of Medicine, Emory University School of Medicine, Atlanta, GA, USA
3
Division of Nephrology, Bone and Mineral Metabolism, Department of Medicine, University of Kentucky, Lexington, KY, USA
4
Department of Critical Care, Guy’s & St Thomas’ Hospital, London, UK
I ntroduction basis for the optimal timing of initiation of treatment,

the selection of modality and dose of RRT, and RRT cir-
Acute kidney injury (AKI) remains a common condition in cuit considerations to optimize RRT circuit performance
hospitalized patients, especially amongst the critically ill in and function. Unfortunately, this chapter will not be
the intensive care unit (ICU), affecting approximately able to address many other important factors to consider
50–60% of ICU patients [1], and it is an independent risk when delivering RRT for AKI such as medication dosing
factor for mortality [2–7]. Unfortunately, the strategies to and nutrition support in RRT, transitions between thera-
prevent AKI, aside from prerenal azotemia, contrast- pies and when to stop acute RRT, or modality-specific
associated AKI, and pigment-induced AKI, remain limited. RRT prescription details. Fortunately, there are many
Recently, several trials have suggested that a renal- other resources and studies available that address these
protective strategy based on results of urinary biomarkers issues [11–17].
may be able to prevent AKI and mitigate AKI risk in some
clinical situations [8, 9]. However, there remain no effec-
tive pharmacologic interventions for the treatment of most Renal Replacement Therapies in AKI
forms of intrinsic AKI. As a result, renal replacement ther-
apy (RRT) remains the cornerstone of care in patients with RRT is at first a misnomer, suggesting that its role is to
severe AKI and the use of RRT for dialysis-dependent AKI replace kidney function when glomerular filtration rate (or
(AKI-D) continues to increase steadily [1, 10]. Over the creatinine clearance) has essentially ceased. Rather, RRT
past three decades there has been a proliferation of modali- should instead be considered a supportive therapy (i.e.
ties of RRT available for the management of AKI to now renal support therapy) in situations where the metabolic
include intermittent hemodialysis (IHD), acute peritoneal and fluid demands exceed native kidney function [18, 19].
dialysis (PD), continuous renal replacement therapy In addition to fluid removal and clearance of waste prod-
(CRRT), and “hybrid” therapies (prolonged-intermittent ucts, RRT may also limit the negative, distant effects of AKI
renal replacement therapy [PIRRT]). Despite the increas- on nonrenal organs, including respiratory and cardiac dys-
ing technological sophistication of RRT, key management function. Recent advances have made the provision of RRT
issues, such as the optimal timing of initiation, selection of easier and safer, but it is important to acknowledge that
modality, and dosing of RRT, remain controversial, and the RRT only replaces excretory function (solute clearance and
clinical complexity and heterogeneity of patients with fluid removal) but does not provide or replace any of the
severe AKI has made clinical trials addressing these ques- myriad other kidney functions (i.e. tubular resorption of
tions challenging. electrolytes and nutrients, endocrine regulation, or meta-
This chapter focuses on key issues surrounding the bolic functions). Finally, clinicians must recognize that
use of RRT in the treatment of AKI, summarizing the patients with AKI are distinctly different from those with
indications for RRT initiation, the current evidence end-stage renal disease (ESRD). The sudden, rapid decline
186 Renal Replacement Therapy in Acute Kidney Injury
of kidney function in AKI and multiorgan failure does not and micronutrients, alterations in drug pharmacokinetics
permit the body’s adaptive responses seen in those with (PK), and dialysis-induced hypothermia to name a few
progressive chronic kidney disease. (Table 13.1).
Intermittent RRT is usually preferred in patients who are
hemodynamically more stable and able to tolerate greater
Modalities
fluctuations in fluid and metabolic status. IHD also pro-
RRT utilizes an extracorporeal circuit containing an artifi- vides greater minute-to-minute small solute clearance,
cial membrane (hemodialysis [HD] or hemofiltration) or which makes it far more effective in situations where life-
the patient’s own peritoneal membrane (peritoneal dialysis) threatening issues require emergent correction such as
(see Table 13.1). Hemodialysis refers to the removal of sol- severe hyperkalemia with marked conduction abnormali-
ute and water across a semipermeable membrane by estab- ties (i.e. complete heart block), acute poisonings, or pro-
lishing and maintaining a concentration gradient between found metabolic acidemia. Intermittent RRT also allows
the blood flowing through the filter and the countercurrent more active physical therapy and rehabilitation by provid-
flow of an electrolyte containing dialysate solution across ing time off from therapy.
the other side of the membrane (principle of diffusion) [21]. Hybrid therapies are variations of RRT that are usually
In hemofiltration, water and dissolved solutes are trans- provided daily for 6–18 hours. They offer some advantages
ferred across a semipermeable membrane by means of a of both intermittent and CRRT, and are often employed
transmembrane pressure gradient (principle of convec- during the transition period from CRRT to intermittent
tion) [21]. Hemodiafiltration incorporates the removal of RRT. There are multiple terminologies in the literature for
solute and water across a semipermeable membrane by hybrid therapies, with the most common being SLED and
both a concentration and transmembrane pressure gradient PIRRT [22].
(diffusion and convection) [21]. PD uses the patient’s peri- Acute PD requires the insertion of a peritoneal dialysis
toneum as a membrane across which fluids, electrolytes, catheter. It has theoretical advantages in that hemodynamic
and small molecules are exchanged by diffusion. disturbances are rare and anticoagulation is not required.
Further characterization of RRT is based on the dura- PD provides continuous clearance, but fluid removal rate
tion. As such, there are continuous and intermittent forms and degree of solute clearance cannot be controlled as tightly
of RRT. CRRT includes continuous veno-venous hemofil- as with IHD or CRRT. In addition, the risks of respiratory
tration (CVVH), continuous veno-venous hemodialysis compromise from increased intra-abdominal pressure,
(CVVHD), continuous veno-venous hemodiafiltration pleural-peritoneal leak and intra-abdominal infection make
(CVVHDF), and PD; intermittent techniques include IHD it generally less suitable in the acute setting in adults, how-
and newer “hybrid” therapies, such PIRRT, which has also ever acute PD remains a commonly used option in neonates
been called sustained low-efficiency dialysis (SLED) and resource-limited settings. In patients with abdominal
(Figure 13.1). pathology, PD is usually contraindicated.
CRRT can be delivered as CVVH, CVVHD, or CVVHDF.
Historically, diffusion, the clearance mode in hemodialysis,
Choice of Modality
favors the removal of solutes with a lower molecular weight
In clinical practice, the choice of modality depends on the (MW) whereas convection (which underlies the process of
characteristics of the patient, availability of machines, and hemofiltration) is more efficient at removing both low-and
expertise of the clinical team [20]. Intermittent modalities higher MW substances as it permits the transfer of mole-
are generally associated with greater fluctuations in meta- cules that can be accommodated by the size of the mem-
bolic and fluid status and an increased risk of intradialytic brane pore. Indeed, there is some evidence that clearance of
hemodynamic compromise (i.e. hypotension) whereas vancomycin (with MW approximately 1500 Da) can be
CRRT provides more sustained fluid removal and solute improved with hemofiltration [23]. There is debate about
clearance, and is associated with less fluctuation whether there is a clinical benefit afforded by hemofiltra-
(Figure 13.1 and Table 13.1). As such, CRRT is recom- tion in the setting of AKI [24]. The potential benefit of
mended for patients who do not tolerate fluctuations in removing a wider spectrum of solutes that are too large to
fluid status or shifts in osmotically active molecules such be removed by hemodialysis (for instance inflammatory
as those patients with hemodynamic instability or acute cytokines), would make hemofiltration an attractive ther-
brain injury [20]. The potential drawbacks of CRRT apy for certain types of AKI, for instance sepsis-associated
include greater restriction of mobility, continuous expo- AKI. However, small studies demonstrate that this differ-
sure to the extracorporeal circuit, need for anticoagula- ence in middle- or large-molecule clearance is low with
tion, often unrecognized losses of amino acids, vitamins, modern high-flux membranes [25, 26]. Additionally, as
Table 13.1 Characteristics of different RRT modalities used in AKI.
Modality
Hybrid RRT
Parameter IHD SLED/EDD/PIRRT CRRT PD
Duration (h) 4–6 6–16 24 24

Frequency Daily/alternate days Daily/alternate days Daily Daily
Solute transport Diffusion Diffusion or convection or Diffusion or convection or both Diffusion
both
Blood flow (ml/min) 200–350 100–300 100–250 NA
Dialysate flow (ml/min) 300–800 200–300 0–50 25–40
Urea clearance (ml/min) 150–180 90–140 20–45 15–35
Access Central venous dialysis Central venous dialysis Central venous dialysis catheter Peritoneal dialysis catheter
catheter catheter
Need for anticoagulation Usually but not absolutely Usually but not absolutely Yes No
necessary necessary
Fluctuations of osmotically active ++ + Less; can occur in case of treatment No
metabolites interruptions
Fluctuations of fluid status ++ + Less; can occur in case of treatment No
interruptions
Effect on ICP in patients with acute Increase Potential increase Usually no change No change
brain injury
Effect on serum concentrations of Major fluctuations Some fluctuations Less fluctuations Less fluctuations
renally cleared drugs
Infection risk Line infection/bacteremia Line infection/bacteremia Line infection/bacteremia Exit site infection; tunnel
infection; peritonitis
Loss of nutrients into dialysate/filtrate Yes Yes Yes Yes
ICP, intracerebral pressure; RRT, renal replacement therapy; SLED, slow extended dialysis; EDD, extended daily dialysis; PIRRT, prolonged intermittent renal replacement therapy.
Source: Adapted from Bagshaw et al. [20].
c13.indd 187 09-12-2022 15:22:44

hemodynamically tenuous patients to receive CRRT rather

SLED
IHD
PIRRT
PD CRRT than IHD, along with the inherent limitations of nonrand-
omized study designs, confounded many of the early obser-
continuous vational studies that compared continuous and intermittent
intermittent therapies. There are at least seven randomized controlled
trials (RCTs) comparing CRRT and IHD (Table 13.2)
Clearance +++ ++ +
per hour
[29–35]. In a multicenter RCT of 166 patients with AKI,
Mehta et al. observed hospital mortality rates of 65.5% in
Fluid status patients randomized to CRRT, compared to 47.6% in
patients randomized to IHD (P < 0.02) [29]. However,
Fluctuations in
urea/NH3/Na+ imbalanced randomization resulted in patients in the
CRRT arm having greater severity of illness and a higher
Figure 13.1 Renal Replacement Therapy Modalities in Acute rate of liver failure, both of which were independently
Kidney Injury.
associated with increased mortality. After covariate adjust-
ment, the investigators found no difference in mortality
attributable to modality of RRT. Interpretation of this study
would be expected based on increased filtration fractions,
is confounded further by a high rate of crossover between
CRRT circuit survival times are generally lower with higher
treatment modalities in both study arms. In a single-center
rates of convection (CVVH) [23, 27]. Thus, any difference in
RCT involving 80 patients, Augustine et al. reported more
minute-to-minute middle- and large-molecule clearance
effective fluid removal and greater hemodynamic stability
may be negated with less cumulative time on CRRT with
associated with CVVHD compared to IHD, but there was
higher rates of convection. This effect may be mitigated by
no difference in survival [31]. Similarly, Uehlinger et al.
CVVHDF combining both diffusion and convection, but
observed no difference in survival in 70 patients rand-
there is no clinical data that this mode is superior. Indeed, a
omized to CVVHDF compared to 55 patients assigned to
meta-analysis comparing hemofiltration and hemodialysis
IHD [32].
for AKI did not demonstrate any reliable differences in
The Hemodiafe study was a multicenter RCT conducted
patient survival or kidney recovery [28].
in 21 ICUs in France which demonstrated no difference in
It is commonly believed that CRRT is superior to IHD in
mortality in 184 patients randomized to IHD compared to
critically ill patients but clinical trials have failed to confirm
175 patients randomized to CVVHDF [33]. This study is
this [20, 29–37]. The propensity for seriously ill and more
particularly notable for the fact that IHD was delivered to
Table 13.2 Randomized controlled trials comparing modalities of RRT.
Randomized controlled Number of CRRT Mortality of CRRT Number of control Mortality of control Unadjusted odds of death in
trial patients group (%) patients group (%) control group (95% CI)
CRRT versus IHD

Mehta et al. [29] 84 66 82 48 0.63 (0.3–1.4)
John et al. [30] 10 70 20 70 1.0 (0.1–6.6)
Augustine et al. [31] 40 68 40 70 1.12 (0.4–3.2)
Uehlinger et al. [32] 70 47 55 51 1.16 (0.5–2.5)
Vinsonneau et al. [33] 175 67 184 68 1.05 (0.6–1.7)
Lins et al. [34] 172 58.1 144 62.5 NA
Schefold et al. [35] 122 45.4 128 52.4 1.32 (0.80–2.19)
CRRT versus PD
Phu et al. [38] 34 15 36 47 5.1
CRRT versus SLED
Abe et al. [39] 30 37 30 17 NA
RRT, renal replacement therapy; CRRT, continuous renal replacement therapy; IHD, intermittent hemodialysis; PD, peritoneal dialysis; SLED,
sustained low-efficiency dialysis; CI, confidence interval; NA, not available.
Indications for Acute Renal Replacement Therap 189
patients despite marked hemodynamic instability, with lit- clearances were equivalent but less heparin was needed
tle crossover between treatment groups. However, in the and acidosis was corrected faster in patients treated with
Hemodiafe study, IHD delivery was tailored to minimize PIRRT [47]. In another RCT in 60 patients comparing
risk of hemodynamic compromise – specifically, initial CVVHDF to 6–8 hours of PIRRT, there was no difference in
blood flow was limited to 250 ml/min, dialysate flow to ICU or 30-day mortality, but survivors in the PIRRT group
500 ml/min (both limiting speed of small solute clearance), had higher renal recovery rates and a shorter length of stay
dialysate sodium concentration was increased to 150 mEq/l in ICU [39]. Finally, a systematic review and meta-analysis
(high dialysate sodium concentration is known to mini- including seven RCTs and 10 observational studies con-
mize risk of intradialytic hypotension), and dialysate tem- cluded that PIRRT was associated with similar outcomes to
perature was decreased to 35.0°C to increase peripheral CRRT [48].
vascular resistance and vascular refilling. Mean IHD treat- There is a limited number of studies comparing PD to
ment duration was 5.2 hours and mean net fluid removal CRRT in adults. Two studies reported better results with
during each IHD session was only 2.2 l [33]. In the United CRRT but the clearance delivered by PD was low [49, 50].
States and many other countries, such optimized, tailored, One RCT found that CVVH was superior to PD in infection-
and gentle IHD treatment is less common and challenging associated AKI but the predominance of malaria as the
to deliver in those with AKI-D, making it difficult to extrap- cause of AKI limits the study’s generalizability [38].
olate these results to daily practice. In summary, an understanding of the benefits and
The single-center CONVINT (Contininous versus inter- drawbacks of the different types of RRT is essential.
mittent renal replacement therapy Trial) trial was a pro- Current data do not show superiority of any specific
spective RCT comparing daily IHD with CVVH in 252 modality of renal support in patients with AKI. Rather
critically ill patients [35]. Survival rates at 14 days after RRT than viewing these modalities as competing, we recom-
were 39.5% (IHD) versus 43.9% (CVVH) (P = 0.50). All- mend viewing them as complementary therapies.
cause hospital and 30-day mortality rates were not different Modality selection for an individual patient requires a
between the two groups (all P > 0.5). No differences were consideration of the clinical characteristics of patients,
observed in days on RRT, vasopressor days, days on ventila- the goals of RRT on a given day, provider expertise, and
tor, or length of stay in hospital [35]. locally available technology and staffing resources.
The propensity for hemodynamic instability and intra- Importantly, guidelines do endorse the use of CRRT in
dialytic hypotension during IHD has led to the suggestion those with (or at risk for) increased intracranial pressure
that CRRT may be associated with an increased likelihood as data does indicate that IHD causes a greater increase in
for recovery of renal function [40–42]. However, multiple intracerebral pressure (ICP) during the session compared
meta-analyses comparing outcomes between these modali- to CRRT [51, 52]. Finally, as complimentary therapies,
ties of RRT have not found any convincing differences in the choice of modality may need to be adjusted during the
any relevant patient-centered outcomes [36, 37, 43–46]. course of critical illness as a given patient’s clinical status
One meta-analysis included both randomized and nonran- and renal support needs evolve.
domized studies and concluded that weaknesses in the
quality of the studies significantly limited meaningful
comparisons between modalities [43]. Based on weighting I ndications for Acute Renal
of the studies using an a priori assessment of study quality Replacement Therapy
and comparability of severity of illness between treatment
arms, the authors suggested that CRRT might be associated RRT for AKI (AKI-D) is initiated in approximately 15–20%
with a lower relative mortality risk. Subsequent meta- of all critically ill patients [1]. The incidence of AKI-D in
analyses differed in their criteria for including studies, noncritically ill hospitalized patients is less well estab-
resulting in minor differences in the calculated odds ratios lished and is widely variable across hospital settings and
(ORs) for mortality, but there were no differences in mor- populations. Acute RRT for symptomatic hyperkalemia (or
tality associated with use of IHD or CRRT [36, 37, 44, 45]. other severe electrolyte disorders), severe acidemia, symp-
The most recent systematic review including 21 RCTs and tomatic uremia, marked fluid overload leading to severe
prospective cohort studies also concluded that RRT modal- hypoxic acute respiratory failure, and toxic ingestions
ity was not associated with in-hospital mortality or dialysis remain classic indications. However, there is wide diver-
dependence [46]. gence between providers about the severity of a given
Studies comparing other types of RRT are limited. derangement (i.e. degree of hyperkalemia, blood urea
In a small trial of 39 patients randomized to either CVVH nitrogen [BUN], pH, or fluid overload) that prompts dialy-
or 12-hour PIRRT, cardiovascular tolerability and urea sis initiation [53].
The Kidney Disease Improving Global Outcomes Timing of Dialysis: Definition

(KDIGO) AKI guidelines [54] recommend that acute RRT
Throughout the literature, conventional/delayed/late dial-
should be initiated for “life threatening changes in fluid,
ysis initiation is defined as starting dialysis in an individ-
electrolyte, and acid-base balance exist.” However, it is
ual with advanced AKI (according to serum creatinine or
important to note that KDIGO does not define what these
urine-based criteria) when the patient reaches a prede-
specific life-threatening levels are. The KDIGO guidelines
fined “life-threatening” indication for dialysis – sympto-
go on to state that the clinician should “consider the
matic azotemia, hyperkalemia, severe acidemia, volume
broader clinical context, the presence of conditions that can
overload, and worsening hypoxia. Conversely, early/
be modified with RRT, and trends of laboratory tests – rather
preemptive dialysis has had a wide range of definitions in
than single BUN and creatinine thresholds alone – when
different studies: lower serum BUN levels, earlier stages of
making the decision to start RRT” [54]. This puts the onus
AKI, and shorter time to initiation of dialysis. This has
back on the clinician teams to determine indications for
introduced a significant heterogeneity in the literature as
acute RRT initiation in an individualized, patient-centered,
to how this early/preemptive initiation has been applied.
goal-driven approach.
Finally, almost all trials comparing timing of dialysis have
Renal support therapy in AKI has several goals: (i) return
been performed in the ICU setting, presumably where the
to and maintain fluid, electrolyte, acid-base, and solute
mortality impacts of a certain strategy may have the big-
homeostasis; (ii) to prevent complications of uremia; and
gest effect.
(iii) to allow other supportive measures (such as antimicro-
bial or other drug therapy and nutrition support) to pro-
ceed without limitations in vulnerably acutely ill patients.
Conventional/Delayed/Late RRT
However, acute RRT is not without risk and inherent com-
plications – most having to do with RRT access (vascular or In patients with AKI, it remains difficult to predict the tra-
peritoneal) complications (infections and traumatic) and jectory for renal function and AKI recovery. Traditional
circuit-related factors such as need for anticoagulation, tools such as BUN, creatinine, and urine output trends are
circuit-associated blood loss requiring transfusions, circuit helpful but have limitations. Some with advanced AKI will
failures limiting effective RRT delivery, and extracorporeal recover renal function quickly enough to never to require
removal of medications such antibiotics (resulting in inad- acute RRT. As mentioned above, acute RRT also carries
equate dosing) and nutrients. Therefore, initiating RRT risks including mechanical or infectious complications
must weigh the possible (or desired) benefits of therapy from the dialysis vascular access catheter, bleeding, loss of
against the risks. essential substances (e.g. nutrients, drugs), and increased
cost, to name a few.
Isolated azotemia is rarely the only complication of AKI
in hospitalized patients; encouragingly, studies evaluating
Timing of Dialysis Initiation the timing of initiation based purely on a BUN level (high
or low) have shown no benefit to starting dialysis ear-
Decisions surrounding whether and when to initiate acute lier [19, 55, 57]. This suggests that in those cases with iso-
RRT continue to be amongst the most challenging and vex- lated azotemia, a more deliberate, delayed approach to
ing issues in the hospitalized and/or critically ill patient dialysis is likely safe.
over the last 60 years [1, 18, 19, 53–55], particularly in More recently, the approach to studying RRT timing
patients in whom imminent life-threatening complications has centered on AKI staging according to criteria from the
are present. Acute Kidney Injury Network (AKIN) and KDIGO. The
In-hospital mortality amongst patients with AKI-D two largest trials to date were the Artificial Kidney
remains an all-too-frequent outcome afflicting 40–60% Initiation in Kidney Injury (AKIKI) trial published in
of patients [1, 37]. In a retrospective analysis of more 2016 [58] and the Initiation of Dialysis Early Versus
than 10 000 ICU patients with AKI, Liborio et al. demon- Delayed in the Intensive Care Unit (IDEAL-ICU) trial
strated that the excess hospital mortality at all stages and published in 2018 [59].
severities of AKI seems driven by the complications Both of these were multicenter trials enrolling critically
associated with AKI (hyperkalemia, metabolic acidosis, ill patients exclusively in ICUs in France [58, 59]. They had
and >5% fluid overload) rather than AKI itself [56]. In opposite hypotheses (no benefit of early RRT initiation for
theory, timely initiation of RRT might prevent the devel- AKIKI and benefit of early RRT initiation for IDEAL-ICU)
opment of AKI complications, thereby possibly improv- and enrolled adult patients with AKI who did not have
ing outcomes. specific objective criteria for RRT initiation only if/when
Timing of Dialysis Initiatio 191
they developed KDIGO stage 3 AKI based on creatinine or Limitations of both the AKIKI and IDEAL-ICU trial
urine output criteria [58, 59]. AKIKI did require that sub- designs should be noted. Both trials excluded patients in
jects have a clinical diagnosis of acute tubular necrosis whom there were objective criteria for earlier initiation of
(ATN), while IDEAL-ICU did not [58, 59]. Furthermore, renal support, such as severe volume overload, hyper-
both trials required patients to be critically ill – IDEAL- kalemia, and metabolic acidosis. While some have criticized
ICU requiring septic shock (i.e. on vasopressors) and both trials for significant use of intermittent RRT, it should
AKIKI requiring shock on vasopressors and/or invasive be noted that utilization was equivalent in both treatment
mechanical ventilation [58, 59]. Both randomized patients arms, that this represented usual care in the health systems
who had stage 3 AKI to early versus delayed dialysis with where the trials were conducted, and that prior trials, such
those in the early group starting dialysis within 6 hours as the Hemodiafe trial [33], demonstrated no improvement
(AKIKI) or 12 hours (IDEAL-ICU) of reaching stage 3 in mortality associated with CRRT. The studies also did not
AKI [58, 59]. Those in the late groups started RRT when address the question of whether a strategy of even earlier
they developed either a prespecified end point (i.e. hyper- initiation of therapy would provide a survival benefit.
kalemia, severe acidemia, progressive fluid overload, or
advancing hypoxemia) or if patients remained oligo-anuric
Early/Preemptive RRT: Theoretical Rationale
without signs of renal recovery at 48 (IDEAL-ICU) [59] or
72 hours (AKIKI) [58]. As Liborio et al. [56] demonstrated, hyperkalemia, aci-
AKIKI enrolled 619 patients and found no difference in demia, and fluid overload remain the primary life-
the primary outcome of death at 60 days (48.5% in the early threatening complications of AKI. Thus, it has long been
group vs. 49.7% in the delayed group, P = 0.79) [58]. postulated that dialysis initiation prior to development of
Furthermore, although while almost all patients in the life-threatening AKI complications may improve ICU and
early group received dialysis, only 50% of patients in the AKI outcomes. Over the last 10 years, while imperfect,
delayed group required initiation of dialysis [58]. IDEAL- data have emerged that add fuel to this debate: namely, is
ICU enrolled 488 patients but was stopped after the second survival improved when early or preemptive RRT is started
interim analysis based on the assessment of the safety in an effort to prevent the development of emergent,
board that it was futile to continue the study because com- immediately life-threatening indications for dialysis?
pletion of enrollment was unlikely to change the results of Earlier RRT initiation may allow for prevention of fluid
the trial significantly [59]. In the final analyses of the overload or improved fluid management, minimizing risk
IDEAL-ICU study, 90-day mortality was 58% in the early of hyperkalemia, and/or improved acid-base and azotemia
group and 54% in the delayed group (P = 0.38) [59]. Similar control [18, 60]. Mitigating and minimizing these AKI
to the AKIKI study, not all subjects in the IDEAL-ICU trial complications may have an important impact on nonrenal
received RRT, with 97% in the early group receiving RRT organ function such as improved pulmonary function,
while only 62% in the delayed group got RRT [59]. Of the faster liberation from mechanical ventilation, and
remaining 38% of the delayed group who did not start RRT, improved vascular and cardiac function.
75% recovered renal function without requiring RRT (70 of It is widely accepted in the critical care literature that
the total 242 patients in the delayed group) [59]. fluid overload portends poor ICU prognosis and associates
In both trials, AKIKI and IDEAL ICU, mortality was low- with organ dysfunction and impairment of organ and
est in those patients who were randomized to delayed RRT patient recovery through a variety of putative mechanisms.
but never received it due to recovery of renal function, and This association is independent of kidney functional status
mortality was highest in patients who were randomized to and more pronounced in patients with AKI, where fluid
delayed RRT and actually started RRT. For clinicians man- overload is consistently associated with mortality and lower
aging patients with AKI, this constitutes a major challenge renal recovery rates [61–64]. Studies by Bouchard et al. [63]
because at any point in time it is difficult to predict whether and Vaara et al. [64] demonstrated that the presence of fluid
renal function will recover in the following 48 hours or not. overload at the time of acute RRT initiation for AKI doubles
While these two trials found no evidence of a survival ben- the risk of death. Additionally, both studies showed a dose-
efit associated with earlier initiation of RRT, they also iden- response relationship whereby a higher percentage of fluid
tify a need for better risk-stratification tools to differentiate overload led to progressive increase risk of death [63, 64].
between patients whose renal function will recover sponta- When defining early versus delayed acute RRT on the basis
neously, so that RRT can be avoided, and patients who will of time from ICU admission to RRT initiation (within 2 or
have progressive AKI, so that RRT can be timely started >2 days from ICU admission), in a retrospective review of
before the onset of fluid overload and serious metabolic the SOAP (Sepsis Occurrence in Acutely Ill Patients Study)
derangements. study, Payen et al. showed improved survival in the early
RRT initiation group and that the delayed group had higher [BUN >100 mg/dl, potassium >6 mEq/l, magnesium
cumulative fluid balance at the time of RRT initiation [65]. >8 mEq/l, severe oliguria, or diuretic-resistant organ
In patients with AKI due to ATN requiring acute RRT, a edema]) [69]. As a single-center study, the total number
higher degree of fluid overload at RRT initiation was associ- enrolled was lower than either AKIKI or IDEAL-ICU,
ated with worse kidney recovery (RRT-dependence) at with ELAIN enrolling 231 subjects, 112 in the early group
1 year [61]. Similary, Woodward et al. showed that in criti- and 119 in the delayed group [69]. The primary outcome
cally ill patients with AKI requiring CRRT, fluid overload was 90-day mortality, which was 39.3% in the early group
>10% at the time of CRRT initiation was independently and 54.7% in the delayed group (P = 0.03), with a hazard
associated with major adverse kidney events (composite of ratio for death of 0.66 (95% confidence interval [CI] 0.45–
mortality, RRT-dependence or inability to recover 50% of 0.97) for the early group [69]. However, in ELAIN, the
baseline eGFR (estimated glomerular filtration rate) if not median time difference from meeting full eligibility crite-
on RRT) evaluated up to 90 days after discharge [66]. ria to RRT initiation was only 21 hours (95% CI
However, it should be recognized that while these observa- 18–24 hours) [69]. At a minimum, this begs the question
tional data are hypothesis-generating, there have been no as to whether a 15% absolute risk reduction for death in
randomized trials demonstrating that early initiation of the early RRT arm of the study can realistically be plausi-
RRT to optimize fluid overload decreases mortality or pro- ble with such a short time difference between early versus
motes organ recovery. late RRT initiation strategy.
Early Acute RRT Initiation Timing of Dialysis Initiation: Trial Limitations

Prior meta-analyses including RCTs, prospective observa- As illustrated above, there are important limitations in
tional trials, and retrospective trials have suggested an many of the trials performed to date. Additionally, almost
improved survival with earlier initiation of dialysis [67, all trials have used some version of a creatinine- or urea-
68]. Following these meta-analyses and published nearly based definition of AKI as an inclusion criterion for the
simultaneously with the AKIKI trial in 2016, Zarbock study. Given that fluid balance, acidemia, or electrolyte
et al. in the Effect of Early vs Delayed Initiation of Renal disturbances often drive decision-making and that
Replacement Therapy on Mortality in Critically Ill azotemia is rarely the sole criterion for RRT initiation in
Patients With Acute Kidney Injury [ELAIN] trial [69] per- the ICU, enrolling patients into clinical trials regarding
formed a single-center RCT comparing preemptive versus the timing of RRT solely on the basis of meeting urea-or
conventional dialysis initiation for AKI in ICU patients. creatinine-based AKI definitions excludes patients from
There are some important trial design differences between consideration who do not meet these AKI definitions yet
ELAIN, AKIKI, and IDEAL-ICU that warrant discussion. still have indications for RRT such as diuretic-resistant
Most importantly, the inclusion criteria for ELAIN began fluid overload. No RCTs to date have evaluated RRT ini-
with patients who had developed KDIGO stage 2 AKI tiation on the basis of fluid balance or diuretic respon-
with evidence of ATN despite optimal fluid resuscita- siveness alone.
tion [69]. Participants were required to have ongoing cat- Second, significant heterogeneity exists in study design
echolamine infusion requirements at time of enrollment, and definitions of early versus late RRT: what is “early” in
severe sepsis, refractory fluid overload, or a nonrenal one trial may be “late” in another. For example, the early
Sequential Organ Failure Assessment (SOFA) score of 2 dialysis group in AKIKI [58] and IDEAL-ICU [59] and the
or higher [69]. This difference in AKI stage at enrollment late group in ELAIN [69] were determined by the same cri-
(stage 2 in ELAIN and stage 3 in AKIKI/IDEAL-ICU) terion: stage 3 AKI without complications.
suggests that patients in ELAIN may indeed have received Third, the mode of RRT at initiation varies widely
“preemptive” dialysis in the early RRT group. Additionally, between trials and even within some trials. While the RRT
the study mandated that all patients receive CRRT at the options are varied between studies, the choice of RRT
time of RRT initiation with clear, established guidelines modalities is generally balanced between early and late
to guide transition to intermittent modes of RRT or for groups within a given trial in most circumstances. However,
discontinuation of dialysis. Finally, nearly all patients in relative rates of use of intermittent and continuous thera-
both arms actually received dialysis. pies as initial RRT modalities should be considered in eval-
In ELAIN, subjects were randomized to the early RRT uating the applicability of trial results to individual practice
group (starting within 8 hours of reaching stage 2 AKI settings where modality utilization may differ.
and enrollment) or the delayed RRT group (starting The largest multicenter, multinational trial, entitled
within 12 hours of stage 3 AKI or an absolute indication Standard vs. Accelerated Initiation of RRT in Acute Kidney
Dose of Renal Replacement Therapy in AK 193
Injury (STAART-AKI), is still enrolling but does bear some without dialysis. Chawla et al. showed that failure to
of these same study design limitations. However, in gen- respond to a formal furosemide stress test (FST) predicted
eral, the inclusion criteria for STAART-AKI starts with stage 3 AKI and the need for acute RRT (area under the
stage 2 AKI, requires equipoise by the healthcare team in receiver operating characteristic curve, 0.87 and 0.86,
regards to RRT need, and does not mandate initiation of respectively) [73]. In a pilot trial that randomized FST non-
RRT solely by stage 3 AKI criteria. responders to early versus conventional RRT initiation,
Lumlertgul et al. demonstrated that in ICU patients with
AKI, those who responded to the FST had a very low rate of
Timing of Dialysis: Summary
acute RRT needs (only 6/44 responders required RRT) [74].
Unfortunately, data remain inconclusive to clearly guide Renal biomarkers may prove useful to predict RRT needs.
clinical decisions at the bedside. Recognizing this limita- Urine tissue inhibitor of metalloproteinase 2 and insulin-
tion, the KDIGO AKI criteria suggest that clinicians should like growth factor-binding protein 7 (TIMP-2*IGFBP-7)
“consider the broader clinical context, the presence of con- can risk-stratify those at high risk for progression to severe
ditions that can be modified with RRT, and the trends of forms of AKI, with higher values (>2.0) strongly associated
laboratory tests. . . when making a decision to start with development of stage 3 AKI [75–77]. Research is
RRT” [54]. However, one should keep in mind that these needed to determine if TIMP-2*IGFBP-7 results or trends
guidelines predate the publication of AKIKI, ELAIN, and over time can discriminate patients who will and will not
IDEAL-ICU trials. While KDIGO AKI guideline review require RRT, as well as hard outcomes such as death or
and revisions are ongoing, it would seem that these more RRT-dependence.
recent studies do not argue substantially to change from
the recommendation to “consider the broader clinical con-
text of the individual patient” when weighing whether or ose of Renal Replacement
D
not to timely initiate acute RRT. Therapy in AKI
In 2017, Feng et al. published a meta-analysis of nine
RCTs on the timing of dialysis, which included the AKIKI The goal of RRT for AKI management is to optimize solute
trial but not IDEAL-ICU, that found no statistical differ- control and fluid balance according to the specific needs of
ence between trials [70]. More recently, Pasin et al. reported the patient. It is important to recognize that the dose of
a metanalysis including ELAIN, AKIKI, and IDEAL-ICU RRT is a dynamic metric that should be individualized
that concluded that early initiation of RRT in critically ill according to the clinical context of the patient, but should
patients with AKI does not provide a clinically relevant also adhere to best evidence-based practice once the patient
advantage when compared with late RRT initiation [71]. achieves a steady-state of solute and volume control.
Importantly, this study also points out the heterogeneity in
the “early” and “late” initiation of RRT definitions and the
Intermittent Hemodialysis Dose in AKI
need for specific studies in special populations such as sep-
sis or post cardiac surgery as well as the examination of There are limited data regarding the “adequate” per treat-
performance and quality metrics of different RRT modali- ment dose of IHD in AKI. Paganini et al. retrospectively
ties in future interventional studies. evaluated mortality in relation to the delivered dose of HD
The Acute Dialysis Quality Initiative (ADQI) recom- in critically ill patients with AKI [78]. Among patients at
mends a similar, individualized approach to RRT initiation either the low or high spectrum of severity of illness, HD
whereby the demands on renal function and renal capacity dose had no impact on patient outcome. However, in
are weighed with RRT initiated when significant discord- patients with an intermediate severity of illness score, the
ance is seen between renal capacity/capability and delivery of HD dose in excess of the 50th percentile (Kt/V
demands on kidney function [18]. Critically ill patients approximately 1) was associated with lower mortality risk
have high “renal demands,” frequently with significant than the delivery of lower doses of HD [78]. While these
burdens attributable to disease-related acidemia, fluid data are consistent with evidence relating dose of IHD to
management, small solutes and electrolytes, and medica- outcomes in patients with end-stage kidney disease receiv-
tion and toxin clearance. However, admittedly it remains ing maintenance hemodialysis, it should be recognized
difficult to determine which patients may benefit from that these observational data cannot differentiate between
early application of RRT versus a more patient, deliberate the dose of HD per se and patient-specific factors that lim-
strategy with close observation [72]. Additionally, it ited the delivered dose of therapy as the mediator of the
remains challenging to predict which patients will need increased mortality associated with the lower delivered
acute RRT and which are destined recover renal function dose of HD.
In a study evaluating the effect of treatment frequency of modalities of CRRT (Table 13.3). There is no consensus in
IHD on outcomes among patients with AKI, Schiffl et al. relation to which patient’s weight to use (admission weight,
alternately assigned 160 critically ill patients with ATN to ideal weight, or current weight) but we tend to use current
daily or alternate-day IHD and found that patients who weight to account for the acute changes in volume of distri-
received daily IHD had both lower mortality 14 days after bution from fluid expansion (during critical illness and
discontinuation of RRT (28% vs. 46%, P = 0.01) and shorter resuscitation for shock) and systemic inflammation.
duration of AKI (9 ± 2 vs. 16 ± 6 days, P = 0.001) [79]. There are more data regarding the appropriate dosing of
However, it is important to note that the dose of IHD deliv- CRRT in the literature (Table 13.4). Early work by Ronco
ered to the alternate-day treatment group was low (mean et al. randomized 435 patients to one of three CVVH doses,
delivered Kt/Vurea 0.94 ± 0.11). This resulted in a markedly defined by ultrafiltration rates of 20, 35, and 45 ml/kg/h [2].
elevated time-averaged BUN concentration, which may Mortality was markedly lower in the intermediate- and
have been the reason that this group had a higher incidence high-dose arms (43% and 42%, respectively) compared to
of complications, including gastrointestinal bleeding, altera- the low-dose arm (59%, P < 0.001). The absence of a sur-
tions in mental status, and sepsis [79]. In addition, the vival benefit of high-dose therapy compared to the inter-
alternate-day IHD group had more episodes of intradialytic mediate dose argued against a linear relationship between
hypotension when compared to the daily IHD group: 25 ± 5 dose and outcome. However, it is also possible that a sur-
vs. 5 ± 2, respectively. Nonetheless, while this study demon- vival benefit from high-dose CRRT was blunted by a com-
strated that increasing the delivered dose of IHD from a very peting mortality that was CRRT-dose dependent (i.e.
low dose by increasing the frequency of treatment improves inadequate antimicrobial drug dosing). Similarly, Saudan
survival, one cannot conclude that increasing the frequency et al. reported an overall improved mortality when dialysate
of a more “adequate” alternate-day or a thrice-weekly sched- (CVVHDF) was added to a fixed dose of CVVH thereby
ule to daily therapy would improve survival. increasing the total CRRT dose [86]. In this study, a total of
When examining data related to dose of IHD for AKI, it 102 patients were randomized to receive CVVH with a
is important to recognize that several studies have estab- mean ultrafiltration rate of 25 ± 5 ml/kg/h, and 104 patients
lished that there is a significant discrepancy between pre- were randomized to CVVHDF with a mean ultrafiltration
scribed and successfully delivered dose, therefore more rate of 24 ± 6 ml/kg/h and a mean dialysate flow rate of
frequent monitoring of the delivered dose of IHD with pre- 18 ± 5 ml/kg/h. The addition of diffusive clearance was
and post-BUN is recommended [79–81]. associated with an increase in 28-day survival from 39% to
In summary, when evaluating the dose of IHD, both the 59% (P = 0.03) and increased 90-day survival from 34% to
dose per treatment (Kt/Vurea) and the treatment frequency 59% (P = 0.0005). In contrast, Bouman et al. did not observe
need to be considered. Special considerations in critically ill improvement in mortality with higher doses of CRRT (20
patients with AKI should be noted, such as different urea vs. 48 ml/kg/h) [85]. However, the overall study mortality
generation rates and urea volume of distribution due to vol- of less than 30% suggests that the enrolled patients were
ume overload. The KDIGO consensus guidelines recom- not fully representative of most critically ill patients
mend a delivered Kt/Vurea of 3.9 per week when using IHD with AKI requiring RRT. Finally, Tolwani et al. reported no
for the management of AKI [54]. However, this recommen- difference in outcomes in 200 patients randomized to
dation is based on the arithmetic sum of a Kt/Vurea of 1.3 per
individual HD treatment (assuming three HD sessions per
week and supported by the average weekly Kt/Vurea dose Table 13.3 Total effluent dose according to the different
modalities of CRRT.
delivered in the less-intensive arm of the ATN trial [82]) but
not on urea kinetic modeling [83]. Therefore, a more precise
CRRT modality Total effluent dose (ml/h)
recommendation is that in the absence of specific medical
indications (e.g. severe hyperkalemia, acute drug intoxica- CVVH or Pre-filter replacement fluid rate (ml/h) + post-
tions, volume overload, etc.) in patients with AKI, there is no CVVHF filter replacement fluid rate (ml/h) + fluid
additional benefit to providing HD more frequently than removal rate (ml/h)
every other day, with a delivered single-pool Kt/Vurea of >1.2 CVVHD Dialysate rate (ml/h) + fluid removal rate (ml/h)
per treatment (urea-reduction ratio [URR] >0.67) [84]. CVVHDF Pre-filter replacement fluid rate (ml/h) + post-
filter replacement fluid rate (ml/h) + fluid
removal rate (ml/h) + dialysate rate (ml/h)
CRRT Dose for AKI
CVVH or CVVHF, continuous veno-venous hemofiltration; CVVHD,
In the case of CRRT, the total effluent rate (ml/kg/h) is continuous veno-venous hemodialysis; CVVHDF, continuous
used to determine the dose according to the different veno-venous hemodiafiltration.
Dose of Renal Replacement Therapy in AK 195
Table 13.4 Interventional studies examining higher vs. standard intensity dose of RRT.
Study Participants Interventions Population Mortality AKI recovery
Ronco [2] 425 CVVH 75% post surgical, 12% septic 59% vs. 43% vs.
20 vs. 35 vs. 45 ml/kg/h 42% (P < 0.002)
Bouman [85] 106 CRRT 58% post cardiac surgery, 31.2% vs. 25.7% No difference
20 vs. 48 ml/kg/h 100% respiratory failure, (P = ns)
100% inotrope or pressors
Saudan [86] 206 CVVHF 60% septic 61% vs. 41%
25 ml/kg/h vs. (P = 0.03)
CVVHDF
42 ml/kg/h
Tolwani [87] 200 CVVHDF 54% septic, 77.5% respiratory 44% vs. 51%
20 vs. 35 ml/kg/h failure (P = ns)
Palevsky 1124 CVVHDF 63% septic, 80.6% respiratory 51.5% vs. 53.6% No difference
(ATN) [82] 20 vs. 35 ml/kg/h failure (P = 0.47)
IHD
3/wk HD vs. 6/wk HD
Bellomo 1508 CVVHDF 49.4% septic, 73.9% 44.7% vs. 44.7% No difference
(RENAL) [88] 25 vs. 40 ml/kg/h respiratory failure (P = 0.99)
Joannes-Boyau [89] 140 SVHF 35 vs. HVHF 70 ml/kg/h 100% septic, 97% respiratory 59.2% vs. 62.1% No difference
failure (P = 0.94)
Combes [90] 224 HVHF 80 ml/kg/h + CVVHDF Post cardiac surgery with 36% vs. 36% No difference
vs. SOC with CVVHDF if needed severe shock (P = ns)
CVVH or CVVHF, continuous veno-venous hemofiltration; CVVHDF, continuous veno-venous hemodiafiltration; HVHF, high-volume
hemofiltration; IHD, intermittent hemodialysis; SOC, standard of care; SVHF, standard-volume hemofiltration. P = ns, nonsignificant.
CVVHDF at 20 vs. 35 ml/kg/h of total effluent dose [87]. therapy (OR 1.09, 95% CI 0.86–1.40, P = 0.47). There was
Survival to ICU discharge or 30 days, whichever was ear- also no difference in the duration of RRT, recovery of kid-
lier, was 49% in the patients randomized to the higher dose ney function, or the course of nonrenal organ failure [82].
group, compared to 56% of patients randomized to the The RENAL study randomized 1508 critically ill
lower dose group. patients with AKI to a higher intensity CVVHDF (n = 721,
Subsequently, two large multicenter studies evaluated total effluent dose of 40 ml/kg/h) vs. a lower intensity
dosing of RRT in AKI: the VA/NIH Acute Renal Failure CVVHDF (n = 743, total effluent dose of 25 ml/kg/h)
Trial Network (ATN) study [82] and the Randomized without an option for IHD/PIRRT as initial RRT modal-
Evaluation of Normal versus Augmented Level (RENAL) ity [88]. There was no difference in the primary end-point
Replacement Therapy study [88]. The ATN study utilized a of 90-day mortality (44.7% in each group, P = 0.99). In
strategy that included both HD in hemodynamically stable addition, no differences in 28-day mortality, RRT-
patients and either CVVHDF or SLED/PIRRT in hemody- dependence, total days of RRT, or nonrenal organ failure
namically unstable patients [82]. In the intensive manage- were found [88]. A post hoc study of extended follow-up
ment strategy, HD and SLED/PIRRT were provided daily (median of 43.9 months) did not reveal a difference in
(except Sunday), and CVVHDF was provided with a total mortality or RRT-dependence between these two inten-
effluent dose of 35 ml/kg/h; in the less-intensive strategy, sity strategies of CRRT dosing [91].
HD and SLED/PIRRT were provided every other day Joannes-Boyau et al. conducted a multicenter rand-
(except Sunday), and CVVHDF was dosed at 20 ml/kg/h. In omized trial including 137 critically ill patients with septic
both treatment arms, the target delivered single-pool shock and AKI to evaluate even higher doses of continuous
Kt/Vurea for HD and SLED/PIRRT was 1.2–1.4 per treat- hemofiltration [89]. The study tested standard-volume
ment, with an actual delivered dose during HD of hemofiltration (SVHF, n = 71, 35 ml/kg/h) vs. high-volume
1.32 ± 0.36. Sixty-day all-cause mortality was 53.6% in the hemofiltration (HVHF, n = 66, 70 ml/kg/h) for a 96-hour
563 patients randomized to intensive therapy, compared to period based on the hypothesis that HVHF may improve
51.5% in the 561 patients randomized to the less-intensive hemodynamics and mortality in critically ill septic patients
by improving removal of sepsis-related pro-inflammatory changes, and procedures, imagining studies or surgery
cytokines, but the results showed no difference in the pri- that require temporarily stopping CRRT). Based on the
mary end-point of 28-day mortality (SVHF 59.2% vs. HVHF research in CRRT dose, for most critically ill AKI patients,
62.1%, P = 0.94). There was also no difference in secondary an effective delivered dose 20–25 ml/kg/h is recommended
outcomes that included 60-day and 90-day mortality, dura- [11, 54, 94]. However, to account for this CRRT down time,
tion of mechanical ventilation, duration of RRT, kidney it is reasonable to prescribe 25–30 ml/kg/h of total effluent
recovery, and length of hospital stay [89]. A subsequent dose for CRRT. Finally, it is critical to recognize that the
metanalysis by Clark et al., including four randomized tri- effluent dose of CRRT should be individualized according
als, concluded that insufficient evidence exists of a thera- to the specific needs of the critically ill patient, who may
peutic benefit for routine use of HVHF for critically ill benefit from higher intensity CRRT in specific scenarios
septic patients with AKI [92]. In a more recent study, such as severe metabolic acidosis, severe hyperkalemia, or
Combes et al. conducted a multicenter randomized trial severe hypercatabolic states that require higher intensity
including patients with post-cardiac surgery severe shock solute control. Therefore, it is fundamental to develop a
to test early HVHF (n = 112, 80 ml/kg/h × 48 hours, fol- systematic way of iterative assessment and adjustment
lowed by standard-dose CVVHDF until resolution of shock of RRT goals according to the clinical status of the
and recovery of kidney function) vs. standard of care patient [95].
with delayed CVVHDF only for persistent/severe AKI
(n = 112) [90]. Again, there was no difference in 30-day
PIRRT/SLED Dose in AKI
mortality between groups (36% for both groups). There was
also no difference in the secondary outcomes of 60-day and Dosing of dialysis in PIRRT is less well established. It
90-day mortality, ventilator-free days, kidney recovery, and remains unclear which index is best to guide dose in PIRRT
length of hospitalization [90]. However, it is important to (URR, Kt/Vurea, effluent flow) [96]. As with IHD and CRRT,
note that 43% of the subjects in the standard of care arm PIRRT therapy should be individualized to patients needs to
did not receive any RRT. In addition, patients in the HVHF accomplish the specified daily RRT goal (correction of
arm had more frequent hypophosphatemia, metabolic hyperkalemia, acidemia control, fluid removal). Similarly,
alkalosis, and thrombocytopenia [90]. there are no data to support prolonged exposure to high-
In summary, although some small single-center studies dose PIRRT. One option to determine a PIRRT prescription
have suggested that increased intensity of RRT is associ- when transitioning from a CRRT device is to first decide on
ated with improved outcomes, this approach is not sup- the effluent dose that one would use for that given patient on
ported by data generated from large multicenter studies. CRRT (i.e. 25 ml/kg/h, 35 ml/kg/h, etc.), then calculate the
Differences between study results most likely reflect differ- total daily effluent needs for the patient based on that CRRT
ences among study designs (delivered RRT dose, timing of dose (i.e. 30 ml/kg/h × 80 kg = 2400 ml/h × 24 h = 72 000 ml/
RRT initiation, etc.) and study populations (acuity of criti- day = 72 l/day). Finally, divide total effluent needs by the
cal illness, presence of sepsis, etc.). Wang et al. conducted a number of hours that PIRRT is planned to give hourly efflu-
meta-analysis of seven randomized clinical trials address- ent flow rates on PIRRT (72 l/day/10 h = 7.2 l/h = 120 ml/
ing higher versus standard intensity dose of RRT and con- min) [22]. If using a standard HD device, customizing the
cluded that higher intensity RRT does not affect mortality PIRRT prescription to achieve a Kt/Vurea of ~0.7–0.9 per
outcomes [93]. However, the authors highlighted that treatment for a total of six treatments per week can provide
more participants assigned to higher intensity RRT a weekly Kt/Vurea of 3.0–3.5 based on urea kinetic mode-
remained dependent on RRT at 28 days (29.7% for the ling [83], which is above RRT dosing thresholds although
higher intensity group vs. 24.9% for the standard intensity data are still limited to provide more conclusive recommen-
group, relative risk of 1.15, 95% CI 1.00–1.33, P = 0.05) [93]. dations with the use of PIRRT.
Nonetheless, one should note that kidney recovery (RRT- It is important to remember that PIRRT will have widely
dependence) was a secondary outcome in all the interven- variable impacts on drug dosing of dialyzable drugs, espe-
tional studies included in this meta-analysis and there may cially important for antimicrobial agents. Duration of
be ascertainment bias of the outcome as higher intensity PIRRT, flow rates in PIRRT, and fluid removal during
RRT lowers solute burden and thus possibly reduces solute- PIRRT will all impact a given drug’s pharmacokinetics
driven urine output, which is one of the cornerstone met- (PK). There are expanding PK data for many antimicrobi-
rics of kidney recovery examination. als during PIRRT; however, these data are generally based
As seen with IHD, there is usually a discrepancy between on small PK studies and require that PIRRT be operated
prescribed versus true delivered CRRT dose due to CRRT under the exact operating characteristics that are described
down time (circuit failures, CRRT therapy fluid bag in the published research protocol.
RRT Circuit Consideration 197
RRT Circuit Considerations group and only 9% of the shorter catheter group. Mean
CRRT circuit life was prolonged by 6.5 hours (24 vs
When prescribing and managing acute RRT in AKI, there 17.5 hours, P = 0.001) and dialysis delivery was more effec-
are many factors beyond the RRT modality, timing of RRT tive (91% of dialysis dose delivered vs 81%, I = 0.001) in the
initiation, and RRT dose to consider and that do often longer catheter group [97]. There were no differences in
affect whether RRT application proves successful at achiev- complication rates, including rates of atrial fibrillation. RA
ing the specified daily goal. These circuit-based factors (or caval atrial junction) insertion is safe with current soft
include dialysis vascular access, use of convective versus catheter materials and design and there is a very low rate of
diffusive clearance, type of membrane, circuit anticoagula- vascular or atrial erosions due to dialysis catheters [97].
tion (especially for CRRT), medication dosing, and many Our and other centers’ experience mirror the results of the
others. We will conclude by addressing some of these top- Morgan study. As a result, we recommend use of catheters
ics briefly. that are placed as far distally as possible, with a goal that
the tip be at the caval atrial junction or in the right atrium
when using the IJ insertion site. Catheters placed in the
Dialysis Vascular Access
right IJ vein are preferred as they have a straighter intravas-
Hemodialysis vascular access is of paramount importance cular course and are associated with fewer complications;
when providing acute RRT for AKI. Insertion is often the however we differ from KDIGO in that we favor properly
rate-limiting step to RRT initiation and poorly functioning positioned left IJ vein catheters over catheters in the femo-
vascular access is often the foil of the acute RRT procedure. ral veins.
Given that a well-functioning vascular access is vital for The KDIGO preference of femoral over left IJ sites is
RRT circuit function and, ultimately, being able to achieve based on older trials where left IJ catheters were unques-
the goal-based acute RRT, care must be taken to select the tionably too short [54]. For example, the Cathedia study
optimal site, appropriate catheter size and length, depth group found comparable small solute clearance and similar
and angle of insertion. If the nephrologist is not the proce- catheter and dialysis circuit function comparing IJ venous
duralist for catheter insertion, at a minimum the providers (primarily right) to femoral venous access [98]. However,
should discuss and agree on these points prior to the proce- this study did not specify catheter length or size and 100%
dure to improve the odds of a well-functioning access on of IJ catheters were less than 20 cm in length (usually
initial insertion. 16 cm) and no comment was made regarding ultimate
Data to guide catheter insertion remains limited. KDIGO catheter tip position. This suggests that the performance of
recommends a nontunneled temporary dialysis catheter the IJ catheters may have been worse than had they used a
for RRT in AKI [54]. However, there are cases where AKI longer catheter. Furthermore, in this study, 30% of cathe-
recovery is delayed in whom a tunneled longer-term dialy- ters in the IJ group were placed on the left side, where they
sis catheter may be appropriate to lower the rate of still used relatively short catheters (<20 cm), which likely
catheter-related infection. Additionally, KDIGO recom- contributed to worse left IJ catheter performance com-
mends varying lengths of catheters depending on site of pared to the right IJ and femoral sites [98].
insertion (internal jugular [IJ] versus femoral) and pre- Finally, in a separate study, the Cathedia study group
ferred sites, but these guidelines are not based on evidence compared infection rates between IJ and femoral vascular
and, in our opinion, are inaccurate as the catheter length access sites, demonstrating that catheter-related infection
recommendations are too short and the guidelines eschew rates were similar between IJ and femoral sites [99].
the left IJ site [54]. However, prespecified subgroup analysis did suggest that
Morgan et al. [97] performed a randomized trial on IJ infection rates were different in the group with the highest
dialysis catheter length in AKI requiring CRRT comparing quartile body mass index (BMI) which was BMI > 28 kg/
shorter versus longer catheters with the goal of achieving a m2 [99]. Thus, caution is warranted when selecting cathe-
catheter tip in the right atrium (RA) in the long catheter ter site for those with BMI > 28 kg/m2, which is common in
group and superior vena cava position (SVC) in the short US-based ICUs.
group with the primary end-point being CRRT circuit life.
Catheter lengths were determined by site and group, with
Dialyzer Membranes
the short catheter group getting 15 cm catheters placed in
right IJ veins and 20 cm in left IJ veins (per KDIGO guide- In the modern era, all membranes used for acute RRT in
lines) and the long catheter group getting 20 cm catheters AKI are high-permeability, high-flux, biocompatible mem-
in right IJ veins and 24 cm in left IJ veins. Positioning of the branes, especially for CRRT, where each device manufac-
tip in the RA was achieved in 81% of the longer catheter turer generally has a limited selection of membranes.
With IHD and PIRRT (and with some CRRT machines), with anticoagulation is improved when using heparin,
the clinician is theoretically free to select from a wider regional citrate anticoagulation (RCA), direct thrombin
array of membranes available in a given country. However, inhibitors, and other strategies [101–107]. Systemic antico-
in practice, a given institution only carries a few options agulation with heparin, low-molecular weight heparins,
and these are almost always the standard high-flux, highly and direct thrombin inhibitors obviously carries risk of sys-
permeable, biocompatible, artificial, noncellulose based temic bleeding. RCA is well proven in CRRT and has been
membranes. The typical membrane materials are polyacry- shown to have superior circuit survival effects when com-
lonitrile (AN69), polyarylethersulfone (PAES), and poly- pared to both heparin and no anticoagulation with no sys-
ethersulfone (PES). There are no data suggesting that one temic bleeding risk and few complications that are easily
type of membrane provides improved outcomes. prevented (i.e. systemic hypocalcemia) with attention to
There are copious preclinical and human case series established protocols [101–107]. KDIGO endorses the use
reported from outside of the Unites States on specialized of RCA in CRRT for patients both with and without
membranes and adsorption devices for selective cytokine, increased bleeding risks [54].
endotoxin or other toxin removal, especially in sepsis, but
these specialized membranes are generally not approved
for use outside of research studies. Summary
AKI requiring RRT is common and increasing among hos-

Dialysis Fluids/Solutions and Buffers
pitalized patients, especially among patients with concom-
With all modes of RRT (IHD, PIRRT, CRRT) there is a wide itant critical illness. Generally, prolonged therapies (CRRT
variety of commercially available dialysis solutions for and/or PIRRT) are preferred for the most vulnerable and
both dialysate and replacement fluids with varying electro- critically ill patients. Care must be taken to individualize
lyte concentrations. Solutions (especially for CRRT) can the timing of RRT initiation and the dose of RRT to accen-
alternatively be customized and compounded in a given tuate the benefits of RRT while minimizing the potential
hospital. However, there is a risk of compounding errors harms such as anticoagulation, impacts on drug clearance
and bacterial contamination inherent with compounding (especially antimicrobial agents), and nutrition. Finally, we
large quantities of CRRT fluids [100]. As a result, we rec- strongly encourage a multidisciplinary goal-driven
ommend the use of commercial solutions. Most commer- approach to each day’s RRT endeavors. The specific goals
cially available solutions are primarily bicarbonate-buffered for RRT (CRRT, PIRRT, IHD, PD) should be reassessed
solutions, with near abandonment of lactate or acetate as each day and the RRT plan should be designed to best
the primary buffer over the last 20 years. accomplish the targeted goals (rapid correction of severe
hyperkalemia, tailored fluid removal, etc.). Accountability
toward achieving the desired goals of therapy should be
Anticoagulation Strategies
assured through multidisciplinary quality improvement
An important advantage of IHD and PIRRT is that the use programs [95]. RRT in patients with AKI continues to be
of anticoagulation is usually not required. However, anti- associated with a high mortality rate, but an individual-
coagulation for CRRT circuit survival is strongly encour- ized, dynamic, and goal-driven renal support program
aged by KDIGO and we agree [54]. CRRT circuit survival could optimize patient outcomes.
R
eferences
Hoste EA, Bagshaw SM, Bellomo R, et al. (2015).

1 outcome in critically ill patients. Crit. Care Med. 30 (9):
Epidemiology of acute kidney injury in critically ill 2051–2058.
patients: the multinational AKI-EPI study. Intensive Care 4 Clermont G, Acker CG, Angus DC, et al. (2002). Renal
Med. 41 (8): 1411–1423. failure in the ICU: comparison of the impact of acute
2 Ronco C, Bellomo R, Homel P, et al. (2000). Effects of renal failure and end-stage renal disease on ICU
different doses in continuous veno-venous haemofiltration outcomes. Kidney Int. 62 (3): 986–996.
on outcomes of acute renal failure: a prospective 5 Chertow GM, Christiansen CL, Cleary PD, et al. (1995).
randomised trial. Lancet 356 (9223): 26–30. Prognostic stratification in critically ill patients with
3 Metnitz PG, Krenn CG, Steltzer H, et al. (2002). Effect of acute renal failure requiring dialysis. Arch. Intern. Med.
acute renal failure requiring renal replacement therapy on 155 (14): 1505–1511.
Reference 199
6 Guerin C, Girard R, Selli JM, et al. (2000). Initial versus 19 Karakala, N. and Tolwani, A.J. (2018). Timing of renal
delayed acute renal failure in the intensive care unit. A replacement therapy for acute kidney injury. J. Intensive
multicenter prospective epidemiological study. Rhone- Care Med. 885066618774257. 2019;34(2):94–103.
Alpes Area Study Group on Acute Renal Failure. Am. J. 20 Bagshaw SM, Darmon M, Ostermann M, et al. (2017).
Respir. Crit. Care Med. 161 (3 Pt 1): 872–879. Current state of the art for renal replacement therapy
7 Maher ER, Robinson KN, Scoble JE, et al. (1989). in critically ill patients with acute kidney injury.
Prognosis of critically-ill patients with acute renal failure: Intensive Care Med. 43 (6): 841–854.
APACHE II score and other predictive factors. Q. J. Med. 21 Tolwani, A. (2012). Continuous renal-replacement
72 (269): 857–866. therapy for acute kidney injury. N. Engl. J. Med. 367 (26):
8 Gocze I, Jauch D, Gotz M, et al. (2018). Biomarker-guided 2505–2514.
intervention to prevent acute kidney injury after major 22 Edrees, F., Li, T., and Vijayan, A. (2016). Prolonged
surgery: the prospective randomized BigpAK study. Ann. intermittent renal replacement therapy. Adv. Chronic
Surg. 267 (6): 1013–1020. Kidney Dis. 23 (3): 195–202.
9 Meersch M, Schmidt C, Hoffmeier A, et al. (2017). 23 Ricci, Z; Ronco, C; Bachetoni, A, et al. (2006). Solute removal
Prevention of cardiac surgery-associated AKI by during continuous renal replacement therapy in critically ill
implementing the KDIGO guidelines in high risk patients patients: convection versus diffusion. Crit. Care 10 (2): R67.
identified by biomarkers: the PrevAKI randomized 24 Wald, R; Friedrich, JO; Bagshaw, SM, et al. (2012). Optimal
controlled trial. Intensive Care Med. 43 (11): 1551–1561. Mode of clearance in critically ill patients with Acute
10 Hsu RK, McCulloch CE, Dudley RA, Lo LJ, Hsu CY. Kidney Injury (OMAKI) – a pilot randomized controlled
(2013). Temporal changes in incidence of dialysis- trial of hemofiltration versus hemodialysis: a Canadian
requiring AKI. J. Am. Soc. Nephrol. 24 (1): 37–42. Critical Care Trials Group project. Crit. Care 16 (5): R205.
11 Connor, M.J. Jr. and Karakala, N. (2017). Continuous 25 Brunet, S; Leblanc, M; Geadah, D, et al. (1999). Diffusive
renal replacement therapy: reviewing current best practice and convective solute clearances during continuous renal
to provide high-quality extracorporeal therapy to critically replacement therapy at various dialysate and ultrafiltration
ill patients. Adv. Chronic Kidney Dis. 24 (4): 213–218. flow rates. Am. J. Kidney Dis. 34 (3): 486–492.
12 Demirjian S, Teo BW, Guzman JA, et al. (2011). 26 Hofmann, C.L. and Fissell, W.H. (2010). Middle-molecule
Hypophosphatemia during continuous hemodialysis is clearance at 20 and 35 ml/kg/h in continuous venovenous
associated with prolonged respiratory failure in patients hemodiafiltration. Blood Purif. 29 (3): 259–263.
with acute kidney injury. Nephrol. Dial. Transplant. 26 27 Razavi, SA; Still, MD; White, SJ, et al. (2014). Comparison
(11): 3508–3514. of circuit patency and exchange rates between 2 different
13 Gervasio, J.M., Garmon, W.P., and Holowatyj, M. (2011). continuous renal replacement therapy machines. J. Crit.
Nutrition support in acute kidney injury. Nutr. Clin. Pract. Care 29 (2): 272–277.
26 (4): 374–381. 28 Friedrich, JO; Wald, R; Bagshaw, SM, et al. (2012).
14 Honore PM, De Waele E, Jacobs R, et al. (2013). Hemofiltration compared to hemodialysis for acute
Nutritional and metabolic alterations during continuous kidney injury: systematic review and meta-analysis. Crit.
renal replacement therapy. Blood Purif. 35 (4): 279–284. Care 16 (4): R146.
15 Sgambat, K. and Moudgil, A. (2016). Carnitine deficiency 29 Mehta RL, McDonald B, Gabbai FB, et al. (2001). A
in children receiving continuous renal replacement randomized clinical trial of continuous versus
therapy. Hemodial. Int. 20 (1): 63–67. intermittent dialysis for acute renal failure. Kidney Int. 60
16 Taylor BE, McClave SA, Martindale RG, et al. (2016). (3): 1154–1163.
Guidelines for the provision and assessment of nutrition 30 John S, Griesbach D, Baumgartel M, et al. (2001). Effects
support therapy in the adult critically ill patient: Society of continuous haemofiltration vs intermittent
of Critical Care Medicine (SCCM) and American Society haemodialysis on systemic haemodynamics and
for Parenteral and Enteral Nutrition (A.S.P.E.N.). Crit. splanchnic regional perfusion in septic shock patients: a
Care Med. 44 (2): 390–438. prospective, randomized clinical trial. Nephrol. Dial.
17 Wiesen P, Van Overmeire L, Delanaye P, et al. (2011). Transplant. 16 (2): 320–327.
Nutrition disorders during acute renal failure and renal 31 Augustine JJ, Sandy D, Seifert TH, et al. (2004). A
replacement therapy. J. Parenter. Enteral Nutr. 35 (2): randomized controlled trial comparing intermittent with
217–222. continuous dialysis in patients with ARF. Am. J. Kidney
18 Ostermann M, Joannidis M, Pani A, et al. (2016). Patient Dis. 44 (6): 1000–1007.
selection and timing of continuous renal replacement 32 Uehlinger DE, Jakob SM, Ferrari P, et al. (2005).
therapy. Blood Purif. 42 (3): 224–237. Comparison of continuous and intermittent renal
replacement therapy for acute renal failure. Nephrol. 44 Tonelli, M., Manns, B., and Feller-Kopman, D. (2002).
Dial. Transplant. 20 (8): 1630–1637. Acute renal failure in the intensive care unit: a systematic
33 Vinsonneau C, Camus C, Combes A, et al. (2006). review of the impact of dialytic modality on mortality
Continuous venovenous haemodiafiltration versus and renal recovery. Am. J. Kidney Dis. 40 (5): 875–885.
intermittent haemodialysis for acute renal failure in 45 Pannu N, Klarenbach S, Wiebe N, et al. (2008). Renal
patients with multiple-organ dysfunction syndrome: a replacement therapy in patients with acute renal failure:
multicentre randomised trial. Lancet 368 (9533): a systematic review. JAMA 299 (7): 793–805.
379–385. 46 Nash DM, Przech S, Wald R, O’Reilly D. (2017). Systematic
34 Lins RL, Elseviers MM, Van der Niepen P, et al. (2009). review and meta-analysis of renal replacement therapy
Intermittent versus continuous renal replacement modalities for acute kidney injury in the intensive care unit.
therapy for acute kidney injury patients admitted to the J. Crit. Care 41: 138–144.
intensive care unit: results of a randomized clinical trial. 47 Kielstein JT, Kretschmer U, Ernst T, et al. (2004). Efficacy
Nephrol. Dial. Transplant. 24 (2): 512–518. and cardiovascular tolerability of extended dialysis in
35 Schefold JC, von Haehling S, Pschowski R, et al. (2014). critically ill patients: a randomized controlled study. Am.
The effect of continuous versus intermittent renal J. Kidney Dis. 43 (2): 342–349.
replacement therapy on the outcome of critically ill 48 Zhang L, Yang J, Eastwood GM, Zhu G, Tanaka A,
patients with acute renal failure (CONVINT): a Bellomo R. (2015). Extended daily dialysis versus
prospective randomized controlled trial. Crit. Care 18 (1): continuous renal replacement therapy for acute kidney
R11. injury: a meta-analysis. Am. J. Kidney Dis. 66 (2):
36 Rabindranath K, Adams J, Macleod AM, Muirhead N. 322–330.
(2007). Intermittent versus continuous renal replacement 49 Gangji, A.S., Rabbat, C.G., and Margetts, P.J. (2005).
therapy for acute renal failure in adults. Cochrane Database Benefit of continuous renal replacement therapy in
Syst. Rev. 3: CD003773. subgroups of acutely ill patients: a retrospective analysis.
37 Bagshaw SM, Berthiaume LR, Delaney A, Bellomo R. Clin. Nephrol. 63 (4): 267–275.
(2008). Continuous versus intermittent renal replacement 50 Swartz RD, Bustami RT, Daley JM, et al. (2005).
therapy for critically ill patients with acute kidney injury: Estimating the impact of renal replacement therapy
a meta-analysis. Crit. Care Med. 36 (2): 610–617. choice on outcome in severe acute renal failure. Clin.
38 Phu NH, Hien TT, Mai NT, et al. (2002). Hemofiltration Nephrol. 63 (5): 335–345.
and peritoneal dialysis in infection-associated acute renal 51 Davenport, A. (2009). Continuous renal replacement
failure in Vietnam. N. Engl. J. Med. 347 (12): 895–902. therapies in patients with acute neurological injury.
39 Abe M, Okada K, Suzuki M, et al. (2010). Comparison of Semin. Dial. 22 (2): 165–168.
sustained hemodiafiltration with continuous venovenous 52 Lund A, Damholt MB, Wiis J, et al. (2018). Intracranial
hemodiafiltration for the treatment of critically ill pressure during hemodialysis in patients with acute brain
patients with acute kidney injury. Artif. Organs 34 (4): injury. Acta Anaesthesiol. Scand. 2019;63(4):493–499.
331–338. 53 Uchino S, Bellomo R, Morimatsu H, et al. (2007).
40 Manns B, Doig CJ, Lee H, et al. (2003). Cost of acute renal Continuous renal replacement therapy: a worldwide
failure requiring dialysis in the intensive care unit: practice survey. The beginning and ending supportive
clinical and resource implications of renal recovery. Crit. therapy for the kidney (B.E.S.T. kidney) investigators.
Care Med. 31 (2): 449–455. Intensive Care Med. 33 (9): 1563–1570.
41 Jacka, M.J., Ivancinova, X., and Gibney, R.T. (2005). 54 Kidney Disease Improving Global Outcomes (KDIGO)
Continuous renal replacement therapy improves renal Acute Kidney Injury (AKI) Workgroup (2012). KDIGO
recovery from acute renal failure. Can. J. Anaesth. 52 (3): clinical practice guideline for acute kidney injury. Kidney
327–332. Int. 2 (Suppl. 1): 1–138.
42 Schneider AG, Bellomo R, Bagshaw SM, et al. (2013). 55 Liu KD, Himmelfarb J, Paganini E, et al. (2006). Timing
Choice of renal replacement therapy modality and of initiation of dialysis in critically ill patients with acute
dialysis dependence after acute kidney injury: a kidney injury. Clin. J. Am. Soc. Nephrol. 1 (5): 915–919.
systematic review and meta-analysis. Intensive Care Med. 56 Liborio AB, Leite TT, Neves FM, et al. (2015). AKI
39 (6): 987–997. complications in critically ill patients: association with
43 Kellum JA, Angus DC, Johnson JP, et al. (2002). mortality rates and RRT. Clin. J. Am. Soc. Nephrol. 10 (1):
Continuous versus intermittent renal replacement 21–28.
therapy: a meta-analysis. Intensive Care Med. 28 (1): 57 Gettings, L.G., Reynolds, H.N., and Scalea, T. (1999).
29–37. Outcome in post-traumatic acute renal failure when
Reference 201
continuous renal replacement therapy is applied early vs. 70 Feng YM, Yang Y, Han XL, et al. (2017). The effect of
late. Intensive Care Med. 25 (8): 805–813. early versus late initiation of renal replacement therapy
58 Gaudry, S; Hajage, D; Schortgen, F, et al. (2016). Initiation in patients with acute kidney injury: a meta-analysis with
strategies for renal-replacement therapy in the intensive trial sequential analysis of randomized controlled trials.
care unit. N. Engl. J. Med. 375 (2): 122–133. PLoS One 12 (3): e0174158.
59 Barbar SD, Clere-Jehl R, Bourredjem A, et al. (2018). 71 Pasin, L., Boraso, S., and Tiberio, I. (2019). Early initiation
Timing of renal-replacement therapy in patients with of renal replacement therapy in critically ill patients: a
acute kidney injury and sepsis. N. Engl. J. Med. 379 (15): meta-analysis of randomized clinical trials. BMC
1431–1442. Anesthesiol. 19 (1): 62.
60 Bagshaw, S.M. and Wald, R. (2017). Strategies for the 72 Haines, R.W., Kirwan, C.J., and Prowle, J.R. (2018).
optimal timing to start renal replacement therapy in Continuous renal replacement therapy: individualization
critically ill patients with acute kidney injury. Kidney Int. of the prescription. Curr. Opin. Crit. Care 24 (6): 443–449.
91 (5): 1022–1032. 73 Chawla LS, Davison DL, Brasha-Mitchell E, et al. (2013).
61 Heung, M; Wolfgram, DF; Kommareddi, M, et al. (2012). Development and standardization of a furosemide stress
Fluid overload at initiation of renal replacement therapy is test to predict the severity of acute kidney injury. Crit.
associated with lack of renal recovery in patients with acute Care 17 (5): R207.
kidney injury. Nephrol. Dial. Transplant. 27 (3): 956–961. 74 Lumlertgul N, Peerapornratana S, Trakarnvanich T, et al.
62 Neyra, JA; Li, X; Canepa-Escaro, F, et al. (2016). (2018). Early versus standard initiation of renal
Cumulative fluid balance and mortality in septic patients replacement therapy in furosemide stress test non-
with or without acute kidney injury and chronic kidney responsive acute kidney injury patients (the FST trial).
disease. Crit. Care Med. 44 (10): 1891–1900. Crit. Care 22 (1): 101.
63 Bouchard J, Soroko SB, Chertow GM, et al. (2009). 75 Kashani K, Al-Khafaji A, Ardiles T, et al. (2013).
Fluid accumulation, survival and recovery of kidney Discovery and validation of cell cycle arrest biomarkers
function in critically ill patients with acute kidney in human acute kidney injury. Crit. Care 17 (1): R25.
injury. Kidney Int. 76 (4): 422–427. 76 Vijayan A, Faubel S, Askenazi DJ, et al. (2016). Clinical
64 Vaara ST, Korhonen AM, Kaukonen KM, et al. (2012). use of the urine biomarker [TIMP-2] x [IGFBP7] for acute
Fluid overload is associated with an increased risk for kidney injury risk assessment. Am. J. Kidney Dis. 68 (1):
90-day mortality in critically ill patients with renal 19–28.
replacement therapy: data from the prospective FINNAKI 77 Alge, J.L. and Arthur, J.M. (2015). Biomarkers of AKI: a
study. Crit. Care 16 (5): R197. review of mechanistic relevance and potential
65 Payen D, de Pont AC, Sakr Y, et al. (2008). A positive fluid therapeutic implications. Clin. J. Am. Soc. Nephrol. 10
balance is associated with a worse outcome in patients (1): 147–155.
with acute renal failure. Crit. Care 12 (3): R74. 78 Paganini EP, Tapolyai M, Goormastic M, et al. (1996).
66 Woodward, CW; Lambert, J; Ortiz-Soriano, V, et al. Establishing a dialysis therapy/patient outcome link in
(2019). Fluid overload associates with major adverse intensive care unit acute dialysis for patients with acute
kidney events in critically ill patients with acute kidney renal failure. Am. J. Kidney Dis. 28 (5): S81–S89.
injury requiring continuous renal replacement therapy. 79 Schiffl, H., Lang, S.M., and Fischer, R. (2002). Daily
Crit. Care Med. Crit Care Med 2019;47(9):e753–e760. hemodialysis and the outcome of acute renal failure. N.
67 Karvellas CJ, Farhat MR, Sajjad I, et al. (2011). A Engl. J. Med. 346 (5): 305–310.
comparison of early versus late initiation of renal 80 Overberger, P., Pesacreta, M., and Palevsky, P.M. (2007).
replacement therapy in critically ill patients with acute Management of renal replacement therapy in acute
kidney injury: a systematic review and meta-analysis. kidney injury: a survey of practitioner prescribing
Crit. Care 15 (1): R72. practices. Clin. J. Am. Soc. Nephrol. 2 (4): 623–630.
68 Wang, X. and Jie Yuan, W. (2012). Timing of initiation of 81 Evanson JA, Himmelfarb J, Wingard R, et al. (1998).
renal replacement therapy in acute kidney injury: a Prescribed versus delivered dialysis in acute renal failure
systematic review and meta-analysis. Ren. Fail. 34 (3): patients. Am. J. Kidney Dis. 32 (5): 731–738.
396–402. 82 Palevsky PM, Zhang JH, O’Connor TZ, et al. (2008).
69 Zarbock A, Kellum JA, Schmidt C, et al. (2016). Effect of Intensity of renal support in critically ill patients with
early vs delayed initiation of renal replacement therapy acute kidney injury. N. Engl. J. Med. 359 (1): 7–20.
on mortality in critically ill patients with acute kidney 83 Diaz-Buxo, J.A. and Loredo, J.P. (2006). Standard Kt/V:
injury: the ELAIN randomized clinical trial. JAMA 315 comparison of calculation methods. Artif. Organs 30 (3):
(20): 2190–2199. 178–185.
4 Liang, K.V., Zhang, J.H., and Palevsky, P.M. (2019). Urea

8 best for dialysis dose quantification? Int. J. Artif. Organs
reduction ratio may be a simpler approach for 30 (3): 235–243.
measurement of adequacy of intermittent hemodialysis 97 Morgan D, Ho K, Murray C, et al. (2012). A randomized
in acute kidney injury. BMC Nephrol. 20 (1): 82. trial of catheters of different lengths to achieve right
85 Bouman CS, Oudemans-Van Straaten HM, Tijssen JG, atrium versus superior vena cava placement for
et al. (2002). Effects of early high-volume continuous continuous renal replacement therapy. Am. J. Kidney
venovenous hemofiltration on survival and recovery of Dis. 60 (2): 272–279.
renal function in intensive care patients with acute renal 98 Parienti, JJ; Mégarbane, B; Fischer, MO, et al. (2010).
failure: a prospective, randomized trial. Crit. Care Med. 30 Catheter dysfunction and dialysis performance according
(10): 2205–2211. to vascular access among 736 critically ill adults requiring
86 Saudan P, Niederberger M, De Seigneux S, et al. (2006). renal replacement therapy: a randomized controlled
Adding a dialysis dose to continuous hemofiltration study. Crit. Care Med. 38 (4): 1118–1125.
increases survival in patients with acute renal failure. 99 Parienti JJ, Thirion M, Megarbane B, et al. (2008).
Kidney Int. 70 (7): 1312–1317. Femoral vs jugular venous catheterization and risk of
87 Tolwani AJ, Campbell RC, Stofan BS, et al. (2008). nosocomial events in adults requiring acute renal
Standard versus high-dose CVVHDF for ICU-related replacement therapy: a randomized controlled trial.
acute renal failure. J. Am. Soc. Nephrol. 19 (6): 1233–1238. JAMA 299 (20): 2413–2422.
88 Bellomo R, Cass A, Cole L, et al. (2009). Intensity of 100 Barletta, JF; Barletta, GM; Brophy, PD, et al. (2006).
continuous renal-replacement therapy in critically ill Medication errors and patient complications with
patients. N. Engl. J. Med. 361 (17): 1627–1638. continuous renal replacement therapy. Pediatr. Nephrol.
89 Joannes-Boyau O, Honore PM, Perez P, et al. (2013). 21 (6): 842–845.
High-volume versus standard-volume haemofiltration for 101 Hetzel GR, Schmitz M, Wissing H, et al. (2010).
septic shock patients with acute kidney injury (IVOIRE Regional citrate versus systemic heparin for
study): a multicentre randomized controlled trial. anticoagulation in critically ill patients on continuous
Intensive Care Med. 39 (9): 1535–1546. venovenous haemofiltration: a prospective randomized
90 Combes A, Brechot N, Amour J, et al. (2015). Early multicentre trial. Nephrol. Dial. Transplant.
high-volume hemofiltration versus standard care for 2011;26(1):232–9.
post-cardiac surgery shock. The HEROICS Study. Am. J. 102 Kutsogiannis DJ, Gibney RT, Stollery D, Gao J. (2005).
Respir. Crit. Care Med. 192 (10): 1179–1190. Regional citrate versus systemic heparin anticoagulation
91 Gallagher M, Cass A, Bellomo R, et al. (2014). Long-term for continuous renal replacement in critically ill
survival and dialysis dependency following acute kidney patients. Kidney Int. 67 (6): 2361–2367.
injury in intensive care: extended follow-up of a 103 Monchi M, Berghmans D, Ledoux D, et al. (2004).
randomized controlled trial. PLoS Med. 11 (2): e1001601. Citrate vs. heparin for anticoagulation in continuous
92 Clark E, Molnar AO, Joannes-Boyau O, et al. (2014). venovenous hemofiltration: a prospective randomized
High-volume hemofiltration for septic acute kidney study. Intensive Care Med. 30 (2): 260–265.
injury: a systematic review and meta-analysis. Crit. Care 104 Morabito S, Pistolesi V, Tritapepe L, et al. (2014). Regional
18 (1): R7. citrate anticoagulation for RRTs in critically ill patients
93 Wang Y, Gallagher M, Li Q, et al. (2018). Renal with AKI. Clin. J. Am. Soc. Nephrol. 9 (12): 2173–2188.
replacement therapy intensity for acute kidney injury and 105 Stucker F, Ponte B, Tataw J, et al. (2015). Efficacy and
recovery to dialysis independence: a systematic review safety of citrate-based anticoagulation compared to
and individual patient data meta-analysis. Nephrol. Dial. heparin in patients with acute kidney injury requiring
Transplant. 33 (6): 1017–1024. continuous renal replacement therapy: a randomized
94 Connor, M.J., Jr. and Tolwani, A. (2018). Prescription of controlled trial. Crit. Care 19: 91.
continuous renal replacement therapy in acute kidney 106 Tolwani AJ, Prendergast MB, Speer RR, et al. (2006). A
injury in adults. In UpToDate, (ed. S. Motwani). practical citrate anticoagulation continuous venovenous
95 Neyra, J.A. and Goldstein, S.L. (2018). Optimizing renal hemodiafiltration protocol for metabolic control and
replacement therapy deliverables through multidisciplinary high solute clearance. Clin. J. Am. Soc. Nephrol. 1 (1):
work in the intensive care unit. Clin. Nephrol. Clin Nephrol 79–87.
2018;90(1):1–5. 107 Tolwani, A.J. and Wille, K.M. (2009). Anticoagulation
96 Ratanarat, R., Permpikul, C., and Ronco, C. (2007). Renal for continuous renal replacement therapy. Semin. Dial.
replacement therapy in acute renal failure: which index is 22 (2): 141–145.
203
Primary Diseases of the Kidney

Part 3
205
14
Renal Biopsy
Robert MacGinley1,2 and Emily J. See3,4
1
Department of Renal Medicine, Eastern Health, Box Hill, Australia
2
Renal Medicine Department, Eastern Health Clinical School, Monash University, Melbourne, Australia
3
Department of Intensive Care, Austin Health, Melbourne, Australia
4
School of Medicine, University of Melbourne, Melbourne, Australia
G
RADE Recommendations usually self-limiting and rarely life-threatening. Several
risk factors for complications have been identified, includ-
Introduction ing uncontrolled hypertension, coagulopathy, advanced
age, and reduced kidney function. The evidence behind
The prevalence of kidney disease is rising, with chronic Renal biopsy is seen in table 14.1.
kidney disease (CKD) affecting more than 1 in 10 individu-
als in Australia, and acute kidney injury (AKI) complicat- Indications
ing more than 130,000 hospital admissions each year [1].
Performing a renal biopsy is warranted if it is needed to
Parenchymal renal diseases (glomerular, tubular, or vascu-
diagnose or guide the treatment of a suspected disease
lar) are leading causes of both CKD and AKI (when pre-
which is associated with significant morbidity and/or mor-
and postrenal causes have been excluded), and renal biopsy
tality, if the natural history of the suspected disease can be
remains the gold standard method for their diagnosis,
improved with treatment, and if the treatment for the sus-
assessment of prognosis, and management. There have
pected disease is appropriate for and acceptable to the
been many advances in renal biopsy techniques since its
patient. Renal biopsy is useful to assess the chronicity and
inception that have greatly enhanced the success and
activity of a disease, while serial renal biopsy may be per-
safety of this procedure. These have included modifica-
formed to assess the natural history or response to treat-
tions to patient positioning, introduction of real-time imag-
ment. Although there are no defined criteria for performing
ing guidance, and the use of automated biopsy needles.
a renal biopsy, there are many accepted indications:
Performing a kidney biopsy has been shown to change the
prebiopsy clinical diagnosis in 44% of cases and to change ●● Nephrotic syndrome is the most frequent indication for
therapy in 30% of cases [2]. renal biopsy. In this setting, biopsy provides information
The procedure itself involves the removal of tissue from for the diagnosis, staging, and activity of disease [4].
the renal cortex for histopathological examination by a spe- ●● Persistent proteinuria >1 g per day or persistent pro-
cialist pathologist. The percutaneous approach is standard, teinuria <1 g per day in the presence of haematuria or
and the procedure is typically guided by ultrasound and renal dysfunction are also common reasons to perform
performed under local anesthesia, with the patient in the a renal biopsy. In the presence of haematuria, biopsy
prone position. Both coaxial and noncoaxial needle tech- should only be undertaken after malignancy has been
niques have been described. Renal biopsy has been found excluded.
to be safe and efficacious, with an acceptably low risk of ●● Systemic diseases with suspected renal involvement (e.g.
complications and with sufficient tissue being obtained in systemic lupus erythematosus, multiple myeloma, amy-
more than 95% of cases [3]. The most common complica- loid) often necessitate renal biopsy to guide decisions
tions are minor and major bleeding, both of which are surrounding prognosis and treatment.
206 Renal Biopsy
Table 14.1 Recommendations for renal biopsies.
Strength of
Treatment Recommendation Certainty of evidence recommendation
Education We recommend that patients and their carers be provided Low or very low Weak [1, 2]
with relevant information regarding the renal biopsy
procedure, including why they are having one, and the
pre-and post-biopsy complications. It should be provided with
clarity and cover the topic in a way that is appropriate for their
level of health literacy.
Pre-biopsy medication We recommend that aspirin is continued in patients with high High Strong [3]
cardiovascular risk or ceased 3 (to reduce major bleeding) or
7 days prior (to reduce minor bleeding) in patients with low
cardiovascular risk. In patients receiving anticoagulation,
bridging anticoagulation is recommended in patients with
mechanical valves and in those with a high thrombotic risk.
We suggest that desmopressin should be used sparingly and Low or very low
with caution.
Positioning for renal We recommend that native renal biopsies be performed in a Moderate Weak [4, 5]
biopsy prone position, that transplant renal biopsies be performed in
a supine position, and that biopsies in patients who are obese
or who have difficulty lying prone (pregnancy or respiratory
difficulty) be performed in the in a supine anterolateral
position.
Imaging We recommend that native renal biopsies be performed using Moderate (US guidance) Weak [6]
real-time ultrasound guidance, with the use of computed Low (CT guidance)
tomography localization being used for difficult biopsies.
Biopsy needles We recommend the use of spring-loaded 16-gauge needles for Moderate Moderate [7]
native and transplant biopsies.
Post-biopsy care We recommend strict bed rest following a renal biopsy and a Moderate Moderate [8]
6-h observation period in patients at low risk of bleeding.
Patients at high risk of bleeding should be observed overnight.
Routine post-biopsy imaging or hemoglobin measurement is Low (post biopsy tests)
not recommended.
Management of We recommend the use or radiological (glue, gel or coiling) or Low Weak
bleeding complications surgical management to arrest persistent post-biopsy bleeding.
GRADE assessment of the certainty of the evidence (9): High: This research provides a very good indication of the likely effect. The likelihood
that the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood that
the effect will be substantially different* is moderate. Low: This research provides some indication of the likely effect. However, the likelihood
that it will be substantially different* is high. Very low: This research does not provide a reliable indication of the likely effect. The likelihood
that the effect will be substantially different* is very high. *Substantially different = a large enough difference that it might affect decision-making.
●● Unexplained AKI is an indication for renal biopsy, biopsy to assess for evidence of allograft rejection, drug
following exclusion of prerenal or postrenal disease. toxicity, and recurrence of primary disease.
Glomerulonephritis, interstitial nephritis, and systemic
diseases with renal involvement can all present with The decision of whether to proceed with renal biopsy in a
AKI. diabetic patient can be particularly challenging. Diabetes has
●● Unexplained CKD may prompt investigation with renal become the most common cause of CKD in the western
biopsy, although the diagnostic yield is typically low and world [5] and diabetic nephropathy can coexist with other
the risk of bleeding complications is Higher than biopsy causes of parenchymal renal disease. It is currently difficult to
of a non disease kidney due to possible smaller and predict the degree of diabetic renal injury based on clinical or
harder to biopsy kidneys. Biopsy in this setting can be laboratory findings alone. New-onset nephrotic-range protein-
useful in assessing the likelihood of disease recurrence uria or hematuria or a rapidly progressive decline in renal
following future kidney transplantation. function are common indications for renal biopsy in this popu-
●● Transplant dysfunction (impaired biochemical function, lation. A recent meta-analysis [6] examining the predictors of
proteinuria or hematuria) is another indication for renal nondiabetic kidney disease in 4876 patients across 48 studies
Procedur 207
(range from 3% to 82.9% incidence) found that higher sys- P

rocedure
tolic blood pressure, worse diabetes control (worse diabetic
control), longer diabetes duration, and the presence of Technique
retinopathy reduced the positive predictive value of renal
biopsy in diagnosing nondiabetic renal disease. Percutaneous renal biopsies are commonly performed
under local anesthetic. Occasionally, in the pediatric setting
Contraindications or in anxious individuals, conscious sedation or general
anesthesia is used, which requires the patient to be fasted
Absolute contraindications to percutaneous kidney biop- and monitored closely during and after the procedure.
sies include coagulopathy and uncontrolled hypertension Lignocaine (1% or 2%) is infiltrated from the skin to the
(>160/100 mmHg). Additional relative contraindications renal capsule, and the biopsy needle is advanced along the
include the presence of small kidneys, hydronephrosis, anesthetized tract. The patient is asked to pause at full
pyelonephritis, abnormal anatomy (e.g. polycystic kidney inspiration such that the lower pole of the kidney is in con-
disease), a solitary kidney, technical difficulties (patients tact with the needle tip, and the triggering mechanism of
who are obese, unable to lie flat, or uncooperative), and the the needle is then released. The lower pole of the left kidney
special circumstances of pregnancy, amyloid kidney, and is usually selected for biopsy because it tends to be more
the pediatric setting. caudal and superficial than the right kidney, is away from
When a contraindication to percutaneous renal biopsy the major vessels and organs, and is easiest for right-handed
exists, alternative methods of obtaining renal tissue can be operators. Between one and three cores of tissue are sam-
considered, including open, laparoscopic, and transjugular pled and the procedure time is generally between 15 and
approaches. These differ in terms of their safety profile and 30 minutes. No fewer than seven glomeruli and one artery
sample adequacy. Open and laparoscopic biopsies are asso- are required for a sample to be considered adequate, while
ciated with a low risk of bleeding and excellent tissue ade- optimally 10 glomeruli and two arteries should be provided
quacy, but require general anesthesia and longer recovery for analysis. Real-time evaluation for adequacy by a pathol-
time. Sample adequacy is lower with transjugular renal ogist or scientist confirms adequate tissue while minimiz-
biopsy and there is a risk of contrast-induced nephropathy. ing the number of passes.
Although the reported complication rates of transjugular
biopsies are similar to those seen in the percutaneous
approach, the baseline risk of individuals clearly differs Positioning
significantly such that direct comparison of complications Although the percutaneous native renal biopsies were tra-
rates between the approaches cannot be performed. ditionally performed in an upright position, they are now
routinely performed with the patient in the prone position,
Epidemiology often with a pillow or sandbag under the umbilicus, which
Accurate estimation of the frequency of renal biopsies is used to splint the kidney. The lateral decubitus position
across the world is particularly challenging. The capacity is another possible approach, which has not been found to
of a country to perform kidney biopsies, and the number of be different to prone positioning in terms of procedure
kidney biopsies performed, will depend on their healthcare time or risk of complications [7], but may be more comfort-
infrastructure and the availability of trained practitioners able for some patients, especially those who are unable to
(nephrologists, radiologists, scientists, pathologists). The lie prone (e.g. during pregnancy). Transplant biopsies are
Global Kidney Health Atlas recently reported that patho- routinely performed in the supine position; the supine
logical services for renal biopsy were routinely available for anterolateral approach may be appropriate in obese
23% of countries, but were rarely or never available in 24% patients. In the setting of kidney transplants, consideration
and 14% of countries, respectively. In a recent systematic should be given to whether the transplanted kidney is
review of national biopsy registries, regional databases, located in the intraperitoneal (e.g. simultaneous kidney
and single-center cohorts, Fiorentino et al. [6] also found pancreas transplant) or extraperitoneal space.
that there was significant variation in the rate of renal
biopsies between countries, ranging from 10.8 per million
Imaging Guidance
population per year in Romania and Serbia to 215 per mil-
lion population per year in the United States and Australia. Historically, renal biopsies were done blind or by palpa-
Although hypothesis-generating, their analysis was poten- tion. However, with the widespread availability and ease of
tially confounded by their consideration of different time use of imaging techniques, ultrasound-guided renal biopsy
periods (before year 2000 and after), reporting and coding has become the standard of care for native and transplant
bias, and heterogeneity of the indications for biopsy. kidney biopsies. Ultrasound may be used for real-time
208 Renal Biopsy
guidance or for surface marking and approximating needle Practitioner

direction and depth. Computed tomography (CT)-guided
Although renal biopsy can be performed by a nephrologist,
renal biopsy may be considered preferable in obese patients,
radiologist, or other practitioner, it is well accepted that
patients with complicated anatomy, and those with inade-
experienced proceduralists have lower complication rates.
quate kidney imaging on ultrasound since it may offer better
In a survey of practitioners in Europe, renal biopsies were
resolution and tissue contrast. The Caring for Australasians
performed in 60% of centers by nephrologists, in 30% of
with Renal Impairment (CARI) guidelines [8] have recently
centers by radiologists, and in 10% of centers by both [11].
summarized the evidence comparing imaging modalities,
In many countries, competency in performing renal biop-
reporting that ultrasound guidance is superior to blinded
sies is a compulsory component of nephrology training.
renal biopsy with respect to the number of glomeruli, num-
Two-thirds of nephrology trainees in the United States
ber of passes, and risk of bleeding complications, but there
have learned renal biopsy, but few perform it in their con-
are insufficient data directly comparing CT and ultrasound
sultant practice. Both Australia and the United Kingdom
guidance. Many of the published studies have been con-
have recommended that learning skills in renal biopsy be
founded by the use of different needle types and historical
part of nephrology training. There is currently no formal
cohorts.
accreditation process for nonproceduralists.
Needle Size
Although a number of needle sizes are available, the major-
P
reprocedure Care
ity of renal biopsies are performed using 14-gauge (internal
diameter 900–1000 μm), 16-gauge (internal diameter 600–
Prebiopsy Evaluation
700 μm), or 18-gauge (internal diameter 300–400 μm) nee-
dles. Both 14-gauge and 16-gauge needles are used in The prebiopsy evaluation is an essential component of
adults, while 18-gauge needles are generally reserved for the renal biopsy procedure and should include a detailed
the pediatric setting since the internal diameter of the nee- clinical history and examination in addition to targeted
dle is only marginally larger than an adult glomerulus laboratory testing and imaging to evaluate the risk of
(200–250 μm) [9, 10]. When selecting a needle size, bleed- bleeding. Relevant aspects of the clinical history relate to
ing risk and sample adequacy are the key considerations. knowledge and previous experience of renal biopsy,
The conclusion from the recently published CARI guide- mobility and ability to lie prone, psychological condition,
lines was that 16-gauge needles obtained more diagnostic and level of comprehension (e.g. anxiety, confusion, lan-
tissue with the smallest number of passes (predominantly guage barrier). Enquiring about the patient’s current and
compared to 18-gauge needles) and with the smallest num- recent medications, as well as previous bleeding difficulty
ber of minor and major bleeding complications (predomi- following procedures (e.g. dental work) or a history of a
nantly compared to 14-gauge needles). bleeding disorder, is important to assess their risk of
bleeding. In all cases, written informed consent must be
obtained from the patient or their legally authorized rep-
Sample Preparation
resentative prior to the procedure. This may necessitate
Preparation of the renal tissue for processing requires provi- the use of interpreters.
sion of clinical information to the scientist or pathologist and The recently published CARI guidelines [12] have high-
careful handling of the tissue. Normal saline should be used lighted the significant psychosocial impact of renal biopsy
to wash the tissue from the needle. The samples are subse- on patients and caregivers, and a qualitative study by the
quently placed in the appropriate fixatives and transport same group identified the need for better communication,
media for light microscopy, immunofluorescence, and elec- education, and pre-and post-procedure psychological sup-
tron microscopy. If no anatomical pathology scientist is pre- port. Similar studies in other percutaneous biopsies
sent, specimens may be placed in normal saline for transport (including breast, cervical, liver, and prostate) have
(assuming sample examination will occur within a couple of reported similar findings. All patients should receive psy-
hours). For samples being sent to a referral center, tissue can chological preparation prior to their procedure to ensure a
be divided and prepared prior to dispatch. Fresh tissue is patient-centered, holistic approach to care and to prevent
required for immunofluorescence to preserve antigenic the adverse physiological sequelae of anxiety, including
activity. A sample for electron microscopy should be placed hypertension, a heightened perception of pain, and a delay
in 3% glutaraldehyde, if available. The remainder of the tis- in recovery time. Ideally, patients should be provided with
sue should be placed in formalin, for light microscopy only. information regarding the procedure itself, the sensations
Preprocedure Car 209
the patient can expect to feel (e.g. sharpness, numbness), associated with an increased risk of major bleeding events
and the expected post-procedure recovery (e.g. mobiliza- (e.g. need for transfusion, interventional, or surgical proce-
tion, return to normal activities). dure). Therefore, the current recommendation is to con-
tinue aspirin in patients at high risk of cardiovascular
events, including those with a history of coronary stent
Investigations
(particularly within 3 months of bare metal stent or
Investigations should include a full blood count to quantify 12 months of drug-eluting stent insertion), symptomatic
hemoglobin and platelet count, a coagulation profile to myocardial ischemia or peripheral vascular disease (includ-
assess activated partial thromboplastin time and prothrom- ing patients with a peripheral stent), or previous ischemic
bin time, and an ultrasound to diagnose anatomic abnor- stroke. In patients at low risk for a cardiovascular event,
malities, including a solitary kidney, small or polycystic aspirin should be withheld for 3 or 7 days to prevent major
kidneys, hydronephrosis, and vascular abnormalities. and minor bleeding complications, respectively. There are
Although abnormal coagulation should be corrected prior limited data examining the safety of continuing clopidogrel
to biopsy, there is in sufficient evidence to inform the spe- at the time of biopsy, and most guidelines recommend that
cific international normalized ratio (INR) and the platelet it should be withheld for 7 days prior. Aspirin and clopi-
threshold at which proceeding with renal biopsy is safe. In dogrel can be recommenced 24–48 hours after an uncom-
most published studies, patients undergoing renal biopsy plicated procedure since most complications will have
were required to have an INR below or equal to 1.5, and a occurred within this time.
platelet count greater than 50 000/μl. However, appropriate
values should be individualized, taking into account other
Anticoagulants
factors associated with the risk and consequences of bleed-
ing, such as the use of antiplatelet and anticoagulant agents, In patients who are taking oral anticoagulants there is a
and the presence of hypertension, uremia, complex anat- need to balance the bleeding risk against the thrombosis
omy, and anemia. Transfusion should be considered if risk. Warfarin should be ceased 5 days prior to renal biopsy
serum hemoglobin is<80 g/l and is recommended if <70 g/l. in all patients. Bridging anticoagulation should be used in all
The use of bleeding time in assessing the risk of bleeding patients with a mechanical valve, antiphospholipid syn-
complications remains controversial. It is a poor predictor drome, high-risk thrombophilia, or venous thromboembo-
of the risk of surgical bleeding and its clinical benefit in the lism within 3 months. The role of bridging anticoagulation is
setting of percutaneous renal biopsy has not been stud- less clear in atrial fibrillation. In low-risk patients, interrup-
ied. [13]. Although a correlation between bleeding time tion of warfarin without bridging anticoagulation appears to
and “clinical” bleeding in patients with uremia has been be noninferior to the use of bridging anticoagulation for the
reported, prior knowledge of bleeding time has not been prevention of thromboembolism [17]. However, these
shown to reduce bleeding complications in native or trans- results may not be generalizable to high-risk atrial fibrilla-
plant biopsies in the limited available studies [14, 15]. tion patients, in whom the decision should be individualized
according to patient risk. For patients who receive bridging
anticoagulation, low molecular weight heparin should be
Antiplatelet Agents
withheld for 24 hours, while unfractionated heparin should
The use of antiplatelet medications is very common in the be ceased 4–6 hours prior to biopsy. Anticoagulation can be
renal biopsy population. In many centers, these agents are recommenced 24–48 hours following an uncomplicated pro-
routinely discontinued between 5 and 7 days prior to the cedure. For patients requiring hemodialysis, the biopsy
procedure. However, in a patient with preexisting cardio- should be performed at least 6 hours after treatment (if hepa-
vascular disease this approach increases the risk of a peri- rin is administered) and the use of intradialytic anticoagula-
procedure cardiovascular event [16], and results in delayed tion should be avoided the following day. It is important to
diagnosis and treatment. Specifically, cessation of aspirin note that there are no data on the effect of new oral antico-
has been shown to increase the risk of stroke by three- to agulants on renal biopsy complication rates.
fourfold and to increase the relative risk of recurrent
myocardial infarction by 40%, especially when performed
Desmopressin
soon after a myocardial infarct or percutaneous coronary
intervention. A heightened risk of bleeding in the context of uremia is
Although the continuation of aspirin has been associ- well described [18]. It is likely that many factors contribute
ated with an increased risk of minor bleeding complica- to this bleeding propensity, including dysfunctional von
tions (e.g. hemoglobin drop 1.0 g/dl), it has not been Willebrand factor (vWF), platelet membrane abnormalities
210 Renal Biopsy
and impaired aggregation, anemia, and increased prostacy- P

ost-procedure Care
clin and nitric oxide levels. The use of procoagulants prior
to biopsy (e.g. desmopressin acetate) has been suggested as The post-biopsy period is critical to detection of major
a potential strategy to lessen minor and major bleeding. complications of renal biopsy, including hemodynamic
Desmopressin is a synthetic analog of antidiuretic hor- compromise, macroscopic haematuria, urinary tract
mone, which increases platelet adhesion to the blood ves- obstruction, and severe loin pain. Although mild pain or
sel wall by augmenting plasma levels of factor VII and discomfort is expected post biopsy, severe pain over the site
vWF. It has been shown to normalize bleeding time in may indicate bleeding. Based on the findings of early stud-
uraemia for up to 4–8 hours (Vigano GL, Mannucci PM, ies [22], patients were historically observed overnight fol-
Lattuada A, et al. Subcutaneous desmopressin (DDAVP) lowing a renal biopsy. However, several advances in the
shortens the bleeding time in uremia. Am J Hematol. renal biopsy technique have subsequently enhanced the
1989;31(1):32–5.] from CARI) when delivered intrave- safety of the procedure. The optimal timing of observation
nously at a dose of 0.3 μg/kg over a period of after biopsy in the modern day is debatable and should be
20–30 minutes. guided by the likelihood of complications. In general, at
Despite its use in many units, there is little evidence to least 4–8 hours of observation are recommended following
support the clinical benefit of desmopressin. A single rand- an uncomplicated biopsy in a low-risk patient, but as many
omized controlled trial of 162 patients undergoing native as one-third of complications may occur after 8 hours [14].
renal biopsy compared the use of desmopressin to placebo, Almost all complications are detected by 24 hours [23]. For
showing a 16.8% reduction in minor bleeding, but no differ- patients at high risk of complications, and for those who
ence in post-biopsy hemoglobin [19]. Of note, all patients in live remotely, overnight observation is often warranted.
this study had preserved renal function and normal coagu- During this observation time, vital signs are monitored
lation, so the generalizability of their findings is unclear. frequently and every voided urine sample is examined for
Moledina et al. conducted an observational study of 159 macroscopic hematuria. Unless contraindicated due to
hospitalized adults who underwent native kidney biopsy in fluid restriction, a high fluid intake is encouraged to reduce
the United States to investigate acute kidney disease. After the risk of clot formation in the urinary tract. Following
adjustment for blood urea nitrogen, desmopressin was asso- discharge, patients should be instructed to seek medical
ciated with a lower risk of blood transfusion (odds ratio attention in the setting of escalating pain, recurrence or
0.24, 95% confidence interval [CI] 0.06–0.88). In contrast, a worsening of hematuria, or inability to pass urine. Most
retrospective study of 22 renal biopsies in children showed published protocols also suggest a period of abstinence
no difference between desmopressin and placebo. The most from intense physical labor, exercise or heavy lifting for
serious adverse event of desmopressin (severe hypona- 1–2 weeks after a biopsy, but there is no evidence to support
tremia) is reported in 1 in 10 000 cases, but other symptoms, this recommendation.
including hypotension, tachycardia, facial flushing, nausea, The utility of routine post-biopsy hemoglobin testing or
and abdominal pain, can also occur [20, 21]. imaging in predicting clinical complications has not been
Other treatments, including tranexamic acid, recombi- demonstrated. The presence and size of post-biopsy hema-
nant erythropoietin to achieve a hematocrit >30%, conju- tomas on routine imaging does not correlate with the
gated estrogens, and cryoprecipitate, have not been development of clinically significant complications [24],
formally investigated in the setting of renal biopsy. The although the absence of a hematoma has a high negative
impact of dialysis on uremic bleeding is also unclear. predictive value for bleeding. Routine haemoglobin testing
and imaging following biopsy are therefore of limited util-
ity in patients who are otherwise asymptomatic.
Bleeding Prediction Scores
Mejia-Vilet et al. recently developed a score to predict bleed-
ing complications using a high-risk population of hospital- C
omplications
ized patients undergoing renal biopsy in Mexico. They found
that CKD features on ultrasound, blood urea nitrogen The most common complication after renal biopsy is bleed-
>50 mg/dl, haemoglobin <11 g/l, and platelet count <150/ ing, which may manifest as hematuria, perinephric hema-
mm3 were independent predicters of bleeding complications toma, loin pain, hypotension, or a drop in hemoglobin. Most
following renal biopsy. The discriminative power of their minor bleeding can be managed conservatively, but moder-
model was supported in a small, prospective validation ate bleeding, indicated by severe pain, large perinephric
cohort, but external validation has not been undertaken. hematoma or hypotension, should be initially managed with
Complication 211
blood transfusion. In the setting of severe bleeding, with tions tend to be lower than in native biopsy: hematoma in
hemodynamic instability and failure to respond to conserva- 1%, major and minor bleeding in 2–3%, hemoglobin decline
tive measures, radiological intervention or surgical manage- in <1%, and graft loss in <0.04% [29, 30].
ment is required to secure hemostasis.
Estimating the risk of bleeding complications after biopsy
is made difficult by differences in how bleeding is defined Monoclonal Gammopathies and Paraprotein
and diagnosed between studies, by variation in patient Disease
selection, technique, and needle type, and by likely report- It has been suggested in historical cohorts that patients with
ing bias. A recent meta-analysis of 118 064 biopsies monoclonal gammopathies and paraprotein diseases are
(87 studies selected) showed complication rates included at increased bleeding risk (Colm C. Magee Nephrology
perinephric hematoma in 11% of the kidney biopsies, mac- Dialysis Transplantation, Volume 21, Issue 12, December
roscopic haematuria in 3.5%, bleeding requiring blood 2006, Pages 3601–3602, https://doi.org/10.1093/ndt/gfl578).
transfusion in 1.6%, and interventions to stop bleeding in However, large cohort studies have not supported this,
0.3%. Death attributed to native kidney biopsy was rare, except in reporting a higher risk of hemoglobin decrease
occurring in only an estimated 0.06% of all biopsies [25]. >1 g/dl [31–33].
Complication rates were higher in hospitalized patients and
those with AKI. Recent further reassurance that outpatient
biopsy is safe when a low-risk patient group is selected can Pregnancy
be seen in a large single-center uncontrolled study of the
outcomes of 824 outpatient kidney biopsies (448 native and In the setting of pregnancy, the risk of biopsy must be
326 transplant kidney) with both a minor or major compli- weighed against any potential benefit of diagnosis and treat-
cation identified in 11.4% of patients within 4–6 hours post ment, especially considering that treatment options may be
biopsy [26]. An older meta-analysis of almost 10 000 biop- limited due to teratogenicity. Renal blood flow is increased in
sies [27] reported that macroscopic hematuria occurred in pregnancy, which may increase complication rates, and mod-
3.5% (95% CI 0.3–14.5%) of patients, while 0.9% (95% CI ification of biopsy technique may be necessary (e.g. patient
0.4–1.5%) of patients received a blood transfusion and only positioning). A recent systematic review of 39 studies (mostly
0.6% (0.4–0.8%) required angiography. The incidence of per- retrospective case series) of renal biopsies performed during
inephric hematoma is likely to be much higher than other pregnancy or in the perinatal period [34] found that the
reported complications. Based on studies performing CT- complication rate was 7% during pregnancy, compared to
guided biopsies in a historical cohort, as many as 91% of only 1% after delivery. All major complications occurred dur-
patients have evidence of a hematoma on imaging [28]. ing gestational weeks 23–28. Studies in which biopsy was
In the older meta-analysis, the study-level predictors of performed for suspected glomerulonephritis or preeclampsia
bleeding complications included the use of 14-gauge nee- reported that therapy was altered in 66% of cases. Bleeding
dles (compared with 16-gauge and 18-gauge), higher serum Complications after Pediatric Kidney Biopsy- A systematic
creatinine (>2.0 mg/dl), AKI in >10% of cases, a baseline Review and Meta-analysis .Charles D. Varnell, Hillarey K.
hemoglobin <12 g/dl, a mean age >40 years, and systolic Stone and Jeffrey A. Welge Clinical Journal of the American
blood pressure >130 mmHg [27]. The safe blood pressure Society of NephrologyJanuary 2019, 14 (1) 57-65; DOI:
threshold to prevent bleeding is debatable, with most stud- https://doi.org/10.2215/CJN.05890518.
ies and guidelines recommending a maximum acceptable
systolic blood pressure of 140 or 160 mmHg.
Pediatrics
Other rare complications include arteriovenous fistula
formation (0.1–0.5%), infection, pneumothorax, and injury Renal biopsy in the pediatric setting has the added difficulty
to adjacent organs, including the liver, spleen, pancreas, of small kidneys and variable patient cooperation. A recently
and colon. Nephrectomy is required after <0.1% of cases published systematic review of 23 studies including 5504
and the risk of death is similarly low. renal biopsies reported that between 11% and 18% of pediat-
ric patients develop a hemotoma after biopsy, but only 0.9%
require a blood transfusion. As few as 0.7% of biopsies
Transplant Biopsies
resulted in an additional intervention due to post-biopsy
There are many single-center studies in the literature complications. Meta-regression suggested that the use of
addressing the safety of transplant biopsy in the contempo- real-time ultrasound guidance did not modify the risk of
rary setting of spring-loaded needles and ultrasound hematoma or requirement for blood transfusion or further
localization and guidance. The reported rates of complica- intervention. In general, pediatric biopsies are performed
212 Renal Biopsy
with a spring-loaded device and 18-gauge needle [35]. The key information that can be used in conjunction with
number of passes should be 3 in 80% of cases. There does biochemical and urinary testing. In modern-day practice
not appear to be a difference in the rate of complications with automated, spring-loaded needles and ultrasound
between inpatient and outpatient procedures. guidance, it has very low complication rates and high
diagnostic yield, altering patient management in many
C
onclusion cases. Developing a patient-centred approach to renal
biopsy, with appropriate education and support, will be
Renal parenchymal biopsy remains an invaluable diag- the next step to further improving the safety and accept-
nostic and prognostic tool for nephrologists, providing ability of the procedure.
R
eferences
1 Australian Institute of Health and Welfare (2015). Acute 11 (2015). Amore a on behalf of ESPN WG and ERA-
kidney injury in Australia: a first national snapshot. Cat. EDTAWG. Nephrol Dial Transplant 30 (3): iii104–iii105.
no. PHE 190. Canberra: AIHW. 12 Champion de Crespigny, P., See, E., Lopez-Vargas, P. et al.
2 Turner, M.W., Hutchinson, T.A., Barré, P.E. et al. (1986). (in press). Biopsy education for consumers – CARI
A prospective study on the impact of the renal biopsy in guideline. Nephrology. Identifying and integrating patient
clinical management. Clin. Nephrol. 26 (5): 217–221. and caregiver perspectives in clinical practice guidelines
3 Korbet, S.M. (2002). Percutaneous renal biopsy. Semin. for percutaneous renal biopsy Talia Gutman, Pamela
Nephrol. 22 (3): 254–267. Lopez-Vargas, Karine E Manera, Jonathan C Craig,
4 Bhagavatula, S.K. and Shyn, P.B. (2017). Radiol. Clin. Martin Howell, David Tunnicliffe, Laura J James, Rob
North Am. 55: 359–371. MacGinley, Emily See, Jeffrey Wong, David Voss … See all
5 Gonzalez Suarez, M.L., Thomas, D.B., Barisoni, L., and authors First published: 23 May 2018 https://doi.
Fornoni, A. (2013). Diabetic nephropathy: is it time yet org/10.1111/nep.13406.
for routine kidney biopsy? World J. Diabetes 4 (6): 13 Peterson, P., Hayes, T.E., Arkin, C.F. et al. (1998). Arch.
245–255. Surg. 133: 134–139.
6 Fiorentino, M., Bolignano, D., Tesar, V. et al. (2017). Renal 14 Whittier, W.L. and Korbet, S.M. (2004). Timing of
biopsy in patients with diabetes: a pooled meta-analysis complications in percutaenous renal biopsy. Journal of
of 48 studies. Nephrol. Dial. Transplant. 32: 97–110. the American Society of Nephrology (JASN) 15: 142–147.
7 Gesualdo, L., Cormio, L., Stallone, G. et al. (2008). 15 Stiles, K.P., Hill, C., LeBrun, C.J. et al. (2001). Renal
Percutaneous ultrasound-guided renal biopsy in supine biopsy. J. Nephrol. 14: 275–279.
antero-lateral position : a new approach for obese and 16 Collet, J.P., Montalescot, G., Blanchet, B. et al. (2004).
non-obese patients. Nephrol. Dial. Transplant. 23 (3): Impact of prior use or recent withdrawal of oral
971–976. antiplatelet agents on acute coronary syndromes.
8 Rob MacGinley, Paul J Champion De Crespigny, Pamela Circulation 110: 2361–2367.
Lopez-Vargas, Karine Manera, Solomon Menahem, John 17 Perioperative Bridging Anticoagulation in Patients with
Saunders, Emily See, David Voss, Jeffrey Wong First Atrial Fibrillation Douketis, J.D., Spyropoulos, A.C.,
published: 06 September 2019, https://doi.org/10.1111/ Kaatz, S. et al. (2015). N. Engl. J. Med. 373 (9): 823–833.
nep.13662. 18 Hedges, S.J., Dehoney, S.B., Hooper, J.S. et al. (2007).
9 Riehl, J., Maigatter, S., Kierdorf, H. et al. (1994). Evidence-based treatment recommendations for uremic
Percutaneous renal biopsy: comparison of manual and bleeding. Nat. Clin. Pract. Nephrol. 3: 138–153.
automated puncture techniques with native and 19 Manno, C., Bonifati, C., Torres, D.D. et al. (2011).
transplanted kidneys J. Riehl, Sabine Maigatter, H. Desmopressin acetate in percutaneous ultrasound-guided
Kierdorf, H. Schmitt, N. Maurin, H. G. Sieberth kidney biopsy: a randomised control trial. Am. J. Kidney
Nephrology Dialysis Transplantation, Volume 9, Issue 11, Dis. 57 (6): 850–855.
1994, Pages 1568–1574, https://doi.org/10.1093/ 20 Ferring pharaceuticals product inofrmation (2017).
ndt/9.11.1568. MINIRIN Nasal Drops. MINIRIN/OCTOCTIM Injections.
10 Lynn, K.L., Wong, K.M., Welsh, J. et al. (1997). Comparison 21 Canavese, C., Salomone, M., Pacitti, A. et al. (1989).
of automated percutaneous kidney biopsy using needles of Redcuced response of uraemic bleeding time to repeated
different gauge. Nephrology 3 (2): 211–214. doses of desmopressin. Lancet 334 (8664): 675–676.
Reference 213
2 Marwah, D.S. and Korbet, S.M. (1996). Timing of

2 30 Fereira, L.C., Karras, A., Martinex, F. et al. (2004).
complications in percutaneous renal biopsy: what is the Complications of protocol renal biopsy. Transplantation
optimal period of observation? Am. J. Kidney Dis. 28: 47–52. 77: 1475–1476.
23 Prasad, N., Kumar, S., Manjunath, R. et al. (2015). 31 Eiro, M., Katoh, T., and Watanabe, T. (2005). Risk factors
Real-time ultrasound-guided percutaneous renal biopsy for bleeding compli-cations in percutaneous renal biopsy.
with needle guide by nephrologists decreases post-biopsy Clin. Exp. Nephrol. 9: 40–45.
complications. Clin. Kidney J. 8: 151–156. 32 Soares, S.M., Fervenza, F.C., Lager, D.J. et al. (2008).
24 Ishikawa, E., Nomura, S., Hamaguchi, T. et al. (2009). Bleeding complications after transcutaneous kidney
Ultrasonography as a predictor of overt bleeding after biopsy in patients with systemic amyloidosis: single-
renal biopsy. Clin. Exp. Nephrol. 13: 325–331. center experience in 101 patients. Am. J. Kidney Dis.
25 Poggio, E., McClelland, R., Hansen, S. et al. (2020). 52: 1079–1083.
Systematic review and meta-analysis of native kidney biopsy 33 Fish, R., Pinney, J., Jain, P. et al. (2010). The incidence of
complications. Clin. J. Am. Soc. Nephrol. 15: 1595–1602. major hemorrhagic complications after renal biopsies in
26 Aaltonen, S. et al. (2020). Outpatient Kidney Biopsy: patients with monoclonal gammopathies. Clin. J. Am.
A Single Centre Experience and Review of Literature. Soc. Nephrol. 5: 1977–1980.
Nephron. 144: 14–20. 34 Piccoli, G.B., Daidola, G., Attini, R. et al. (2013).
27 Corapi, K.M., Chen, J.L., Balk, E.M., and Gordon, C.E. Kidney biopsy in pregnancy: evidence for counselling?
(2012). Bleeding complications of native kidney biopsy: a A systematic narrative review. BJOG: An International
systematic review and meta-analysis. Am. J. Kidney Dis. Journal of Obstetrics and Gynaecology
60 (1): 62–73. 120: 412–427.
28 Ralls, P.W., Barakos, J.A., Kaptein, E.M. et al. (1987). 35 Franke, M., Kramarczyk, A., Taylan, C. et al. (2014).
Renal biopsy-related hemor-rhage: frequency and Ultrasound-guided percutaneous renal biopsy in 295
comparison of CT and sonography. J. Comput. Assisted children and adolescents: role of ultrasound and
Tomogr. 11: 1031–1034. analysis of complications. PLoS One 10.137/journal.
29 Furness, P.N., Philpott, C.M., Chorbadjian, M.T. et al. pone.0114737.
(2003). Protocol biopsy of the stable renal transplant: a 36 Magee C, Vella JP, Tormey WP, Walshe JJ. Multiple
multicentre study of methds and complication rates. myeloma and renal failure: one center’s experience.
Transplantation 76: 969–973. Renal Failure 1998; 20: 597–60.
214
15
Minimal Change Disease and Focal Segmental Glomerulosclerosis in Adults

Yuemiao Zhang and Jicheng Lv
Peking University First Hospital, Beijing, China
I ntroduction D
efinition
Minimal change disease (MCD) and focal segmental glomer- MCD

ulosclerosis (FSGS) are the two key forms of idiopathic
The definition of MCD has not changed. It presents as
nephrotic syndrome (INS). In 1999, evidence-based recom-
nephrotic syndrome of sudden onset. In adults proteinu-
mendations for treating adult INS were published in Kidney
ria is less selective than in children. Hypertension may be
International [1]. In 2012, the Kidney Disease: Improving
present. A functional impaired glomerular filtration rate
Global Outcomes (KDIGO) Clinical Practice Guideline for
is common at onset and normalizes with remission. Light
Glomerulonephritis published its first guideline on glomeru-
microscopy and immunofluorescence show normal glo-
lar diseases [2]. A review of the literature shows that our cur-
meruli. Electron microscopy discloses diffuse foot process
rent clinical and pathogenic knowledge of MCD and FSGS
flattening and no immune deposits.
has evolved. Increasing studies report that in addition to idi-
opathic cases, MCD and FSGS could be caused by multiple
factors, such as virus infection, allergies, Hodgkin disease, FSGS
and medications. As such, MCD and FSGS denote histologic
findings, rather than two single diseases. Patients with genetic The onset of nephrotic syndrome can be more insidious than
mutations affecting podocyte biology, including slit dia- in MCD, except in the so-called glomerular tip lesion, a vari-
phragm structure, actin cytoskeleton of podocytes, and foot ant whose response to treatment is comparable to that of
process structure, may develop their first episode of nephrosis MCD [3–5]. In other forms, an explosive onset also seems to
as an adult when it does not respond to immunosuppressive entail a response to treatment closer to MCD than in more
treatment. This indicates that some patients with mutations progressive forms, although considerable overlap exists [6].
and who fail immunosuppression may explain some of the Hypertension and renal insufficiency are not unusual from
variability in observed treatment effects in trials. Although the onset. Proteinuria is not selective. A detailed discussion of
corticosteroids and calcineurin inhibitors (CNIs) are still the the histopathology of FSGS is beyond the scope of this chapter
cornerstone treatments, new insights into pathogenesis and but is available elsewhere [7–9]; it is not covered further here
novel treatment options have been developed, targeting the because there are no implications for specific treatment [7].
pathogenic process with more efficacy and fewer adverse
effects. For example, rituximab (RTX), a monoclonal antibody
Remission
targeting B cells, is an efficient and safe alternative. A number
of new agents targeting pathogenic factors, such as CD80, Complete Remission
have now been tried in the treatment of MCD and FSGS There is general agreement regarding the definition
(Table 15.1, Table 15.2). of complete remission (CR): 24 hours proteinuria of
This chapter examines the evidence for management of 0.3 g, with serum albumin of 3.5 g/dl and stable renal
MCD and FSGS in adults. function.
Epidemiolog 215
Table 15.1 Treatment of adult minimal change disease (MCD).
Certainty of Strength of
Treatment Recommendation evidence recommendation
Prednisone/ We recommend that corticosteroids be given for initial treatment of Low Strong (use of
prednisolone nephrotic syndrome. We suggest prednisone or prednisolone be given at a corticosteroids)
daily single dose of 1 mg/kg (maximum 80 mg) or alternate-day single Weak (dose and
dose of 2 mg/kg (maximum 120 mg). The initial high dose of duration)
corticosteroids should be maintained for a minimum period of 4 weeks if
complete remission is achieved, and for a maximum period of 16 weeks if
complete remission is not achieved. In patients who remit, we suggest that
corticosteroids be tapered slowly over a total period of up to 6 months
after achieving remission. For infrequent relapses, we suggest using the
same initial dose and duration of corticosteroids.
Calcineurin For patients with relative contraindications or intolerance to high-dose Low Weak
inhibitors (CNIs) corticosteroids, we suggest short-term intravenous methylprednisolone
(0.8 mg/kg/day for 10 days) + tacrolimus monotherapy (0.05 mg/kg/day for
16–20 weeks) subsequently tapering over approximately 18 weeks. We
suggest oral CNIs for relapsed patients despite cyclophosphamide, or
patients of childbearing age. Cyclosporine initiates at 3–5 mg/kg/day or
tacrolimus 0.05–0.1 mg/kg/day in two equally divided doses. Following
3 months of stable remission, tapered to reach the minimum dose that
maintains remission, for 1–2 years.
Cyclophosphamide We suggest oral cyclophosphamide 2–2.5 mg/kg/day at a single dose as Low Weak

tolerated for 8 weeks for patients with frequently relapsing MCD and with
relative contraindications or intolerance to high-dose corticosteroids.
Mycophenolate We suggest oral mycophenolate mofetil for patients intolerant to Low Weak
mofetil corticosteroids, cyclophosphamide, and/or CNIs. 500–1000 mg twice daily
for 1–2 years.
Rituximab We suggest using rituximab for patients with frequently relapsing or Low Weak
corticosteroid-resistant MCD.
GRADE assessment of the certainty of the evidence: High: This research provides a very good indication of the likely effect. The likelihood
that the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood
that the effect will be substantially different* is moderate. Low: This research provides some indication of the likely effect. However, the
likelihood that it will be substantially different* is high. Very low: This research does not provide a reliable indication of the likely effect. The
likelihood that the effect will be substantially different* is very high. *Substantially different = a large enough difference that it might affect
decision-making.
Partial Remission Steroid Dependent

Partial remission (PR) has been variably defined among
Steroid dependent (SD) usually referrers to two consecu-
publications. An acceptable definition is 24 hours protein-
tive relapses during corticosteroid therapy, or within
uria of >0.3 and <3 g and/or a decrease of at least 50% from
14 days of ceasing therapy.
the baseline level, along with a rise of serum albumin of
3 g/dl and stable renal function.
E
pidemiology
Relapse
Clinically, FSGS and MCD share numerous similarities,
The definition of relapse also varies among publications. In but show different onset age, treatment responses, and
general it is defined by recurrence of massive proteinuria prognosis (Table 15.3). MCD is the most common cause of
(24 hours proteinuria of >3.5 g). Frequent relapse (FR) is INS, accounting for 70–90% in children under 10 years old
generally defined as more than two relapses observed over and 10–25% in the adult population with an increase in
the last 6 months or more than four relapses observed over percentage of FSGS [10]. In addition to age, ethnicity is an
any 12-month interval. important predictor of the percentage of nephrotic patients
216 Minimal Change Disease and Focal Segmental Glomerulosclerosis in Adults
Table 15.2 Treatment of adult focal segmental glomerulosclerosis (FSGS).
Prednisone/ We recommend that corticosteroids be given for initial treatment of Low Strong (use of
prednisolone nephrotic syndrome. We suggest prednisone or prednisolone be given at a corticosteroids)
daily single dose of 1 mg/kg/day (maximum 80 mg) or alternate-day Weak (dose and
single dose of 2 mg/kg (maximum 120 mg). The initial high dose of treatment duration)
corticosteroids should be maintained for a minimum period of 4 weeks if
complete remission is achieved and for a maximum period of 16 weeks if
complete remission is not achieved. In patients who remit, we suggest
that corticosteroids be tapered slowly over a total period of up to
6 months after achieving remission.
Calcineurin We suggest oral cyclosporine for steroid-resistant patients. Cyclosporine Moderate Weak
inhibitors initiates at 3–5 mg/kg/day in two equally divided doses. Following
4–6 months of stable remission, tapered to reach the minimum doge that
maintains remission, for 1–2 years.
Mycophenolate We suggest oral mycophenolate mofetil + high-dose dexamethasone for Low Weak
mofetil patients with steroid-resistant FSGS intolerant to cyclosporine.
GRADE assessment of the certainty of the evidence: High: This research provides a very good indication of the likely effect. The likelihood that
the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood that the
effect will be substantially different* is moderate. Low: This research provides some indication of the likely effect. However, the likelihood that
it will be substantially different* is high. Very low: This research does not provide a reliable indication of the likely effect. The likelihood that the
effect will be substantially different* is very high. *Substantially different = a large enough difference that it might affect decision-making.
Table 15.3 Clinical features of minimal change disease (MCD) and focal segmental glomerulosclerosis (FSGS).
MCD FSGS
Age prevalence ● 70–90% in children <10 years old ● No

● 10–25% in adults
Male/female ratio ● 2 : 1 in children ● 1 : 1
● 1 : 1 in adults
Secondary forms ● Rare ● Frequent
Familial aggregation ● Rare ● Frequent
Light microscope ● Almost normal ● Focal and segmental sclerosis
Immunofluorescence ● IgM± ● IgM±
● C3± ● C3±
● IgA±
● IgG±
Electron microscope ● Diffuse foot process effacement ● Diffuse foot process effacement
Treatment resistance ● Rare ● Frequent
Frequent relapse ● Frequent ● Rare
Progression to end stage renal disease ● Rare ● Frequent
C3, complement 3; IgA, immunoglobulin A; IgG, immunoglobulin G; IgM, immunoglobulin M. Renal deposition of C3, IgA, IgG and I gM were
semi-quantified using immunofluorescence staining, grading as 0 (lack of deposits), 1 + (trace), 2 + (weak), 3 + ( moderate), and 4 + (strong).
with MCD. The percentage of nephrotic patients with common in Asia (6.9%) [11, 12]. In recent years, the inci-
MCD is highest in Asian and Caucasian populations, while dence and the prevalence of nephrotic FSGS have steadily
the percentage of nephrotic patients with underlying FSGS risen [13, 14].
is highest in African-American populations. FSGS is the Genetic predisposition is proposed in patients with FSGS
most common form of glomerular histology in North because of their familial aggregation and the geographic
American (19.1%), the second most common in Europe variation in prevalence. Patients of black African ancestry
(14.9%) and Latin America (15.8%), and the fifth-most are twice as likely to develop FSGS than patients of white
Pathophysiolog 217
European origin [18]. This is true of most histological vari- likely to be systemic disorders, where circulating factors
ants, especially the cellular variants [19, 20]. The exception and/or immune dysfunction are the pathogenic culprit,
is the glomerular tip lesion [3, 4]. The response to corticos- subsequently causing podocyte injury (Figure 15.1).
teroid treatment in black people is distinctly worse than in
white people [21], although the overall outcome of the dis-
ease in each race follows the same trend. In recent years, Putative Circulating Permeability Factors
familial cases with MCD have also been described [22], sug-
gesting a possible genetic origin of the disease. MCD and suPAR
FSGS are characterized by diffuse foot process effacement. Soluble urokinase-type plasminogen activator receptor
As the intricacies of podocyte cell biology have been unrave- (suPAR) has been suggested to be one of the pathogenic
led, increasing causative genes regulating slit diaphragm circulating factors [33–35]. The urokinase plasminogen
structure (nephrin, NPHS2, CD2AP) and actin cytoskeleton activator surface receptor (uPAR) is a membrane-bound
of podocytes (MYO1E, myocin IIA, α-actinin-4) have been glycosyl-phosphatidylinisotol anchored protein present on
identified [23–26]. The role of high-risk APOL1 genotypes in multiple cells, including podocytes. In podocytes, uPAR
the development of glomerulosclerosis is still under investi- could activate αvβ3 integrin signaling, inducing foot pro-
gation. Genetic testing for pediatric nephrotic syndrome and cess effacement and proteinuria [33]. However, subsequent
adult FSGS is controversial, and it can be considered for studies have suggested that suPAR may be a novel prognos-
patients with congenital and infantile forms of nephrotic tic biomarker for chronic kidney disease, but does not
syndrome (children <1 year of age) or less than the two appear to represent the permeability factor in FSGS [36].
forms but with steroids resistant nephrotic syndrome
(SRNS), nephrotic syndrome associated with other syndro- CLCF1
mic features, or familial forms of SRNS/FSGS [27]. Genetic Cardiotrophin-like cytokine factor 1 (CLCF1) is another
FSGS is typically resistant to corticosteroids. The value of candidate circulating factor which was detected in plasma
routine implementation of genetic testing in adults remains from patients with recurrent FSGS. CLCF1 is a member of
to be resolved in the future [28–30]. the IL6 family of cytokines. CLCF1 circulates as a heterodi-
meric cytokine, combining with cytokine receptor-like fac-
tor 1 (CRLF1) or soluble ciliary neurotrophic factor receptor
P
athophysiology alpha (sCNTFRa), which is an effector of the Janus kinase
(JAK)/signal transducer and activator of transcription
MCD and FSGS are both examples of pathogenic mecha- (STAT) signaling [37]. In podocytes, CLCF regulates the glo-
nisms that primarily affect the podocyte (podocytopathies). merular filtration barrier function through the JAK/STAT
They are not characterized by immune deposits, but there pathway [38]. In isolated rat glomeruli, CLCF1 increased
may be a role for circulating factors in the pathogenesis of albumin permeability (Palb) and an anti-CLCF1 monoclo-
both diseases. In recurrent FSGS following kidney trans- nal antibody was able to block the CLCF1-induced increase
plantation, early renal biopsy shows widespread foot pro- in Palb. Western blot analysis showed that CLCF1 upregu-
cess effacement on electron microscopy, hallmarks of lated phosphorylation of STAT3 in glomeruli. Inhibitors of
MCD, while repeated biopsy suggests evolution to an FSGS JAK2 and STAT3 significantly blocked the effect of CLCF1
lesion [15] or recovery to be normal when translated into a or FSGS serum on Palb. As such, albuminuria in FSGS might
diabetic nephropathy patient [16]. It is an ongoing debate be related to qualitative or quantitative changes in the
whether MCD and FSGS represent different stages of one CLCF1-CRLF1 complex, and JAK2 or STAT3 inhibitors may
disease or are different disease entities [17]. be novel therapeutic agents to treat FSGS.
The role of circulating factors has long been suspected in
pathogenesis of FSGS due to several clinical observations, Anti-CD40 Antibody
including the recurrence of FSGS in post-transplant patients The beneficial effects of RTX in MCD and FSGS suggest the
(about 30% in adults), the induction of proteinuria in exper- role of autoantibodies in disease pathogenesis. By screening
imental animals by plasma fractions from FSGS patients, autoantibodies in pretransplant sera from 10 patients with
and the improvement of disease course after treatment with recurrent FSGS and 10 without recurrent FSGS, auto-anti-
plasma exchange (PE), but with little success for searching bodies against 7 target proteins, including CD40 molecule,
the responsible factors [31]. In addition, the therapeutic protein tyrosine phosphatase receptor O, tumor necrosis
benefits of immunosuppressive agents and RTX which factor (TNF) receptor superfamily member 6, Chorionic
inhibit cell-mediated immunity and the coincidence of INS gonadotropin β, Apoliprotein 2, P2Y purinoceptor 11 and
in Hodgkin’s disease indicate the role of immune disorder Small nuclear retinoid × receptor α, were identified and
in MCD and FSGS [32]. Taken together, MCD and FSGS are validated. Among them, anti-CD40 antibody alone had the
Rituximab Abatacept
Steroids
CD80
IL13
CTLA4
CNIs
CLCF SMPDL3b
Apoptosis
Th2 JAK-STAT
Podocyte-specific-proteins
β3 integrin
suPAR
T IL17
Th17
B Steroids Treg
MMF
CTX
Rituximab CNIs
suPAR MCD
ApoA1b FSGS
anti-CD40 antibody MCD+FSGS
CLCF
Figure 15.1 Pathogenesis of MCD and FSGS. The podocyte injury caused by circulating factors (suPAR, CLCF, ApoA1b, anti-CD40
antibody) and cytokines (IL13, IL17) secreted by abnormal interactions between T cells and B cells is crucial in disease pathogenesis.
Except for their systemic effects, immunosuppressive agents (steroids, MMF, CTX, CNIs) and targeting treatment options (rituximab,
abatacept) also show a direct effect on podocyte. Podocyte-related markers and products (CD80, suPAR) can be detected in the urine of
patients with MCD and FSGS, providing a noninvasive technique for gathering information about the disease diagnosis and progression.
Items mainly involved in MCD are shown in red, in FSGS are shown in green, and in both MCD and FSGS are shown in yellow.
best accuracy of 0.77 in predicting FSGS recurrence. During mal models have implicated T helper type 2 (Th2)-derived
the phase of remission in recurrent FSGS patients after cytokines, particularly interleukin (IL)-13. In rats, sys-
treatment of RTX plus cyclosporin (CsA), anti-CD40 anti- temic overexpression of IL-13 results in albuminuria,
body decreased accordingly. CD40 was stained on podo- hypoalbuminemia, and, on kidney biopsy, up to 80%
cytes in patients with FSGS but not in normal human podocyte foot process fusion by electron microscopy with
kidneys. Anti-CD40 antibody could cause podocyte injury similar pathologic findings as human MCD. In addition,
directly by inducing cytoskeletal reorganization both in cul- T cells from patients with nephrotic syndrome spontane-
tured human podocytes and in wild-type mice. The inhibi- ously produce IL-13, while B cells express the IL-13 recep-
tion effects of monoclonal CD40-blocking antibodies, tor. T cells from patients with relapsed MCD have
monoclonal suPAR clocking antibodies, or inhibitor of αvβ3 increased expression of IL-13 compared with those from
suggest an interaction between anti-CD40 antibodies and patients in remission.
suPAR/β3 integrin pathway [39]. Independent validation in
other recurrent FSGS cohorts is needed. microRNA
microRNAs (miRNAs) are important regulators of gene
IL-13 expression that suppress their target genes. miR-193a was
The identity of the glomerular permeability factor in significantly upregulated in isolated glomeruli from FSGS
MCD has not been determined in humans. Data from ani- patients. Mechanistically it inhibits the expression of WT1,
Prognosi 219
a transcription factor and master regulator of podocyte dif- mechanism for proteinuria. It actives T cells by binding to
ferentiation and homeostasis, which in turn leads to down- receptor CD28, but inhibits T cell activation by binding to
regulation of genes crucial for the architecture of podocytes, CTLA-4, cytotoxic T lymphocyte-associated antigen-4
such as podocalyxin (PODXL) and nephrin (NPHS1) [55]. expressed on regulatory T cells. CTLA-4 also downregu-
The cause of the increased expression of miR-193a in podo- lates CD80 and CD86 expression on APC, Antigen-
cytes of FSGS patients and the miRNA profiles in FSGS presenting cells. In α3 integrin −/−, and nephrin −/− mice,
and MCD still needs to be addressed. CD80 is upregulated. In wild-type mouse models, injection
of lipopolysaccharide (LPS) leads to CD80 upregulation in
podocytes and proteinuria due to direct action of Toll-like
Immune Disorders
receptor (TLR) ligands on podocytes. Adding tumor necro-
T Cells sis factor α (TNFα) to podocytes causes CD80 upregulation,
Abnormal T cell dysfunction has long been implicated in which further interacts with Neph1 causing slit diaphragm
the pathogenesis of MCD and FSGS. However, no consen- disruption [51, 52]. CD80-deficient mice are protected
sus has yet been reached regarding a single cell subset. from LPS-mediated podocyte injury. IL13 or microbial
Some studies have shown elevated levels of IL2, some have products via TLRs could be factors that induce CD80
suggested a relative predominance of a Th2 cytokine expression on podocytes.
response (IL4), while others reported activation of Th1 A two-hit podocyte immune disorder was postulated in
cytokine response with increased interferon-γ. Urinary MCD [53]. Initially, CD80 is induced by a microbial agent
excretion of Th1 and Th2 cytokines cannot discriminate and/or T-cell cytokine. Impaired production of soluble
FSGS from MCD. However, urinary Th1 cytokines IL12 was CTLA4 by decreased circulating Tregs then induces persis-
associated with both global sclerosis and interstitial fibrosis, tent expression of CD80, which causes persistent proteinu-
and IL2 and GM-CSF, Granulocyte-Macrophage Colony ria. Administration of abatacept, an inhibitor of CD80,
Stimulating Factor were significantly increased in FSGS appears to cure patients with severe nephrotic syndrome
patients resistant to treatment. Th2 cytokines, especially due to primary FSGS or recurrent FSGS after
IL4, IL5, IL-8, IL10, and IL13, are implicated in frequent transplantation [54].
relapses of MCD [40, 41]. With the detection of new subsets
of T cells, an increased Th17/Treg ratio was determined in
adult patients with MCD and FSGS compared to normal Others
subjects, which was associated with proteinuria [42]. Th17 Some other molecules participating in podocytopathy
cells are strong candidate drivers for steroid response in have also been suggested, such as nonmuscle myosin
immune diseases [43], and the Th17/Treg imbalance heavy chain-IIA (NMMHC-IIA) and RAP1GAP. NMMHC-
returned to normal after effective corticosteroids ther- IIA is located primarily at cell body and primary processes
apy [42]. Dominance of infiltrating Th17 cells over Treg of podocytes. Expression levels of NMMHC-IIA were sig-
cells in the renal interstitial was also observed [44]. nificantly decreased in puromycin aminonucleoside-
treated rats and patients with idiopathic FSGS [56].
B Cells Increased podocyte expression of RAP1GAP, RAP1
The benefits of B cell-depleting therapy with RTX in MCD GTPase Activating Protein was associated with loss of
add a potential role of B cell in the pathogenesis of MCD. activated β1 integrin in vivo and in vitro by a mechanism
RTX performs its effects through suppressing the that involves loss of RAP1-mediated activation of β1 inte-
interactions between B cells and T cells, and restoring Treg grin. Furthermore, preventing elevation of RAP1GAP lev-
cell populations to keep the Th17/Treg balance normal [45– els in injured podocytes maintained β1 integrin–mediated
49]. It also presents direct protective effects on podocyte adhesion and prevented cellular detachment [57].
cytoskeleton stabilization in a sphingomyelin phosphodi-
esterase acid-like 3b (SMPDL-3b)-dependent manner [50].
It partially prevented SMPDL-3b and ASMase downregula- P
rognosis
tion that was observed in podocytes treated with the sera of
patients with recurrent FSGS. Response to steroid therapy was the most important factor
for preservation of renal function. In general, the majority
of patients with MCD respond to steroids (80% sensitive
CD80
rate with remission in 74%) and have preserved renal func-
CD80/B7-1, a transmembrane protein usually expressed on tion for the long term, while FSGS is related to less steroid
B cells, nature killer cells, and antigen-presenting cells, responsiveness (58% sensitive rate and 30% remission
also expressed on podocytes, which was proposed as a rate) [91]. However, more than two-thirds of adult patients
with MCD experienced relapses, and up to a third of patients T

reatment
may become frequent relapsers or steroid-dependent.
Younger onset age, steroid without cyclophosphamide treat- MCD
ment, and shorter treatment duration were independent risk
factors for relapse [92]. Age is also a factor for steroid-related Corticosteroids
adverse events in MCD patients [93]. For MCD patients with Corticosteroids remain the mainstay of first-line treat-
FR or FSGS patients with less favorable outcome, there is ment of MCD in adults. There are still no large rand-
increased early mortality owing to a higher risk of implica- omized controlled trials (RCTs) that provide evidence on
tions, including infections, thromboembolic events, and the optimal agent, dose, and duration of corticosteroids
acute kidney injury (AKI). that should be given. The evidence of using corticosteroids
is based largely on extrapolation from RCTs in chil-
dren [101, 102] and observational studies in adults [103–
Infection 110]. The drug of choice is oral prednisone or prednisolone
Infectious complications are commonly seen in nephrotic Box 15.1. As compared with MCD in children, adults tend
patients due to mass proteinuria and/or use of immuno- to respond more slowly and the responsiveness is also less
suppressive agents. Pneumonia is mostly common seen, predictable. Prednisone was suggested to be used as a rec-
and primary peritonitis, pneumonia, and cellulitis have ommended dose of 1 mg/kg/day, not exceeding 80 mg,
also been reported [94, 95]. Accordingly, antibiotics until CR (minimum 4 weeks to a maximum of 16 weeks).
should be applied as per regional practice. In addition, In the elderly some clinicians use a dose of 2 mg/kg given
hepatitis B virus reactivation, tuberculosis, and parasite on an alternate-day basis, although no specific study has
infection should be evaluated, if necessary, during addressed the issue of similar efficacy and better tolerabil-
immunosuppression. ity. The issue of corticosteroid toxicity is well-recognized,
but although weight gain, diabetes, and osteonecrosis are
constantly cited regarding prolonged treatment with corti-
Thromboembolic Events costeroids in adult MCD, very few published studies have
Thromboembolic event is one of the most severe complica- provided specific long-term data on the rates of these
tions in NS. Hypercoagulable state in patients with NS complications.
could easily leads to venous thromboembolism (VTE) or Considering the adverse effects of long-term usage of
arterial thromboembolism (ATE), the most common of high-dose corticosteroids, studies have examined enhanced
which are deep vein thrombosis, renal vein thrombosis, and corticosteroid therapy (short-term intravenous methyl-
pulmonary embolism [96]. In a large retrospective cohort prednisolone with a low-dose of prednisone and/or another
study, Mahmoodi et al. evaluated the VTE or ATE events in immunosuppressant) [109]. Two very early small RCTs
298 consecutive patients with NS [97]. In the 10-year fol- compared intravenous methylprednisolone with or with-
low-up period, annual incidences of VTE and ATE were out subsequent oral prednisone to oral prednisone
1.02% (95% confidence interval [CI] 0.68–1.46) and 1.48% alone [111]. In one study, Yeung found no difference
(95% CI 1.07–1.99), respectively. However, over the first between intravenous methylprednisolone plus oral pred-
6 months of follow-up, these rates were as high as 9.85% and nisone compared with oral prednisone alone for CR (rela-
5.52%, respectively. The ratio of proteinuria to serum albu- tive risk [RR] 0.74, CI 0.50–1.08) in 18 adult patients with
min was associated with VTE, while eGFR and multiple MCD. In the other study, Imbasciati observed that oral
classic risk factors for atherosclerosis, such as hypertension, prednisone, compared with short-course intravenous
diabetes, and smoking, were predictors of ATE.
Box 15.1 Corticosteroid Agents (Potency Relative to
AKI Prednisone)
AKI is a common complication of adult patients with MCD Oral agents

(up to 20–25%) [98] and patients with FSGS 39% [91]. It is an ●● Prednisone [1]
●● Prednisolone [1]
independent factor that delays the time to complete remis-
sion [99]. MCD-NS patients complicated with AKI had Intravenous agents
a more severe hypercoagulable state [100], which in turn ●● Methylprednisolone [1.25]
increases the risk of thromboembolic events. For patients Subcutaneous agents
complicated with AKI, contemporary renal replacement ●● ACTH (Adrenocorticotropic hormone)
therapy could be used, combining with corticosteroids.

Treatmen 221
methylprednisolone, increased the number of subjects Alkylating Agents

who achieved CR (RR 4.95, CI 1.15–21.26) in 22 patients Only a few patients have been treated with alkylating agents
with MCD. Given the limited number of patients, these (cyclophosphamide or chlorambucil) in the initial treat-
two studies might be underpowered and inferiority of ment of MCD [106, 117, 118] Box 15.2. Alkylating agents
intravenous methylprednisolone could not be excluded. are indicated in multirelapsing cases more than for SR
Subsequent observational studies suggested that intrave- forms (opinion) [103], although true SR is rare in true MCD,
nous methylprednisolone and prednisolone use was that is, cases in which a repeat renal biopsy carried out after
significantly associated with early remission and lower 4 months of unsuccessful corticosteroid therapy still shows
incidence of relapse compared with prednisolone use MCD. In most cases the second histopathologic examina-
alone [108, 109], which was suggested in an Italian RCT in tion discloses lesions of FSGS that had been overlooked or
pediatric patients with MCD [112]. not sampled on the first biopsy, or had not yet appeared at
In a recent multicenter RCT, tacrolimus (TAC) mono- the onset of NS, which explains the initial failure of corti-
therapy or oral prednisone (1 mg/kg/day, maximum 80 mg/ costeroids [103, 105, 106, 119, 120].
day) after short-term intravenous methylprednisolone Despite the lack of studies specifically related to adult
(0.8 mg/kg/day) for 10 days showed remission rates of multirelapsers, it is likely that cytotoxic agents result in
98.3% (55/56) and 96.2% (51/53), respectively. Although prolonged remissions, as shown in RCTs in children.
there was no statistically significant difference in remission Cyclophosphamide, at 2–3 mg/kg/day, is more commonly
rates, mean time to remission, and relapse rates between used than chlorambucil, without clear evidence of the
the two groups, adverse events occurred more frequently in superiority of one over the other. It should be administered
the prednisone group than in the TAC group (128 vs. orally and not in the form of intravenous pulses, and
81) [113]. In a retrospective study, treatment with intrave- presumably, considering the experience acquired with
nous methylprednisolone (0.5 or 1.0 g/day for 3 days) fol- childhood onset cases, for 8 weeks. Alkylating agents entail
lowed by oral cyclosporine (2–3 mg/kg/day) and low dose known hazards of hematologic, infectious, and oncologic
of prednisolone (30 mg/day) showed 100% complete remis- side effects, but quantifying these risks is not possible given
sion rate compared to 85.7% in patients treated with intra- the available data. In men and women the risk of sterility is
venous methylprednisolone followed by oral prednisolone significant if the course is prolonged, and the risk increases
(0.4–0.8 mg/kg/day) and 69.2% in patients treated with oral with increasing age at the time of onset of the therapy.
prednisolone (0.6–1.0 mg/kg/day) alone during the
9-month follow-up. The adverse effects caused by predni-
CNIs
solone were fewer [114].
Although more than 90% of adult patients with MCD CsA
achieve CR, about 50% of them experience relapses dur- CsA has been used in adults with MCD since 1986 [1] Box
ing the disease course [94, 103, 104, 107, 110]. For 15.3. Its side effects include hypertension, tremor, gingival
patients with infrequent relapse, optimal corticosteroid hyperplasia, and hypertrichosis. Conversely, CsA has the
dose for relapse is still unclear. The KDIGO guidelines advantage of not being cytotoxic, does not induce sterility,
recommend the same initial dose and duration. One and entails fewer metabolic disturbances. These reasons
observational study in children and one in adults sug- may justify its use as first-line treatment in patients at
gested lower dose corticosteroid therapy might be helpful risk of diabetes, osteonecrosis, or infertility. Aside from clini-
to improve the quality of life of patients [110, 115]. For cal scenarios in which corticosteroids are contraindicated
patients with FR or SD, corticosteroid-sparing agents
such as cyclosporin (CsA), cyclophosphamide, and
Box 15.2 Alkylating Agents
mycophenolate mofetil (MMF) are usually needed (dis-
cussed in the following sections). A small observational
study tested the effects of adrenocorticotropic hormone Cyclophosphamide
(ACTH) gel (80 units subcutaneously twice weekly) in 15 Chlorambucil
subjects with resistant glomerular diseases (MCD = 2,
FSGS = 3) for 6 months. One subject with resistant FSGS
achieved CR and one subject with resistant MCD Box 15.3 Calcineurin Inhibitors
achieved PR but relapsed within 4 weeks of stopping
ACTH [116]. The evidence should be studied further in Cyclosporine (CsA)
controlled trials against currently available therapies for Tacrolimus (TAC)
resistant disease.
[121], CsA first-line monotherapy has been used in many respectively [127]. In one small observational study, TAC
case series (level IV studies) with very low dose [122]. One (0.05–0.1 mg/kg/day) combined with prednisone (0.5 mg/
such study [121] included 86 adult patients with MCD and kg/day) showed a remission rate of 90.9% (10/11) with a
NS that were resistant to conventional therapy or were SD CR rate of 72.7% (8/11) in CsA-resistant and -dependent
or multirelapsers. They were treated with the Sandimmune adult MCD patients within 2 months [128].
formulation at a dose of 5.18 ± 0.94 mg/kg/day. Superior
efficacy was reported in SD patients (CR in 73% of cases MMF
and PR in 14%) compared to those who were SR (30% and Studies of MMF in the treatment of MCD in adults are few
26%, respectively). Serum creatinine levels remained sta- and are limited to case series and reports [129–131].
ble during CsA treatment. These patients were not all CsA Considering the encouraging results obtained in childhood
dependent, as ∼20–25% were progressively tapered off nephrosis, MMF may have a place in the treatment of adult
after 2 years without relapsing [123]. Combination therapy MCD, but so far evidence either supporting or refuting its
with CsA and corticosteroid agents might further enhance use is absent.
the efficacy and with less corticosteroid toxicity as men-
tioned above [114], suggesting an additional benefit of RTX
lower exposure to corticosteroids. A RCT performed in 52 RTX, a chimeric monoclonal antibody directed against
adult patients with the first relapse of MCD noted that CD20, is a potential effective and safe treatment in adults
remission was achieved sooner in patients treated with with MCD Box 15.4. The use of RTX in adults with FR and
CsA plus 0.8 mg/kg/day prednisone compared to patients immunosuppression-dependent MCD has been reported
receiving only 1 mg/kg/day prednisone [124]. in case reports [132–134] and small observational
One of the main safety concerns of CsA therapy in INS is studies [135–138] in reducing relapses and sparing immu-
the potential for nephrotoxicity. This is a class effect com- nosuppression. A prospective, multicenter, off–on trial
mon to all CNIs, including TAC [125, 126] Box 15.3. evaluated the efficacy of RTX and rate of recurrences in a
Previous studies indicated that the dose and type of pri- cohort of 10 children and 20 adults with FR MCD and
mary renal disease are two elements to keep in mind when FSGS after administration of RTX and discontinuation of
interpreting a decline in renal function [123]. immunosuppressants. One year after, all 30 patients were
A problem that has appeared since 1999 is that of formu- in remission and half of them never presented any recur-
lation. Most publications mention “cyclosporine” and do rence during the follow-up. The total number of relapses,
not specify whether they are discussing Sandimmune or the per-patient annualized relapse rate, the per-patient cor-
Neoral or a generic form of the drug, which may be impor- ticosteroid maintenance median dose, and the median
tant because it is known that the better pharmacokinetic cumulative dose to achieve relapse remission were reduced
profile of Neoral may lead to better efficacy and hence a significantly after treatment with RTX. Renal function
distinct reduction of dose [125] below the maximum of improved in all subjects, especially in children and in FSGS
5 mg/kg/day recommended for Sandimmune [123]. subgroups [139]. Similar results were reported extending
the observation period to more than 2 years, demonstrating
TAC a long-lasting effect of RTX [136, 140–142]. One very small
Given the FR after withdrawal and the risk of renal toxicity observational study has reported the efficacy of RTX as a
after long-term therapy with CsA, there are several studies first-line therapy in six adult-onset MCD cases [143].
focus on detecting the efficacy of TAC to induce CR in In future RCTs, detecting the efficacy of RTX as both first-
adults with MCD. TAC, like CsA, is a CNI that inhibits line and second-line treatment with extended follow-up
T-cell-driven elaboration of cytokines. It shows more will add important information regarding its optimal treat-
potent cytokine suppression and seems to cause less toxic- ment, relapse, and safety.
ity than CsA. In a recent multicenter RCT, TAC monother-
apy following short-term intravenous methylprednisolone
(0.8 mg/kg/day) for 10 days showed comparable remission
rates, remission rates, mean time to remission, and relapse Box 15.4 Monoclonal Antibodies
rates with the conventional oral prednisone regimen, but
fewer adverse events [113]. In one pilot study, the cumula- Rituximab
tive CR rates of patients treated with oral TAC (0.05 mg/kg Fresolimumab
twice daily) and low-dose prednisolone (0.5 mg/kg/day) Adalimumab
were 7.7% (1/14), 64.2% (9/14), 71.3% (10/14), 92.9% Abatacept
(13/14), and 100% (14/14) at 1, 2, 4, 8, and 20 weeks
Treatmen 223
Summary FSGS
MCD is not uncommon in adults and may assume a multire-
All forms of FSGS require treatment, irrespective of the
lapsing course. The majority of patients respond to corticos-
histopathologic variant [4, 5, 8, 144]. The goal of therapy
teroids with CR, but this may require 4 months of treatment,
is to reduce, if not eliminate, proteinuria. Inducing even
compared with 6–8 weeks in children, with the attendant
a PR reduces the rate of deterioration in renal func-
hazards of corticosteroid toxicity. Enhanced corticosteroid
tion [6]. Corticosteroids remain the first-line treatment.
therapy (short-term of intravenous methylprednisolone
Response to corticosteroids is also the best indicator of a
with a low dose of prednisone and/or with another immu-
better long-term prognosis. In case of failure, a panoply
nosuppressant) might be alternative option to reduce corti-
of immunosuppressive drugs have been tried, including
costeroid toxicity. In selected cases, low-dose CNIs can be
new immunosuppressive agents used in solid organ
considered an alternative to corticosteroids, with very little
transplantation. A few recent publications on RTX [145]
nephrotoxicity. SR may be due to an initially missed diagno-
indicate that the anti-CD20 monoclonal antibody may
sis of FSGS, and resistance should be considered an indica-
inconsistently obtain remission in cases of FSGS, espe-
tion to verify the diagnosis by a repeat kidney biopsy. The
cially when the NS relapses after renal transplantation.
place of MMF in the treatment of adult MCD remains to be
determined. For patients with FR or conventional treatment-
resistant, RTX is a promising option, although the current Corticosteroids
evidence is limited to observational studies. The strength of Current consensus remains that corticosteroids with full dose
evidence supporting these statements is low or very low, and extended period of time are the first-line of therapy in
being largely based on case series and observational studies nephrotic FSGS. In the absence of any RCT data, the evidence
(RCTs tending to revise treatment recommendations are mainly based on several observational studies with remission
summarized in Table 15.4). rates ranging from 30% to 60% [20, 146, 147]. The initial
Table 15.4 Summary of findings: treatment of adult minimal change disease (MCD).
No. of Absolute effect per Relative risk Certainty of the

participants 1000 patients treated (95% confidence evidence
Outcomes (no. of studies) for 1 year (95% CI) interval) (GRADE) Conclusion
Intravenous methylprednisolone/TAC vs. intravenous methylprednisolone/prednisone [113]

Remission rate 109 (1) Not estimable 1.02 (0.96–1.09) Low TAC is noninferior to prednisone after
(CR or PR) ●○○○ short-term intravenous
methylprednisolone
Relapse rate 106 (1) Not estimable 0.93 (0.62–1.39) Low TAC and prednisone show similar
●○○○ relapse rates
Adverse event 109 (1) Not estimable 0.83 (0.71–0.98) Low Adverse events occur more frequently in
rate ●○○○ the prednisone group
CsA/prednisolone (0.8 mg/kg/day) vs. prednisolone (1 mg/kg/day) [124]
Remission rate 52 (1) Not estimable 1.25 (1.01–1.55) Low Low-dose CsA allows the prednisolone
(CR or PR) ●○○○ dose to be reduced as compared to
therapy with prednisolone alone
Relapse rate 52 (1) Not estimable 1.20 (0.41–3.47) Low Combining with or without CsA show
●○○○ similar relapse rates
Adverse event — — — — Adverse event was not analyzed
rate
CI, confidence interval; CR, complete remission; CsA, cyclosporin; PR, partial remission; TAC, tacrolimus. GRADE assessment of the certainty
of the evidence: High: This research provides a very good indication of the likely effect. The likelihood that the effect will be substantially
different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood that the effect will be substantially
different* is moderate. Low: This research provides some indication of the likely effect. However, the likelihood that it will be substantially
different* is high. Very low: This research does not provide a reliable indication of the likely effect. The likelihood that the effect will be
substantially different* is very high. *Substantially different = a large enough difference that it might affect decision-making. Pooled analyses were
not performed due to the heterogeneity of the data.
response to treatment in adults with FSGS was analyzed by c orticosteroids [121]. Combination of CsA and corticos-
Korbet et al. [144, 148, 149]. The trend before 1980 was to treat teroids remains the mainstay of treatment, with good tol-
for a short time and with low doses of corticosteroids. The erability when CsA dose is low. A retrospective
highest CR rates, >30%, were observed in cases treated for observational study compared high-dose oral prednisone
>5 months, and the lowest, 20%, were seen in patients (1 mg/kg/day) with low-dose prednisone (0.5 mg/kg/day)
treated for 2 months. Thus, corticosteroid treatment must be in combination with CsA (3 mg/kg/day) or azathioprine
sufficiently long. Full-dose (1 mg/kg/day) prednisone is given (2 mg/kg/day) as initial treatments. Low-dose prednisone
for 8–12 weeks, followed in case of remission, even partial, by was given to 16 patients with obesity, bone disease, or
slow tapering over months to avoid a “rebound” relapse [150]. mild diabetes. Remission rates were comparable; 63% for
CR portends a favorable outcome, comparable to that of prednisone (n = 9), 80% for prednisone plus
corticosteroid-responsive MCD [18]. However, a minority of azathioprine (n = 6), and 86% for prednisone plus
patients with nephrotic FSGS achieve stable remission after CsA (n = 10) [155]. Using the Glomerular Disease
tapering corticosteroids to complete withdrawal. Most cases Collaborative Network (GDCN) Registry, Laurin
are SD, usually to a high threshold dose. This poses a serious et al. [156] showed that early use of immunosuppressive
problem, as pursuing indefinite treatment with a high- therapy with CNIs and/or corticosteroids was associated
maintenance dose leads to corticosteroid side effects. with improved renal outcome compared with no immu-
Corticosteroid dependency or resistance leads to the need to nosuppression in 458 adults and children with primary
consider other treatment options aimed at reducing proteinu- FSGS (HR 0.49, 0.28–0.86). Although there was a trend
ria, but which trade corticosteroid toxicity for other toward greater survival among patients treated with
complications. CNIs, the superiority of CNIs over corticosteroids alone
In a pilot study, 24 nephrotic patients with idiopathic remained unproven.
FSGS, including six SD and 15 SR, were treated with ACTH For SD or SR patients, two RCTs noted the efficacy of
gel (median dose was 80 units injected subcutaneously twice CsA [157, 158]. Remission rates are 69% and 60%, but
weekly). After treatment, 29% (7/24) patients showed remis- relapse after CsA withdrawal occurred in 61% and 69%,
sion. All remitters had SR (n = 5) or SD (n = 2) FSGS. ACTH respectively. In the former study [157], CsA (3.5 mg/kg/
treatment might be an alternative option for SR or SD day) was combined with low-dose prednisone (0.15 mg/
patients with FSGS, although the response rate is low [151]. kg/day). Long-term renal function was significantly bet-
In a observational study, 20 patients of post-transplant recur- ter preserved in the CsA group, suggesting that low-dose
rent and de novo FSGS resistant to conventional therapy CsA-induced PR reduces the rate of progression of
with therapeutic PE and RTX received ACTH, and 50% nephrotic FSGS patients to renal insufficiency. A compar-
(10/20) of patients achieved CR or PR [152]. ison was made in 226 patients treated for SR-NS (the
majority were FSGS with some cases of MCD), with 127
Alkylating Agents receiving CsA alone and 103 receiving CsA plus corticos-
The best indication for use of an alkylating agent is the case teroids. The proportions of patients with CR, PR, and fail-
of a SD or multirelapsing patient, which is rare in nephrotic ure were, respectively, 14% versus 24%, 12% versus 24%,
FSGS. Some studies included patients treated with a combi- and 74% versus 52%. Thus, CsA combined with low-dose
nation of cyclophosphamide or chlorambucil and corticos- corticosteroids can be considered both an immunosup-
teroids. A RCT in 57 patients by Heering et al. [153] included pressive drug and a steroid-sparing agent to treat resistant
23 patients treated for 6 months with corticosteroids and forms of FSGS.
chlorambucil. They achieved 17% CR and 48% PR. Five of the Another case series [121] was based on 68 patients
23 progressed to end-stage renal disease (ESRD). The contri- treated with 5.18 ± 0.94 mg/kg/day of CsA (Sandimmune
butions of other studies to treatment recommendations formulation). The results on NS were 14 CR (21%), 19 PR
about alkylating agents are uncertain, such as Martinelli (28%), and 35 failures (51%). Other case series have com-
et al.’s case series [154]. In short, there is no strong evidence prised patients treated with CsA (among other immuno-
to recommend alkylating agents in the treatment of FSGS. suppressants) and support the beneficial effect of CsA for
inducing at least PR in nephrotic FSGS.
CNIs TAC
There are no RCTs for this medication in FSGS. So far, the
CsA experience acquired is mainly in CsA-resistant or -dependent
CsA has been used for three decades in the treatment of patients. Another small observational study used TAC mon-
FSGS. It has been shown that its efficacy is greatly otherapy as initial treatment in six patients, and all six
increased by a combination with low-dose patients achieved remission after 6.5 ± 5.9 months [159].
Treatmen 225
Sirolimus tion) or complete (<0.3 g/24 h) remission of proteinuria.

Autophagy was suggested to be involved in maintaining This small underpowered study has not provided any con-
cellular health in podocyte and autophagic exhaustion clusions yet [173].
was suggested to be central deriver of podocyte injury in
FSGS [160]. However, translating this finding into clinical Adalimumab
practice dampened the enthusiasm, since treatment with In the Novel Therapies for Resistant FSGS (FONT) phase II
mammalian target of rapamycin (mTOR) inhibition, study, 21 patients with resistant FSGS were assigned to
sirolimus, appears to be ineffective and potentially harm- adalimumab (an anti-TNFα monoclonal antibody), glac-
ful. The current experience with this drug is controversial tose or standard medical therapy. While none of the
and inadequate to allow recommendations on its use in adalimumab-treated subjects achieved the primary out-
FSGS [161]. come, two subjects in the galactose and two in the standard
medical therapy arm had a 50% reduction in proteinuria
without a decline in eGFR [174].
MMF
The current experience regarding MMF in the treatment of
Abatacept
FSGS seems more or less encouraging [162–164]. A recent
Considering the role of CD80 in pathogenesis of MCD, an
RCT compared CsA to the combination of MMF and high-
inhibitor of CD80, abatacept, was suggested to be used in
dose dexamethasone in children and young adults with
disease treatment. Abatacept (CTLA-4 immunoglobin) is a
SD-FSGS [165]. No difference in treatment effectiveness
soluble fusion protein that consists of the extracellular
was shown between the two groups, and adverse effects
domain of human CTLA-4 linked to the modified Fc por-
were comparable. However, the limitations of this study,
tion of human IgG1. It significantly reduced the proteinu-
regarding the limited number of patients (2.5 patients per
ria in RTX-resistant recurrent FSGS after transplantation
center) and the heterogeneous population (children and
and in one patient with SR primary FSGS with CD80 stain-
adults, different ethnity origin), make it difficult to indicate
ing of podocytes in kidney-biopsy specimens [175].
that CsA treatment might be superior.
However, the effectiveness of abatacept was not replicated
in subsequent studies [176, 177]. Larger RCTs are required
Monoclonal Antibodies to determine its efficacy and safety in the future.
RTX
The efficacy of RTX in adult-onset FSGS remains conflict Summary
[166]. Case series suggest RTX therapy could increase the Nephrotic FSGS, irrespective of its histopathologic variant,
CR or PR rates, decrease the relapse rate, and stop the con- is always an indication for a course of full-dose corticoster-
comitant therapy to include corticosteroids and immuno- oids, preferably prednisone (opinion), for a minimum of
suppressive agents in patients with a severe disease course, 4 months prior to labeling the SD patient (opinion). CR
a failure of previously used immunosuppressive regimens, portends an excellent prognosis. PR is beneficial, both clin-
and kidney transplantation recipients with recurrent ically and in terms of preservation of renal function. In the
FSGS [134, 167–171]. In a 2-year observational study, a case of SD, the use of a combination of CsA and low-dose
high dose of RTX (eight weekly doses of 375 mg/m2) failed corticosteroids obtains partial or, more rarely, CR in about
to improve proteinuria in seven out of eight patients [172]. 50–70% of patients, with a definite advantage regarding
A multicenter off–on trial primarily evaluated the effects of long-term preservation of renal function. Among the vari-
RTX followed by immunosuppressive agent withdrawal on ous other immunosuppressants tried in the treatment of
disease recurrence in 10 children and 20 adults MCD/ FSGS, MMF has the advantage of no nephrotoxicity, but
MesGN (n = 22) or FSGS. Overall, the evidence of RTX in the studies to date, although positive, have been small and
FSGS is mainly drawn from case reports or in a retrospec- pilot in nature. With our understanding of disease patho-
tive manner with various treatment modes, and several genesis, clinical trials of targeting therapy options with
questions remain to be answered, such as dose, duration, more efficacy and fewer adverse effects are ongoing (RCTs
and long-term follow-up outcomes. tending to revise treatment recommendations are summa-
rized in Table 15.5).
Fresolimumab
In a phase 2 double-blind, placebo-controlled, randomized
study of Fresolimumab in patients with SD primary FSGS, Plasmapheresis
patients received placebo (n = 10), fresolimumab 1 mg/kg The rationale for treating nephrotic FSGS with PE is
(n = 14), and fresolimumab 4 mg/kg (n = 12), respectively. based on the concept of a glomerular permeability factor,
However, none of the patients achieved partial (50% reduc- postulated by Shalhoub in 1974 [178], and despite this
Table 15.5 Summary of findings: adult focal segmental glomerulosclerosis (FSGS).
No. of Absolute effect per Certainty of

participants 1000 patients treated Relative risk the evidence
Outcomes (no. of studies) for 1 year (95% CI) (95% CI) (GRADE) Conclusion
CsA vs. MMF/dexamethasone [165]

Remission rate 138 (1) Not estimable 1.38 (0.91–2.09) Low CsA and MMF/dexamethasone show
(CR or PR) ●○○○ similar remission rate in patients with
steroid-resistant FSGS
Relapse rate 48 (1) Not estimable 2.14 (0.84–5.45) Low CsA and MMF/dexamethasone show
●○○○ similar relapse rate in patients with
Adverse event 138 (1) Not estimable Not estimable Low CsA and MMF/dexamethasone show
rate ●○○○ similar adverse event rate in patients with
CsA/prednisolone vs. intravenous methylprednisolone [158]
Remission rate 25 (1) Not estimable 1.69 (0.92–3.12) Low There was no significant difference in
(CR or PR) ●○○○ remission rate between the two groups
Relapse rate 15 (1) Not estimable Not estimable Absent A high relapse rate (69%) occurs after CsA
withdrawal
Adverse event 25 (1) Not estimable Not estimable Absent A high adverse event rate (25%) receiving
rate occurs in the intravenous
methylprednisolone group
CsA/low-dose prednisone vs. prednisone [157]
Remission rate 49 (1) Not estimable 8.85 (1.22–63.92) Moderate There was a significant increase in the
(CR or PR) ●●○○ number of patients with remission in the
CsA/low-dose prednisone group
Relapse rate 35 (1) Not estimable Not estimable Absent A high relapse rate (60%) occurs in CsA/
low-dose prednisone group by week 78
Adverse event 49 (1) Not estimable Not estimable Absent Nausea and vomiting were reported in one
rate patient after in the CsA/low-dose
prednisone group
CI, confidence interval; CR, complete remission; CsA, cyclosporin; MMF, mycophenolate mofetil; PR, partial remission; TAC, tacrolimus.
GRADE assessment of the certainty of the evidence: High; This research provides a very good indication of the likely effect. The likelihood that
the effect will be substantially different* is low. Moderate; This research provides a good indication of the likely effect. The likelihood that the
effect will be substantially different* is moderate. Low; This research provides some indication of the likely effect. However, the likelihood that
it will be substantially different* is high. Very low; This research does not provide a reliable indication of the likely effect. The likelihood that the
effect will be substantially different* is very high. *Substantially different = a large enough difference that it might affect decision-making. Pooled
analyses were not performed due to the heterogeneity of the data.
factor being the subject of numerous publications, much there was only poor evidence to support its use, and no
remains uncertain about it [179]. There is no evidence significant additional data have been added since that
that PE modifies the course of FSGS, outside of the par- time. A systematic review and meta-analysis of 77 case
ticular case of relapse of NS in a kidney transplant recipi- reports and case series noted that 71% of patients achieved
ent [144]. Based on the concept of a “circulating factor,” CR or PR after treatment with PE [181].
PE has been advocated to remove it, reduce proteinuria, Similarly, plasma protein adsorption on columns coated
and avoid the progression to renal insufficiency, but there with staphylococcal protein A [182] does not seem to
is little evidence for or against its use. In 2001 Bosch and remove specifically the elusive circulating factor responsi-
Wendler [180] concluded, in a review of extracorporeal ble for the proteinuria, as this experimental protocol also
plasma treatment in primary and recurrent FSGS, that reduces proteinuria in other glomerulopathies [183].
Reference 227
Pregnancy during pregnancy continues to be an unresolved issue. In

summary, CsA seems to be the less toxic agent to treat a
INS occurring during pregnancy poses a unique situation.
case of INS during pregnancy, but the data are weak.
The majority of the data relative to the effects of immuno-
suppressive treatment in pregnancy are based on post-
kidney transplantation reports. High-dose corticosteroids
entail a risk to the fetus [184]. Alkylating agents are con- C
onclusion
traindicated. In the CsA era, case reports, center reports,
and registry data have supported the concept that success- Both similarities and differences regarding the clinical
ful pregnancy outcomes are possible for mother and new- characteristics, pathogenesis, treatment responses, and
born in the presence of stable graft function and outcomes were shown between MCD and idiopathic FSGS.
well-controlled maternal comorbidities, such as hyperten- There is still a debate on whether they are different pheno-
sion. While the risks of prematurity and low birth weight types of the same disease or two entities. Corticosteroids
are greater than those seen in the general population, there and CNIs remain the mainstay treatment in MCD and idi-
has been no increase in the malformation rate of new- opathic FSGS, while RTX seems to be a promising alterna-
borns. Successful pregnancy outcomes have been reported tive. Clinicopathological studies and experimental models
in nonrenal CsA recipients as well [185–187]. The available support the pathogenesis that MCD and FSGS are podocyte
experience with MMF has been limited to organ transplan- disorders caused by circulating factors and/or immune
tation [185–187]. Two newborns with malformations were dysfunction. Lifting the veil of these factors and the under-
noted in a limited case series with MMF exposure, but lying mechanisms provides diagnostic, therapeutic, and
other factors may have also been at play. The use of MMF prognostic insights.
R
eferences
Bargman, J.M. (1999). Management of minimal lesion

1 9 D’Agati, V.D., Alster, J.M., Jennette, J.C. et al. (2013).
glomerulonephritis: evidence-based recommendations. Association of histologic variants in FSGS clinical trial
Kidney Int. (Suppl. 70): S3–S16. with presenting features and outcomes. Clin. J. Am. Soc.
2 KDIGO Work Group (2012). KDIGO clinical practice Nephrol. 8 (3): 399–406.
guideline for glomerulonephritis. Kidney Int. (Suppl. 2): 10 Tune, B.M. and Mendoza, S.A. (1997). Treatment of the
139–274. idiopathic nephrotic syndrome: regimens and outcomes
3 Howie, A.J., Pankhurst, T., Sarioglu, S. et al. (2005). in children and adults. J. Am. Soc. Nephrol. 8 (5):
Evolution of nephrotic-associated focal segmental 824–832.
glomerulosclerosis and relation to the glomerular tip 11 Kitiyakara, C., Eggers, P., and Kopp, J.B. (2004). Twenty-
lesion. Kidney Int. 67 (3): 987–1001. one-year trend in ESRD due to focal segmental
4 Haas, M. (2005). The glomerular tip lesion: what does it glomerulosclerosis in the United States. Am. J. Kidney
really mean? Kidney Int. 67 (3): 1188–1189. Dis. 44 (5): 815–825.
5 Stokes, M.B., Markowitz, G.S., Lin, J. et al. (2004). 12 O’Shaughnessy, M.M., Hogan, S.L., Thompson, B.D.
Glomerular tip lesion: a distinct entity within the minimal et al. (2018). Glomerular disease frequencies by race, sex
change disease/focal segmental glomerulosclerosis and region: results from the International Kidney
spectrum. Kidney Int. 65 (5): 1690–1702. Biopsy Survey. Nephrol. Dial. Transplant. 33 (4):
6 Troyanov, S., Wall, C.A., Miller, J.A. et al. (2005). Focal and 661–669.
segmental glomerulosclerosis: definition and relevance of 13 Braden, G.L., Mulhern, J.G., O’Shea, M.H. et al. (2000).
a partial remission. J. Am. Soc. Nephrol. 16 (4): 1061–1068. Changing incidence of glomerular diseases in adults. Am.
Toronto Glomerulonephritis Registry Group J. Kidney Dis. 35 (5): 878–883.
7 Korbet, S.M. (2012). Treatment of primary FSGS in adults. 14 Dragovic, D., Rosenstock, J.L., Wahl, S.J. et al. (2005).
J. Am. Soc. Nephrol. 23 (11): 1769–1776. Increasing incidence of focal segmental
8 Chun, M.J., Korbet, S.M., Schwartz, M.M., and Lewis, E.J. glomerulosclerosis and an examination of demographic
(2004). Focal segmental glomerulosclerosis in nephrotic patterns. Clin. Nephrol. 63 (1): 1–7.
adults: presentation, prognosis, and response to therapy of 15 DH, I.J., Farris, A.B., Goemaere, N. et al. (2008). Fidelity
the histologic variants. J. Am. Soc. Nephrol. 15 (8): and evolution of recurrent FSGS in renal allografts. J. Am.
2169–2177. Soc. Nephrol. 19 (11): 2219–2224.
6 Gallon, L., Leventhal, J., Skaro, A. et al. (2012).

1 30 Rosenberg, A.Z. and Kopp, J.B. (2017). Focal segmental
Resolution of recurrent focal segmental glomerulosclerosis. Clin. J. Am. Soc. Nephrol. 12 (3):
glomerulosclerosis after retransplantation. N. Engl. J. 502–517.
Med. 366 (17): 1648–1649. 31 Konigshausen, E. and Sellin, L. (2016). Circulating
17 Maas, R.J., Deegens, J.K., Smeets, B. et al. (2016). permeability factors in primary focal segmental
Minimal change disease and idiopathic FSGS: glomerulosclerosis: a review of proposed candidates.
manifestations of the same disease. Nat. Rev. Nephrol. 12 Biomed Res. Int. 2016: 3765608.
(12): 768–776. 32 Reggiani, F. and Ponticelli, C. (2016). Focal segmental
18 Korbet, S.M. (1999). Clinical picture and outcome of glomerular sclerosis: do not overlook the role of immune
primary focal segmental glomerulosclerosis. Nephrol. response. J. Nephrol. 29 (4): 525–534.
Dial. Transplant. 14 (Suppl 3): 68–73. 33 Wei, C., Moller, C.C., Altintas, M.M. et al. (2008).
19 Fujihara, C.K., Mattar, A.L., Vieira, J.M. Jr. et al. (2002). Modification of kidney barrier function by the urokinase
Evidence for the existence of two distinct functions for receptor. Nat. Med. 14 (1): 55–63.
the inducible NO synthase in the rat kidney: effect of 34 Wei, C., El Hindi, S., Li, J. et al. (2011). Circulating
aminoguanidine in rats with 5/6 ablation. J. Am. Soc. urokinase receptor as a cause of focal segmental
Nephrol. 13 (9): 2278–2287. glomerulosclerosis. Nat. Med. 17 (8): 952–960.
20 Schwartz, M.M., Evans, J., Bain, R., and Korbet, S.M. 35 Wei, C., Trachtman, H., Li, J. et al. (2012). Circulating
(1999). Focal segmental glomerulosclerosis: prognostic suPAR in two cohorts of primary FSGS. J. Am. Soc.
implications of the cellular lesion. J. Am. Soc. Nephrol. 10 Nephrol. 23 (12): 2051–2059.
(9): 1900–1907. 36 Hayek, S.S., Sever, S., Ko, Y.A. et al. (2015). Soluble
urokinase receptor and chronic kidney disease. N. Engl. J.
21 Crook, E.D., Habeeb, D., Gowdy, O. et al. (2005). Effects
Med. 373 (20): 1916–1925.
of steroids in focal segmental glomerulosclerosis in a
37 Senaldi, G., Stolina, M., Guo, J. et al. (2002). Regulatory
predominantly African-American population. Am. J. Med.
effects of novel neurotrophin-1/b cell-stimulating factor-3
Sci. 330 (1): 19–24.
(cardiotrophin-like cytokine) on B cell function. J.
22 Chehade, H., Cachat, F., Girardin, E. et al. (2013). Two
Immunol. 168 (11): 5690–5698.
new families with hereditary minimal change disease.
38 Sharma, M., Zhou, J., Gauchat, J.F. et al. (2015). Janus
BMC Nephrol. 14: 65.
kinase 2/signal transducer and activator of transcription
23 Barua, M., Stellacci, E., Stella, L. et al. (2014). Mutations
3 inhibitors attenuate the effect of cardiotrophin-like
in PAX2 associate with adult-onset FSGS. J. Am. Soc.
cytokine factor 1 and human focal segmental
Nephrol. 25 (9): 1942–1953.
glomerulosclerosis serum on glomerular filtration barrier.
24 Lim, B.J., Yang, J.W., Do, W.S., and Fogo, A.B. (2016).
Transl. Res. 166 (4): 384–398.
Pathogenesis of focal segmental glomerulosclerosis. J.
39 Delville M, Sigdel TK, Wei C, et al. (2014). A circulating
Pathol. Transl. Med. 50 (6): 405–410.
antibody panel for pretransplant prediction of FSGS
25 Liu, J. and Wang, W. (2017). Genetic basis of adult-onset recurrence after kidney transplantation. Sci. Transl. Med.
nephrotic syndrome and focal segmental 6(256):256ra136.
glomerulosclerosis. Front. Med. 11 (3): 333–339. 40 Stangou, M., Spartalis, M., Daikidou, D.V. et al. (2017).
26 Lepori, N., Zand, L., Sethi, S. et al. (2018). Clinical and Impact of Tauh1 and Tauh2 cytokines in the progression
pathological phenotype of genetic causes of focal of idiopathic nephrotic syndrome due to focal segmental
segmental glomerulosclerosis in adults. Clin. Kidney J. 11 glomerulosclerosis and minimal change disease. J.
(2): 179–190. Nephropathol. 6 (3): 187–195.
27 Rovin, B.H., Caster, D.J., Cattran, D.C. et al. (2019). 41 Lai, K.W., Wei, C.L., Tan, L.K. et al. (2007).
Management and treatment of glomerular diseases (part Overexpression of interleukin-13 induces minimal-
2): conclusions from a Kidney Disease: Improving Global change-like nephropathy in rats. J. Am. Soc. Nephrol. 18
Outcomes (KDIGO) Controversies Conference. Kidney (5): 1476–1485.
Int. 95 (2): 281–295. 42 Liu, L.L., Qin, Y., Cai, J.F. et al. (2011). Th17/Treg
28 D’Agati, V.D., Kaskel, F.J., and Falk, R.J. (2011). Focal imbalance in adult patients with minimal change
segmental glomerulosclerosis. N. Engl. J. Med. 365 (25): nephrotic syndrome. Clin. Immunol. 139 (3): 314–320.
2398–2411. 43 Schewitz-Bowers, L.P., Lait, P.J., Copland, D.A. et al.
29 Chen, Y.M. and Liapis, H. (2015). Focal segmental (2015). Glucocorticoid-resistant Th17 cells are selectively
glomerulosclerosis: molecular genetics and targeted attenuated by cyclosporine A. Proc. Natl. Acad. Sci. USA.
therapies. BMC Nephrol. 16: 101. 112 (13): 4080–4085.
Reference 229
4 Salcido-Ochoa, F., Hue, S.S., Haase, D. et al. (2017).

4 focal segmental glomerulosclerosis. Nephrol. Dial.
Analysis of T cell subsets in adult primary/idiopathic Transplant. 28 (12): 2993–3003.
minimal change disease: a pilot study. Int. J. Nephrol. 57 Potla, U., Ni, J., Vadaparampil, J. et al. (2014). Podocyte-
2017: 3095425. specific RAP1GAP expression contributes to focal
45 Sfikakis, P.P., Boletis, J.N., Lionaki, S. et al. (2005). segmental glomerulosclerosis-associated glomerular
Remission of proliferative lupus nephritis following B cell injury. J. Clin. Invest. 124 (4): 1757–1769.
depletion therapy is preceded by down-regulation of the 58 Segarra, A., Jatem, E., Quiles, M.T. et al. (2014).
T cell costimulatory molecule CD40 ligand: an open-label Diagnostic value of soluble urokinase-type plasminogen
trial. Arthritis Rheum. 52 (2): 501–513. activator receptor serum levels in adults with idiopathic
46 Stasi, R., Cooper, N., Del Poeta, G. et al. (2008). Analysis nephrotic syndrome. Nefrologia 34 (1): 46–52.
of regulatory T-cell changes in patients with idiopathic 59 Segarra, A., Jatem, E., Quiles, M.T. et al. (2014). Value of
thrombocytopenic purpura receiving B cell-depleting soluble urokinase receptor serum levels in the differential
therapy with rituximab. Blood 112 (4): 1147–1150. diagnosis between idiopathic and secondary focal
47 Sfikakis, P.P., Souliotis, V.L., Fragiadaki, K.G. et al. (2007). segmental glomerulosclerosis. Nefrologia 34 (1): 53–61.
Increased expression of the FoxP3 functional marker of 60 Huang, J., Liu, G., Zhang, Y.M. et al. (2013). Plasma
regulatory T cells following B cell depletion with soluble urokinase receptor levels are increased but do not
rituximab in patients with lupus nephritis. Clin. distinguish primary from secondary focal segmental
Immunol. 123 (1): 66–73. glomerulosclerosis. Kidney Int. 84 (2): 366–372.
48 Saadoun, D., Rosenzwajg, M., Landau, D. et al. (2008). 61 Li, F., Zheng, C., Zhong, Y. et al. (2014). Relationship
Restoration of peripheral immune homeostasis after between serum soluble urokinase plasminogen activator
rituximab in mixed cryoglobulinemia vasculitis. Blood receptor level and steroid responsiveness in FSGS. Clin. J.
111 (11): 5334–5341. Am. Soc. Nephrol. 9 (11): 1903–1911.
49 Tokunaga, M., Fujii, K., Saito, K. et al. (2005). Down- 62 Meijers, B., Maas, R.J., Sprangers, B. et al. (2014). The
regulation of CD40 and CD80 on B cells in patients with soluble urokinase receptor is not a clinical marker for
life-threatening systemic lupus erythematosus after focal segmental glomerulosclerosis. Kidney Int. 85 (3):
successful treatment with rituximab. Rheumatology 636–640.
(Oxford) 44 (2): 176–182. 63 Franco Palacios, C.R., Lieske, J.C., Wadei, H.M. et al.
50 Fornoni, A., Sageshima, J., Wei, C. et al. (2011). (2013). Urine but not serum soluble urokinase receptor
Rituximab targets podocytes in recurrent focal segmental (suPAR) may identify cases of recurrent FSGS in kidney
glomerulosclerosis. Sci. Transl. Med. 3 (85): 85ra46. transplant candidates. Transplantation 96 (4): 394–399.
51 Shimada, M., Ishimoto, T., Lee, P.Y. et al. (2012). Toll-like 64 Maas, R.J., Wetzels, J.F., and Deegens, J.K. (2014). Serum
receptor 3 ligands induce CD80 expression in human suPAR concentrations in patients with focal segmental
podocytes via an NF-kappaB-dependent pathway. glomerulosclerosis with end-stage renal disease. Kidney
Nephrol. Dial. Transplant. 27 (1): 81–89. Int. 85 (3): 711.
52 Jain, N., Khullar, B., Oswal, N. et al. (2016). TLR- 65 Maas, R.J., Deegens, J.K., and Wetzels, J.F. (2013). Serum
mediated albuminuria needs TNFalpha-mediated suPAR in patients with FSGS: trash or treasure? Pediatr.
cooperativity between TLRs present in hematopoietic Nephrol. 28 (7): 1041–1048.
tissues and CD80 present on non-hematopoietic tissues in 66 Schlondorff, D. (2014). Are serum suPAR determinations
mice. Dis. Model Mech. 9 (6): 707–717. by current ELISA methodology reliable diagnostic
53 Shimada, M., Araya, C., Rivard, C. et al. (2011). Minimal biomarkers for FSGS? Kidney Int. 85 (3): 499–501.
change disease: a “two-hit” podocyte immune disorder? 67 Zhang, Q., Zeng, C., Fu, Y. et al. (2012). Biomarkers of
Pediatr. Nephrol. 26 (4): 645–649. endothelial dysfunction in patients with primary focal
54 Haraldsson, B. (2013). A new era of podocyte-targeted segmental glomerulosclerosis. Nephrology (Carlton) 17
therapy for proteinuric kidney disease. N. Engl. J. Med. (4): 338–345.
369 (25): 2453–2454. 68 Zhao, B., Han, H., Zhen, J. et al. (2018). CD80 and
55 Gebeshuber, C.A., Kornauth, C., Dong, L. et al. (2013). CTLA-4 as diagnostic and prognostic markers in adult-
Focal segmental glomerulosclerosis is induced by onset minimal change disease: a retrospective study. PeerJ
microRNA-193a and its downregulation of WT1. Nat. 6: e5400.
Med. 19 (4): 481–487. 69 Novelli, R., Gagliardini, E., Ruggiero, B. et al. (2016). Any
56 Miura, K., Kurihara, H., Horita, S. et al. (2013). Podocyte value of podocyte B7-1 as a biomarker in human MCD
expression of nonmuscle myosin heavy chain-IIA and FSGS? Am. J. Physiol. Renal Physiol. 310 (5):
decreases in idiopathic nephrotic syndrome, especially in F335–F341.
70 Garin, E.H., Mu, W., Arthur, J.M. et al. (2010). Urinary 83 Perez, V., Lopez, D., Boixadera, E. et al. (2017).
CD80 is elevated in minimal change disease but not in focal Comparative differential proteomic analysis of minimal
segmental glomerulosclerosis. Kidney Int. 78 (3): 296–302. change disease and focal segmental glomerulosclerosis.
71 Cara-Fuentes, G., Wei, C., Segarra, A. et al. (2014). CD80 BMC Nephrol. 18 (1): 49.
and suPAR in patients with minimal change disease and 84 Kalantari, S., Nafar, M., Samavat, S. et al. (2014). Urinary
focal segmental glomerulosclerosis: diagnostic and prognostic biomarkers in patients with focal segmental
pathogenic significance. Pediatr. Nephrol. 29 (8): glomerulosclerosis. Nephrourol. Mon. 6 (2): e16806.
1363–1371. 85 Hao, X., Liu, X., Wang, W. et al. (2013). Distinct metabolic
72 Ling, C., Liu, X., Shen, Y. et al. (2015). Urinary profile of primary focal segmental glomerulosclerosis
CD80 levels as a diagnostic biomarker of minimal change revealed by NMR-based metabolomics. PLoS One 8 (11):
disease. Pediatr. Nephrol. 30 (2): 309–316. e78531.
73 Smeets, B., Stucker, F., Wetzels, J. et al. (2014). Detection 86 Ramezani, A., Devaney, J.M., Cohen, S. et al. (2015).
of activated parietal epithelial cells on the glomerular tuft Circulating and urinary microRNA profile in focal
distinguishes early focal segmental glomerulosclerosis segmental glomerulosclerosis: a pilot study. Eur. J. Clin.
from minimal change disease. Am. J. Pathol. 184 (12): Invest. 45 (4): 394–404.
3239–3248. 87 Zhang, W., Zhang, C., Chen, H. et al. (2014). Evaluation
74 Froes, B.P., de Almeida Araujo, S., Bambirra, E.A. et al. of microRNAs miR-196a, miR-30a-5P, and miR-490 as
(2017). Is CD44 in glomerular parietal epithelial cells a biomarkers of disease activity among patients with FSGS.
pathological marker of renal function deterioration in Clin. J. Am. Soc. Nephrol. 9 (9): 1545–1552.
primary focal segmental glomerulosclerosis? Pediatr. 88 Cai, X., Xia, Z., Zhang, C. et al. (2013). Serum microRNAs
Nephrol. 32 (11): 2165–2169. levels in primary focal segmental glomerulosclerosis.
75 Lausecker, F., Tian, X., Inoue, K. et al. (2018). Vinculin is Pediatr. Nephrol. 28 (9): 1797–1801.
required to maintain glomerular barrier integrity. Kidney 89 Zhang, C., Zhang, W., Chen, H.M. et al. (2015). Plasma
Int. 93 (3): 643–655. microRNA-186 and proteinuria in focal segmental
76 Li, J.S., Chen, X., Peng, L. et al. (2015). Angiopoietin- glomerulosclerosis. Am. J. Kidney Dis. 65 (2): 223–232.
like-4, a potential target of tacrolimus, predicts earlier 90 Xiao, B., Wang, L.N., Li, W. et al. (2018). Plasma
podocyte injury in minimal change disease. PLoS One 10 microRNA panel is a novel biomarker for focal segmental
(9): e0137049. glomerulosclerosis and associated with podocyte
77 Cara-Fuentes, G., Segarra, A., Silva-Sanchez, C. et al. apoptosis. Cell Death Dis. 9 (5): 533.
(2017). Angiopoietin-like-4 and minimal change disease. 91 Moura, L.R., Franco, M.F., and Kirsztajn, G.M. (2015).
PLoS One 12 (4): e0176198. Minimal change disease and focal segmental
78 Motojima, M., Matsusaka, T., Kon, V., and Ichikawa, I. glomerulosclerosis in adults: response to steroids and risk
(2010). Fibrinogen that appears in Bowman’s space of of renal failure. J. Bras. Nefrol. 37 (4): 475–480.
proteinuric kidneys in vivo activates podocyte Toll-like 92 Lee, H., Yoo, K.D., Oh, Y.K. et al. (2016). Predictors of
receptors 2 and 4 in vitro. Nephron Exp. Nephrol. 114 (2): relapse in adult-onset nephrotic minimal change disease.
e39–e47. Medicine (Baltimore) 95 (12): e3179.
79 Sorensen, I., Susnik, N., Inhester, T. et al. (2011). 93 Shinzawa, M., Yamamoto, R., Nagasawa, Y. et al. (2013).
Fibrinogen, acting as a mitogen for tubulointerstitial Age and prediction of remission and relapse of
fibroblasts, promotes renal fibrosis. Kidney Int. 80 (10): proteinuria and corticosteroid-related adverse events in
1035–1044. adult-onset minimal-change disease: a retrospective
80 Craciun, F.L., Ajay, A.K., Hoffmann, D. et al. (2014). cohort study. Clin. Exp. Nephrol. 17 (6): 839–847.
Pharmacological and genetic depletion of fibrinogen 94 Huang, J.J., Hsu, S.C., Chen, F.F. et al. (2001). Adult-
protects from kidney fibrosis. Am. J. Physiol. Renal onset minimal change disease among Taiwanese: clinical
Physiol. 307 (4): F471–F484. features, therapeutic response, and prognosis. Am. J.
81 Wang, Y., Zheng, C., Xu, F., and Liu, Z. (2016). Urinary Nephrol. 21 (1): 28–34.
fibrinogen and renal tubulointerstitial fibrinogen deposition: 95 McIntyre, P. and Craig, J.C. (1998). Prevention of serious
discriminating between primary FSGS and minimal change bacterial infection in children with nephrotic syndrome.
disease. Biochem. Biophys. Res. Commun. 478 (3): 1147–1152. J. Paediatr. Child Health 34 (4): 314–317.
82 Samejima, K., Nakatani, K., Suzuki, D. et al. (2012). 96 Singhal, R. and Brimble, K.S. (2006). Thromboembolic
Clinical significance of fibroblast-specific protein-1 complications in the nephrotic syndrome:
expression on podocytes in patients with focal segmental pathophysiology and clinical management. Thromb. Res.
glomerulosclerosis. Nephron Clin. Pract. 120 (1): c1–c7. 118 (3): 397–407.
Reference 231
97 Mahmoodi, B.K., ten Kate, M.K., Waanders, F. et al. 110 Ozeki, T., Ando, M., Yamaguchi, M. et al. (2018).
(2008). High absolute risks and predictors of venous and Treatment patterns and steroid dose for adult minimal
arterial thromboembolic events in patients with change disease relapses: a retrospective cohort study.
nephrotic syndrome: results from a large retrospective PLoS One 13 (6): e0199228.
cohort study. Circulation 117 (2): 224–230. 111 Palmer, S.C., Nand, K., and Strippoli, G.F. (2008).
98 Meyrier, A. and Niaudet, P. (2018). Acute kidney injury Interventions for minimal change disease in adults with
complicating nephrotic syndrome of minimal change nephrotic syndrome. Cochrane Database Syst. Rev. (1):
disease. Kidney Int. 94 (5): 861–869. CD001537.
99 Komukai, D., Hasegawa, T., Kaneshima, N. et al. (2016). 112 Imbasciati, E., Gusmano, R., Edefonti, A. et al. (1985).
Influence of acute kidney injury on the time to complete Controlled trial of methylprednisolone pulses and low
remission in adult minimal change nephrotic syndrome: dose oral prednisone for the minimal change nephrotic
a single-centre study. Nephrology (Carlton) 21 (10): syndrome. Br. Med. J. (Clin. Res. Ed.) 291 (6505):
887–892. 1305–1308.
100 Huang, M.J., Wei, R.B., Su, T.Y. et al. (2016). Impact of 113 Li, X., Liu, Z., Wang, L. et al. (2017). Tacrolimus
acute kidney injury on coagulation in adult minimal monotherapy after intravenous methylprednisolone in
change nephropathy. Medicine (Baltimore) 95 (46): adults with minimal change nephrotic syndrome. J. Am.
e5366. Soc. Nephrol. 28 (4): 1286–1295.
101 Gipson, D.S., Massengill, S.F., Yao, L. et al. (2009). 114 Fujiwara, A., Hirawa, N., Kobayashi, Y. et al. (2015).
Management of childhood onset nephrotic syndrome. Efficacy of cyclosporine combination therapy for
Pediatrics 124 (2): 747–757. new-onset minimal change nephrotic syndrome in
102 Hodson, E.M., Willis, N.S., and Craig, J.C. (2007). adults. Clin. Exp. Nephrol. 19 (2): 240–246.
Corticosteroid therapy for nephrotic syndrome in 115 Raja, K., Parikh, A., Webb, H., and Hothi, D. (2017). Use
children. Cochrane Database Syst. Rev. (4): of a low-dose prednisolone regimen to treat a relapse of
CD001533. steroid-sensitive nephrotic syndrome in children.
103 Mak, S.K., Short, C.D., and Mallick, N.P. (1996). Pediatr. Nephrol. 32 (1): 99–105.
Long-term outcome of adult-onset minimal-change 116 Bomback, A.S., Canetta, P.A., Beck, L.H. Jr. et al. (2012).
nephropathy. Nephrol. Dial. Transplant. 11 (11): Treatment of resistant glomerular diseases with
2192–2201. adrenocorticotropic hormone gel: a prospective trial.
104 Waldman, M., Crew, R.J., Valeri, A. et al. (2007). Adult Am. J. Nephrol. 36 (1): 58–67.
minimal-change disease: clinical characteristics, 117 Al-Khader, A.A., Lien, J.W., and Aber, G.M. (1979).
treatment, and outcomes. Clin. J. Am. Soc. Nephrol. 2 Cyclophosphamide alone in the treatment of adult
(3): 445–453. patients with minimal change glomerulonephritis. Clin.
105 Korbet, S.M., Schwartz, M.M., and Lewis, E.J. (1988). Nephrol. 11 (1): 26–30.
Minimal-change glomerulopathy of adulthood. Am. J. 118 Uldall, P.R., Feest, T.G., Morley, A.R. et al. (1972).
Nephrol. 8 (4): 291–297. Cyclophosphamide therapy in adults with minimal-
106 Nolasco, F., Cameron, J.S., Heywood, E.F. et al. (1986). change nephrotic syndrome. Lancet 1 (7763):
Adult-onset minimal change nephrotic syndrome: a 1250–1253.
long-term follow-up. Kidney Int. 29 (6): 1215–1223. 119 Nakayama, M., Katafuchi, R., Yanase, T. et al. (2002).
107 Tse, K.C., Lam, M.F., Yip, P.S. et al. (2003). Idiopathic Steroid responsiveness and frequency of relapse in
minimal change nephrotic syndrome in older adults: adult-onset minimal change nephrotic syndrome. Am. J.
steroid responsiveness and pattern of relapses. Nephrol. Kidney Dis. 39 (3): 503–512.
Dial. Transplant. 18 (7): 1316–1320. 120 Fujimoto, S., Yamamoto, Y., Hisanaga, S. et al. (1991).
108 Shinzawa, M., Yamamoto, R., Nagasawa, Y. et al. Minimal change nephrotic syndrome in adults: response
(2014). Comparison of methylprednisolone plus to corticosteroid therapy and frequency of relapse. Am.
prednisolone with prednisolone alone as initial J. Kidney Dis. 17 (6): 687–692.
treatment in adult-onset minimal change disease: a 121 Meyrier, A. (2003). Treatment of idiopathic nephrosis by
retrospective cohort study. Clin. J. Am. Soc. Nephrol. 9 immunophillin modulation. Nephrol. Dial. Transplant.
(6): 1040–1048. 18 (Suppl 6) vi: 79–86.
109 Zhao, L., Cheng, J., Zhou, J. et al. (2015). Enhanced 122 Matsumoto, H., Nakao, T., Okada, T. et al. (2001). Initial
steroid therapy in adult minimal change nephrotic remission-inducing effect of very low-dose cyclosporin
syndrome: a systematic review and meta-analysis. monotherapy for minimal-change nephrotic syndrome
Intern. Med. 54 (17): 2101–2108. in Japanese adults. Clin. Nephrol. 55 (2): 143–148.
123 Meyrier, A., Noel, L.H., Auriche, P., and Callard, P. 136 Iwabuchi, Y., Takei, T., Moriyama, T. et al. (2014).
(1994). Long-term renal tolerance of cyclosporin A Long-term prognosis of adult patients with steroid-
treatment in adult idiopathic nephrotic syndrome. dependent minimal change nephrotic syndrome
Collaborative Group of the Societe de Nephrologie. following rituximab treatment. Medicine (Baltimore) 93
Kidney Int. 45 (5): 1446–1456. (29): e300.
124 Eguchi, A., Takei, T., Yoshida, T. et al. (2010). Combined 137 Papakrivopoulou, E., Shendi, A.M., Salama, A.D. et al.
cyclosporine and prednisolone therapy in adult patients (2016). Effective treatment with rituximab for the
with the first relapse of minimal-change nephrotic maintenance of remission in frequently relapsing
syndrome. Nephrol. Dial. Transplant. 25 (1): 124–129. minimal change disease. Nephrology (Carlton) 21 (10):
125 Serkova, N. and Christians, U. (2003). Transplantation: 893–900.
toxicokinetics and mechanisms of toxicity of 138 King, C., Logan, S., Smith, S.W., and Hewins, P.
cyclosporine and macrolides. Curr. Opin. Invest. Drugs 4 (2017). The efficacy of rituximab in adult frequently
(11): 1287–1296. relapsing minimal change disease. Clin. Kidney J.
126 Filler, G. (2005). How should microemulsified 10 (1): 16–19.
cyclosporine A (Neoral) therapy in patients with 139 Ruggenenti, P., Ruggiero, B., Cravedi, P. et al. (2014).
nephrotic syndrome be monitored? Nephrol. Dial. Rituximab in steroid-dependent or frequently relapsing
Transplant. 20 (6): 1032–1034. idiopathic nephrotic syndrome. J. Am. Soc. Nephrol. 25
127 Kim, Y.C., Lee, T.W., Lee, H. et al. (2012). Complete (4): 850–863.
remission induced by tacrolimus and low-dose 140 Bruchfeld, A., Benedek, S., Hilderman, M. et al. (2014).
prednisolone in adult minimal change nephrotic Rituximab for minimal change disease in adults:
syndrome: a pilot study. Kidney Res. Clin. Pract. 31 (2): long-term follow-up. Nephrol. Dial. Transplant. 29 (4):
112–117. 851–856.
128 Xu, D., Gao, X., Bian, R. et al. (2017). Tacrolimus 141 Munyentwali, H., Bouachi, K., Audard, V. et al. (2013).
improves proteinuria remission in adults with Rituximab is an efficient and safe treatment in adults
cyclosporine A-resistant or -dependent minimal change with steroid-dependent minimal change disease. Kidney
disease. Nephrology (Carlton) 22 (3): 251–256. Int. 83 (3): 511–516.
129 Day, C.J., Cockwell, P., Lipkin, G.W. et al. (2002). 142 Brown, L.C., Jobson, M.A., Payan Schober, F. et al.
Mycophenolate mofetil in the treatment of resistant (2017). The evolving role of rituximab in adult minimal
idiopathic nephrotic syndrome. Nephrol. Dial. change glomerulopathy. Am. J. Nephrol. 45 (4): 365–372.
Transplant. 17 (11): 2011–2013. 143 Fenoglio, R., Sciascia, S., Beltrame, G. et al. (2018).
130 Levin, M.L. (2002). Mycophenolate mofetil treatment Rituximab as a front-line therapy for adult-onset
for primary glomerular diseases. Kidney Int. 62 (4): 1475. minimal change disease with nephrotic syndrome.
131 Norona, B., Valentin, M., Gutierrez, E., and Praga, M. Oncotarget 9 (48): 28799–28804.
(2004). Treatment of steroid-dependent minimal 144 Korbet, S.M. (2002). Treatment of primary focal
change-nephrotic syndrome with mycophenolate segmental glomerulosclerosis. Kidney Int. 62 (6):
mofetil. Nefrologia 24 (1): 79–82. 2301–2310.
132 Amemiya, N., Takei, T., Kojima, C. et al. (2011). 145 Ahmed, M.S. and Wong, C.F. (2008). Rituximab and
Induction of remission following a single dose of nephrotic syndrome: a new therapeutic hope? Nephrol.
rituximab alone in a patient with minimal change Dial. Transplant. 23 (1): 11–17.
nephrotic syndrome. Clin. Exp. Nephrol. 15 (6): 933–936. 146 Ponticelli, C., Villa, M., Banfi, G. et al. (1999). Can
133 Takei, T. and Nitta, K. (2011). Rituximab and minimal prolonged treatment improve the prognosis in adults
change nephrotic syndrome: a therapeutic option. Clin. with focal segmental glomerulosclerosis? Am. J. Kidney
Exp. Nephrol. 15 (5): 641–647. Dis. 34 (4): 618–625.
134 Kronbichler, A., Konig, P., Busch, M. et al. (2013). 147 Alexopoulos, E., Stangou, M., Papagianni, A. et al.
Rituximab in adult patients with multi-relapsing/ (2000). Factors influencing the course and the response
steroid-dependent minimal change disease and focal to treatment in primary focal segmental
segmental glomerulosclerosis: a report of 5 cases. Wien. glomerulosclerosis. Nephrol. Dial. Transplant. 15 (9):
Klin. Wochenschr. 125 (11–12): 328–333. 1348–1356.
135 Takei, T., Itabashi, M., Moriyama, T. et al. (2013). Effect 148 Korbet, S.M., Schwartz, M.M., and Lewis, E.J. (1994).
of single-dose rituximab on steroid-dependent minimal- Primary focal segmental glomerulosclerosis: clinical
change nephrotic syndrome in adults. Nephrol. Dial. course and response to therapy. Am. J. Kidney Dis. 23
Transplant. 28 (5): 1225–1232. (6): 773–783.
Reference 233
49 Korbet, S.M. (1995). Management of idiopathic

1 162 Choi, M.J., Eustace, J.A., Gimenez, L.F. et al. (2002).
nephrosis in adults, including steroid-resistant Mycophenolate mofetil treatment for primary
nephrosis. Curr. Opin. Nephrol. Hypertens. 4 (2): glomerular diseases. Kidney Int. 61 (3): 1098–1014.
169–176. 163 Cattran, D.C., Wang, M.M., Appel, G. et al. (2004).
150 Meyrier, A. (2005). Treatment of focal segmental Mycophenolate mofetil in the treatment of focal
glomerulosclerosis. Expert Opin. Pharmacother. 6 (9): segmental glomerulosclerosis. Clin. Nephrol. 62 (6):
1539–1549. 405–411.
151 Hogan, J., Bomback, A.S., Mehta, K. et al. (2013). 164 Segarra, A., Amoedo, M.L., Martinez Garcia, J.M. et al.
Treatment of idiopathic FSGS with adrenocorticotropic (2007). Efficacy and safety of ’rescue therapy’ with
hormone gel. Clin. J. Am. Soc. Nephrol. 8 (12): mycophenolate mofetil in resistant primary
2072–2081. glomerulonephritis – a multicenter study. Nephrol. Dial.
152 Alhamad, T., Manllo Dieck, J., Younus, U. et al. (2019). Transplant. 22 (5): 1351–1360.
ACTH gel in resistant focal segmental 165 Gipson, D.S., Trachtman, H., Kaskel, F.J. et al. (2011).
glomerulosclerosis after kidney transplantation. Clinical trial of focal segmental glomerulosclerosis in
Transplantation 103 (1): 202–209. children and young adults. Kidney Int. 80 (8):
153 Heering, P., Braun, N., Mullejans, R. et al. (2004). 868–878.
Cyclosporine A and chlorambucil in the treatment of 166 Fernandez-Fresnedo, G., Segarra, A., Gonzalez, E. et al.
idiopathic focal segmental glomerulosclerosis. Am. J. (2009). Rituximab treatment of adult patients with
Kidney Dis. 43 (1): 10–18. steroid-resistant focal segmental glomerulosclerosis.
154 Martinelli, R., Pereira, L.J., Silva, O.M. et al. (2004). Clin. J. Am. Soc. Nephrol. 4 (8): 1317–1323.
Cyclophosphamide in the treatment of focal segmental 167 Audard, V., Kamar, N., Sahali, D. et al. (2012).
glomerulosclerosis. Braz. J. Med. Biol. Res. 37 (9): Rituximab therapy prevents focal and segmental
1365–1372. glomerulosclerosis recurrence after a second renal
155 Goumenos, D.S., Tsagalis, G., El Nahas, A.M. et al. transplantation. Transpl. Int. 25 (5): e62–e66.
(2006). Immunosuppressive treatment of idiopathic 168 Ramachandran, R., Rajakumar, V., Duseja, R. et al.
focal segmental glomerulosclerosis: a five-year (2013). Successful treatment of adult-onset collapsing
follow-up study. Nephron Clin. Pract. 104 (2): c75–c82. focal segmental glomerulosclerosis with rituximab. Clin.
156 Laurin, L.P., Gasim, A.M., Poulton, C.J. et al. (2016). Kidney J. 6 (5): 500–502.
Treatment with glucocorticoids or calcineurin inhibitors 169 Marasa, M., Cravedi, P., Ruggiero, B., and Ruggenenti, P.
in primary FSGS. Clin. J. Am. Soc. Nephrol. 11 (3): (2014). Refractory focal segmental glomerulosclerosis in
386–394. the adult: complete and sustained remissions of two
157 Cattran, D.C., Appel, G.B., Hebert, L.A. et al. (1999). A episodes of nephrotic syndrome after a single dose of
randomized trial of cyclosporine in patients with rituximab. BMJ Case Rep. 2014.
steroid-resistant focal segmental glomerulosclerosis. 170 Cho, J.H., Lee, J.H., Park, G.Y. et al. (2014). Successful
North America Nephrotic Syndrome Study Group. treatment of recurrent focal segmental
Kidney Int. 56 (2226): 2220–2226. glomerulosclerosis with a low dose rituximab in a
158 Braun, N., Schmutzler, F., Lange, C. et al. (2008). kidney transplant recipient. Ren. Fail. 36 (4): 623–626.
Immunosuppressive treatment for focal segmental 171 Garrouste, C., Canaud, G., Buchler, M. et al. (2017).
glomerulosclerosis in adults. Cochrane Database Syst. Rituximab for recurrence of primary focal segmental
Rev. (3): CD003233. glomerulosclerosis after kidney transplantation: clinical
159 Duncan, N., Dhaygude, A., Owen, J. et al. (2004). outcomes. Transplantation 101 (3): 649–656.
Treatment of focal and segmental glomerulosclerosis in 172 Roccatello, D., Sciascia, S., Rossi, D. et al. (2017).
adults with tacrolimus monotherapy. Nephrol. Dial. High-dose rituximab ineffective for focal segmental
Transplant. 19 (12): 3062–3067. glomerulosclerosis: a long-term observation study. Am.
160 Kawakami, T., Gomez, I.G., Ren, S. et al. (2015). J. Nephrol. 46 (2): 108–113.
Deficient autophagy results in mitochondrial 173 Vincenti, F., Fervenza, F.C., Campbell, K.N. et al. (2017).
dysfunction and FSGS. J. Am. Soc. Nephrol. 26 (5): A phase 2, double-blind, placebo-controlled,
1040–1052. randomized study of fresolimumab in patients with
161 Tumlin, J.A., Miller, D., Near, M. et al. (2006). A steroid-resistant primary focal segmental
prospective, open-label trial of sirolimus in the glomerulosclerosis. Kidney Int. Rep. 2 (5): 800–810.
treatment of focal segmental glomerulosclerosis. Clin. J. 174 Trachtman, H., Vento, S., Herreshoff, E. et al. (2015).
Am. Soc. Nephrol. 1 (1): 109–116. Efficacy of galactose and adalimumab in patients with
resistant focal segmental glomerulosclerosis: report of segmental glomerulosclerosis: a systematic review and
the font clinical trial group. BMC Nephrol. 16: 111. meta-analysis of 77 case-reports and case-series. BMC
175 Cattran, D., Neogi, T., Sharma, R. et al. (2003). Serial Nephrol. 17 (1): 104.
estimates of serum permeability activity and clinical 182 Dantal, J., Bigot, E., Bogers, W. et al. (1994). Effect of
correlates in patients with native kidney focal segmental plasma protein adsorption on protein excretion in
glomerulosclerosis. J. Am. Soc. Nephrol. 14 (2): 448–453. kidney-transplant recipients with recurrent nephrotic
176 Alachkar, N., Carter-Monroe, N., and Reiser, J. (2014). syndrome. N. Engl. J. Med. 330 (1): 7–14.
Abatacept in B7-1-positive proteinuric kidney disease. 183 Esnault, V.L., Besnier, D., Testa, A. et al. (1999). Effect of
N. Engl. J. Med. 370 (13): 1263–1264. protein A immunoadsorption in nephrotic syndrome of
177 Garin, E.H., Reiser, J., Cara-Fuentes, G. et al. (2015). various etiologies. J. Am. Soc. Nephrol. 10 (9):
Case series: CTLA4-IgG1 therapy in minimal change 2014–2017.
disease and focal segmental glomerulosclerosis. Pediatr. 184 Nijland, M.J. (2003). Fetal exposure to corticosteroids:
Nephrol. 30 (3): 469–477. how low can we go? J. Physiol. 549 (Pt 1): 1.
178 Shalhoub, R.J. (1974). Pathogenesis of lipoid nephrosis: 185 Bar Oz, B., Hackman, R., Einarson, T., and Koren, G.
a disorder of T-cell function. Lancet 2 (7880): 556–560. (2001). Pregnancy outcome after cyclosporine therapy
179 Meyrier, A. (2005). Mechanisms of disease: focal during pregnancy: a meta-analysis. Transplantation 71
segmental glomerulosclerosis. Nat. Clin. Pract. Nephrol. (8): 1051–1055.
1 (1): 44–54. 186 Armenti, V.T. (2004). Immunosuppression and
180 Bosch, T. and Wendler, T. (2001). Extracorporeal plasma teratology: evolving guidelines. J. Am. Soc. Nephrol. 15
treatment in primary and recurrent focal segmental (10): 2759–2760.
glomerular sclerosis: a review. Ther. Apher. 5 (3): 187 Armenti, V.T., Radomski, J.S., Moritz, M.J. et al. (2003).
155–160. Report from the National Transplantation Pregnancy
181 Kashgary, A., Sontrop, J.M., Li, L. et al. (2016). The role Registry (NTPR): outcomes of pregnancy after
of plasma exchange in treating post-transplant focal transplantation. Clin. Transpl.: 131–141.
235
16
Membranous Nephropathy
Katie Trinh1,2 and Bhadran Bose1,2
1
Department of Nephrology, Nepean Hospital, Kingswood, NSW, Australia
2
University of Sydney, Sydney, NSW, Australia
I ntroduction for example proteinuria of >8 g/24 h. The remaining two-

thirds of patients who do not undergo spontaneous remis-
Idiopathic membranous nephropathy is a common sion generally divide equally into either those with
immune-mediated glomerular disease and remains the persistent proteinuria who will maintain renal function
leading cause of nephrotic syndrome (NS) in Caucasian long term or those patients who will progress to kidney fail-
adults. Secondary forms may account for up to one-third of ure. In Caucasian patients with nephrotic syndrome, 10-
cases and are associated with autoimmune diseases (e.g. year kidney survival of <70% has been reported [2, 5–8].
systemic lupus erythematosus), infections (e.g. hepatitis B Because of its high incidence rate, membranous
and C), medications (e.g. nonsteroidal anti-inflammatory nephropathy remains the second or third leading cause of
drugs [NSAIDs], d-penicillamine, gold), and neoplasias end-stage kidney disease (ESKD) among the primary glo-
(e.g. carcinomas) [1]. The association with malignancy merulonephritis types [9]. Even patients who do not pro-
increases with age, reaching up to 20% in patients over the gress but remain nephrotic are at an increased risk for
age of 60. Because idiopathic and secondary forms have life-threatening thromboembolic and cardiovascular
similar clinical presentations, the designation of idiopathic events. A rapid change in either the degree of proteinuria
is made only after ruling out secondary causes by a careful or the rate of loss of renal function, especially in a previ-
history, physical examination, and laboratory evaluation. It ously stable patient, should raise the possibility of a
is crucial to rule out secondary causes of membranous superimposed condition, for example interstitial nephri-
nephropathy because the management is directed toward tis, crescentric glomerulonephritis, sepsis, or renal vein
removing or correcting the underlying cause. thrombosis.
N
atural History C
linical Manifestations
Membranous nephropathy is a chronic disease, with spon- Membranous nephropathy affects patients of all ages and
taneous remission and relapses. There is great variability in races, but it is more common in men than women by a 2 : 1
the rate of disease progression, and the natural course is dif- (male:female) to 3 : 1 ratio. Idiopathic membranous
ficult to assess in part due to heterogeneous criteria for indi- nephropathy has a peak incidence during the fourth and
cation of biopsy and genetic and geographic disease fifth decades of life, and is relatively uncommon in patients
biology [2–4]. Spontaneous remissions occur in up to 30% of under 20 years. At presentation, 60–70% of patients will
cases, usually in the first 2 years after presentation, but may have nephrotic syndrome. The presence of microscopic
occur any time. The proportion of patients going into spon- hematuria is common (30–40%), but macroscopic hematu-
taneous remission is much lower when patients are selected ria and red cell casts are rare and suggest a different histo-
who have higher grades of proteinuria at presentation, pathology. At presentation, the great majority of patients
236 Membranous Nephropathy
are normotensive, with only 10–20% having hypertension P

redicting Factors
and only a small fraction of patients (<10%) having signifi-
cant renal insufficiency. Risk of thromboembolism is very Evaluating the prognosis is critical in making decisions
high in the first 6 months [10]. The incidence of venous regarding treatment [26–28]. Individual factors, such as
thromboembolism is 9.85% and arterial thrombembolism advanced age, male gender, and selected biopsy findings
is 5.52%. The risk is higher if serum albumin <2.8 g/dl [11]. (e.g. degree of interstitial fibrosis, glomerulosclerosis, vas-
cular damage, or glomeruli with focal segmental glomeru-
losclerosis), are predictors of prognosis and/or response to
P
athogenesis immunosuppressive therapy [29]. In addition, the degree
of proteinuria may also predict those who are most likely to
Phospholipase A2 receptor progress. Pei et al. observed a 47% risk for progression in
In the past decade, the understanding of the pathogenesis patients with proteinuria of >4 g/24 h for longer than
of idiopathic membranous nephropathy has substantially 18 months and a 66% risk in patients with proteinuria of
improved. In 2009, phospholipase A2 receptor (PLA2R) >8 g/24 h for more than 6 months [30]. The degree of renal
was identified as the major antigen responsible for autoan- impairment at presentation has also been found to corre-
tibody binding in idiopathic membranous nephropa- late with long-term renal survival.
thy [12]. PLA2R is a transmembrane receptor that is highly PLA2R antibodies appear to correlate with disease activ-
expressed in glomerular podocytes and was initially identi- ity, response to therapy, and prognosis, which may allow
fied in 70% patients with idiopathic membranous nephrop- them to be used as biomarkers for assessing treatment effi-
athy [12]. Subsequent studies from various cohorts have cacy [21, 31–33]. Additionally, higher antibody levels are
shown that PLA2R antibodies are positive in 50–80% of linked to a higher risk of declining renal function, suggest-
patients with idiopathic membranous nephropathy [13– ing that these affected individuals may benefit from earlier
18]. PLA2R antibodies are uncommon in patients with initiation of immunosuppression [21]. Conversely, favora-
membranous nephropathy associated with malignancies. ble outcomes have been shown in patients who do not have
There is genetic polymorphism in PLA2R-associated mem- PLA2R antibodies. Spontaneous remission rates are higher
branous nephropathy, which is linked with HLA-DQA1 in patients with low (negative or lowest tertile) antibody
risk alleles [19, 20]. Furthermore, the presence of HLA levels (79%, HR 2.72, 95% CI 1.22–6.09, P = 0.02) compared
DQA1*05 : 01 and DQB1*02 : 01 alleles is associated with to patients with higher antibody levels [33]. In another
higher PLA2R antibody levels [21]. study of 28 seronegative patients, 91% had spontaneous
remission (7 out of 10 had complete remission [CR]) [34].
Thrombospondin Type-1 Domain-Containing 7A

R
esponse Measurements
Thrombospondin type-1 domain-containing 7A (THSD7A),
like PLA2R, is a protein highly expressed in podocytes and The goal of treatment is to achieve either CR or partial
was identified in European and North American patients remission (PR). Definition of CR or PR is reported differ-
with anti-PLA2R-negative idiopathic membranous nephrop- ently by different studies. However, the definition reported
athy but not in healthy controls or patients with other glo- by Kidney Disease Improving Global Outcome (KDIGO)
merular diseases [22]. It occurs in 2.5–5% of all patients with seems to be more widely agreed upon [35]. KDIGO defines
idiopathic nephropathy, which corresponds to 8–14% of CR as urinary protein excretion <0.3 g/day (urinary protein
patients who are seronegative for anti-PLA2R antibodies. A to creatinine ratio < 300 mg/g or < 30 mg/mmol) accompa-
recent meta-analysis of 10 studies involving 4121 patients nied by a normal serum albumin concentration and a nor-
showed prevalence of THSD7A was low at 3% (95% CI 2–4%) mal serum creatinine. PR is defined as urinary protein
of all patients with idiopathic membranous nephropathy, excretion <3.5 g/day (urinary protein to creatinine ratio
which corresponded with 10% (95% CI 6–15%) of anti-PLA2R < 3500 mg/g or < 350 mg/mmol) and a 50% or greater
antibody negative patients [23]. However, this meta-analysis reduction from peak values accompanied by an improve-
was limited by a small number of studies and sample size. ment or normalization of the serum albumin concentra-
Cancer may be more common in patients with THSD7A tion and stable serum creatinine.
antibodies [24]. Further studies to elucidate the role of About 30% of membranous nephropathy cases will relapse
THSD7A as a marker of prognosis and response to therapy subsequent to CR [36]. The great majority who do, however,
are required. Antibodies against PLA2R and THSD7A coex- will relapse to subnephrotic-range proteinuria and will have
ist but only occasionally [25]. stable long-term function. PR has been also recognized as a
Treatmen 237
specific outcome. A review of 350 nephrotic patients with Blood Pressure

membranous nephropathy found that the 10-year renal sur- In patients with proteinuric renal disease, including mem-
vival was 100% in the CR group, 90% in the PR group, and branous nephropathy, there is high certainty evidence that
45% in the no-remission group [37]. Patients with CR or PR the target for blood pressure is 125/75 mmHg [42].
have a similar rate of decline: −1.5 ml/min/year in the CR Angiotensin-converting enzyme inhibitor (ACEi) and/or
group and −2 ml/min/year in the PR group. In contrast, the angiotensin receptor blockers (ARBs) are effective antihy-
no-remission group lost glomerular filtration rate (GFR) at a pertensive agents that can reduce proteinuria and slow
rate of −10 ml/min/year. Thus, both CR and PR appear to be progression of renal disease in both diabetic and nondia-
excellent predictors of long-term renal survival. betic chronic nephropathy patients, and for these reasons
Anti-PLA2R antibody levels have been shown to decline in they are the preferred agents to treat hypertension in pro-
response to immunosuppressive therapy. In an observa- teinuric renal diseases. In the Modification of Diet in Renal
tional study, high-risk patients receiving oral cyclophospha- Disease study, patients with proteinuria of >1 g/day had a
mide or mycophenolate mofetil in combination with significantly better outcome if their blood pressure was
corticosteroids had a rapid decline in anti-PLA2R antibodies reduced to 125/75 mmHg [43].
following treatment [38]. Baseline antibody levels did not Patients need to be instructed to follow a low-salt diet
predict response to treatment but patients who were because high salt intake (e.g. 200 mg Na or 4.6 g sodium/
antibody-negative 5 years following treatment were more day) can significantly impair the beneficial effects of angio-
likely remain in persistent remission [38]. In another pro- tensin II blockade [44].
spective study of 101 patients who received immunosup-
pression, antibody levels fell by 69–81% (depending on the Lipid Lowering
assay) and proteinuria fell by 38.8% within 3 months [39]. Statins may have a synergistic antiproteinuric effect when
The decline in proteinuria continued up until 12 months, combined with ACEi, but this effect is small and mainly
whereas antibody levels remained within low range. observed in patients with proteinuria of <3 g/24 h. When
A prospective trial showed that patients treated with ritux- used in combination with high-dose cyclosporine (CsA),
imab who showed an immunologic response were more statins may increase the risk of rhabdomyolysis [45].
likely to achieve remission [40]. The decline in anti-PLA2R
antibodies preceded the corresponding reduction in protein- Thromboembolism Prophylaxis
uria, which may allow response to rituximab to be predicted Prophylactic anticoagulation has been shown in retrospec-
and subsequent dosing to be determined. In another pro- tive reviews to be beneficial in reducing fatal thromboem-
spective trial of patients with persistent nephrotic syndrome bolic episodes in nephrotic patients with membranous
treated with rituximab, lower antibody titers at baseline and nephropathy without a concomitant increase in the risk of
full antibody depletion at 6 months following rituximab bleeding [46]. In general, membranous nephropathy
treatment strongly predicted remission (HR 7.90, 95% CI patients who are severely nephrotic (proteinuria of >10 g/
2.54–24.60, P < 0.001) [41]. Similarly, in each case of CR day and serum albumin of <2.5 g/dl) are candidates for
(n = 25), antibody depletion preceded clinical remission. On anticoagulation.
average, a 50% reduction in antibody titer preceded a 50%
reduction in proteinuria by 10 months.
Immunosuppressive Therapy
Several treatment strategies, including a variety of immu-
T
reatment
nosuppressive agents, have been shown to be at least par-
tially successful in reducing proteinuria in membranous
We can rationally assign patients to nonimmunosuppres-
nephropathy [47]. The available evidence is presented
sive therapy or to immunosuppressive therapy according
according to the risk group (e.g. low, medium, and high)
to their risk for renal disease progression. Table 16.1 sum-
that the patients in the studies most closely represent.
marizes GRADE recommendations for the treatment of
Table 16.2 summarizes the different risk groups.
membranous nephropathy.
Treatment of Low-risk Patients

Nonimmunosuppressive Therapy
Evidence to support this approach comes from published
Nonimmunosuppressive therapy is based on controlling validation studies and from data on the clinical relevance
edema, dietary protein intake, blood pressure, and hyper- of PR [37, 48]. Treatment should be nonimmunosuppressive
lipidemia. Interventions are indicated based on evidence therapy only, given the excellent prognosis of this group of
available in adults with proteinuria renal disease. patients when untreated.
Table 16.1 GRADE recommendations for treatment of membranous nephropathy.
Blood pressure lowering We recommend using ACEi or ARB to achieve a target blood pressure of Moderate– Strong
125/75 mmHg. ACEi or ARB also help to reduce proteinuria and slow high
the progression of kidney disease.
Low-salt diet We suggest a low-salt diet because high salt intake (e.g. 200 mg Na or Low Weak
4.6 g sodium/day) can significantly impair the beneficial effects of
angiotensin II blockade
Lipid-lowering therapy Statins combined with ACEi have synergistic effect in reducing Low Weak
proteinuria
Thromboembolism We suggest prophylactic anticoagulation in patients with proteinuria Low Weak
prophylaxis 10 g/day and serum albumin below 2.5 g/dl as they are at high risk of
fatal thromboembolism
Nonimmunosuppressive We suggest initial treatment with nonimmunosuppressive therapy Moderate Strong
therapy comprising of ACEi or ARB, low-salt diet and statins for 6 months before
starting immunosuppressive therapy. However, patients with high-risk
of progression will need to start immunosuppressive therapy earlier.
Cytotoxics and We recommend a 6-month course of corticosteroids with oral cytotoxic High Strong
corticosteroids agent as first-line therapy
Calcineurin inhibitor with We recommend calcineurin inhibitor as first-line therapy or for patients High Strong
or without corticosteroids who cannot tolerate cytotoxic agents. There is high risk of relapse once
calcineurin inhibitors are ceased hence might need long-term treatment
with calcineurin inhibitor.
Mycophenolate mofetil with We suggest mycophenolate mofetil with corticosteroids in patients Low Weak
or without corticosteroids refractory to other agents. There is high risk of relapse.
Rituximab We suggest rituximab as first-line therapy or for patients with refractory Low Weak
disease
Table 16.2 Risk assessment for progressive decline in kidney function.
Low risk Medium risk High risk
Patients in the low-risk group are Patients in the medium-risk group are Patients in the high-risk group are
categorized by a <5% risk for defined by normal renal function and characterized by progressive loss of renal
progression over 5 years of observation persistent proteinuria between 4 and 8 g/24 h function and/or by persistent high-grade
They are defined by normal renal over 6 months of observation despite the proteinuria of 8 g/24 h during the 6 months
function and proteinuria of 4 g/24 h institution of maximum conservative therapy of observation
over a 6-month observation period
Treatment of Medium-risk Patients effective in the treatment of patients with idiopathic

Patients in this group should be treated with nonimmuno- membranous nephropathy and preserved renal
suppressive treatment for at least 6 months before being function, with benefits maintained well beyond the
considered for immunosuppressive therapy. 1-year treatment period, although relapse rates
approached 35% at 2 years. The long-term adverse
Corticosteroids Corticosteroid therapy alone may provide
effects of these cytotoxic agents, in particular, effects on
little or no benefit for induction of remission or
fertility and also malignancy are the major drawbacks
preservation of kidney function even after the data are
to the universal application of this form of therapy. The
adjusted to include only patients with proteinuria at entry
risk of malignancy is not increased for patients treated
of >3.5 g/24 h (Table 16.3).
with cumulative cyclophosphamide doses of 36 g but
Cytotoxic Agents Combined with increases significantly in patients with cumulative
Corticosteroids Cyclophosphamide and chlorambucil cyclophosphamide doses of 36 g [53]. These studies are
in combination with corticosteroids appear to be summarized in Table 16.4 [8, 54–69].
Table 16.3 Corticosteroid treatment in idiopathic membranous nephropathy.
Author Year Type of study Risk group n Treatment regimens Follow-up (months) Outcomes/comments
Collaborative Study of 1979 RCT Medium 72 Prednisone 100–150 mg 23 10 of 38 controls, but only 1 of 34
the Adult Idiopathic PO on alternate given prednisone had SCr > 5 mg/
Nephrotic Sydrome days × 8–12 weeks vs. dl or died
(CSAINS) [49] placebo Rapid decline of renal function in
controls considered unexpected
Cattran [50] 1989 RCT Medium 158 Prednisone 125–150 mg 48 Proportion of patients with CR
PO on alternate similar in the two groups
days × 6 months vs. No differences in annual change
placebo in CrCl between the two groups
No sustained difference in
proteinuria
Cameron [51] 1990 RCT High 107 Prednisolone 45 mg/m2 52 No difference in remission rates
on alternate for NS in either short (6 and
days × 8 weeks vs. 12 months) or long (48 months)
placebo term
No differences in rates of renal
function decline
Short [52] 1987 Prospective High 15 MTP 1 g IV × 5 days 32 Serum creatinine fell by a mean of
nonrandomized followed by 46%
study prednisolone 100 mg, In 10 patients the beneficial effect
75 mg 50 mg, then 25 mg was sustained, but in three it had
on alternate days for reversed by 6 months
4 weeks, then dose In the other two patients the
decreasing by a further progressive decline of renal
5 mg each month function was not influenced
CR, complete remission; CrCl, creatinine clearance; IV, intravenous; MTP, methylprednisolone; NS, nephrotic syndrome; PO, per oral; RCT, randomized controlled trial; SCr, serum creatinine.
c16.indd 239 09-12-2022 15:23:39

Table 16.4 Cytotoxic treatment in idiopathic membranous nephropathy.
Mean Follow-up
Author Year Type of study Risk group n Treatment regimens (months) Outcomes/comments
Ponticelli [54] 1984 RCT Medium 67 MTP 1 g IV × 3 days followed by MTP 0.4 mg/kg 31 At the end of follow-up 23 of 32 treated
PO × 27 days, on months 1, 3, and 5 and CHL patients (72%) were in CR or PR, as
(0.2 mg/kg/day) on months 2, 4, and 6 vs. compared with 9 out of 30 controls (30%).
symptomatic therapy
Ponticelli [55] 1992 RCT Medium 92 MTP 1 g IV × 3 days followed by MTP 0.4 mg/kg 48 At 3 years, 66% remission in MTP/CHL
PO × 27 days on months 1, 3, and 5 and CHL group vs. 40% in MTP group. At 4 years, the
(0.2 mg/kg/day) on months 2, 4, and 6 vs. MTP difference was no longer statistically
alone significant.
Ponticelli [56] 1995 RCT Medium 67 MTP 1 g IV × 3 days followed by prednisone 120 92% probability of renal survival in treated
0.5 mg/kg/day × 27 days on months 1, 3, and 5 and compared with 60% in the control group. 8%
CHL (0.2 mg/kg/day) on months 2, 4, and 6 vs. of treated patients vs. 40% of the untreated
symptomatic therapy group reached ESRD.
Ponticelli [57] 1998 RCT Medium 87 MTP 1 g IV × 3 days followed by MTP 0.4 mg/kg/ 36 CR/PR in 82% of patients on MTP/CHL vs.
day PO × 27 days on months 1, 3, and 5 93% on MTP/CYC (P = NS). Side effects
alternating with either CHL 0.2 mg/kg/day or lower in the MTP/CYC group.
CYC 2.5 mg/kg/day on months 2, 4, and 6
Jha [58] 2007 RCT Medium 93 MTP 1 g IV × 3 days followed by prednisolone 120 CR/PR in 72% of patients on MTP/CYC vs.
0.5 mg/kg/day PO × 27 day on months 1, 3, and 5 35% in control group
and CYC 2 mg/kg/day on months 2, 4, and 6 vs.
conservative therapy
Murphy [59] 1992 RCT Medium 40 CYC 1.5 mg/kg × 6 months + DIP/W24 × 2 years Treatment group had less proteinuria. No
vs. symptomatic therapy differences in renal function at 2 years
Donadio [60] 1974 RCT Medium 22 CYC 1.5–2.5 mg/kg PO × 1 year vs. symptomatic 12 No benefit of CYC on proteinuria, renal
therapy function, or histology
Falk [61] 1992 RCT High 26 Prednisone 2 mg/kg on alternate days × 8 weeks, 29 No impact of CYC on renal function, level
tapered over 4 weeks, vs. CYC 0.5–1 g/m2 IV of proteinuria, or progression to ESRD
monthly × 6 months + MTP (7 mg/kg) × 3 followed
by prednisone (2 mg/kg/alternate days) for
8 weeks, then tapered over next 4 weeks
West [62] 1987 Case-control High 26 CYC 2 mg/kg × 20 ± 4 months ± prednisone vs. 49 CYC associated with an increased rate of
study prednisone or symptomatic therapy remission of NS and better preservation of
renal function. Adverse effects were
significant.
Jindal [63] 1992 Case-control High 9 CYC (1–2 mg/kg/day) for a mean of 23 ± 4 months 64–83 CR in 4/9 and PR in 5/9 in CYC group. One
study (8–54) ± prednisone (6/9 patients; dose 33 ± 9 mg/ patient in the CYC group and 10 patients in
day) compared to 17 controls receiving control group reached ESRD. Four relapses
prednisone, maximum dose 50–125 mg alternate in 3 treated patients, and 3 of 4 responded to
days in 9/17 and 40–80 mg in 6/17, for a mean of repeat therapy. Of the 7 controls who did
20 ± 4 months not reach ESRD, only 2 had persistent NS.
c16.indd 240 09-12-2022 15:23:39

Torres [64] 2002 Case-control High 19 Prednisone (1 mg/kg/day month 1, 0.5 mg/kg/day 48 At the end of follow-up, 65% of patients in
study month 2, 0.5 mg/kg/day months 3–6) together the control group had reached ESRD, 10%
with CHL 0.15 mg/kg/day for the first 14 weeks had advanced renal failure, and 25% had
vs. historic control group on symptomatic therapy died. Majority of treated patients showed
stabilization or improvement of renal
function.
Bruns [65] 1991 Prospective High 11 CYC, 100 mg/day in all patients + prednisone 24–25 Serum creatinine decreased in 10 patients
nonrandomized (60–100 mg alternate days) in 10/11 patients for on combined therapy by 6 months and
1 year remained stable in 7/8 followed long-term.
Proteinuria decreased from 11.9 to 2.3 g/day.
Similar course observed in the 1 patient on
CYC alone.
Warwick [66] 1994 Prospective High 21 Prednisolone 125 mg alternate days on months 1, 39 Renal function stable or improved in 11/21
nonrandomized 3, and 5 and CHL 10 mg/day on months 2, 4, and patients (52%). Of these 11, PR in 4 and CR
6. Later patients received MTP 1 g iv × 3 at the in 2. Serum creatinine >5.7 mg/dl or ESRD
start first cycle of treatment. Four patients were in 6 patients (29%). 3 patients died. Side
retreated. effects + significant complications related to
therapy in >50% of patients.
Branten [67] 1998 Prospective High 32 MTP 1 g IV × 3 days, followed by prednisone 26–38 Renal function improved in both groups but
nonrandomized (0.5 mg/kg/day months 1, 3, and 5) and CHL the improvement was short-lived in the
(0.15 mg/kg/day months 2, 4, and 6), or CYC CHL group. Remissions of proteinuria more
(1.5–2 mg/kg/day for 1 year) + steroids in a frequent after CYC treatment (15/17 vs.
comparable dose 5/15). CHL associated with more side
effects.
Du 2004 Prospective High 65 MTP 1 g IV × 3 days at months 1, 3, and 5, and 51 Renal function improved or stabilized in all
Buf-Vereijken [68] nonrandomized prednisone 0.5 mg/kg/48 h for 6 months + CYC patients. At the end of follow-up, CR in 16,
(1.5–2 mg/kg/day) × 12 months PR in 31, 4 progressed to ESRD, and 5
patients had died. Overall survival 86% after
5 years.
Mathieson [69] 1988 Prospective High 8 MTP 1 g IV × 3 days, 0.5 mg/kg/day PO × 27 days, 8 Renal function improved in 6, stabilized in
nonrandomized then CHL 0.15–0.2 mg/kg × 28 days × 3 cycles 1, and deteriorated in 1 patient. Proteinuria
fell from 15 to 2 g/day. Side effects were
severe.
CHL, chlorambucil; CR, complete remission; CYC, cyclophosphamide; DIP/W, dipyridamole/warfarin; IV, intravenous; MTP, methylprednisolone; PO, per oral; PR, partial remission; RCT,
randomized controlled trial.
c16.indd 241 09-12-2022 15:23:39

Cyclosporine A CsA can induce a remission (CR or PR) of evidence for mycophenolate mofetil is based on
NS in 50–60% of patients. Prolonged low-dose CsA nonrandomized studies (Table 16.6) [79–81]. Based on low
(~1.5 mg/kg/day) could be considered for long-term certainty evidence from uncontrolled studies,
maintenance of patients who achieve CR or PR, especially mycophenolate mofetil has uncertain benefits in high-risk
in patients at high risk for relapse [70]. It is important to membranous nephropathy. Based on a study including
emphasize that although reduction of proteinuria usually historical controls treated with cyclophosphamide, relapses
occurs within a few weeks, the majority of CR occurred may occur more frequently with mycophenolate mofetil
after more than 6 months of treatment. On the other hand, therapy [81].
if after 3–4 months of CsA therapy at adequate doses
proteinuria is not significantly reduced, it is unlikely that Rituximab Rituximab is a chimeric monoclonal antibody
the therapy will be effective. These studies are summarized against CD20+ B cells and as membranous nephropathy
in Table 16.5 [72–76]. is an autoimmune disease, rituximab has been studied in
the treatment of patients with membranous nephropathy.
Tacrolimus Similar to patients treated with CsA, Prospective uncontrolled studies have been performed
maintenance of remission may require prolonged use of evaluating the efficacy of rituximab in idiopathic
low-dose tacrolimus (TAC). In a single randomized membranous nephropathy and are summarized in
controlled trial in 25 patients with normal kidney function Table 16.7. These studies have led to the hypothesis that
and nephrotic range proteinuria, tacrolimus monotherapy rituximab may modify disease progression in membranous
incurred complete remission in 94% of patients prescribed nephropathy.
tacrolimus compared to 35% in the control group [77]. A Two randomized controlled studies provide evidence for
small number of patients (six patients in the control group rituximab in membranous nephropathy. In the small (75
and one in the TAC group) reached the secondary end participants) Evaluate Rituximab Treatment for Idiopathic
point of a 50% increase in serum creatinine, resulting in Membranous Nephropathy (GEMRITUX) study, rituximab
low certainty evidence for this outcome. (375 mg/m2 on days 1 and 8) may have induced a higher
rate of disease remission during extended follow-up (up to
Treatment of High-risk Patients 18 months) compared to standard therapy alone [82].
Corticosteroids The Membranous Nephropathy Trial Of Rituximab
Randomized, controlled trial evidence is available in a (MENTOR) study (NCT01180036) is a multicenter rand-
small number of patients with high-risk disease. Based on omized controlled trial in 126 participants comparing the
low certainty evidence, corticosteroid therapy alone may efficacy and safety of rituximab to CsA in medium- to
offer little or no benefit for preserving kidney function or high-risk patients with idiopathic membranous nephrop-
lowering proteinuria (Table 16.3) [49–52]. athy [91]. Patients with proteinuria 5 g/day following a
minimum of 3 months conservative nonimmunosuppres-
Cytotoxic Agents Combined with Corticosteroids Prednisolone sive therapy were randomized to receive IV rituximab
combined with chlorambucil may retard progression of (days 1 and 15, then repeated at 6 months) or oral CsA
kidney failure compared to steroid therapy and supportive (3.5 mg/kg/day for 12 months). Based on results in a con-
treatment alone [78]. The addition of CsA to steroid therapy is ference proceeding, participants who received rituximab
likely to have similar clinical effects compared to chlorambucil, more frequently achieved CR or PR (odds ratio 6.0, 95%
although may incur fewer serious adverse events. CI 2.7–13.2).
A summary of the studies conducted in this group of Table 16.7 summarizes the studies using rituximab in
patients is presented in Table 16.4 [54–69]. membranous nephropathy [41, 82–92]. Together these
studies suggest rituximab could be considered as first-line
Cyclosporine There has been only one controlled trial with treatment for membranous nephropathy, but we need
17 participants evaluating CsA in patients with high-grade larger trials with hard endpoints such as ESKD to change
proteinuria and progressive renal failure [72]. CsA was our current practice. Most importantly, so far there is no
associated with significant reduction in proteinuria and published study comparing rituximab with the current
prevented loss (slope) of renal function [70–75]. A first-line therapy, i.e. combination of cytotoxics and
summary of studies is presented in Table 16.5. corticosteroids.
Mycophenolate Mofetil There has been a paucity of studies Eculizumab Eculizumab is a humanized anti-C5
using mycophenolate mofetil for membranous nephropathy monoclonal antibody designed to prevent the cleavage of
leading to low or very low certainty evidence. Current C5 into its proinflammatory by-products. In a randomized
Table 16.5 Cyclosporin A treatment in idiopathic membranous nephropathy.
Follow-up
Cattran [71] 2001 RCT Medium 51 Prednisone (0.15 mg/kg/day) plus CsA 19.5 75% of patients in CsA group vs. 22% in control
(3–4 mg/kg/day) vs. placebo plus group achieved PR or CR. Relapse rate 43% in
prednisone × 26 weeks. CsA group vs. 40% in placebo by week 52.
Cattran [72] 1995 RCT High 17 CsA 3.5 mg/kg/day × 12 months vs. placebo 21 CsA associated with slower rate of decline in
renal function. Sustained remission of
proteinuria in 6/8 CsA patients.
Alexopoulos [70] 2006 RCT Medium 51 CsA 2–3 mg/kg/day and prednisone vs. CsA 26 After 12 months of treatment, 26 patients in the
(same dose) alone for 12 months. combination group and 17 patients in the CsA
Responders were placed on long-term alone group had a CR or PR of proteinuria. Daily
low-dose CsA 1–1.5 mg/kg/day and CsA dose was higher in nonrelapsers in both
prednisone vs. CsA alone. groups while relapsers in both groups had lower
CsA trough levels
Ambalavanan [73] 1996 Prospective Medium 41 CsA 4–5 mg/kg/day for 3–6 months. Lupus 18 CsA lowered median proteinuria by 56% from
nonrandomized serology (+) in 12 patients. Retreatment in 7.3 to 3.2 g/24 h. In 6 patients with declining
20 of 31 patients who relapsed after GFR during prolonged CsA therapy, a repeat
stopping CsA. biopsy showed more prominent immune
deposits and thicker GBM.
Rostoker [74] 1993 Prospective High 15 CsA 4–5 mg/kg × 12–30 months prednisone 40 CR or PR in 11/15 (73%) patients. Relapse in 3/9
nonrandomized 1–2 mg/kg/day × 2 months on CsA withdrawal, but the relapse remain
sensitive to CsA. All patients had received
corticosteroids (1 mg/kg/day × 2 months) prior to
enrollment.
DeSanto [75] 1987 Case series Medium- 5 CsA 7 mg/kg/day for 6 months plus 8 All had failed prior cytotoxic therapy. Prompt
Low MTP 1–0.3 mg/kg/day for 1 month, reduced remission of proteinuria in 4/5 patients. No
to 0.3–0.15 mg/kg on the second month, renal failure.
and then down to 0.15 mg/kg for 4 months
CR, complete remission; CsA, cyclosporin A; GBM, glomerular basement membrane; MTP, methylprednisolone; PR, partial remission; RCT, randomized controlled trial.
c16.indd 243 09-12-2022 15:23:39

Table 16.6 Mycophonolate mofetil in medium-to high-risk idiopathic membranous nephropathy.
Author Year Type of study Risk group n Treatment regimens Median Follow-up Outcomes
Miller [79] 2000 Retrospective study Medium–high 16 Mycophenolate mofetil 500–2000 mg/ 8 months 6 patients had a 50% reduction in
day for 8 months their proteinuria, 2 had a minor
reduction in proteinuria, 4 had no
change, 3 withdrew because of
significant adverse effects, and 1
stopped treatment. There were no
significant changes in mean serum
creatinine, or serum albumin levels.
Choi [80] 2000 Retrospective study Medium–high 15 Mycophenolate mofetil 500–1000 mg 12 months There was 61% reduction of proteinuria
twice daily for a mean of 12 months (7.8 to 2.3 g/24 h; P = 0.001), with 8
combined with steroids patients having PR and 2 patients CR.
Renal function improved in 3 of 6
patients with kidney failure.
Branten [81] 2007 Clinical trial with Medium–high 32 Mycophenolate mofetil 1000 mg twice 12 mo Cumulative incidences of remission of
historic controls. a day for 12 months (intervention). proteinuria at 12 months were 66% in
Cyclophosphamide, 1.5 mg/kg/day for the mycophenolate mofetil group vs.
12 months (historical control). Both 72% in the cyclophosphamide group
groups also received methylpredisone (P = 0.3). There were more relapses in
IV 1 g three times at months 1, 3, and mycophenolate mofetil-treated group.
5, followed by oral predisone at 0.5 mg/
kg every other day for 6 months, with
subsequent tapering.
c16.indd 244 09-12-2022 15:23:39

Table 16.7 Rituximab treatment in idiopathic membranous nephropathy.
Median
Follow-up
Dahan [82] 2017 RCT Medium– 75 RTX 375 mg/m2 on day 1 and 8 + NIAT vs. 17.0 At 6 months, 13 of 37 (35.1%) treated with
high NIAT alone RTX + NIAT achieved remission vs. 8 of 38
(21.1%) controls (P = 0.21). Significantly higher
rates of PLA2R antibody depletion at 3 and
6 months in RTX + NIAT group.
Fiorentino [83] 2016 Prospective Medium– 38 RTX 375 mg/m2 monthly × 6 15 29 of 38 (76.3%) achieved remission: 15 (39.5%)
nonrandomized high CR and 14 (36.8%) PR. Proteinuria significantly
reduced. Renal function stable. No significant
adverse events.
Bagchi [84] 2018 Retrospective High 21 RTX 500 mg × 2 doses 7–10 days apart ± 3rd 13 13 of 21 (61.9%) achieved remission: 4 (19.05)
dose after 4–6 weeks if CD19 not depleted CR and 9 (42.9%) PR. One patient relapsed
after achieving PR. Renal survival was
significantly better in responders (P = 0.0037).
Moroni [85] 2017 Prospective Medium–- 34 RTX 375 mg/m2 × 1 dose (n = 18) or × 2 23.9 (mean) At 12 months, 5 (14.7%) CR, 10 (29.4%) PR, and
nonrandomized high 2 weeks apart (n = 16) 19 (55.8%) no response. Outcome similar for
one vs. two doses.
Waldman [86] 2016 Prospective High 13 CsA 3 mg/kg/day for 6 months then tapered 41 (mean) By 6 months 85% achieved remission
nonrandomized by 50 mg/day every 3 weeks plus RTX (CR + PR). By 12 months, 54% achieved CR. 2
1000 mg day 1 and 15, then after 6 months relapsed by 24 months. Treatment well
when CD19+ B cell count 5 cells/μl tolerated.
Ruggenenti [41] 2015 Prospective High 132 RTX 375 mg/m2 weekly × 4 30.8 84 of 132 (63.6%) achieved remission
nonrandomized (CR + PR), 43 (32.6%) achieved CR. Anti-
PLA2R antibody depletion preceded remission.
Ruggenenti [87] 2003 Prospective Medium 8 RTX 375 mg/m2 weekly × 4 Not specified Significant reduction in proteinuria. Renal
nonrandomized function stabilized. 2 achieved CR and 3
achieved PR.
Ruggenenti [88] 2012 Prospective High 100 RTX 375 mg/m2 weekly × 4 29 (median) At end of follow-up, 65% achieved CR or PR.
nonrandomized Median time to remission 7.1 months. Renal
function improved in those who achieved CR.
(Continued)
c16.indd 245 09-12-2022 15:23:39

Median
Follow-up
Fervenza [89] 2010 Prospective Medium 20 RTX 375 mg/m2 weekly × 4, repeated at Not specified Proteinuria reduced from 11.9 g/24 h to 4.2 and
nonrandomized 6 months 2.0 g/24 h at 12 and 24 months, respectively. At
24 months, 4 of 18 achieved CR, 12 of 18
achieved PR, and 1 relapsed. Remission rates
higher than fortnightly dosing.
Fervenza [90] 2008 Prospective High 14 RTX 1000 mg on days 1 and 15, repeated at 12 At 6 months, 4 achieved PR. At 12 months, 2
nonrandomized 6 months if proteinuria >3 g/24 h and CD9+ B achieved CR and 6 PR.
cell >15 cells/μl
Fervenza [91] 2015 RCT High 130 RTX 1000 mg on days 1 and 15, repeated at 24 At 24 months, CR or PR in RTX arm was 62.5%
(results from 6 months vs. CsA 3.5–5 mg/kg/day for vs. 20.6% in the CsA arm. Treatment failure
abstract) 6 months higher in CsA group compared to RTX group
(79.4% vs. 37.5%).
Rojas-Rivera [92] 2015 RCT High MTP 1 g IV days 1–3 then MTP p.o. 0.5 mg/ Awaited
kg/day for days 4–30 on months 1, 3, and 5
and cyclophosphamide PO 2.0 mg/kg/day for
30 days on months 2, 4, and 6 vs. tacrolimus
0.05 mg/kg/day for 6 months, then tapered to
withdrawal by 9 months + RTX 1 g at day 180.
CR, complete remission; CsA, cyclosporine; IV, intravenous; MTP, methylprednisolone; NIAT, nonimmunosuppressive antiproteinuric therapy; PO, per oral; PR, partial remission; RCT,
randomized controlled trial; RTX, rituximab.
c16.indd 246 09-12-2022 15:23:39

Table 16.8 Adrenocorticotrophic hormone treatment in idiopathic membranous nephropathy.
Median
Author Year Level of Evidence Risk Group N Treatment regimens Follow-up Outcomes
Berg [94] 1999 Prospective Medium– 14 ACTH 1 mg intramuscular. The dose of ACTH 70 days Urinary albumin excretion dropped
uncontrolled high was increased from one injection every other from 4830 mg/day (3660–15877) to
week to three injections per week during a 1040 mg/day (630–3750). Serum
period of 8 weeks. creatinine went up from 135 μmol/l
(98–314) to 110 μmol/l (98–210)
Ponticelli [96] 2006 RCT Medium 32 Group A (16 patients) received 22 months CR or PR in group A was 93% (CR 5
methylprednisolone(1 g) administered IV on 3 and PR 10 patients) and 87% in group
consecutive days, and then orally 0.4 mg/kg B achieved CR or PR (10 CR and 4
for 27 days on months 1, 3, and 5. On months PR).
2, 4, and 6, received either chlorambucil There was no significant difference in
(0.2 mg/kg/day orally) or cyclophosphamide either treatment group.
(2.5 mg/kg/day orally).
Group B (16 patients) received tetracosactide
(a synthetic analogue of ACTH),
intramuscular injection of 1 mg between 7 : 00
and 9 : 00 a.m. Administration of ACTH was
increased from 1 injection every other week
to 2 injections per week for a total treatment
period of 1 year.
Hladunewich [97] 2014 RCT (nonblinded Medium 20 Patients received either 40 or 80 units of 12 months In both groups combined, the mean
dose finding ACTH. The dose of ACTH was increased from proteinuria improved significantly
study) one injection every other week to two from 9.1 ± 3.4 g/day to 3.9 ± 4.2 g/day
injections per week for total of 12 weeks at 1 year of follow-up
ACTH, adrenocorticotrophic hormone; CR, complete remission; PR, partial remission; RCT, randomized controlled trial.
c16.indd 247 09-12-2022 15:23:39

controlled trial 200 patients with membranous nephropathy There are no standard or universal first-line specific thera-
were treated every 2 weeks with two different intravenous peutic options for idiopathic membranous nephropathy.
dose regimens and compared to a placebo group over a There is moderate to high certainty evidence that blood
total of 16 weeks [93]. In low certainty evidence, eculizumab pressure lowering therapy aiming for a blood pressure
did not have a detectable effect on proteinuria or renal 125/75 or below will retard CKD progression. There is low
function compared to placebo. certainty evidence that statins, salt restriction, and prophy-
lactic anticoagulation have a role in the treatment of mem-
branous nephropathy.
A
drenocorticotropic Hormone In patients who are at low risk of progression, therapy
focused on blood pressure lowering should suffice, although
Synthetic adrenocorticotropic hormone has been studied
long-term follow-up is needed to ensure that there is no dis-
by various investigators as a treatment option for idiopathic
ease progression or worsening of proteinuria.
membranous nephropathy [94–97]. These studies have
Patients at medium or high risk are candidates for addi-
been summarized in Table 16.8. However, these studies
tional immunosuppressive therapy. Patients with persis-
have small sample size with only short-term follow up.
tent nephrotic-range proteinuria are at increased risk for
There is low-certainty evidence that adrenocorticotropic
cardiovascular and thromboembolic complications.
hormone has a significant effect on proteinuria.
Proteinuria is an inducer of kidney injury and plays a
Patients with severe renal insufficiency (serum creati-
major role in the development of progressive tubular
nine of 3 mg/dl) are less likely to benefit from immuno-
injury, interstitial fibrosis, and subsequent loss in GFR. The
suppression therapy, and the risk of treatment is
higher the sustained level of proteinuria, the more likely
significantly higher, so these patients should be considered
the development of end-stage renal disease. Therefore,
for conservative therapy only and with plans made for
even if the main benefit of immunosuppressive therapy is
transplantation in the future.
to accelerate the induction of a remission, it may still have
value in the long term. A treatment algorithm that com-
Disease and Treatment Summary bines the predictive factors and best evidence for immuno-
suppressive therapy is presented in Figure 16.1. Current
In conclusion, control of nephrotic syndrome, specifically first-line therapy is corticosteroids combined with cyclo-
with CR or PR, is strongly associated with renal survival and phosphamide or chlorambucil (which may be associated
a slower rate of chronic kidney disease (CKD) progression. with a higher frequency of adverse events). Calcineurin
Membranous Nephropathy Treatment Algorithm
Low risk Medium risk High risk

Normal renal function Normal renal function and Progressive renal dysfunction and/or
Proteinuria ≤ 4 g/day proteinuria 4–8 g/day proteinuria ≥ 8 g/day
(± anti-PLA2R levels low or negative) despite 6 months maximum despite 2–3 months maximum
conservative therapy conservative therapy
(± high anti-PLA2R titre levels)
Conservative therapy*
Consider one of these three
ACEi/ARB Maintain BP<120/75 mm Hg agents.
Monitor renal function and proteinuria Cytotoxic + CNI ± Rituximab ± If no response to one agent,
corticosteroids corticosteroids CNI consider alternate agent
No response to all
three agents
ACTH/MMF
Figure 16.1 Membranous nephropathy treatment algorithm. ⋆Conservative treatment involves the use of ACEi ± ARB blocker to
maintain blood pressure < 125/75 mmHg, lipid control with HMG-CoA reductase inhibitor, dietary protein restriction (0.6–0.8 g/kg ideal
body weight/day), dietary NaCl intake (goal is 2–3 g Na) to optimize antiproteinuric effects of ACEi and ARBs, smoking cessation, and
attempt to reduce obesity, if present. Abbreviations: anti-PLA2R, phospholipase A2 receptor antibody; ACEi, angiotensin converting
enzyme inhibitor; ARB, angiotensin II receptor blocker; CNI, calcineurin inhibitor; ACTH, adrenocorticotrophic hormone; MMF,
mycophenolate mofetil.
Reference 249
inhibitors combined with low-dose prednisone may be mab may provide CR or PR at 24 months. Patients who do
considered for patients who prefer or are intolerant of first- not respond well or relapse after a first course of immuno-
line therapy. Long-term therapy may be required due to the suppression therapy may benefit from a second course of
high incidence of relapse. Emerging data suggest rituxi- immunosuppression [98].
R
eferences
1 Glassock, R.J. (1992). Secondary membranous idiopathic membranous nephropathy. N. Engl. J. Med. 361
glomerulonephritis. Nephrol. Dial. Transplant. 7 (Suppl (1): 11–21.
1): 64–71. 13 Cattran, D.C. and Brenchley, P.E. (2017). Membranous
2 Donadio, J.V. Jr., Torres, V.E., Velosa, J.A. et al. (1988). nephropathy: integrating basic science into improved
Idiopathic membranous nephropathy: the natural history clinical management. Kidney Int. 91 (3): 566–574.
of untreated patients. Kidney Int. 33 (3): 708–715. 14 De Vriese, A.S., Glassock, R.J., Nath, K.A. et al. (2017).
3 Erwin, D.T., Donadio, J.V. Jr., and Holley, K.E. (1973). A proposal for a serology-based approach to membranous
The clinical course of idiopathic membranous nephropathy. J. Am. Soc. Nephrol. 28 (2): 421–430.
nephropathy. Mayo Clin. Proc. 48 (10): 697–712. 15 Francis, J.M., Beck, L.H. Jr., and Salant, D.J. (2016).
4 Glassock, R.J. (2003). Diagnosis and natural course of Membranous nephropathy: a journey from bench to
membranous nephropathy. Semin. Nephrol. 23 (4): bedside. Am. J. Kidney Dis. 68 (1): 138–147.
324–332. 16 Debiec, H. and Ronco, P. (2016). Immune response
5 Hladunewich, M.A., Troyanov, S., Calafati, J., and against autoantigen PLA2R is not gambling: implications
Cattran, D.C., Metropolitan Toronto Glomerulonephritis for pathophysiology, prognosis, and therapy. J. Am. Soc.
Registry (2009). The natural history of the non-nephrotic Nephrol. 27 (5): 1275–1277.
membranous nephropathy patient. Clin. J. Am. Soc. 17 Ronco, P. and Debiec, H. (2015). Pathophysiological
Nephrol. 4 (9): 1417–1422. advances in membranous nephropathy: time for a shift in
6 Zucchelli, P., Ponticelli, C., Cagnoli, L., and Passerini, P. patient’s care. Lancet 385 (9981): 1983–1992.
(1987). Long-term outcome of idiopathic membranous 18 Sinico, R.A., Mezzina, N., Trezzi, B. et al. (2016).
nephropathy with nephrotic syndrome. Nephrol. Dial. Immunology of membranous nephropathy: from animal
Transplant. 2 (2): 73–78. models to humans. Clin. Exp. Immunol. 183 (2):
7 Schieppati, A., Mosconi, L., Perna, A. et al. (1993). 157–165.
Prognosis of untreated patients with idiopathic 19 Stanescu, H.C., Arcos-Burgos, M., Medlar, A. et al. (2011).
membranous nephropathy. N. Engl. J. Med. 329 (2): 85–89. Risk HLA-DQA1 and PLA(2)R1 alleles in idiopathic
8 Maisonneuve, P., Agodoa, L., Gellert, R. et al. (2000). membranous nephropathy. N. Engl. J. Med. 364 (7):
Distribution of primary renal diseases leading to 616–626.
end-stage renal failure in the United States, Europe, and 20 Lv, J., Hou, W., Zhou, X. et al. (2013). Interaction between
Australia/New Zealand: results from an international PLA2R1 and HLA-DQA1 variants associates with
comparative study. Am. J. Kidney Dis. 35 (1): 157–165. anti-PLA2R antibodies and membranous nephropathy.
9 du Buf-Vereijken, P.W., Branten, A.J., and Wetzels, J.F. J. Am. Soc. Nephrol. 24 (8): 1323–1329.
(2005). Idiopathic membranous nephropathy: outline and 21 Kanigicherla, D., Gummadova, J., McKenzie, E.A. et al.
rationale of a treatment strategy. Am. J. Kidney Dis. 46 (6): (2013). Anti-PLA2R antibodies measured by ELISA
1012–1029. predict long-term outcome in a prevalent population of
10 Mahmoodi, B.K., ten Kate, M.K., Waanders, F. et al. patients with idiopathic membranous nephropathy.
(2008). High absolute risks and predictors of venous and Kidney Int. 83 (5): 940–948.
arterial thromboembolic events in patients with 22 Tomas, N.M., Beck, L.H., Meyer-Schwesinger, C. et al.
nephrotic syndrome: results from a large retrospective (2014). Thrombospondin type-1 domain-containing 7A in
cohort study. Circulation 117 (2): 224–230. idiopathic membranous nephropathy. N. Engl. J. Med. 371
11 Lionaki, S., Derebail, V.K., Hogan, S.L. et al. (2012). (24): 2277–2287.
Venous thromboembolism in patients with 23 Ren, S., Wu, C., Zhang, Y. et al. (2018). An update on
membranous nephropathy. Clin. J. Am. Soc. Nephrol. 7 clinical significance of use of THSD7A in diagnosing
(1): 43–51. idiopathic membranous nephropathy: a systematic
12 Beck, L.H., Bonegio, R.G.B., Lambeau, G. et al. (2009). review and meta-analysis of THSD7A in IMN. Ren. Fail.
M-type phospholipase A2 receptor as target antigen in 40 (1): 306–313.
4 Hoxha, E., Wiech, T., Stahl, P.R. et al. (2016). A

2 membranous nephropathy: definition and relevance of a
mechanism for cancer-associated membranous partial remission. Kidney Int. 66 (3): 1199–1205.
nephropathy. N. Engl. J. Med. 374 (20): 1995–1996. 38 Bech, A.P., Hofstra, J.M., Brenchley, P.E., and Wetzels, J.F.
25 Larsen, C.P., Cossey, L.N., and Beck, L.H. (2016). (2014). Association of anti-PLA(2)R antibodies with
THSD7A staining of membranous glomerulopathy in outcomes after immunosuppressive therapy in idiopathic
clinical practice reveals cases with dual autoantibody membranous nephropathy. Clin. J. Am. Soc. Nephrol. 9
positivity. Mod. Pathol. 29 (4): 421–426. (8): 1386–1392.
26 Cattran, D. (2005). Management of membranous 39 Hoxha, E., Thiele, I., Zahner, G. et al. (2014).
nephropathy: when and what for treatment. J. Am. Soc. Phospholipase A2 receptor autoantibodies and clinical
Nephrol. 16 (5): 1188–1194. outcome in patients with primary membranous
27 Glassock, R.J. (2004). The treatment of idiopathic nephropathy. J. Am. Soc. Nephrol. 25 (6): 1357–1366.
membranous nephropathy: a dilemma or a conundrum? 40 Beck, L.H., Fervenza, F.C., Beck, D.M. et al. (2011).
Am. J. Kidney Dis. 44 (3): 562–566. Rituximab-induced depletion of anti-PLA2R
28 Marx, B.E. and Marx, M. (1999). Prediction in idiopathic autoantibodies predicts response in membranous
membranous nephropathy. Kidney Int. 56 (2): 666–673. nephropathy. J. Am. Soc. Nephrol. 22 (8): 1543.
29 Wehrmann, M., Bohle, A., Bogenschutz, O. et al. (1989). 41 Ruggenenti, P., Debiec, H., Ruggiero, B. et al. (2015).
Long-term prognosis of chronic idiopathic membranous Anti-phospholipase A2 receptor antibody titer predicts
glomerulonephritis. An analysis of 334 cases with post-rituximab outcome of membranous nephropathy.
particular regard to tubulo-interstitial changes. Clin. J. Am. Soc. Nephrol. 26 (10): 2545.
Nephrol. 31 (2): 67–76. 42 Whelton P, K., Carey R, M., Aronow W, S. et al. (2018).
30 Pei, Y., Cattran, D., and Greenwood, C. (1992). Predicting 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/
chronic renal insufficiency in idiopathic membranous ASPC/NMA/PCNA guideline for the prevention,
glomerulonephritis. Kidney Int. 42 (4): 960–966. detection, evaluation, and management of high blood
31 Oh, Y.J., Yang, S.H., Kim, D.K. et al. (2013). pressure in adults: a report of the American College of
Autoantibodies against phospholipase A2 receptor in Cardiology/American Heart Association Task Force on
Korean patients with membranous nephropathy. PLoS Clinical Practice Guidelines. Hypertension 71 (6):
One 8 (4): e62151. e13–e115.
32 Hoxha, E., Harendza, S., Pinnschmidt, H. et al. (2014). 43 Klahr, S., Levey, A.S., Beck, G.J. et al. (1994). The effects
PLA2R antibody levels and clinical outcome in patients of dietary protein restriction and blood-pressure control
with membranous nephropathy and non-nephrotic range on the progression of chronic renal disease. Modification
proteinuria under treatment with inhibitors of the of Diet in Renal Disease Study Group. N. Engl. J. Med. 330
renin-angiotensin system. PLoS One 9 (10): e110681. (13): 877–884.
33 Timmermans, S.A., Abdul Hamid, M.A., Cohen 44 Vegter, S., Perna, A., Postma, M.J. et al. (2012). Sodium
Tervaert, J.W. et al., Limburg Renal Registry (2015). intake, ACE inhibition, and progression to ESRD. J. Am.
Anti-PLA2R antibodies as a prognostic factor in Soc. Nephrol. 23 (1): 165–173.
PLA2R-related membranous nephropathy. Am. J. 45 Keogh, A., Macdonald, P., Kaan, A. et al. (2000). Efficacy
Nephrol. 42 (1): 70–77. and safety of pravastatin vs simvastatin after cardiac
34 Hoxha, E., Harendza, S., Pinnschmidt, H.O. et al. (2015). transplantation. J. Heart Lung Transplant. 19 (6):
Spontaneous remission of proteinuria is a frequent event 529–537.
in phospholipase A2 receptor antibody-negative patients 46 Sarasin, F.P. and Schifferli, J.A. (1994). Prophylactic oral
with membranous nephropathy. Nephrol. Dial. anticoagulation in nephrotic patients with idiopathic
Transplant. 30 (11): 1862–1869. membranous nephropathy. Kidney Int. 45 (2): 578–585.
35 Radhakrishnan, J. and Cattran, D.C. (2012). The KDIGO 47 Perna, A., Schieppati, A., Zamora, J. et al. (2004).
practice guideline on glomerulonephritis: reading Immunosuppressive treatment for idiopathic
between the (guide)lines – application to the individual membranous nephropathy: a systematic review. Am. J.
patient. Kidney Int. 82 (8): 840–856. Kidney Dis. 44 (3): 385–401.
36 Ponticelli, C., Passerini, P., Altieri, P. et al. (1992). 48 Cattran, D.C., Pei, Y., Greenwood, C.M. et al. (1997).
Remissions and relapses in idiopathic membranous Validation of a predictive model of idiopathic
nephropathy. Nephrol. Dial. Transplant. 7 (Suppl 1): membranous nephropathy: its clinical and research
85–90. implications. Kidney Int. 51 (3): 901–907.
37 Troyanov, S., Wall, C.A., Miller, J.A. et al., Toronto 49 Coggins, C.H. (1979). A controlled study of short-term
Glomerulonephritis Registry Group (2004). Idiopathic prednisone treatment in adults with membranous
Reference 251
nephropathy. Collaborative Study of the Adult Idiopathic 60 Donadio, J.V. Jr., Holley, K.E., Anderson, C.F., and Taylor,
Nephrotic Syndrome. N. Engl. J. Med. 301 (24): W.F. (1974). Controlled trial of cyclophosphamide in
1301–1306. idiopathic membranous nephropathy. Kidney Int. 6 (6):
50 Cattran, D.C., Delmore, T., Roscoe, J. et al. (1989). A 431–439.
randomized controlled trial of prednisone in patients 61 Falk, R.J., Hogan, S.L., Muller, K.E., and Jennette, J.C.
with idiopathic membranous nephropathy. N. Engl. J. (1992). Treatment of progressive membranous
Med. 320 (4): 210–215. glomerulopathy. A randomized trial comparing
51 Cameron, J.S., Healy, M.J., and Adu, D. (1990). The cyclophosphamide and corticosteroids with
Medical Research Council trial of short-term high-dose corticosteroids alone. The Glomerular Disease
alternate day prednisolone in idiopathic membranous Collaborative Network. Ann. Intern. Med. 116 (6):
nephropathy with nephrotic syndrome in adults. The 438–445.
MRC Glomerulonephritis Working Party. Q. J. Med. 74 62 West, M.L., Jindal, K.K., Bear, R.A., and Goldstein, M.B.
(274): 133–156. (1987). A controlled trial of cyclophosphamide in patients
52 Short, C.D., Solomon, L.R., Gokal, R., and Mallick, N.P. with membranous glomerulonephritis. Kidney Int. 32 (4):
(1987). Methylprednisolone in patients with membranous 579–584.
nephropathy and declining renal function. Q. J. Med. 65 63 Jindal, K., West, M., Bear, R., and Goldstein, M. (1992).
(247): 929–940. Long-term benefits of therapy with cyclophosphamide
53 Faurschou, M., Sorensen, I.J., Mellemkjaer, L. et al. and prednisone in patients with membranous
(2008). Malignancies in Wegener’s granulomatosis: glomerulonephritis and impaired renal function. Am. J.
incidence and relation to cyclophosphamide therapy Kidney Dis. 19 (1): 61–67.
in a cohort of 293 patients. J. Rheumatol. 35 (1): 64 Torres, A., Dominguez-Gil, B., Carreno, A. et al. (2002).
100–105. Conservative versus immunosuppressive treatment of
54 Ponticelli, C., Zucchelli, P., Imbasciati, E. et al. (1984). patients with idiopathic membranous nephropathy.
Controlled trial of methylprednisolone and chlorambucil Kidney Int. 61 (1): 219–227.
in idiopathic membranous nephropathy. N. Engl. J. Med. 65 Bruns, F.J., Adler, S., Fraley, D.S., and Segel, D.P. (1991).
310 (15): 946–950. Sustained remission of membranous glomerulonephritis
55 Ponticelli, C., Zucchelli, P., Passerini, P., and Cesana, after cyclophosphamide and prednisone. Ann. Intern.
B. (1992). Methylprednisolone plus chlorambucil as Med. 114 (9): 725–730.
compared with methylprednisolone alone for the 66 Warwick, G.L., Geddes, C.G., and Boulton-Jones, J.M.
treatment of idiopathic membranous nephropathy. The (1994). Prednisolone and chlorambucil therapy for
Italian Idiopathic Membranous Nephropathy idiopathic membranous nephropathy with progressive
Treatment Study Group. N. Engl. J. Med. 327 (9): renal failure. Q. J. Med. 87 (4): 223–229.
599–603. 67 Branten, A.J., Reichert, L.J., Koene, R.A., and Wetzels,
56 Ponticelli, C., Zucchelli, P., Passerini, P. et al. (1995). A J.F. (1998). Oral cyclophosphamide versus chlorambucil
10-year follow-up of a randomized study with in the treatment of patients with membranous
methylprednisolone and chlorambucil in membranous nephropathy and renal insufficiency. Q. J. Med. 91 (5):
nephropathy. Kidney Int. 48 (5): 1600–1604. 359–366.
57 Ponticelli, C., Altieri, P., Scolari, F. et al. (1998). A 68 du Buf-Vereijken, P.W., Branten, A.J., and Wetzels, J.F.,
randomized study comparing methylprednisolone plus Membranous Nephropathy Study Group (2004).
chlorambucil versus methylprednisolone plus Cytotoxic therapy for membranous nephropathy and
cyclophosphamide in idiopathic membranous renal insufficiency: improved renal survival but high
nephropathy. J. Am. Soc. Nephrol. 9 (3): 444–450. relapse rate. Nephrol. Dial. Transplant. 19 (5):
58 Jha, V., Ganguli, A., Saha, T.K. et al. (2007). A 1142–1148.
randomized, controlled trial of steroids and 69 Mathieson, P.W., Turner, A.N., Maidment, C.G. et al.
cyclophosphamide in adults with nephrotic syndrome (1988). Prednisolone and chlorambucil treatment in
caused by idiopathic membranous nephropathy. J. Am. idiopathic membranous nephropathy with deteriorating
Soc. Nephrol. 18 (6): 1899–1904. renal function. Lancet 2 (8616): 869–872.
59 Murphy, B.F., McDonald, I., Fairley, K.F., and Kincaid- 70 Alexopoulos, E., Papagianni, A., Tsamelashvili, M. et al.
Smith, P.S. (1992). Randomized controlled trial of (2006). Induction and long-term treatment with
cyclophosphamide, warfarin and dipyridamole in cyclosporine in membranous nephropathy with the
idiopathic membranous glomerulonephritis. Clin. nephrotic syndrome. Nephrol. Dial. Transplant. 21 (11):
Nephrol. 37 (5): 229–234. 3127–3132.
1 Cattran, D.C., Appel, G.B., Hebert, L.A. et al. (2001).

7 85 Moroni, G., Depetri, F., Del Vecchio, L. et al. (2017).
Cyclosporine in patients with steroid-resistant Low-dose rituximab is poorly effective in patients with
membranous nephropathy: a randomized trial. Kidney primary membranous nephropathy. Nephrol. Dial.
Int. 59 (4): 1484–1490. Transpl. 32 (10): 1691–1696.
72 Cattran, D.C., Greenwood, C., Ritchie, S. et al. (1995). 86 Waldman, M., Beck, L.H., Braun, M. et al. (2016).
A controlled trial of cyclosporine in patients with Membranous nephropathy: pilot study of a novel regimen
progressive membranous nephropathy. Canadian combining cyclosporine and rituximab. Kidney Int. Rep. 1
Glomerulonephritis Study Group. Kidney Int. 47 (4): (2): 73–84.
1130–1135. 87 Ruggenenti, P., Chiurchiu, C., Brusegan, V. et al. (2003).
73 Ambalavanan, S., Fauvel, J.P., Sibley, R.K., and Myers, Rituximab in idiopathic membranous nephropathy: a
B.D. (1996). Mechanism of the antiproteinuric effect of one-year prospective study. J. Am. Soc. Nephrol. 14 (7):
cyclosporine in membranous nephropathy. J. Am. Soc. 1851–1857.
Nephrol. 7 (2): 290–298. 88 Ruggenenti, P., Cravedi, P., Chianca, A. et al. (2012).
74 Rostoker, G., Belghiti, D., Ben Maadi, A. et al. (1993). Rituximab in idiopathic membranous nephropathy.
Long-term cyclosporin A therapy for severe idiopathic J. Am. Soc. Nephrol. 23 (8): 1416.
membranous nephropathy. Nephron 63 (3): 335–341. 89 Fervenza, F.C., Abraham, R.S., Erickson, S.B. et al.
75 DeSanto, N.G., Capodicasa, G., and Giordano, C. (1987). (2010). Rituximab therapy in idiopathic membranous
Treatment of idiopathic membranous nephropathy nephropathy: a 2-year study. Clin. J. Am. Soc. Nephrol. 5
unresponsive to methylprednisolone and chlorambucil (12): 2188.
with cyclosporin. Am. J. Nephrol. 7 (1): 74–76. 90 Fervenza, F.C., Cosio, F.G., Erickson, S.B. et al. (2008).
76 Cattran, D.C. (2001). Idiopathic membranous Rituximab treatment of idiopathic membranous
glomerulonephritis. Kidney Int. 59 (5): 1983–1994. nephropathy. Kidney Int. 73 (1): 117–125.
77 Praga, M., Barrio, V., Juarez, G.F., and Luno, J. (2007). 91 Fervenza, F.C., Canetta, P.A., Barbour, S.J. et al. (2015).
Grupo Espanol de Estudio de la Nefropatia M. Tacrolimus A Multicenter randomized controlled trial of rituximab
monotherapy in membranous nephropathy: a versus cyclosporine in the treatment of idiopathic
randomized controlled trial. Kidney Int. 71 (9): 924–930. membranous nephropathy (MENTOR). Nephron 130 (3):
78 Howman, A., Chapman, T.L., Langdon, M.M. et al. 159–168.
(2013). Immunosuppression for progressive membranous 92 Rojas-Rivera, J., Fernandez-Juarez, G., Ortiz, A. et al.
nephropathy: a UK randomised controlled trial. Lancet (2015). A European multicentre and open-label
381 (9868): 744–751. controlled randomized trial to evaluate the efficacy of
79 Miller, G., Zimmerman, R. 3rd, Radhakrishnan, J., and sequential treatment with TAcrolimus-rituximab versus
Appel, G. (2000). Use of mycophenolate mofetil in steroids plus cyclophosphamide in patients with primary
resistant membranous nephropathy. Am. J. Kidney Dis. 36 MEmbranous nephropathy: the STARMEN study. Clin.
(2): 250–256. Kidney J. 8 (5): 503–510.
80 Choi, M.J., Eustace, J.A., Gimenez, L.F. et al. (2002). 93 Appel, G., Nachman, P., Hogan, S. et al. (2002).
Mycophenolate mofetil treatment for primary glomerular Eculizumab (C5 complement inhibitor) in the treatment
diseases. Kidney Int. 61 (3): 1098–1114. of idiopathic membranous nephropathy (abstract). J. Am.
81 Branten, A.J., du Buf-Vereijken, P.W., Vervloet, M., and Soc. Nephrol. 13: –668A.
Wetzels, J.F. (2007). Mycophenolate mofetil in idiopathic 94 Berg, A.L., Nilsson-Ehle, P., and Arnadottir, M. (1999).
membranous nephropathy: a clinical trial with Beneficial effects of ACTH on the serum lipoprotein
comparison to a historic control group treated with profile and glomerular function in patients with
cyclophosphamide. Am. J. Kidney Dis. 50 (2): 248–256. membranous nephropathy. Kidney Int. 56 (4):
82 Dahan, K., Debiec, H., Plaisier, E. et al. (2017). Rituximab 1534–1543.
for severe membranous nephropathy: a 6-month trial 95 Berg, A.L. and Arnadottir, M. (2004). ACTH-induced
with extended follow-up. J. Am. Soc. Nephrol. 28 (1): 348. improvement in the nephrotic syndrome in patients with
83 Fiorentino, M., Tondolo, F., Bruno, F. et al. (2016). a variety of diagnoses. Nephrol. Dial. Transplant. 19 (5):
Treatment with rituximab in idiopathic membranous 1305–1307.
nephropathy. Clin. Kidney J. 9 (6): 788–793. 96 Ponticelli, C., Passerini, P., Salvadori, M. et al. (2006).
84 Bagchi, S., Subbiah, A.K., Bhowmik, D. et al. (2018). A randomized pilot trial comparing methylprednisolone
Low-dose rituximab therapy in resistant idiopathic plus a cytotoxic agent versus synthetic
membranous nephropathy: single-center experience. adrenocorticotropic hormone in idiopathic membranous
Clin. Kidney J. 11 (3): 337–341. nephropathy. Am. J. Kidney Dis. 47 (2): 233–240.
Reference 253
7 Hladunewich, M.A., Cattran, D., Beck, L.H. et al. (2014).

9 98 du Buf-Vereijken, P.W. and Wetzels, J.F. (2004). Efficacy
A pilot study to determine the dose and effectiveness of of a second course of immunosuppressive therapy in
adrenocorticotrophic hormone (H.P. Acthar(R) Gel) in patients with membranous nephropathy and persistent or
nephrotic syndrome due to idiopathic membranous relapsing disease activity. Nephrol. Dial. Transplant. 19
nephropathy. Nephrol. Dial. Transplant. 29 (8): 1570–1577. (8): 2036–2043.
254
17
IgA Nephropathy in Adults and Children

See Cheng Yeo1, Hernán Trimachi2, and Jonathan Barratt3
1
Department of Renal Medicine, Tan Tock Seng Hospital, Singapore
2
Nephrology Service and Kidney Transplant Unit, Hospital Británico de Buenos Aires, Buenos Aires, Argentina
3
The John Walls Renal Unit, Leicester General Hospital, and Department of Cardiovascular Sciences, University of Leicester, Leicester United Kingdom
I ntroduction the management of patients with IgAN. Discussions of the

treatment of IgAV or secondary forms of IgAN are not
Immunoglobulin A nephropathy (IgAN) is the most included in this chapter.
common pattern of glomerulonephritis (GN) in all coun-
tries where renal biopsy is routinely practiced and is an
important cause of end-stage kidney disease (ESKD) at E
pidemiology
all ages [1, 2]. IgAN is a mesangial proliferative glomeru-
lonephritis characterized by the predominant deposition No clinical presentation is pathognomonic of IgAN, includ-
of IgA in the glomerular mesangium. The degree of his- ing the archetypal young male with episodic visible hema-
topathologic iewwwnjury is heterogeneous, and this is turia following an upper respiratory tract infection. Also,
reflected in the varied progression and severity of clinical although a number of abnormalities in circulating IgA and
presentation seen in this disease [3]. Closely associated its production have been reported in IgAN patients, cohorts
with IgAN is IgA vasculitis (IgAV) (also known as are heterogeneous with respect to these abnormalities,
Henoch–Schönlein purpura), a small vessel systemic vas- making their diagnostic utility poor [7]. Therefore, a diag-
culitis characterized by small blood vessel deposition of nosis of IgAN currently requires a renal biopsy, which as
IgA predominantly affecting the skin, joints, gut, and described below also provides important prognostic infor-
kidney. The nephritis of IgAV is also characterized by mation. This requirement for a renal biopsy means the true
mesangial IgA deposition and may be histologically incidence and prevalence of IgAN are difficult to deter-
indistinguishable from IgAN [4]. mine. Indeed, the prevalence of mesangial IgA deposits
In this chapter, we summarize the published studies found at post mortem in unselected cases is surprisingly
according to their level of evidence and provide recom- high: 4–16% [8, 9]. Reported prevalence rates for IgAN vary
mendations for the common clinical situations that con- across the world, ranging from 5% in the Middle East [10]
front the nephrologist treating patients with idiopathic to 38% in Europe [11], and reaching 50% in China and
IgAN (Figure 17.1). In each section we describe the results Japan [12, 13]. Asian people who live in the United States
of therapeutic approaches to these clinical situations in have a fourfold higher incidence of IgAN compared to
adults with IgA nephropathy, followed by a pediatric per- Caucasian Americans of European ancestry, and a sixfold
spective for which we review the results of similar thera- higher prevalence than African-Americans while Northern
peutic trials in children and adolescents. Some of these Europeans with IgAN have a two to four times greater risk
reports have been reviewed previously [6]. Whenever pos- of progressing to end-stage renal disease (ESRD) compared
sible, the pediatric experience with each treatment will be to Southern Europeans [14]. There is also a disparity in
compared with evidence-based conclusions drawn from gender prevalence of IgAN: in Europe and North America
adult studies. In this way, we will attempt to facilitate com- there is a 3 : 1 male-to-female distribution, while this ratio
parison between pediatric and adult recommendations for is 1 : 1 in East and South East Asia [15, 16].
Prognosi 255
P
athophysiology mechanisms (Figure 17.2). There may be acute, severe
immune and inflammatory injury producing crescent for-
Recurrence of IgAN after transplantation, and also the mation, or crescentic IgA nephropathy, which may be ame-
rare cases of resolution of IgA deposits in transplanted nable to intensive immunosuppression. Alternatively,
kidneys from donors with IgAN, supports the hypothesis acute kidney injury can occasionally occur with mild glo-
that mesangial IgA is derived from a pathogenic IgA frac- merular injury when heavy glomerular hematuria leads to
tion within the circulating pool of serum IgA [7, 17, 18]. tubular occlusion and/or damage by red blood cells. This is
What defines this pathogenic IgA fraction is incompletely a reversible phenomenon, and recovery of renal function
understood, but there is evidence for the importance of occurs with supportive measures.
low-affinity, poorly O-galactosylated polymeric IgA1 mol-
ecules forming circulating IgA immune complexes with a
propensity for both mesangial deposition and mesangial Pediatric IgAN
cell activation [19, 20]. The lack of a complete understand-
Children with progressive forms of IgAN will often not
ing of the pathogenesis of IgAN has resulted in there still
develop ESRD until they are adults [24, 25]. Thus, most
being no treatment known to modify mesangial deposi-
pediatric trials must rely on surrogate measures, as dis-
tion of IgA. Available treatment options are mostly
cussed in the next section. Some reports of childhood
directed at downstream immune and inflammatory events
IgAN have concluded that the risk of progressive renal
in the glomerulus and the tubulo-interstitium that may
failure is very low [26–28], whereas others have shown
lead to renal scarring. It is therefore likely that these are
that a significant number of pediatric patients with IgAN
generic treatments with potential benefit in other chronic
will progress to ESRD [24, 25, 29]. It has been estimated
glomerular diseases.
that as many as 30% will progress to ESRD in the United
States [25], but for Japanese children only 11–20% are
expected to progress to ESRD [24, 28, 29]. However, simi-
P
rognosis
lar to many adult studies, predictions are based on differ-
ent selection criteria for renal biopsy candidates [28, 30].
Adult IgAN
In addition, when progressive disease occurs, it is often
The natural history of IgAN has now been well defined in insidious, resulting in considerable difficulty in assessing
a number of large series with prolonged follow-up [21]. outcomes over the short term. This also presents a signifi-
Fewer than 10% of all patients with IgAN have complete cant obstacle when designing clinical trials to evaluate
resolution of urinary abnormalities [22] and episodes of the effect of therapeutic interventions within a feasible
visible hematuria become less frequent with time after time period (i.e. 2–5 years).
diagnosis, although the majority of patients will still have
persistent invisible hematuria. All patients have the poten-
tial for slowly progressive chronic kidney disease leading
Prognostic Factors
eventually to ESRD. Approximately 25–30% of any cohort
will require renal replacement therapy within 20–25 years Many studies have identified features at presentation that
of presentation. From the first renal symptom, on average, mark a poor prognosis (Table 17.1) [21]: proteinuria more
1.5% of patients with IgAN have been calculated to reach than 1 g/24 h, raised serum creatinine, and hypertension.
ESRD per year [23], but the observed risk varies with the The severity of proteinuria has been shown to correlate
diagnostic approach. Centers with a low threshold for renal with extent of glomerular lesions [31]. In one study, 98% of
biopsy for patients with mild urine abnormalities, particu- patients presenting with proteinuria of <1 g/24 h had at
larly those in countries where urine screening programs least a 15-year renal survival [32]. Another analysis identi-
are established, will likely diagnose IgAN in a larger fied those with proteinuria of more than 1 g/24 h and serum
number of patients with mild disease and good prognosis creatinine more than 1.7 mg/dl. The 7-year renal survival
(length bias) and at an earlier stage in their disease (lead- was 99% if both values were below this threshold, 87% if
time bias), thus favorably influencing the apparent progno- either one was above the cutoff, and 21% if both were above
sis for the cohort. the cutoff [33]. Episodic visible hematuria does not entail a
Rarely, acute kidney injury can complicate preexisting poor prognosis. It is likely that this observation reflects the
IgAN, and evaluation should include a further renal biopsy variation in time between clinical presentation and diagno-
unless renal function improves rapidly with supportive sis, which can hinder the accurate interpretation of the
measures. Acute kidney injury develops by two distinct natural history.
256 IgA Nephropathy in Adults and Children
Table 17.1 Prognostic markers at presentation.
Prognosis category Clinical finding(s) Histopathological findings
Worse prognosis Increasing age Light microscopy

Duration of preceding symptoms Mesangial hypercellularity
Severity of proteinuria Endocapillary hypercellularity
Hypertension Segmental glomerulosclerosis
Renal impairment Tubular atrophy/interstitial fibrosis
Increased body mass index Cellular/fibrocellular crescents
Serum uric acid Immunofluorescence
Capillary loop IgA deposits
Ultrastructure
Capillary wall electron-dense deposits
Mesangiolysis
GBM abnormalities
Good prognosis Recurrent visible hematuria Minimal light microscopic abnormalities
No effect on prognosis Gender Intensity of IgA deposits
Serum IgA level Codeposition of mesangial IgG, IgM, or C3
GBM, glomerular basement membrane.
he Oxford Classification of IgA

T whom IgA deposition will lead to substantial and sustained
Nephropathy glomerular injury. An approach incorporating information
on blood pressure (BP), proteinuria, and estimated glomer-
Renal biopsy provides important independent prognostic ular filtration rate (eGFR) at time of biopsy along with the
information about the renal outcome of patients with IgAN. Oxford MEST-C score has recently been developed by the
The recently updated Oxford Classification of IgAN [34] International IgA Nephropathy Network (IIGANN) which
identifies five morphological features that correlate strongly allows individual risk stratification at the time of biopsy
with clinical outcome, independent of known clinical risk and is available at the QxMD website (https://qxmd.com/
factors at the time of diagnosis, including the presence of calculate/ calculator_499/international-igan-prediction-
hypertension, reduced glomerular filtration rate (GFR), and tool-adults).
degree of proteinuria as well as proteinuria and blood pres-
sure (BP) at follow-up. These features are mesangial hyper- T
reatment
cellularity (M), endocapillary hypercellularity (E), segmental
glomerulosclerosis (S), crescent formation (C), and tubular We have reviewed the English literature published on the
atrophy/interstitial fibrsosis (T). The presence of multiple treatment of IgAN since 1976. Despite the prevalence of
pathological features (M, E, C, S, and/or T) in combination IgAN, published randomized controlled trials (RCTs) are
results in additive risk of kidney disease progression. The few in number, and even recent RCTs are not always suf-
predictive value of these biopsy features is similar in both ficiently powered to provide definitive information on
adults and children. Since its publication, the Oxford tested interventions. In part, this is because IgAN is a
Classification of IgAN has been validated in different patient slowly progressive disease, making it necessary to study
populations from North America, Europe, and Asia, and is large numbers of patients for prolonged periods of time to
now widely accepted as the histopathological scoring system determine the efficacy of any therapeutic intervention.
of choice for IgAN [35–38]. Another consequence of the slowly progressive nature of
The various prognostic factors described may be inform- IgAN is that for many of the trials now published, patient
ative for groups of patients but do not have sufficient accu- recruitment occurred at a time when the management of
racy to predict the prognosis of an individual patient with progressive glomerular disease was less clearly defined
complete confidence. More refined methods of assessing than it is now. These caveats are even more relevant for
disease activity are required to define those patients in reports dealing with therapeutic interventions in children
All Patients with IgA 257
with IgAN. Although the most important outcome indica- with proteinuria of >1.0 g/l/1.73 m2/day should be treated
tor for all patients with IgAN is based on deterioration of with the maximal tolerated dose of either an ACEi or ARB
the GFR, the period from diagnosis to ESRD in patients to reduce proteinuria as much as possible. Furthermore,
with onset in childhood may be decades and may be highly all patients should have an individual assessment of their
heterogeneous. Thus, most studies of therapy for pediatric cardiovascular risk, particularly those with progressive
IgAN have relied on surrogate markers, such as deteriora- chronic kidney disease, hypertension, and dyslipidemia,
tion in renal biopsy findings [39, 40] and/or decline in the and treated accordingly (reviewed in Chapter 6).
amount of proteinuria or hematuria [40–42]. Significant
deterioration in renal function (i.e. 40–50% reduction of Recommendations for Children
GFR or doubling of serum creatinine concentration) is the
surrogate end point most associated with progression to It is also important to maintain pediatric patients with
ESRD, although this occurs infrequently in patients par- IgAN in a normotensive range, utilizing age-, gender-, and
ticipating in pediatric trials of IgAN [42]. height-appropriate norms for BP [43] (strong recommen-
dation, high level of evidence). ACEi and ARB are the first
choice for treatment of hypertension and/or proteinuria
All Patients with IgAN (strong recommendation, high level of evidence).
Recommendations for Adults Evidence in Adults

All adult patients with IgAN should have a BP of <125/75 In common with virtually all kidney diseases, hyperten-
to 130/80 mmHg (strong recommendation, high level of sion is a risk factor for progression in IgAN and should be
evidence). The antihypertensive agents of choice should be treated early in the course of the disease [21, 44]. There
angiotensin converting enzyme inhibitors (ACEi) and is, however, limited RCT evidence devoted specifically to
angiotensin II receptor blockers (ARB) (strong recommen- IgAN concerning a target BP required to preserve renal
dation, high level of evidence). All normotensive patients function. In one 3-year RCT of 49 patients with IgAN, the
Biopsy proven IgA nephropathy

Proteinuria
Adults < 1 g/24 h 1 to 3 g/24 h > 3 g/24 h

Pediatrics PCR males < 0.8 [g/g]; females, < 0.6 0.8–2.0 > 2.0
Clinically Clinically
Non-nephrotic Nephrotic syndrome
OR AND
Renal biopsy Renal biopsy
Target BP <125/75 Significant light Minimal light
(or < 90th percentile in pediatric patients) microscopic changes microscopic changes
ideally with ACEI and/or ARB
Management of individual
cardiovascular risk Target BP <125/75 Treat as for minimal change
(or < 90th percentile in pediatric patients) disease with prednisone in
ideally with ACEI and/or ARB both children and adults using
age appropriate regimens
Management of individual
cardiovascular risk
Consider prednisone if progressive decline in

Adults GFR and/or proteinuria >1 g/24 hr despite
achieving target BP and taking ACEI and ARB
Consider prednisone +/– azathioprine if progressive
Pediatrics decline in GFR and/or PCR > 0.8 males [0.6 females]
despite achieving target BP and taking ACEI and/or ARB
Figure 17.1 Flowchart for the management of IgA nephropathy.

achieved mean BP of 129/70 mmHg stabilized GFR over Evidence in Children

3 years, whereas patients with an achieved mean BP of
A placebo-controlled RCT conducted in 23 centers in five
136/76 mmHg showed an average decline in GFR of
European countries that evaluated benazepril (0.2 mg/kg/
13 ml/min over 3 years [45]. As with other proteinuric
day) in 66 children and young adults <35 years of age pro-
glomerular diseases, the antiproteinuric effect of antihy-
vided evidence to support the use of ACEi in pediatric
pertensive treatment appears to predict renoprotection,
patients [54]. The patient cohort included 29 patients
and so therapy should be titrated not only to target BP
18 years of age. Seventy-four percent of the patients had
values but also to maximize reduction of proteinuria.
their initial renal abnormalities when they were 18 years
There is increasing evidence that the antihypertensive
old. The overall allocation of patients was 32 in the benazepril
strategy of choice should be maximal renin–angiotensin
arm and 34 in the placebo arm. There was a significant ben-
system (RAS) blockade, to both achieve a target BP of
efit with benazepril in preventing progression of renal dis-
<125/75 to 130/80 mmHg and to minimize proteinu-
ease in these patients (defined as a reduction in GFR by 30%
ria [46] (Table 17.2). Retrospective data from the Toronto
versus baseline or worsening of proteinuria to 3.5 g/l/1.73 m2/
GN registry have shown that patients with IgAN treated
day). Such progression was seen in 9/34 (26%) of the placebo
with an ACEi to control BP had a lower rate of annual
group but only 1/32 (3%) of the benazepril group.
loss of renal function than similar patients treated with
Furthermore, at the end of the trial, the number of patients
alternative antihypertensives [47]. This is supported by
with proteinuria of <0.5 g/l/1.73 m2/day in the benazepril
an RCT of 44 patients that demonstrated an additional
group was significantly higher than the number in the pla-
benefit of an ACEi (enalapril) on progressive kidney dis-
cebo group (13/32 [40%] vs. 3/34 [9%]; P = 0.0002). This was
ease in IgAN despite equivalent BP control [50]. This
evident in the groups of children (6/10 [60%] vs. 2/19 [10%])
benefit was believed to have arisen from the additional
and adults (7/12 [58%] vs. 1/15 [7%]) in this trial.
reduction in proteinuria seen in the ACEi-treated group.
Similarly, a more recent and larger RCT of 109 Asian
patients showed benefit with an ARB (valsartan) both in atients with Recurrent Visible
P
proteinuria reduction and in retarding the rate of renal
Hematuria
deterioration, although the investigators were unable to
demonstrate a significant improvement in their primary
Recommendation for Adults and Children
end points of doubling of serum creatinine or ESRD at
2 years [51]. Patients with recurrent visible hematuria require no spe-
The safety of using ACEi in combination with an ARB is cific additional intervention (strong recommendation,
unclear [55, 56] and therefore should be avoided in IgAN. high level of evidence).
Table 17.2 Evidence for use of RAS blockade in IgANa.
Author [reference] Design n Treatment Follow-up (months) Results
Cattran [47] NRCT 115 ACEi 29 ↑ GFR, ↓ proteinuria
Coppo [48] NRCT 27 Captopril 26 ↑ GFR, ↓ proteinuria
Maschio [49] RCT 39 Fosinopril 9 ↓ proteinuria
Kanno [45] RCT 49 Benazapril and amlodipine 36 GFR
Praga [50] RCT 44 Enalapril 74 Slower ↓ GFR, ↓ proteinuria
Li [51] RCT 109 Valsartan 24 slower ↓ GFR, ↓ proteinuria
Russo [52] NRCT 8 ACEi and losartan 3 ↓ proteinuria
Song [53] NRCT 14 Ramipril and candersartan 6 ↓ proteinuria
Coppo [54] RCT 66 Benazepril 38 Slower ↓ GFR, ↓ proteinuria
GFR, glomerular filtration rate; NRCT, nonrandomized controlled trial.

a
Overall quality of evidence is moderate: consistent effects, variable quality of design and reporting, moderate-sized and small studies,
surrogate outcomes.
Patients with Rapidly Declining GF 259
Evidence in Adults the seven children with IgAN who received medication only.
However, it is of interest that significant improvement also
Episodes of visible hematuria are self-limiting and can be
followed tonsillectomy in five of the eight children with
provoked by a range of mucosal, most commonly respira-
other glomerulonephritis.
tory, infections. There is no evidence supporting prophy-
These limited data on the effect of tonsillectomy in chil-
lactic use of antibiotics, even in the minority of patients in
dren with IgAN are similar to the reports in adults. Most
whom recurrent episodes are provoked by bacterial tonsil-
of the studies that have been reported in both groups are
litis. Tonsillectomy is still favored as therapy in some
retrospective in nature, are based on unvalidated surro-
regions of the world, notably Japan. Although several ret-
gate markers for outcome, and are confounded by co-
rospective studies of the effectiveness of tonsillectomy
interventions. There is no convincing evidence that
have been published, most contain small numbers of
tonsillectomy will prevent progressive renal disease in
patients, are nonrandomized, uncontrolled trials, and
children or adults, and so this procedure should not be
have generated conflicting data. In one Japanese study
done (weak recommendation, low level of evidence).
tonsillectomy was shown to be an independent factor in
predicting remission in 329 patients followed for a mini-
mum of 3 years [57]. A second retrospective study from
atients with Isolated Invisible
P
Japan looked at the outcome in 118 patients followed over
20 years, of which 48 underwent tonsillectomy [58].
Hematuria and Proteinuria of <1 g/24 h
Benefit from the tonsillectomy only became apparent
Recommendation for Adults and Children
10 years after initial diagnosis. The concomitant use of
other treatment modalities and changing therapeutic Adult patients with isolated invisible hematuria and pro-
goals during the follow-up period make these data difficult teinuria of <1 g/24 h require no specific additional inter-
to interpret. A retrospective study from Germany of 55 vention (strong recommendation). There are insufficient
patients in whom 16 had a tonsillectomy suggested no data to establish a specific evidence-based recommenda-
benefit of tonsillectomy at 10 years [59]. A retrospective tion for children in this category, but we suggest that such
study of 112 Chinese patients, of whom 54 underwent ton- patients require no additional therapy if their urinary pro-
sillectomy, similarly showed no difference in renal sur- tein (grams)/creatinine (grams) ratio is <0.6 (boys) or <0.8
vival at 130 months [60]. The only prospective RCT of (girls) (strong recommendation, high level of evidence).
tonsillectomy combined with steroids versus steroids
alone has reported 12-month outcome data and while
Evidence
there was a small improvement in hematuria and protein-
uria in the tonsillectomy group there was no difference in Available data suggest that most patients presenting with
the rate of eGFR decline between the groups [61]. proteinuria of <1 g/24 h have at least a 15-year renal sur-
vival [32]. It is generally accepted that these patients
require no additional treatment, although they should
Evidence in Children receive regular follow-up. It is, however, important to note
that while a threshold for proteinuria of 1 g/24 h is com-
As for adults, there are no prospective clinical trials evaluat-
monly used to identify those at increased risk of progres-
ing the role of tonsillectomy in children with IgAN. In 1996,
sion (and therefore warranting treatment), this is an
Tomioka et al. reported that 13 of 15 children with IgAN
arbitrary value and the risk attributable to proteinuria is
who underwent tonsillectomy had improved urinalyses,
almost certainly a continuum. Interventions that may
with six of them going into remission [62]. More recently,
lower proteinuria further, and are therefore likely renopro-
changes in the levels of hematuria and proteinuria were
tective, for example RAS blockade, have not been tested in
assessed by Sanai et al. in eight children treated with “medi-
this setting in either adults or children.
cation” combined with tonsillectomy compared to seven
“control” children treated with “medication” [63]. These
patients were a small subset of a larger patient cohort (ages
Patients with Rapidly Declining GFR
3–13 years) that included five children with Henoch–
Schonlein purpura nephritis and eight with “other” types of
Recommendation for Adults
glomerulonephritis. No details were given regarding the
“medication” that was employed. The authors reported that Unexplained acute kidney injury complicating preexist-
both proteinuria and hematuria improved in five of eight ent IgAN requires evaluation with a kidney biopsy
patients in the tonsillectomy group compared to only one of (Figure 17.2). Crescentic IgAN associated with active
Biopsy proven IgA nephropathy

&
rapid unexplained decline in GFR
Re-biopsy
Tubular necrosis Active crescentic glomerulonephritis

&
Minimal glomerular injury
Minimal tubulointerstitial scarring Significant tubulointerstitial scarring
[ >25% global glomerulosclerosis and/or
>40% tubulo-Interstitial fibrosis and tubular atrophy]
Supportive treatment ADULTS Target BP <125/75 (or < 90th

Treat as for small vessel vasculitis with percentile in pediatric patients)
cyclophosphamide and prednisone ideally with ACEI and/or ARB
CHILDREN Management of individual

Treat with IV pulse methylprednisolone cardiovascular risk
followed by oral prednisone
Figure 17.2 Flowchart for the evaluation and management of patients with previously diagnosed IgA nephropathy who present with
a rapid unexplained decline in renal function.
glomerular inflammation and deteriorating renal func- Evidence in Adults

tion in the absence of significant chronic damage should
Acute kidney injury may occur as the first presentation of
be treated with induction therapy comprising cyclo-
IgAN with little preceding renal insult or may be superim-
phosphamide and corticosteroids followed by mainte-
posed on a background of preexistent disease with variable
nance therapy of azathioprine and corticosteroids at
degrees of chronic glomerular and tubulo-interstitial scar-
doses similar to those used for the treatment of anti-
ring. Even if the diagnosis of IgAN has previously been
neutrophil cytoplasmic antibody-positive (ANCA) small
established, evaluation should include renal biopsy to dis-
vessel vasculitis (weak recommendation, very low level
tinguish between acute kidney injury due to acute tubular
of evidence).
necrosis, which should be self-limiting with supportive
treatment, and crescentic IgAN, which may be amenable to
Recommendation for Children intensive immunosuppression.
Crescentic IgAN has a less favorable prognosis, even with
Although there are no controlled trials of treatment regi- immunosuppressive therapy, than other forms of crescentic
mens for children with crescentic IgAN, we recommend glomerulonephritis, such as ANCA-associated small vessel
that pediatric patients with rapidly progressive disease and vasculitis; cumulative published cases suggest that renal sur-
crescents in 50% of their glomeruli be treated with a vival in crescentic IgAN is only 50% at 1 year and 20% at
2-week course of intravenous (IV) methylprednisone 5 years [21, 64]. Evidence of chronic glomerular and tubulo-
pulses (six doses of 1 g/1.73 m2 every other day), followed interstitial injury with scarring usually predicts a poor
by 1 month of daily prednisone (1 mg/kg/day) and then response to intensive immunosuppression. There is no pub-
alternate-day prednisone for 2–3 months (weak recom- lished systematic definition of the degree of chronicity that
mendation, very low level of evidence). Additional therapy predicts poor response to immunosuppressive treatment,
with cyclophosphamide should be considered in patients but we do not recommend such an approach if a representa-
who fail to respond to the pulse therapy (weak recommen- tive renal biopsy shows >40% tubulo-interstitial fibrosis
dation, very low level of evidence). and/or >25% global glomerulosclerosis.
Patients with Nephrotic Syndrom 261
Table 17.3 Evidence for management of crescentic IgAN with acute kidney injurya.
Author [reference] Design n Treatment Follow-up (months) Results
Lai [65] NRCT 2 Plasma exchange 12 Unsustained slower ↓ GFR

Nicholls [66] NRCT 13 Plasma exchange Unsustained slower ↓ GFR
Roccatello [67] NRCT 9 Plasma exchange, CP and steroids 60 Unsustained slower ↓ GFR
Roccatello [68] NRCT 20 CP and steroids 60 Slower ↓ GFR
McIntyre [69] NRCT 9 CP and steroids 17 ↑ GFR, ↓ proteinuria
Tumlin [70] NRCT 24 CP and steroids 36 Stabilized GFR, ↓ proteinuria
CP, cyclophosphamide; GFR, glomerular filtration rate; NRCT, nonrandomized controlled trial.
a
Overall quality very low: inconsistent effects, nonrandomized, very small studies, surrogate outcomes.
A number of case series have been published that indi- biopsy, and six had recovered clinically, although one
cate good preservation of renal function when using patient was still hypertensive.
treatment regimens similar to those recommended for
renal vasculitis, usually with high-dose corticosteroids
and cyclophosphamide, and in some cases plasma Patients with Nephrotic Syndrome
exchange (Table 17.3). However, all of these series are
small nonrandomized studies, most using historical con- Recommendation for Adults and Children
trols. There has still been no RCT of these treatments in In patients presenting with nephrotic syndrome, preserved
crescentic IgAN, and response to treatment is not uni- renal function, and minimal glomerular injury evident on
form. In addition, comparisons across studies are difficult light microscopy, a trial of high-dose corticosteroids using
because published reports use varying definitions of cres- a regimen appropriate for minimal change disease in IgAN
centic IgAN. Some include cases where crescents are should be considered (strong recommendation, very low
seen, but others include acute injury where the glomeru- level of evidence). However, there is no evidence to support
lar tuft is not intense and renal function is not deteriorat- prolonged exposure to corticosteroids if there is not a
ing. One report indicated that there is a subset of prompt response, nor for their use in nephrotic syndrome
crescentic IgAN with circulating ANCA antibodies which in the presence of structural glomerular damage.
respond well to immunosuppression [71].
Evidence in Adults
Evidence in Children In many patients with IgAN, nephrotic syndrome-range pro-
As in adults, the prognosis for children with crescentic teinuria is a manifestation of significant structural glomeru-
IgAN is also poor according to most reports [72]. However, lar damage and progressive renal dysfunction. However, a
in the largest pediatric experience reported to date, Niaudet small minority of both adults and children have nephrosis
et al. described a very aggressive and successful approach with minimal glomerular change on renal biopsy, although
to 12 children aged 8–14 years with crescentic IgAN, 10 of there are also IgA deposits, and proteinuria remits promptly
whom had crescents in 50% of their glomeruli [73]. The in response to corticosteroids. In these patients, two com-
patients received IV methylprednisone in a dose of mon glomerular diseases may coincide: minimal change
1 g/1.73 m2 every other day, followed by 1 month of daily nephrotic syndrome and IgAN [74, 75]. The only RCT of cor-
prednisone (1 mg/kg/day) and then alternate-day pred- ticosteroids in nephrotic IgAN confirmed this approach,
nisone for 2–3 months. Three of the patients received a sec- since there was remission of proteinuria only in patients
ond course of pulse methylprednisone, whereas three with minimal glomerular change on light microscopy [76].
others received cyclophosphamide. Although uncon- More recent RCTs of corticosteroids in IgAN have excluded
trolled, the authors described very good results after a fol- those with nephrotic-range proteinuria, so there is little evi-
low-up period of 1–9 years because none of the children dence to inform treatment choices for nephrotic IgAN with
progressed to ESRD, nine had improved histology on repeat significant histologic glomerular injury.
Evidence in Children (>1 g/24 h) despite tight BP control (<125/75 mmHg) and

maximal RAS blockade (independent of BP), a 6-month
A number of reports have also described the association of
treatment course of corticosteroids may reduce proteinuria
mesangial IgA deposition in association with steroid-
and stabilize kidney function (weak recommendation, low
responsive nephrotic syndrome in children [77–79]. In at
level of evidence). Available evidence does not support a
least two of the patients described in the SPNSG report,
role for mycophenolate mofetil (MMF) in Caucasian
there was a temporal dissociation between the onset and
patients (weak recommendation, very low level of evi-
course of nephrotic syndrome and the later development
dence), although evidence from a single Chinese study sug-
of clinical features of IgAN [77].
gests MMF may be useful as a steroid sparing agent in
selected cases of IgAN in Chinese patients. We cannot rec-
ommend the use of cyclophosphamide (strong recommen-
atients with Slowly Progressive
P dation, low level of evidence), fish oil (weak
Renal Impairment recommendation, very low level of evidence), and other
agents at the present time (Table 17.4).
Recommendations for Adults
Tight BP control to a target of <125/75 mmHg and evalua- Recommendations for Children
tion of individual cardiovascular risk are essential in this
group of patients (strong recommendation, high level of Based upon the limited evidence currently available, we
evidence). In patients with persistent proteinuria are unable to make a specific recommendation regarding
Table 17.4 Treatment for IgA nephropathy.
Strength of
Angiotensin converting We suggest angiotensin converting enzyme inhibitors or High (decrease in eGFR Strong
enzyme inhibitors or angiotensin II receptor blockers be used to lower blood and proteinuria)
angiotensin II receptor pressure to <125/75 to 130/80 and/or maximize
blockers reduction of proteinuria for patients with IgAN
Tonsillectomy We do not suggest tonsillectomy to be considered Low (decrease in eGFR Weak
routinely in patients with IgAN and proteinuria)
Corticosteroids We suggest in patients with progressive decline in eGFR Low (decrease in eGFR Weak
and/or persistent proteinuria (>1 g/24 h) despite tight and proteinuria)
BP control (<125/75 mmHg) and maximal RAS
blockade, a 6-month treatment course of corticosteroids
may reduce proteinuria and stabilize kidney function
Cyclophosphamide and We suggest crescentic IgAN associated with active Low (decrease in eGFR Weak
azathioprine glomerular inflammation and deteriorating renal and proteinuria)
function in the absence of significant chronic damage
should be treated with induction therapy comprising
cyclophosphamide and corticosteroids followed by
maintenance therapy of azathioprine and corticosteroids
at doses similar to those used for the treatment of
ANCA-positive small vessel vasculitis
Mycophenolate mofetil We do not suggest mycophenolate mofetil be used in Very low (decrease in Weak
patients with IgAN eGFR and proteinuria)
Fish oil We do not suggest that fish oil be used in patients with Very low (decrease in Weak
IgAN eGFR and proteinuria)
eGFR, estimated glomerular filtration rate; IgAN, IgA nephropathy; BP, blood pressure; RAS, renin angiotensin system; ANCA, anti-neutrophil
cytoplasmic antibody.
GRADE assessment of the certainty of the evidence [5]: High: this research provides a very good indication of the likely effect. The likelihood
that the effect will be substantially different* is low. Moderate: this research provides a good indication of the likely effect. The likelihood that
the effect will be substantially different* is moderate. Low: this research provides some indication of the likely effect. However, the likelihood
Corticosteroids in Adult 263
the use of fish oil supplements for treatment of IgAN in 60 ml/min or greater received corticosteroid monotherapy
the pediatric patient, although preliminary data from for 6 months, and patients with an eGFR of 30–59 ml/min
the North American IgA Nephropathy Trials indicate received cyclophosphamide for 3 months followed by aza-
that such therapy may be efficacious in reducing pro- thioprine plus oral prednisolone. At 3 years there was no
teinuria in such patients [77]. We recommend that a difference in either the number of patients who had a
trial of alternate-day prednisone be considered in pedi- decrease in eGFR of at least 15 ml/min or the annual rate
atric patients who have deterioration of GFR or persis- of decline in eGFR between the two groups. The glucocor-
tent proteinuria (weak recommendation, very low level ticoid monotherapy arm produced a transient reduction in
of evidence). If prednisone alone is unhelpful, we rec- proteinuria in patients with relatively well-preserved GFR,
ommend therapy combining corticosteroids with aza- but the reduction in proteinuria did not translate into pro-
thioprine (weak recommendation, very low level of tection against GFR decline. Adverse events (severe infec-
evidence) tions, impaired glucose tolerance, and weight gain of >5 kg
in the first year of treatment) were more common in the
immunosuppression group and one patient in the immu-
Evidence in Adults and Children nosuppression group died of sepsis.
Patients at risk of progressive renal dysfunction are typically
those with hypertension, proteinuria of >1 g/24 h, reduced
GFR at the time of diagnosis, and high Oxford scores in the Corticosteroids in Adults
diagnostic renal biopsy (M1, E1, S1, C1/2, and T1/2).
Specific treatment strategies in this group of patients remain The review of immunosuppressive treatments for IgAN by
contentious. Progression is usually slow and therefore large the Cochrane Renal Group identified six RCTs of sufficient
studies with prolonged follow-up are necessary to evaluate quality to be included in their meta-analysis of corticoster-
new treatment strategies in these patients. Reported RCTs oid treatment in IgAN (Table 17.5) [86]. This analysis sug-
have tested interventions intended to slow immune gests that corticosteroid therapy may be effective in
and inflammatory events implicated in progressive IgAN, reducing proteinuria (six trials, 263 patients) and reducing
including corticosteroids, cyclophosphamide, and MMF. risk of ESRD (six trials, 341 patients), although the meta-
Because of the long duration required to identify with con- analysis was unable to evaluate the influence of RAS block-
fidence the benefit of interventions, it is inevitable that ade or achieved BP in the analysis. Follow-up in a large
recruitment into a number of these studies goes back Italian study of corticosteroid treatment has now reached
10 years or more, to a time when the generic approach to 10 years and the investigators report impressive benefit of
progressive glomerular disease was less well defined, so treatment in reducing proteinuria and preventing
that BP targets and the use of RAS blockade are variable in ESRD [87]. However, the high-dose corticosteroid regi-
these studies. men, with pulse methylprednisone (1 g daily for 3 days at
induction and beginning of months 2 and 4) and alternate-
day oral prednisone (0.5 mg/kg) for 6 months, is regarded
I mmunosuppressive Treatments by many physicians as likely to carry considerable toxicity,
in Adults even though none was reported by the investigators.
Notably, RAS blockade was only used in a minority of
The value of immunosuppressive treatment, in addition to patients in this study, although equally distributed among
best supportive therapy, in IgAN remains uncertain. The the participants, and achieved BP was not in line with cur-
Supportive Versus Immunosuppressive Therapy for rent recommendations. Another RCT of corticosteroids
Progressive IgA Nephropathy (STOP-IgAN) trial was a (20 mg/day induction and 5 mg/day maintenance) from
multicenter, open-label, RCT in which immunosuppres- Japan in which BP control was tight even though RAS
sive therapy added to comprehensive supportive therapy blockade was not used showed only a modest reduction in
was compared with comprehensive supportive therapy proteinuria with no protection of GFR [83]. It is unclear
alone in IgAN [80]. During a 6-month run-in period, whether this lack of renoprotection was due to the lower
patients received optimal comprehensive supportive care. dose of corticosteroid or a genuine lack of effect in patients
Patients with persistent urinary protein excretion of managed to current BP targets. In a meta-analysis, consist-
>0.75 g/day were assigned randomly to continue compre- ing of nine studies and 536 patients, Lv et al. found that
hensive supportive care alone or comprehensive support- steroids therapy was associated with lower risk of develop-
ive care plus immunosuppressive therapy for 3 years. In the ing kidney failure and reduction in proteinuria, but a 55%
immunosuppression group, patients with an eGFR of higher risk for adverse events [88].
Table 17.5 RCTs included in the Cochrane Renal Group meta-analysis evaluating the benefit of corticosteroids in IgANa.
Author [reference] n Treatment Follow-up (months) Results
Julian [81] 35 Pred, 60 mg/alt day (3/12) 6–24 Stabilized GFR, ↓ proteinuria

Controls: no treatment
Kobayashi [82] 46 Pred, 40 mg/day (tapering over 7/12) 120 Stabilized GFR, ↓ proteinuria
Katafuchi [83] 90 Pred, 20 mg/day (tapering to 5 mg/day over 24/12) 60 No effect on GFR decline, ↓
Controls: dipyridamole 150–300 mg/day proteinuria
Lai [76] 34 Pred, 40–60 mg/day (halved at 2/12 for further 2/12) 38 Stabilized GFR, ↓ proteinuria
Pozzi [84] 86 MP, 1 g IV 3 times then 0.5 mg/kg/day for 6/12 60 Stabilized GFR, ↓ proteinuria
Shoji [85] 21 Pred, 0.8 mg/kg/day (tapering to 10 mg alt day for 12/12) 13 Stabilized GFR, ↓ proteinuria
Controls: dipyridamole 300 mg/day (12/12)
GFR, glomerular filtration rate; MP, methylprednisone; Pred, prednisone.

a
Overall quality low: inconsistent effects, variable quality of RCT design and reporting, small studies, surrogate outcomes.
The Therapeutic Evaluation of Steroids in IgA in this study would not be expected to have progressive
Nephropathy Global (TESTING) study was a multicenter, disease.
double-blind placebo-controlled RCT in which the efficacy Waldo et al. compared the outcomes of 13 children with
and safety of methylprednisolone (0.6–0.8 mg/kg/day, maxi- IgAN in Alabama followed for 4–10 years after they
mum 48 mg/day) was assessed in patients with IgAN [89]. At received alternate-day prednisone for 2 years with 15 chil-
entry proteinuria was >1 g/24 h and eGFR of 20–120 ml/min dren in Tennessee who received no steroid therapy [41]. All
after at least 3 months of BP control with RAS blockade. The of the patients had either proteinuria of >1 g/m2/day or
study almost exclusively recruited form China and was ter- renal biopsies showing more than a minimal degree of
minated prematurely due to an excess of serious adverse interstitial fibrosis, tubular atrophy, or glomerular sclero-
events, mostly serious infections, including two deaths in sis. None of the 13 treated patients progressed to ESRD,
the methylprednisolone group. While there did appear to be compared to five of 15 of the untreated patients (P = 0.04).
a potential renal benefit of oral methylprednisolone with a At last follow-up, 12 of 13 treated patients had no hematu-
decrease in the primary composite outcome of ESRD, death ria and normal protein excretion (P < 0.001 compared with
resulting from kidney failure, or a 40% decrease in eGFR, a nontreated historic control patients).
definitive conclusion could not be made because of early
termination of the study. A low-dose TESTING study
(ClinicalTrials.gov identifier NCT01560052) is now recruit- yclophosphamide with/without
C
ing using half-dose methylprednisolone with co-trimoxazole Azathioprine in Adults
prophylaxis and is expected to report in 2021/2.
The use of cyclophosphamide in patients at very high risk
of progression (ESRD predicted in all cases within 5 years)
Corticosteroids in Children is supported by a single study. Patients received cyclophos-
phamide (1.5 mg/kg/day for 3 months) followed by azathi-
The short-term effects of prednisone on proteinuria and oprine (1.5 mg/kg/day) in conjunction with high-dose
hematuria were examined by Welch et al. in a group of prednisone (40 mg/day induction, 10 mg/day maintenance)
children with IgAN. Twenty patients were randomized to and were followed for at least 2 years [91]. Notably, BP con-
either placebo or prednisone (2 mg/kg/day, maximum trol and use of RAS blockade in this trial fell outside cur-
80 mg) for 2 weeks, followed by the same dose on alternate rent recommendations. Previous RCTs of cyclophosphamide
days for 10 weeks [90]. After a 12-week washout period, the in less severe, slowly progressive IgAN have shown no con-
treatments were reversed in each subject. No difference in sistent benefit [92], and this is supported by the STOP-
the severity of hematuria was reported after treatment with IgAN trial and a Cochrane Renal Group meta-analysis,
prednisone compared to placebo. However, most of the which failed to show any significant renal survival benefit
subjects had only mild histologic changes, and the subjects from those RCTs incorporating cyclophosphamide,
MMF and Mizoribine in Adults and Childre 265
c yclosporine, or other cytotoxic agents, although there was nisone, azathioprine, heparin, warfarin, and dipyridamole;
a significant reduction in daily proteinuria [80, 93]. These group 2 heparin, warfarin, and dipyridamole [40]. The
studies were, however, insufficiently powered to exclude prednisone dose was 2 mg/kg/day (maximum 80 mg) for
any effect on progression with certainty. 4 weeks, reduced over 8 weeks, then maintained at 1 mg/kg
every other day for 21 months. The azathioprine dose was
2 mg/kg/day. Although there was no significant change in
yclophosphamide with/without
C GFR in either group, group 1 patients had a significant
Azathioprine in Children reduction of proteinuria but those in group 2 did not. There
was also a significant decrease in glomerular IgA staining in
The efficacy of a 1-year combined course of prednisone and follow-up biopsies in group 1 patients but not in group 2.
azathioprine was evaluated in 10 children with IgAN by These follow-up biopsies showed progression of glomerular
Andreoli and Bergstein [94]. Prednisone at 60 mg/m2 (maxi- sclerosis in control patients but not in those receiving pred-
mum 60 mg) daily for 8 weeks was followed by the same dose nisone and azathioprine. More recently, in 2006, Yoshikawa
every other day for 10 months. Azathioprine (2–3 mg/kg/day) et al. described the results of a second RCT that compared
was given for 12 months. Proteinuria decreased significantly the effects of combination therapy using prednisone, aza-
after treatment and serial renal biopsies showed that the activ- thioprine, warfarin, and dipyridamole in 40 children versus
ity score improved, but the chronicity score was unchanged. prednisone alone in 40 children over 24 months [96]. Both
The level of microscopic hematuria also improved. treatment regimens were associated with remarkable
In 1994, Murakami et al. evaluated the efficacy of a improvement in proteinuria. The percentage of glomeruli
6-month course of prednisolone at 10–15 mg on alternate showing sclerotic changes was unchanged from baseline in
days, cyclophosphamide at 1 mg/kg/day, and dipyridamole the patients receiving combination therapy but significantly
at 5 mg/kg/day in 17 patients who had proteinuria of >1 g/ higher in those receiving only prednisone. However, the use
m2/day plus histologic risk factors for progressive disease of ACEi/ARB was prohibited in this trial, hence it is not
compared to 21 patients with similar features who received clear whether such aggressive immunosuppressive therapy
the same regimen plus warfarin for 3 months [95]. The would be the first choice of therapy in clinical practice.
dipyridamole therapy was subsequently continued in all
patients until the patient had 1 g/m2/day proteinuria.
Both groups of patients showed significant improvement MF and Mizoribine in Adults
M
in proteinuria but the chronic histology indices on post and Children
therapy biopsies (14 patients) showed persistent signs of
chronic disease, as with Andreoli and Bergstein [94]. MMF has been used in six major trials for IgAN (Table 17.6).
Addionally, follow-up studies showed rebound deteriora- Two studies reported no benefit from MMF (2 g/day) in
tion of proteinuria after 5–6 years. Caucasian patients either at risk of progression (hyperten-
In 1999, Yoshikawa et al. reported an RCT evaluating sive and/or proteinuria of >1 g/24 h and/or reduced GFR
2 years of therapy in two groups of children: group 1 pred- within 5 years of diagnosis) [97] or with more advanced
Table 17.6 Evidence for the use of MMF in IgANa.
Author [reference] Design n MMF treatment Follow-up (months) Results
Maes [97] RCT 34 2 g/day for 36/12 36 No effect on GFR decline, no effect on

proteinuria
Chen [98] RCT 93 1–1.5 g/day for 6/12 1 g/day for 18 No effect on GFR decline, ↓ proteinuria
6/12 0.75 g/day for 6/12
Tang [99] RCT 40 2 g/day for 6/12 18 No effect on GFR decline, ↓ proteinuria
Frisch [100] RCT 32 2 g/day for 12/12 24 No effect on GFR decline, no effect on
proteinuria
Hogg [101] RCT 44 25-36 mg/kg/day for 12/12 6 (early No effect on proteinuria
termination)
Hou [102] RCT 176 1.5 g/day 12 No effect on proteinuria
GFR, glomerular filtration rate; RCT, randomized controlled trial.

a
disease (mean serum creatinine at entry, 2.6 mg/dl) [100]. Fish Oil in Adults

Both of these studies achieved rigorous BP control with use
of an ACEi. In two separate studies in Chinese patients A number of studies have evaluated the role of fish oils
with less advanced renal impairment, MMF (1–2 g/day) did (eicos-apentanoic acid and docosahexanoic acid) in IgAN
reduce proteinuria over an 18-month follow-up period, but (Table 17.7). An RCT of 106 patients with proteinuria of
neither study demonstrated a change in rate of renal >1 g/24 h and impaired renal function at enrollment (60%
decline [98, 99]. Again, both studies achieved tight BP con- also hypertensive) found those treated with fish oil had a
trol with ACE inhibition. A more recent study examined slower rate of decline in GFR at both 2 and 5 years [107,
the use of MMF versus placebo in Caucasian adults and 108]. This effect appeared independent of the dose
children in addition to lisinopril/losartan plus highly puri- used [109]. These results have not been replicated in other
fied omega-3 fatty acid. The trial was terminated early RCTs studying similar patient cohorts [42, 105, 106]. A
because of a lack of benefit [101]. In a Chinese study, MMF meta-analysis (three trials, 175 patients [105–107]) failed to
and low-dose prednisolone (0.4–0.6 mg/kg per day) was detect a benefit of fish oils on renal outcome in IgAN [110].
compared with full dose prednisolone (0.8–1 mg/kg per Furthermore, a recent RCT showed no benefit following
day). Both regimens had equal efficacy in terms of remis- 2 years of treatment with fish oil compared to placebo [42],
sion rates at 6 and 12 months [102]. However, the MMF although a subsequent post hoc analysis of the data in the
and steroid group suffered significantly fewer adverse study revealed a dose-dependent decrease in proteinuria in
events, suggesting MMF could be used as a steroid sparing the fish oil group [108]. In a Cochrane meta-analysis, con-
agent in Chinese patients with IgAN. sisting of four studies with 256 participants, there was no
A retrospective study in Japan showed that mizoribine, overall benefit of fish oil, as compared to placebo or no
which blocks purine synthesis in a manner similar to treatment, on renal outcomes, including change in serum
MMF, resulted in a significant reduction in proteinuria creatinine, creatinine clearance, and decrease in
when given to 20 pediatric patients in combination with proteinuria [111].
prednisone, warfarin, and dipyridamole [103]. This was
significantly better than the reduction in proteinuria seen
in 21 historic control patients who were given only pred- R
ituximab
nisone, warfarin, and dipyridamole, or in 20 historic
control patients who also received IV pulses of methyl- The role of rituximab in the treatment of patients with IgA
prednisone. Follow-up renal biopsies in the mizoribine- nephropathy was evaluated in an open-label randomized
treated patients showed no progression of chronic lesions, study of 34 patients. Patients were randomly assigned to
whereas the other two sets of patients had a significant receive two doses of rituximab 1 g administered 2 weeks
increase in the chronicity index. A recent pilot study apart and an identical course 6 months later, or no rituxi-
using mizoribine, again in combination with prednisone, mab. Despite successful depletion of CD19+ B cells, there
warfarin, and dipyridamole, confirmed the efficacy and were no differences in the change in proteinuria or change
safety of this regimen in treating children with IgAN [104]. in renal function from baseline between the two groups [112].
Table 17.7 Evidence for the use of fish oil in IgANa.
Author [reference] Design n EPA + DHA daily doses (g) Follow-up (months) Results
Bennett [105] RCT 37 1.8 + 1.2 24 No effect on GFR decline
Pettersson [106] RCT 32 3.3 + 1.8 6 ↓ GFR, no effect on proteinuria
Donadio [107, 108] RCT 106 1.8 + 1.2 24 and 60 Slower ↓ GFR
Donadio [109] RCT 73 3.8 + 2.9 24 No difference between high-and

1.9 + 1.5 low-dose EPA–DHA
Hogg [42] RCT 96 1.88 + 1.48 24 No effect on GFR decline, ↓

proteinuria
DHA, docosahexanoic acid; EPA, eicosapentanoic acid; GFR, glomerular filtration rate; RCT, randomized controlled trial.
a
Reference 267
O
ther Therapies (ClinicalTrials.gov identifier NCT03643965), fostam-
atinib (ClinicalTrials.gov identifier NCT02112838),
Warfarin, urokinase, antiplatelet agents, phenytoin, sparsentan, OMS721 (ClinicalTrials.gov identifier
sodium cromoglycate, dietary gluten restriction, and a low- NCT03608033), LNP023 (ClinicalTrials.gov identifier
antigen content diet have all been assessed for the treat- NCT03373461), blisibimod (ClinicalTrials.gov identifier
ment of IgAN and have not been shown to affect renal NCT02062684), and atacicept (ClinicalTrials.gov identi-
outcomes [92]. fier NCT02808429), while of potential interest for this
The use of newer immunosuppressive agents, such as group of patients, should be considered experimental
the targeted release formulation of budesonide therapies at this point.
R
eferences
1 Wyatt, R.J. and Julian, B.A. (2013). IgA nephropathy. N. 13 Zhou, F.D., Zhao, M.H., Zou, W.Z. et al. (2009). The
Engl. J. Med. 368 (25): 2402–2414. changing spectrum of primary glomerular diseases
2 Barratt, J. and Feehally, J. (2005). IgA nephropathy. J. Am. within 15 years: a survey of 3331 patients in a single
Soc. Nephrol. 16 (7): 2088–2097. Chinese Centre. Nephrol. Dial. Transplant. 24 (3):
3 Berthoux, F.C., Mohey, H., and Afiani, A. (2008). Natural 870–876.
history of primary IgA nephropathy. Semin. Nephrol. 28 14 Kiryluk, K., Li, Y., Sanna-Cherchi, S. et al. (2012).
(1): 4–9. Geographic differences in genetic susceptibility to IgA
4 Davin, J.C., Ten Berge, I.J., and Weening, J.J. (2001). What nephropathy: GWAS replication study and geospatial risk
is the difference between IgA nephropathy and Henoch- analysis. PLos Genet. 8 (6): e1002765.
Schonlein purpura nephritis? Kidney Int. 59 (3): 823–834. 15 Radford, M.G. Jr., Donadio, J.V. Jr., Bergstralh, E.J., and
5 Hultcrantz, M., Rind, D., Akl, E.A. et al. (2017). The Grande, J.P. (1997). Predicting renal outcome in IgA
GRADE Working Group clarifies the construct of nephropathy. J. Am. Soc. Nephrol. 8 (2): 199–207.
certainty of evidence. J. Clin. Epidemiol. 87: 4–13. 16 Feehally, J. and Cameron, J.S. (2011). IgA nephropathy:
6 Wyatt, R.J. and Hogg, R.J. (2001). Evidence-based progress before and since Berger. Am. J. Kidney Dis. 58
assessment of treatment options for children with IgA (2): 310–319.
nephropathies. Pediatr. Nephrol. 16 (2): 156–167. 17 Floege, J. (2004). Recurrent IgA nephropathy after renal
7 Barratt, J., Smith, A.C., Molyneux, K., and Feehally, J. transplantation. Semin. Nephrol. 24 (3): 287–291.
(2007). Immunopathogenesis of IgAN. Semin. 18 Silva, F.G., Chander, P., Pirani, C.L., and Hardy, M.A.
Immunopathol. 29 (4): 427–443. (1982). Disappearance of glomerular mesangial IgA
8 Sinniah, R. (1983). Occurrence of mesangial IgA and IgM deposits after renal allograft transplantation.
deposits in a control necropsy population. J. Clin. Pathol. Transplantation 33 (2): 241–246.
36 (3): 276–279. 19 Novak, J., Julian, B.A., Tomana, M., and Mestecky, J.
9 Waldherr, R., Rambausek, M., Duncker, W.D., and Ritz, E. (2008). IgA glycosylation and IgA immune complexes in
(1989). Frequency of mesangial IgA deposits in a the pathogenesis of IgA nephropathy. Semin. Nephrol. 28
non-selected autopsy series. Nephrol. Dial. Transplant. 4 (1): 78–87.
(11): 943–946. 20 Yeo, S.C., Cheung, C.K., and Barratt, J. (2018). New
10 Demircin, G., Delibas, A., Bek, K. et al. (2009). A insights into the pathogenesis of IgA nephropathy.
one-center experience with pediatric percutaneous renal Pediatr. Nephrol. 33 (5): 763–777.
biopsy and histopathology in Ankara, Turkey. Int. Urol. 21 D’Amico, G. (2004). Natural history of idiopathic IgA
Nephrol. 41 (4): 933–939. nephropathy and factors predictive of disease outcome.
11 Working Group of the International Ig ANN, the Renal Semin. Nephrol. 24 (3): 179–196.
Pathology Study, Roberts, I.S., Cook, H.T. et al. (2009). 22 Costa, R.S., Droz, D., and Noel, L.H. (1987). Long-
The Oxford classification of IgA nephropathy: pathology standing spontaneous clinical remission and glomerular
definitions, correlations, and reproducibility. Kidney Int. improvement in primary IgA nephropathy (Berger’s
76 (5): 546–556. disease). Am. J. Nephrol. 7 (6): 440–444.
12 Sugiyama, H., Yokoyama, H., Sato, H. et al. (2011). Japan 23 Ibels, L.S. and Gyory, A.Z. (1994). IgA nephropathy:
Renal Biopsy Registry: the first nationwide, web-based, analysis of the natural history, important factors in the
and prospective registry system of renal biopsies in Japan. progression of renal disease, and a review of the
Clin. Exp. Nephrol. 15 (4): 493–503. literature. Medicine (Baltimore) 73 (2): 79–102.
4 Kusumoto, Y., Takebayashi, S., Taguchi, T. et al. (1987).

2 IgA nephropathy in adult chinese patients. Am. J. Kidney
Long-term prognosis and prognostic indices of IgA Dis. 60 (5): 812–820.
nephropathy in juvenile and in adult Japanese. Clin. 39 Tanaka, H., Waga, S., and Yokoyama, M. (1998). Age-
Nephrol. 28 (3): 118–124. related histologic alterations after prednisolone therapy
25 Wyatt, R.J., Kritchevsky, S.B., Woodford, S.Y. et al. (1995). in children with IgA nephropathy. Tohoku J. Exp. Med.
IgA nephropathy: long-term prognosis for pediatric 185 (4): 247–252.
patients. J. Pediatr. 127 (6): 913–919. 40 Yoshikawa, N., Ito, H., Sakai, T. et al. (1999). A controlled
26 Kher, K.K., Makker, S.P., and Moorthy, B. (1983). IgA trial of combined therapy for newly diagnosed severe
nephropathy (Berger’s disease) – a clinicopathologic childhood IgA nephropathy. The Japanese Pediatric IgA
study in children. Int. J. Pediatr. Nephrol. 4 (1): 11–18. Nephropathy Treatment Study Group. J. Am. Soc.
27 Michalk, D., Waldherr, R., Seelig, H.P. et al. (1980). Nephrol. 10 (1): 101–109.
Idiopathic mesangial IgA-glomerulonephritis in 41 Waldo, F.B., Wyatt, R.J., Kelly, D.R. et al. (1993).
childhood description of 19 pediatric cases and review of Treatment of IgA nephropathy in children: efficacy of
the literature. Eur. J. Pediatr. 134 (1): 13–22. alternate-day oral prednisone. Pediatr. Nephrol. 7 (5):
28 Nozawa, R., Suzuki, J., Takahashi, A. et al. (2005). 529–532.
Clinicopathological features and the prognosis of IgA 42 Hogg, R.J., Lee, J., Nardelli, N. et al. (2006). Clinical trial
nephropathy in Japanese children on long-term to evaluate omega-3 fatty acids and alternate day
observation. Clin. Nephrol. 64 (3): 171–179. prednisone in patients with IgA nephropathy: report from
29 Yoshikawa, N., Ito, H., Yoshiara, S. et al. (1987). Clinical the Southwest Pediatric Nephrology Study Group. Clin. J.
course of immunoglobulin A nephropathy in children. J. Am. Soc. Nephrol. 1 (3): 467–474.
Pediatr. 110 (4): 555–560. 43 National High Blood Pressure Education Program
30 Hogg, R.J., Silva, F.G., Wyatt, R.J. et al. (1994). Prognostic Working Group on High Blood Pressure in C,
indicators in children with IgA nephropathy--report of Adolescents (2004). The fourth report on the diagnosis,
the Southwest Pediatric Nephrology Study Group. Pediatr. evaluation, and treatment of high blood pressure in
Nephrol. 8 (1): 15–20. children and adolescents. Pediatrics 114 (2 Suppl 4th
31 Neelakantappa, K., Gallo, G.R., and Baldwin, D.S. (1988). Report): 555–576.
Proteinuria in IgA nephropathy. Kidney Int. 33 (3): 44 Katafuchi, R., Takebayashi, S., and Taguchi, T. (1988).
716–721. Hypertension-related aggravation of IgA nephropathy: a
32 Bailey, R.R., Lynn, K.L., Robson, R.A. et al. (1994). Long statistical approach. Clin. Nephrol. 30 (5): 261–269.
term follow up of patients with IgA nephropathy. N. Z. 45 Kanno, Y., Okada, H., Saruta, T., and Suzuki, H. (2000).
Med. J. 107 (976): 142–144. Blood pressure reduction associated with preservation of
33 Frimat, L., Briancon, S., Hestin, D. et al. (1997). IgA renal function in hypertensive patients with IgA
nephropathy: prognostic classification of end-stage renal nephropathy: a 3-year follow-up. Clin. Nephrol. 54 (5):
failure. L’Association des Nephrologues de l’Est. Nephrol. 360–365.
Dial. Transplant. 12 (12): 2569–2575. 46 Jafar, T.H., Stark, P.C., Schmid, C.H. et al. (2003).
34 Trimarchi, H., Barratt, J., Cattran, D.C. et al. (2017). Progression of chronic kidney disease: the role of blood
Oxford classification of IgA nephropathy 2016: an update pressure control, proteinuria, and angiotensin-converting
from the IgA Nephropathy Classification Working Group. enzyme inhibition: a patient-level meta-analysis. Ann.
Kidney Int. 91 (5): 1014–1021. Intern. Med. 139 (4): 244–252.
35 Coppo, R., Troyanov, S., Bellur, S. et al. (2014). Validation 47 Cattran, D.C., Greenwood, C., and Ritchie, S. (1994).
of the Oxford classification of IgA nephropathy in cohorts Long-term benefits of angiotensin-converting enzyme
with different presentations and treatments. Kidney Int. inhibitor therapy in patients with severe immunoglobulin
86 (4): 828–836. a nephropathy: a comparison to patients receiving
36 Katafuchi, R., Ninomiya, T., Nagata, M. et al. (2011). treatment with other antihypertensive agents and to
Validation study of oxford classification of IgA patients receiving no therapy. Am. J. Kidney Dis. 23 (2):
nephropathy: the significance of extracapillary 247–254.
proliferation. Clin. J. Am. Soc. Nephrol. 6 (12): 2806–2813. 48 Coppo, R., Amore, A., Gianoglio, B. et al. (1993).
37 Herzenberg, A.M., Fogo, A.B., Reich, H.N. et al. (2011). Angiotensin II local hyperreactivity in the progression of
Validation of the Oxford classification of IgA IgA nephropathy. Am. J. Kidney Dis. 21 (6): 593–602.
nephropathy. Kidney Int. 80 (3): 310–317. 49 Maschio, G., Cagnoli, L., Claroni, F. et al. (1994). ACE
38 Zeng, C.H., Le, W., Ni, Z. et al. (2012). A multicenter inhibition reduces proteinuria in normotensive patients
application and evaluation of the Oxford classification of with IgA nephropathy: a multicentre, randomized,
Reference 269
placebo-controlled study. Nephrol. Dial. Transplant. 9 (3): 62 Tomioka, S., Miyoshi, K., Tabata, K. et al. (1996). Clinical
265–269. study of chronic tonsillitis with IgA nephropathy treated
50 Praga, M., Gutierrez, E., Gonzalez, E. et al. (2003). by tonsillectomy. Acta Otolaryngol. Suppl. 523: 175–177.
Treatment of IgA nephropathy with ACE inhibitors: a 63 Sanai, A. and Kudoh, F. (1996). Effects of tonsillectomy in
randomized and controlled trial. J. Am. Soc. Nephrol. 14 children with IgA nephropathy, purpura nephritis, or
(6): 1578–1583. other chronic glomerulonephritides. Acta Otolaryngol.
51 Li, P.K., Leung, C.B., Chow, K.M. et al. (2006). Hong Suppl. 523: 172–174.
Kong study using valsartan in IgA nephropathy (HKVIN): 64 Tumlin, J.A. and Hennigar, R.A. (2004). Clinical
a double-blind, randomized, placebo-controlled study. presentation, natural history, and treatment of crescentic
Am. J. Kidney Dis. 47 (5): 751–760. proliferative IgA nephropathy. Semin. Nephrol. 24 (3):
52 Russo, D., Pisani, A., Balletta, M.M. et al. (1999). Additive 256–268.
antiproteinuric effect of converting enzyme inhibitor and 65 Lai, K.N., Lai, F.M., Leung, A.C. et al. (1987). Plasma
losartan in normotensive patients with IgA nephropathy. exchange in patients with rapidly progressive idiopathic
Am. J. Kidney Dis. 33 (5): 851–856. IgA nephropathy: a report of two cases and review of
53 Song, J.H., Lee, S.W., Suh, J.H. et al. (2003). The effects of literature. Am. J. Kidney Dis. 10 (1): 66–70.
dual blockade of the renin-angiotensin system on urinary 66 Nicholls, K., Becker, G., Walker, R. et al. (1990). Plasma
protein and transforming growth factor-beta excretion in exchange in progressive IgA nephropathy. J. Clin. Apher. 5
2 groups of patients with IgA and diabetic nephropathy. (3): 128–132.
Clin. Nephrol. 60 (5): 318–326. 67 Roccatello, D., Ferro, M., Coppo, R. et al. (1995). Report
54 Coppo, R., Peruzzi, L., Amore, A. et al. (2007). IgACE: a on intensive treatment of extracapillary
placebo-controlled, randomized trial of angiotensin- glomerulonephritis with focus on crescentic IgA
converting enzyme inhibitors in children and young nephropathy. Nephrol. Dial. Transplant. 10 (11):
people with IgA nephropathy and moderate proteinuria. 2054–2059.
J. Am. Soc. Nephrol. 18 (6): 1880–1888. 68 Roccatello, D., Ferro, M., Cesano, G. et al. (2000). Steroid
55 Fried, L.F., Emanuele, N., Zhang, J.H. et al. (2013). and cyclophosphamide in IgA nephropathy. Nephrol.
Combined angiotensin inhibition for the treatment of Dial. Transplant. 15 (6): 833–835.
diabetic nephropathy. N. Engl. J. Med. 369 (20): 69 McIntyre, C.W., Fluck, R.J., and Lambie, S.H. (2001).
1892–1903. Steroid and cyclophosphamide therapy for IgA
56 Mann, J.F., Schmieder, R.E., McQueen, M. et al. (2008). nephropathy associated with crescenteric change: an
Renal outcomes with telmisartan, ramipril, or both, in effective treatment. Clin. Nephrol. 56 (3): 193–198.
people at high vascular risk (the ONTARGET study): a 70 Tumlin, J.A., Lohavichan, V., and Hennigar, R. (2003).
multicentre, randomised, double-blind, controlled trial. Crescentic, proliferative IgA nephropathy: clinical and
Lancet 372 (9638): 547–553. histological response to methylprednisolone and
57 Hotta, O., Miyazaki, M., Furuta, T. et al. (2001). intravenous cyclophosphamide. Nephrol. Dial.
Tonsillectomy and steroid pulse therapy significantly Transplant. 18 (7): 1321–1329.
impact on clinical remission in patients with IgA 71 Haas, M., Jafri, J., Bartosh, S.M. et al. (2000). ANCA-
nephropathy. Am. J. Kidney Dis. 38 (4): 736–743. associated crescentic glomerulonephritis with mesangial
5 8 Xie, Y., Nishi, S., Ueno, M. et al. (2003). The efficacy IgA deposits. Am. J. Kidney Dis. 36 (4): 709–718.
of tonsillectomy on long-term renal survival in 72 Welch, T.R., McAdams, A.J., and Berry, A. (1988). Rapidly
patients with IgA nephropathy. Kidney Int. 63 (5): progressive IgA nephropathy. Am. J. Dis. Child. 142 (7):
1861–1867. 789–793.
59 Rasche, F.M., Schwarz, A., and Keller, F. (1999). 73 Niaudet, P., Murcia, I., Beaufils, H. et al. (1993). Primary
Tonsillectomy does not prevent a progressive course in IgA nephropathies in children: prognosis and treatment.
IgA nephropathy. Clin. Nephrol. 51 (3): 147–152. Adv. Nephrol. Necker Hosp. 22: 121–140.
6 0 Chen, Y., Tang, Z., Wang, Q. et al. (2007). Long- 74 Clive, D.M., Galvanek, E.G., and Silva, F.G. (1990).
term efficacy of tonsillectomy in Chinese patients Mesangial immunoglobulin A deposits in minimal change
with IgA nephropathy. Am. J. Nephrol. 27 (2): nephrotic syndrome: a report of an older patient and
170–175. review of the literature. Am. J. Nephrol. 10 (1): 31–36.
61 Komatsu, H., Fujimoto, S., Hara, S. et al. (2008). Effect of 75 Furuse, A., Hiramatsu, M., Adachi, N. et al. (1985).
tonsillectomy plus steroid pulse therapy on clinical Dramatic response to corticosteroid therapy of nephrotic
remission of IgA nephropathy: a controlled study. Clin. J. syndrome associated with IgA nephropathy. Int. J. Pediatr.
Am. Soc. Nephrol. 3 (5): 1301–1307. Nephrol. 6 (3): 205–208.
6 Lai, K.N., Lai, F.M., Ho, C.P., and Chan, K.W. (1986).
7 89 Lv, J., Zhang, H., Wong, M.G. et al. (2017). Effect of oral
Corticosteroid therapy in IgA nephropathy with methylprednisolone on clinical outcomes in patients
nephrotic syndrome: a long-term controlled trial. Clin. with IgA nephropathy: the TESTING randomized
Nephrol. 26 (4): 174–180. clinical trial. JAMA 318 (5): 432–442.
77 The Southwest Pediatric Nephrology Study Group (1985). 90 Welch, T.R., Fryer, C., Shely, E. et al. (1992). Double-
Association of IgA nephropathy with steroid-responsive blind, controlled trial of short-term prednisone therapy
nephrotic syndrome. A report of the Southwest Pediatric in immunoglobulin A glomerulonephritis. J. Pediatr. 121
Nephrology Study Group. Am. J. Kidney Dis. 5 (3): (3): 474–477.
157–164. 91 Ballardie, F.W. and Roberts, I.S. (2002). Controlled
78 Saint-Andre, J.P., Simard, C., Spiesser, R., and Houssin, A. prospective trial of prednisolone and cytotoxics in
(1980). Nephrotic syndrome in a child, with minimal progressive IgA nephropathy. J. Am. Soc. Nephrol. 13 (1):
glomerular lesions and mesangial IgA deposits. Nouv. 142–148.
Presse. Med. 9 (8): 531–532. 92 Barratt, J. and Feehally, J. (2008). Chapter 16 -IgA
79 Sinnassamy, P. and O’Regan, S. (1985). Mesangial IgA nephropathy and Henoch-Schönlein Purpura. In:
deposits with steroid responsive nephrotic syndrome: Therapy in Nephrology and Hypertension, 3e (ed. C.S.
probable minimal lesion nephrosis. Am. J. Kidney Dis. 5 Wilcox), 172–186. Philadelphia: W.B. Saunders.
(5): 267–269. 93 Samuels, J.A., Strippoli, G.F., Craig, J.C. et al. (2004).
80 Rauen, T., Eitner, F., Fitzner, C. et al. (2015). Intensive Immunosuppressive treatments for immunoglobulin A
supportive care plus immunosuppression in IgA nephropathy: a meta-analysis of randomized controlled
nephropathy. N. Engl. J. Med. 373 (23): 2225–2236. trials. Nephrology (Carlton) 9 (4): 177–185.
81 Julian, B.A. and Barker, C. (1993). Alternate-day 94 Andreoli, S.P. and Bergstein, J.M. (1989). Treatment of
prednisone therapy in IgA nephropathy. Preliminary severe IgA nephropathy in children. Pediatr. Nephrol. 3
analysis of a prospective, randomized, controlled trial. (3): 248–253.
Contrib. Nephrol. 104: 198–206. 95 Murakami, K., Yoshioka, K., Akano, N. et al. (1994).
82 Kobayashi, Y., Hiki, Y., Kokubo, T. et al. (1996). Steroid Combined therapy in children and adolescents with IgA
therapy during the early stage of progressive IgA nephropathy. Nihon Jinzo Gakkai Shi 36 (1): 38–43.
nephropathy. A 10-year follow-up study. Nephron 72 96 Yoshikawa, N., Honda, M., Iijima, K. et al. (2006).
(2): 237–242. Steroid treatment for severe childhood IgA
83 Katafuchi, R., Ikeda, K., Mizumasa, T. et al. (2003). nephropathy: a randomized, controlled trial. Clin. J. Am.
Controlled, prospective trial of steroid treatment in IgA Soc. Nephrol. 1 (3): 511–517.
nephropathy: a limitation of low-dose prednisolone 97 Maes, B.D., Oyen, R., Claes, K. et al. (2004).
therapy. Am. J. Kidney Dis. 41 (5): 972–983. Mycophenolate mofetil in IgA nephropathy: results of a
84 Pozzi, C., Bolasco, P.G., Fogazzi, G.B. et al. (1999). 3-year prospective placebo-controlled randomized study.
Corticosteroids in IgA nephropathy: a randomised Kidney Int. 65 (5): 1842–1849.
controlled trial. Lancet 353 (9156): 883–887. 98 Chen, X., Chen, P., Cai, G. et al. (2002). A randomized
85 Shoji, T., Nakanishi, I., Suzuki, A. et al. (2000). Early control trial of mycophenolate mofeil treatment in
treatment with corticosteroids ameliorates proteinuria, severe IgA nephropathy. Zhonghua Yi Xue Za Zhi 82
proliferative lesions, and mesangial phenotypic (12): 796–801.
modulation in adult diffuse proliferative IgA 99 Tang, S., Leung, J.C., Chan, L.Y. et al. (2005).
nephropathy. Am. J. Kidney Dis. 35 (2): 194–201. Mycophenolate mofetil alleviates persistent proteinuria
86 Vecchio, M., Bonerba, B., Palmer, S.C. et al. (2015). in IgA nephropathy. Kidney Int. 68 (2): 802–812.
Immunosuppressive agents for treating IgA 100 Frisch, G., Lin, J., Rosenstock, J. et al. (2005).
nephropathy. Cochrane Database Syst. Rev. (8) (Art. No.: Mycophenolate mofetil (MMF) vs placebo in patients
CD003965). with moderately advanced IgA nephropathy: a double-
87 Pozzi, C., Andrulli, S., Del Vecchio, L. et al. (2004). blind randomized controlled trial. Nephrol. Dial.
Corticosteroid effectiveness in IgA nephropathy: Transplant. 20 (10): 2139–2145.
long-term results of a randomized, controlled trial. J. Am. 101 Hogg, R.J., Bay, R.C., Jennette, J.C. et al. (2015).
Soc. Nephrol. 15 (1): 157–163. Randomized controlled trial of mycophenolate
88 Lv, J., Xu, D., Perkovic, V. et al. (2012). Corticosteroid mofetil in children, adolescents, and adults
therapy in IgA nephropathy. J. Am. Soc. Nephrol. 23 (6): with IgA nephropathy. Am. J. Kidney Dis. 66 (5):
1108–1116. 783–791.
Reference 271
02 Hou, J.H., Le, W.B., Chen, N. et al. (2017).

1 Mayo Nephrology Collaborative Group. N. Engl. J. Med.
Mycophenolate mofetil combined with prednisone 331 (18): 1194–1199.
versus full-dose prednisone in IgA nephropathy with 108 Donadio, J.V. Jr., Grande, J.P., Bergstralh, E.J. et al.
active proliferative lesions: a randomized controlled (1999). The long-term outcome of patients with IgA
trial. Am. J. Kidney Dis. 69 (6): 788–795. nephropathy treated with fish oil in a controlled trial.
103 Kawasaki, Y., Hosoya, M., Suzuki, J. et al. (2004). Mayo Nephrology Collaborative Group. J. Am. Soc.
Efficacy of multidrug therapy combined with Nephrol. 10 (8): 1772–1777.
mizoribine in children with diffuse IgA nephropathy in 109 Donadio, J.V. Jr., Larson, T.S., Bergstralh, E.J., and
comparison with multidrug therapy without mizoribine Grande, J.P. (2001). A randomized trial of high-dose
and with methylprednisolone pulse therapy. Am. J. compared with low-dose omega-3 fatty acids in
Nephrol. 24 (6): 576–581. severe IgA nephropathy. J. Am. Soc. Nephrol. 12 (4):
104 Yoshikawa, N., Nakanishi, K., Ishikura, K. et al. (2008). 791–799.
Combination therapy with mizoribine for severe 110 Strippoli, G.F., Manno, C., and Schena, F.P. (2003). An
childhood IgA nephropathy: a pilot study. Pediatr. “evidence-based” survey of therapeutic options for IgA
Nephrol. 23 (5): 757–763. nephropathy: assessment and criticism. Am. J. Kidney
105 Bennett, W.M., Walker, R.G., and Kincaid-Smith, P. Dis. 41 (6): 1129–1139.
(1989). Treatment of IgA nephropathy with 111 Reid, S., Cawthon, P.M., Craig, J.C. et al. (2011).
eicosapentanoic acid (EPA): a two-year prospective trial. Non-immunosuppressive treatment for IgA
Clin. Nephrol. 31 (3): 128–131. nephropathy. Cochrane Database Syst. Rev. (3) (Art. No.:
106 Pettersson, E.E., Rekola, S., Berglund, L. et al. (1994). CD003962).
Treatment of IgA nephropathy with omega-3- 112 Lafayette, R.A., Canetta, P.A., Rovin, B.H. et al. (2017).
polyunsaturated fatty acids: a prospective, double-blind, A randomized, controlled trial of rituximab in IgA
randomized study. Clin. Nephrol. 41 (4): 183–190. nephropathy with proteinuria and renal dysfunction. J.
107 Donadio, J.V. Jr., Bergstralh, E.J., Offord, K.P. et al. Am. Soc. Nephrol. 28 (4): 1306–1313.
(1994). A controlled trial of fish oil in IgA nephropathy.
272
18
Membranoproliferative Glomerulonephritis
Fabrizio Fabrizi1 and Piergiorgio Messa1,2
1
Division of Nephrology, Maggiore Hospital and IRCCS Foundation, Milano, Italy
2
University School of Medicine, Milano, Italy
I ntroduction M
PGN: Histology
The term membranoproliferative glomerulonephritis The morphological pattern of MPGN includes a marked
(MPGN) refers to a morphological pattern of injury com- and diffuse thickening of the glomerular capillary walls.
mon to a heterogeneous group of diseases; it is also The thickening can be more prominent in some glomeruli
known as “mesangiocapillary glomerulonephritis” or and in some capillary loops than in others. Periodic acid
“lobular glomerulonephritis.” Light microscopy exami- Schiff and silver methenamine stains show that the thick-
nation shows glomerular injury which consists of ened glomerular capillary walls sometimes have two base-
increased mesangial matrix accompanied by an increase ment membranes with a clear nonargyrophylic region
in the cellularity of the mesangium and capillary-wall between them. The double contour is termed duplication
remodeling. These changes result in lobular accentua- of the glomerular basement membrane; interposition of
tion of the glomerular tufts. mesangial cells or trapping of circulating cells between
Based on the results provided by electron microscopy, various layers of basement membrane commonly occurs.
MPGN has been conventionally categorized as primary In some capillaries, the duplication of the glomerular base-
(idiopathic) type I, type II, or type III. The most common ment membrane results in multiple laminations. The
form of MPGN is MPGN I, which is characterized by sub- thickening of the capillary walls, the mesangial expansion
endothelial deposits (often presumed to be immune com- (increase in matrix and cellularity in mesangium), and the
plexes); MPGN type III has both subendothelial and occurrence of subendothelial deposits lower the glomeru-
subepithelial deposits. MPGN type II has dense deposits lar capillary lumen. These abnormalities are most often dif-
in the glomerular basement membrane (“dense deposit fuse and generalized, and lead to a simplification of the
disease”) [1]. architecture of the glomeruli and an accentuation of the
A more recent classification is based on the composition lobulation of the glomerular capillaries. Extracapillary
of the deposits, as highlighted by immunofluorescence crescent formation can complicate the histological picture
staining [2–4]. This classification results from an enhanced in approximately 10% of patients. These are frequently
understanding of the role of complement in the pathogen- small and focal crescents. Morphological changes in the
esis of MPGN, and treatments targeting specific diseases tubules and interstitium typically reflect the changes noted
are under way. This chapter reviews the most important in the glomeruli. Arterial and arterioles are affected in
advances regarding pathogenesis, prognosis, and current cases with renal failure and long-standing hypertension.
therapeutic options for MPGN and some additional MPGN- The strong association between MPGN and hypocomple-
associated conditions. mentemia has been highlighted by West and coworkers in
Immune-complex-mediated MPG 273
1965 and led to the adoption of the term “hypocomple- may be classified into dense-deposit disease (DDD) and C3
mentemic persistent glomerulonephritis” for this subset of glomerular disease (C3GN). Cases of the former MPGN type
glomerular disorders [5]. II (DDD) and some cases of MPGN type III (Strife and
Anders type) are included in the subset of complement-
mediated MPGN.
PGN: Novel Classification and
M An additional category is nonimmune-complex medi-
Pathophysiology ated/noncomplement-mediated MPGN. Glomerular
lesions consistent with MPGN are very infrequent and
According to the novel classification, three types of MPGN confined to a few patients of chronic thrombotic microan-
have been suggested: (i) immune-complex-mediated MPGN, giopathies, de novo chronic transplant nephropathy, radi-
(ii) complement-mediated MPGN, and (iii) nonimmune- ation nephritis, or drug-associated thrombotic angiopathy.
complex mediated/noncomplement mediated MPGN [2–4] In these cases, immunofluorescence is typically negative
(Figure 18.1). for immunoglobulins and complement, and electron-
Patients with immune-complex-mediated MPGN (IC- dense deposits are not detected by electron microscopy
MPGN) typically show at kidney biopsy immunoglobulin (Figure 18.1).
and complement on immunofluorescence staining.
Immune-complex mediated MPGN is further subdivided
depending on the Ig structure (i.e. monoclonal IgG heavy I mmune-complex-mediated MPGN
chain or polyclonal IgG, as in lupus nephritis). When an
autoimmune disease such as lupus is the underlying disor- Immune-complex-mediated MPGN is typically a conse-
der, MPGN is characterized by a “full house” pattern of quence of the deposition of immune complexes in the
immunoglobulin deposition including IgG, IgM, IgA, C1q, glomeruli owing to persistent antigenemia, with antigen-
C3, and kappa/lambda chains [6]. Immune-complex- antibody immune complexes forming as a result of chronic
mediated MPGN includes most of those cases formerly con- infections, elevated levels of circulating immune com-
sidered MPGN type I and some cases of MPGN type III plexes due to autoimmune diseases, or paraproteinemias
(Burkholder type). The absence of marked immunoglobulin due to monoclonal gammopathies. Then, the immune
staining on immunofluorescence microscopy distinguishes complexes activate the classical complement pathway and
MPGN due to alternative-pathway dysfunction from this is clinically reflected in a normal or mildly decreased
immune-complex-mediated MPGN. Indeed, complement- serum C3 concentration and a low C4 level. The most fre-
mediated MPGN is typically immunoglobulin-negative but quent underlying disorders linked with immune-complex
complement-positive on immunofluorescence studies; it is MPGN are infections and these include hepatitis B [7] or C
associated with dysregulation of the alternative pathway of virus (with or without cryoglobulinemia) [8, 9], bacterial,
complement. On the basis of electron microscopy, or protozoal agents (Table 18.1). In addition, autoimmune
complement-mediated MPGN or C33 glomerulopathy (C3G) disorders involved in the pathogenesis of MPGN are
MEMBRANOPROLIFERATIVE GLOMERULONEPHRITIS (MPGN)
Ig(+) C3(+) C3(+)

Immune complex Complement Absent staining
mediated MPGN mediated MPGN
Infections Monoclonal Autoimmune defects HUS Malignant

Viral Hepatitis B/C gammopathy Autoantibody against (healing phase) hypertension
Bacterial regulator factors of
Others complement
Autoimmune diseases Genetic defects Radiation

SLE Complement regulatory gene mutations Nephritis
Rheumatoid arthritis Mutations in the genes encoding C3 and CFB Others
Figure 18.1 Classification of membranoproliferative glomerulonephritis and etiologies.

274 Membranoproliferative Glomerulonephritis
Table 18.1 Infectious agents associated with MPGN. called Meltzer’s triad. The dermatologic manifestations of
MC can vary from simple cutaneous palpable purpura to
Bacterial infections complex confluent lesions (Figure 18.2), including ulcera-
Escherichia coli tion of the skin.
Leptospirosis Although extrarenal signs of mixed cryoglobulinemic
Pseurodomonas aeruginosa vasculitis usually precede the kidney manifestations, often
Salmonella typhi
by years, in 29% of cases kidney and extrarenal involve-
ment are concurrent. Kidney involvement in mixed cryo-
Staphylococcus (aureus, albus)
globulinemia occurs in 8–58% of patients, and in a minority
Streptococcus (group A, viridans)
of cases kidney disease can be the first manifestation of
Viral infections MC. More than half the patients have proteinuria and/or
Hepatitis C hematuria only. A “nephritic” syndrome is diagnosed in
Hepatitis B around 20% of cases. Hypertension is common and may be
Hepatitis E severe, affecting >50% of patients at the time of diagnosis.
Coxsackie virus In 10% of patients, acute oliguric kidney failure is the first
Parvovirus indicator of kidney disease. Often both nephrotic proteinu-
ria and active urinary sediment with elevated serum creati-
Parasitic infestation
nine are present simultaneously. Of particular importance
Schistosoma mansoni or haematobium
is the development of kidney disease because patients with
Toxoplasmosis cryoglobulinemic nephritis have a poor prognosis, mainly
Plasmodium falciparum or malariae because of a high incidence of infections, end-stage liver
Other infectious organisms and cardiovascular diseases.
Candida albicans Roccatello et al. [8] evaluated 146 patients with cryoglo-
Coxiella bulinemic nephritis, of whom 87% (n = 127) were HCV
positive, and noted type II cryoglobulins (IgG/IgM-k) in
rheumathoid arthritis and systemic lupus erythematosus.

Additional causes of MPGN are monoclonal gammopa-
thies (mostly of uncertain significance) and malignancies
(renal cell carcinoma, non-Hodgkin lymphoma) [10–14].
The diagnosis of idiopathic MPGN is made after all of
the secondary causes are excluded; idiopathic MPGN is
considered now a rare entity and the majority of cases of
immune-complex-mediated MPGN are secondary to the
etiologies reported above.
H
CV-associated Immune-complex-
mediated MPGN
Hepatitis C virus (HCV) may instigate mixed cryoglobu-

linemia and the most significant accompanying kidney
lesion is type I MPGN, usually occurring in the context of
type II mixed cryoglobulinemia. Patients with type II MC
show monoclonal IgM k (or IgG or IgA) having rheumatoid
factor (RF) activity and polyclonal Ig (mainly IgG). After
the identification of HCV, it has been recognized as the
cause of 80–90% of mixed cryoglobulinemia. In the absence
of an identified etiologic factor (<5% of all mixed cryoglo-
bulinemia), cryoglobulinemic vasculitis is defined as
essential or idiopathic. The most common symptoms of Figure 18.2 Confluent purpura (right inferior limb: distal
MC syndrome are weakness, arthralgias, and purpura, also portion thigh and proximal leg) in HCV-related MPGN.
Complement-mediated MPGN or C3 Glomerulopath 275
74.4% of cases; the rest had type III cryoglobulins. The have undetectable circulating C3 because their C3 is
most common histologic pattern was diffuse MPGN (83%). persistently consumed by an uncontrolled complement
Survival at 10 years was only 30%, with cardiovascular dis- activation (alternative pathway) [18].
ease the cause of death in >60% of patients. Additional It has been recently discovered that mutations in genes
causes of death included infections (10%), liver failure encoding proteins of the alternative complement pathway
(19%), and neoplasia (3%). Another survey by Tarantino (AP) occur in more than 50% of patients with atypical
et al. (n = 105 patients) gave different results, with deaths hemolytic uremic syndrome (aHUS) and in approxi-
caused by infections (21%) and liver failure (19%) nearly as mately 20% of patients with MPGN [16]. Complement
common as those caused by cardiovascular diseases gene abnormalities associated with aHUS most com-
(29%) [15]. Differences in antibiotics or immunosuppres- monly result in complement dysregulation restricted to
sive agents could explain these differences. the cell surface, whereas complement activation in the
fluid phase is more common in most genetic cases of
MPGN [16]. These acquisitions have highlighted the
C
omplement-mediated MPGN or C3 role of genetic variations in the complement system
Glomerulopathy in determining predisposition to glomerular injury.
Hyperactivation of the alternative pathway of the comple-
C3G encompasses glomerular diseases associated with per- ment system as observed in deposit dense disease (DDD)
sistent activation of the alternative complement path- or C3 glomerulonephritis (GN) has been often associated
way [3, 16, 17]. Kidney biopsy findings are characterized by with multiple mutations in inhibitors of the alternative
the occurrence of glomerular deposits, mostly composed of pathway, including factor H, factor I, CD46 (membrane
C3, and no staining for immunoglobulins or for compo- cofactor protein [MCP]), and the factor–H-related protein
nents of the classical pathway of complement activation. family (CFHR, 2, 3, 4, and 5). CFHRs are a group of
The complement system includes three pathways, the closely related proteins controlling the activity of factor
classical, lectin and alternative pathways, and patients with H; it appears now that CFHRs are able to antagonize the
C3G have abnormal control of the alternative pathway of CFH action. Many patients with C3G show failure of CFH
complement (Figure 18.3). The most important protein in to control the activation of the alternative pathway in the
the circulation that controls the alternative pathway is circulation, thus they have low levels of serum C3 due to
complement factor H (CFH); mice engineered to lack CFH exaggerated consumption.
CLASSICAL COMPLEMENT MANNOSE BINDING LECTIN ALTERNATIVE COMPLEMENT

PATHWAY PATHWAY PATHWAY (AP)
C4, C2
C3(H20)Bb
C4bC2a
C3a
C3 convertase C3b
C3
C3bBb
C3a C3 convertase
C3b
‘’Amplification loop’’ C3
C3a
C4bC2aC3b C3b
C5 convertase
(C3b)2Bb
C5 C5 convertase
C5a
C5b Membrane attack Complex
(C5b-9)
C6,C7,C8,C9
Figure 18.3 Complement cascade (the three complement pathways).

An increased activation of the AP has been also linked C3GN have been reported: some capillary loops show
with the occurrence of auto-antibodies against various intramembranous dense deposits and others have suben-
components of the complement system: anti-CFH, anti- dothelial and subepithelial deposits on electron micros-
CFB (anti-complement factor B), anti-C4, or anti-C3b. copy. Patients with DDD have dark, ribbon-like osmiophilic
These antibodies activate the alternative pathway either deposits within the glomerular basement membrane. DDD
by stabilizing C3 convertase (such as C3 nephritic factor, is typically a disease of children and C3 is frequently
C3NF) or its components against CFH. C3NFs are a group depressed (79% in DDD vs. 48% in C3 GN). DDD is fre-
of autoantibodies that bind to the AP C3 convertase quently associated with extrarenal abnormalities such as
(C3bBb), increasing both convertase activity and the partial lypodystrophy or drusen, which is similar to age-
duration of the active forms, making it more stable (from related macular degeneration. Ocular drusen are yellow
minutes to hours) [16, 17]. These autoantibodies can be deposits under the retina and are made of lipids and fatty
detected by using C3 convertase stabilizing activities, proteins. Drusen likely do not cause age-related macular
semiquantitative hemolytic assays, by measuring autoan- degeneration; however, patients with ocular drusen are at
tibody levels. There are various types of C3NFs, including risk of developing age-related macular degeneration. A few
both IgM and IgG, which can be classified by either being patients with DDD have partial lipodystrophy (APL) which
heat sensitive and properdin dependent or heat stable and is characterized by symmetric loss of adipose tissue from
properdin independent. The clinical significance of upper portions of the trunk, face, and arms.
C3Nefs remains unclear. More than 80% of patients with The most common pathological lesion in C3GN is
dense-deposit disease (DDD) and 40–50% with C3 glo- MPGN, but other patients of glomerular damage have been
merulonephritis (C3GN) have C3NFs; however, their reported, including mesangio-proliferative GN, diffuse
association with disease outcome gave conflicting results. endocapillary GN, focal segmental glomerulosclerosis, and
Alternatively, autoantibody can block the activity of path- crescentic GN [20].
way inhibitors (such as antifactor H autoantibody) [17]. A There is a paucity of data regarding the natural history
familial type of C3 glomerulopathy is CFHR5 nephropa- and outcome of C3 glomerulopathy. It is an infrequent
thy, where the mutation is a duplication of the first two renal disorder and its classification has been recently
exons of the CFHR5 gene [19]. updated. The outcome of C3 glomerulopathy has been ret-
The most recent view is that the dysregulation of the AP rospectively addressed in a large (n = 111) cohort of
may be familiar or triggered by external factors. A genetic patients, 87 with C3GN and 24 with DDD [21].
or acquired defect in complement regulatory proteins if Complement-associated gene variants and auto-antibodies
severe enough may lead to the disease process. On the were detected in 24% and 35% of patients, respectively.
other hand, a genetic mutation with incomplete pene- Over an average observation period of 72 months, remis-
trance might need an environmental trigger to start with sion occurred in 38% of patients with C3GN and 25% of
the disease. The most common environmental triggers are patients with DDD. Progression to end-stage renal disease
infections, surgery, autoimmune disease flares, vaccina- (ESRD) was frequent, and no differences between C3GN
tions, pregnancy, and postpartum state. Thus, genetic (39%) and DDD (42%) were found. Multivariate analysis
mutations affecting the regulatory proteins of the AP showed that eGFR at diagnosis and tubular atrophy/inter-
(CFH, I, membrane co-factor protein, CFH related pro- stitial fibrosis were the strongest predictors for progression
teins, among others) and triggers result in the clinical man- to ESRD. Medjeral-Thomas and coworkers in a European
ifestations of DDD and C3GN. cohort of 80 patients with C3 glomerulopathy (21 with
DDD and 59 with C3GN) observed that patients with DDD
were younger, more likely to have low serum C3 levels, and
3G, Dense Deposit Disease,
C more likely to have crescentic GN than patients with
and C3GN C3GN [22]. Median follow-up, censored for ESRD develop-
ment, in the remaining cohort (n = 70) was 28 months
DDD is characterized by osmiophilic deposits that replace (interquartile range 7–84) without any significant differ-
the glomerular basement membrane and also occur in the ence in median follow-up duration between subgroups. At
mesangium; C3G without the characteristic appearances last follow-up, 20 patients progressed to ESRD (47% of
of DDD is called C3 glomerulonephritis (C3GN). C3GN patients with DDD and 23% of patients with C3GN).
has frequently mesangial, subendothelial, and sometimes Cumulative renal survival did not, however, differ between
subepithelial and intramembranous deposits. It has been groups by Kaplan–Meier survival analysis. Age > 16 years
suggested that DDD and C3GN are part of a continuum. (P < 0.04), DDD subtype (P < 0.03), and crescentic GN
Histological features intermediate between DDD and (P < 0.01) were independent predictors of ESRD within the
MPGN: Clinical Presentation and Natural Histor 277
entire cohort. DDD recurs very frequently after renal trans- screening; (ii) lack of stratification of patients into HCV-
plant; it has been calculated that approximately 45% of positive or negative cases; (iii) inclusion of patients treated
patients lose their allograft within 5 years after transplant with various regimens in observational studies of progno-
due to the recurrence of the disease. sis; (iv) lack of information concerning the degree of blood
pressure control and the use of ace-inhibitors or angioten-
sin receptor blockers; (v) lack of detailed information on
PGN: Clinical Presentation and
M concomitant renal lesions affecting outcomes (e.g. crescen-
Natural History tic involvement); and (vi) failure to stratify patients accord-
ing to the presence and persistence of complement
The clinical manifestations of patients with MPGN are regulatory dysfunction, thrombotic microangiopathy, or
highly variable and similar to those reported in other types monoclonal immunoglobulin deposition disease. For the
of glomerular diseases. Patients can show microscopic last, the recent changes in the classification of MPGN with
hematuria with or without mild proteinuria. Proteinuria in the identification of some additional MPGN-associated
the nephrotic range or even full brown nephrotic syndrome conditions are an additional impediment.
with or without hypertension and low glomerular filtration The analysis of the long-term renal survival (5- and 10-
rate (GFR) may occur. It may also present with macro- year after diagnosis) for idiopathic and untreated MPGN
scopic hematuria with or without a fast worsening of the can be in part reconstructed from the medical literature.
renal function (particularly when crescents are present Placebo or untreated arms of the reported randomized con-
at histology). The discovery of hypocomplementemia trolled trials (RCTs) of specific therapies can give detailed
(reduced levels of C3 and/or C4 or lowered C/H50) is a fre- information on this point. For example, in the open-label
quent feature which characterizes most MPGN forms in RCT of therapy for MPGN in children and adults reported
comparison with the other idiopathic glomerular diseases by Cattran et al. in 1985, the control group (n = 32), who
(minimal change disease, focal and segmental glomerulo- received no specific therapy, displayed a 2-year kidney sur-
sclerosis, membranous nephropathy, and IgA nephropathy vival of 85% in MPGN type I and 90% in MPGN type II. All
seldom, if ever, display hypocomplementemia). of these patients had an initial creatinine clearance of
Clinical factors which are associated with a progressive <80 ml/min/1.73 m2 and proteinuria of >2 g/day at base-
course leading to ESRD include persistence of nephrotic- line [23]. Of note, 8 of 32 control patients in this study
range proteinuria (> 3 g/g of creatinine in a spot urine sam- obtained a spontaneous clinical remission. In the rand-
ple in an adult), the presence of arterial hypertension (AH) omized, placebo-controlled trial of treatment of MPGN
(blood pressure > 140/90 mmHg), and raised serum creati- reported by Tarshish et al. the actuarial renal survival in
nine at baseline. In addition, marked hematuria (> 500 000 the placebo group (n = 40) was only 12% at 130 months,
erythrocytes/ml of urine not centrifuged) is strongly asso- and a progressive decline of renal function was noted in
ciated with the presence of crescents, which confer a less about 60% of the placebo group [24]. In the study by
favorable prognosis. Patients with MPGN who have intact Tarshish and coworkers 52% had MPGN type I, 17% had
kidneys at baseline and show persistent non-nephrotic pro- MPGN type II, and 30% had MPGN type III or other types
teinuria seldom progress to ESRD. of MPGN. Donadio et al. carried out a randomized, double-
Kidney biopsy can provide some pathological features at blind placebo-controlled trial of therapy for MPGN type I.
the time of diagnosis which play a role in the outcome. An The average decline in GFR was 19.6 ml/min/year/1.73 m2
unfavorable course of MPGN is often associated with some in the placebo group [25]. A total of 9 of 19 patients in the
kidney changes, including extensive crescent disease placebo group developed ESRD after 33 months. The
(>50% involvement of glomeruli), extensive glomerulo- authors suggested that the poor outcome observed in the
sclerosis, tubulo-interstitial fibrosis, and tubular atrophy. placebo group could be related to the long duration of dis-
Such findings often are strongly associated with an eleva- ease prior to entry into the randomized trial.
tion of serum creatinine levels. Evidence coming from observational studies can also
A contemporary analysis of the natural history of MPGN give some insight into the natural history of MPGN. In the
(unmodified by any specific therapy) and all of its varieties study by Chan et al. 27 (59%) of 46 patients with MPGN
is extraordinarily difficult, and multiple reasons exist for (including 20% with hepatitis B surface infection) devel-
this. The overall long-term renal survival (5 and 10 years oped or died of ESRD within a follow-up period averaging
after diagnosis) for untreated idiopathic MPGN can be 60 months, with no difference in outcome between treated
reconstructed from reports in the literature. Some factors (n = 19) and untreated (n = 27) patients [26]. Cansick et al.
that limit the accuracy of these predictions include the fol- assessed the long-term outcome of 53 children with MPGN
lowing: (i) lead-time bias for patients discovered during (58% type I, 26% type II, and 16% type III and others).
The median follow-up was 3.3 years (range, 0–17 years). function, oral steroids (prednisone, 1 mg/kg/day of body
Twenty-nine of the 53 children were treated with corticos- weight) should be given for 12–16 weeks. Tapering of ster-
teroids with an uncontrolled design. Overall renal survival oids should last for 6–8 months, but should be faster in
including the treated patients and patients without patients with resistant disease. Cyclophoshamide should
nephrotic syndrome was 92% at 5 years and 83% at 10 years. be considered according to the clinical response to ster-
In the group of patients with nephrotic syndrome at dis- oids. In cases of deterioration of kidney function, oral ster-
covery, the mean renal survival was 8.9 years compared to oids (according to the aforementioned dosage scheme)
13.6 years when nephrotic syndrome was absent [27]. should be tried. If there is no response, oral cyclophos-
In summary, it is difficult on the grounds of the current phamide (1–2 mg/kg body weight adjusted to eGFR) for
medical literature to establish a consistent pattern of long- 4–6 months or rituximab (RTX) (1 g × 2 on day 1 and 15)
term outcome for patients with MPGN who did not receive should be considered. In patients with rapidly progressive
therapy. Overall renal survival of MPGN type I patients glomerulonephritis, with or without nephrotic syndrome,
among treated and untreated patients averages about 60% rescue therapy with intravenous steroids (methypredniso-
at 10 years (range 50–80%), with adults having a less favora- lone, 0.5–1 g/day for three consecutive days) has been
ble outcome. Concomitant hepatitis B or C virus infection suggested, followed by oral prednisone (1 mg/kg/day
likely confers a worse prognosis. body weight) and cyclophosphamide.
Therapies based on ACEIs and ARBs are safe and have
some efficacy in proteinuric patients. Some case series
have been recently published regarding the treatment of
T
reatment idiopathic MPGN with alternative drugs, such as mycophe-
nolate mofetil (MMF) (Table 18.3) [36–42], calcineurin
Idiopathic Immune-complex MPGN
inhibitors [43, 44], and monoclonal antibody (rituximab
As mentioned above, the term idiopathic MPGN in adults and imatinib) (Table 18.4) [45–50]. These drugs have been
or children refers to cases of MPGN with an unknown able to give some benefit to patients with idiopathic MPGN
cause. The Kidney Disease Improving Global Outcomes but we need RCTs on this point. Several questions remain
recommends making the diagnosis of idiopathic MPGN unanswered, including dose and duration of therapy with
only after exclusion of any other identifiable cause [28]; the these drugs, their role (first-line therapies or not), and their
histological pattern of type I MPGN is most typical among safety profile.
patients with idiopathic MPGN. Also, the MPGN histologi-
cal pattern is characterized by various clinical pictures. It
HBV-or HCV-associated MPGN
appears that idiopathic MPGN is more frequent in less-
developed countries, where infectious diseases are more Patients with hepatitis B virus (HBV)-associated MPGN
common than among developed world. The management should receive antiviral therapy with nucleos(t)ide analogs
of secondary MPGN requires treatment of the primary dis- (NUCs) and/or immunosuppressive therapy, according to
ease, often in combination with immunosuppression. the clinical presentation of the disease (Figure 18.4). Many
The evidence regarding the treatment of idiopathic patients with HCV-associated MPGN also have mixed cryo-
MPGN is not robust for numerous reasons. A few RCTs globulinemia and they may also have overt chronic liver
have been published, but all these studies were conducted disease. All patients presenting with MPGN type I are now
a long time ago and do not conform to modern-day stand- routinely tested for anti-HCV antibody (and HBsAg) and
ards of excellence. Thus, many early reports of treatment often are also examined for HCV RNA by reverse polymer-
of idiopathic MPGN probably included cases now catego- ase chain reaction. Thus, it should not be difficult to make
rized as secondary MPGN. Only six RCTs (involving fewer a differential diagnosis between idiopathic forms of MPGN
than 270 patients with MPGN) have been reported for and those secondary to HCV or HBV.
MPGN to date (Table 18.2) [23–25, 29–33]. The discovery of HCV and a better understanding of the
In short, the evidence currently coming from RCTs is mechanisms of disease has provided the opportunity to con-
weak (very low to low level), as already emphasized by trol HCV-induced glomerular disease (including MPGN)
others [34, 35]. For patients with mild disease, i.e. hematu- using targeted approaches: (i) antiviral therapy, supported
ria and low-level proteinuria (<500 mg/day), and normal by the notion that HCV infection stimulates the synthesis of
or nearly normal kidney function, RAS inhibition with immune complexes and the ensuing glomerular disease; (ii)
angiotensin-converting enzyme inhibitors (ACEIs) and B-cell depletion therapy targeting B cells that produce cryo-
angiotensin II receptor blockers (ARBs) should be made. globulins; and (iii) nonspecific immunosuppressive therapy
In patients with nephrotic syndrome and intact kidney targeting inflammatory cells at the glomerular level.
Treatmen 279
Table 18.2 Prospective randomized clinical trials and controlled (or cross-over) clinical trials for the treatment of MPGN.
Authors MPGN,
(reference year) Patients, n type Study, design Study, details Results
Tiller et al. (1981) 39 (adults) I and II RCT Cyclophospha-mide, Not conclusive

warfarin, dypiridamole vs.
placebo
Zimmerman 18 (adults) I Prospective, cross-over Warfarin and dipyridamole Paired analysis: better
et al. (1983) vs. no therapy preservation of kidney function
in the treatment year
Donadio et al. 40 (adults) I Prospective, randomized, Dypiridamole and aspirin Slower deterioration of kidney
(1984) double-blind, placebo- vs. placebo function
controlled study
Mota-Hernandez 18 (children) I Double-blind controlled Prednisone vs. placebo Prednisone reduces ESRD
et al. (1985)
Cattran et al. 59 (adults I and II RCT Cyclophospha-mide, Treatment ineffective in altering
(1985) and coumadin, dypiridamole vs. the natural history of type I
children) no therapy MPGN
Tarshish et al. 80 (children) I, II, Randomized, double- Prednisone or lactose was Renal survival rate was better in
(1992) and III blinded, placebo- given every other day as a patients on prednisone than
controlled clinical trial single morning dose those receiving lactose (P = 0.07)
Zauner et al. 18 (adults) I and II RCT Aspirin and dypiridamole Lower proteinuria, no impact on
(1994) vs. no therapy creatinine
Giri et al. (2002) 30 (adults) I Comparative study ACEIs vs. non-ACEIs ACEIs reduce proteinuria and
improve serum creatinine
ACEIs, angiotensin-converting enzyme inhibitors; ESRD, end-stage renal disease; MPGN, membranoproliferative glomerulonephritis; RCT,
randomized controlled trial.
Table 18.3 Treatment of idiopathic MPGN with mycophenolate mofetil (adults).
Author Treatment Kidney,

(reference year) Patients, n Age Drug duration (months) Outcome type
Levin et al. (2002) 1 56 MMF/low-dose steroids 12 Proteinuria and SCr decrease Native
Choi et al. (2002) 1 78 MMF 11 Proteinuria decrease Native
Jones et al. (2004) 5 32.1 ± 11.6 MMF/low-dose steroids 6 Proteinuria decrease Native
Segarra et al. (2007) 15 NA MMF 12 Partial remission of Native
proteinuria (n = 8)
Yuan et al. (2010) 13 NA MMF/low-dose steroids 12 Proteinuria decrease Native
Mori et al. (2013) 1 26 MMF/low-dose steroids NA Proteinuria decrease Native
Koratala et al. 1 22 MMF 2 Proteinuria decrease Native
(2018)
MMF, mycophenolate mofetil; MPGN, membranoproliferative glomerulonephritis; SCr, serum creatinine.
The information in the literature regarding the manage- Several clinical studies regarding the antiviral therapy of
ment of HCV-associated MPGN is limited; the incidence of HCV-associated glomerular disease (including HCV-related
the disease is probably low even if epidemiological studies MPGN) have been published, some based on IFN [51–53]
on this point are lacking. No prospective RCTs to establish and others that adopted direct-acting antiviral agents
evidence-based recommendations to treat MPGN associated (DAAs) [54, 55]. The evidence generated from these studies
with HCV exist. We urgently need this evidence as renal dis- suggests that the advent of interferon-free antiviral therapy
ease is now considered a major cause of morbidity and mor- based on DAAs represents a real advance even in so far
tality in mixed cryoglobulinemic vasculitis induced by HCV. “difficult-to-treat” settings such as HCV-induced MPGN
Table 18.4 Treatment of idiopathic MPGN with alternative drugs (children and adults).
Author (reference year) Patients, n Age Drug Treatment duration Outcome Kidney, type
Haddad et al. (2007) 2 8 (7–9) TAC/low-dose steroids 20 months (n = 1) Complete proteinuria remission Native
4 months (n = 1)
Bagheri et al. (2008) 18 27 ± 9 CsA/low-dose steroids and NA Complete (n = 12) or partial Native
ACEIs proteinuria remission (n = 5)
Sugiura et al. (2011) 1 48 RTX 375 mg/m2 × 1 Partial remission of proteinuria Native
Wallace et al. (2012) 1 60 Imatinib/low-dose steroids 2 + 10 months Proteinuria and SCr decrease Noninfection-related
Cryoglobulinemic MPGN
Dillon et al. (2012) 6 57 ± 19 RTX 1 g × 2 (days 1 and 15) Complete (n = 2) and partial (n = 4) Cryoglobulinemic MPGN (n = 2)
proteinuria reduction
Kong et al. (2013) 2 49.5 RTX 375 mg/m2 × 2 (n = 1) Complete (n = 1) and partial (n = 1) Native
375 mg/m2 × 1 (n = 1) proteinuria reduction
Farooqui et al. (2015) 1 58 RTX 375 mg/m2 × 2 Complete remission of proteinuria Living-unrelated donor kidney
transplant
Yango et al. (2019) 1 NA RTX/plasmapheresis NA Proteinuria and SCr decrease Living donor kidney transplant
ACEIs, angiotensin-converting enzyme inhibitors; CsA, cyclosporine; MPGN, membranoproliferative glomerulonephritis; NA, not applicable; RTX, rituximab; SCr, serum creatinine; TAC,
tacrolimus.
c18.indd 280 09-12-2022 15:24:29

Treatmen 281
Therapeutic strategies in HBV-and HCV-induced membranoproliferative

(MPGN) glomerulonephritis
Non-nephrotic proteinuria and Rapidly progressive glomerulonephritis

intact kidney function with or without nephrotic syndrome
Nephrotic proteinuria and/or

impaired kidney function
Corticosteroids, cyclophosphamide, rituximab,

Antiviral agents (NUCs or DAAs) at dose adjusted for eGFR,
plasma-exchange and concomitant NUCs or DAAs
ACEIs and/or ARBs
(at dose adjusted for eGFR)
Figure 18.4 Treatment of HBV-and HCV-associated MPGN according to the clinical presentation.
(and other HCV-induced glomerular diseases). These stud- At the present time the cumulative evidence from the
ies have an observational and uncontrolled design; in addi- published studies is not satisfactory; clinical studies or
tion, the size is small. The extensive use of DAAs for the anecdotal reports have suggested that IFN-based regimens
antiviral therapy of HCV in the adult general population will were provided with low efficacy and high toxicity in
make the frequency of HCV-associated MPGN even more patients with HCV-related GN (including HCV-associated
uncommon in the near future. MPGN) compared to those with intact kidneys. The evi-
Some studies regarded the treatment of HCV-associated dence from some observational studies reported that the
glomerular disease (including many cases of HCV- treatment of HCV-associated glomerular disease (includ-
associated MPGN); others assessed the antiviral treatment ing HCV-associated MPGN) with DAAs has given a fre-
of HCV-associated cryoglobulinemic disease, but the details quency of SVR of 95–100%, much greater than that
concerning the cryoglobulinemic glomerulonephritis obtained with IFN-based therapy (35–50%) [54–56].
(including cryoglobulinemic MPGN) were not reported As mentioned in these recent series, there were some
clearly [56]. Of note, the sustained viral response (i.e. the clinical nonresponders or relapsers despite SVR and this
clearance of HCV RNA from serum which persists at least highlights the role of immunosuppressive agents in the
3 months after completing antiviral therapy with DAAs) is management of HCV-associated MPGN (and other HCV-
often accompanied by improvement of glomerular abnor- related glomerular diseases). Only preliminary data exist
malities (reduction of proteinuria, haematuria, and serum on the immunosuppressive agents, i.e. micophenolate
creatinine) and this effect has been achieved in the absence mofetil [57, 58], rituximab [59], or immunomodulatory
of concomitant immunosuppressive therapies. However, therapy (low-dose ribavirin) [60] for HCV-related glomer-
clinical improvement after antiviral agents is sometimes ular disease.
incomplete and this is in keeping with the notion that circu-
lating cryoglobulins may be an independent risk factor for
C3 Glomerulopathy
clinical nonresponse to antiviral treatments. In addition,
remote onset of glomerular disease may be another expla- The optimal treatment for C3G remains unclear and vari-
nation. In other cases, glomerular and/or cryoglobulinemic ous approaches have been suggested, at least from a theo-
disease relapsed over a long follow-up despite sustained retical point of view. C3G was grouped with other forms of
virological response (SVR) and it is well known that cryo- MPGN and this hampered the opportunity to make specific
globulinemic vasculitis shows an own course, regardless of recommendations for C3G.
the presence of trigger agents such as HCV (or HBV). As with any GN, blockade of the RAS remains part of
However, the absence of HCV RNA from serum allows the therapy with lipid-lowering therapy.
adoption of immunosuppressive therapies with benefits for Patients with mutations involving deficiency of regula-
glomerular changes and no concern on replication of HCV tory proteins of the alternative complement pathway may
(Figure 18.3). benefit from replacement of factors with plasma therapy.
Patients might need lifelong plasma therapy. In contrast, generation of the pro-thrombotic and cytolytic terminal
plasma therapy is ineffective in patients with mutations of complex (C5b-9, membrane attack complex [MAC]).
the membrane cofactor protein (MCP, also known as Successful blockade of C5 in C3GN should prevent C5a
CD46) since it is a membrane-bound protein and replen- formation, resulting in lower infiltration of leukocytes. A
ishing plasma with this protein would not offer protection major side effect of EZ therapy is recurrent meningococcal
to the endothelial cells. In patients with gain of function infections (incidence less than 1%); patients are mandated
mutations in complement activating proteins (such as fac- to receive the meningococcal vaccination in advance to
tor B or mutant C3 convertase resistant to CFH), plasma therapy to reduce infection risk.
therapy may be detrimental since it provides additional A small prospective study and a limited number of case
substance to activate AP. series or case reports have shown some efficacy of eculi-
Plasma exchange replenishes plasma with deficient pro- zumab in this disease (Table 18.6) [70–79]. These case
teins, removing autoantibodies or mutant proteins that reports are clearly subject to publication bias (studies pro-
activate the AP. The response to PE remains short lived and viding positive results are selected for reporting). The
rebound complement activation is noted within 48 hours. introduction of EZ has changed the natural history of
Theoretically, immunosuppressive therapy (based on aHUS; studies addressing the activity of EZ in patients
corticosteroids, MMF, and rituximab) should reduce or with C3G have given disparate results, especially com-
eliminate autoantibody production. However, the effec- pared with results of EZ in patients with aHUS. It appears
tiveness of this approach appears unclear and no RCTs that treatment with EZ in patients with C3G is beneficial
supporting this approach have been made; most of the evi- in patients with mild/advanced renal damage, and is able
dence concerning RTX is based on a few case reports [61– to lower not only proteinuria but also to improve kidney
65] (Table 18.5). Treatment of active disease with MMF function. It is clear that treatment with EZ requires fur-
and corticosteroids has shown promise in two retrospective ther assessment in patients with C3G, and prospective
case series [66–67], but was not found to be effective in a studies are those we urgently need. Appropriate candi-
third report regarding patients with more severe baseline dates to therapy with EZ are patients with C3G who have
kidney disease [68]. A high recurrence of the original dis- a relatively short duration of the disease, have shown
ease after solid organ transplant has been noted in patients active inflammatory kidney abnormalities (crescents and
who are already immunosuppressed [69]. endocapillary proliferation), and have reported a recent
The advances regarding the current classification have increase in serum creatinine level and/or proteinuria [79].
coincided with the availability of the first complement The evidence based on pathology is limited and suggests
inhibitor, eculizumab (EZ), which is a humanized monoclo- that EZ lowers endo-capillary proliferation and inflam-
nal antibody directed against C5. EZ binds to the C5 comple- matory cell infiltration in some patients. These findings
ment protein and inhibits its cleavage into pro-inflammatory are in accordance with animal data which showed that
and pro-thrombotic C5a and C5b. This binding prevents the anti-C5 therapy effectively targeted the inflammatory
Table 18.5 Rituximab for C3 glomerulopathy.
Author Patients, n
(reference year) (age, years) Kidney RTX, protocol Other therapy Outcome Diagnosis
Daina et al. (2012) 1 (22) Native 375 mg/m2 ×1 Steroids No effect of RTX on renal DDD
Eculizumab function or proteinuria PR
RAS blockade with eculizumab
Rousset-Rouvière 1 (8) Native 375 mg/m2 ×2 Steroids Acute renal failure after RTX DDD
et al. (2014) MMF CR with eculizumab
Eculizumab
RAS blockade
Giaime 1 (34) Native 700 mg ×4 weekly RAS blockade Remission of proteinuria with DDD
et al. (2015) (repeated after 18 mo) RTX
Payette 1 (5) Native 375 mg/m2 ×4 Steroids No effect of RTX on proteinuria DDD
et al. (2015) MMF PR with eculizumab
Eculizumab
CR, complete response; DDD, dense-deposit disease; MMF, mycophenolate mofetil; PR, partial response; RAS, ; RTX, rituximab.
Summar 283
Table 18.6 Eculizumab for C3 glomerulopathy in native and transplant kidneys (adults).
EZ treatment, Response to EZ
Author (reference year) Age C3G, type duration Outcome Kidney, type (histopathology)
McCaughan et al. (2012) 29 DDD (n = 1) 10 weeks SCr improved PCR Graft NA
improved
Daina et al. (2012) 22 DDD (n = 1) 12 months SCr improved Native NA
PCR improved
Bomback et al. (2012) 27.2 (20–42) DDD (n = 3) 12 months SCr improved (n = 2) Graft (n = 3) Improved (3/5)
C3GN (n = 3) Proteinuria improved Native (n = 3) Worsened (2/5)
(n = 1)
Gurkan et al. (2013) 21 C3GN (n = 1) 53 weeks SCr improved Graft Worsened
PCR improved
Inman et al. (2015) 38 C3GN (n = 1) >5 months SCr improved Native NA
PCR improved
Welte et al. (2018) 29 (14–59) C3GN (n = 5) 19 (3–28) SCr improved (n = 2) Graft (n = 3) Worsened (2/2)
DDD (n = 2) PCR improved (n = 3) Native (n = 4)
Sahin et al. (2018) 53 C3GN (n = 1) 9 months SCR improved Graft NA
PCR improved
Moog et al. (2018) 59 DDD (n = 1) 12 months SCR improved Graft Worsened
PCR improved
Garg et al. (2018) 51 C3GN (n = 1) 3 years SCR improved Graft Improved
PCR improved
Kaartinen et al. (2018) 37 C3GN (n = 1) 6 months Poor response Graft Worsened
Le Quintrec et al. (2018) 43 (18–65) C3G (n = 13) 8 (2–51) Global clinical Graft (n = 3) Improved
months response (n = 4) Native (n = 4)
Partial clinical
response (n = 1)
No response (n = 8)
C3G, C3 glomerulopathy; C3GN, C3 glomerulonephritis; DDD, dense-deposit disease; EZ, eculizumab; NA, not applicable; PCR, urine protein/
creatinine ratio; SCr, serum creatinine.
component of C3 glomerulopathy [80]. The current peptide) are under way and promise to change more radi-
approach to C3G is mostly based on case studies and cally than EZ the management of C3G.
expert opinion; trial data are currently lacking. Long-term
use of EZ is required in this disease due to the inherent
risk of recurrence but this hampered by the cost burden S
ummary
(use of EZ for 1 year has been associated with an approxi-
mate cost of US$600 000). Recent advances have been made in the last decade regard-
The pathophysiology of C3G remains unclear and is ing the understanding of the multiple pathogenetic path-
probably more complex than that of aHUS. The levels of ways of MPGN. This has given us the possibility to expand
complement dysregulation at the site of the C3 and C5 the therapeutic armamentarium against MPGN. However,
convertase have great variability and this could explain at this time there is insufficient high-quality evidence
the heterogeneity in response to EZ therapy in patients regarding effective and safe therapies for MPGN. Idiopathic
with C3G. Patients with C3G who show important activa- or primary immune-complex MPGN is clearly an uncom-
tion of C5 convertase will benefit from therapy with EZ; mon condition and an extensive evaluation is required to
on the contrary, patients with C3G who have significant exclude the etiologies reported above. The low frequency of
dysregulation at the level of C3 convertase will not show MPGN could at least partially explain the paucity of con-
dramatic benefit. trolled trials on this issue.
Some clinical trials using targeted complement inhibition Adults with idiopathic immune-complex MPGN show-
with novel agents (i.e. compstatin, which is a C3 inhibitory ing nephrotic syndrome and progressive decline of kidney
function should receive an oral course of cyclophospha- We need prospective trials or cohort studies to assess the
mide, or MMF, or oral/intravenous corticosteroids. The ini- efficacy of EZ and its superiority over conventional
tial therapy should be limited to less than 6 months approaches (i.e. cyclophosphamide, corticosteroids, and
(evidence level, 2C). MMF). In patients with C3G due to genetic defects, fresh
The management of C3 glomerulopathy is not based on frozen plasma is indicated to replace the insufficient plasma
RCTs as the diseases (C3GN and DDD) are very uncommon. concentration of CFH or any other related protein. Patients
Only small case series and case reports have been made. In with C3G and rapidly progressive GN should receive rescue
patients with C3 glomerulopathy who present mild disease, therapy (pulses of intravenous corticosteroids and oral
i.e. haematuria, low-grade proteinuria (<0.5 g/day), and cyclophosphamide or MMF). Methylprednisolone pulses
intact kidney function, management should include RAS (0.5–1 g/day) for 3 days followed by oral prednisone (1 mg/
blockade with ACEIs and ARBs (evidence level, 2D). Patients kg/day body weight and slow tapering over 2–3 months) and
with C3G and moderate or severe renal disease (greater levels cyclophosphamide. This approach is not specifically tar-
of proteinuria, nephrotic proteinuria, and/or kidney dys- geted to DDD or C3GN (evidence level, 2D).
function) should receive treatment according to the underly- Patients with HCV-induced MPGN showing stable kid-
ing etiology. Plasma exchange, rituximab, steroids, and ney function and/or non-nephrotic proteinuria should
eculizumab should be considered in patients with C3G receive initially antiviral therapy with DAAs. (evidence
where antibody-stimulating AP have been identified (evi- level, 1B) Patient with HCV-induced MPGN showing cryo-
dence level, 2C). Treatment with EZ allows targeted comple- globulinemic flare, nephrotic syndrome, or progressive kid-
ment inhibition and has considerably improved the ney failure should receive antiviral therapy with DAAs and
prognosis of some patients with C3G; however, the C5 concomitant immunosuppressive therapy (and/or plasma-
blockade does not seem to significantly alter the course of exchange) (evidence level, 1B). Rituximab should be con-
the disease in many cases. The therapeutic regimen considered the first-line immunosuppressive agent in patients
sisted of 900 mg EZ, given intravenously, weekly for 4 con- with HCV-related MPGN, even in those who do not respond
secutive weeks, and then 1200 mg/2 weeks for 12 months [68]. to antiviral therapy (evidence level, 1B).
R
eferences
Glassock, R. (1997). Membranoproliferative

1 7 Viganò, M., Martin, P., Cappelletti, M., and Fabrizi, F.
glomerulonephritis. In: Treatment of Primary (2014). HBV-associated cryoglobulinemic vasculitis:
Glomerulonephritis (eds. C. Ponticelli and remission after antiviral therapy with entecavir. Kidney
R. Glassock), 218–233. Oxford: Oxford Medical Blood Press Res. 39: 65–73.
Publications. 8 Roccatello, D., Fornasieri, A., Giachino, O. et al. (2007).
2 Sethi, S. and Fervenza, F. (2012). Membranoproliferative Multicenter study on hepatitis C virus-related
glomerulonephritis-a new look at an old entity. N. Engl. J. cryoglobulinemic glomerulonephritis. Am. J. Kidney Dis.
Med. 366: 1119–1131. 49: 69–82.
3 Lionaki, S., Gakiopoulou, H., and Boletis, J. (2016). 9 Fabrizi, F., Plaisier, E., Saadoun, D. et al. (2013). Hepatitis
Understanding the complement–mediated glomerular C virus infection, mixed cryoglobulinemia, and kidney
diseases: focus on membranoproliferative glomerulonephritis disease. Am. J. Kidney Dis. 61: 623–637.
and C3 glomerulonepathies. APMIS 124: 725–735. 10 Rault, R., Holley, J., Banner, B., and el-Shahawy, M.
4 Sanghera, P., Ghanta, M., Ozay, F. et al. (2017). Kidney (1992). Glomerulonephritis and non-Hodgkin’s
diseases associated with alternative complement pathway lymphoma: a report of two cases and review of the
dysregulation and potential treatment options. Am. J. Med. literature. Am. J. Kidney Dis. 20: 84–89.
Sci. 354: 533–538. 11 Rennke, H. (1995). Secondary membranoproliferative
5 West, C., Mc Adams, A., McConville, J. et al. (1963). glomerulonephritis. Kidney Int. 47: 643–656.
Hypocomplementemia and normocomplementemic 12 Ahmed, M., Solangi, K., Abbi, R., and Adler, S. (1997).
persistent (chronic) glomerulonephritis: clinical and Nephrotic syndrome, renal failure, and renal malignancy:
pathological characteristics. J. Pediatr. 67: 1089–1112. an unusual tumor-associated glomerulonephritis. J. Am.
6 Zand, I., Fervenza, F.C., Nasr, S., and Sethi, S. (2014). Soc. Nephrol. 8: 848–852.
Membranoproliferative glomerulonephritis associated with 13 Vedder, A., Weening, J., and Krediet, R. (2004).
autoimmune diseases. J. Nephrol. 27: 165–171. Intracapillary proliferative glomerulonephritis due to
Reference 285
heavy chain deposition disease. Nephrol. Dial. Transplant. mesangiocapillary (membranoproliferative)

19: 1302–1304. glomerulonephritis. Nephrol. Dial. Transplant. 19:
14 Nasr, H., Markowitz, G., Stokes, M. et al. (2004). 2769–2777.
Proliferative glomerulonephritis with monoclonal IgG 28 Rovin, B., Caster, D., Cattran, D. et al.; for Conference
deposits: a distinct entity mimicking immune-complex Participants (2019). Management and treatment of
glomerulonephritis. Kidney Int. 65: 85–96. glomerular diseases (part 2): conclusions from a Kidney
15 Tarantino, A., Campise, R., Banfi, G. et al. (1995). Disease: Improving Global Outcomes (KDIGO)
Long-term predictors of survival in essential mixed Controversies Conference. Kidney Int. 95: 281–295.
cryoglobulinemic glomerulonephritis. Kidney Int. 47: 29 Tiller, D., Clarkson, A., and Mathew, T. (1981). A
618–623. prospective randomized trial of the use of
16 Noris, M. and Remuzzi, G. (2017). Genetics of immune- cyclosphophamide, dipyridamole and warfarin in
mediated glomerular diseases: focus on complement. membranous and membranoproliferative
Semin. Nephrol. 37: 447–463. glomerulonephritis. In: Proceedings of the 8th
17 Servais, A., Noel, L., Roumenina, L. et al. (2012). International Congress of Nephrology (eds. R. Robinson,
Acquired and genetic complement abnormalities play a R. Glassock, C. Tisher, et al.), 345–351. Karger, Basel.
critical role in dense deposit disease and other C3 30 Zimmermann, S., Moorthy, A., Dreher, W. et al. (1983).
glomerulopathies. Kidney Int. 82: 454–464. Prospective trial of warfarin and dipyridamole in patients
18 Pickering, M., Cook, H., Warren, J. et al. (2002). with membranoproliferative glomerulonephritis. Am. J.
Uncontrolled C3 activation causes membranoproliferative Med. 75: 920–927.
glomerulonephritis in mice deficient in complement 31 Mota-Hernandez, F., Gordillo-Paniaqua, G., Munoz-
factor H. Nat. Genet. 31: 424–428. Arizpe, R. et al. (1985). Prednisone versus placebo in
19 Medjeral-Thomas, N., Malik, T., Patel, M. et al. (2014). A membranoproliferative glomerulonephritis: long-term
novel CHFR5 fusion protein causes C3 glomerulopathy in clinicopathological correlations. Int. J. Pediatr. Nephrol.
a family without Cypriot ancestry. Kidney Int. 85: 933–937. 6: 25–28.
20 Pickering, M., D’Agati, V., Nester, C. et al. (2013). C3 32 Zauner, I., Bohler, J., Braun, N. et al. (1994). Effect of
glomerulopathy: consensus report. Kidney Int. 84: aspirin and dypiridamole on proteinuria in idiopathic
1079–1089. membranoproliferative glomerulonephritis: a multicentre
21 Bomback, A., Santoriello, D., Avasare, R. et al. (2018). C3 prospective clinical trial. Collaborative
glomerulonephritis and dense deposit disease share a Glomerulonephritis Therapy Study Groups (CGTS).
similar disease course in a large United States cohort of Nephrol. Dial. Transplant. 9: 619–622.
patients with C3 glomerulopathy. Kidney Int. 93: 977–985. 33 Giri, S., Mahajan, S., Sen, R., and Sharma, A. (2002).
22 Medjeral-Thomas, N., O’Shaughnessy, M., O’Regan, J. Effects of angiotensin converting enzyme inhibitor on
et al. (2014). C3 glomerulopathy: clinicopathological renal function in patients of membranoproliferative
features and predictors of outcome. Clin. J. Am. Soc. glomerulonephritis with mild to moderate renal
Nephrol. 9: 46–53. insufficiency. J. Assoc. Physicians India 50: 1245–1249.
23 Cattran, D., Cardella, C., Roscoe, J. et al. (1985). Results 34 Donadio, J. and Offord, K. (1989). Reassessment of
of a controlled drug trial in membranoproliferative treatment results in membranoproliferative
glomerulonephritis. Kidney Int. 27: 436–441. glomerulonephritis, with emphasis on life-table analysis.
24 Tarshish, P., Bernstein, J., Tobin, J., and Edelmann, C. Am. J. Kidney Dis. 14: 445–451.
(1992). Treatment of mesangiocapillary 35 Schena, F. (1999). Primary glomerulonephritides with
glomerulonephritis with alternate-day prednisone—a nephrotic syndrome. J. Nephrol. 12 (Suppl. 2): S125–S130.
report of the International Study of Kidney Disease in 36 Levin, M. (2002). Mycophenolate mofetil treatment of
Children. Pediatr. Nephrol. 6: 123–130. primary glomerular diseases [letter]. Kidney Int. 621: 1475.
25 Donadio, J., Anderson, C., Mitchell, J. et al. (1984). 37 Choi, M., Eustace, J., Gimenez, L. et al. (2002).
Membranoproliferative glomerulonephritis. A Mycophenolate mofetil treatment of primary glomerular
prospective clinical trial of platelet-inhibitor therapy. N. diseases. Kidney Int. 61: 1098–1114.
Engl. J. Med. 310: 1421–1426. 38 Jones, G., Juszczak, M., Kingdon, E. et al. (2004).
26 Chan, M., Chan, K., Chan, P. et al. (1989). Adult-onset Treatment of idiopathic membranoproliferative
mesangiocapillary glomerulonephritis: a disease with a glomerulonephritis with mycophenolate mofetil and
poor prognosis. Q. J. Med. 72: 599–607. steroids. Nephrol. Dial. Transplant. 19: 160–164.
27 Cansick, J., Lennon, R., Cummins, C. et al. (2004). 39 Segarra, A., Amoedo, M., Martinez Garcia, J. et al. (2007).
Prognosis, treatment and outcome of childhood Efficacy and safety of ‘rescue therapy’ with
mycophenolate mofetil in resistant primary cryoglobulinemic membranoproliferative

glomerulonephritis-a multi center study. Nephrol. Dial. glomerulonephritis. Am. J. Kidney Dis. 43: 617–623.
Transplant. 22: 1351–1360. 53 Garini, G., Allegri, L., Iannuzzella, F. et al. (2007).
40 Yuan, M., Zou, J., Zhang, X. et al. (2010). Combination HCV-related cryoglobulinemic glomerulonephritis:
therapy with mycophenolate mofetil and prednisone in implications of antiviral and immunosoppressive therapy.
steroid-resistant idiopathic membranoproliferative Acta Biomed. 78: 51–59.
glomerulonephritis. Clin. Nephrol. 73: 354–359. 54 Sise, M., Bloom, K., Wisocky, J. et al. (2016). Treatment of
41 Mori, Y., Ihara, K., Yamaguchi, W. et al. (2013). A case of hepatitis C virus-associated mixed cryoglobulinemia with
combination therapy with MMF and steroids for direct-acting antiviral agents. Hepatology 63: 408–417.
idiopathic membranoproliferative glomerulonephritis. 55 Fabrizi, F., Aghemo, A., Lampertico, P. et al. (2018).
Nihon Jinzo Gakkai Shi 55: 177–184. Immunosuppressive and antiviral treatment of hepatitis
42 Koratala, A., Malpartida, F., Loy, J., and Zeng, X. (2018). C virus-associated glomerular disease: A long-term
Idiopathic membranoproliferative glomerulonephritis follow-up. Int. J. Artif. Organs 41: 306–318.
treated with mycophenolate mofetil. Clin. Case Rep. 6: 56 Cacoub, P., Vautier, M., Desbois, A. et al. (2017).
2287–2288. Effectiveness and cost of hepatitis C virus
43 Haddad, M., Lau, K., and Butani, L. (2007). Remission of cryoglobulinaemia vasculitis treatment: from
membranoproliferative glomerulonephritis type I with interferon-based to direct-acting antivirals era. Liver
the use of tacrolimus. Pediatr. Nephrol. 22: 1787–1791. Int. 37: 1805–1813.
44 Bagheri, N., Nemati, E., Rahbar, K. et al. (2008). 57 Reed, M., Alexander, G., Thiru, S., and Smith, K. (2001).
Cyclosporine in the treatment of membranoproliferative Hepatitis C-associated glomerulonephritis-a novel
glomerulonephritis. Arch. Iran Med. 11: 26–29. approach [letter]. Nephrol. Dial. Transplant. 16:
45 Wallace, E., Fogo, A., and Schulman, G. (2012). Imatinib 869–871.
therapy for non-infection related type II cryoglobulinemia 58 Colucci, G., Manno, C., Grandaliano, G., and Schena, F.
with membranoproliferative glomerulonephritis. Am. J. (2011). Cryoglobulinemic membranoproliferative
Kidney Dis. 59: 122–125. glomerulonephritis: beyond conventional therapy. Clin.
46 Sugiura, H., Takei, T., Itabashi, M. et al. (2011). Effect of Nephrol. 75: 374–379.
single-dose rituximab on primary glomerular diseases. 59 Roccatello, D., Sciascia, S., Baldovino, S. et al. (2016).
Nephron Clin. Pract. 117: c98–c105. Improved (4 plus 2) rituximab protocol for severe cases of
47 Dillon, J., Hladunewich, M., Haley, W. et al. (2012). mixed cryoglobulinemia: a 6-year observational study.
Rituximab therapy for type I membranoproliferative Am. J. Nephrol. 43: 251–260.
glomerulonephritis. Clin. Nephrol. 77: 290–295. 60 Pham, H., Feray, C., Samuel, D. et al. (1998). Effects of
48 Kong, W., Swaminathan, R., and Irish, A. (2013). Our ribavirin on hepatitis C-associated nephrotic syndrome
experience with rituximab therapy for adult-onset in four liver transplant recipients. Kidney Int. 54:
primary glomerulonephritis and review of the literature. 131–1319.
Int. Urol. Nephrol. 45: 795–802. 61 Daina, E., Noris, M., and Remuzzi, G. (2012). Eculizumab
49 Farooqui, M., Alsaad, K., Aloudah, N., and Alhamdan, H. in a patient with dense-deposit disease. N. Eng. J. Med.
(2015). Treatment-resistant recurrent 366: 1161–1163.
membranoproliferative glomerulonephritis in renal 62 Rousset-Rouviere, C., Cailliez, M., Garaix, F. et al. (2014).
allograft responding to rituximab: case report. Transplant. Rituximab fails where eculizumab restores renal function
Proc. 47: 823–826. in C3nef-related DDD. Ped. Nephrol. 29: 1107–1111.
50 Yango, A., Fischbach, B., Ruiz, R. et al. (2019). Treatment 63 Giaime, P., Daniel, L., and Burtey, S. (2015). Remission of
of recurrent, post-kidney transplant C3 glomerulopathy with rituximab as only
membranoproliferative glomerulonephritis with immunosuppressive therapy. Clin. Nephrol. 83: 57–60.
plasmapheresis and rituximab: a case report and 64 Payette, A., Patey, N., Dragon-Durey, M. et al. (2015). A
literature review. Clin. Nephrol. 91: 52–58. case of C3 glomerulonephritis successfully treated with
51 Fabrizi, F., Pozzi, C., Farina, M. et al. (1998). Hepatitis C eculizumab. Ped. Nephrol. 30: 1033–1037.
virus infection and acute or chronic glomerulonephritis: 65 Rudnicki, M. (2017). Rituximab for treatment of
an epidemiological and clinical appraisal. Nephrol. Dial. membranoproliferative glomerulonephritis and C3
Transplant. 13: 1991–1997. glomerulopathies. BioMed Res. Int. 2017: 2180508.
52 Alric, L., Plaisier, E., Thebault, S. et al. (2004). Influence 66 Rabasco, C., Cavero, T., Roman, E. et al., for the Spanish
of antiviral therapy in hepatitis C virus-associated Group for the Study of Glomerular Diseases (GLOSEN)
Reference 287
(2015). Effectiveness of mycophenolate mofetil in C3 glomerulopathy with eculizumab. Iranian J. Kidney Dis.
glomerulonephritis. Kidney Int. 88: 1153–1160. 12: 315–318.
67 Avarase, R., Canetta, P., Bomback, A. et al. (2018). 75 Moog, P., Jost, P., and Buttner-Herold, M. (2018).
Mycophenolate mofetil in combination with steroids for Eculizumab as salvage therapy for recurrent monoclonal
treatment of C3 glomerulopathy. A case series. Clin. J. gammopathy-induced C3 glomerulopathy in a kidney
Am. Soc. Nephrol. 13: 406–413. allograft. BMC Nephrol. 19: 106.
68 Caliskan, Y., Torun, E., Tiryaki, T. et al. (2017). 76 Garg, N., Zhang, Y., Nicholson-Weller, A. et al. (2018). C3
Immunosuppressive treatment in C3 glomerulopathy: is glomerulonephritis secondary to mutations in factor H
it really effective ? Am. J. Nephrol. 46: 96–107. and I: rapid recurrence in deceased donor kidney
69 McCaughan, J., O’Rourke, D., and Courtney, A. (2012). transplant effectively treated with eculizumab. Nephrol.
Recurrent dense deposit disease after renal Dial. Transplant. 33: 2260–2265.
transplantation: an emerging role for complementary 77 Kaartinen, K., Martola, L., Raisanen-Solokowski, A., and
therapies. Am. J. Transplant 12: 1046–1051. Meri, S. (2018). Recurrent allograft C3 glomerulonephritis
70 Bomback, A., Smith, R., Barile, G. et al. (2012). and unsuccessful eculizumab treatment. Clin. Immunol.
Eculizumab for dense deposit disease and C3 187: 104–106.
glomerulonephritis. Clin. J. Am. Soc. Nephrol. 7: 748–756. 78 Le Quintrec, M., Lapeyraque, A., Lionet, A. et al. (2018).
71 Gurkan, S., Fyfe, B., Weiss, L. et al. (2013). Eculizumab Patterns of clinical response to eculizumab in patients
and recurrent C3 glomerulonephritis. Pediatr. Nephrol. with C3 glomerulopathy. Am. J. Kidney Dis. 72: 84–92.
28: 1975–1981. 79 Zuber, J., Fakhouri, F., Roumenina, L. et al. (2012). On
72 Inman, M., Prater, G., Fatima, H., and Wallace, E. (2015). behalf of the French study group for aHUS/C3G. Use of
Eculizumab-induced reversal of dialysis-dependent eculizumab for atypical uraemic syndrome and C3
kidney failure from C3 glomerulonephritis. Clin. Kidney J. glomerulopathies. Nat. Rev. Nephrol. 8: 643–657.
8: 445–448. 80 Pickering, M., Warren, J., Rose, K. et al. (2006).
73 Welte, T., Arnold, F., Kappes, J. et al. (2018). Treating C3 Prevention of C5 activation ameliorates spontaneous and
glomerulopathy with eculizumab. BMC Nephrol. 19: 7. experimental glomerulonephritis in factor H-deficient
74 Sahin, H., Oguz, E., Akoglu, H. et al. (2018). Successful mice. Proc. Natl. Acad. Sci. USA 103: 9649–9654.
treatment of posttransplant recurrent complement C3
288
19
Autosomal Dominant Polycystic Kidney Disease

Gopala Rangan1,2, Annette Wong1,2, Alexandra Munt1,2, Irene Sangadi1,2, Sayan Saravanabavan1,2,
and Jennifer Zhang1,2
1
Department of Renal Medicine, Westmead Hospital, Western Sydney Local Health District, Sydney, Australia
2
Michael Stern Laboratory for Polycystic Kidney Disease, Westmead Institute for Medical Research, University of Sydney, Sydney, Australia
I ntroduction was 9.3 cases per 10 000 cases in two large population
sequencing databases [17]. Taken together, these data indi-
Autosomal dominant polycystic kidney disease (ADPKD) cate that the true incidence of ADPKD in the general popu-
is a progressive chronic kidney disease (CKD) that in ~90% lation is probably 1:1000 but that the overall incidence of
of cases has four features: (i) autosomal dominant pattern symptomatic ADPKD leading to end-stage kidney disease
of inheritance in first-degree relatives; (ii) mutation in one (ESKD) is approximately 1:2500.
of the causative genes (PKD1, PKD2 or others); (iii) varia-
ble expressivity but complete age-dependent penetrance of
multiple renal cysts that fulfill the Pei-Ravine ultrasound
P
athophysiology
criteria; and (iv) variable systemic complications (extra-
ADPKD is due to heterozygous germ-line mutations in
renal cysts in solid organs, vascular aneurysms, connective
PKD1 (85% of cases, located on chromosome 16p13.3, 46
tissue defects, such as hernia) [1–11] (Table 19.1).
exon-length), PKD2 (15%, located on chromosome 4q21,
15 exon-length), or rarely additional genes (<1:1000, such
as GANAB or DNAJB11). Approximately 10% of patients
E
pidemiology
have no mutation detected in either PKD1 or PKD2 (no
mutation detected). The mutations are inherited in an
Incidence and Prevalence of ADPKD in the
autosomal dominant manner but occasionally may
General Population
involve other complex genetic mechanisms (e.g. mosai-
ADPKD occurs uniformly throughout the world. Disease cism or bi-allelic transmission of PKD1). PKD1 and PKD2
mutations occurred early in modern human evolution. encode two cellular membrane proteins: polycystin-1 and
Determining the prevalence of ADPKD is difficult as the polycystin-2, respectively. The formation of the kidney
condition is clinically silent in up to 50% of patients. The cysts starts in early life due to random postnatal reduc-
Danish and Olmstead cohort studies (which included tions in PKD1 or PKD2 dose in the principal cells of the
autopsy specimens) performed in the 1950s and 1980s, collecting duct, as a result of loss of heterozygosity (LOH,
respectively, showed that the lifetime risk in the general pop- “second hits”) or other poorly defined stochastic factors.
ulation was 1:1000, with an annualized incidence of 2.75 per In cohort studies, mutations in PKD1 are associated with
100 000 person-years [12, 13]. More recent studies performed more severe renal disease than PKD2, as defined by an
in selected geographic areas in the 2010s suggested a lower earlier onset of ESKD (50–55 vs. 75–80 years old) and
prevalence (4.76 per 10 000, Modena, Italy [14], 3.96 per twice the number of renal cysts. The clinical phenotype of
10 000 European Union [15], 5.73 per 10 000, southwest ADPKD has high intra- and interfamilial heterogeneity
Germany [16]). An analysis performed in 2018 showed that due to environmental (epigenetic) mechanisms and mod-
the frequency of high-confidence mutations in PKD1/PKD2 ifying genes [1].
Pathophysiolog 289
Table 19.1 Treatment of ADPKD.
Diagnosis We recommend confirming diagnosis using the Pei-Ravine ultrasound criteria. High Strong
We recommend genetic testing in selected clinical circumstances and in High Strong [5]
consultation with a clinical geneticist.
Assessment of We recommend evaluating factors associated with high risk for ESKD in an Moderate Strong
renal prognosis individual patient to guide patient education, frequency of clinical follow-up, and
treatment.
Genetic We suggest referral to a clinical geneticist and genetic counselor if detailed advice Low Weak [5]
counseling on family planning and pre-implantation genetic diagnosis is sought by patients
and/or if genetic testing is being considered to inform diagnosis.
Screening We recommend screening all first-degree adults at risk of ADPKD with renal High Strong
at-risk of ultrasound.
ADPKD We suggest that parents be informed of the advantages/disadvantages of screening Low Weak
by renal ultrasound for ADPKD in children <18 years old who are at risk (50%) of
ADPKD.
Fluid intake We suggest consuming fluids as advised by the treating physician. Low Weak [6]
Blood pressure All patients with ADPKD should receive either an ACEI or ARB for first-line High Strong [7]
lowering pharmacological treatment of hypertension (>130/80 mmHg) and the stepwise use
of a α/β cardio-selective blocker, dihydropyridine calcium channel blocker or a
diuretic either as second-line agents or in combination with first-line drugs.
Disease- We do not recommend the use of somatostatin analogs or target of rapamycin High Weak [7]
modifying complex 1 inhibitors, and to use shared decision-making regarding the use of
treatments vasopressin receptor antagonists in patients with “rapid progression” as defined by
a historical decline in the eGFR.
Chronic kidney We suggest that chronic pain is evaluated during clinic visits and that management Low Weak [8]
pain should be stepwise, involving nonpharmacological and pharmacological
treatments, and minimally invasive and invasive therapies.
Kidney cyst We suggest that suspected renal cyst infection is treated with empirical antibiotic Low Weak
infection therapy and renal cyst aspiration if refractory to treatment.
Kidney cyst We suggest that renal cyst hemorrhage be treated with bed rest and analgesia as Low Weak
hemorrhage first-line treatment, followed by minimally invasive or invasive therapies if bleeding
is refractory.
End-stage We recommend that patients with ESKD be considered for kidney transplantation High High [9]
kidney disease and evaluated for need for nephrectomy prior to transplantation.
Polycystic liver We suggest that severe polycystic liver disease (total liver volume >1800 ml/m) is Low Weak [10]
disease managed by a multidisciplinary team (hepatologist, hepatobiliary surgeon,
interventional radiologist, nephrologist).
Intracranial We recommend that MR angiography or CT (circle of Willis angiography) be used Moderate Weak [11]
aneurysm for screening and detection of ICAs in asymptomatic individuals at high risk for an
ICA (i.e. a positive family history of SAH/unruptured ICA in at least one first-
degree relative with ADPKD or previous SAH/ICA).
We recommend urgent intracranial imaging in patients who experience a sudden Very low Weak [11]
onset of severe headache or neurological symptoms of concern.
We recommend referral to a neurosurgeon if an ICA is detected, and suggest Very low Weak [11]
treatment in centers which have expertise in endovascular coiling and microsurgery.
GRADE assessment of the certainty of the evidence [12]: High: This research provides a very good indication of the likely effect. The likelihood
ACEI, angiotensin converting enzyme inhibitor; ADPKD, autosomal dominant polycystic kidney disease; ARB, angiotensin receptor blocker;
CT, computed tomography; eGFR, estimated glomerular filtration rate; ESKD, end-stage renal disease ICA, intracranial aneurysm; MR,
magnetic resonance; SAH, subarachnoid hemorrhage.
290 Autosomal Dominant Polycystic Kidney Disease
The growth of kidney cysts is slow, exponential, and has [18, 19]. Renal ultrasound is the preferred imaging modal-
interindividual variability (~2–10% per year) leading to kid- ity as it is safe, reliable, inexpensive, and easily accessible.
ney enlargement by mid-life (mean 1 kg/kidney vs. 0.2 kg/ The Pei-Ravine criteria use kidney cyst number according
kidney in health). The expanding kidney cysts compress and to age (Table 19.2). If the cyst number fails to meet Pei-
destroy normal kidney tissue, causing chronic pain, hyper- Ravine criteria (“indeterminate”), a follow-up ultrasound
tension, and renal impairment. The growth of renal cysts is in 1–2 years is suggested [19].
mediated by chloride-driven fluid secretion (via the cystic
fibrosis transmembrane regulator [CFTR] chloride channel)
Renal Ultrasound in the Absence of a Family
and proliferation (via multiple signal transduction path-
History
ways, including TORC1). in vitro arginine vasopressin (AVP)
increases intracellular cyclic adenosine monophosphate In the absence of a positive family history, there are no
(cAMP), CFTR-mediated trans-epithelial fluid secretion, evidenced-based ultrasound criteria to make a diagnosis of
and proliferation of cyst-forming renal epithelial (Madin– ADPKD but expert opinion is that at least 10 renal cysts in
Darby canine kidney cells [MDCK] cells), implicating it as each kidney are required [19–23]. In this situation or if the
an important nongenetic factor in exacerbating renal cyst ultrasound is indeterminate, other supporting evidence from
growth during the postnatal period. Thus, disease-modifying the history, physical examination, and ultrasound features
treatments to inhibit proliferation (e.g. sirolimus, a TORC1 (such as the presence of hepatic cysts and increased kidney
inhibitor; metformin, an AMP-activated protein kinase size) might provide evidence to support the diagnosis [19].
[AMPK] inhibitor), “kidney cyst-specific growth factors”
(e.g. vasopressin), and other pathways have been investi-
Genetic Testing
gated in clinical trials [1].
We do not recommend universal genetic testing in all
patients with ADPKD using current technologies (high
D
iagnosis certainty of evidence, strong recommendation) [20]
because (i) the results are inconclusive or negative in 10%
Renal Ultrasound in Patients with a Positive
of patients with typical ADPKD or as high as 50% with a
Family History
clinical phenotype that is atypical [21]; (ii) genetic testing
We recommend confirming the diagnosis with renal ultra- is 10 times more expensive than kidney ultrasound
sound using the Pei-Ravine ultrasound criteria in all [18–21]; and (iii) genotyping does not correlate with phe-
patients at risk of ADPKD (positive family history of auto- notypic severity. We recommend restricting the use of
somal dominant pattern of inheritance in first-degree fam- genetic testing use to specific clinical situations in consul-
ily members) (high certainty, strong recommendation) tation with a clinical geneticist (high certainty, strong
Table 19.2 Pei-Ravine’s criteria for the diagnosis of ADPKD by renal ultrasounda [18,19].
Positive family history Number of kidney cysts needed to make Number of kidney cysts to
Age in years (genotype unknown) the diagnosis of ADPKD exclude the diagnosis of ADPKD
15–29 Yes 3 (unilateral or bilateral); Normal ultrasound does not

PPV = 100%, sensitivity = 81.7% exclude diagnosis
<40 Yes 3 (unilateral or bilateral); Normal ultrasound does not
PPV = 100%, sensitivity = 82–96% exclude diagnosis
40–59 Yes 4 (at least two in each kidney); <2; NPV = 100%,
PPV = 100%, sensitivity = 90% specificity = 98.2%
60 Yes 8 (at least four in each kidney); <2
PPV = 100%, sensitivity = 100%
Any ageb No 10 cysts in each kidney with renal N/A
enlargement ± hepatic cysts
AKPKD, autosomal dominant polycystic kidney disease; NPV, negative predictive value; PPV, positive predictive value.
a
Criteria based on expert opinion.
b
Criteria based on the use of conventional 3–5 MHz ultrasound probe with cyst size typically being above 1 cm in diameter. CT, MRI, and more
sensitive ultrasound probes detect smaller cysts and therefore the above criteria are not applicable to these imaging modalities.
Source: Based on Mai et al. [19].
Prognosi 291
recommendation): (i) the exclusion of disease in poten- Table 19.4 Risk factors associated with high risk for ESKD.
tial living related kidney donors under the age of 40 years
old (as a normal renal ultrasound and computed tomog- Clinical factors
raphy [CT] scan does not exclude ADPKD in this set- Age >35 years old
ting) (Table 19.3); (ii) during pre-implantation genetic Family history of ESKD
diagnosis [24]; and/or (iii) in the evaluation of a patient Age of at least one family member with ESKD <55 years old
(suggest mutation in PKD1) (vs. >70 in PKD2)
with an atypical clinical phenotype (i.e. negative family
Onset of hypertension<35 years old
history, equivocal renal imaging, and unusual clinical
Episode of hematuria <30 years old
phenotype, especially in syndromic forms or early-onset
Laboratory investigators
disease) [20].
Albuminuria
Renal function
Method of Genetic Testing in ADPKD The gold-standard for ●● Baseline serum creatinine>92 μmol/l
genetic testing in ADPKD has been long-range polymerase ●● Stage 2 CKD by 30 years old, stage 3 by 50 years old or stage
chain reaction (LR-PCR) amplification followed by Sanger 4/5 or ESKD by 55 years old
sequencing of all exons. However, this method is labor- ●● Annual rate of decline in eGFR >5 ml/min in 1 year or
intensive and is being superseded by next-generation >2.5 ml/min/year over 5 years

sequencing (whole-genome sequencing, targeted exome Imaging
sequencing) [20]. Renal length on ultrasound >16.5 cm
CKD, chronic kidney disease; eGFR, estimated glomerular filtration
rate; ESKD, end-stage kidney disease; PKD1, polycystic kidney
disease gene-1; PKD2, polycystic kidney disease gene-2.
P
rognosis
We recommend evaluating factors associated with high development and not validated; and (iv) many novel
risk for ESKD in an individual patient to guide patient molecular biomarkers (e.g. urinary monocyte chemoat-
education, frequency of clinical follow-up, and treatment tractant protein-1, copeptin) are under investigation, but
(Table 19.4) (moderate certainty of evidence, strong rec- none has superseded standard methods of clinical history,
ommendation). However, the certainty of estimating risk estimated glomerular filtration rate (eGFR), and imaging.
in an individual is low because (i) there is wide clinical
heterogeneity in the severity of kidney disease (even
Clinical History
within the same family); (ii) the accuracy of current meth-
ods to estimate the risk of development of ESKD are poor ●● Age: Patients older than 35 years old at presentation have
and based on a limited prospective cohort studies; (iii) a 1.7-fold increased risk of developing CKD stage 3A
disease-specific imaging biomarkers (Ht-TKV by magnetic (sensitivity 0.60 and specificity 0.66) [25].
resonance imaging [MRI]) used in research are not avail- ●● Family history: A three-generation family pedigree is rec-
able in clinical practice and statistical prediction models ommended in all ADPKD patients: (i) confirm the auto-
(Mayo Subclassification for TKV) are in the early stages of somal dominant pattern of inheritance; (ii) determine
Table 19.3 Summary of diagnostic tests in ADPKD.
Test What does it measure? Comments
Pei-Ravine ultrasound ●● Determines the number of renal cysts ●● Applies only if there is a family history of ADPKD in a
criteria based on age in a population at 50% risk of first-degree relative
having ADPKD ●● Criteria in the absence of family history are based on
●● Validated by genetic testing expert opinion
●● Simple and cost-effective ●● Negative ultrasound excludes ADPKD only after the age
●● May be indeterminate in up to 10% of cases of 40 (important to recall this fact during live-kidney
donor assessment involving family members)
Genetic test ●● Provides direct information of the mutation ●● Diagnostic accuracy in enriched populations is 90% but
type (PKD1 vs. PKD2 vs. no mutation may decline to 50% in atypical or real-world cases
detected or other genes) ●● Analysis is expensive, complex, and time-consuming but
●● May be indeterminate in up to 10–50% of likely to be more useful as workflow and technological
cases depending on the clinical history improvements are reported
the history of ESKD in the family (and age of onset in ●● Clinical events: The onset of hypertension or a urological
family members); (iii) identify other family members event (episode of cyst hemorrhage, cyst infection, and/or
who are at-risk for ADPKD; and (iv) screen for a fam- acute flank pain) below 35 years of age is associated with
ily history of intracranial aneurysms (ICAs). a truncating mutation in PKD1 and with a worse renal
Specifically, a family history of onset ESKD 58 years prognosis [27,28].
or >68 years predicts a mutation in PKD1 (PPV 100%, ●● Perinatal history: Higher birth weight is protective, with
sensitivity 75%) or PKD2 (PPV 100%, sensitivity 78%), every kilogram at birth deferring the age of ESKD onset
respectively [26]. by 1.7 years [29] (Table 19.5).
Table 19.5 Clinical value of methods used to predict renal prognosis in individual patients.
Prognostic test Bias Precision Benefits Harms and cost Value
Clinical information High High Information easy to obtain but Low in harm and cost Low-moderate
(Age, age of ESKD in has low sensitivity and
family, birth weight) specificity
Tests
Genetic testing High High Identifies the causative mutation Low in harm but very high in Low
(Sanger) (PKD1 PKD2/NMD) but not cost
useful for determining prognosis
in an individual patient
Single measurement of Moderate High Estimates the number of kidney Not all patients able to do Moderate
total kidney volume at cysts and is a disease-specific MRI, moderate in cost, and
baseline corrected by marker that can identify early not validated in all population
height risk for ESKD groups (e.g. by ethnicity,
country)
Single measurement of Moderate Moderate Estimates the number of kidney Low in cost but not validated Moderate–high
renal length by cysts and is a disease-specific in all population groups (e.g.
ultrasound at baseline marker that can identify early by ethnicity, country)
risk for ESKD
Serial measurement of Moderate High Measures actual rate of kidney Intervals to measure standard; Low
Ht-TKV by MRI cyst growth not all patients suitable for
MRI, moderate in cost and not
validated in all groups
Single eGFR at baseline Low Moderate Estimates GFR but not a Low in harm and cost High
disease-specific marker and does
not reflect number of
functioning nephrons due to
glomerular hyperfiltration
Serial eGFR changes Moderate Low Estimates the rate of change in Useful for late stage of disease Low in stage 1
eGFR but insensitive in early and not validated CKD and
stage of disease moderate-high in
CKD stages 2–4
Statistical prediction models
Mayo subclass Moderate High Estimates the rate of kidney cyst Not all patients able to do Moderate
(Irazabal) class at growth based on Ht-TKV and MRI, moderate in cost and not
baseline age at the time of the test, and validated in all populations
therefore subclassifies groups and for ESKD
PRO-PKD score Moderate High Composite score based on Genetic testing not performed Low
clinical events and genetic and not validated in all
testing populations
CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; ESKD, end-stage kidney disease; GFR, glomerular filtration rate;
Ht-TKV, height-adjusted total kidney volume; MRI, magnetic resonance imaging; NMD, no mutation detected; PKD1, polycystic kidney disease
gene-1; PKD2, polycystic kidney disease gene-2; PRO-PKD, predicting renal outcomes in autosomal dominant polycystic kidney disease.
Prognosi 293
Investigations Imaging: Total kidney volume (TKV) by MRI is a disease-

specific biomarker useful for evaluating progression in
Laboratory and imaging tests. The performance character-
clinical trials of patients with preserved renal function. In
istics of tests for renal function (serum creatinine and urea;
contrast, in clinical practice, the value of TKV over annual
urine albumin to creatinine ratio) and kidney volume
eGFR measurement in predicting renal prognosis is uncer-
(Ht-TKV by MRI and renal length by ultrasound) to esti-
tain. Current evidence is limited to a single longitudinal
mate the 8-year risk of developing CKD stage 3a are shown
cohort study where baseline height corrected-TKV>650
in Table 19.6 [25]. These data are derived from longitudinal
ml/m only marginally outperformed the baseline serum
follow-up of a prospective cohort of 241 ADPKD patients
creatinine (97 μmmol/L) in predicting CKD stage 3A over
who had an initial creatinine clearance >70 ml/min aged
8 years (Table 19.6) [30]. However, as MRI is not reim-
between 15 and 45 years [25]. In this cohort 30.7% of
bursed in many countries, longitudinal kidney length on
patients reached the study CKD3A during follow-up
ultrasound provides a simple parameter that can be used in
period. These data indicate that baseline serum creatinine
clinical practice. In this regard, in the same cohort study a
has similar performance characteristics to kidney size (as
kidney length greater than 16.5 cm on renal ultrasound had
determined either by MRI or ultrasound) [30].
a similar performance to Ht-TKV in predicting the develop-
eGFR and serum creatinine: The precision of eGFR is
ment of 8-year risk of CKD stage 3A (Table 19.6) [30].
strongly correlated with measured glomerular filtration rate
(GFR) and predicts the subsequent decline in renal func-
tion [31, 32]. The decline in renal function is estimated by the Statistical Prediction Models
rate of change in eGFR using the CKD-EPI formula in serial
Currently two models have been developed: (i) the Irazabal
measurements. ADPKD patients typically display decades of
(Mayo) subclassification [35] and the PRO-PKD score [28]
stable or mildly abnormal GFR until mid to late adulthood
(Tables 19.7 and 19.8). The Mayo subclassification might be
[32]. Once abnormal and compared to other forms of CKD,
useful for prognostic enrichment in clinical trials. However,
the decline in renal function can be rapid (up to 3–4 ml/
in clinical practice, barriers exist to applying both of these
min/1.73 m2) per year depending on kidney enlargement [32].
Lack of validity to define “rapid progressor.” The availability
of tolvaptan in some regions has forced the use of decline in Table 19.7 Irazabal (Mayo) subclassification [35].
eGFR to define patients with “rapid” rates of progression [33].
The latter has been defined as the presence of eGFR <90 ml/ Scoring method (Ht-TKV divided by age − 150 ml) × 100
min/1.73 m2 with either (i) decline in eGFR of >5 ml/

Mayo Estimated kidney growth rates/
min/1.73 m2 in 1 year; or (ii) decline in eGFR of >2.5 ml/
subclass year (%) Risk for ESKD
min/1.73 m2 per year over 5 years [33]. However, the clinical
rationale and effectiveness of this definition to categorize A <1.5 Low
patients in this manner is not supported by long-term out- B 1.5–3.0 Intermediate
come data. Alternatively, Blanchette [34] (in a retrospective C 3.0–4.5 Intermediate
observational cohort analysis; n = 8018), defined rapid pro-
D 4.5%–6.0 High
gression as if one or more of the following were present:
E >6.0 High
hypertension <35 years old, hematuria <30 year, albuminu-
ria; stage 2 CKD by age 30; stage 3 by age 50, and/or stage 4/5 ESKD, Ht-TKV
or ESKD by 55 year (50% vs. 75% 5-year renal survival). Source: Based on Irazabal [35].
Table 19.6 Eight-year risk of developing CKD stage 3a.
Variable AUC (95% CI) Sensitivity Specificity Cut-off point LR of positive test
Serum Cr 0.75 (0.67–0.82) 0.58 0.81 97 μmol/l 3.05

Serum urea 0.76 (0.70–0.83) 0.63 0.79 5.7 mmol/l 3.0
Urine albumin 0.70 (0.61–0.78) 0.66 0.67 30 mg/day 2.0
Ht-TKV (MRI) 0.84 (0.79–0.90) 0.74 0.70 600 ml/m 2.96
Kidney length (US) 0.87 0.795 0.732 630 ml/m 2.97
AUC, area under the curve; Cr, HtTKV, LR, likelihood ratio; MRI, magnetic resonance imaging.
Source: Adapted from [25, 30].
Table 19.8 PRO-PKD score [28].
Scoring method Score interpretation
Parameter Score Score Risk for ESKD <60 years Median age of ESKD
Male 1 point 0–3 Low (NPV = 81.4%) 70.6

Hypertension <35 years old 2 points 4–6 Intermediate 56.9
First urologic event <35 years old 2 points 7–9 High (PPV = 90.9%) 49
Genotype
●● PKD2 mutation 0 point
●● PKD1 mutation (non-truncating) 2 point
●● PKD1 mutation (truncating) 4 points
ESKD, end-stage kidney disease; NPV, negative predictive value; PKD1, polycystic kidney disease gene-1; PKD2, polycystic kidney disease
gene-2; PPV, positive predictive value; PRO-PKD, predicting renal outcomes in autosomal dominant polycystic kidney disease.
Source: Cornec-Le Gall [28].
models as (i) the tests incorporated in the models (Ht-TKV their circumstances. The counseling should include a
for Mayo subclassification and genetic testing for PRO- discussion on the mode of inheritance (as well as com-
PKD) are not part of routine clinical practice; and (ii) the municating this to at-risk family members); the role,
models have not been validated prospectively for the devel- indication, and interpretation of diagnostic testing;
opment of ESKD or in real-world populations. information regarding the natural history, prognosis,
and treatment; and options regarding family planning
and prenatal testing options (including preimplanta-
T
reatment
tion genetic diagnosis). Referral to specialized genetic
services and a genetic counselor can be offered to
Figure 19.1 provides an overview of the management
enhance the care, especially if molecular genetic test-
approach in a patient with ADPKD. The key goals of care in
ing is considered and/or family planning anticipated
ADPKD are to preserve quality of life (QoL) and normal life
(such as preimplantation genetic diagnosis low cer-
expectancy by (i) reducing the lifetime risk of ESKD; (ii) pre-
tainty, weak recommendation). Patients should be
venting and/or minimizing cardiovascular disease; and (iii)
informed about patient support organizations for PKD
treating and managing extrarenal disease manifestations. The
as well as centers where clinical trials in PKD or other
management of ADPKD should be “patient-centered” (i.e.
novel treatments are being conducted.
care in which patients and their families are active partners in
decision-making by being provided with clear and accurate ●● Screening of adults and children at-risk of ADPKD:
knowledge regarding their situation) and “personalized” (i.e. Screening of first-degree adult relatives is recom-
“a disciplined approach to delivering care that responds to mended [36]. Renal ultrasound is the first-line screen-
patients’ circumstances, capabilities, and preferences”). ing test, and patients should receive the appropriate
counseling regarding the advantages and disadvantages
prior to performing predictive testing (high certainty,
Genetic and Psychosocial Care
strong recommendation) (Table 19.9). The optimal
The provision of coordinated genetic, medical, and psycho- approach for screening asymptomatic at-risk children
social care should be the overarching goal of care [27]. Due (<18 years of age) is unknown but the decision should
to the systemic nature of disease manifestations, multidis- be individualized and considered with parental discus-
ciplinary care is typical for most patients with ADPKD and sion (low certainty, weak recommendation) [37]. The
is tailored according to local health services, but the opti- historical view has been to defer screening children
mal model of delivery is not known. until adulthood but this view must be challenged [37,
38] due to the following emerging evidence: (i) at least
Genetic Care one renal cyst may be detected in up to 80% of children
●● Genetic counseling: As there is high intra-and interindi- who are at-risk of ADPKD [39]; (ii) hypertension is pre-
vidual variability in ADPKD, it is important that all sent in up to one-third of children [39] and a systematic
patients receive education that is individualized to suit review showed that ~8% reduced have renal function
Treatmen 295
1. Provide Comprehensive and Multi-disciplinary Genetic and Psychosocial Care
2. Assess if at high-risk for 3. Reduce kidney cyst 4. Evaluate for other

end-stage kidney disease: growth and prevent eGFR kidney complications:
1. Family history of ESKD<58 yrs
decline and hypertension: (i) Plan for ESKD (if eGFR<30 ml/min.1.73m2)
(ii) Onset of hypertension<35 yrs (i) Review lifestyle factors: (ii) Assess for chronic kidney pain
Renal Care
(iii) Hx of urological event<35 yrs Smoking cessation/avoidance (iii) If acute kidney pain, consider:
(iv) Stage 2–3 CKD with declining eGFR: Salt restriction (80–100 m mol/d) Cyst infection or cyst haemorrhage
>5 ml/min/1.73 m2 in 1 yr OR Moderate protein intake (0.75–1.0 g/kg/d) Nephrolithiasis
>2.5 ml/min/1.73 m2/yr over 5 yrs Maintain BMI<25 kg/m2 (diet+exercise) Urinary tract infection
(v) Increased kidney size: (ii) Maintain BP±130/80 with ACEI/ARB

Renal US length>16.5 cm OR (iii) Total cholesterol<4.0 with diet±drug
TKV>650 ml/m OR (iv) HbA1c<53 mmol/mol (7.0%; if diabetic)
Mayo Subclass 1C-E (v) Avoid nephrotoxic drugs and AKI 5. Evaluate risk-benefit of
(vi) Mutation in PKD1 (if known) tolvaptan if high-risk of ESKD
Other Care
6. Evaluate for extra-renal complications

(i) Evaluate for the presence of extra-renal complications
7. Consider referral to a Clinical Trial
(ii) Discuss the screening of ICAs
(iii) Screen for hepatic cysts/assess for polycystic liver disease
Figure 19.1 Overview of the treatment of ADPKD. Source: Adapted and modified from Rangan et al. [1].
Table 19.9 Benefits and disadvantages of screening adults pattern of inheritance and risks during pregnancy. The
at-risk for ADPKD. perinatal risks of pregnancy to a mother with ADPKD
are dependent on the level of kidney dysfunction and/or
Benefits of screening adults Disadvantages of screening hypertension. Young families with ADPKD are increas-
at-risk for ADPKD adults at-risk for ADPKD
ingly interested in PGD. In a survey of 96 UK patients
Reduces uncertainty and False negative result
with ADPKD, the majority (>95%) were concerned about
anxiety requiring re-testing the risk of genetic transmission [42] and most (>80%)
Diagnosis permits early Financial discrimination would consider using PGD [42]. PGD involves in vitro
treatment of complications Employment discrimination fertilization as well as genetic testing and therefore id a
Permits decisions regarding Changes in social and family costly procedure, but it is considered an option by
lifestyle and employment interactions patients with ADPKD with a family history of severe dis-
Permits discussions on Psychological impact of ease [24]. The application of PGD requires expert genetic
reproductive options diagnosis
counseling given the large phenotypic variability in
Allows family members to be
informed ADPKD and complexities in genetic testing. Patients
Identify potential unaffected should be referred to specialist centers with previous
living-related kidney donors experience and expertise in in vitro fertilization and
molecular diagnostics [24]. The long-term outcome of
PGD in ADPKD is unknown and will be informed by an
[40]; and (iii) subclinical manifestations of ADPKD
ongoing observational cohort study underway in China
begins in childhood, and early detection and treatment
(ESPERANCE trial; NCT02948179).
may slow renal cyst growth [41] with the potential to
eliminate the lifetime risk of ESKD and other extrarenal
Acute and Chronic Psychological Care
manifestations in high-risk families [22] [24].
The prevalence of psychological issues (anxiety, depres-
●● Family planning and preimplantation genetic diagnosis sion, denial, anger) in ADPKD is higher than in to the
(PGD): Couples with ADPKD should be informed of the general population [27, 43]. Acute psychological symptoms
are worst at the time of diagnosis when it comes as a Dietary protein intake: A subgroup analysis of ADPKD
“shock” to patients (some patients describe being told the patients (n = 141, Study A; GFR between 25 and 55 ml/
diagnosis as a receiving a “bomb”) [27, 43, 44]. The origins min1/1.73 m2) in the Modification of Renal Disease study
of emotional distress in ADPKD are complex but include showed that a low protein diet ( 0.6 g protein/kg/day) com-
the hereditary nature of the disease (with patients calling it pared to usual protein intake (1.3 g protein/kg/day) did not
an “evil heritage” and experiencing guilt from transmitting alter the decline in renal function (5.9 ml/min/year) [55].
the disease to future generations), the uncertainty of future Current evidence does not support any specific dietary pro-
prognosis regarding the kidney disease, chronic pain and tein intake for patients with ADPKD beyond standard rec-
lack of effective treatment options, worry about financial ommendations in CKD (no recommendation).
factors (future employment status and eligibility for life
insurance), impact on family planning, and the changes in Fluid Intake
body image due to kidney and liver enlargement [45–47]. We suggest consuming fluids whenever and in whatever
Psychological factors may negatively influence treatment volume desired or as advised by the treating physician
adherence due to a sense of helplessness and denial. The (low certainty of evidence, weak recommendation). The
evidence for psychological care in ADPKD is of low cer- volume of fluid that patients with ADPKD “should” drink
tainty but strongly recommended and includes (i) perform- is a topical question that has been asked by both neph-
ing an initial evaluation using screening questions; (ii) rologists and patients. It has been hypothesized that
determining the severity; and (iii) initiating appropriate maintaining a fluid intake of ~3 l per day may reduce the
management [48, 49]. The goal of care is to recognize and progression of renal cyst growth in ADPKD by suppress-
minimize psychological morbidity, and foster patient ing AVP release [56]. However, there is no high-level evi-
empowerment and self-motivation [50]. dence to support this as a clinical recommendation [6].
Furthermore, the volume of fluid required to suppress
AVP is also likely to vary because it is dependent on the
First-line Treatments to Prevent and/or
amount of dietary solute intake that a patient con-
Minimize Renal and Cardiovascular Disease
sumes [56]. To date, the evidence for the hypothesis is
All patients with ADPKD should receive first-line (or pri- limited and based on short-term studies (less than
mary) treatment (dietary salt minimization, optimal caloric 2 months in duration) which show that it is possible to
energy intake) [51], treatment of hypertension with either reduce serum copeptin at consumption volumes that
angiotensin converting enzyme (ACE) or angiotensin were tolerable to patients. However, the long-term effi-
receptor blocker (ARB), and avoidance of acute renal cacy (on renal function decline and renal cyst growth),
injury by minimizing exposure to nephrotoxins (low cer- safety, and feasibility of this approach are not known. The
tainty evidence, strong recommendation). In patients with only available long-term data in humans is a small non-
low to moderate risk of renal disease progression (which randomized observational cohort study (n = 33) which
may be up to 50-75% of patients) first-line interventions showed that the progression of TKV or decline in eGFR in
may curtail the lifetime development of ESKD and cardio- patients who drank 2.5–3.0 l per day was not different to
vascular morbidity if effectively implemented in young those who consumed lower amounts over a 1-year
adults soon after diagnosis [52]. period [57]. This study was nonrandomized, of insuffi-
cient duration and power as well as being an observation
Diet Modification cohort design, therefore further data are needed [6].
Dietary salt restriction: In the HALT-PKD trial, each 18 mEq Caffeine intake is not contraindicated in ADPKD patients
increase in averaged urinary sodium excretion was associ- (strong recommendation, low certainty evidence) [6].
ated with a 0.43% per year rise in ht-TKV in patients with
eGFR>60 ml/min/1.73 m2 and with a faster decline in renal Blood Pressure Control, Vascular Dysfunction,
function (−0.09 ml/min/1.73 m2) in patients with eGFR and Cardiovascular Disease
between 25 and 60 ml/min/1.73 m2 [53]. In addition, in a Blood pressure lowering: Angiotensin converting enzyme
small randomized controlled trial (RCT) (n = 34) of 2 weeks inhibitors or angiotensin II receptor blockers are preferred
in duration in ADPKD patients with eGFR>60 ml/ first-line agents for blood pressure control in (high certainty
min/1.73 m2, an ~30% reduction in urinary sodium intake evidence, strong recommendation) [7]. There is limited
(as determined by 24-hour urine sodium from ~3 to ~2 g/ evidence to guide second-line treatment choice (alone if
day) together with an ~40% increase in fluid intake caused intolerant to angiotensin converting enzyme inhibitor or in
a small but signification reduction in serum copeptin combination) but the stepwise use of a cardioselective
(~1 pmol/l) (a surrogate marker of AVP) [54]. blocker dihydropyridine calcium channel blocker, and/or
Treatmen 297
or a diuretic is suggested [2]. The KDIGO Controversy ●● Repeat screening: Follow-up interval screening is recom-
Conference on ADPKD suggested a target blood pressure mended only in patients with a family history of ICA or
(BP) of 130/80 mmHg in patients with macroalbuminuria subarachnoid hemorrhage, or those with previous ICA
( 25 mg/mmol in males and 35 mg/mmol in females) or or rupture, at intervals between 2 and 5 years, or accord-
140/90 mmHg if no macroalbuminuria for ADPKD ing to risk factors such as hypertension, smoking, “high-
patients [3]. The HALT-PKD trial showed that a target BP risk” occupation or prior to major elective surgery [3].
of 95/60 to 110/75 mmHg (vs. 120/70 to 130/80) was associ- The incidence of de novo aneurysms is 1.4% patient/
ated with a small (1.1% per year) but significant reduction year [64]. Repeat interval screening in low-risk asymp-
in TKV in young ADPKD patients (15–49 years old with tomatic patients with a negative initial scan is not
eGFR>1.73 m2) over an 8-year period. This was associated recommended (low certainty of evidence, weak
with a 16% increase in dizziness and light-headedness in recommendation) [3].
the low BP target group. Based on this data, the KHA-CARI
ADPKD Guidelines suggest a BP target of at least Serum cholesterol and lipid-lowering drugs: We suggest
130/80 mmHg, and that a lower target BP could be consid- lipid-lowering drugs are prescribed to prevent cardiovascu-
ered in selected patients [2, 7]. lar death based on evidence recommendations in CKD
Cardiac manifestations: The prevalence of cardiac valvu- rather than to specifically reduce kidney cyst growth (low
lar abnormalities (especially mitral valve prolapse/incom- certainty of evidence, weak recommendation). In a study
petence) and cardiomyopathy are increased in ADPKD [58, of 110 pediatric ADPKD patients (8–22 years), pravastatin
59]. The incidence of left ventricular hypertrophy and val- mildly attenuated the increase in TKV over 3 years (21% vs.
vular calcification increases as renal function declines [59]. 30% from baseline, P = 0.02) independent of serum choles-
However, echocardiography is not recommended unless terol [41], but this was not confirmed in a post hoc analysis
physical examination is abnormal [2]. of the HALT trial [65].
Intracranial aneurysms Physical Activity

●● Screening: We recommend that magnetic resonance Patients should be educated that the risk of renal cyst hem-
angiography or CT (circle of Willis angiography) is used orrhage is increased if the kidney volume is high (e.g. if the
for the screening of ICAs in asymptomatic individuals longitudinal length is >15–16 cm) and these patients
at high risk for an ICA (moderate certainty of evidence, should preferably avoid high-risk contact sports (e.g.
weak recommendation). However, the decision to per- boxing, rugby, American football) (low certainty, weak
form universal/systematic screening (i.e. to screen eve- recommendation).
ryone with ADPKD at diagnosis) or to perform targeted
Avoidance of Acute Kidney Injury
screening for of those with risk factors (i.e. screen those
ADPKD patients should avoid or minimize their expo-
at high-risk, family history of SAH, presence of neuro-
sure to nephrotoxins (NSAIDs, aminoglycosides, radio-
logical symptoms, prior to major surgical procedures) is
contrast) or procedures that increase the risk of acute
controversial [60], and this likely explains the higher
kidney injury as it may increase the risk of renal cyst growth
than expected screening rate in ADPKD patients (~30%
(low certainty of evidence, strong recommendation) [66]
undergo screening for ICAs within 2 years of diagnosis
(Table 19.10).
of ADPKD) [61]. In this regard, a recent cohort analysis
of French patients suggested that screening should be
Disease-modifying Treatments to Prevent and/
systematic rather than targeted [62]. Interestingly, clin-
or Minimize Renal Disease
ical practice guidelines led by nephrologists recom-
mend performing targeted screening for ICAs whereas Disease-modifying treatments may be defined as those for
professionals from the stroke community favor a uni- use in selected patients at high risk for ESKD (possibly
versal approach [11, 63]. In counseling patients, the ~25% of patients) and/or who exhibit features of rapid
risks and benefits of screening should be carefully dis- progression (e.g. “rapid progressor” or other prognostic
cussed, noting that when positive, small aneurysms of factors) despite receiving maximal first-line management.
the anterior circulation have a low risk of rupture [64]. The high likelihood of adverse events and/or low uncer-
Overall, complications related to the treatment of tainty of evidence are reasons for their exclusion as first-
unruptured aneurysms may be as high as 15.6% [64]. line treatments.
Cerebral CT angiogram with low-dose contrast or cere- Vasopressin receptor antagonists in patients at high risk
bral magnetic resonance angiography is the standard for progression of ESKD: The current evidence based on
screening method [11]. two RCTs does not support the routine use vasopressin
Table 19.10 Summary of TEMPO 3:4 and REPRISE RCTs [23, 67].
Name Study population Study design n Primary endpoint Secondary endpoint Adverse events
TEMPO 3:4 18–50 years old 3 years 1445 TKV: increase 2.8%/year Slower decline in kidney Tolvaptan associated with
TKV > 750 ml Phase 3 double-blind, in the tolvaptan group vs. function (reciprocal of the aquaresis and hepatic and
eGFR>60 ml/min placebo (2:1), 5.5%/year in the placebo serum creatinine level, higher discontinuation rate
multicenter −2.61 (mg per milliliter)/ (23% vs. 14% in the placebo
year vs. -3.81 (mg per group)
milliliter)/year; P < 0.001);
lower rates of worsening
kidney function (2 vs. 5
events per 100 person-year,
P < 0.001) and kidney pain
(5 vs. 7 events per 100
person-years of follow-up
P = 0.007)
REPRISE 18–55 years old (eGFR 25–65 ml/ 1 year, phase 3, 1370 Change from baseline in Reversible increases in the
min/1.73 m2) and 56–65 (eGFR randomized the estimated GFR was ALT (to >3 times normal
25–44 ml/min/1.73 m2), and ability withdrawal, −2.34 ml per minute per range) 5.6% in tolvaptan
to tolerate tolvaptan after an 8-week multicenter, placebo- 1.73 m in the tolvaptan vs. group vs. 1.2% in the placebo
pre-randomization period which controlled, double- −3.61 ml per minute per group
included sequential placebo and blind trial (1:1) 1.73 m in the placebo;
tolvaptan run-in phases P < 0.001
Sources: Torres et al. [23, 67].
0005152407.INDD 298 09-12-2022 15:25:43

Treatmen 299
receptor antagonists in ADPKD and to use shared decision- ulcers), insufficient drug delivered to target tissue, and trial
making with patients if there is “rapid progression” as design, as therapy was initiated when kidneys had reached
defined by a historical decline in the eGFR. [23, 67]. their near-maximal volume [73].
Tolvaptan is a specific vasopressin receptor antagonist
that inhibits the binding of AVP at the V2 receptor (a GS-
Kidney Pain
protein-coupled receptor) of the cortical collecting duct.
In two RCTs, treatment with tolvaptan reduced the rate of Chronic pain (>4–6 weeks in duration): The optimal man-
decline in eGFR by approximately 1ml/min/1.73m2/year agement of chronic kidney pain is not certain [8]. Mild
compared to placebo [23, 67], but the limitations of this diffuse discomfort in the loin, abdomen, or lower back,
data should be considered: (i) the effect size on eGFR was present on most days, is common and reported by 60% of
small and apparent only after discontinuation, suggesting patients with “large” kidneys due to their physical effect
a role for hemodynamic and other cofounding factors [68]; in the abdominal cavity and pressure on the lower lumbar
and (ii) the long-term effect of tolvaptan on preventing spine. In a cross-sectional analysis of the HALT study
ESKD is assumed and evidence is lacking [68]. In addi- cohort (n = 1043) back pain was reported in ~50% of
tion, mechanistic questions remain: (i) preclinical studies patients and was unrelated to kidney volume when the
in pck rats suggest that both increased water intake and eGFR was greater than 60 ml/min/1.73 m2 (unless the Ht-
vasopressin receptor antagonist could have similar effi- TKV exceeded 1000 ml/m) whereas patients with
cacy on reducing the kidney to body weight ratio by eGFR<60 were more likely to report abdominal fullness
~30% [69, 70]; and (ii) in vitro studies support a direct role and pain that was disabling. The DIPAK cohort analysis
for inhibition of AVP on renal cyst growth, but it is not (n = 309) showed that up to 30% of patients have associ-
known if human renal cysts express V2 receptor. ated gastrointestinal symptoms (such as heartburn, nau-
The main adverse effects of tolvaptan are aquaresis sea, vomiting, and early satiety) due to combined kidney
(mean 5–7 l per day at maximal dose) and 5% of patients and liver volume. In a systematic review of patient per-
may develop reversible drug-induced liver injury. Thus, spectives on living with ADPKD, renal pain was described
patients receiving tolvaptan must have liver function tests as volatile and incapacitating, which disrupted daily liv-
monitored every 1-3 months while on therapy (low cer- ing and impaired ability to establish long-term life
tainty of evidence, strong recommendation). goals [47]. Hence, screening for chronic pain should be
Somatostatin analogs (octreotide, lanreotide): routine during all clinic visits. There is no high-level evi-
Somatostatin is a neuropeptide produced by dence to recommend the optimal approach for managing
gastrointestinal-derived delta-cells that attenuates intesti- chronic pain. However, in refractory cases, a stepwise
nal motility, absorption, and proliferation [71, 72]. In approach consisting of nonpharmacological (behavioral
humans, the ALADIN study (an RCT of 75 patients) modification, physiotherapy, ice/heating pads,) and phar-
showed that oectroetide reduced the increase in TKV at macological (three-step analgesic ladder) treatments fol-
1 year but not at 3 years compared to no treatment [71, 72]. lowed by minimally invasive (nerve block) and invasive
The DIPAK-1 study which was a 2.5-year RCT comparing (cyst aspiration, cyst fenestration, nephrectomy/hepatec-
lanreotide + standard care (120 mg SC 4 weekly, n = 153) tomy) therapies is suggested (low certainty of evidence,
with standard care. The results showed that lanreotide did weak recommendation) [74, 75]. A nonrandomized
not alter the decline in eGFR (−3.56 vs. −3.46 ml/ single-center study showed that a multidisciplinary
min/1.73 m2 per year) but reduced the rate of increase in sequential protocol involving nerve block can relieve pain
TKV (4.2% vs. 5.6%) [71, 72]. These data as well as the high in up to 80% of patients (n = 60) with chronic refractory
risk of gastrointestinal side-effects do not support the use pain [76].
of somatostatin analogs in ADPKD. Acute pain: The incidence of acute pain in ADPKD is
TORC1 inhibitors (everolimus, sirolimus): Animal studies less common than chronic pain, and data from the
conclusively demonstrated that TORC1 inhibitors suppress TEMPO 3:4 study suggested an average incidence of
the proliferation of cystic epithelial cells and reduce renal seven episodes per 100 years. The sudden onset of acute
cyst growth. However, two clinical trials did not confirm pain requires evaluation with history, physical examina-
the same efficacy in humans [73]. This disparity could be tion, and renal imaging (generally by CT). The differential
explained by: patient dropouts due to side effects (mouth diagnosis includes:
●● Kidney cyst hemorrhage and rupture: We suggest that refractory. Positron emission tomography can be used
renal cyst hemorrhage be treated with bed rest and as second-line investigation if the diagnosis is uncer-
analgesia as first-line treatment, followed by minimally tain but has low specificity and is expensive. Suspected
invasive or invasive therapies if bleeding is refractory cyst infection should be treated with lipid- and cyst-
(low certainty of evidence, weak recommendation). permeable antibiotics (e.g. norfloxacin or co-
Kidney cyst hemorrhage is characterized by sudden trimoxazole) for an extended duration (4–8 weeks). A
onset of sharp localized and unilateral abdominal pain systematic review found that risk factors for failure of
either secondary to contact sports, blunt trauma or may initial therapy (~39%) were large cysts (diameter
be spontaneous (as cysts are surrounded by neoangio- >5 cm), ESKD, and atypical organisms [80]. Safety data
genesis). In most cases (~80%) cyst hemorrhage bleed from the DIPAK trial show that lanreotide is associated
into the urinary collecting system causes macroscopic with higher risk of hepatic cyst infection for reasons
hematuria, and less commonly bleeding is retroperito- not understood.
neal. The risk of cyst hemorrhage increases with kidney ●● Effect of tolvaptan on acute pain episodes: A post hoc
size, especially when kidney length exceeds 15 cm [77]. analysis of the TEMPO 3:4 study showed that tolvaptan
Bleeding into the collecting system tends to self-resolve, use was associated with a reduced incidence kidney pain
with symptomatic episodes lasting 3–5 days, and man- events when compared to placebo (10.1% vs. 16.8%,
agement entails bed rest, increased fluid intake, and P < 0.001) [67]. This effect is largely due to a decrease in
analgesia. Rarely, selective renal artery embolization is acute pain secondary to urinary tract infection, kidney
required if retroperitoneal bleeding is refractory or stones, and hematuria, and it is not known if prescribed
catheter irrigation is required if there is ureteric clot water intake could also be similar.
obstruction [78].
●● Nephrolithiasis: Kidney stones occur in ~30% of ADPKD
patients (2-fold higher than the general population) Management of Extrarenal Complications
and are usually composed of uric acid or calcium oxa-
late [79]. They occur because of reduced citrate excre- The frequency of extrarenal manifestations is shown in
tion (an inhibitor of stone formation) and altered Table 19.11. Psychological morbidity (as discussed earlier)
tubular urine flow. Clinical features resemble classical followed by the cardiovascular and gastrointestinal sys-
renal colic with or without macroscopic hematuria and tems are the most commonly affected extra-renal areas.
urological management is the same as in the general Patients with ADPKD were historically suspected to have
population [79]. an increased incidence of malignancy, but data to support
●● Infection: Urinary tract infections (UTIs) (upper: pyelo- this hypothesis is inconsistent.
nephritis; lower: cystitis, urethritis) occur at least once
in ~30–50% of ADPKD patients. The characteristic Hepatic Cysts and Polycystic Liver Disease
features are an acute onset of unilateral renal pain, asso- Liver cysts and hepatomegaly occur in up to 80% of
ciated with fever, chills, and rigors, and/or macroscopic patients, but generally remain asymptomatic, with normal
hematuria with a positive urine culture and pyuria liver function tests in ~95% of patients [1, 10]. Approximately
(mostly gram-negative enteric organisms). The mecha- 5% of patients develop severe polycystic liver disease (PLD,
nisms and management of ADPKD-associated UTI are defined as a height-corrected liver volume >1800 ml/m)
the same as for the general population. By comparison, predominantly (~80%) in females. The increased risk in
an infected cyst (either kidney or hepatic) has a suba- females is associated with a history of multiple pregnancies
cute onset, nonspecific symptoms, low-grade fever, and and exposure to hormone replacement therapy, suggesting
evidence of systemic inflammation (elevated white cell a causative link to estrogen exposure. In these cases, we
count, C-reactive protein). The urine culture is negative suggest that severe PLD be managed by a multidisciplinary
and there may be acute tenderness corresponding to the team (hepatologist, hepatobiliary surgeon, interventional
location of the infected cyst. Features on a CT scan that radiologist, nephrologist; low certainty of evidence, weak
favor cyst infection (vs. hemorrhage) include increase recommendation). Short-term clinical trials suggest that
in cyst wall thickness or rarely the presence of gas somatostatin analogs (octreotide, lanreotide, pasiretoide)
within a cyst. There are no evidence-based guidelines reduce symptoms of PLD with a small effect on reducing
for the optimal management of cyst infection. hepatic cyst growth rate, but long-term outcomes are not
Microbiological confirmation can be difficult, as the known. Rarely, liver transplantation (due to abdominal dis-
urine culture is often negative. Occasionally, CT-guided tension and portal hypertension) is required in severe
aspiration of an infected cyst is required if antibiotic PKD [1, 10].
Reference 301
Table 19.11 Frequency of extrarenal manifestations of ADPKD. appropriate treatment options to manage ESKD. A recent
meta-analysis (n = 14 673) showed that peritoneal dialysis-
Frequency of patients related complications (technique failure, peritonitis, her-
nia) in ADPKD patients were similar to those in non-ADPKD
Psychological issues patients [81]. As for other types of CKD, kidney transplan-
Anxiety, depression, anger and denial High tation is the preferred modality for the management of
Cardiovascular system ESKD and patients should be evaluated for need for
Cardiac valve abnormalities 25% nephrectomy prior to transplantation (high certainty of evi-
(mitral valve prolapse)
dence, strong recommendation). Transplantation may be
Hypertension 80–100%
associated with a ~2-fold increased risk for developing new-
Left ventricular hypertrophy 20–40%
onset diabetes compared to other causes of CKD [9].
Gastrointestinal system
Liver cysts 94% >35 years old
Diverticular disease 50–83% in ESKD
Hernia patients C
onclusion
Pancreatic cysts 45%
Splenic cysts 9–26% At present there is minimal evidence that treatment with
Choledochal cysts 2.7% disease-modifying agents (such as vasopressin receptor
Common bile duct dilation Rare antagonists) improves outcomes in ADPKD due to uncer-
40% tainty regarding efficacy, absence of patient-centered end-
Neurological points, and the high risk of significant risk adverse events.
Intracranial aneurysms 9–12% On the other hand, it is possible that first-line treatments
Arachnoid cysts 8–12% (as described in this chapter), if instituted consistently in
Spinal meningeal cysts 1.7% early life, could reduce lifetime risk of ESKD in moderate-
Other organ complications to low-risk patients. For high-risk patients, numerous clini-
Seminal vesicle cysts 40% cal trials are in development and understanding of disease
Bronchiectasis 37% pathogenesis and diagnostics is proceeding at an exponen-
Thyroid cysts Not clearly defined tial rate. Furthermore, previously neglected areas of treat-
Risk of malignancy Controversial, not ment, such as optimal genetic and psychosocial care, ICAs,
defined and polycystic liver disease, are also gaining much deserved
attention. The challenge for basic and clinical researchers
Management of ESKD in ADPKD will be to discover, select, and recommend the most prom-
ising interventions that have acceptable side effects (given
Late-stage 4–5 CKD should be managed as for other types that they might need to be taken for decades) using the
of CKD [9]. Either peritoneal dialysis or hemodialysis are most efficient methods in clinical trial design.
R
eferences
Rangan, G.K., Tchan, M.C., Tong, A. et al. (2016). Recent

1 care: European ADPKD forum and multispecialist
advances in autosomal-dominant polycystic kidney roundtable participants. Nephrol. Dial. Transplant. 33:
disease. Intern. Med. J. 46 (8): 883–992. 563–573.
2 Chebib, F.T. and Torres, V.E. (2018). Recent advances in 5 Tchan, M., Savige, J., Patel, C. et al. (2015). KHA-CARI
the management of autosomal dominant polycystic kidney autosomal dominant polycystic kidney disease guideline:
disease. Clin. J. Am. Soc. Nephrol. 13: 1765–1776. genetic testing for diagnosis. Semin. Nephrol. 35 (6): 545–9.e2.
3 Chapman, A.B., Devuyst, O., Eckardt, K.U. et al. (2015). 6 Campbell, K.L., Rangan, G.K., Lopez-Vargas, P., and Tong,
Autosomal-dominant polycystic kidney disease (ADPKD): A. (2015). KHA-CARI autosomal dominant polycystic
executive summary from a kidney disease: improving kidney disease guideline: diet and lifestyle management.
global outcomes (KDIGO) controversies conference. Semin. Nephrol. 35 (6): 572–81.e17.
Kidney Int. 88 (1): 17–27. 7 Mallett, A., Lee, V.W., Mai, J. et al. (2015). KHA-CARI
4 Harris, T., Sandford, R., de Coninck, B. et al. (2017). autosomal dominant polycystic kidney disease guideline:
European ADPKD forum multidisciplinary position pharmacological management. Semin. Nephrol. 35 (6):
statement on autosomal dominant polycystic kidney disease 582–9.e17.
8 Savige, J., Tunnicliffe, D.J., and Rangan, G.K. (2015). 20 Patel, C., Tchan, M., Savige, J. et al. (2015). KHA-CARI
KHA-CARI autosomal dominant kidney disease autosomal dominant polycystic kidney disease guideline:
guideline: management of chronic pain. Semin. Nephrol. genetics and genetic counseling. Semin. Nephrol. 35 (6):
35 (6): 607–11.e3. 550–6.e1.
9 Lee, V.W., Tunnicliffe, D.J., and Rangan, G.K. (2015). 21 Mallawaarachchi, A.C., Hort, Y., Cowley, M.J. et al.
KHA-CARI autosomal dominant polycystic kidney (2016). Whole-genome sequencing overcomes
disease guideline: management of end-stage kidney pseudogene homology to diagnose autosomal dominant
disease. Semin. Nephrol. 35 (6): 595–602.e12. polycystic kidney disease. Eur. J. Hum. Genet. 24 (11):
10 Savige, J., Mallett, A., Tunnicliffe, D.J., and Rangan, G.K. 1584–1590.
(2015). KHA-CARI autosomal dominant polycystic 22 Grantham, J.J. (2015). Rationale for early treatment of
kidney disease guideline: management of polycystic liver polycystic kidney disease. Pediatr. Nephrol. 30 (7):
disease. Semin. Nephrol. 35 (6): 618–22.e5. 1053–1062.
11 Lee, V.W., Dexter, M.A., Mai, J. et al. (2015). KHA-CARI 23 Torres, V.E., Chapman, A.B., Devuyst, O. et al. (2017).
autosomal dominant polycystic kidney disease guideline: Tolvaptan in later-stage autosomal dominant polycystic
management of intracranial aneurysms. Semin. Nephrol. kidney disease. N. Engl. J. Med. 377 (20): 1930–1942.
35 (6): 612–7.e20. 24 Murphy, E.L., Droher, M.L., DiMaio, M.S., and Dahl, N.K.
12 Dalgaard, O. (1957). Bilateral polycystic disease of the (2018). Preimplantation genetic diagnosis counseling in
kidneys; a follow-up of two hundred and eighty-four autosomal dominant polycystic kidney disease. Am. J.
patients and their families. Acta Med. Scand. (Suppl 328): Kidney Dis. 72: 866–872.
1–255. 25 Chapman, A.B., Bost, J.E., Torres, V.E. et al. (2012).
13 Iglesias, C.G., Torres, V.E., Offord, K.P. et al. (1983). Kidney volume and functional outcomes in autosomal
Epidemiology of adult polycystic kidney disease, Olmsted dominant polycystic kidney disease. Clin. J. Am. Soc.
County, Minnesota: 1935-1980. Am. J. Kidney Dis. 2 (6): Nephrol. 7 (3): 479–486.
630–639. 26 Barua, M., Cil, O., Paterson, A.D. et al. (2009). Family
14 Solazzo, A., Testa, F., Giovanella, S. et al. (2018). The history of renal disease severity predicts the mutated gene
prevalence of autosomal dominant polycystic kidney in ADPKD. J. Am. Soc. Nephrol. 20 (8): 1833–1838.
disease (ADPKD): a meta-analysis of European literature 27 Perez Dominguez, T.S., Rodriguez Perez, A., Buset Rios,
and prevalence evaluation in the Italian province of N. et al. (2011). Psychonephrology: psychological aspects
Modena suggest that ADPKD is a rare and in autosomal dominant polycystic kidney disease.
underdiagnosed condition. PLoS One 13 (1): e0190430. Nefrologia 31 (6): 716–722.
15 Spithoven, E.M., Kramer, A., Meijer, E. et al. (2014). 28 Cornec-Le Gall, E., Audrezet, M.P., Rousseau, A. et al.
Renal replacement therapy for autosomal dominant (2016). The PROPKD score: a new algorithm to predict
polycystic kidney disease (ADPKD) in Europe: prevalence renal survival in autosomal dominant polycystic kidney
and survival--an analysis of data from the ERA-EDTA disease. J. Am. Soc. Nephrol. 27 (3): 942–951.
registry. Nephrol. Dial. Transplant. 29 (Suppl 4): 29 Orskov, B., Christensen, K.B., Feldt-Rasmussen, B., and
iv15–iv25. Strandgaard, S. (2012). Low birth weight is associated
16 Neumann, H.P., Jilg, C., Bacher, J. et al. (2013). with earlier onset of end-stage renal disease in Danish
Epidemiology of autosomal-dominant polycystic patients with autosomal dominant polycystic kidney
kidney disease: an in-depth clinical study for South- disease. Kidney Int. 81 (9): 919–924.
Western Germany. Nephrol. Dial. Transplant. 28 (6): 30 Bhutani, H., Smith, V., Rahbari-Oskoui, F. et al. (2015). A
1472–1487. comparison of ultrasound and magnetic resonance
17 Lanktree, M.B., Haghighi, A., Guiard, E. et al. (2018). imaging shows that kidney length predicts chronic
Prevalence estimates of polycystic kidney and liver kidney disease in autosomal dominant polycystic kidney
disease by population sequencing. J. Am. Soc. Nephrol.: disease. Kidney Int. 88 (1): 146–151.
2593–2600. 31 Shen, C., Landsittel, D., Irazabal, M.V. et al. (2017).
18 Pei, Y., Obaji, J., Dupuis, A. et al. (2009). Unified criteria Performance of the CKD-EPI equation to estimate GFR
for ultrasonographic diagnosis of ADPKD. J. Am. Soc. in a longitudinal study of autosomal dominant polycystic
Nephrol. 20 (1): 205–212. kidney disease. Am. J. Kidney Dis. 69 (3): 482–484.
19 Mai, J., Lee, V.W., Lopez-Vargas, P., and Vladica, P. (2015). 32 Yu, A.S.L., Shen, C., Landsittel, D.P. et al. (2019).
KHA-CARI autosomal dominant polycystic kidney Long-term trajectory of kidney function in autosomal-
disease guideline: imaging approaches for diagnosis. dominant polycystic kidney disease. Kidney Int. 95 (5):
Semin. Nephrol. 35 (6): 538–44.e17. 1253–1261.
Reference 303
33 Gansevoort, R.T., Arici, M., Benzing, T. et al. (2016). polycystic kidney disease. Pain Manag. Nurs. 10 (3):
Recommendations for the use of tolvaptan in autosomal 134–141.
dominant polycystic kidney disease: a position statement 46 Manjoney, D.M. and McKegney, F.P. (1978). Individual
on behalf of the ERA-EDTA working groups on inherited and family coping with polycystic kidney disease: the
kidney disorders and European renal best practice. harvest of denial. Int. J. Psychiatry Med. 9 (1): 19–31.
Nephro.l Dial. Transplant. 31 (3): 337–348. 47 Tong, A., Rangan, G.K., Ruospo, M. et al. (2015). A
34 Blanchette, C.M., Liang, C., Lubeck, D.P. et al. (2015). painful inheritance-patient perspectives on living with
Progression of autosomal dominant kidney disease: polycystic kidney disease: thematic synthesis of
measurement of the stage transitions of chronic kidney qualitative research. Nephrol. Dial. Transplant. 30 (5):
disease. Drugs Context 4: 212275. 790–800.
35 Irazabal, M.V., Rangel, L.J., Bergstralh, E.J. et al. (2015). 48 Levenson J. (2018). Psychological factors affecting other
Imaging classification of autosomal dominant polycystic medical conditions: Management. In UpToDate J.
kidney disease: a simple model for selecting patients for Dimsdale and D. Solomon (Eds) https://www.uptodate.
clinical trials. J. Am. Soc. Nephrol. 26 (1): 160–172. com/contents/psychological-factors-affecting-other-
36 Mai, J., Lee, V.W., Lopez-Vargas, P. et al. (2015). KHA- medical-conditions-management?source=related_link
CARI autosomal dominant polycystic kidney disease (accessed 22 May 2021).
guideline: monitoring disease progression. Semin. 49 Levenson J. (2018). Psychological factors affecting other
Nephrol. 35 (6): 565–71.e18. medical conditions: Clinical features, assessment, and
37 Harris, T. (2017). Is it ethical to test apparently “healthy” diagnosis. In UpToDate J. Dimsdale and D. Solomon
children for autosomal dominant polycystic kidney (Eds) https://www.uptodate.com/contents/
disease and risk Medicalizing thousands? Front. Pediatr. psychological-factors-affecting-other-medical-
5: 291. conditions-clinical-features-assessment-and-diagnosis
38 De Rechter, S., Breysem, L., and Mekahli, D. (2017). Is (accessed 22 May 2021).
autosomal dominant polycystic kidney disease becoming 50 Tong, A., Tunnicliffe, D.J., Lopez-Vargas, P. et al. (2016).
a pediatric disorder? Front. Pediatr. 5: 272. Identifying and integrating consumer perspectives in
39 Massella, L., Mekahli, D., Paripovic, D. et al. (2018). clinical practice guidelines on autosomal-dominant
Prevalence of hypertension in children with early-stage polycystic kidney disease. Nephrology (Carlton) 21 (2):
ADPKD. Clin. J. Am. Soc. Nephrol. 13 (6): 874–883. 122–132.
40 Marlais, M., Cuthell, O., Langan, D. et al. (2016). 51 Nowak, K.L., You, Z., Gitomer, B. et al. (2018).
Hypertension in autosomal dominant polycystic kidney Overweight and obesity are predictors of progression in
disease: a meta-analysis. Arch. Dis. Child. 101 (12): early autosomal dominant polycystic kidney disease. J.
1142–1147. Am. Soc. Nephrol. 29 (2): 571–578.
41 Cadnapaphornchai, M.A., George, D.M., McFann, K. 52 Furlano, M., Loscos, I., Marti, T. et al. (2018). Autosomal
et al. (2014). Effect of pravastatin on total kidney volume, dominant polycystic kidney disease: clinical assessment
left ventricular mass index, and microalbuminuria in of rapid progression. Am. J. Nephrol. 48 (4): 308–317.
pediatric autosomal dominant polycystic kidney disease. 53 Torres, V.E., Abebe, K.Z., Schrier, R.W. et al. (2017).
Clin. J. Am. Soc. Nephrol. 9 (5): 889–896. Dietary salt restriction is beneficial to the management of
42 Swift, O., Vilar, E., Rahman, B. et al. (2016). Attitudes in autosomal dominant polycystic kidney disease. Kidney
patients with autosomal dominant polycystic kidney Int. 91 (2): 493–500.
disease toward prenatal diagnosis and preimplantation 54 Amro, O.W., Paulus, J.K., Noubary, F., and Perrone, R.D.
genetic diagnosis. Genet. Test Mol. Biomarkers 20 (12): (2016). Low-Osmolar diet and adjusted water intake for
741–746. vasopressin reduction in autosomal dominant polycystic
43 Perez-Dominguez, T., Rodriguez-Perez, A., Garcia-Bello, kidney disease: a pilot randomized controlled trial. Am. J.
M.A. et al. (2012). Progression of chronic kidney disease. Kidney Dis. 68 (6): 882–891.
Prevalence of anxiety and depression in autosomal 55 Klahr, S., Breyer, J.A., Beck, G.J. et al. (1995). Dietary
dominant polycystic kidney disease. Nefrologia 32 (3): protein restriction, blood pressure control, and the
397–399. progression of polycystic kidney disease. Modification of
44 Levy, M. (2001). How well do we manage and support Diet in Renal Disease Study Group. J. Am. Soc. Nephrol. 5
patients and families with dominantly inherited renal (12): 2037–2047.
disease? Nephrol. Dial. Transplant. 16 (1): 1–4. 56 Torres, V.E., Bankir, L., and Grantham, J.J. (2009). A case
45 Heiwe, S. and Bjuke, M. (2009). “An evil heritage”: for water in the treatment of polycystic kidney disease.
interview study of pain and autosomal dominant Clin. J. Am. Soc. Nephrol. 4 (6): 1140–1150.
7 Higashihara, E., Nutahara, K., Tanbo, M. et al. (2014).

5 polycystic kidney disease in the PCK rat. J. Am. Soc.
Does increased water intake prevent disease progression Nephrol. 17 (8): 2220–2227.
in autosomal dominant polycystic kidney disease? 70 Wang, X., Gattone, V. 2nd, Harris, P.C., and Torres, V.E.
Nephrol. Dial. Transplant. 29 (9): 1710–1719. (2005). Effectiveness of vasopressin V2 receptor
58 Chebib, F.T., Hogan, M.C., El-Zoghby, Z.M. et al. (2017). antagonists OPC-31260 and OPC-41061 on polycystic
Autosomal dominant polycystic kidney patients may be kidney disease development in the PCK rat. J. Am. Soc.
predisposed to various cardiomyopathies. Kidney Int. Rep. Nephrol. 16 (4): 846–851.
2 (5): 913–923. 71 Caroli, A., Perico, N., Perna, A. et al. (2013). Effect of
59 Bardaji, A., Martinez-Vea, A., Valero, A. et al. (2001). longacting somatostatin analogue on kidney and cyst
Cardiac involvement in autosomal-dominant polycystic growth in autosomal dominant polycystic kidney disease
kidney disease: a hypertensive heart disease. Clin. (ALADIN): a randomised, placebo-controlled,
Nephrol. 56 (3): 211–220. multicentre trial. Lancet 382 (9903): 1485–1495.
60 Flahault, A., Trystram, D., Fouchard, M. et al. (2016). 72 Meijer, E., Visser, F.W., van Aerts, R.M.M. et al. (2018).
Screening for Unruptured intracranial aneurysms in Effect of lanreotide on kidney function in patients with
autosomal dominant polycystic kidney disease: a survey autosomal dominant polycystic kidney disease: the
of 420 nephrologists. PLoS One 11 (4): e0153176. DIPAK 1 randomized clinical trial. JAMA 320: 2010–2019.
61 Wilkinson, D.A., Burke, J.F., Nadel, J.L. et al. (2018). A 73 Myint, T.M., Rangan, G.K., and Webster, A.C. (2014).
large database analysis of rates of aneurysm screening, Treatments to slow progression of autosomal dominant
elective treatment, and subarachnoid hemorrhage in polycystic kidney disease: systematic review and meta-
patients with polycystic kidney disease. Neurosurgery 85: analysis of randomized trials. Nephrology (Carlton) 19 (4):
E266–E274. 217–226.
62 Flahault, A., Trystram, D., Nataf, F. et al. (2018). 74 Casteleijn, N.F., Visser, F.W., Drenth, J.P. et al. (2014). A
Screening for intracranial aneurysms in autosomal stepwise approach for effective management of chronic
dominant polycystic kidney disease is cost-effective. pain in autosomal-dominant polycystic kidney disease.
Kidney Int. 93 (3): 716–726. Nephrol. Dial. Transplant 29 (Suppl 4): iv142–iv153.
63 Thompson, B.G., Brown, R.D. Jr., Amin-Hanjani, S. et al. 75 Hogan, M.C. and Norby, S.M. (2010). Evaluation and
(2015). Guidelines for the management of patients with management of pain in autosomal dominant polycystic
unruptured intracranial aneurysms: a guideline for healthcare kidney disease. Adv. Chronic Kidney Dis. 17 (3): e1–e16.
professionals from the American Heart Association/ 76 Casteleijn, N.F., van Gastel, M.D., Blankestijn, P.J. et al.
American Stroke Association. Stroke 46 (8): 2368–2400. (2017). Novel treatment protocol for ameliorating refractory,
64 Cagnazzo, F., Gambacciani, C., Morganti, R., and Perrini, chronic pain in patients with autosomal dominant
P. (2017). Intracranial aneurysms in patients with polycystic kidney disease. Kidney Int. 91 (4): 972–981.
autosomal dominant polycystic kidney disease: prevalence, 77 Mabillard, H., Srivastava, S., Haslam, P. et al. (2017).
risk of rupture, and management. A systematic review. Large retroperitoneal haemorrhage following cyst rupture
Acta Neurochir. (Wien) 159 (5): 811–821. in a patient with autosomal dominant polycystic kidney
65 Brosnahan, G.M., Abebe, K.Z., Rahbari-Oskoui, F.F. et al. disease. Case Rep. Nephrol. 2017: 4653267.
(2017). Effect of statin therapy on the progression of 78 Hajjar, K., Bou Chebl, R., Kanso, M., and Abou, D.G.
autosomal dominant polycystic kidney disease. A (2018). Autosomal dominant polycystic kidney disease
secondary analysis of the HALT PKD trials. Curr. and minimal trauma: medical review and case report.
Hypertens Rev. 13 (2): 109–120. BMC Emerg. Med. 18 (1): 38.
66 Franco Palacios, C., Keddis, M.T., Qin, D. et al. (2011). 79 Mallett, A., Patel, M., Tunnicliffe, D.J., and Rangan, G.K.
Acute kidney injury in ADPKD patients with pneumonia. (2015). KHA-CARI autosomal dominant polycystic
Int. J. Nephrol. 2011: 617904. kidney disease guideline: management of renal stone
67 Torres, V.E., Chapman, A.B., Devuyst, O. et al. (2012). disease. Semin. Nephrol. 35 (6): 603–6.e3.
Tolvaptan in patients with autosomal dominant polycystic 80 Lantinga, M.A., Casteleijn, N.F., Geudens, A. et al. (2017).
kidney disease. N. Engl. J. Med. 367 (25): 2407–2418. Management of renal cyst infection in patients with
68 Sans-Atxer, L. and Joly, D. (2018). Tolvaptan in the autosomal dominant polycystic kidney disease: a systematic
treatment of autosomal dominant polycystic kidney review. Nephrol. Dial. Transplant. 32 (1): 144–150.
disease: patient selection and special considerations. Int. 81 Boonpheng, B., Thongprayoon, C., Wijarnpreecha, K. et al.
J. Nephrol. Renovasc. Dis. 11: 41–51. (2018). Outcomes of patients with autosomal dominant
69 Nagao, S., Nishii, K., Katsuyama, M. et al. (2006). polycystic kidney disease on peritoneal Dialysis: a
Increased water intake decreases progression of meta-analysis. Nephrology (Carlton) 24: 638–646.
305
20
Urinary Tract Infection

Suetonia C. Palmer1 and Giovanni F.M. Strippoli2
1
Department of Medicine, University of Otago Christchurch, Christchurch, New Zealand
2
Department of Emergency and Organ Transplantation University of Bari Bari, Italy and School of Public Health University of Sydney Sydney, NSW, Australia
I ntroduction E
pidemiology
Urinary tract infection is one of the most common bacterial Infection of the lower urinary tract accounts for between
infections around the globe, and is an important cause of 1% and 3% of all ambulatory assessments in the United
morbidity, hospitalization, and antibiotic resistance. Urinary Kingdom and approximately 0.5% of visits in the United
tract infection can be upper or lower; lower urinary tract States [2, 3]. Urinary tract infection leads to 10–20% of anti-
infection refers to infection of the bladder (or cystitis). biotic prescriptions in ambulatory care [4]. The diagnosis
Asymptomatic bacteriuria refers to the presence of >105 of a urinary tract infection is suggested by the presence of
colony-forming units in the urine without physical signs of typical urinary symptoms. Confirmation of uropathogenic
infection. Uncomplicated lower urinary tract infection is bacterial species at >100 colony-forming units/ml in the
usually a self-limiting illness. Symptom duration is reduced urine indicates infection when symptoms are present.
by treatment with antibiotics to which the infective patho- Leucocyte esterase or nitrite positivity on dipstick is highly
gen is sensitive. Urinary tract infections occur due to pre- sensitive in detecting pyuria but may not accurately rule
dominantly bowel-related organisms, most commonly out infection [5]. If the dipstick is negative in the presence
Escherichia coli, while other common bacterial etiologies of symptoms, urinary culture is appropriate to evaluate fur-
include Staphylococcus and less frequently Enterobacteriaceae ther for the diagnosis. Lower urinary tract infections mostly
(Proteus mirabilis and Klebsiella spp.). Lower urinary tract occur through migration of gut-related bacteria into the
infection is frequently identified by specific urinary tract bladder via the urethra [6]. E. coli is the most likely
symptoms combined with urine dipstick testing, with or causative organism, occurring in 70–95% of cases, with
without subsequent microbiological confirmation. Older Staphylococcus saprophyticus occurring in about 5–10% of
adults and children may present with atypical symptoms. cases and occasionally Enterobacteriaceae spp. such as
Children often present with fever and other symptoms such Proteus and Enterobacteriaceae [7]. Asymptomatic bacteri-
as lethargy, vomiting, and poor feeding. uria is considered present when there are no urinary symp-
In this chapter, we summarize the evidence and provide toms and >105 colony-forming units of pathogenic bacterial
recommendations for the treatment and prevention of species are detected in the urine.
uncomplicated lower urinary tract infection in nonpreg- Urinary tract infections are more common in boys before
nant women and in children and men (Table 20.1). We the age of 6 months, occur at similar rates for boys and girls
describe the evidence for therapeutic approaches to at 2 years (8%), occur in 1.7% of girls and boys at 7 years, and
asymptomatic bacteriuria. Discussions of the treatment in 11% and 4% of girls and boys by 16 years [8]. Women are
of complicated urinary tract infection in all adults and approximately 30 times more likely to experience lower uri-
children (including upper tract infection), and uncompli- nary tract infection than men, with almost half of women
cated urinary tract infection in pregnant women are not experiencing one or more infections during their lifetime [9].
included in this chapter. The risk is markedly increased after the onset of sexual
306 Urinary Tract Infection
Table 20.1 Treatment for uncomplicated urinary tract infections.
Strength of
Uncomplicated lower urinary tract infection in women

Antibiotics We suggest that lower urinary tract infection is treated with nitrofurantoin or High (symptomatic Strong
trimethoprim/sulfamethoxazole as first-line therapy. Other drug classes may relief); moderate
have similar effects (beta-lactams), although resistance may be higher. (bacterial cure)
Fluoroquinolones have no advantage over other antibiotic groups and may
increase resistance. We suggest that 3 days of antibiotic therapy is similar to
5–10 days for achieving symptomatic cure for uncomplicated urinary tract
infection while longer treatment can be used for bacterial cure in patients for
whom this is important.
Recurrent lower urinary tract infection in women
Antibiotics We suggest that continuous antibiotic prophylaxis (trimethoprim/ Low (recurrent Weak
sulfamethoxazole, nitrofurantoin, cephalosporin) be used to prevent lower urinary tract
urinary tract infection although the effect is limited to the active antibiotic infections)
intake period, the optimal duration is uncertain, and side effects are common.
Cranberries We do not suggest cranberries be used to prevent urinary tract infections. High (recurrent Strong
urinary tract
infections)
Probiotics We do not suggest probiotics be used to prevent urinary tract infections. Low (recurrent Weak
urinary tract
infections)
Lower urinary tract infection in children and adolescents
Antibiotics We suggest children older than 3 months with lower urinary tract infection be Low (symptomatic Weak
treated with trimethoprim/sulfamethoxazole, trimethoprim, or relief, bacterial cure)
nitrofurantoin as first line. We suggest that 3 days of antibiotic therapy is
similar to 5–10 days for achieving symptomatic cure for uncomplicated
urinary tract infection while longer treatment can be used for bacterial cure
in patients for whom this is important.
Prevention We do not suggest long-term antibiotics be used in children with one or more Moderate (recurrent Moderate
previous urinary tract infections, except in young infants or children with urinary tract
renal abnormalities. We do not suggest that cranberries or probiotics be used infection, antibiotic
to prevent recurrent infection. resistance)
Uncomplicated lower urinary tract infection in men
Antibiotics We suggest that lower urinary tract infection in men is treated with Very low (symptom Weak
nitrofurantoin as first-line therapy. Other drug classes may have similar and bacterial cure)
effects (trimethoprim/sulfamethoxazole, beta-lactams).
Asymptomatic bacteriuria
Antibiotics We do not suggest that antibiotics be used to treat asymptomatic bacteriuria. Moderate Moderate
Screening for asymptomatic bacteriuria in nonpregnant adults and in (development of
children is not recommended. urinary tract
infection, side
effects)
GRADE assessment of the certainty of the evidence [1]: High: This research provides a very good indication of the likely effect. The likelihood that
the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood that the effect
will be substantially different* is moderate. Low: This research provides some indication of the likely effect. However, the likelihood that it will be
substantially different* is high. Very low: This research does not provide a reliable indication of the likely effect. The likelihood that the effect will
be substantially different* is very high. *Substantially different = a large enough difference that it might affect decision-making.
activity. Recurrent urinary tract infections are those that P

athophysiology
occur after appropriate treatment of a previous infection and
are usually due to reinfection rather than relapse. One- Urinary tract infection occurs due to the migration of uro-
quarter of women experience a further infection within pathic bowel-related organisms (predominantly E. coli) that
6 months of a sentinal urinary tract infection [10]. colonize the periurethral opening and which subsequently
Prognosi 307
ascend to the bladder [6]. The bacteria then bind directly to infection have been published. These reviews focus on
epithelial cells (facet cells) in the bladder wall, which then antibiotic therapy and some nonantibiotic interventions,
may internalize the adherent organisms [11]. Uropathogenic including cranberries and probiotics as prophylaxis. Most
bacteria may neutralize intracellular lysosomes and enter studies of urinary tract infection treatment have used
the bladder epithelial cell cytoplasm. Subsequently, bacteria symptomatic cure, bacteriological cure (short-term and
form clonal intracellular bacterial communities within facet long-term), isolation of resistant pathogens, time to symp-
cells and evade innate immune responses when encased tom resolution, days of work loss, and adverse events, par-
within a biofilm [12]. Protein secretion and induction of ticularly rash and gastrointestinal effects, as critical
host gene expression can alter the immune response to blad- outcomes. Evidence is most available for nonpregnant
der wall bacterial invasion and replication. E. coli or other women, with a small number of trials involving children,
bacteria may develop resistance to neutrophil action through and a paucity of data for outcomes in men with urinary
morphological changes and reinvade surrounding and pre- tract infection. Some men (<10%) have been included in
viously unexposed bladder wall epithelial cells. trials of women with infection. Trials have evaluated dura-
tion of antibiotic therapy, route of administration includ-
ing oral versus intravenous, and the long-term use of
P
rognosis antibiotics to prevent urinary tract infection. There is a lim-
ited evidentiary base for evaluation of nonpharmacological
Adult Urinary Tract Infection
interventions.
Uncomplicated urinary tract infection is associated with a
good prognosis in the long term. Among women with an
Treatment of Urinary Tract Infection
infection sensitive to antibiotics, severe symptoms are
expected to last for approximately 3 days, although they Recommendations for Adults
may continue for an additional 1.5 days in women with an Adult patients with uncomplicated urinary tract infection
antimicrobial-resistant organism and somewhat longer should receive oral nitrofurantoin or trimethoprim/sul-
overall if antibiotics are not prescribed [13]. Patients may famethoxazole for 3 days (strong recommendation, high
wait almost 5 days before seeking a medical consultation level of evidence). Antibiotic therapy may be reasonably
for symptoms which have a moderate impact on activities delayed by 48 hours in adults without an indication for
of daily life, with minimal ongoing impact on usual activi- immediate treatment (moderate recommendation; moder-
ties at 10 days [14]. Recurrence of urinary traction infection ate level of evidence). Beta-lactam therapy is likely to be
is most commonly reinfection with the same organism. equivalent to other first-line antibiotic classes, although
Following a first urinary tract infection, approximately 25% local organism resistance may lead to a lower preference of
of women will have a culture-confirmed recurrence within this treatment option. Fluoroquinolones are effective but
6 months and 2.7% may have a second recurrence within associated with higher risks of antibiotic resistance. A
that time frame [15, 16]. Asymptomatic urinary tract infec- longer duration of antibiotic treatment (5–10 days) is rec-
tion is associated with increased risk of urinary tract infec- ommended for women for whom bacteriological eradica-
tion with symptoms, although antimicrobial therapy does tion might be important (planning pregnancy, recurrent,
not reduce this risk [17]. and painful lower urinary tract infections) and for all men.
Pediatric Urinary Tract Infection Recommendations for Children

Children with uncomplicated urinary tract infection
In a systematic review, it was reported that between 3% and
should receive oral antibiotics for 3 days (weak recommen-
15% of children have demonstrable renal parenchymal
dation, low level of evidence). Appropriate antibiotic
changes at 1–2 years after a first urinary tract infection [18].
choices include trimethoprim/sulfamethoxazole, trimeth-
However, it is likely that most children have no cortical
oprim, nitrofurantoin, amoxycillin/clavulanate, or a ceph-
renal defects (>80%) and are unlikely to develop new renal
alosporin. Short treatment courses (3 days) have been
damage despite recurrent infection, while the remainder
shown to be as effective as longer duration of therapy
have focal structural abnormalities that may represent con-
(5–10 days).
genital or acquired injury [18, 19].
Evidence in Adults
Treatment
The different antimicrobial treatments for acute uncom-
Several Cochrane systematic reviews of interventions for plicated lower urinary tract infection have been evaluated
treatment and prevention of uncomplicated urinary tract in 21 randomized controlled trials involving over 6000
participants [20]. Trimethoprim-sulfamethoxazole is as effects are common (weak recommendation, low certainty
effective as fluoroquinolones, beta-lactam drugs, and evidence). Adults do not benefit from cranberries or probi-
nitrofurantoin based on short-term and long-term resolu- otics to prevent urinary tract infections.
tion of symptoms. Rash is more commonly experienced
with trimethoprim/sulfamethoxazole compared to nitro- Recommendations for Children
furantoin and fluoroquinolones. Due to higher resistance Children do require antibiotic treatment to prevention uri-
to beta-lactams in some areas, first-line therapy decisions nary tract infection (moderate recommendation, moderate
might include consideration of the potential for adverse certainty evidence).
events, local microbial patterns, and patient preferences.
In a single randomized trial in 300 nonpregnant women, Evidence in Adults
five different management approaches (empirical antibi- In 10 trials involving 430 women, microbiological and clin-
otics, delayed antibiotics [48 hours], targeted antibiotics ical recurrence were reduced during the active prophylaxis
based on a symptom score, a dipstick result, or a positive phase of therapy for 6–12 months duration. In general, the
result on a mid-stream urine analysis) achieved similar methodological quality of the contributing trials was poor,
symptom control. Antibiotic therapy based on dipstick reducing certainty in the results. In two studies, there was
results or delayed prescribing may reduce antibiotic no detectable difference between prophylactic antibiotic
use [21]. In evidence from 32 trials (9605 participants), use and placebo after completion of antibiotic therapy [24].
there was no difference between 3-day and a 5–10-day In older women with recurrent urinary tract infection,
antibiotic regimen, although the bacteriological failure three small trials compared antibiotics with vaginal estro-
rate may be higher with 3-day therapy [22]. Adverse gen therapy, probiotics, or d-mannose powder, antibiotic
events were not surprisingly more common in the 5–10- therapy similarly reduced microbiologically confirmed uri-
day treatment groups. Evidence in men is largely derived nary tract infection episodes during treatment, but there
from trials involving men as a small proportion of the was no apparent effect at 3 months after treatment cessa-
study population and is very uncertain. tion, although microbiological resistance may have been
increased [25].
Evidence in Children Cranberries release proanthocyanidins that prevent E.
Sixteen randomized controlled trials involving 1116 chil- coli adherence to bladder wall epithelia [26]. In a Cochrane
dren were included in a Cochrane systematic review of review published in 2008, it appeared that cranberries
antibiotics for treatment children with lower urinary tract might decrease symptomatic urinary tract infections in
infection [23]. The methodological quality of most studies women, and especially those with recurrent infections [27].
resulted in high or unclear risk of bias for many aspects of In an updated review that included 24 studies (4473 par-
trial conduct. This resulted in lower certainty in the evi- ticipants), cranberries did not prevent symptomatic uri-
dence and a weak recommendation. A lack of data reduced nary tract infection in women with recurrent infection,
confidence in understanding whether different classes of older women, children, pregnant women, or those with
antibiotics have different effects, based on outcomes such urinary tract abnormalities [28]. Participants experienced a
as persistent bacteriuria, recurrence, or reinfection. There high drop-out rate, indicating that cranberry juice may not
was no evidence that a longer course of treatment (10 days) be acceptable over the longer term. In nine studies (735
had different effects than a single dose, except for a higher participants) at high risk of bias involving women and girls
rate of initial bacterial eradication. Children and young with recurrent urinary tract infection, there was no evi-
people in the included studies were aged between 2 weeks dence that probiotics prevented symptomatic urinary tract
and 19 years. infection [29].
Evidence in Children
Prevention of Urinary Tract Infection
A Cochrane review published in 2019 included 16 studies
Recommendations for Adults involving 2036 children at often high risk of methodological
We suggest in women with two or more documented bias [30]. When analysis was limited to higher quality stud-
urinary tract infections in the previous 12 months that ies, there was evidence that antibiotics led to a modest reduc-
continuous antibiotic prophylaxis (trimethoprim/sul- tion in symptomatic recurrence in children with and without
famethoxazole, nitrofurantoin, cephalosporin) might be vesico-ureteric reflux. Prophylactic antibiotics possibly
used to prevent lower urinary tract infection although increased microbiological resistance. There was no evidence
the effect is limited to the active antibiotic intake period, in two Cochrane reviews that cranberries or probiotics pre-
the optimal duration of treatment is uncertain, and side vented urinary tract infection in children [28, 29].
Reference 309
Treatment of Asymptomatic Bacteriuria no evidence that screening for asymptomatic bacteriuria is

Nonpregnant adults and children with asymptomatic bac- effective [32].
teriuria and a normal or noninstrumented urinary tract do
not require antibiotic therapy. Nonpregnant adults do not Evidence in Children
benefit from screening for asymptomatic bacteriuria. There is insufficient evidence to be certain whether antibi-
otic therapy benefits children in the long term. The impact
Evidence in Adults on kidney injury or harmful effects of providing treatment
In a Cochrane view that included nine studies and 1614 are uncertain [33].
participants, antibiotic therapy of asymptomatic bacteriu-
ria had no detectable effect on symptomatic urinary tract Patient Perspectives
infection or complications [31]. Studies were generally of In qualitative analysis, women with symptomatic urinary
high or moderate methodological quality. Antibiotics tract infection preferred to avoid antibiotic therapy and
increase the likelihood of bacteriological cure, at the would consider nonantibiotic-based approaches. Prior
expense of increased adverse events. Limited data were experiences of antibiotics led to a greater acceptance of this
available for understanding the impact of antibiotic ther- approach, while a “delayed antibiotic” strategy was pre-
apy on microbial resistance. In data from a randomized ferred by most women but was viewed as lack of validation
controlled trial and a prospective cohort study, there was for symptoms and being taken seriously by some [34].
R
eferences
Hultcrantz, M., Rind, D., Akl, E.A. et al. (2017). The

1 10 Foxman, B., Gillespie, B., Koopman, J. et al. (2000). Risk
GRADE Working Group clarifies the construct of certainty factors for second urinary tract infection among college
of evidence. J. Clin. Epidemiol. 87: 4–13. women. Am. J. Epidemiol. 151 (12): 1194–1205.
2 Shapiro, D., Hicks, L., Pavia, A., and Hersh, A. (2014). 11 Martinez, J., Mulvey, M., Schilling, J. et al. (2000). Type 1
Antibiotic prescribing for adults in ambulatory care in the pilus-mediated bacterial invasion of bladder epithelial
USA, 2007-09. J. Antimicrob. Chemother. 69 (1): 234–240. cells. EMBO J. 19 (2812): 2803–2812.
3 Butler, C.C., Hawking, M.K., Quigley, A., and McNulty, 12 Andreson, G., Palermo, J., Schilling, J. et al. (2003).
C.A. (2015). Incidence, severity, help seeking, and Intracellular bacterial biofilm-like pods in urinary tract
management of uncomplicated urinary tract infection: a infections. Science 301 (5629): 105–107.
population-based survey. Br. J. Gen. Prac. 65 (639): 13 Little, P., Merriman, R., Turner, S. et al. (2010).
e702–e707. Presentation, pattern, and natural course of severe
4 Lundborg, C., Olsson, E., and Mölstad, S., Swedish Study symptoms, and role of antibiotics and antibiotic
Group on Antibiotic Use (2002). Antibiotic prescribing in resistance among patients presenting with suspected
outpatients: a 1-week diagnosis-prescribing study in 5 uncomplicated urinary tract infection in primary care:
counties in Sweden. Scand. J. Infect. Dis. 359: 442–448. observational study. BMJ 340: b5633.
5 Hurlbut Tr and Littenberg, B. (1991). The diagnostic 14 Nickel, J., Lee, J., Grantmyre, J., and Polygenis, D. (2005).
accuracy of rapid dipstick tests to predict urinary tract Natural history of urinary tract infection in a primary
infection. Am. J. Clin. Pathol. 96 (5): 582–588. care environment in Canada. Can. J. Urol. 12 (4):
6 Yamamoto, S., Tsukamoto, T., Terai, A. et al. (1997). 2728–2737.
Genetic evidence supporting the fecal-perineal-urethral 15 Ikäheimo, R., Siitonen, A., Heiskanen, T. et al. (1996).
hypothesis in cystitis caused by Escherichia coli. J. Urol. 157 Recurrence of urinary tract infection in a primary care
(3): 1127–1129. setting: analysis of a 1-year follow-up of 179 women. Clin.
7 Ronald, A. (2002). The etiology of urinary tract infection: Infect. Dis. 22 (1): 91–99.
traditional and emerging pathogens. Am. J. Med. 113 (1 16 Foxman, B. (2011). Recurring urinary tract infection:
Suppl 1): 14–19. incidence and risk factors. Am. J. Public Health 80 (3):
8 Jakobsson, B., Esbjörner, E., and Hansson, S. (1999). 331–333.
Minimum incidence and diagnostic rate of first urinary 17 Hooton, T.M., Scholes, D., Stapleton, A.E. et al. (2000). A
tract infection. Pediatrics 104 (2 Pt 1): 222–226. prospective study of asymptomatic bacteriuria in sexually
9 Foxman, B. and Brown, P. (2003). Epidemiology of urinary active young women. N. Engl. J. Med. 343 (14): 992–997.
tract infections: transmission and risk factors, incidence, 18 Shaikh, N., Ewing, A.L., Bhatnagar, S., and Hoberman, A.
and costs. Infect. Dis. Clin. North Am. 17: 227–241. (2010). Risk of renal scarring in children with a first
urinary tract infection: a systematic review. Pediatrics 126 implications for Escherichia coli-uroepithelial attachment.
(6): 1084–1091. J. Med. Food 2: 259–270.
19 Soylu, A., Demir, B.K., Turkmen, M. et al. (2008). 27 Jepson, R. and Craig, J. (2008). Cranberries for preventing
Predictors of renal scar in children with urinary infection urinary tract infections. Cochrane Database Syst. Rev. (1)
and vesicoureteral reflux. Pediatr. Nephrol. 23 (12): (Art. No.: CD001321).
2227–2232. 28 Jepson, R., Williams, G., and Craig, J. (2012). Cranberries
20 Zalmanovici Trestioreanu, A., Green, H., Paul, M. et al. for preventing urinary tract infections. Cochrane
(2010). Antimicrobial agents for treating uncomplicated Database Syst. Rev. (10) (Art. No.: CD001321).
urinary tract infection in women. Cochrane Database 29 Schwenger, E., Tejani, A., and Loewen, P.S. (2015).
Syst. Rev. (10) (Art. No.: CD007182). Probiotics for preventing urinary tract infections in adults
21 Little, P., Moore, M.V., Turner, S. et al. (2010). and children. Cochrane Database Syst. Rev. (12) (Art. No.:
Effectiveness of five different approaches in management CD008772).
of urinary tract infection: randomised controlled trial. 30 Williams, G. and Craig, J. (2019). Long-term antibiotics
BMJ 340: c199. for preventing recurrent urinary tract infection in
22 Milo, G., Katchman, E., Paul, M. et al. (2010). Duration of children. Cochrane Database Syst. Rev. (4) (Art. No.:
antibacterial treatment for uncomplicated urinary tract CD001534).
infection in women. Cochrane Database Syst. Rev. (2) 31 Zalmanovici Trestioreanu, A., Lador, A., Sauerbrun-
(Art. No.: CD004682). Cutler, M., and Leibovici, L. (2015). Antibiotic treatment
23 Fitzgerald, A., Mori, R., Lakhanpaul, M., and Tullus, K. for asymptomatic bacteriuria. Cochrane Database Syst.
(2012). Antibiotics for treating lower urinary tract Rev. (4) (Art. No.: CD009534).
infection in children. Cochrane Database Syst. Rev. (8) 32 Lin, K. and Fajardo, K., and US Preventative Services
(Art. No.: CD006857). Task Force (2008). Screening for asymptomatic
24 Albert, X., Huertas, I., Pereiro, I. et al. (2004). Antibiotics bacteriuria in adults: evidence for the US Preventive
for preventing recurrent urinary tract infection in Services Task Force reaffirmation recommendation
non-pregnant women. Cochrane Database Syst. Rev. (3) statement. Ann. Intern. Med. 149 (1): W20–W24.
(Art. No.: CD001209). 33 Fitzgerald, A., Mori, R., and Lakhanpaul, M. (2012).
25 Ahmen, H., Davies, F., Francis, N. et al. (2017). Long- Antibiotics for covert bacteriuria in children. Cochrane
term antibiotics for prevention of recurrent urinary tract Database Syst. Rev. (2) (Art. No.: CD006943).
infection in older adults: systematic review and meta- 34 Leydon, G., Turner, S., and Smith, H., and on behalf of
analysis of randomised trials. BMJ Open 7: e015233. the UTIS team (2010). Women’s views about management
26 Pinzón-Arango, P., Liu, Y., and Camesano, T. (2009). Role and cause of urinary tract infection: qualitative interview
of cranberry on bacterial adhesion forces and study. BMJ 340: c279.
311
21a
Toxic Nephropathies: Environmental Agents and Metals

Joseph B. Pryor1, William M. Bennett2, and Ali Olyaei3
1
Department of Internal Medicine, University of Washington, Seattle, WA, USA
2
Legacy Transplant Services, Legacy Good Samaritan Medical Center, Portland, OR, USA
3
Department of Pharmacy, Oregon State University/Oregon Health & Science University, Portland, OR, USA
I ntroduction lead poisoning usually occurred because of the ingestion of

lead-based paint in deteriorated housing. Fanconi syn-
Environmental nephrotoxins include materials found pre- drome (aminoaciduria and phosphaturia) has been
dominantly in the workplace [1–3] (Table 21a.1). Many of reported in children with acute lead toxicity. This finding is
these chemicals have no biological value for the treatment consistent with transient proximal tubular reabsorptive
of diseases and are toxic to many tissues and cells. The clin- defects [8]. Fanconi syndrome is regularly reproducible in
ically important industrial nephrotoxins include lead, mer- experimental animals exposed to sufficiently high doses of
cury, cadmium, silica, a variety of (sometimes poorly lead. Lead-induced Fanconi syndrome is reversible with
defined) organic hydrocarbons (aliphatic, aromatic, and chelating therapy (CaNa2–EDTA) and removal from
halogenated), uranium, chromium, and arsenic [4]. Most exposure [9].
of these nephrotoxins cause biological exposure through In acute lead toxicity states, there is a clear relationship
inhalation and ingestion of contaminated food and drinks. between plasma level and toxic exposure. The diagnosis
Although low concentrations of heavy metals are present and treatment are much more difficult in the long-term,
in the environment naturally, there is a greater exposure to low-dose exposure of lead in adults. In addition, many
these nephrotoxins in industrial and developed countries. comorbid conditions might mask the signs and symptoms
For many of these nephrotoxins [5], it should be recog- of chronic lead toxicity [5, 10]. Chronic lead exposure is
nized that there are no randomized controlled trials for the also associated with hypertension, gout, and tubulo-
management of kidney diseases caused by these environ- interstitial nephritis, which is histologically indistinguish-
mental agents. The cornerstone of treatment is prevention able from hypertensive nephrosclerosis. In the past, lead
and limiting further exposure (Table 21a.2). nephropathy was identified in lead workers or those work-
ing with moonshiner (illegal use of distilled liquor) after
chronic exposure [2]. Blood lead in these workers was often
above 40 μg/dl, which is the exposure limit promulgated in
H
eavy Metals
the federal occupational lead standard. It was widely
assumed that blood lead levels of 60 μg/dl or higher, sus-
Lead
tained for many years, were required to induce lead
Acute lead poisoning continues to rise in most developed nephropathy. However, recent epidemiologic evidence
countries and in particular in socio-economically deprived using blood and bone lead measurements (by noninvasive
communities [6]. Acute lead poisoning causing abdominal in vivo K X-ray fluorescence) indicate that the adverse
pain, anemia, and encephalopathy may occur days or effects of lead on blood pressure and kidney function occur
weeks after exposure to lead. Blood lead levels exceeding at much lower levels than previously recognized. Ninety-
70 μg/dl (current mean blood lead in the USA is <2 μg/dl) five percent of lead in the body is retained in bone, with a
are associated with acute lead toxicity [7]. Acute childhood biological half-life approximating two decades, whereas
312 Toxic Nephropathies: Environmental Agents and Metals
Table 21a.1 Environmental nephrotoxins.
Pathology
Substance ATN TIN G Comments
Metals
Pb + + − Fanconi syndrome, tubular proteinuria, hypertension, gout
Cd − + − Fanconi syndrome, Ca++ wasting, tubular proteinuria
a
Hg + + + G → immune, genetic control, nephrotic syndrome; HgCl2 → ATN
As + + − AsH3, hemolysis
Cra + + − Cr+++ → tubular proteinuria, nasal perforation, cancer;
Cr+++++ → ATN
U + + − Tubular proteinuria
Si − − + Immune adjuvant, immune complexes, ANCA, Wegener’s granulomatosis
Organics
CCl4 + + − Liver damage, alcohol enhanced, dry cleaner
Toluene − − − Tubular proteinuria, glue sniffing
Solvents − − + Immune G, tubular proteinuria
Selected drugsb
Aminoglycosides + + − ARF after 8 days, lysosomal accumulation S1 and S2
Amphotericin B − + − Distal RTA, K+↓, Mg++↓, Uosm ↓, tubular calcification
Penicillins − + − Allergic component, acute, after 15 days, interstitial edema, lymphocytic infiltration
Vancomycin + + − ARF
Sulfonamides − + − Intra-luminal crystals, kidney stones, ARF, microhematuria, soluble in alkaline urine
Acyclovir − + − Intraluminal crystal deposition, flank pain, microhematuria
Cyclosporin A Tacrolimus + + − ARF, vasomotor, proximal tubule vacuolization, striped fibrosis, arteriolar hyalinosis
Gold salts − + + Membranous glomerulopthy, gold in tubular epithelium, separate tubular disease
d-penicillamine − − − Membranous glomerulopthy, vasculitis
cis-platinum + + − ARF, usually reversible
Mitomycin − + − Thrombosis, hemolytic uremic syndrome
ANCA, anti-neutrophil cytoplasmic antibody; ARF, acute renal failure; ATN, acute tubular necrosis; G, glomerulopathy; RTA, renal tubular
acidosis; TIN, tubular interstitial nephritis; Uosm, urine osmolality.
a
Depends on valance and organic form.
b
Often in clinically complex settings with multicausal renal damage.
blood lead has a biological half-life of about 30 days [11]. also reported a significant association between low-level
Therefore, bone lead reflects cumulative lead absorption, lead exposure and serum creatinine [7, 12]. The adverse
whereas blood lead primarily reflects recent exposure. effects of low-level lead exposure on renal function are fur-
Blood lead and bone lead predict blood pressure even ther supported by longitudinal observations. Because lead
when both are within the range traditionally considered does not accumulate in patients with kidney failure of non-
“normal.” Epidemiologic evidence that low-level lead lead etiology, these observations point to low-level lead
absorption (blood lead of <10 μg/dl) has deleterious effects exposure as a contributor to hypertension and kidney
on blood pressure and renal function was obtained from disease.
National Health and Nutrition Evaluation Survey III, Bone lead levels were found to be significant predictors
1988–1994. Analysis of data on over 15 000 Americans of hypertension in community-exposed men [13]. The
showed that hypertensive people had significantly higher mean blood lead level in these men was 6 μg/dl. Similarly,
blood lead levels (4.21 vs. 3.30 μg/dl) than people without bone lead levels were found to be a significant predictor of
hypertension [11, 12]. Two large, cross-sectional studies hypertension in nurses [14]. A study in pregnant women
Heavy Metal 313
Table 21a.2 Treatment of environmental exposures. averaging 7 g over 3 months, significantly reduced the rate
of progression of kidney failure compared to untreated dia-
Metals Treatment betic controls. These results suggest that reducing even
very low levels of lead may improve GFR in a variety of
Pb CaNa2–EDTA, dimercaprol, succimer, nonlead-related kidney diseases. Moreover, if the benefi-
or penicillamine
cial effect of CaNa2–EDTA is, in fact, due to the removal of
Cd EDTA and supportive measure
lead from the body, these observations indicate that there is
Hg1 Dimercaprol, succimer no threshold for the deleterious effects of lead on
As Dimercaprol, succimer kidneys [18].
Cr1 Supportive measures and avodiance However, the mechanism of the beneficial effect of
U Supportive measures and avodiance CaNa2–EDTA on the kidneys is not clear. Improvement
Si Supportive measures and avodiance and management of comorbid conditions may play an
important role in the improvement of hypertension and
Organics
reduced GFR. Prevention remains the most effective
CCl4 Supportive measures and avodiance method of treatment of lead poisoning. Chelation tran-
Toluene Supportive measures and avodiance siently increases the rate of removal of lead, causing the
Solvents Supportive measures and avodiance urinary excretion of up to several milligrams of lead in
heavily exposed individuals within a few days. If, on the
CaNa2–EDTA, calcium ethylenediaminetetraacetic acid.
other hand, excessive exposure is terminated, the expected
urinary excretion of 100–200 μg of lead/day will achieve a
with a geometric mean blood lead level of 1.9 μg/dl found negative balance of 35–70 mg/year. Because a lifetime of
increases in bone lead were associated with an increased occupational exposure can result in bone lead stores of
risk of pregnancy hypertension [15]. Increases in both 500 mg or more, a course of chelation therapy (e.g. 1–2 g
diastolic and systolic blood pressures in pregnant women CaNa2–EDTA) eliminates only about 0.2–0.4% of the body
were significantly associated with blood lead, and the burden, whereas terminating exposure reduces bone stores
major portion of the effect was found with blood lead levels by about 10% per year. Although it is not feasible to elimi-
under 5 μg/dl. These observations on adverse effects at nate all lead intake, similar calculations for individuals
blood lead levels below 10 μg/dl in diverse groups indicate with low-level exposure indicate that, in the long term, pre-
that, even at levels traditionally considered acceptable, vention is more effective than chelation in lowering the
important adverse health effects occur. body lead stores. In summary, chelation therapy is justified
Treatment of lead-induced hypertension and kidney fail- in cases of symptomatic lead poisoning or when the blood
ure with calcium ethylenediaminetetraacetic acid (CaNa2– lead exceeds about 70 μg/dl, but when no symptom end
EDTA; also called calcium disodium versenate or edetate point is clearly defined, chelation for blood lead of <70 μg/
calcium disodium) may improve the glomerular filtration dl is not usually justified [8].
rate (GFR) or reduce the rate of progression of kidney fail-
ure. However, there is no clinical data to support the revers-
Mercury
ibility of established interstitial nephritis after treatment
with CaNa2–EDTA [16, 17]. It is reasonable to do an EDTA Exposure to mercury can be through inhalation, oral, or
lead mobilization study and 24-hour urine collection in dermal. Mercury is a natural element with atomic weight
patients with unknown causes of hypertension, gout, and of 200.6. There are many form of organic and inorganic
potential exposure to lead. Therapeutic benefits have been mercury in the pharmaceutical industry. Mercury is unique
noted in patients with acute kidney injury from much in that it exists as a liquid at room temperature. Inorganic
lower levels of lead exposure. Lin et al. [11] treated patients mercury mostly accumulates in the kidney while organic
with chronic kidney disease (CKD) from lead exposure mercury mostly accumulates in the central nervous sys-
(mean creatinine 2.1 mg/dl, mean blood lead 5.3 μg/dl) tem. A number of studies have suggested that patients/ani-
with CaNa2–EDTA, 4–13 g intravenously, over 2 years. The mals with reduced renal function are at greater risk of
treated group had a modest improvement in GFR averag- mercury renal toxicity [19]. The salt of mercury thimerosal
ing 3.4 ml/min, whereas untreated controls had a decrease is still used in vaccines as a preservative. The toxicity of
in GFR of 1.0 ml/min during this period. The same labora- mercury depends on both its chemical form and the route
tory demonstrated that low levels of lead absorption corre- of absorption. Although preferentially accumulated in the
late significantly with the rate of progression of kidney kidney, neurologic disease, but not kidney disease, regu-
failure in type 2 diabetics and that CaNa2–EDTA therapy, larly follows exposure to elemental mercury. Once in the
environment, elemental mercury undergoes biotransfor- among the nonsmoking population [24]. However, in the
mation to both organic and inorganic salts, which are smoking population, inhalation is the primary route of expo-
absorbed by living organisms and thus enter the food sure as tobacco plants accumulate Cd in high concentra-
chain. Mercury at the cellular level accumulates at proxi- tions. Verdugo et al. investigated the concentration of Cd in
mal tubular cells via OAT-1 and 3. Mercury may alter the cigarettes and found that 1–2 μg of Cd were present in each
function and structure of nephrons. After exposure to a cigarette, of which 5% is absorbed by the lungs [25–27].
toxic dose of mercury exposure, mitochondrial swelling Once Cd is absorbed systemically, it accumulates in mul-
and dilation of the endoplasmic reticulum of proximal tiple organs, including bone and liver, like other heavy
tubular cell can be identified [20]. metals, but also in the kidney. As much as 50% of the total
In addition to diuresis and acute tubular necrosis, mer- body stores are found in the kidney, which explains the sig-
cury has been sporadically reported to cause nephrotic syn- nificant renal adverse effects associated with Cd [28, 29].
drome in what has usually been considered an idiosyncratic Cadmium deposition in the bone results in a long half-life,
response. Observations in rats may provide a framework requiring decades for complete elimination. Complete
for understanding mercury-induced glomerular disease in elimination, however, is difficult because of the chronic
humans. In 1971, Bariety et al. [21] reported that multiple low-dose exposure that may initially go unnoticed until
subcutaneous injections of HgCl2 in specific rat strains, in large burdens accumulate.
doses too small to produce tubular necrosis, induced mem- While renal handling of Cd is complex and not fully
branous nephropathy. Kidney disease characterized by glo- understood, it is known that Cd has a high affinity for sulf-
merular deposition of immune complexes and heavy hydrl moieties, and is suspected to be transported along
proteinuria developed in about 2 months. Subsequent stud- with organic anions/cations and amino acids or oligopep-
ies showed that the immune response is actually biphasic; ties [30]. Additionally Cd utilizes other metal transports
immune complex deposition is preceded by antiglomerular such as divalent metal transporter (DMT1) and zinc trans-
basement membrane antibody and complement deposi- porters such as ZIP8 and ZIP14 to cross cell membranes.
tion. The response to mercury in the rat is under precise Despite these routes, it is suspected that most Cd enters the
genetic control. As little as 0.05 mg/kg of body weight will proximal tubular epithelial cells after being transformed to
elicit immunologically mediated glomerular disease in Cd-metallothionein [MT] by the liver. Cd-MT is a nontoxic
selected strains. As in humans, mercury-induced glomeru- byproduct that usually accumulates in the liver, but may be
lar disease in rats is self-limited [22]. released through Cd-induced hepatocellular necrosis,
There are no randomized clinical trials for the treatment resulting in serum Cd-MT, which is subsequently absorbed
of acute tubular necrosis or nephrotic syndrome due to mer- by the proximal tubular epithelium via endocytosis [30,
cury. Chelation therapy has been employed for excessive 31]. Once in the kidney, Cd disassociates from MT in the
mercury exposure, often defined as more than 5 μg Hg/dl in lysosomes where it may lead to renal adverse effects.
blood or 50 μg Hg/l in urine, but evidence for the effective- Accumulation of Cd in the kidney typically causes polyu-
ness of chelation therapy is lacking. Traditional use of BAL ria, reduced GFR, and Fanconi syndrome [28].
(British antilewisite, dimercaprol) for chelating mercury is Thought rarely tested, cadmium monitoring can be
currently being replaced by its soluble congeners, dimercap- done through urine and or blood assays. Serum levels are
topropane sulfonic acid (unithiol or Dimaval) and succimer helpful for acute exposures (within the last month),
(dimercaptosuccinic acid, Chemet). Dimercaptopropane whereas urinary sampling is utilized when chronic expo-
sulfonic acid is the chelator of choice for inorganic mercury sure is suspected. Given that most people are exposed
intoxication, whereas succimer is the agent of choice for chronically at low dose, urinary excretion of Cd is a more
organic mercury [22]. Case reports suggest that in the pres- reliable indicator, especially when it comes to renal and
ence of severe kidney failure the mercury-chelate complex total body burden [23, 32]. Monitoring for kidney damage
can be removed by hemodialysis or hemofiltration. typically involves nonspecific signs including urinary bio-
markers, β2-microglobulin, N-acetyl-β-d -glucosamidase
(NAG), kidney injury moleculre-1 (KIM-1), and cystatin
Cadmium
C [28], the latter being a more reliable biomarker for Cd-
Cadmium (Cd) is an environmentally prevalent heavy metal induced damage [33]. Prolonged Cd exposure typically
and pollutant that is known to have multiple adverse effects manifests as proximal tubular microvilli damage, result-
on the human body and is most commonly found in batter- ing in reduced transporters and reabsorption. If severe
ies, plastics, pigments, and fertilizers [23]. Given its role in enough, Fanconi syndrome may be diagnosed, which is
fertilizers, Cd can accumulate in both plants and animals. characterized by aminoaciduria, glucosuria, hypercalciu-
As a result, diet becomes the primary route of exposure ria, and hyperphosphaturia. While kidney damage has
Silico 315
multiple negative downstream effects, nonrenal adverse effluents, the syndrome, known as itai-itai or “ouch-
events occur such as decreased synthesis of activated vita- ouch” disease, primarily afflicted postmenopausal, mul-
min D (1,25-dihydroxycholecalciferol) and liver damage. tiparous women. Sustained deficiencies in iron, zinc,
Multiple epidemiological studies have documented a calcium, and vitamin D rendered these women particu-
dose-related correlation to renal insufficiency, but given larly vulnerable to cadmium toxicity [45]. Women with
the high incidence and prevalence of CKD in the world, itai-itai disease tended to have reduced GFR, anemia,
one must have a high clinical suspicion to diagnose Cd- lymphopenia, and hypotension as well as osteomalacia.
induced CKD. This most typically occurs in places with They exhibited a waddling gait, short stature, anemia, gly-
high incidence of Cd in the environment such as Sri cosuria, and elevated serum alkaline phosphatase levels.
Lanka [34]. As CKD develops and GFR declines, there is Hypertension was absent. β2-microglobulin excretion
a reduced capacity to clear the nephrotoxin, exacerbating exceeded the normal maximum (1 mg/g of creatinine) by
preexisting renal disease [35–38]. 100-fold, and GFR was substantially reduced in the most
Excretion occurs very slowly given the distribution and severely affected individuals. Long-term follow-up stud-
deposition in the bone, but risk factors such as age, gender, ies showed that excessive urinary excretion of the low-
and diabetes decrease excretion. One study reported molecular-weight protein β2-microglobulin predicts the
decreased urinary Cd was associated with diabetic rats later development of kidney failure in patients with itai-
compared to nondiabetic rats [4]. Although not fully itai disease and that kidney damage progresses even after
understood, some studies have shown decreased insulin exposure has ceased. Succimer is effective for chelation in
expression in Cd exposure, but this is confounded because acute cadmium poisoning [22]. No agent has been found
both Cd and diabetes cause renal disease, which has been to be effective for mobilizing hepatic or renal stores of
shown to decrease excretion. Interestingly, studies assess- cadmium [2, 46].
ing the differential burdens of Cd between men and women
found women to have greater burdens compared to men.
Further investigation revealed that both blood and urinary S
ilicon
Cd levels correlate with age and iron stores. On average
women tend to have lower iron stores than men and as a Silicon is a semimetal found as the dioxide (SiO2, silicon
result of the lower iron stores, they upregulate DMT1 in an dioxide) in 28% of the earth’s crust. It has been reported
attempt to absorb more iron. This also facilitates Cd uptake to induce interstitial nephritis by direct deposition of
through the same transporter, ultimately predisposing crystalline material in the renal parenchyma [47] and by
those with low iron to increased Cd uptake [39–41]. immunologic mechanisms acting as an adjuvant to stim-
Once cadmium has been detected in the body at toxic lev- ulate the immune response [48]. Tubular proteinuria is
els, treatment is primarily through chelation therapy with found in workers exposed to silica dust. The odds of a
succimer or EDTA. Further research still has to be done, but sandblaster developing end-stage renal disease is 3.8
other metals such as zinc and magnesium have been shown compared to matched controls. In the accelerated form
to limit renal toxicity. While the mechanism of action is of silicosis known as silicoproteinosis, silicon dust
likely multifactorial, it has been shown to prevent enzyme appears to be indirectly responsible for rapidly progres-
alteration as well as saturate metal transporters. This hin- sive, immune complex-mediated focal glomerulosclero-
ders the further uptake of Cd into both the cell and the body. sis [49]. In addition to severe pulmonary disease, these
While no agents are particularly effective for removing Cd patients develop an overwhelming autoimmune response
from the liver, zinc has been shown to help transform Cd to that frequently includes lupus erythematosus [23] or
the nontoxic form (Cd-MT) and reduce cellular apoptosis rheumatoid arthritis (Caplan’s syndrome). Glomerular
induced by Cd [42–44]. Overall, to diagnose Cd toxicity, one disease, sometimes in association with silica-induced
must have a high suspicion for it as exposure is commonly systemic sclerosis, systemic lupus erythematosus, and
low dose and chronic. Despite limited interventions, diag- small vessel vasculitis, has also been described as a result
nosing cadmium toxicity enables environmental assessment of exposure to silica dust independent of silicosis [24].
in an attempt to reduce exposure and limit further damage. Antineutrophil cytoplasmic antibody (c-ANCA)-positive
Wegener’s granulomatosis has been associated with
Itai-Itai Disease exposure to silica dust as well as to silica-containing
In Japan, a painful bone disease associated with pseudof- compounds, such as grain dust. No specific therapies
ractures due to cadmium-induced renal calcium wasting have been reported for silica-induced glomerular disease
was recognized in the 1950s. Attributed to local contamina- other than those in current use for immunologically
tion of food staples by river water polluted with industrial medicated glomerular disease [3].
Solvent Nephropathy pulmonary hemorrhage, i.e. Goodpasture’s syndrome, but

Halogenated hydrocarbons have often been implicated in later reports of solvent nephropathy have included many
the induction of acute tubular necrosis or Fanconi syn- different types of glomerulonephritis [50].
drome in both humans and experimental animals. Low- The etiologic role of solvents remains controversial
level occupational absorption by inhalation of volatile because the dose and exact chemical composition of indus-
hydrocarbons or absorption through the skin may also trial solvents are usually unknown. Moreover, of the thou-
induce tubular proteinuria, which does not necessarily sig- sands of workers exposed, very few develop immunologically
nify the presence of clinical kidney failure. mediated glomerular disease. The genetic and environmen-
At least 40 case-control studies have examined the rela- tal factors that make specific individuals susceptible to sol-
tionship between glomerulonephritis and exposure to vent nephropathy have not been delineated. The experimental
organic solvents. A number of these studies concluded that mercury-induced immunologically mediated glomerular
patients with chronic glomerulonephritis had been exposed disease in rodents described above may provide a model for
to organic solvents (aliphatic and aromatic) more fre- understanding solvent nephropathy in humans. No specific
quently than patients with other diseases. Initially, solvent therapies have been recommended for solvent nephropathy.
nephropathy was associated with antiglomerular basement Avoidance of exposure to volatile hydrocarbons and their
membrane antibody-mediated glomerulonephritis and derivatives remains an essential preventive approach.
R
eferences
1 Perazella, M.A. (2018). Pharmacology behind common 11 Lin, J.L., Lin-Tan, D.T., Hsu, K.H., and Yu, C.C. (2003).
drug nephrotoxicities. Clin. J. Am. Soc. Nephrol. 13 (12): Environmental lead exposure and progression of chronic
1897–1908. renal diseases in patients without diabetes. N. Engl. J
2 Orr, S.E. and Bridges, C.C. (2017). Chronic kidney disease Med. 348 (4): 277–286.
and exposure to nephrotoxic metals. Int. J. Mol. Sci. 18 (5): 12 Staessen, J.A., Lauwerys, R.R., Buchet, J.P. et al. (1992).
1039. Impairment of renal function with increasing blood lead
3 Steenland, K., Sanderson, W., and Calvert, G.M. (2001). concentrations in the general population. The Cadmibel
Kidney disease and arthritis in a cohort study of Study Group. N. Engl. J Med. 327 (3): 151–156.
workers exposed to silica. Epidemiology 12 (4): 13 Cheng, Y., Schwartz, J., Sparrow, D. et al. (2001). Bone
405–412. lead and blood lead levels in relation to baseline blood
4 Lauwerys, R., Bernard, A., and Cardenas, A. (1992). pressure and the prospective development of
Monitoring of early nephrotoxic effects of industrial hypertension: the Normative Aging Study. Am. J.
chemicals. Toxicol. Lett. 64–65 Spec No: 33–42. Epidemiol. 153 (2): 164–171.
5 Sabath, E. and Robles-Osorio, M.L. (2012). Renal health 14 Korrick, S.A., Hunter, D.J., Rotnitzky, A. et al. (1999).
and the environment: heavy metal nephrotoxicity. Lead and hypertension in a sample of middle-aged
Nefrologia 32 (3): 279–286. women. Am. J. Public Health 89 (3): 330–335.
6 Linakis, J.G. (Jan 1995). Childhood lead poisoning. Rhode 15 Rothenberg, S.J., Kondrashov, V., Manalo, M. et al. (2002).
Island Med. 78 (1): 22–26. Increases in hypertension and blood pressure during
7 Kim, R., Rotnitsky, A., Sparrow, D. et al. (1996). A pregnancy with increased bone lead levels. Am. J.
longitudinal study of low-level lead exposure and Epidemiol. 156 (12): 1079–1087.
impairment of renal function. The Normative Aging 16 Gracia, R.C. and Snodgrass, W.R. (2007). Lead toxicity
Study. JAMA 275 (15): 1177–1181. and chelation therapy. Am. J Health Syst. Pharm. 64 (1):
8 Sachdeva, C., Thakur, K., Sharma, A., and Sharma, K.K. 45–53.
(2018). Lead: tiny but mighty poison. Indian J. Clin. 17 Smith, D. and Strupp, B.J. (2013). The scientific basis for
Biochem. 33 (2): 132–146. chelation: animal studies and lead chelation. J. Med.
9 Barnett, L.M.A. and Cummings, B.S. (2018). Toxicol. 9 (4): 326–338.
Nephrotoxicity and renal pathophysiology: a 18 Ding, Y., Vaziri, N.D., and Gonick, H.C. (Feb 1998).
contemporary perspective. Toxicol. Sci. 164 (2): 379–390. Lead-induced hypertension. II. Response to sequential
10 Payton, M., Hu, H., Sparrow, D. et al. (1993). Relation infusions of L-arginine, superoxide dismutase, and
between blood lead and urinary biogenic amines in nitroprusside. Environ. Res. 76 (2): 107–113.
community-exposed men. Am. J. Epidemiol. 138 (10): 19 Hu, X.F., Singh, K., and Chan, H.M. (2018). Mercury
815–825. exposure, blood pressure, and hypertension: a systematic
Reference 317
review and dose-response meta-analysis. Environ. Health pro-inflammatory signaling pathways and programed cell
Perspect. 126 (7): 076002. death. Arch. Toxicol. 89 (7): 1057–1070.
20 Khan, H., Singh, R.D., Tiwari, R. et al. (2017). Mercury 33 Cherian, M.G. and Nordberg, M. (1983). Cellular
exposure induces cytoskeleton disruption and loss of adaptation in metal toxicology and metallothionein.
renal function through epigenetic modulation of MMP9 Forensic Toxicol. 28 (1–2): 1–15.
expression. Forensic Toxicol. 386: 28–39. 34 Liu, J., Liu, Y., Habeebu, S.S., and Klaassen, C.D. (1999).
21 Bariety, J., Druet, P., Laliberte, F., and Sapin, C. (1971). Metallothionein-null mice are highly susceptible to the
Glomerulonephritis with γ-and β1C-globulin deposits hematotoxic and immunotoxic effects of chronic CdCl2
induced in rats by mercuric chloride. Am. J. Pathol. 65 (2): exposure. Toxicol. Appl. Pharmacol. 159 (2): 98–108.
293–302. 35 Galazyn-Sidorczuk, M., Brzoska, M.M., Jurczuk, M., and
22 Andersen, O. (2004). Chemical and biological Moniuszko-Jakoniuk, J. (2009). Oxidative damage to
considerations in the treatment of metal intoxications by proteins and DNA in rats exposed to cadmium and/or
chelating agents. Mini Rev. Med. Chem. 4 (1): 11–21. ethanol. Chem. Biol. Interact. 180 (1): 31–38.
23 Odabaei, G., Chatterjee, D., Jazirehi, A.R. et al. (2004). 36 Matovic, V., Buha, A., Ethukic-Cosic, D., and Bulat, Z.
Raf-1 kinase inhibitor protein: structure, function, (Apr 2015). Insight into the oxidative stress induced by
regulation of cell signaling, and pivotal role in apoptosis. lead and/or cadmium in blood, liver and kidneys. Food
Adv. Cancer Res. 91: 169–200. Chem. Toxicol. 78: 130–140.
24 Singh, R.D., Tiwari, R., Khan, H. et al. (2015). Arsenic 37 So, K.Y. and Oh, S.H. (2016). Cadmium-induced heme-
exposure causes epigenetic dysregulation of IL-8 oxygenase-1 expression plays dual roles in autophagy and
expression leading to proneoplastic changes in kidney apoptosis and is regulated by both PKC-delta and PKB/
cells. Toxicol. Lett. 237 (1): 1–10. Akt activation in NRK52E kidney cells. Forensic Toxicol.
25 Sinha, M., Manna, P., and Sil, P.C. (2008). Arjunolic acid 370: 49–59.
attenuates arsenic-induced nephrotoxicity. 38 Chen, Z., Gu, D., Zhou, M. et al. (2016). Regulatory role
Pathophysiology 15 (3): 147–156. of miR-125a/b in the suppression by selenium of
26 Verdugo, M., Ogra, Y., and Quiroz, W. (2016). cadmium-induced apoptosis via the mitochondrial
Mechanisms underlying the toxic effects of antimony pathway in LLC-PK1 cells. Chem. Biol. Interact. 243:
species in human embryonic kidney cells (HEK-293) and 35–44.
their comparison with arsenic species. J. Toxicol. Sci. 41 39 Mueller, P.W., Price, R.G., and Finn, W.F. (1998). New
(6): 783–792. approaches for detecting thresholds of human
27 Roy, A., Manna, P., and Sil, P.C. (Oct 2009). Prophylactic nephrotoxicity using cadmium as an example. Environ.
role of taurine on arsenic mediated oxidative renal Health Perspect. 106 (5): 227–230.
dysfunction via MAPKs/ NF-kappaB and mitochondria 40 Thomas, L.D., Elinder, C.G., Wolk, A., and Akesson, A.
dependent pathways. Free Radical Res. 43 (10): 995–1007. (2014). Dietary cadmium exposure and chronic kidney
28 Nordberg, G.F., Goyer, R., and Nordberg, M. (Apr 1975). disease: a population-based prospective cohort study of
Comparative toxicity of cadmium-metallothionein and men and women. Int. J. Hyg. Environ. Health 217 (7):
cadmium chloride on mouse kidney. Arch. Pathol. 99 (4): 720–725.
192–197. 41 Jayatilake, N., Mendis, S., Maheepala, P., and Mehta, F.R.
29 Thévenod, F. (2003). Nephrotoxicity and the proximal (2013). Chronic kidney disease of uncertain aetiology:
tubule. Nephron Physiol. 93 (4): 87–93. prevalence and causative factors in a developing country.
30 Cui, X. and Okayasu, R. (2008). Arsenic accumulation, BMC Nephrol. 14: 180.
elimination, and interaction with copper, zinc and 42 Mortensen, M.E., Wong, L.Y., and Osterloh, J.D. (2011).
manganese in liver and kidney of rats. Food Chem. Smoking status and urine cadmium above levels
Toxicol. 46 (12): 3646–3650. associated with subclinical renal effects in US adults
31 Olsson, I.M., Bensryd, I., Lundh, T. et al. (2002). without chronic kidney disease. Int. J. Hyg. Environ.
Cadmium in blood and urine – impact of sex, age, dietary Health 214 (4): 305–310.
intake, iron status, and former smoking-association of 43 Sabolic, I., Ljubojevic, M., Herak-Kramberger, C.M., and
renal effects. Environ. Health Perspect. 110 (12): Brown, D. (Dec 2002). Cd-MT causes endocytosis of
1185–1190. brush-border transporters in rat renal proximal tubules.
32 Gong, X., Ivanov, V.N., Davidson, M.M., and Hei, T.K. Am. J Physiol. Ren. Physiol. 283 (6): F1389–F1402.
(2015). Tetramethylpyrazine (TMP) protects against 44 Huang, M., Choi, S.J., Kim, D.W. et al. (2009). Risk
sodium arsenite-induced nephrotoxicity by suppressing assessment of low-level cadmium and arsenic on the
ROS production, mitochondrial dysfunction, kidney. J. Toxicol. Environ. Health A 72 (21–22): 1493–1498.
5 Nishijo M, Nakagawa H, Suwazono Y, Nogawa K, Kido T.

4 48 Parks, C.G., Conrad, K., and Cooper, G.S. (1999).
Causes of death in patients with Itai-itai disease suffering Occupational exposure to crystalline silica and
from severe chronic cadmium poisoning: a nested case– autoimmune disease. Environ. Health Perspect. 107 (Suppl
control analysis of a follow-up study in Japan. BMJ Open 5): 793–802.
2017;7(7):e015694. 49 Osorio, A.M., Thun, M.J., Novak, R.F. et al. (Mar 1987).
46 Satarug, S. (2018). Dietary cadmium intake and its effects Silica and glomerulonephritis: case report and review of
on kidneys. Toxics 6 (1): 15. the literature. Am. J. Kidney Dis. 9 (3): 224–230.
47 Dobbie, J.W. and Smith, M.J. (Jan 1982). Silicate 50 Qin, W., Xu, Z., Lu, Y. et al. (2012). Mixed organic
nephrotoxicity in the experimental animal: the missing solvents induce renal injury in rats. PLoS One 7 (9):
factor in analgesic nephropathy. Scot. Med. J. 27 (1): 10–16. e45873.
319
21b
Toxic Nephropathies: Nonsteroidal Anti-inflammatory Drugs
Joseph B. Pryor1, William M. Bennett2, and Ali Olyaei3
1
Department of Internal Medicine, Oregon Health and Science University, School of Medicine, Portland, USA
2
Department of Nephrology, Legacy Transplant Services, Legacy Good Samaritan Medical Center, Portland, USA
3
Department of Pharmacy, Oregon State University/Oregon Health & Science University, Portland, USA
I ntroduction that are involved in the production of prostaglandins and

thromboxane A2, while selective COX-2 inhibitors only
Nonsteroidal anti-inflammatory drugs (NSAIDs) are one inhibit inducible COX-2. Originally, three COX-2-selective
of the most commonly prescribed groups of drugs for the NSAIDs, namely celecoxib, rofecoxib, and valdecoxib, were
treatment of pain and inflammation. It is estimated that, approved. They were designed with the hope of exhibiting
worldwide, more than 30 million people take NSAIDs anti-inflammatory properties while lacking the deleterious
daily [1]. The precise incidence of renal disease related to effects on homeostatic functions mediated by COX-1 activa-
NSAIDs is difficult to estimate due to the heterogeneity of tion such as gastrointestinal ulcers and bleeding; however,
the population. In general, approximately 1–3% of the increased myocardial infarctions and ischemic strokes were
patients taking NSAIDS may manifest one or more kind reported with all agents when used at higher than recom-
of renal injury from acute or chronic use of NSAIDS [2]. mended dosage compared to placebo or other NSAIDs [7].
Thus, the widespread use of NSAIDs means that renal In the Vigor study, a 79% greater risk of death or serious car-
complications are observed relatively frequently in clini- diovascular events was found in one treatment group com-
cal practice. Complications include hypertension, edema, pared with the other, rofecoxib (P = 0.007) [8], and increased
salt and water retention, acute tubular necrosis, acute cardiovascular events in patients who underwent coronary
interstitial nephritis, hyperkalemia, as well as acute and artery bypass surgery and received valdecoxib [9]. This led to
chronic kidney failure [3]. In addition, NSAIDs may their withdrawal from the market in 2004 and 2005, respec-
worsen preexisting hypertension, heart failure, and electively. Recently, the results of the PRECISION study were
trolyte imbalance and cause acceleration and progression published [10]. Among patients requiring a moderate dose
of kidney disease [4, 5]. of daily NSAID treatment, the incidence of cardiovascular
In addition to renal side effects, NSAIDs also have sig- disease, cardiovascular death, nonfatal myocardial infarc-
nificant gastrointestinal complications such as ulcers and tion, and nonfatal stroke for celecoxib [200 mg daily] was
bleeding. Furthermore, the use of NSAIDS at higher doses noninferior to ibuprofen and naproxen. However, the inci-
is associated with increased risk of cardiac events, such as dence of renal events was significantly lower with celecoxib
myocardial infarctions, acute coronary syndromes, and versus ibuprofen.
strokes. Low-dose aspirin irreversibly inhibits platelet COX-1,
Two different iso-enzymes of cyclooxygenase (1 and 2) thereby blocking the synthesis of thromboxane A2, a pro-
have been identified. Both enzymes play an important role thrombotic factor, and has little effect on endothelial
in platelet aggregation, gastrointestinal integrity, and cardio- COX-2 derived prostacyclin [11]. Prostacyclin inhibits
vascular health [6]. It has been hypothesized that nonselec- platelet aggregation and leads to vasodilatation. Due to its
tive NSAIDs inhibit both constitutive cyclooxygenase 1 selectivity, low-dose aspirin is the drug of choice for the
(COX-1) and inducible COX-2, the rate-limiting enzymes prevention of thrombotic events, even in patients with
320 Toxic Nephropathies: Nonsteroidal Anti-inflammatory Drugs
underlying kidney disease, without the risk of jeopardizing aintenance of Renal Blood Flow
M
prostaglandin-dependent kidney function. By contrast, and Glomerular Filtration Rate
higher doses of aspirin can inhibit COX-2-derived prosta-
cyclin production. Prostacyclin is mainly COX-2 derived NSAIDs produce negligible effects on the renal hemodynam-
and is inhibited by both nonselective NSAIDs and COX-2- ics of most young and healthy euvolemic subjects [24, 25].
selective NSAIDs. Therefore, a potential cardiovascular However, in hemodynamic compromised subjects, NSAIDS
risk is seen with nonselective NSAIDs as well as COX-2- are associated with number of renal syndromes. Prostaglandins
selective drugs [12–14]. maintain renal blood flow and glomerular filtration rate
despite vasoconstrictor stimuli, such as leukotrienes, throm-
Actions of Renal Prostaglandins boxane A2, angiotensin II, vasopressin, endothelin, and
catecholamines. Though vasoconstrictive, catecholamines
Both COX-1 and COX-2 are expressed extensively in the kid- stimulate local prostaglandin production, resulting in a feed-
ney. COX-2 activities have been identified mostly in the mac- back loop between vasoconstrictors and vasodilatory prosta-
ula densa and medulla while COX-1 works in the glomerulus glandins [26–28]. PGF2α-like peroxidation products also have
and both cortical and medullary collective ducts [15, 16]. major vasoconstrictive effects [29]. Thus, in patients with
The kidney produces the vasodilator prostaglandins PGE2, underlying ischemic or inflammatory renal injury, the addi-
PGF2α, and PGI2 as well as vasoconstrictive thromboxane tion of a nonselective or COX-2-selective NSAID not only
A2. These autocoids, synthesized and metabolized by the decreases the production of vasodilatory prostaglandins but
kidney, autoregulate renal blood flow, renin release, tubular also results in the nonenzymatic formation of vasoconstrictor
ion transport, and water metabolism [17, 18]. PGI2, which is metabolites of arachidonic acid, further jeopardizing renal
mainly present in the afferent arteriole and glomerulus, blood flow and glomerular filtration. However, in the pres-
plays a major role in controlling glomerular hemodynam- ence of salt depletion, an ineffective circulating plasma vol-
ics [19]. In contrast, PGE2, which is predominantly produced ume, or under conditions characterized by high circulating
in the collecting tubule and within the interstitium, regu- levels of vasoconstrictor hormones, NSAIDs may be nephro-
lates medullary hemodynamics [18]. toxic. Such conditions include cirrhosis, hypovolemia, cardiac
disease, renal disease, septic shock, advanced age, diuretic
use, diabetes mellitus, and following surgery [30].
ffects on COX-2 NSAIDs on Renal
E In elderly, salt-replete subjects, both indomethacin and
Prostaglandins rofecoxib decreased sodium excretion, but only indomethacin
reduced the glomerular filtration rate [20]. Celecoxib, like
All NSAIDS, nonselective or selective, prevent prostaglan- rofecoxib, affects renal function in selected groups of subjects.
din synthesis in the kidneys. The urinary excretion of PGE2 Whelton et al. compared celecoxib with naproxen in 29
and 6-keto-PGF1a, the stable metabolite of PGI2, reflects healthy elderly subjects in a single-blind, randomized, crosso-
the renal synthesis of PGE2 and PGI2, respectively. In ver study [31]. Subjects were given either celecoxib at 200 mg
healthy older adults, rofecoxib reduces baseline urinary twice daily for 5 days followed by celecoxib at 400 mg twice
6-keto-PGF1α by 47%, comparable to the 53% reduction daily for the next 5 days or they received naproxen at 500 mg
induced by indomethacin [20]. In another study, rofecoxib twice daily for 10 days. After a 7-day washout, subjects were
reduced urinary PGE2 and 6-keto-PGF1α excretion in crossed over to the other regimen. Glomerular filtration rate
healthy volunteers by approximately 40–50%, similar to fell more with naproxen than with celecoxib, although uri-
that induced by meloxicam or diclofenac [21]. This inhibi- nary excretion of prostaglandin E2, 6-keto-PGF1α, and sodium
tion has a profound effect on hemodynamic compensatory was comparable. In another study involving salt-depleted
mechanism in the kidneys. elderly subjects, rofecoxib and indomethacin induced a com-
Furthermore, excretion of urinary 6-keto-PGF1α was parable reduction of glomerular filtration rate [32]. These
comparable in response to celecoxib and traditional studies illustrate that COX-2 inhibitors and nonselective
NSAIDs [22]. In a trial of healthy elderly volunteers con- NSAIDs have similar effects on renal hemodynamics.
suming normal amounts of sodium, multiple doses of
twice-daily celecoxib reduced PGE2 and 6-keto-PGF1α
excretion to the same degree as naproxen, by approximately Postoperative Use of NSAIDs
65 and 80%, respectively [23]. These data suggest that the
COX-2 isoform plays an important role in renal prostaglan- Lee and coworker in a meta-analysis of 19 trials (n = 1024)
din biosynthesis. It is thus likely that COX-2 inhibitors will demonstrated that the postoperative use of NSAIDs in
impact renal function just as nonselective NSAIDs do. adults with normal preoperative renal function resulted in
Clinical Syndromes Associated with Nonselective and COX-2-Selective NSAIDs 321
Table 21b.1 Patients at risk for renal impairment. trials, there are no differences in renal side effects between
COX-2-selective inhibitors and nonselective inhibitors. In
Heart failure; EF less than 40% the PRESION study, the serum creatinine level of 2.0 mg/
Liver disease dl (177 μmol/l), an increase of verified serum creatinine
Patients with SCr > 1.5 mg/dl level of 0.7 mg/dl (62 μmol/l) from baseline or doubling of
Elderly population the baseline serum creatinine, or hyperkalemia (defined as
Dehydration (protracted-several days)
>6 mmol/dl) with 50% elevation in serum creatinine and
initiation of hemodialysis or peritoneal dialysis were
Concomitant use of nephrotoxins (cyclosporine, tacrolimus,
aminoglycoside, etc.) comparable between naproxen and celecoxib while much
higher for ibuprofen (0.5% vs. 0.9% P = 0.004) [10]
SCr, serum creatinine.
a 16 ml/min fall in creatinine clearance (95% confidence enin Release and Potassium

R
interval [CI] 5–28) and a fall in potassium excretion of Homeostasis
38 mmol/day (95% CI 19–56) on the first day after surgery
compared with placebo [33]. No cases of postoperative PGE2, PGI2, and arachidonic acid are potent stimuli of renin
acute kidney failure requiring dialysis were reported. The release [15]. Both nonselective and COX-2-selective NSAIDs
conclusions were that NSAIDs caused a clinically unim- can inhibit renin secretion and under some circumstances
portant reduction in kidney function in the early postop- lead to hyporeninemia and hypoaldosteronism with atten-
erative period in patients with normal kidney function and dant hyperkalemia. This is particularly common in patients
that these drugs should not be withheld in such patients. with preexisting renal impairment [38–40]. Inhibition of
Others have reported, however, an overall incidence of prostaglandin synthesis can also lead to hyperkalemia by
postoperative renal insufficiency of 18% after major sur- decreasing distal tubular flow rate and sodium delivery, both
gery, with a subsequent hospital mortality rate of 13% [34]. of which limit potassium secretion [41, 42].
In general, NSAIDs should be withheld from patients at
risk for kidney injury or patients with preexisting kidney
disease pre- and postoperative because of concerns about Natriuresis and Diuresis
their effects on renal function [3] (Table 21b.1).
Renal prostaglandins are natriuretic and diuretic. They
inhibit sodium and chloride reabsorption in the proximal
Renal Dysfunction in COX-2 Trials and distal nephrons as well as in the loop of Henle [43, 44].
In addition, renal prostaglandins reduce the renal cortico-
In one study, renal adverse events were reported in 24.3% medullary solute gradient and antagonize the action of
of 144 patients receiving celecoxib and in 30.8% of 143 vasopressin in vivo [45]. Although prostaglandins acutely
patients receiving diclofenac and omeprazole. Kidney influence salt and water excretion, they do not regulate it
failure (defined as a rise in serum creatinine to above under normal conditions.
200 μmol/l) was seen in 5.6 and 6.3% of patients,
respectively [35]. Overall, renal adverse events were more
common in patients with renal impairment (celecoxib Clinical Syndromes Associated with
51.4%, diclofenac plus omeprazole 40.7%). In another
Nonselective and COX-2-
study, adverse effects related to kidney function occurred
in about 1% of naproxen-and rofecoxib-treated patients [8].
Selective NSAIDs
In a review of randomized clinical trials lasting 2 weeks or
cute Renal Impairment and Acute Tubular
A
more involving celecoxib, a rise in serum creatinine was
Necrosis
seen in 0.7% of patients treated with celecoxib and in 1.2%
of patients treated with diclofenac (P < 0.05) [36]. In a Under circumstances where there is poor renal perfusion
meta-analysis of data from company clinical trial reports, with high renin levels, nonselective and COX-2-selective
there was no difference in the incidence of renal adverse NSAIDs can reduce glomerular filtration rate, resulting in
events (defined as an increase in serum creatinine >1.3 acute kidney failure. This complication has been reported
times the upper limit of normal) in 15 319 patients treated with most NSAIDs but only rarely with aspirin. Kidney
with celecoxib (0.3%) or other NSAIDs (0.5%) (relative risk failure has also been described after administration of topi-
[RR] 0.78, 95% CI 0.48–1.3) [37]. In summary, from these cal and intramuscular NSAIDs [46–48].
A multicenter study in France examined the incidence tubular damage coexists with tubulo-interstitial infiltrate
and subsequent outcome of patients with drug-induced predominantly of T-lymphocytes and, to a lesser extent,
acute kidney failure [49]. Of the 398 patients with acute monocytes/macrophages, B lymphocytes, plasma cells,
kidney failure, 147 (36.9%) had taken NSAIDs. One-third and eosinophils [57, 65]. Rarely, a granulomatous intersti-
of them required dialysis, and 71.4% recovered or regained tial nephritis is seen [66]. Immunofluorescence micros-
previous renal function. A renal biopsy obtained in 25 copy is usually negative or nonspecific. The predominance
patients with NSAID-associated kidney failure disclosed of T-lymphocytes in the interstitial infiltrate has been
acute tubular necrosis and acute interstitial nephritis in 21 taken to indicate that T-lymphocyte activation mediates
patients and either minimal change nephropathy or this syndrome, rather than a humoral mechanism, as
chronic renal damage in four. A nested case–control study seen in other forms of drug-induced acute interstitial
using the United Kingdom General Practice Research nephritis [56, 57]. Inhibition of renal COX has also been
Database reported that current users had an RR of develop- incriminated in the genesis of NSAID-induced acute
ing acute kidney failure of 3.2 (95% CI 1.8–5.8) compared tubulo-interstitial nephritis. The resulting stimulation of
with non-NSAID users [50]. This increased risk was higher the lipoxygenase pathway of arachidonic acid metabolism
in patients with heart failure, hypertension, or diabetes. produces leukotrienes, which are potent chemotactic fac-
Thus, although renal side effects from NSAID use are rela- tors for lymphocytes. Recovery of renal function may be
tively rare, renal damage can be irreversible and the out- only partial [65], and chronic interstitial fibrosis may pro-
come can be fatal. Renal function usually improves upon gress to chronic renal failure [67]. Prednisolone has been
drug withdrawal, although in some cases permanent renal successfully used in anecdotal reports, but there is no
damage may occur [51]. conclusive evidence that corticosteroids hasten the reso-
Acute kidney failure and hyperkalemia have been lution of the renal lesion [2].
observed after the administration of COX-2-selective inhib-
itors to patients with risk factors for NSAID-induced acute
renal insufficiency, including underlying chronic renal Glomerulonephritis
impairment and volume depletion [52]. Acute kidney fail-
ure was also reported in a patient with a kidney transplant Membranous nephropathy with nephrotic syndrome may
4 weeks after starting rofecoxib [53]. occur as an idiosyncratic reaction to various classes of
NSAIDs [68–70]. The temporal association with the intake
of NSAIDs, the prompt and complete recovery after drug
A
cute Tubulo-Interstitial Nephritis discontinuation, and the absence of recurrent disease may
help distinguish NSAID-associated membranous nephropa-
NSAIDs of different classes have been associated with thy from the idiopathic form [71]. As with NSAIDs, glomer-
acute tubulo-interstitial nephritis and kidney failure [2, 51, ulopathies with the nephrotic syndrome can occur with
54, 55]. Acute allergic tubulo-interstitial nephritis due to COX-2 inhibitors [63]. Membranous nephropathy with
NSAIDs is much less common than hemodynamic kidney acute interstitial nephritis secondary to celecoxib has been
failure. The patients are often elderly and may have taken described [58].
the offending agent for months or years before the develop-
ment of acute interstitial nephritis. Clinical evidence of an
allergic reaction, such as fever, rash, arthralgia, eosino- Renal Papillary Necrosis
philia, or eosinophiluria, is uncommon. Of note, proteinu-
ria, often in the nephrotic range, may occasionally appear, Renal papillary necrosis has been infrequently reported in
especially in fenoprofen-induced tubulo-interstitial nephri- patients treated with ibuprofen, indomethacin, phenylb-
tis [54, 56, 57]. Cases of interstitial nephritis have been utazone, fenoprofen, or mefenamic acid [51, 67, 72–74] or
reported with both celecoxib and rofecoxib [58–64]. In two with paracetamol [75, 76]. One case of celecoxib-related
cases, interstitial nephritis was associated with glomeru- renal papillary necrosis has been reported [77].
lopathies, one case with minimal change disease [62] and
the other one with membranous nephropathy [58]. Thus,
there is little evidence that suggests a major difference Chronic Kidney Disease
between NSAIDs and COX-2 inhibitors in the incidence of
acute interstitial nephritis. Sandler et al. evaluated the risk for chronic kidney disease
NSAID-induced acute tubulo-interstitial nephritis is associated with regular use of nonaspirin NSAIDs in 554
formally diagnosed by renal biopsy. A patchy acute patients with newly diagnosed chronic renal dysfunction [5].
Hyperkalemia and Hyporeninemic Hypoaldosteronis 323
They found a twofold-increased risk for chronic kidney dis- Whelton et al. performed a post hoc analysis on the renal
ease in patients with a history of previous daily use of safety of celecoxib, incorporating more than 50 clinical
NSAIDs (adjusted odds ratio [OR] 2.1, 95% CI 1.1–4.1). The studies with more than 13 000 subjects [23]. The most
increased risk was predominantly limited to men older common events, peripheral edema (2.1%), hypertension
than 65 years, for whom the OR was 10.0 (95% CI 1.2–82.7) (0.8%), and exacerbation of preexisting hypertension
after adjusting for use of other analgesics. The NSAID- (0.6%), were not dose or time-related. Their incidence and
associated risk was also greater among those with a history profile were similar to those of nonselective NSAIDs. A
of conditions that might indicate an enhanced susceptibil- similar post hoc analysis of rofecoxib revealed peripheral
ity to the effects of NSAIDs, including previous myocardial edema in 3.8% of patients [90].
infarction, congestive heart failure, heavy alcohol con- Whelton et al. also compared the effects of celecoxib at
sumption (as a surrogate for cirrhosis), or diuretic use. 200 mg and rofecoxib at 25 mg over a 6-week period in 810
These observations were confirmed in a case-control study hypertensive patients with osteoarthritis, aged over
of 716 patients with end-stage renal failure and 361 con- 65 years [91]. Edema developed in nearly twice as many
trols [6]. In this study, a high cumulative intake of NSAIDs rofecoxib-treated than celecoxib treated patients (9.5% vs.
(>5000 tablets) was associated with a 4.5-fold excess risk of 4.9%, P = 0.014). Systolic blood pressure increased signifi-
end-stage renal failure, although the CI was wide (1.0–19.5) cantly in 17% of rofecoxib-treated patients, compared
and, curiously, this excess risk was not seen when an aver- with 11% of celecoxib-treated patients (P = 0.032). In con-
age annual intake of NSAIDs was examined. Other studies clusion, celecoxib induces edema less frequently and
of NSAID usage in hospitalized patients [78, 79] and also results in smaller rises in blood pressure than rofecoxib.
cohort studies, however, did not show this association [80, A meta-analysis of COX-2 inhibitors and their effects on
81]. The reasons for these discrepant findings are unclear. blood pressure showed that they were associated with a
On balance, it seems likely that chronic NSAID use may be nonsignificant higher risk of causing hypertension com-
associated with a slightly increased risk for the develop- pared with placebo (RR 1.61, 95% CI 0.91–2.84) or nonse-
ment of chronic kidney failure. Some patients with chronic lective NSAIDs (RR 1.25, 95% CI 0.87–1.78) [92]. Thus,
kidney failure rely on prostaglandin-mediated vasodilata- both NSAIDs and COX-2 inhibitors can raise blood
tion to maintain renal blood flow [82–84]. Addition of pressure, especially in hypertensive, elderly patients,
NSAIDs may cause further deterioration of renal func- and there is no substantial evidence to suggest that
tion [49, 82, 84, 85]. COX-2 inhibitors are safer in this respect.
Pavlicevic et al., in a prospective clinical study, investi-
gated 88 hypertensive patients being treated with lisino-
Salt and Water Retention pril/hydrochlorothiazide and amlodipine. The effects of
addition of ibuprofen, acetaminophen, or piroxicam were
NSAID therapy may aggravate the sodium retention recorded over a 3-month period. A significant systolic
induced by renal hypoperfusion in heart failure, cirrhosis, blood pressure increase of 7.7–9.9% (P < 0.001) was
or nephrotic syndrome [3]. Hyponatremia may occur if reported when ibuprofen and piroxicam were also given
water retention is disproportionate to sodium reten- with lisinopril/hydrochlorothiazide and NSAIDs [93].
tion [86], especially when thiazide diuretics are given
simultaneously [2].
yperkalemia and Hyporeninemic
H
Hypoaldosteronism
H
ypertension
NSAIDs may cause hyperkalemia, and this is seen more
Two large meta-analyses encompassing more than 90 commonly in patients with chronic kidney failure, diabe-
studies have demonstrated that NSAIDs may increase tes mellitus, and type IV tubular acidosis through previ-
blood pressure, especially in previously hypertensive ously outlined mechanisms [39, 94]. Increases in
patients [87, 88]. NSAIDs elevate supine mean blood potassium levels are also expected to occur with
pressure by 5 mmHg [88], a rise known to increase COX-2 inhibitors. Indeed, rofecoxib raises serum potas-
hypertension-related morbidity and mortality [89]. This sium levels by more than 0.8 mM, with a similar incidence
complication is of importance in the elderly, who are fre- as NSAIDs [95]. In a retrospective cohort study of 202
quently prescribed NSAIDs for musculoskeletal disorders patients using propensity scores methods the incidence of
and also have a high prevalence of other chronic disor- hyperkalemia was compared between nonselective
ders, including hypertension. NSAIDS versus COX-2 selective NSAIDs. Compared to
patients prescribed nonselective NSAIDs, those pre- C

onclusions
scribed a selective COX-2 inhibitor had a higher risk of
serum potassium increase greater than 5 mEq/l (OR 2.56, The increasing use of NSAIDs both by prescription and
95%CI 1.03–6.36). The use of NSAIDs in particular over the counter has increased the prevalence of nephro-
COX-2 inhibitors should be avoided in patients at risk for toxicity. A history of NSAID use should be sought in all
hyperkalemia and in patients taking other drugs known patients with unexplained impairment of renal function
to decrease renal potassium excretion, such as potassium- and/or proteinuria. In patients with volume depletion or
sparing diuretics, angiotensin-converting enzyme inhibi- decreased organ perfusion, the use of NSAIDs should be
tors, and β-blockers. avoided. NSAIDs should not be prescribed in patients with
chronic renal impairment or with a functioning kidney
transplant. Patients with an NSAID-induced, interstitial
herapeutic Use of NSAIDs in
T nephritis or papillary necrosis should not be given NSAIDs
Nephrotic Syndrome again. In some individuals who have developed NSAID-
induced acute kidney failure and who have recovered kid-
NSAIDs reduce proteinuria in patients with nephrotic syn- ney function, NSAIDs may be reintroduced if clinically
drome [40, 96], probably by reducing renal blood flow and necessary, provided that the risk factors for enhanced sus-
glomerular filtration rate [97]. The occurrence of irreversi- ceptibility have been corrected and that renal function is
ble kidney failure in patients with a nephrotic syndrome closely monitored. COX-2-selective inhibitor use requires
treated with NSAIDs [55] suggests great caution for the use the same cautions as with traditional NSAIDs.
of these drugs in this clinical setting.
References
Hatt, K.M., Vijapura, A., Maitin, I., and Cruz, E. (2018).

1 8 Bombardier, C., Laine, L., Reicin, A. et al. (2000).
Safety considerations in prescription of NSAIDS for Comparison of upper gastrointestinal toxicity of rofecoxib
musculoskeletal pain: a narrative review. PM R 10 (12): and naproxen in patients with rheumatoid arthritis.
1404–1411. VIGOR study group. N. Engl. J. Med. 343: 1520–1528, 2.
2 Zhang, X., Donnan, P.T., Bell, S., and Guthrie, B. (2017). 9 Nussmeier, N.A., Whelton, A.A., Brown, M.T. et al.
Non-steroidal anti-inflammatory drug induced acute kidney (2005). Complications of the COX-2 inhibitors parecoxib
injury in the community dwelling general population and and valdecoxib after cardiac surgery. N. Engl. J. Med. 352:
people with chronic kidney disease: systematic review and 1081–1091.
meta-analysis. BMC Nephrol. 18: 256. 10 Nissen, S.E., Yeomans, N.D., Solomon, D.H. et al. (2016).
3 Kovic, S.V., Vujovic, K.S., Srebro, D. et al. (2016). Cardiovascular safety of Celecoxib, naproxen, or
Prevention of renal complications induced by non- ibuprofen for arthritis. N. Engl. J. Med. 375: 2519–2529.
steroidal anti-inflammatory drugs. Curr. Med. Chem. 23: 11 FitzGerald, G.A. (2008). Translational therapeutics at the
1953–1964. platelet vascular interface. Summary. Arterioscler.
4 Ljungman, C., Kahan, T., Schioler, L. et al. (2017). Non- Thromb. Vasc. Biol. 28: s51–s52.
steroidal anti-inflammatory drugs and blood pressure 12 Walker, C. and Biasucci, L.M. (2018). Cardiovascular
control in patients treated for hypertension: results from safety of non-steroidal anti-inflammatory drugs revisited.
the Swedish primary care cardiovascular database. Blood Postgrad. Med. 130: 55–71.
Press. 26: 220–228. 13 Karger, C., Machura, K., Schneider, A. et al. (2018).
5 White, W.B. (2007). Cardiovascular risk, hypertension, and COX-2-derived PGE2 triggers hyperplastic renin
NSAIDs. Curr. Pain Headache Rep. 11: 428–435. expression and hyperreninemia in aldosterone synthase-
6 Mitchell, J.A. and Kirkby, N.S. (2019). Eicosanoids, deficient mice. Pflugers Arch. -Eur. J. Physiol. 470:
prostacyclin and cyclooxygenase in the cardiovascular 1127–1137.
system. Br. J. Pharmacol. 176(8): 1038–1050. 14 Araujo, M. and Welch, W.J. (2009). Cyclooxygenase
7 Fanelli, A., Ghisi, D., Aprile, P.L., and Lapi, F. (2017). 2 inhibition suppresses tubuloglomerular feedback: roles
Cardiovascular and cerebrovascular risk with nonsteroidal of thromboxane receptors and nitric oxide. Am. J. Physiol.
anti-inflammatory drugs and cyclooxygenase 2 inhibitors: Renal Physiol. 296: F790–F794.
latest evidence and clinical implications. Ther. Adv. Drug 15 Moreno, C., Llinas, M.T., Rodriguez, F. et al. (2016).
Saf. 8: 173–182. Nitric oxide, prostaglandins and angiotensin II in the
Reference 325
regulation of renal medullary blood flow during volume 30 Garella, S. and Matarese, R.A. (1984). Renal effects of
expansion. J. Physiol. Biochem. 72: 1–8. prostaglandins and clinical adverse effects of non-
16 Curiel, R.V. and Katz, J.D. (2013). Mitigating the steroidal anti-inflammatory drugs. Medicine 63:
cardiovascular and renal effects of NSAIDs. Pain Med. 165–181.
(Malden, Mass) 14 (Suppl. 1): S23–S28. 31 Whelton, A., Schulman, G., Wallemark, C. et al. (2000).
17 Schnermann, J. and Briggs, J.P. (1981). Participation of Effects of celecoxib and naproxen on renal function in the
renal cortical prostaglandins in the regulation of elderly. Arch. Intern. Med. 160 (10): 1465–1470.
glomerular filtration rate. Kidney Int. 19: 802–815. 32 Swan, S.K., Rudy, D.W., Lasseter, K.C. et al. (2000). Effect
18 Dunn, M.J. (1984). Nonsteroidal anti-inflammatory drugs of cyclooxygenase-2 inhibition on renal function in
and renal function. Annu. Rev. Med. 35: 411–428. elderly persons receiving a low salt diet. A randomised,
19 Patrono, C., Ciabattoni, G., Remuzzi, G. et al. (1985). controlled trial. Ann. Intern. Med. 133 (1): 1–9.
Functional significance of renal prostacyclin and 33 Lee, A., Cooper, M.C., Craig, J.C. et al. (2004). Effects of
thromboxane A2 production in patients with systemic nonsteroidal anti-inflammatory drugs on postoperative
lupus erythematosus. J. Clin. Invest. 76: 1011–1018. renal function in adults with normal renal function.
20 Catella-Lawson, F., McAdam, B., Morrison, B.W. et al. Cochrane Database Syst. Rev. 2: CD002765.
(1999). Effects of specific inhibition of cyclooxygenase-2 34 Hou, S., Bushinsky, D., Wish, J. et al. (1983). Hospital-
on sodium balance, hemodynamics and vasoactive acquired renal insufficiency: a prospective study. Am. J.
eicosanoids. J. Pharmacol. Exp. Ther. 289: 735–741. Med. 1983 (74): 243–248.
21 Van Hecken, A., Schwartz, J.I., Depré, M. et al. (2000). 35 Chan, F.K., Hung, L.C., Suen, B.Y. et al. (2002). Celecoxib
Comparative inhibitory activity of rofecoxib, meloxicam, versus diclofenac and omeprazole in reducing the risk of
diclofenac, ibuprofen and naproxen on COX-2 versus recurrent ulcer bleeding in patients with arthritis. N.
COX-1 in healthy volunteers. J. Clin. Pharmacol. 40: Engl. J. Med. 347 (26): 2104–2110.
1109–1120. 36 Silverstein, F.E., Faich, G., Goldstein, J.L. et al. (2000).
22 McAdam, B.F., Catella-Lawson, F., Mardini, I.A. et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal
(1999). Systemic biosynthesis of prostacyclin by anti-inflammatory drugs for osteoarthritis and
cyclooxygenase (COX)-2: the human pharmacology of a rheumatoid arthritis. The CLASS study: a randomized
selective inhibitor of COX-2. Proc. Natl Acad. Sci. USA 96: controlled trial. Celecoxib long-term arthritis safety study.
272–277. JAMA 284 (10): 1247–1255.
23 Whelton, A., Maurath, C.J., Verburg, K.M., and Geis, 37 Moore, R.A., Derry, S., Makinson, G.T., and McQuay, H.J.
G.S. (2000). Renal safety and tolerability of celecoxib, a (2005). Tolerability and adverse events in clinical trials of
novel cyclooxygenase-2 inhibitor. Am. J. Ther. 7 (3): celecoxib in osteoarthritis and rheumatoid arthritis:
159–175. systematic review and meta-analysis of information from
24 Donker, A.J.M., Arisz, L., Brentjens, J.R.H. et al. (1976). company clinical trial reports. Arthritis Res. Ther. 7 (3):
The effect of indomethacin on kidney function and R644–R665.
plasma renin activity in man. Nephron 17: 288–296. 38 Goldzer, R.C., Coodley, E.L., Rosner, M.J. et al. (1980).
25 Muther, R.S. and Bennett, W.M. (1980). Effect of aspirin Hyperkalaemia associated with indomethacin. Arch.
on glomerular filtration rate in normal humans. Ann. Intern. Med. 141: 802–804.
Intern. Med. 92: 386–387. 39 Galler, M., Folkert, V.W., and Schlondorff, D. (1981).
26 Dibona, G.F. (1986). Prostaglandins and non-steroidal Reversible acute renal insufficiency and hyperkalemia
anti-inflammatory drugs: effects on renal following indomethacin therapy. JAMA 246: 154–155.
haemodynamics. Am. J. Med. 80 (Suppl 1A): 12–21. 40 Brater, D., Harris, C., Redfern, J., and Gertz, B. (2001).
27 Scharschmidt, L.A., Simonson, M.S., and Dunn, M.J. Renal effects of COX-2 selective inhibitors. Am. J.
(1986). Glomerular prostaglandins, angiotensin II, and Nephrol. 2001 (21): 1–15.
nonsteroidal antiinflammatory drugs. Am. J. Med. 81 41 Tannen, R.L. (1986). Potassium in cardiovascular and
(2B): 30–42. renal medicine, arrhythmias, myocardial infarction and
28 Pelayo, J.C. (1988). Renal adrenergic effector hypertension. In: Drug Interactions Causing
mechanisms: glomerular sites for prostaglandin Hyperkalaemia (eds. P.K. Whelton, A. Whelton and W.G.
interaction. Am. J. Phys. 254 (23): F184–1F90. Walker). New York: Marcel Dekker.
29 Takahashi, K., Nammour, T.M., Fukunaga, M. et al. 42 Field, M.J. and Giebisch, G. (1989). Mechanisms of
(1992). Glomerular action of a free radical-generated segmental potassium reabsorption and secretion. In: The
novel prostaglandin, 8-epi-prostaglandin F2 alpha in the Regulation of Potassium Balance (eds. D.W. Seldin and G.
rat. J. Clin. Invest. 90: 136–141. Giebisch). New York: Raven.
3 Stokes, J.B. (1979). Effect of prostaglandin E2 on chloride

4 Immunological characterisation of the infiltrate. Am. J.
transport across the rabbit thick ascending limb of Henle. Med. 1984: 1006–1012.
Selective inhibition of the medullary portion. J. Clin. 58 Alper, A.B.J., Meleg-Smith, S., and Krane, N.K. (2002).
Invest. 64: 495–502. Nephrotic syndrome and interstitial nephritis associated
44 Kinoshita, Y., Romero, J.C., and Knox, F. (1989). Effect of with celecoxib. Am. J. Kidney Dis. 40: 1886–1890.
renal interstitial infusion of arachidonic acid on proximal 59 Alim, N., Peterson, L., Zimmerman, S., and Updike, S.
sodium reabsorption. Am. J. Phys. 26: F237–F242. (2003). Rofecoxib-induced acute interstitial nephritis.
45 Lum, G.M., Aisenberg, G.A., Dunn, M.J. et al. (1977). Am. J. Kidney Dis. 41: 720–721.
in vivo effect of indomethacin to potentiate the renal 60 Brewster, U.C. and Perazella, M.A. (2004). Acute
medullary cyclic AMP response to vasopressin. J. Clin. tubulointerstitial nephritis associated with celecoxib.
Invest. 59: 8–13. Nephrol. Dial. Transplant. 18: 1017–1018.
46 Pearce, C.J., Gonzalez, F.M., and Wallin, J.D. (1993). 61 Chow, K.M., Szeto, C.C., Li, P., and Lai, F.M. (2003).
Renal failure and hyperkalemia associated with Acute interstitial nephritis and COX-2 inhibition. Hosp.
ketorolac tromethamine. Arch. Intern. Med. 153: Med. 64: 429.
1000–1002. 62 Demke, D., Zhao, S., and Arellano, F.M. (2001).
47 Smith, K., Halliwell, R.M.T., Lawrence, S. et al. (1993). Interstitial nephritis associated with celecoxib. Lancet
Acute renal failure associated with intramuscular 358: 1726–1727.
ketorolac. Anaesth. Intensive Care 21: 700–703. 63 Markowitz, G.S., Falkowitz, D.C., Isom, R. et al. (2003).
48 O’Callaghan, C.A., Andrews, P.A., and Ogg, C.S. (1994). Membranous glomerulopathy and acute interstitial
Renal disease and use of topical non-steroidal anti- nephritis following treatment with celecoxib. Clin.
inflammatory drugs. Br. Med. J. 308: 110–111. Nephrol. 59: 137–142.
49 Kleinknecht, D., Landais, P., and Goldfarb, B. (1986). 64 Rocha, J.L. and Fernando-Alonso, J. (2001). Acute
Analgesic and non-steroidal anti-inflammatory drug- tubulointerstitial nephritis associated with the selective
associated acute renal failure: a prospective collaborative COX-2 enzyme inhibitor, rofecoxib. Lancet 357:
study. Clin. Nephrol. 25: 275–281. 1946–1947.
50 Huerta, C., Castellsague, J., Varas-Lorenzo, C., and Garcia 65 Cameron, J.S. (1988). Allergic interstitial nephritis:
Rodriguez, L.A. (2005). Nonsteroidal anti-inflammatory clinical features and pathogenesis. QJM 66 (250):
drugs and risk of ARF in the general population. Am. J. 97–115.
Kidney Dis. 45 (3): 531–539. 66 Schwartz, A., Krause, P.H., Keller, T. et al. (1988).
51 Carmichael, T. and Shankel, S.W. (1985). Effects of Granulomatous interstitial nephritis after non-steroidal
non-steroidal anti-inflammatory drugs on prostaglandins anti-inflammatory drugs. Am. J. Nephrol. 8: 410–416.
and renal function. Am. J. Med. 78: 992–1000. 67 Adams, D.H., Howie, A.J., Micheal, J. et al. (1986).
52 Parazella, M. and Eras, J. (2000). Are selective Non-steroidal anti-inflammatory drugs and renal failure.
COX-2 inhibitors nephrotoxic. Am. J. Kidney Dis. 35 (5): Lancet i: 57–60.
937–940. 68 Campistol, J.M., Galofre, J., Botey, A. et al. (1989).
53 Wolf, G., Porth, J., and Stahl, R.A. (2000). Acute renal Reversible membranous nephropathy associated with
failure associated with rofecoxib. Ann. Intern. Med. 133 diclofenac. Nephrol. Dial. Transplant. 4: 393–395.
(5): 394. 69 Tattersall, J., Greenwood, R., and Farrington, K. (1992).
54 Brezin, J.H., Katz, S.M., Schwartz, A.B., and Chinitz, J.L. Membranous nephropathy associated with diclofenac
(1979). Reversible renal failure and nephrotic syndrome (letter). Postgrad. Med. J. 68 (799): 392–393.
assiciated with nonsteroidal anti-inflammatory drugs. N. 70 Grcevska, L., Polenakovi, M., Ferluga, D. et al. (1993).
Engl. J. Med. 310: 1271–1273. Membranous nephropathy with severe tubulointerstitial
55 Abraham, P.A. and Keane, W.F. (1984). Glomerular and and vascular changes in a patient with psoriatic arthritis
interstiti disease induced by nonsteroidal anti- treated with non-steroidal anti-inflammatory drugs. Clin.
inflammatory drugs. Am. J. Nephrol. 4: 1–6. Nephrol. 39: 250–253.
56 Finkelstein, A., Fraley, D.S., Stachura, I. et al. (1982). 71 Radford, M.G., Holley, K.E., Grande, J.P. et al. (1996).
Fenoprofen nephropathy: lipoid nephrosis and interstitial Reversible membranous nephropathy associated with the
nephritis: a possible T lymphocyte disorder. Am. J. Med. use of nonsteroidal antiinflammatory drugs. JAMA 276:
72: 81–87. 466–469.
57 Bender, W.L., Whelton, A., Beschorner, W.E. et al. (1984). 72 Munn, E., Lynn, K.L., and Bailey, R.R. (1976). Renal
Interstitial nephritis, proteinuria and renal failure caused papillary necrosis following regular consumption of
by non-steroidal anti-inflammatory drugs. NSAIDs. Aust. NZ J. Med. 95: 213–214.
Reference 327
3 Shah, G.M., Muhalwas, K.K., and Winer, R.L. (1981).

7 87 Pope, J.E., Anderson, J.J., and Felson, D.T. (1994).
Renal papillary necrosis due to ibuprofen. Arthritis A meta-analysis of the effects of nonsteroidal anti-
Rheum. 24: 1208–1210. inflammatory drugs on blood pressure. Arch. Intern. Med.
74 Segasothy, M., Thyaparan, A., Kamal, A., and Sivalingam, 21: 289–300.
S. (1987). Mefanamic acid nephropathy. Nephron 45: 88 Johnson, A.G., Nguyen, T.V., and Day, R.O. (1994). Do
156–157. nonsteroidal anti-inflammatory drugs affect blood
75 Krikler, D.M. (1967). Paracetamol and the kidney. Br. J. pressure? Ann. Intern. Med. 121: 289–300.
Med. 2: 615. 89 Collins, R., Peto, R., Godwin, J., and MacMahon, S.
76 Master, D.R. and Krikler, D.M. (1973). Analgesic (1990). Blood pressure, stroke, and coronary heart
nephropathy associated with paracetamol. Proc. R. Soc. disease. Part 2. Short-term reductions in blood pressure:
Med. 66: 904. overview of randomised drug trials in their
77 Akhund, L., Quinet, R.J., and Ishaq, S. (2003). Celecoxib- epidemiological context. Lancet 335: 827–838.
related renal papillary necrosis. Arch. Intern. Med. 163: 90 Whelton, A. (2001). Renal aspects of treatment with
114–115. conventional nonsteroidal anti-inflammatory drugs
78 Fox, D.A. and Jick, H. (1984). Non-steroidal anti- versus cyclooxygenase-2 inhibitor. Am. J. Med. 110 (Suppl
inflammatory drugs and renal disease. JAMA 151: 1): 33–42.
1299–1300. 91 Whelton, A., Fort, J.G., Puma, J.A. et al. (2001).
79 Beard, K., Perera, D.R., and Jick, H. (1988). Drug-induced Cyclooxygenase-2 specific inhibitors and cardiorenal
parenchymal renal disease in outpatients. J. Clin. function: a randomized, controlled trial of celecoxib and
Pharmacol. 28 (5): 431–435. refecoxib in older hypertensive osteoarthritis patients.
80 Rexrode, K.M., Buring, J.E., Glynn, R.J. et al. (2001). Am. J. Ther. 8 (2): 85–95.
Analgesic use and renal function in men. JAMA 286 (3): 92 Aw, T.J., Haas, S.J., Liew, D., and Krum, H. (2005).
315–321. Meta-analysis of cyclooxygenase-2 inhibitors and their
81 Curhan, G.C., Knight, E.L., Rosner, B. et al. (2004). effects on blood pressure. Arch. Intern. Med. 165 (5):
Lifetime nonnarcotic analgesic use and decline in renal 490–496.
function in women. Arch. Intern. Med. 164 (14): 93 Pavlicevic, I., Kuzmanic, M., Rumboldt, M., and
1519–1524. Rumboldt, Z. (2008). Interaction between
82 Ciabattoni, G., Cinotti, G.A., and Pierucci, A. (1984). antihypertensives and NSAIDs in primary care: a
Effects of sulindac and ibuprofen in patients with controlled trial. Can. J. Clin. Pharmacol. J. Can. de
chronic glomerular disease: evidence for the Pharmacol. Clin. 15: e372–e382.
dependence of renal function on prostacyclin. N. Engl. J. 94 Kutyrina, I.M., Androsova, S.O., and Tareyeva, I.E. (1979).
Med. 310: 279–288. Indomethacin-induced hyporeninaemic
83 Patrono, C. and Pierucci, A. (1986). Renal effects hypoaldosteronism. Lancet i: 785.
of nonsteroidal antiinflammatory drugs in chronic 95 Findling, J.W., Beckstrom, D., Rawsthorne, L. et al.
glomerular disease. Am. J. Med. 82 (Suppl 2B): (1980). Indomethacin-induced hyperkalaemia in
71–83. three patients with gouty arthritis. JAMA 244:
84 Brandstetter, R.D. and Mar, D.D. (1978). Reversible 1127–1128.
oliguric renal failure associated with ibuprofen treatment. 96 Donker, A.J.M., Brentjens, J.R.H., van der Hem, G.K., and
Br. Med. J. 2: 1194–1195. Arisz, L. (1978). Treatment of the nephrotic syndrome
85 Tan, S.Y., Shapiro, R., and Kish, M.A. (1979). Reversible with indomethacin. Nephron 22: 374–381.
acute renal failure induced by indomethacin. JAMA 241: 97 Tiggeler, R.G.W.L., Hulme, B., and Wijdeveld, P.G.A.B.
2732–2733. (1979). Effect of indomethacin on glomerular
86 Blum, M. and Aviram, A. (1980). Ibuprofen-induced permeability in the nephrotic syndrome. Kidney Int. 16:
hyponatraemia. Rheumatol. Rehabil. 19: 258–259. 312–321.
329
Secondary Diseases of the Kidney

Part 4
331
22
Hypertension
Swapnil Hiremath
Department of Medicine, University of Ottawa, Ottawa, Canada
I ntroduction As portrayed in the prevention paradox also, the majority

of the adverse outcomes may indeed occur in the low to
The kidneys are inextricably interwoven with high blood intermediate risk, asymptomatic individuals [2]. Whether
pressure (BP). In his second paper published in 1836, while it is labeled as a disease or a condition, the global burden of
describing the natural course of nephritis, Richard Bright hypertension is high, estimated at 20 526 per 100 000, with
mentions the observation of a hard pulse along with a significant associated attributable mortality and disabil-
detailed description of the accompanying headache, loss of ity [3]. Thus, treatment of hypertension, i.e. lowering of BP,
vision, seizures, and dyspnoea [1]. These cardiovascular would be to prevent adverse clinical outcomes. For patients
changes were then thought to be secondary to the kidney with chronic kidney disease (CKD), these could be either
disease, though it is reported that Bright acknowledged the cardiovascular or renal outcomes. Though BP is often ban-
possibility of reverse causation. He did suggest that the died about as the most important risk factor for worsening
“altered quality” of the blood in renal disease affected the kidney function, the evidence for lowering BP and decreas-
peripheral vasculature in such a way that an increased ing progression of kidney disease is rather limited. It is true
force was necessary for propelling the blood, a remarkably that at very high levels of BP there exists an association of
accurate observation. In 1856, Ludwig Traube reasoned BP and kidney outcomes. However, in the modern era
that a shrunken kidney will first decrease the amount of where the decisions are more to do with whether systolic
liquid which is removed from the arterial system as urinary BP should be lowered to <140, <130 or even to <120, the
secretion, as a result of which the pressure of the arterial outcomes that matter are mostly the cardiovascular rather
system (“Aortenspannung”) must increase, as good a than renal. The relationship of BP with cardiovascular
description of renal parenchymal hypertension as one (CV) outcomes is stronger than for progression of kidney
could have done at the time [1]. BP measurement became disease. Hence, unsurprisingly, lowering of stroke and
more feasible in the late nineteenth century, and in heart failure drives most of the CV outcome reduction with
1874 Mahomed described high arterial pressure which was intensive BP lowering, and prevention of end stage kidney
present before albuminuria in patients with Bright’s dis- disease (ESKD) is not one of them [4].
ease, and advanced the view that high BP could be leading
to contracted kidneys [1]. Thus, right from the beginning, Blood Pressure Measurement
hypertension as a cause, or a consequence of kidney dis-
ease has been a matter of contention. An underappreciated aspect of the management of hyper-
tension is proper BP measurement. Inaccurate measure-
ment will result in inappropriate treatment. While the
W
hy Treat Hypertension? research gold standard of BP measurement is intra-arterial
continuous BP monitoring, several practical alternatives
Hypertension is called the silent killer in the lay media are suitable for use in clinical practice. A “casual” meas-
precisely because it is asymptomatic for the vast majority. urement performed in a noisy atmosphere using improper
332 Hypertension
technique and with inadequate rest is perfectly useless and masked hypertension, nocturnal dipping pattern, and pro-
should be abandoned [5]. At the other end, we have 24- vide an assessment of variability.
hour ambulatory blood pressure monitoring (ABPM) Another out-of-office technique is home BP measure-
which provides multiple measurements throughout the ment. Modern home BP devices use an oscillometric tech-
day and night, thus providing an assessment of the overall nique, and are cheap and easy to use. They offer similar
BP load, diurnal pattern, and an assessment of out of office advantages to the ABPM method, in providing multiple
BP values. However, it is not available in all jurisdictions, measures for an overall BP load, potentially minimizing
and cost as well as logistics may hinder universal adoption. white coat effect and picking up masked effect. Caveats of
A properly done resting office measurement can go a long home BP measurement include the accuracy of these
way, but manual methods are notoriously unreliable, with devices, as multiple studies have reported a large propor-
human error, poor technique, terminal digit preference, tion of home BP monitors being inaccurate [12, 13]. Lastly,
and such issues being an unsurmountable barrier despite they do not provide a picture of nocturnal pattern and vari-
educational efforts [6]. Automated oscillometric blood ability as ABPM does. They do, however, empower patients,
pressure (AOBP) offers a suitable compromise, especially and have been reported to result in improved BP control,
newer devices which can specify a period of rest before ini- potentially by overcoming therapeutic inertia [14].
tiating measurement and with repeat measurements pro- Wearable BP devices are the next frontier, with one
viding an average of two to five readings [7, 8]. Comparative device being approved recently in the United States.
data suggests AOBP use decreases the misdiagnosis of Potentially, these will provide richer data about BP and
white coat hypertension and, more importantly, it approxi- empower patients immensely. Certain caveats about their
mates proper resting BP measurement (as is done in clini- accuracy and the ability to deal with the sheer amount of
cal trials) more than a casual measurement [7]. Recent information and possible resultant false alarms will have to
hypertension trials, including Heart Outcomes Prevention be worked out before widespread adoption [15].
Evaluation – 3 (HOPE-3), Action to Control Cardiovascular To summarize, out-of-office measurement, preferably
Risk in Diabetes (ACCORD) and the Systolic Blood with ABPM followed by home BP measurement, is the pre-
Pressure Intervention Trial (SPRINT) all used a form of ferred way of diagnosing and following up hypertension
AOBP, and further validation of AOBP with clinical out- management. Office measurement with AOBP is a reason-
comes should be considered hence unnecessary [9–11]. able alternative when this option is not routinely available.
Nevertheless, AOBP should not be construed as a replace- Table 22.1 provides highlights of different BP measure-
ment for ABPM. The ABPM can allow for diagnosis of ment techniques.
Table 22.1 Different BP measurement methods and their advantages and disadvantages.
Technique Advantages Disadvantages Comments
Casual Easy Rarely done correctly Should be abandoned

Misdiagnosis of true BP
AOBP Eliminates many human errors More time required Used in recent trials
Can specify rest period Workflow may be slowed Devices vary with respect to
Multiple measurements Cannot diagnose masked period of rest, number of
Minimizes white coat effect hypertension or diurnal pattern measurements
ABPM Diagnose white coat and Cost “Gold standard”

masked effect Inconvenience Strong association with
Assess BP variability outcomes
Assess nocturnal BP pattern Outcome trial with ABPM-
Assess overall BP load based strategy ongoing
Home Inexpensive Accuracy variable

Easy Cannot assess nocturnal
Can diagnose white coat and pattern
masked effect
Empowers self-management
AOBP, automated oscillometric blood pressure; ABPM, ambulatory blood pressure monitoring; BP, blood pressure.
Why Treat Hypertensio? 333
Blood Pressure Variability absence of nocturnal dipping is thought to be caused by

heightened sympathetic activity, increased sodium sensi-
A brief note about BP variability is necessary given its
tivity and volume status, and disturbances of the sleep
reported association with clinical outcomes. Though we still
activity cycle. These factors are all clearly more common in
do not know enough about the causes and the cure, we do
CKD patients, thus the finding of increased nondipping is
know more about the consequences of high BP variability.
not surprising. More important is whether nondipping sta-
Variability is variably defined based on the time frame.
tus can be corrected and whether this correction decreases
Short-term variability refers to that seen on a 24-hour ABPM
the adverse risk profile observed in these patients.
whereas long-term refers to visit-to-visit variability.
One group has reported success with nighttime dosing of
Numerically, variability is often defined on the basis of a
BP medications in CKD patients, which lowered nocturnal
large standard deviation or a coefficient of variation that
BP sufficiently to restore normal dipping pattern in an addi-
reflects an overall dispersion of the values around the mean,
tional 30% of the patients compared to controls, but also
but not specifically the order of these dispersed values. The
resulted in an improved survival [22]. Interpretation of these
average real variability represents a potentially more reliable
fantastic results (which reported a hazard ratio of 0.28 for
representation of true variation [16]. Long-term variability
major adverse cardiovascular events [MACEs]) from the
refers to visit-to-visit variation, as measured on the basis of
Monitorización Ambulatoria de la Presión Arterial y Eventos
office or out-of-office measurements. An increased short-
Cardiovasculares, i.e. the Ambulatory Blood Pressure
term, as well as long-term, variability is associated with a
Monitoring and Cardiovascular Events (MAPEC) study,
correspondingly increased risk of cardiovascular events.
should be tempered with the lack of effect on nocturnal BP
Specifically in CKD, visit-to-visit variability has been
seen in two other trials [23]. Moreover, the reduction in
reported to be associated with a 20% higher relative risk of
MACE outcomes reported in the MAPEC study with bed-
death and a 90% higher risk of hemorrhagic stroke (though
time dosing (but not additional BP lowering medications) is
with no association for ischemic stroke, heart failure or
higher than the reduction reported in the HOPE study with
ESKD) [17]. Similarly, CKD itself has been associated with
ramipril compared to placebo (hazard ratio [HR] 0.28 in
short-term variability, with the caveat that the absolute vari-
MAPEC and HR 0.78 in HOPE) [20]. Nevertheless, bedtime
ability may seem higher in CKD purely because individuals
dosing is a simple and inexpensive intervention that could
with CKD are more likely to have higher BPs [18]. Certain
be applied to most BP lowering drugs with the exception of
extreme cases are caused by autonomic dysfunction; how-
diuretics.
ever in the vast majority we still do not have a good mecha-
Since sleep apnea is associated with nondipping status,
nistic explanation of this phenomenon. Additionally, a
correction of sleep apnea with a device such as continuous
search for strategies that could potentially mitigate this vari-
positive airway pressure (CPAP) could presumably lower
ability and whether these will also reduce the higher risk of
nocturnal BP. From a meta-analysis of four randomized
adverse outcomes is fertile ground for clinical research.
controlled trials (RCTs), the effect of CPAP therapy was
quite small on nighttime BP, with a mean reduction of
Nocturnal Pattern
1.9 mmHg in systolic BP and 1.5 mmHg in diastolic BP,
The normal pattern of BP has a diurnal variation with lower insufficient to significantly alter dipping status [24]. Not
BP at night and a fall of about 15% being commonly surprisingly, CPAP therapy has also failed to show a benefit
observed [19]. The absence of this nocturnal dipping is overall for MACE or mortality, though other benefits, espe-
defined as an average nocturnal systolic BP being 10% or cially on sleep quality and quality of life, do exist [25].
more lower than the daytime average. This absence of noc-
turnal dipping (i.e. being a “nondipper”) is associated with
Masked Hypertension
an increased risk of cardiovascular events and mortality in
the general population. In the CKD literature, the effects are Masked hypertension deserves a separate mention since it
less clear, since nondipping is also associated with other follows, as a corollary of improper BP measurement, that it
comorbid conditions (e.g. diabetes, proteinuria) which are is underrecognized and potentially undertreated. Masked
themselves also associated with poor outcomes, thus mak- effect, in general, refers to an out-of-office BP measurement
ing causation somewhat confounded [20]. An exception is a that is higher than the office measurement. Masked hyper-
study from Italy, which did take proteinuria and comorbidi- tension refers to the phenomenon when BP may be normal
ties into account and reported an increased hazard of death (or controlled) in the office, but elevated out of office. An
and ESKD amongst nondippers compared to dippers [21]. additional wrinkle is whether this masked hypertension is
The prevalence of nondipper status is quite high in CKD, present, but controlled, or uncontrolled. See Table 22.2 for a
with reports estimating it at 60–80%. Mechanistically, breakdown of the somewhat bewildering terminology. The
334 Hypertension
Table 22.2 Different facets of masked hypertension with a proposed labelling scheme.
Masked, untreated Masked uncontrolled Masked controlled

Possible label Normotension Masked effect hypertension hypertension hypertension
Office BP Normal Normal Normal Normal Normal

Out-of-office BP 10 mmHg of office >10 mmHg of office >10 mmHg of office >10 mmHg of office >10 mmHg of office
BP BP BP BP BP
Out-of-office BP Yes Yes No No No
controlled?
On treatment? No No No Yes Yes
Possible label Normotension Masked effect Masked, untreated Masked uncontrolled Masked controlled
hypertension hypertension hypertension
BP, blood pressure.
exact biological explanation for this phenomenon is not sat- sodium, more potassium, and other dietary interventions),
isfactorily elucidated yet, though observational studies sug- more exercise and weight loss, and lower alcohol intake.
gest associations with age, sex, increased sympathetic These are well-established interventions on the basis of
activity, and certain personality types [26]. RCTs and systematic reviews of RCTs, mostly in the gen-
This phenomenon matters in CKD because the preva- eral population. Most of these can be extrapolated to the
lence of masked hypertension in CKD is higher than in CKD population, with a few considerations, as summa-
the general population. Most reports peg the prevalence rized in Table 22.4 [34, 35].
at about 20–30% with slightly higher numbers (43–50%)
Sodium: A good systematic review from the World
reported in two cohorts with a high proportion of African-
Health Organization (WHO) reports a reduction in BP of
American individuals [21, 27–33]. More important than
about 3.4 mmHg in systolic BP and 1.5 mmHg in dias-
mere prevalence numbers is whether masked hyperten-
tolic BP in trials of reducing sodium intake [36]. The
sion is associated with adverse outcomes. Masked hyper-
mean reduction of sodium intake in these trials was an
tension is indeed associated with surrogate outcomes
impressive 1.7 g/day (45% from baseline). Though these
such as increased left ventricular mass index and protein-
trials were in the general population, one could presume
uria, and also with a higher risk of cardiovascular out-
that the contribution of sodium intake to hypertension
comes (see Table 22.3 for a summary of the reported
would be even higher in the CKD population. One small
literature). Perhaps even more clinically relevant is
RCT from Australia of 20 patients did report a signifi-
whether treatment of masked hypertension mitigates the
cant decrease in BP with sodium restriction, of
risk of these adverse outcomes or just represents a more
−10/4 mmHg, in patients with CKD (glomerular filtra-
aggressive phenotype with the masked hypertension
tion rate [GFR] 15–59 ml/min) [37]. Another small
being an epiphenomenon? An ongoing randomized trial,
crossover RCT of 58 patients with stage 3 or 4 CKD also
the MASTER trial (MASked-unconTrolled hypERtension
reported a decrease of 10.7 mmHg in 24-hour ABPM,
Management Based on Office BP or on Out-of-office,
accompanying a 57 mmol decrease in 24-hour urinary
Ambulatory BP Measurement; NCT02804074) might fill
sodium and a mean weight loss of 2.3 kg [38].
some of these lacunae, randomizing 1240 patients to an
Potassium: Another WHO systematic review highlighted
office versus ABPM guided strategy. The sample size is
the evidence that increasing potassium intake (whether
powered for surrogate outcomes (albuminuria and left
by diet or by using a supplement) also resulted in lower
ventricular mass index) and not hard clinical outcomes,
BP, on the order of −3.5 mmHg in systolic BP and
nevertheless it will be helpful to validate the practice of
−2.0 mmHg in diastolic BP [39]. Increasing potassium
diagnosing and treating masked hypertension.
intake, needless to say, should be done more carefully and
extrapolating to CKD, especially advanced CKD, where
there might be a limit to the tubular potassium excretion,
Treatment of Hypertension in CKD may not be wise. An ongoing RCT of a potassium supple-
ment (40 mmol of potassium chloride, potassium citrate
Nonpharmacological Measures
or placebo) in 400 patients with advanced CKD (GFR
Nonpharmacological measures that have been reported to 15–44 ml/min) will provide more data on the efficacy and
be efficacious in hypertension include changes in diet (less safety of potassium in the CKD population [40].
Treatment of Hypertension in CK 335
Table 22.3 Estimates of masked hypertension in the literature and the association with clinical outcomes.
Prevalence of ABPM profiles
Normotension/
controlled Masked uncontrolled
hypertension hypertension
White coat Sustained

Study Cohort details hypertension hypertension Comments
Pogue et al. n = 617 37% 43% Masked and sustained

(AASK) [28] 100% Black 2% 18% hypertension associated with LV
0% diabetes hypertrophy and proteinuria
Mean age 60 years
62% men
Mean GFR 44 ml/min/1.73 m2
29% proteinuria1
Gorostidi et al. [27] n = 5693 15% 7%
0% Black 29% 50%
33% diabetes
Mean age 67 years
58% men
5% proteinuria2
Iimuro et al. [155] n = 1057 38% 31%
0% Black 6% 26%
35% diabetes
Mean age 61 years
63% men
89% proteinuria3
Minutolo et al. [156] n = 489 17% 15% Masked and sustained
0% Black 22% 47% hypertension associated with
36% diabetes mortality, dialysis and
cardiovascular events
Mean age 64 years
59% men
Mean proteinuria 0.3 g/day
Drawz et al. (CRIC n = 1492 49% 28% Masked and sustained
study) [30] 39% Black 4% 19% hypertension associated with
42% diabetes greater LV mass index
Mean age 63 years
56% men
50% proteinuria2
Agarwal et al. [31] n = 333 10% 50%
17% Black <1% 39%
66% diabetes
Mean age 69 years
98% men
Mean ACR 0.33 g/g
(Continued)
336 Hypertension
Prevalence of ABPM profiles
Normotension/
controlled Masked uncontrolled
hypertension hypertension
White coat Sustained

Study Cohort details hypertension hypertension Comments
Wang et al. [32] n = 588 25% 21% Masked and sustained HT

0% Black 10% 45% associated with greater
16% diabetes mortality, CV and renal events
Mean age 43 years
57% men
Mean proteinuria 1.7 g/day
Kushiro et al. n = 4346 38% 29% Masked, white coat and
(HONEST 0% Black 9% 24% sustained HT associated with
study) [33] 29% diabetes greater CV events
Mean age 71 years
50% men
Scheppach n = 300 41% 18% Increased LV mass index (white
et al. [157] 0% Black 11% 30% coat, masked and sustained
24% diabetes hypertension)
Mean age ~58 years Increased PWV (masked and
sustained HT)
60% men
Mean GFR ~ 51 ml/
min/1.73 m2
AASK, African American Study of Kidney Disease; ACR, albumin to creatinine ratio; CRIC, Chronic Renal Insufficiency Cohort study; GFR,
glomerular filtration rate; HONEST, Home BP measurement with Olmesartan Naive patients to Establish Standard Target blood pressure; LV,
left ventricular.
Table 22.4 Major nonpharmacological measures in management of hypertension in CKD.
Intervention Expected BP decrease Evidence in CKD? Comments
Sodium reduction −3.4/1.5 mmHg Yes Small RCTs in CKD suggest greater BP lowering (~10 mmHg)
Potassium increase −3.5/2.0 mmHg No Risk of hyperkalemia
Await ongoing RCTs
Alcohol reduction −5.5/4.0 mmHg No Little reason to doubt heterogeneity in CKD
Exercise −4.6/2.6 mmHg Yes Small trials, adherence, effectiveness unknown
DASH diet −11.4/5.5 mmHg No Small feeding RCT of n = 11
Observational data suggests benefit
BP, blood pressure; CKD, chronic kidney disease; DASH, Dietary Approach to Stop Hypertension; RCTs, randomized clinical trials.
DASH diet: The Dietary Approach to Stop Hypertension are high in saturated fat, such as fatty meats, full-fat dairy
(DASH) trial demonstrated a robust reduction in systolic products, and tropical oils such as coconut, palm kernel,
BP of 11.4 mmHg and diastolic BP of 5.5 mmHg [41]. The and palm oils, and limiting sugar-sweetened beverages
DASH diet consists of eating vegetables, fruits, and whole and sweets. Of interest in CKD, the DASH diet provides
grains, including fat-free or low-fat dairy products, fish, approximately 4700 mg potassium corresponding to about
poultry, beans, nuts, and vegetable oils, limiting foods that 120 mmol/day. A small feeding trial of 11 patients with
CKD did not demonstrate a significant increase in serum methylphenidate, and over-the-counter remedies such as
potassium (+0.15 ± 0.28 mmol/l) [42]. Longer-term data cough/cold/allergy preparations, appetite suppressants,
are not available. Observational data suggest an associa- licorice and drugs such as cocaine.
tion between following the DASH pattern of diet and
slower decline of kidney function [43], but there are no
Efficacy and Effectiveness
RCTs which support such an effect. Caution should be
in Nonpharmacological Interventions
exercised in recommending the DASH diet in patients
who may have high normal potassium or being concomi- Several caveats should be taken into account when consid-
tantly started on other agents such as renin angiotensin ering nonpharmacological interventions for management
system (RAS) blockade. of hypertension in CKD. Indeed, patients often favor life-
style modifications and shy away from starting BP lowering
Alcohol: Despite robust epidemiological data supporting
medications, demonstrating pill disutility, sometimes of an
the association of higher alcohol intake and BP, most
extreme nature [49]. In situations when the observed BP is
public health messages do not mention temperance as a
~5 mmHg higher than the target, and in a motivated
means of lowering BP. These data are actually now bol-
patient, it is indeed reasonable to focus on lifestyle changes
stered with high-quality trials which show that reduc-
and review at the next visit. However, effects seen in RCTs
tion of alcohol intake results in a lowering of BP, with an
are likely to be an overestimate of the effects one might see
overall effect of −5.5/−4 mmHg for systolic and diastolic
in real clinical practice. RCTs of diet in the hypertension
BP [44]. Moreover, this meta-analysis also demonstrated
literature are feeding trials (and are not trials of coun-
a dose–response effect, with the strongest effect seen in
seling) and as mentioned in the exercise section, the exer-
those who drank more than six drinks and no significant
cise interventions that have the most effect are supervised
reduction in BP seen from further reduction in those
interventions [50]. A healthier diet is more expensive (esti-
who drank two or fewer drinks per day.
mated at ~US $1.5 daily) [51]. Access to resources and care,
Exercise: Evidence from the general population sup- attention to adherence, and physician–patient interaction
ports a recommendation to exercise in an effort to are all additional determinants of true effectiveness.
decrease BP. Specifically, a systematic review of 21 tri- A neglected area, which is especially important for com-
als reported that a supervised program of aerobic exer- plex interventions such as lifestyle change, is poor health
cise of 30–60 minutes, five times per week, can reduce literacy, which is especially common in CKD [52–54]. In
systolic BP by a pooled estimate of 4.6 mmHg and dias- the setting of poor health literacy, pharmacological therapy
tolic BP by 2.6 mmHg [45]. In addition, we do have data may be considered preferable to these complex and often
in CKD patients from a systematic review which costly lifestyle changes.
reported the effects of exercise (including cardiovascu-
lar training, mixed cardiovascular and resistance train- Level and Drug Choice
ing, resistance-only training, and yoga) [46]. The
review reports a similar overall reduction in BP (mean Over the last few decades, the research has shifted from
difference, 6.1 mmHg in systolic from nine trials and whether high BP should be lowered at all, to how should it
2.3 mmHg in diastolic from 11 trials) as the previous be lowered, and now to how low should it be lowered [55,
review conducted in the general population. However, 56]. With respect to first-line drug to use in hypertension, the
in subgroup analyses, the effects were significant in the seminal trial in the primary care population was the
pooled analysis of trials when exercise training was Antihypertensive and Lipid-Lowering Treatment to Prevent
supervised (and was not significant in trials when it Heart Attack Trial (ALLHAT), which reported that a
was unsupervised) and in trials with high-intensity thiazide-like diuretic (i.e. chlorthalidone) was superior to
exercise training (and not significant in trials with low- alpha-adrenergic antagonists (doxazosin) and just as good as
intensity exercise). Lastly, emerging data also suggest an angiotensin converting enzyme inhibitor (ACE-I, lisino-
that the beneficial effect of exercise on BP is not pril) or a calcium channel blocker (CCB, amlodipine) [57].
restricted to aerobic, but also to dynamic resistance and This and other evidence have lead to the acceptance of one
isometric static exercise [47, 48]. of the so-called A-C-D drugs (i.e. ACEi or angiotensin recep-
tor blocker [ARB], a CCB, or a diuretic) as the choice for
Stop Interfering substances: Interfering substances which first-line BP lowering drug, as well as being the first three
can increase BP include prescription drugs such as non- drug classes that should be used for BP lowering. These
steroidal anti-inflammatory drugs (NSAIDs), corticoster- drugs are also the ones to be preferably combined when one
oids, erythropoiesis-stimulating agents, antidepressants, drug is not enough, as is common in CKD.
338 Hypertension
With respect to the level of BP control, BP control down However, one might consider this as an academic ques-
to 140/90 has been widely accepted. Beyond that, to what is tion, since most CKD patients do require more than one
labeled as “intensive BP lowering”, to systolic BP levels of agent, and RAS blockade being part of that armamentar-
<120 or <130, is still a matter of contentious debate, and is ium is eminently sensible.
discussed below. What is not sensible is the use of an ACEi and an ARB
in combination. In patients at high CV risk, and in
patients with proteinuric diabetic nephropathy, the use of
Pharmacotherapy these in combination did not demonstrate any benefit in
CV or kidney outcomes [60, 61]. Moreover, there was an
Nonpharmacological measures are insufficient for BP low-
excess risk of hyperkalemia and acute kidney injury, and
ering in most CKD patients, most of whom need pharma-
since then dual RAS blockade has not been recommended
cotherapy not just for BP lowering, but in addition for
for use in management of hypertension. Lastly, the direct
specific indications such as the antiproteinuric effect or
renin inhibitor aliskiren might have similar effects as
more specifically for CV protection. A large cohort study
ACEi or ARB. However, the major clinical trials were all
suggests that about 60% of CKD patients receive three or
conducted in combination with either an ACEi or an
more antihypertensive medications, thus making the dis-
ARB, and as with ACEi–ARB combination, these did not
cussion of the specific BP lowering class somewhat redun-
demonstrate any benefit and increased risk of adverse
dant [58]. The discussion below summarizes some of the
events [62]. Monotherapy with aliskiren might have simi-
evidence of the specific medications, while acknowledging
lar effects as ACEi or ARB monotherapy, but this has not
that combination therapy with most of the first-line drugs
been proven and hence aliskiren use may be relegated to
will be the most common scenario.
the rare occasion when RAS blockade is desirable, but
both drugs are not tolerated.
Renin Angiotensin System Blockade
Thiazide and Thiazide-like Diuretics
RAS blockade, consisting of either ACEi or angiotensin
receptor blockade (ARB), has become a mainstay of Thiazide diuretics are the oldest class of BP lowering
hypertension management in CKD. ACEi blocks the con- drugs which are still first-line agents. They act on the Na-
version of angiotensin I to angiotensin II, a potent vaso- Cl transporter in the distal convoluted tubule and acutely
constrictor, and ARBs block the angiotensin II receptors. do cause natriuresis. With chronic therapy, however,
The downstream reduction in aldosterone secretion and extracellular volume comes back to baseline but periph-
sodium reabsorption, coupled with a decrease in angio- eral resistance falls, from an unknown mechanism, but
tensin II medicated vasoconstriction, underlies their BP thought to be related to the initial natriuresis. The main
lowering effect. Extensive trial data demonstrate the ben- thiazide diuretics are hydrochlorothiazide (HCTZ),
efit in CV outcome reduction with these agents. However, Bendroflumethiazide, and chlorothiazide. The thiazide-
a careful examination of how much of their effect is inde- like diuretics, namely indapamide and chlorthalidone,
pendent of BP lowering reveals a slightly complicated are somewhat longer acting, have greater effect on noc-
story. Briefly, when compared against placebo, RAS turnal BP, and are more potent compared to HCTZ, the
blockade is indeed superior in preventing CV outcomes most commonly used thiazide [63]. Moreover, in the
and prolonging survival. However, when compared RCTs that demonstrate a benefit with diuretics, either
against active comparators, i.e. other BP lowering drugs, chlorthalidone or indapamide have been used [10, 57, 64].
then there is no significant difference for hard clinical Even in SPRINT, chlorthalidone was preferentially used
outcomes in patients with heart disease, or with diabetes, in the formulary. Some other RCTs in which HCTZ has
or with nonproteinuric CKD. The major factor in all these been used have failed to report beneficial CV effects [9,
systematic reviews is the inclusion of ALLHAT, which did 65]. However, no head-to-head trial of HCTZ compared to
not report a benefit to using lisinopril over chlorthalidone chlorthalidone exists, though one is ongoing [66].
and, being a large trial, carries a lot of weight in the meta- Additionally the increased potency of thiazide-like diu-
analysis [59]. The only exception, where RAS blockade is retics can result in more adverse events related to electro-
clearly superior to other BP lowering drugs, is in the set- lyte imbalances in particular [67]. For relevance to CKD,
ting of proteinuric diabetic nephropathy. RAS blockade thiazide diuretics were not thought to be effective in
has an additional effect on the intraglomerular pressure severe CKD (GFR < 30). This has been challenged in a
by dilatation of the efferent arteriole, reducing proteinu- recent pilot trial with chlorthalidone and more data is
ria and explaining this additional beneficial effect. awaited [68].
Calcium Channel Blockers of BP by −3.8/−1.8 mmHg with these agents, which also

have clinical outcome data in the form of heart failure and
CCBs bind to the alpha-1 subunit of L-type calcium channels,
CV event reductions [72, 73]. The neprilsyn inhibitor,
lowering cytosolic calcium and resulting in vascular smooth
sacubutril/valsartan, has been reported to improve survival
muscle relaxation and consequent vasodilatation. The nondi-
in patients with congestive heart failure [74]. It has been
hydropyridine CCBs (diltiazem and verapamil) also have an
shown to be safe in a recent trial in CKD patients (GFR
additional negative chronotropic effect on the heart. Because
20–60) and reduced BP more than irbesartan by about
of their additional drug interactions (inhibition of the
5 mmHg [75]. Other classes of drugs which show promise
cytochrome p450 3A4 system) and drug interactions, these
in BP lowering and are under investigation include aldos-
are used less often compared to the dihydropyridine CCBs,
terone synthase inhibitors, ACE-2 activators, aminopepti-
namely nifedipine, amlodipine, and others. These drugs are
dase inhibitors, angiotensin Ii vaccines, and endothelin
mostly hepatically metabolized and do not require dose
antagonists [76].
adjustment, have few metabolic adverse effects, and work
throughout the spectrum of CKD.
Combination Choice
Second-line Drugs Mechanistic considerations and empiric data, along with

pharmacokinetic considerations and adverse effects, are
Spironolactone, doxazosin, and bisoprolol all significantly the factors that go into decisions about drug combinations.
reduced BP compared to placebo, as add-on to the A-C-D From a physiological perspective as well as reviewing trial
combination in patients with resistant hypertension [69]. data, combining RAS blockade with either CCB or diuret-
The highest BP lowering effect was with spironolactone, ics makes sense. Conversely, combining a beta-blocker
and this would be the appropriate fourth agent to use in with RAS blockade does not, given that renin release would
this setting. The trial, however, excluded patients with GFR already be reduced by the beta-blocker. Similarly, combin-
<45 ml/min, and the efficacy of spironolactone for BP lowing a CCB with a diuretic has no specific synergistic advan-
ering is not well established in severe CKD. A small obser- tages (and often the diuretic has been mistakenly added to
vational study does suggest it has BP lowering effects below manage CCB-induced leg edema, better treated with a
a GFR <45, albeit with the additional unsurprising risk of lower dose of the CCB and addition of RAS blockade). For
hyperkalemia [70]. A follow-up of the same Prevention patients with CKD, given the primacy of RAS blockade,
And Treatment of Hypertension With Algorithm-based combining them with either a diuretic or a CCB makes
therapy (PATHWAY) trial also suggested that the efficacy eminent sense, followed by addition of the third agent.
of amiloride may be similar to spironolactone [71]. Since the publication of the PATHWAY study, as well as
Eplerenone is another mineralocorticoid antagonist which observational data supporting the relatively high preva-
does not have the androgenic side effects of spironolac- lence of hyperaldosteronism in this population, a miner-
tone. Alpha- and beta-adrenergic antagonists are the next alocorticoid antagonist is the ideal fourth agent [69].
preferred agents, based on the PATHWAY-2 trial [69]. Beta-blockers, alpha adrenergic antagonists and vasodila-
These suggestions are based on BP lowering and not on tors and central sympatholytic agents come next. Little trial
clinical outcome data. evidence for clinical outcomes exist for these drugs, hence
Centrally acting alpha-agonists (clonidine and guanfa- their relegation here. It is not unusual for their use to be
cine) and direct vasodilators (minoxidil and hydralazine) are common in later stages of CKD, when fears of worsening
also potent in BP lowering and reserved for use after exhaust- creatinine have lead to discontinuation of RAS blockade
ing all above options. Clonidine is available in some coun- and/or diuretics.
tries as a transdermal patch, which can be applied once a
week (especially useful for hemodialysis patients). Though
hydralazine has some evidence (though not in the CKD set- Device Therapy
ting), it requires three to four times a day dosing and minoxi- Device therapy has shown some promising results in recent
dil is preferable for BP lowering. years, but it is far from being ready for clinical use yet. It is
nevertheless important to review the most promising inter-
ventions and any potential role they might have for man-
Newer Agents
agement of hypertension in CKD patients.
Sodium-glucose transporter inhibitors (SGLT2i), which are Sympathetic renal nerve ablation has the most data
perceived as hypoglycemic agents, also have natriuretic and among all device-based interventions. We have known
BP lowering effects. A meta-analysis reports a lowering that lumbar sympathectomy is extremely effective in
340 Hypertension
lowering BP since Adson’s first case report decreasing the P

atient Populations
BP from 250/180 to 170/120 after ventral rhizotomy [77].
However, this procedure was almost always followed by Intensive BP Lowering
impressive side effects, such as loss of sensation, paralytic
ileus, problems with ejaculation, and loss of sweating and RCTs of BP lowering have traditionally been conducted in
the occasional mortality [78]. With the advent of safe BP carefully defined populations based on demographics or
lowering drugs, sympathectomy went rapidly out of favor. comorbidities, e.g. the elderly (Systolic Hypertension in the
This was the state of affairs until a few years ago, when an Elderly Program [SHEP] and Hypertension in the Very
intrepid group reported on the role of renal sympathetic Elderly Trial [HYVET]) [64, 90], diabetes (ACCORD) [11],
nerves in resistant hypertension, followed by the success- African-American with kidney disease (AASK) [91] and so
ful demonstration of percutaneous renal denervation on. These groups are not homogenous when it comes to CV
using a catheter with a dramatic BP fall [79, 80]. The ini- risk, which is the main purpose of treating hypertension.
tial excitement following the first few clinical trials came Hence the benefits (and risks) of BP lowering will vary
crashing down to Earth after the release of the Symplicity even within these populations based on their underlying
HTN-3 trial [81]. This was the first sham controlled and CV risk. On the other hand, stratifying populations based
double-blind RCT, performed in patients with resistant on the underlying CV risk may represent a cleaner view on
hypertension, and showed no difference in BP (~14 mmHg the risks and benefits of BP lowering, especially of inten-
decrease in both groups). However, subsequent soul- sive BP lowering. Indeed, a meta-analysis from the BP
searching and an awareness of the problems inherent in Treatment Trialists Collaboration does show that there is
studying patients with resistant hypertension (with high no heterogeneity on relative risk reduction between trials
variance and varying adherence) lead to the design of dif- with and without CKD or diabetes [92]. The common
ferent RCTs [82]. The resultant RCTs, this time in patients denominator for understanding treatment benefit is the
with mild hypertension, reported a decrease of about underlying baseline CV risk. As shown in Figure 22.1 from
5–6 mmHg and this seems to be the true average BP low- an individual patient level meta-analysis, the absolute risk
ering effect of renal denervation [83, 84]. Safety in the benefit with BP lowering increases with higher level of
short term seems good, though the absence of a renal baseline CV risk [93]. One proposed scheme to think about
sympathetic afferent system has been associated rarely stratifying patient population while thinking about inten-
with shock and acute kidney injury [85]. Do these results sive BP lowering is shown in Figure 22.2 [56]. Needless to
mean renal denervation could help CKD patients? Most say, intensive BP lowering requires careful assessment of
CKD patients do require more than one BP lowering the risk of adverse events, such as electrolyte imbalances
medication, so the advantages of reducing pill burden by and orthostatic hypotension.
one (considering the average effect of −5 mmHg) is small.
Second, while increased sympathetic activity might
Nondiabetic CKD
indeed have a larger role to play in hypertension in
advanced CKD, there is no data on clinical outcomes. Four major RCTs have been conducted in nondiabetic
Small case series do report that denervation is technically CKD with BP lowering targets, as summarized in
feasible in CKD patients, so the benefits will have to be Table 22.5 [10, 91, 94–96]. As can be seen, the earlier tri-
carefully weighed if denervation becomes widely availa- als were not powered for CV outcomes, nor for all-cause
ble [86, 87]. mortality. In one of the trials, the separation between
The other device therapies under development include arms of the BP was also minimal. The largest and most
baroreceptor activation and the arteriovenous fistula crea- consequent RCT in this area is hence the Systolic Blood
tion. Baroreceptor activation therapy in theory modulates Pressure Intervention Trial (SPRINT) [10]. In the main
“deranged sympathovagal balance” but in practice the clin- trial, there was no significant interaction between the
ical trial did not meet safety or efficacy endpoints, though patients with CKD and those without CKD, which is the
further refinements are ongoing [88]. Central iliac arterio- proper method to assess if patients with CKD are differ-
venous anastomosis creation has been reported in a small ent in their response to intensive BP lowering. In addi-
open-label trial, in which ipsilateral venous stenosis devel- tion, the detailed analysis of the CKD subgroup published
oped in a third of the patients (along with significant BP subsequently confirms the benefit of intensive BP lower-
reductions) [89]. Lack of blinding, lack of data on clinical ing for all-cause mortality [94]. A subsequent systematic
outcomes, and serious long-term safety concerns suggest review, including data from 18 RCTs and 15 924 partici-
this is the least promising of all device-based BP lowering pants confirms these findings, reporting 14% relative
interventions. risk reduction in mortality with intensive BP lowering
Patient Population 341
Active Control Mean BP difference, Risk difference (95% CI) Risk ratio (95% CI)
(n/N) (n/N) mm Hg
5-year risk of CVD
<11% 419/10644 623/14836 4.6/3.0 –1.41 (–2.05 to 0.77) 0.82 (0.73 to 0.93)
<11–15% 443/5679 599/6865 6.0/3.2 –1.95 (–3.09 to 0.82) 0.85 (0.75 to 0.96)
15–21% 467/3944 575/4343 7.1/3.2 –2.41 (–4.04 to 0.77) 0.87 (0.78 to 0.98)
>21% 472/2760 569/2846 5.9/3.0 –3.84 (–6.06 to 1.61) 0.85 (0.76 to 0.95)
p(het)=0.004, p(trend)=0.04 p(het)=0.93, p(trend)=0.30
–5 –4 –3 –2 –1 0 0.5 1 1.5
Favours active treatment Favours control Favours active treatment Favours control
Figure 22.1 Effects of blood pressure reduction on absolute and proportional risks of cardiovascular disease for patient groups
defined by different baseline levels of cardiovascular risk. Total sample, n = 51 917. n/N is the number of cases/number at risk. BP
difference is the difference between active and control groups in treatment-induced reduction in systolic/diastolic blood pressures. BP,
blood pressure; CVD, cardiovascular disease; Het, heterogeneity. Source: From Sundstrom et al. [93] used with permission. © 2014,
Elsevier.
(with an achieved systolic BP fall from a mean of 148 to intensive glycemic control group, intensive BP lowering
132 mmHg) [97]. Indeed, the risk of adverse events, increased mortality [100]. Thus it is indeed possible that
mostly electrolyte disorders, do also go up as a conse- intensive BP lowering not only confers a benefit in stroke
quence of adding potent antihypertensive agents, but reduction, but may not increase mortality if done in con-
this risk should be weighed against the benefit of not junction with standard, but not intensive, glycemic
dying. lowering.
Another approach, using a Bayesian analytic model of
conducting a meta-analysis of achieved BP suggests inten-
Diabetic CKD
sive BP lowering is associated with a 10% reduction in all-
There are no trials which specifically included patients cause mortality, a 17% increase in stroke, but with a 20%
with diabetes and CKD, and targeted a certain BP level increase in serious adverse effects [101]. Specifically, it
intervention. From the general population with diabetes, also suggests that these benefits were seen with an
the most relevant trial is Action to Control Cardiovascular achieved BP of 130–135 mmHg, with further reductions
Risk in Diabetes Blood Pressure (ACCORD) [11]. In this only providing additional benefit in stroke reduction.
randomized trial, 4773 patients with diabetes and at high These and other meta-analytic approaches either rely on
CV risk were randomized to two BP targets: systolic BP baseline BP or achieved BP for grouping trials and
<140 mmHg versus systolic BP <120 mmHg. The overall arms [102–104]. They also include RCTs of BP lowering
primary composite outcome was not significantly different agents (e.g. olmesartan or lisinopril-losartan combina-
between the two groups, with a hazard ratio of 0.88 (95% tion) compared with placebo, where the achieved BP
confidence interval [CI] 0.73–1.06). Of all the specific would be higher with placebo, but where the interven-
events included, stroke (which arguably has the strongest tions are no longer used as tested in these RCTS (e.g. dual
link with BP) was significantly lower with the intensive BP blockade) [60]. To finish this discussion, the most recent
target (HR 0.59, 95% CI 0.39–0.89). There was much con- meta-analysis of these reports on the CV effect of a stand-
sternation about the slightly greater number of deaths with ardized decrease of 10 mmHg on outcomes [4]. Overall,
intensive BP lowering (150 versus 144, HR 1.07, 95 CI 0.85– they report a benefit of 10 mmHg BP lowering on CV out-
1.35). Intriguingly, this has lead to many different post hoc comes (including stroke and heart failure) and on all-
analyses, with two major schools of thought being that cause mortality in diabetes, with little heterogeneity
either ACCORD was underpowered (the observed event between trials in patients with and without diabetes (see
rate was about half as expected) or the other showing an Figure 22.3).
effect modification [98]. ACCORD had a factorial design, These analyses, needless to say, are from the general pop-
with the other intervention being intensive blood glucose ulation, and not in CKD, where we must necessarily
lowering, which was found to be harmful compared to extrapolate the evidence. Should we expect outcomes to be
standard glucose control [99]. In a post hoc analysis of different in patients with diabetes and CKD compared to
ACCORD published a decade after the original publica- those without CKD? From a mechanistic perspective,
tion, a significant interaction (P = 0.03) was reported, such patients with diabetes and CKD are possibly different, but
that in the group receiving standard glucose lowering, from an empiric point of view, no interaction has been
intensive BP lowering decreased mortality, whereas in the reported yet from epidemiological studies.
342 Hypertension
Table 22.5 Details of the four major trials targeting BP

thresholds in nondiabetic CKD.
Outcomes: renal, CV
Trial Population and mortality
Elderly
Klahr et al. n = 840 Decline in GFR
Frail (MDRD) [95] Inclusion: 10.7 (9.113.9) ml/
General Age 18–70 years min/3 years in usual
Hypertensive BP arm
Population CrCl <70 ml/min
CKD 12.3 (10.6–14.0) ml/
MAP <125 mmHg min/3 years in low BP
Diabetes arm
Exclusion:
Diabetes,
Proteinuria >10 g/day ESRD or death
Chronic medical RR 0.85 (0.60–1.22)
conditions
(a)
Body weight <80%
or >160% standard
body weight
Low Risk
Wright et al. n = 1096 % risk reduction for
(AASK) [91] Inclusion: GFR slope + ESRD
Intermediate Risk
Age 18–70 years −2 (95% CI −31 to
+20)
GFR 20–65 ml/min
High Risk
(includes CKD) Exclusion:
CV
Diabetes
2.9% (lower BP)
Clinical evidence of versus
heart failure
3.4 % (usual BP)
High Risk of Proteinuria (urine
adverse
PCR > 2.5 g/g)
events Mortality
Serious systemic
illness 1.6% (lower BP) vs.
1.9% (usual BP)
Rugenenti n = 338 Decline in GFR

(b) et al. Inclusion: 0.22 ml/min/mo in
(REIN-2) [96] intensified BP arm
Figure 22.2 (a) This schematic (not to scale) shows the current Age 18–70 years
model of presenting blood pressure targets, with different versus
recommendations for diabetes, chronic kidney disease, and the Proteinuria 1–3 g/day
and CrCl 0.24 ml/min/mo in
elderly, though overlap between these conditions often occurs. usual BP arm
(b) Suggested schematic for classification according to <45 ml/min OR
(P = 0.62)
cardiovascular (cardiovascular) risk for blood pressure targets. Proteinuria >3 g/day
Patients at high risk of adverse events (irrespective of and CrCl
cardiovascular risk) deserve individualized consideration for >70 ml/min CV
whom intensive targets may not apply. Exclusion: 37 (22%) in intensified
Source: From Ruzicka et al. [56] used with permission. © 2017, arm versus
Wolters Kluwer. Diabetes
Overt heart failure 25 (15%) in usual
(NYHA III or IV) arma
Advanced CKD
Recent (<6 months)
Little evidence exists about BP management in advanced MI or CVA Mortality
CKD, once the GFR is less than 20 in particular, since these 2 (1.2%) in intensified
patients have been excluded from the large trials that guide versus
practice. The outcomes of interest are indeed different in 3 (1.8%) in usual arm
these patients, since the specter of dialysis looms much
(Continued)
larger. Intensive BP lowering does cause a decrease in GFR,
which in some patients might tip them over to dialysis.
Hemodialysi 343
Table 22.5 (Continued) measurements do not necessarily provide a proper assess-

ment of the overall BP load. Nevertheless, hypertension is
Outcomes: renal, CV very common in dialysis patients, with an estimated prev-
Trial Population and mortality alence of 70–80% or higher, irrespective of whether predi-
alysis BP or interdialytic measures using ABPM are
Wright et al. n = 9361 overall; In CKD subgroup
(SPRINT) 2646 with CKD used [107, 108].
50% GFR reduction
[10, 94] Inclusion: RR 0.87 (0.36–2.07)
Age >50 years AND Dialysis
●● Clinical or subclinical RR 0.57 (0.19–1.54) P
eritoneal Dialysis
CV disease OR
●● Framingham risk BP measurement in peritoneal dialysis (PD) is less of a
CV
score > 15% OR
RR 0.82 (0.63–1.07) quagmire compared to HD, though we do not have any
●● Age >75 OR
RCTs to guide us regarding the BP level which would best
●● CKD (GFR 20–59 ml/
min) Mortality reduce CV outcomes and mortality in this population.
Exclusion: RR 0.73 (0.53, 1.00) Unlike some early estimates, hypertension is quite com-
GFR <20 ml/min mon in PD, and the overall prevalence is similar to that in
Proteinuria >1 g/day (or HD [109]. High BP also has similar associations with CV
albuminuria >600 mg/ outcomes, however, residual renal function and tech-
day) nique failure are additional important outcomes to con-
Residence in a nursing sider. The circadian pattern of BP in PD is different than
home, dementia
in HD which is not surprising given the nature of and
Standing systolic BP
<110 mmHg
timing of salt and fluid removal between the two modali-
ties [110]. Given their benefit in preserving kidney func-
CVA, cerebrovascular accident; CV, cardiovascular disease; ESRD, tion in nondialysis CKD, one should not be surprised that
end stage renal disease; GFR, glomerular filtration rate; MAP, mean RAS blockade would be useful in this setting to preserve
arterial pressure; MI, myocardial infarction; NYHA, New York Heart
residual function as well [111]. Additional intriguing data
Association; PCR, urine protein:creatinine ratio; RR, relative risk.
Figures in parenthesis refer to 95% confidence intervals. suggests that spironolactone may be useful in this setting
a
Reported as “non-fatal serious adverse events, including myocardial since it not only reduces BP, but increases potassium lev-
infarction, congestive heart failure, stroke and cancer”; els, which is useful since hypokalemia is quite common in
Source: Used with permission from Ruzicka et al. [56] © 2017,
PD patients [112]. With respect to volume management,
Wolters Kluwer.
body impedance analysis guided volume management
Adverse effects from drugs required to lower BP, namely resulted in a modest 2.7 mmHg drop in systolic BP with
electrolyte disorders, are also more common in advanced no impact on clinical outcomes [113]. The use of
CKD. This would specifically also apply to the use of RAS icodextrin-containing solutions is a useful method for
blockade. An ongoing trial in which 410 patients with pro- better volume management and has been reported to
gressive stage 4 or 5 CKD have been randomized to con- lower BP in a few RCTs, without impairing residual func-
tinue or stop ACEi/ARB will answer this question [105]. tion or having the deleterious effect of high glucose solu-
However, if the benefit of intensive BP lowering (i.e. a tions [114, 115]. Reducing dialysate sodium to levels
decrease in all-cause mortality and cardiovascular events) lower than what are commercially available is also
reported in the general population and milder stages of an intriguing approach so far only tried in phase two
CKD applies to this population also, one has to consider trials [116, 117].
this against the real risk of needing to start dialysis from
worsening GFR. This trade-off requires shared decision
making and open discussion about preferences and patient H
emodialysis
important outcomes. Additionally, initiation of dialysis
should not be on the basis of GFR alone, which also might Which BP Measurement to Use?
help some of the decision-making process [106].
Though predialysis and postdialysis BP measurements are
considered most often, they are least useful given events
Dialysis
around dialysis initiation, which may cause a physiological
The foremost question in dialysis populations is that of BP increase in BP. Interdialytic ABPM measurements, which
measurement. In hemodialysis (HD) though many BP can be done over the 44-hour interdialytic interval, provide
measurements are taken during a session, many of those a more accurate assessment of the BP load as well as the
344 Hypertension
Baseline Events Participants Events Participants RR per RR (95% Cl)

Subgroup Studies
SBP Intervention Intervention Control Control 10 mmHg lower SBP
Major Cardiovascular Events

Diabetes 23 145 3999 30796 4007 28977 0.88 [0.82; 0.94]
No Diabetes 18 153 3494 52616 4035 46427 0.75 [0.70; 0.80]
Fixed effect model 0.81 [0.77; 0.85]
Test for interaction p=0.0006
Coronary Heart Disease

Diabetes 19 144 1404 26688 1487 25441 0.88 [0.80; 0.97]
No Diabetes 15 153 1143 41462 1389 35389 0.77 [0.70; 0.86]
Stroke
Diabetes 21 144 1263 29643 1368 28421 0.74 [0.65; 0.84]
No Diabetes 19 153 1691 52920 1978 46852 0.74 [0.67; 0.81]
Heart Failure
Diabetes 13 143 1028 21424 1094 20536 0.84 [0.72; 0.98]
No Diabetes 10 143 263 18578 373 18442 0.75 [0.65; 0.87]
Rental Failure
Diabetes 9 143 760 14549 774 13641 0.88 [0.75; 1.03]
No Diabetes 4 146 592 9921 638 9952 0.92 [0.79; 1.07]
All–cause Mortality
Diabetes 20 144 2324 27562 2309 25782 0.87 [0.79; 0.97]
No Diabetes 17 153 1952 45588 2033 39414 0.83 [0.76; 0.90]
0.5 1 2 3
Figure 22.3 Standardized effect of a 10 mmHg reduction in systolic blood pressure (SBP) on the relative risk (RR) of major
cardiovascular events, coronary heart disease, stroke, heart failure, renal failure, and all-cause mortality stratified by subgroups in
which all (DM) or none (No DM) of the participants had diabetes mellitus at baseline. Source: From Ettehad et al. [4], used with
permission. © 2016, Elsevier.
circadian pattern. Similar data exist about home BP, and moreover rely on predialysis BP measurements. Cross-
these out-of-office measures are also more reproducible sectional data is notorious for susceptibility to selection
and have a stronger association with CV outcomes. bias and especially in HD populations is even more sus-
Early studies on BP and outcomes in hemodialysis ceptible to reverse epidemiology [119]. Patients with
reported a U-shaped curve of BP and survival in hemodi- heart failure, cancer, and patients in the last few months
alysis, and most concerning has been that the nadir of of their life have lower BP, and will also be more likely to
the U was reported at about systolic BP 160 mmHg, much die, independent of the BP. This leads to spurious asso-
higher than conventional thinking about BP targets and ciations which have been widely reported. In contrast, a
thresholds [118]. Although almost all other biological meta-analysis of RCTs in dialysis patients shows a reduc-
variables have a U-shaped relationship with clinical out- tion of CV events with BP lowering [120]. Similarly, a
comes, the precise optimal BP being 160–180 mmHg linear relationship between BP and CV outcomes up to a
would be a huge change from the ~120–140 in nondialy- systolic BP of 110–120 mmHg was reported when out-of-
sis populations. The former has indeed been reported, office BP measurement was used rather than predialysis
but those reports are from cross-sectional data and BP [120, 121].
Pharmacotherap 345
Nonpharmacological Measures P
harmacotherapy
Minimizing dietary sodium intake: The focus of communi-
cation with patients on dialysis is often on fluid intake, a From the limited trial evidence available, discussed
thoroughly useless and futile waste of time and energy. below, the use of beta-blockers as initial agents, followed
Sodium intake drives fluid intake, and efforts to reduce by RAS blockade and CCBs is a reasonable approach to
sodium intake are likely to have a greater effect on BP in be considered.
the dialysis population given the outsize contribution of A few RCTs have been conducted of BP lowering drugs
sodium to BP in this population. Data supporting this is compared to placebo in HD patients. The results are decid-
mostly from observational literature and, as before, the edly mixed, owing potentially to small sample sizes, co-
effectiveness of counseling on change in behavior is interventions, and open label designs. Two small RCTs
unknown. suggest a benefit of beta-blocker use in improving survival.
Maximizing dialytic sodium removal: This can be For RAS blockade, one RCT of fosinopril failed to show a
achieved by adequate ultrafiltration by probing the dry benefit for CV outcomes (P = 0.06), and two small open
weight, especially in patients with high BP even without label RCTs of different ARBs seemed to show a benefit on
any signs of overt volume overload, and by considering CV outcomes [130, 131]. The most rigorous of the RAS
longer or more frequent dialysis in those who do not tol- blockade trials is the most recent, which was open label,
erate ultrafiltration. There is RCT-level evidence for the but had an adequate sample size and where the target BP
former as well as for the latter. These data, however, dem- was similar in both groups, and did not report any differ-
onstrate efficacy in the form of a change in BP, but not on ence between the two arms for CV outcomes or for all-
consequent clinical outcomes [122]. Longer and more fre- cause mortality [132]. Lastly, one RCT of amlodipine,
quent dialysis does not improve quality of life or longev- which enrolled fewer patients and reported fewer events
ity, and may have deleterious effects on vascular access than expected, still reported a CV outcome benefit [133].
and residual renal function, which are important consid- However, achieved systolic BP was about 10 mmHg lower
erations [123–126]. The last mechanism of maximizing in the amlodipine group, hence it is hard to separate the BP
sodium removal is by altering dialysis sodium. effect from the drug effect. Thus, to summarize, there are
Observational data provide mixed evidence in this regard. clear uncertainties about the benefits of these individual
The clinical trial data do suggest that lowering or indi- classes of drugs in HD patients. The existing data do seem
vidualizing dialysate sodium to patient serum sodium is sufficient to elevate beta-blockers, along with RAS block-
an efficacious method of lowering BP [127–129] (see ade and CCBs, to the first line of therapy when needed to
Table 22.6). Given the association of hemodynamic stabil- control BP. These data also suggest that BP lowering with
ity and lower cramps (both patient important outcomes) pharmacotherapy overall is beneficial in decreasing CV
with higher dialysate sodium, higher quality data from an events.
ongoing cluster RCT may help resolve this dilemma Some additional consideration is necessary to consider
(Randomized Evaluation of Sodium Dialysate Levels on pharmacokinetic aspects, including the effect of dialyzability.
Vascular Events [RESOLVE] NCT 02823821). Recent research has revealed that with modern high
Table 22.6 Clinical trials of changing dialytic sodium and blood pressure.
Study Population Design Interventions Results
De Paula n = 27 Non randomized, Dialysate sodium Individualized dialysate sodium

et al. [127] Non diabetes and not prospective, crossover study 138 mmol/l or set to decreased weight gain (− 0.6 kg)
prone to hypotension 3 weeks each patients serum sodium BP similar between two arms
Munoz n = 15 Prospective, crossover trial Dialysate sodium 140 Lower dialysate sodium with
Mendoza [128] In-center nocturnal 12 weeks each (three versus 136/134 versus lower weight gain (0.6 kg) BP
hemodialysis phases) 140 change −8.3 mmHg (±14.9)
Inrig [158] n = 16 Prospective, crossover trial Dialysate sodium set at Lower dialysate sodium lowered
Patients with history of 3 weeks each serum sodium +5 or BP (mean −9.5 mmHg, 95% CI
intradialytic −5 mmol/l −13.4, −6.4 mmHg)
hypertension
BP, blood pressure.

346 Hypertension
flux dialysers, drugs such as bisoprolol, earlier thought to CCBs, RAS blockade, and even thiazide-like diuretics,
be nondialyzable, are indeed removed by dialysis [134]. On are potent in lowering BP in this population [146, 147].
the other hand, despite atenolol being dialyzed out, it did For outcomes, the strongest data support the use of CCBs
show greater efficacy compared to lisinopril in one of the as BP lowering drugs, as reported in this systematic
few RCTs in this area [135]. Thus it is not clear that dialyz- review, and reduced graft loss (relative risk 0.75, 95% CI
ability, or the lack thereof, matters in terms of efficacy in 0.57–0.99) [148]. For RAS blockade, the evidence is
CV outcome reduction [136, 137]. However, the prolonged decidedly mixed when it comes to graft or patient sur-
half-life of certain drugs can be used for more convenient vival. Overall, when combining ACE-inhibitors and
dosing schedules, especially when there is a concern of ARBs as a class, there is no benefit seen against placebo
nonadherence, and direct observed therapy can be pro- or active comparators in this meta-analysis [149].
vided with three times a week dosing in the dialysis Intriguingly, the few ARB trials seem to report better
unit [138]. graft survival than the ACEi RCTs, which may be a
An additional note is warranted about the utility of miner- chance finding or worthy of further dissection or a larger
alocorticoid antagonists, namely spironolactone and replication [150, 151]. Other BP lowering drugs have lit-
eplerenone, which have had resurgence in resistant hyper- tle data on clinical outcomes and their use should be
tension as well as in ESKD. Two trials suggest a fantastic guided based on the BP lowering effect desired, adverse
benefit of these in lowering CV events and mortality, but effects, and pharmacokinetics.
both trials had a relatively small sample size to demonstrate
such an outsize benefit [139, 140]. Another pilot/feasibility
study of 100 patients did not demonstrate any difference, but Guidelines and Variability
did show a higher risk of hyperkalemia, borne out in a sub- There is little agreement when it comes to the level of BP
sequent systematic review [141, 142]. Given these aspects, lowering in the plethora of existing international and
prudence would dictate awaiting the results of two large national guidelines, even when it comes to the CKD sub-
RCTs ongoing in this area (Aldosterone bloCkade for Health population [152–154]. Despite a seeming overabundance
Improvement EValuation in End-stage Renal Disease of clinical trials, the same evidence is rated and weighed
[ACHIEVE] NCT03020303 and ALdosterone Antagonist differently, resulting in this difference of opinion. Points
Chronic HEModialysis Interventional Survival Trial that are common to all guidelines are that the systolic BP
[ALCHEMIST] NCT01848639). goal is <140 mmHg and the diastolic BP goal is <90 mmHg.
The differences mostly arise with respect to how much
Transplantation lower should the BP target be and in whom should the BP
be lowered further. Given that CKD itself is a robust risk
In kidney transplant populations, additional contributors factor for adverse CV outcomes, it would be sensible for
to hypertension include primarily the effect of immuno- intensive BP targets to be applied to this population.
suppressive medications, but additionally the effect of the Previously mentioned caveats of identifying patients at
donor: recipient kidney volume mismatch, acute and high risk of adverse events and discussing risks and bene-
chronic rejection, and transplant renal artery stenosis fits and shared decision making, especially those of weigh-
(often at anastomotic site) [143]. The steroid effect on BP is ing outcomes which matter to the individual patients, do
well known, and calcineurin inhibitors (CNIs) also increase apply. The Kidney Disease Improving Global Outcomes
BP by calcium channel mediated vasoconstriction, renal (KDIGO) taskforce is also reviewing the data and will pre-
sodium retention, and in the long run by worsening kidney sumably come to a conclusion that may be acceptable to all
function. Specifically in the kidney transplant population, in the coming months.
higher BP is associated with poorer patient survival, but
also with more graft failure on the basis of observational
data [144, 145].
R
esearch Needs
With respect to treatment of hypertension, no trials
have examined targets of BP lowering in these popula-
The discussion above highlights several lacunae in the evi-
tions and guidelines have most often followed nontrans-
dence for management of hypertension in CKD patients.
plant CKD data. Given the role of immunosuppression,
minimizing steroids and/or the CNI dose can assist in BP ●● What are optimal BP targets in patients with diabetes
lowering, apart from the other nonpharmacological and CKD?
options discussed above. For specific pharmacotherapy, ●● Is there a benefit with intensive BP lowering in patients
there do exist a few RCTs. The first line of agents, i.e. with severe CKD (GFR < 20)?
Reference 347
●● What BP target should be used in PD patients? HD patients? ●● What strategies can be used for improving adherence to
●● Which outcomes are important for patients and what is BP lowering drugs?
the role of shared decision making for intensive BP ●● What are effective strategies for delivering nonpharma-
lowering in CKD? cological therapies for BP lowering?
R
eferences
1 Harlos, J. and Heidland, A. (1994). Hypertension as cause 12 Ruzicka, M., Akbari, A., Bruketa, E. et al. (2016). How
and consequence of renal disease in the 19th century. accurate are home blood pressure devices in use? A
Am. J. Nephrol. 14 (4–6): 436–442. cross-sectional study. PLoS One 11 (6): e0155677.
2 Rose, G. (1985). Sick individuals and sick populations. 13 Ruzicka, M. and Hiremath, S. (2017). Accuracy—limiting
Int. J. Epidemiol. 14 (1): 32–38. factor of home blood pressure monitors? Am. J.
3 Forouzanfar, M.H., Liu, P., Roth, G.A. et al. (2017). Global Hypertens. 30 (7): 661–664.
burden of hypertension and systolic blood pressure of at 14 Agarwal, R., Bills, J.E., Hecht, T.J., and Light, R.P. (2011).
least 110 to 115 mmHg, 1990–2015. JAMA 317 (2): Role of home blood pressure monitoring in overcoming
165–182. therapeutic inertia and improving hypertension control: a
4 Ettehad, D., Emdin, C.A., Kiran, A. et al. (2016). Blood systematic review and meta-analysis. Hypertension 57 (1):
pressure lowering for prevention of cardiovascular 29–38.
disease and death: a systematic review and meta-analysis. 15 Omron, H. (2018). HeartGuide, the highly-anticipated
Lancet 387 (10022): 957–967. first wearable blood pressure monitor from Omron.
5 Melville, S. and Byrd, J.B. (2018). Out-of-office blood https://www.prnewswire.com/news-releases/
pressure monitoring in 2018. JAMA 320 (17): heartguide-the-highly-anticipated-first-wearable-blood-
1805–1806. pressure-monitor-from-omron-available-for-pre-order-
6 Kallioinen, N., Hill, A., Horswill, M.S. et al. (2017). on-december-20th-300769335.html (accessed 22 May
Sources of inaccuracy in the measurement of adult 2021).
patients’ resting blood pressure in clinical settings: a 16 Mena, L., Pintos, S., Queipo, N.V. et al. (2005). A reliable
systematic review. J. Hypertens. 35 (3): 421–441. index for the prognostic significance of blood pressure
7 Myers, M.G., Godwin, M., Dawes, M. et al. (2011). variability. J. Hypertens. 23 (3): 505–511.
Conventional versus automated measurement of blood 17 Chang, T.I., Tabada, G.H., Yang, J. et al. (2016).
pressure in primary care patients with systolic Visit-to-visit variability of blood pressure and death,
hypertension: randomised parallel design controlled trial. end-stage renal disease, and cardiovascular events in
BMJ 342: d286. patients with chronic kidney disease. J. Hypertens. 34
8 Leung, A.A., Nerenberg, K., Daskalopoulou, S.S. et al. (2): 244–252.
(2016). Hypertension Canada’s 2016 Canadian 18 Tanner, R.M., Shimbo, D., Dreisbach, A.W. et al. (2015).
hypertension education program guidelines for blood Association between 24-hour blood pressure variability
pressure measurement, diagnosis, assessment of risk, and chronic kidney disease: a cross-sectional analysis of
prevention, and treatment of hypertension. Can. J. African Americans participating in the Jackson heart
Cardiol. 32 (5): 569–588. study. BMC Nephrol. 16: 84.
9 Lonn, E.M., Bosch, J., Lopez-Jaramillo, P. et al. (2016). 19 Staessen, J.A., Bieniaszewski, L., O’Brien, E. et al. (1997).
Blood-pressure lowering in intermediate-risk persons Nocturnal blood pressure fall on ambulatory monitoring
without cardiovascular disease. N. Engl. J. Med. 374 (21): in a large international database. The “Ad Hoc” working
2009–2020. group. Hypertension 29 (1 Pt 1): 30–39.
10 Wright, J.T. Jr., Williamson, J.D., Whelton, P.K. et al. 20 Yusuf, S., Sleight, P., Pogue, J. et al. (2000). Effects of an
(2015). A randomized trial of intensive versus standard angiotensin-converting-enzyme inhibitor, ramipril, on
blood-pressure control. N. Engl. J. Med. 373 (22): cardiovascular events in high-risk patients. N. Engl. J.
2103–2116. Med. 342 (3): 145–153.
11 Cushman, W.C., Evans, G.W., Byington, R.P. et al. 21 Minutolo, R., Agarwal, R., Borrelli, S. et al. (2011).
(2010). Effects of intensive blood-pressure control in Prognostic role of ambulatory blood pressure
type 2 diabetes mellitus. N. Engl. J. Med. 362 (17): measurement in patients with nondialysis chronic kidney
1575–1585. disease. Arch. Intern. Med. 171 (12): 1090–1098.
348 Hypertension
2 Hermida, R.C., Ayala, D.E., Mojon, A., and Fernandez,

2 management of blood pressure in CKD. Am. J. Kidney
J.R. (2011). Bedtime dosing of antihypertensive Dis. 63 (6): 869–887.
medications reduces cardiovascular risk in CKD. J. Am. 35 Hiremath, S. (2018). Towards better blood pressure: do
Soc. Nephrol. 22 (12): 2313–2321. non-pharmacological strategies provide the right path? J.
23 Rahman, M., Greene, T., Phillips, R.A. et al. (2013). A Clin. Hypertens. (Greenwich) 20 (3): 528–531.
trial of 2 strategies to reduce nocturnal blood pressure in 36 Aburto, N.J., Ziolkovska, A., Hooper, L. et al. (2013).
blacks with chronic kidney disease. Hypertension 61 (1): Effect of lower sodium intake on health: systematic
82–88. review and meta-analyses. BMJ 346: f1326.
24 Liu, L., Cao, Q., Guo, Z., and Dai, Q. (2016). Continuous 37 McMahon, E.J., Bauer, J.D., Hawley, C.M. et al. (2013). A
positive airway pressure in patients with obstructive sleep randomized trial of dietary sodium restriction in CKD. J.
Apnea and resistant hypertension: a meta-analysis of Am. Soc. Nephrol. 24 (12): 2096–2103.
randomized controlled trials. J. Clin. Hypertens. 38 Saran, R., Padilla, R.L., Gillespie, B.W. et al. (2017). A
(Greenwich) 18 (2): 153–158. randomized crossover trial of dietary sodium restriction
25 Yu, J., Zhou, Z., McEvoy, R.D. et al. (2017). Association of in stage 3-4 CKD. Clin. J. Am. Soc. Nephrol. 12 (3):
Positive Airway Pressure with Cardiovascular Events and 399–407.
Death in adults with sleep Apnea: a systematic review 39 Aburto, N.J., Hanson, S., Gutierrez, H. et al. (2013). Effect
and meta-analysis. JAMA 318 (2): 156–166. of increased potassium intake on cardiovascular risk
26 Terracciano, A., Scuteri, A., Strait, J. et al. (2014). Are factors and disease: systematic review and meta-analyses.
personality traits associated with white-coat and masked BMJ 346: f1378.
hypertension? J. Hypertens. 32 (10): 1987–1992; discussion 40 Gritter, M., Vogt, L., Yeung, S.M.H. et al. (2018). Rationale
92. and Design of a Randomized Placebo-Controlled Clinical
27 Gorostidi, M., Sarafidis, P.A., de la Sierra, A. et al. (2013). Trial Assessing the Renoprotective effects of potassium
Differences between office and 24-hour blood pressure supplementation in chronic kidney disease. Nephron 140
control in hypertensive patients with CKD: a (1): 48–57.
5,693-patient cross-sectional analysis from Spain. Am. J. 41 Appel, L.J., Moore, T.J., Obarzanek, E. et al. (1997). A
Kidney Dis. 62 (2): 285–294. clinical trial of the effects of dietary patterns on blood
28 Pogue, V., Rahman, M., Lipkowitz, M. et al. (2009). pressure. DASH collaborative research group. N. Engl. J.
Disparate estimates of hypertension control from Med. 336 (16): 1117–1124.
ambulatory and clinic blood pressure measurements in 42 Tyson, C.C., Lin, P.H., Corsino, L. et al. (2016). Short-term
hypertensive kidney disease. Hypertension 53 (1): 20–27. effects of the DASH diet in adults with moderate chronic
29 Iimuro, S., Imai, E., Watanabe, T. et al. (2013). Clinical kidney disease: a pilot feeding study. Clin. Kidney J. 9 (4):
correlates of ambulatory BP monitoring among patients 592–598.
with CKD. Clin. J. Am. Soc. Nephrol. 8 (5): 721–730. 43 Rebholz, C.M., Crews, D.C., Grams, M.E. et al. (2016).
30 Drawz, P.E., Alper, A.B., Anderson, A.H. et al. (2016). DASH (dietary approaches to stop hypertension) diet and
Masked hypertension and elevated Nighttime blood risk of subsequent kidney disease. Am. J. Kidney Dis. 68
pressure in CKD: prevalence and association with target (6): 853–861.
organ damage. Clin. J. Am. Soc. Nephrol. 11 (4): 642–652. 44 Roerecke, M., Kaczorowski, J., Tobe, S.W. et al. (2017).
31 Agarwal, R., Pappas, M.K., and Sinha, A.D. (2016). The effect of a reduction in alcohol consumption on
Masked uncontrolled hypertension in CKD. J. Am. Soc. blood pressure: a systematic review and meta-analysis.
Nephrol. 27 (3): 924–932. Lancet Public Health 2 (2): e108–e120.
32 Wang, C., Zhang, J., Li, Y. et al. (2017). Masked 45 Dickinson, H.O., Mason, J.M., Nicolson, D.J. et al. (2006).
hypertension, rather than white-coat hypertension, has a Lifestyle interventions to reduce raised blood pressure: a
prognostic role in patients with non-dialysis chronic systematic review of randomized controlled trials. J.
kidney disease. Int. J. Cardiol. 230: 33–39. Hypertens. 24 (2): 215–233.
33 Kushiro, T., Kario, K., Saito, I. et al. (2017). Increased 46 Heiwe, S. and Jacobson, S.H. (2011). Exercise training for
cardiovascular risk of treated white coat and masked adults with chronic kidney disease. Cochrane Database
hypertension in patients with diabetes and chronic Syst. Rev. 10: CD003236.
kidney disease: the HONEST study. Hypertens. Res. 40 (1): 47 Cornelissen, V.A. and Smart, N.A. (2013). Exercise
87–95. training for blood pressure: a systematic review and
34 Ruzicka, M., Quinn, R.R., McFarlane, P. et al. (2014). meta-analysis. J. Am. Heart Assoc. 2 (1): e004473.
Canadian Society of Nephrology commentary on the 48 Naci, H., Salcher-Konrad, M., Dias, S. et al. (2018). How
2012 KDIGO clinical practice guideline for the does exercise treatment compare with antihypertensive
Reference 349
medications? A network meta-analysis of 391 randomised lipid-lowering treatment to prevent heart attack trial
controlled trials assessing exercise and medication effects (ALLHAT). JAMA 288 (23): 2981–2997.
on systolic blood pressure. Br. J. Sports Med. 2019 60 Fried, L.F., Emanuele, N., Zhang, J.H. et al. (2013).
Jul; 53(14):859–869. doi: 10.1136/bjsports-2018-099921. Combined angiotensin inhibition for the treatment of
Epub 2018 Dec 18. diabetic nephropathy. N. Engl. J. Med. 369 (20):
49 Fontana, M., Asaria, P., Moraldo, M. et al. (2014). 1892–1903.
Patient-accessible tool for shared decision making in 61 Yusuf, S., Teo, K.K., Pogue, J. et al. (2008). Telmisartan,
cardiovascular primary prevention: balancing longevity ramipril, or both in patients at high risk for vascular
benefits against medication disutility. Circulation 129 events. N. Engl. J. Med. 358 (15): 1547–1559.
(24): 2539–2546. 62 Parving, H.H., Brenner, B.M., McMurray, J.J. et al. (2012).
50 Ruzicka, M., Hiremath, S., Steiner, S. et al. (2014). What Cardiorenal end points in a trial of aliskiren for type 2
is the feasibility of implementing effective sodium diabetes. N. Engl. J. Med. 367 (23): 2204–2213.
reduction strategies to treat hypertension in primary care 63 Ernst, M.E., Carter, B.L., Goerdt, C.J. et al. (2006).
settings? A systematic review. J. Hypertens. 32 (7): Comparative antihypertensive effects of
1388–1394; discussion 94. hydrochlorothiazide and chlorthalidone on ambulatory
51 Drewnowski, A., Rehm, C.D., Maillot, M. et al. (2015). and office blood pressure. Hypertension 47 (3): 352–358.
The feasibility of meeting the WHO guidelines for 64 Beckett, N.S., Peters, R., Fletcher, A.E. et al. (2008).
sodium and potassium: a cross-national comparison Treatment of hypertension in patients 80 years of age or
study. BMJ Open 5 (3): e006625. older. N. Engl. J. Med. 358 (18): 1887–1898.
52 Rothman, R.L., Housam, R., Weiss, H. et al. (2006). 65 Jamerson, K., Weber, M.A., Bakris, G.L. et al. (2008).
Patient understanding of food labels: the role of literacy Benazepril plus amlodipine or hydrochlorothiazide for
and numeracy. Am. J. Prev. Med. 31 (5): 391–398. hypertension in high-risk patients. N. Engl. J. Med. 359
53 Taylor, D.M., Fraser, S.D.S., Bradley, J.A. et al. (2017). A (23): 2417–2428.
systematic review of the prevalence and associations of 66 Lederle, F.A., Cushman, W.C., Ferguson, R.E. et al.
limited health literacy in CKD. Clin. J. Am. Soc. Nephrol. (2016). Chlorthalidone versus hydrochlorothiazide: a new
12 (7): 1070–1084. kind of veterans affairs cooperative study. Ann. Intern.
54 Mazarova, A., Hiremath, S., Sood, M.M. et al. (2017). Med. 165 (9): 663–664.
Hemodialysis access choice: impact of health literacy. 67 Dhalla, I.A., Gomes, T., Yao, Z. et al. (2013).
Health Lit. Res. Pract. 1 (3): e136–e144. Chlorthalidone versus hydrochlorothiazide for the
55 Ruzicka, M., Burns, K.D., and Hiremath, S. (2017). treatment of hypertension in older adults: a population-
Precision medicine for hypertension management in based cohort study. Ann. Intern. Med. 158 (6): 447–455.
chronic kidney disease: relevance of SPRINT for 68 Agarwal, R., Sinha, A.D., Pappas, M.K., and Ammous, F.
therapeutic targets in non-diabetic renal disease. Can. J. (2014). Chlorthalidone for poorly controlled hypertension
Cardiol. 33 (5): 611–618. in chronic kidney disease: an interventional pilot study.
56 Ruzicka, M. and Hiremath, S. (2017). Intensive blood Am. J. Nephrol. 39 (2): 171–182.
pressure lowering in chronic kidney disease: the time has 69 Williams, B., MacDonald, T.M., Morant, S. et al. (2015).
come. Curr. Opin. Nephrol. Hypertens. 26 (3): 197–204. Spironolactone versus placebo, bisoprolol, and doxazosin
57 ALLHAT AOaCftACRGTAaL-LTtPHAT (2002). Major to determine the optimal treatment for drug-resistant
outcomes in high-risk hypertensive patients randomized hypertension (PATHWAY-2): a randomised, double-blind,
to angiotensin-converting enzyme inhibitor or calcium crossover trial. Lancet 386 (10008): 2059–2068.
channel blocker vs diuretic: the antihypertensive and 70 Heshka, J., Ruzicka, M., Hiremath, S., and McCormick,
lipid-lowering treatment to prevent heart attack trial B.B. (2010). Spironolactone for difficult to control
(ALLHAT). JAMA 288 (23): 2981–2997. hypertension in chronic kidney disease: an analysis of
58 Muntner, P., Anderson, A., Charleston, J. et al. (2010). safety and efficacy. J. Am. Soc. Hypertens. 4 (6): 295–301.
Hypertension awareness, treatment, and control in adults 71 Brown, M.J., Williams, B., Morant, S.V. et al. (2016).
with CKD: results from the chronic renal insufficiency Effect of amiloride, or amiloride plus
cohort (CRIC) study. Am. J. Kidney Dis. 55 (3): 441–451. hydrochlorothiazide, versus hydrochlorothiazide on
59 The ALLHAT Officers and Coordinators for the ALLHAT glucose tolerance and blood pressure (PATHWAY-3): a
Collaborative Research Group (2002). Major outcomes in parallel-group, double-blind randomised phase 4 trial.
high-risk hypertensive patients randomized to Lancet Diabetes Endocrinol. 4 (2): 136–147.
angiotensin-converting enzyme inhibitor or calcium 72 Baker, W.L., Buckley, L.F., Kelly, M.S. et al. (2017). Effects
channel blocker vs diuretic: the antihypertensive and of sodium-glucose Cotransporter 2 inhibitors on 24-hour
350 Hypertension
ambulatory blood pressure: a systematic review and MED): a randomised, sham-controlled, proof-of-concept
meta-analysis. J. Am. Heart Assoc. 6 (5). trial. Lancet 390 (10108): 2160–2170.
73 Zelniker, T.A., Wiviott, S.D., Raz, I. et al. (2019). 85 Logan, A.G., Diaconita, V., Ing, D.J. et al. (2015). Acute
SGLT2 inhibitors for primary and secondary prevention renal failure after renal denervation. J. Vasc. Interv.
of cardiovascular and renal outcomes in type 2 diabetes: a Radiol. 26 (3): 450–451.
systematic review and meta-analysis of cardiovascular 86 Kiuchi, M.G., Mion, D. Jr., Graciano, M.L. et al. (2016).
outcome trials. Lancet 393 (10166): 31–39. Proof of concept study: improvement of
74 McMurray, J.J., Packer, M., Desai, A.S. et al. (2014). echocardiographic parameters after renal sympathetic
Angiotensin-neprilysin inhibition versus enalapril in denervation in CKD refractory hypertensive patients. Int.
heart failure. N. Engl. J. Med. 371 (11): 993–1004. J. Cardiol. 207: 6–12.
75 Haynes, R., Judge, P.K., Staplin, N. et al. (2018). Effects 87 Kiuchi, M.G., Maia, G.L., de Queiroz Carreira, M.A. et al.
of Sacubitril/valsartan versus Irbesartan in patients (2013). Effects of renal denervation with a standard
with chronic kidney disease. Circulation 138 (15): irrigated cardiac ablation catheter on blood pressure and
1505–1514. renal function in patients with chronic kidney disease
76 Oparil, S. and Schmieder, R.E. (2015). New approaches in and resistant hypertension. Eur. Heart J. 34 (28):
the treatment of hypertension. Circ. Res. 116 (6): 2114–2121.
1074–1095. 88 Bisognano, J.D., Bakris, G., Nadim, M.K. et al. (2011).
77 Adson, A.W. and Brown, G.E. (1934). Malignant Baroreflex activation therapy lowers blood pressure in
hypertension: report of case treated by bilateral section of patients with resistant hypertension: results from the
anterior spinal nerve roots from the sixth thoracic to the double-blind, randomized, placebo-controlled rheos
second lumbar, inclusive. J. Am. Med. Assoc. 102 (14): pivotal trial. J. Am. Coll. Cardiol. 58 (7): 765–773.
1115–1118. 89 Lobo, M.D., Ott, C., Sobotka, P.A. et al. (2017). Central
78 Grimson, K.S. (1941). Total thoracic and partial to Total iliac Arteriovenous anastomosis for uncontrolled
lumbar Sympathectomy and celiac Ganglionectomy in hypertension: one-year results from the ROX CONTROL
the treatment of hypertension. Ann. Surg. 114 (4): HTN trial. Hypertension 70 (6): 1099–1105.
753–775. 90 Perry, H.M. Jr., Davis, B.R., Price, T.R. et al. (2000). Effect
79 Esler, M.D., Krum, H., Sobotka, P.A. et al. (2010). Renal of treating isolated systolic hypertension on the risk of
sympathetic denervation in patients with treatment- developing various types and subtypes of stroke: the
resistant hypertension (the Symplicity HTN-2 trial): a systolic hypertension in the elderly program (SHEP).
randomised controlled trial. Lancet 376 (9756): JAMA 284 (4): 465–471.
1903–1909. 91 Wright, J.T. Jr., Bakris, G., Greene, T. et al. (2002). Effect
80 Krum, H., Schlaich, M., Whitbourn, R. et al. (2009). of blood pressure lowering and antihypertensive drug
Catheter-based renal sympathetic denervation for class on progression of hypertensive kidney disease:
resistant hypertension: a multicentre safety and proof-of- results from the AASK trial. JAMA 288 (19): 2421–2431.
principle cohort study. Lancet 373 (9671): 1275–1281. 92 Xie, X., Atkins, E., Lv, J. et al. (2016). Effects of intensive
81 Bhatt, D.L., Kandzari, D.E., O’Neill, W.W. et al. (2014). A blood pressure lowering on cardiovascular and renal
controlled trial of renal denervation for resistant outcomes: updated systematic review and meta-analysis.
hypertension. N. Engl. J. Med. 370 (15): 1393–1401. Lancet 387 (10017): 435–443.
82 Howard, J.P., Shun-Shin, M.J., Hartley, A. et al. (2016). 93 Sundstrom, J., Arima, H., Woodward, M. et al. (2014).
Quantifying the 3 biases that Lead to unintentional Blood pressure-lowering treatment based on
overestimation of the blood pressure-lowering effect of cardiovascular risk: a meta-analysis of individual patient
renal denervation. Circ. Cardiovasc. Qual. Outcomes 9 (1): data. Lancet 384 (9943): 591–598.
14–22. 94 Cheung, A.K., Rahman, M., Reboussin, D.M. et al. (2017).
83 Kandzari, D.E., Bohm, M., Mahfoud, F. et al. (2018). Effects of intensive BP control in CKD. J. Am. Soc.
Effect of renal denervation on blood pressure in the Nephrol. 28 (9): 2812–2813.
presence of antihypertensive drugs: 6-month efficacy and 95 Klahr, S., Levey, A.S., Beck, G.J. et al. (1994). The effects
safety results from the SPYRAL HTN-ON MED proof-of- of dietary protein restriction and blood-pressure control
concept randomised trial. Lancet 391 (10137): 2346–2355. on the progression of chronic renal disease. Modification
84 Townsend, R.R., Mahfoud, F., Kandzari, D.E. et al. (2017). of diet in renal disease study group. N. Engl. J. Med. 330
Catheter-based renal denervation in patients with (13): 877–884.
uncontrolled hypertension in the absence of 96 Ruggenenti, P., Perna, A., Loriga, G. et al. (2005).
antihypertensive medications (SPYRAL HTN-OFF Blood-pressure control for renoprotection in patients
Reference 351
with non-diabetic chronic renal disease (REIN-2): 108 Agarwal, R. (2011). Epidemiology of interdialytic
multicentre, randomised controlled trial. Lancet 365 ambulatory hypertension and the role of volume excess.
(9463): 939–946. Am. J. Nephrol. 34 (4): 381–390.
97 Malhotra, R., Nguyen, H.A., Benavente, O. et al. (2017). 109 Cocchi, R., Degli Esposti, E., Fabbri, A. et al. (1999).
Association between more intensive vs less intensive Prevalence of hypertension in patients on peritoneal
blood pressure lowering and risk of mortality in chronic dialysis: results of an Italian multicentre study. Nephrol.
kidney disease stages 3 to 5: a systematic review and Dial. Transplant. 14 (6): 1536–1540.
meta-analysis. JAMA Intern. Med. 177 (10): 1498–1505. 110 Rodby, R.A., Vonesh, E.F., and Korbet, S.M. (1994).
98 Perkovic, V. and Rodgers, A. (2015). Redefining Blood pressures in hemodialysis and peritoneal dialysis
blood-pressure targets--SPRINT starts the Marathon. N. using ambulatory blood pressure monitoring. Am. J.
Engl. J. Med. 373 (22): 2175–2178. Kidney Dis. 23 (3): 401–411.
99 Gerstein, H.C., Miller, M.E., Byington, R.P. et al. (2008). 111 Liu, Y., Ma, X., Zheng, J. et al. (2017). Effects of
Effects of intensive glucose lowering in type 2 diabetes. angiotensin-converting enzyme inhibitors and angiotensin
N. Engl. J. Med. 358 (24): 2545–2559. receptor blockers on cardiovascular events and residual
100 Tsujimoto, T. and Kajio, H. (2018). Benefits of intensive renal function in dialysis patients: a meta-analysis of
blood pressure treatment in patients with type 2 randomised controlled trials. BMC Nephrol. 18 (1): 206.
diabetes mellitus receiving standard but not intensive 112 Langote, A., Hiremath, S., Ruzicka, M., and McCormick,
glycemic control. Hypertension 72 (2): 323–330. B.B. (2017). Spironolactone is effective in treating
101 Bangalore, S., Kumar, S., Lobach, I., and Messerli, F.H. hypokalemia among peritoneal dialysis patients. PLoS
(2011). Blood pressure targets in subjects with type 2 One 12 (11): e0187269.
diabetes mellitus/impaired fasting glucose: observations 113 Covic, A., Ciumanghel, A.I., Siriopol, D. et al. (2017).
from traditional and bayesian random-effects meta- Value of bioimpedance analysis estimated “dry weight”
analyses of randomized trials. Circulation 123 (24): in maintenance dialysis patients: a systematic review
2799–2810, 9 p following 810. and meta-analysis. Int. Urol. Nephrol. 49 (12):
102 Emdin, C.A., Rahimi, K., Neal, B. et al. (2015). Blood 2231–2245.
pressure lowering in type 2 diabetes: a systematic review 114 Woodrow, G., Oldroyd, B., Stables, G. et al. (2000).
and meta-analysis. JAMA 313 (6): 603–615. Effects of icodextrin in automated peritoneal dialysis on
103 Brunstrom, M. and Carlberg, B. (2016). Effect of blood pressure and bioelectrical impedance analysis.
antihypertensive treatment at different blood pressure Nephrol. Dial. Transplant. 15 (6): 862–866.
levels in patients with diabetes mellitus: systematic 115 Paniagua, R., Ventura, M.D., Avila-Diaz, M. et al. (2009).
review and meta-analyses. BMJ 352: i717. Icodextrin improves metabolic and fluid management in
104 Thomopoulos, C., Parati, G., and Zanchetti, A. (2017). high and high-average transport diabetic patients. Perit.
Effects of blood-pressure-lowering treatment on Dial. Int. 29 (4): 422–432.
outcome incidence in hypertension: 10 -should blood 116 Rutkowski, B., Tam, P., van der Sande, F.M. et al. (2016).
pressure management differ in hypertensive patients Low-sodium versus standard-sodium peritoneal dialysis
with and without diabetes mellitus? Overview and solution in hypertensive patients: a randomized
meta-analyses of randomized trials. J. Hypertens. 35 (5): controlled trial. Am. J. Kidney Dis. 67 (5): 753–761.
922–944. 117 Khandelwal, M. and Oreopoulos, D.G. (2004). Is there a
105 Bhandari, S., Ives, N., Brettell, E.A. et al. (2016). need for low sodium dialysis solution for peritoneal
Multicentre randomized controlled trial of angiotensin- dialysis patients? Adv. Perit. Dial. 20: 156–162.
converting enzyme inhibitor/angiotensin receptor 118 Zager, P.G., Nikolic, J., Brown, R.H. et al. (1998). “U”
blocker withdrawal in advanced renal disease: the curve association of blood pressure and mortality in
STOP-ACEi trial. Nephrol. Dial. Transplant. 31 (2): hemodialysis patients. Medical directors of dialysis
255–261. clinic, Inc. Kidney Int. 54 (2): 561–569.
106 Nesrallah, G.E., Mustafa, R.A., Clark, W.F. et al. (2014). 119 Kalantar-Zadeh, K., Block, G., Humphreys, M.H., and
Canadian Society of Nephrology 2014 clinical practice Kopple, J.D. (2003). Reverse epidemiology of
guideline for timing the initiation of chronic dialysis. cardiovascular risk factors in maintenance dialysis
CMAJ 186 (2): 112–117. patients. Kidney Int. 63 (3): 793–808.
107 Agarwal, R., Nissenson, A.R., Batlle, D. et al. (2003). 120 Agarwal, R. and Sinha, A.D. (2009). Cardiovascular
Prevalence, treatment, and control of hypertension in protection with antihypertensive drugs in dialysis
chronic hemodialysis patients in the United States. Am. patients: systematic review and meta-analysis.
J. Med. 115 (4): 291–297. Hypertension 53 (5): 860–866.
352 Hypertension
21 Agarwal, R. (2010). Blood pressure and mortality among

1 hypertensive haemodialysis patients. Nephrol. Dial.
hemodialysis patients. Hypertension 55 (3): 762–768. Transplant. 23 (11): 3605–3612.
122 Kotanko, P., Garg, A.X., Depner, T. et al. (2015). Effects 134 Tieu, A., Velenosi, T.J., Kucey, A.S. et al. (2018).
of frequent hemodialysis on blood pressure: results from Beta-blocker Dialyzability in maintenance Hemodialysis
the randomized frequent hemodialysis network trials. patients: a randomized clinical trial. Clin. J. Am. Soc.
Hemodial. Int. 19 (3): 386–401. Nephrol. 13 (4): 604–611.
123 Hall, Y.N., Larive, B., Painter, P. et al. (2012). Effects of 135 Agarwal, R., Sinha, A.D., Pappas, M.K. et al. (2014).
six versus three times per week hemodialysis on Hypertension in hemodialysis patients treated with
physical performance, health, and functioning: frequent atenolol or lisinopril: a randomized controlled trial.
hemodialysis network (FHN) randomized trials. Clin. J. Nephrol. Dial. Transplant. 29 (3): 672–681.
Am. Soc. Nephrol. 7 (5): 782–794. 136 Weir, M.A., Dixon, S.N., Fleet, J.L. et al. (2015). Beta-
124 Jardine, M.J., Zuo, L., Gray, N.A. et al. (2017). A trial of blocker dialyzability and mortality in older patients
extending Hemodialysis hours and quality of life. J. Am. receiving hemodialysis. J. Am. Soc. Nephrol. 26 (4):
Soc. Nephrol. 28 (6): 1898–1911. 987–996.
125 Daugirdas, J.T., Greene, T., Rocco, M.V. et al. (2013). 137 Assimon, M.M., Brookhart, M.A., Fine, J.P. et al. (2018).
Effect of frequent hemodialysis on residual kidney A comparative study of Carvedilol versus Metoprolol
function. Kidney Int. 83 (5): 949–958. initiation and 1-year mortality among individuals
126 Suri, R.S., Larive, B., Sherer, S. et al. (2013). Risk of receiving maintenance Hemodialysis. Am. J. Kidney Dis.
vascular access complications with frequent 72 (3): 337–348.
hemodialysis. J. Am. Soc. Nephrol. 24 (3): 498–505. 138 Ruzicka, M., McCormick, B., Leenen, F.H. et al. (2013).
127 de Paula, F.M., Peixoto, A.J., Pinto, L.V. et al. (2004). Adherence to blood pressure-lowering drugs and
Clinical consequences of an individualized dialysate resistant hypertension: should trial of direct observation
sodium prescription in hemodialysis patients. Kidney therapy be part of preassessment for renal denervation?
Int. 66 (3): 1232–1238. Can. J. Cardiol. 29 (12): 1741, e1–3.
128 Munoz Mendoza, J., Bayes, L.Y., Sun, S. et al. (2011). 139 Matsumoto, Y., Mori, Y., Kageyama, S. et al. (2014).
Effect of lowering dialysate sodium concentration on Spironolactone reduces cardiovascular and
interdialytic weight gain and blood pressure in patients cerebrovascular morbidity and mortality in
undergoing thrice-weekly in-center nocturnal hemodialysis patients. J. Am. Coll. Cardiol. 63 (6):
hemodialysis: a quality improvement study. Am. J. 528–536.
Kidney Dis. 58 (6): 956–963. 140 Lin, C., Zhang, Q., Zhang, H., and Lin, A. (2016).
129 Inrig, J.K., Molina, C., D’Silva, K. et al. (2015). Effect of Long-term effects of low-dose spironolactone on
low versus high dialysate sodium concentration on chronic dialysis patients: a randomized placebo-
blood pressure and endothelial-derived vasoregulators controlled study. J. Clin. Hypertens. (Greenwich) 18
during hemodialysis: a randomized crossover study. Am. (2): 121–128.
J. Kidney Dis. 65 (3): 464–473. 141 Walsh, M., Manns, B., Garg, A.X. et al. (2015). The
130 Takahashi, A., Takase, H., Toriyama, T. et al. (2006). safety of Eplerenone in Hemodialysis patients: a
Candesartan, an angiotensin II type-1 receptor blocker, noninferiority randomized controlled trial. Clin. J. Am.
reduces cardiovascular events in patients on chronic Soc. Nephrol. 10 (9): 1602–1608.
haemodialysis--a randomized study. Nephrol. Dial. 142 Quach, K., Lvtvyn, L., Baigent, C. et al. (2016). The
Transplant. 21 (9): 2507–2512. safety and efficacy of mineralocorticoid receptor
131 Suzuki, H., Kanno, Y., Sugahara, S. et al. (2008). Effect antagonists in patients who require dialysis: a
of angiotensin receptor blockers on cardiovascular systematic review and meta-analysis. Am. J. Kidney Dis.
events in patients undergoing hemodialysis: an 68 (4): 591–598.
open-label randomized controlled trial. Am. J. Kidney 143 Weir, M.R., Burgess, E.D., Cooper, J.E. et al. (2015).
Dis. 52 (3): 501–506. Assessment and management of hypertension in
132 Iseki, K., Arima, H., Kohagura, K. et al. (2013). Effects transplant patients. J. Am. Soc. Nephrol. 26 (6):
of angiotensin receptor blockade (ARB) on mortality 1248–1260.
and cardiovascular outcomes in patients with long-term 144 Kasiske, B.L., Anjum, S., Shah, R. et al. (2004).
haemodialysis: a randomized controlled trial. Nephrol. Hypertension after kidney transplantation. Am. J.
Dial. Transplant. 28 (6): 1579–1589. Kidney Dis. 43 (6): 1071–1081.
133 Tepel, M., Hopfenmueller, W., Scholze, A. et al. (2008). 145 Opelz, G., Wujciak, T., and Ritz, E. (1998). Association
Effect of amlodipine on cardiovascular events in of chronic kidney graft failure with recipient blood
Reference 353
pressure. Collaborative transplant study. Kidney Int. 53 the task force for the management of arterial
(1): 217–222. hypertension of the European Society of Cardiology and
146 Hiremath, S., Fergusson, D., Doucette, S. et al. (2007). the European Society of Hypertension. J. Hypertens. 36
Renin angiotensin system blockade in kidney (10): 1953–2041.
transplantation: a systematic review of the evidence. 153 Whelton, P.K., Carey, R.M., Aronow, W.S. et al. (2018).
Am. J. Transplant. 7 (10): 2350–2360. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/
147 Moes, A.D., Hesselink, D.A., van den Meiracker, A.H. ASPC/NMA/PCNA guideline for the prevention,
et al. (2017). Chlorthalidone versus amlodipine for detection, Evaluation, and Management of High Blood
hypertension in kidney transplant recipients treated Pressure in adults: a report of the American College of
with Tacrolimus: a randomized crossover trial. Am. J. Cardiology/American Heart Association task force on
Kidney Dis. 69 (6): 796–804. clinical practice guidelines. Hypertension 71 (6):
148 Cross, N.B., Webster, A.C., Masson, P. et al. (2009). e13–e115.
Antihypertensive treatment for kidney transplant 154 Nerenberg, K.A., Zarnke, K.B., Leung, A.A. et al. (2018).
recipients. Cochrane Database Syst. Rev. 3: CD003598. Hypertension Canada’s 2018 guidelines for diagnosis,
149 Hiremath, S., Fergusson, D.A., Fergusson, N. et al. risk assessment, prevention, and treatment of
(2017). Renin-angiotensin system blockade and hypertension in adults and children. Can. J. Cardiol. 34
long-term clinical outcomes in kidney transplant (5): 506–525.
recipients: a meta-analysis of randomized controlled 155 Iimuro, S., Imai, E., Watanabe, T. et al. (2013). Chronic
trials. Am. J. Kidney Dis. 69 (1): 78–86. Kidney Disease Japan Cohort Study Group. Clin. J. Am.
150 Ibrahim, H.N., Jackson, S., Connaire, J. et al. (2013). Soc. Nephrol. 8 (5): 721–730.
Angiotensin II blockade in kidney transplant recipients. 156 Minutolo, R., Gabbai, F.B., Agarwal, R. et al. (2014).
J. Am. Soc. Nephrol. 24 (2): 320–327. Assessment of achieved clinic and ambulatory blood
151 Philipp, T., Martinez, F., Geiger, H. et al. (2010). pressure recordings and outcomes during treatment in
Candesartan improves blood pressure control and hypertensive patients with CKD: a multicenter
reduces proteinuria in renal transplant recipients: prospective cohort study. Am. J. Kidney Dis. 64 (5):
results from SECRET. Nephrol. Dial. Transplant. 25 (3): 744–752.
967–976. 157 Scheppach, J.B., Raff, U., Toncar, S. et al. (2018). Blood
152 Williams, B., Mancia, G., Spiering, W. et al. (2018). 2018 pressure pattern and target organ damage in patients
ESC/ESH guidelines for the management of arterial with chronic kidney disease. Hypertension. 72:
hypertension: the task force for the management of 929–936.
arterial hypertension of the European Society of 158 Inrig et al. (2015). Am J Kidney Dis. 2015 Mar; 65(3):464–73.
Cardiology and the European Society of Hypertension: doi: 10.1053/j.ajkd.2014.10.021. Epub 2014 Dec 17.
354
23
Renovascular Disease
Jörg Radermacher
Department of Nephrology, Johannes Wesling Klinikum Minden, UK-RUB, Minden, Germany
I ntroduction renoparenchymatous disease, technical failure, choles-

terol embolism, or radiocontrast toxicity. These aspects
End-stage renal disease requiring renal replacement will be discussed.
therapy puts a major economic burden on the healthcare
system. Atherosclerotic renovascular disease has been Terminology
found in about 12–31% of patients >45 years old with end-
The term renovascular hypertension is reserved for those
stage renal disease [1, 2], constitutes the fastest-growing
patients with improvement, or rarely even normalization, of
group of these patients [3], and is associated with a mean
blood pressure after intervention. The term renovascular
survival time of 25–34 months [4]. In unselected patients
azotemia is used for those patients who have improved renal
with hypertension, the prevalence of renal artery stenosis
function. Frequently, the term ischemic nephropathy is
(RAS) is only about 1–5% [5]. In certain patient popula-
used, but this term does not distinguish whether a patient is
tions, such as those with severe hypertension, refractory
going to be a responder. RAS and renovascular hypertension
hypertension, aortic aneurysm [6], or coronary artery
are frequently used as synonyms in the literature [17].
disease [7], the prevalence is as high as 20–40% [8, 9].
However, the term RAS should be reserved for an anatomi-
Patients with RAS of >50–70% renal artery diameter were
cal description, whereas renovascular hypertension and
frequently advised to undergo correction of the stenosis. The
renovascular azotemia describe the functional relevance of
reason for treatment usually was difficult-to-control hyper-
the stenosis.
tension. However, nowadays kidney failure should also be
present, since patients with both hypertension and impaired
Pathophysiology of RAS
renal function benefit the most from angioplasty. Identifying
patients whose blood pressure and/or renal function will RAS leads to renal ischemia, causing the release of renin
improve postintervention remains difficult. Retrospective from the juxtaglomerular cells of the kidney and resulting
studies have shown a lack of improvement in blood pressure in the conversion of angiotensin I to angiotensin II, subse-
and renal function despite correction of the stenosis in quent release of aldosterone from the adrenal gland, vaso-
20–46% of patients [10–16]. constriction, and sodium and water retention. The
Because angioplasty and surgery can lead to complica- long-term consequences are the buildup of extracellular
tions like cholesterol embolism, persistent acute kidney matrix and collagen IV via angiotensin II-induced increased
failure, and even patient death, these procedures should expression of transforming growth factor β and interstitial
only be performed in patients with a high likelihood of platelet-derived growth factor, resulting in irreversible
benefit. Possible reasons for a lack of treatment effect are parenchymal damage [18–20] also in the nonaffected kid-
correction of trivial stenoses (<60%), lack of a clinical ney. The challenge of treatment of atherosclerotic RAS is to
problem (not difficult to control hypertension, not intervene at a stage before major parenchymal damage has
severely impaired or declining renal function) underlying occurred.
Introductio 355
Should we still Diagnose and Correct RAS? This shows that correction of RAS may still be consid-
ered in patients with RAS and a relevant clinical
At present there are 10 randomized controlled trials [15,
problem.
21–29] (Table 23.1), a metaanalysis [31], and a Cochrane
A recent S2K guideline on diseases of the renal artery
analysis [30], none of which has shown a benefit of angio-
(https://www.awmf.org/uploads/tx_szleitlinien/004-008l_
plasty compared to optimal medical treatment (OMT)
S2k_Erkrankungen-der-Nierenarterien_2020-01_01.pdf)
alone. Some experts suggest we should stop looking for the
describes case scenarios where angioplasty (usually stent
presence of RAS because a benefit of correction of RAS has
assisted) should be avoided or can still be considered
not been shown.
(Table 23.2) [32].
However, relevant patient groups have not been included
in those studies. The second-largest ASTRAL trial [25]
only included patients for whom the treating doctor was Patients who should be Screened for RAS
unsure whether angioplasty would be of help. The largest
Because RAS is present in only about 1% of unselected
and most recent CORAL trial [27] included mostly patients
patients with hypertension, general screening is not advis-
without clinical problems (no severe hypertension, no
able. The clinical signs suggesting renovascular disease
declining renal function). Even if angioplasty were always
include an abdominal bruit, difficult-to-control hyperten-
futile – which it is not as will be shown – the diagnosis of
sion (requiring three or more antihypertensive agents),
RAS would still have to be made. All recent studies found
accelerated hypertension or hypertension that was previ-
a 50% less than expected event rate in the control group
ously well-controlled, worsening of renal function
and attributed that to the OMT consisting of ASS, statins,
4–31 days after introduction of angiotensin converting
and antihypertensives, including a RAS-inhibitor. Treating
enzyme inhibitor (ACEi) or angiontensin II receptor
patients with OMT is class 1A advice in all guidelines and
blocker treatment [10], severe atherosclerosis in other vas-
not looking for the presence of RAS would deny patients
cular beds, otherwise unexplained chronic kidney disease,
this treatment.
hypertension associated with sudden and repeated left
The Salford group included 72 patients in the ASTRAL
heart failure or pulmonary edema, the onset of hyperten-
trial and reported their outcome on a further 467 patients
sion before the age of 30 years in nonobese and 15 years in
with >50% RAS who were not included in ASTRAL [33]. all patients (from fibromuscular renal artery disease) or
Thirty-seven patients had flash pulmonary edema but only after the age of 55 years (from atherosclerotic renal artery
66% received angioplasty. The hazard ratio for death in the disease), and differences in the sizes of the two kidneys. It
angioplasty group was 0.4 (0.2–0.9). Thirty-one patients is important to identify these clinical clues because the
had refractory hypertension and declining renal function. prevalence of RAS increases to 39% among these
The hazard ratio for death in those patients receiving angi- patients [40]. Once RAS is suspected, a screening test is
oplasty was 0.15 (0.02–0.9) and the HR for a cardiovascular used to confirm its presence.
event was 0.23 (0.1–0.6). Vassallo et al. showed similar
findings in 131 patients with 70% stenosis and clinical Screening Methods for RAS
problems compared to 144 patients with Screening tests should be inexpensive, accurate (i.e. with a
the same degree of stenosis but without the clinical low rate of technical failure and high sensitivity and
problem [34]. Median follow-up time was 58 months. specificity), and noninvasive. This excludes angiography,
Revascularitzation was associated with a 50% reduced risk although angiography is still considered the reference
of progression to end-stage kidney disease and cardiovas- standard for detection and quantification of stenosis. A
cular events in patients with rapidly deteriorating renal high interobserver variability for stenosis quantification
function. Furthermore, patients with bilateral RAS and has been shown (κ < 0.4) [41]. One study found a systolic
patients with proteinuria <1 g also had significantly better pressure gradient of 30 mmHg in 13 of 22 renal arteries in
renal and cardiovascular outcomes post revascularization which arteriography had shown only minor degrees of
compared to controls. stenoses (<50%) [42], a pressure gradient generally associ-
Grey et al. reported findings on 39/207 patients with ated with a reduction of the glomerular filtration rate
flash pulmonary edema or recurrent episodes of congestive (GFR) [43]. Angiography may also cause atheromatous
heart failure. After angioplasty blood pressure and renal embolization of the kidneys or renal impairment due to
function improved significantly, and hospitalization for radiocontrast nephrotoxicity.
heart failure was reduced from 2.4 to 0.3 episodes per year Spiral computed tomography angiography (CTA) and
before and after angioplasty [35]. magnetic resonance angiography are noninvasive imaging
Table 23.1 Randomized controlled trials angioplasty vs. no angioplasty.
Follow up Blood pressure Antihypertensive Renal Cardiovascular

Study n Inclusion Exclusion OMT (months) improvement drug classes function disease SE-PTRA Mortality
EMMA 23/26 Stenosis 60–75% Bilat. sten. No 6 Ø (24 h–RR) −0.8 (P = 0.009) Ø NI RR 3.4 Ø
1998 [21] Single kidney
Webster 25/30 Stenosis >50% RASi No 6 Ø (office BP) NI NI NI ↑ NI
1998 [22]
DRASTIC 56/50 Stenosis [30] 50% Creatinine No 3 Ø (office BP) −1.0 (P < 0.001) Ø NI Ø Ø
2000 [15] >2.3 12 −0.6 (P = 0.10)
STAR 64/74 Stenosis [30] 50%; GFR < 15 Yes 24 Ø (office BP) Ø Ø Ø ↑ Ø
2009 [23] GFR < 80 PU >3 g/Tag
BP < 140/90
NITER 28/24 Stenosis >70% Yes 43 Ø (office BP) Ø Ø Ø ↑ Ø
2009 [24] GFR 30-60
ASTRAL 403/403 Stenosis >50% Yes 33.6 Ø (office BP) −0.22 Ø Ø ↑ Ø
2009 [25] Doctor uncertain (P = 0.03)
RASCAD 43/41 Stenosis: 50–80% Yes 12 Ø (office BP) Ø Ø Ø (LVH) Ø Ø
2012 [26] KHK
CORAL [27] 459/472 Stenosis >80% oder Creatinine Yes + 43 −2 mmHg syst Ø Ø Ø Ø
>60% and >4 mg/dl lifestyle (P = 0.03)
PW > 20 mmHg syst
RADAR [28] 34/33 Stenosis >70% Yes 36 Ø 24 h-RR Ø Ø
METRAS [29] ~60 >70% oder <70% with Yes ? Szinti
poststenotic dilatation
n, number of angioplasty + OMT/OMT without angioplasty; Ø, no effect; OMT, optimal medical treatment (AHT (antihypertensive treatment) including RAS inhibitor, ASS, (Acetyl salicylic Acid)
statin); NI, no information; GFR, glomerular filtration rate; PU, proteinuria; gray, primary endpoint; CVD, cardiovascular disease; SE-PTRA, side effect of angioplasty; PW, pressure wire.
Sources: Bavry et al. [31], Jenks et al. [30] and Oberhuber et al. [32].
0005152411.INDD 356 09-12-2022 15:41:47

Introductio 357
Table 23.2 Treatment of atherosclerotic renovascular disease [30–32].
No angioplasty No angioplasty in unilateral or bilateral asymptomatic stenosis High Medium [32]

No angioplasty No angioplasty in patients with proteinuria >1 g/day, biopsy Moderate Weak [32]
proven severe renal disease or resistive index >0.80
Angioplasty Angioplasty can be considered in patients with both refractory Low Weak [32]
hypertension and rapidly declining renal function (also with
unilateral stenosis)
Angioplasty Angioplasty can be considered in patients with flash pulmonary Low Medium [32, 35, 36]
edema or recurring cardiac decompensation even in unilateral
stenosis
Nicotin Nicotine use must be stopped High Strong [32]
Statin All patients with RAS should be treated with a statin to a goal of Moderate Medium [32, 37, 38]
LDL < 70 mg/dl
Diet Follow a healthy diet (salt restriction, moderation of alcohol Moderate Medium [32, 36]
consumption, high consumption of vegetables and fruits) and
reduce weight and perform regular physical exercise
Goal blood pressure Blood pressure should be lowered to <140/90 mmHg High Strong [32]
RAS inhibitor A RAS inhibitor should be given to all patients with unilateral RAS Moderate Medium [32, 39]
RAS inhibitor A RAS inhibitor can be given to patients with bilateral RAS with Low Weak [32]
tight control of s-creatinine and potassium
Aspirin Acetylsalicylic Acid should be given High Strong [32, 36]
HbA1c In patients with RAS and diabetes HbA1c should be 7.5% Moderate Medium [32]
techniques that have high sensitivity and specificity (>95%) and acceptable specificity, and has also been shown to be of
for detecting RAS [44–47], although a more recent study value in identifying patients whose blood pressure will
did not find the same level of accuracy when these meth- improve after correcting the stenotic lesion [49]. This test,
ods were applied in everyday practice [47]. The latter study however, has not been shown to predict an improvement in
reported a sensitivity of 64% and a specificity of 92% for renal function after correction of RAS and it cannot locate
CTA and 62% and 84%, respectively, for magnetic reso- the stenosis or determine its severity [44]. Furthermore, the
nance angiography. These techniques are also limited by sensitivity of this test is reduced to 80% in patients with renal
their high costs and, in the case of spiral CT, the use of insufficiency (GFR of <50 ml/min) or in patients with bilat-
contrast agents. Magnetic resonance is not suitable for eral stenoses or a stenosis in a single functioning kidney [44,
patients with claustrophobia and certain types of metallic 50]. It is particularly important to identify significant sten-
implants. Measurement of the concentrations of renin in oses in these patients because the major rationale for per-
the renal veins have been used to predict the potential suc- forming surgery or angioplasty is to preserve renal function.
cess of surgical revascularization. False-negative and false- In experienced hands Doppler ultrasonography is highly
positive results are common with this technique, and it is sensitive and specific for detecting RAS, and it is rapid and
therefore not recommended as a reliable screening test for inexpensive [51]. Two main approaches are used to detect a
RAS. In theory, the accuracy of renal vein renins is significant RAS of 50–70% [51]. The first (direct) approach
enhanced by using an ACEi (captopril test), which attenu- looks at flow acceleration at the site of the stenosis. This
ates the vasoconstrictive effect of angiotensin II on the approach has good sensitivity and specificity for detecting
efferent arteriole and reduces filtration on the side of the stenoses of 50%. Obesity, excessive bowel gas, or poor
stenosis [44]. The reported sensitivity and specificity after blood flow in the main renal artery can, however, interfere
25–50 mg of captopril, however, are also low [48]. with direct visualization of the renal arteries [51]. The sec-
Captopril scintigraphy (25–50 mg of oral captopril given ond (indirect) approach looks at poststenotic flow phe-
12 hours before isotope) can detect RAS with high sensitivity nomena (tardus and parvus). This approach can be used in
358 Renovascular Disease
nearly all patients but will only detect severe stenoses of Exclusion of Renoparenchymatous Disease
>60%. A combination of both approaches is the most suit- Possibly the main reason for a lack of treatment benefit
able technique for accurate detection of RAS in almost all despite successful correction of RAS is preexisting reno-
patients. In patients with normal or impaired renal func- parenchymatous disease. The two most frequent diseases
tion, we reported no technical failure with this combina- coexisting with or even causing RAS are long-term hyper-
tion method and showed high sensitivity (96.7%) and tension and diabetes leading to hypertensive nephroscle-
specificity (98.0%) for detecting RAS of 50% compared rosis and diabetic glomerulosclerosis [71]. Renal biopsy
with angiography [51]. The value of Doppler ultrasonogra- can predict treatment failure but is usually not indicated
phy to reliably detect stenoses of 50% and more has been in these disease entities [72]. Some readily available clini-
shown in the majority of more recent studies (Table 23.3) cal and angiographic clues as predictors of treatment suc-
and screening for RAS with Doppler ultrasound is favored cess have been suggested [73, 74] (Table 23.4). However,
by most guidelines [36]. none of these parameters is sufficiently sensitive or spe-
In summary, the best screening test is probably the test cific, and contradicting data have been published for
performed most frequently at an individual facility. In many of these. Rapidly deteriorating renal function in the
patients with a GFR of <50 ml/min, CTA (due to radio- presence of RAS has been associated with a favorable
contrast toxicity) and captopril scintigraphy (due to low response after angioplasty [81]; however, follow-up data
sensitivity) should be avoided. on renal function are frequently not available. Some
screening methods have been evaluated regarding predic-
tion of treatment success: measurement of renal vein
Quantification of Stenosis and Estimation renin, captopril scintigraphy, magnetic resonance tomog-
of Functional Relevance raphy, and Doppler ultrasonography. Because a hemody-
Other than detecting the presence of stenosis, Doppler namically relevant RAS should cause increased renin
ultrasonography can also be used to estimate the severity of production, renal vein renin measurements have been
RAS with reliable estimates up to 70% diameter reduction. considered the reference standard for predicting the func-
A good correlation compared to intravascular ultrasound tional relevance of stenoses. The diagnostic accuracy of
has been shown (R = 0.97) [65]. An RI difference >0.05 this test is, however, disappointing, and the requirement
(parvus phenomenon) suggests stenosis of at least 60% and for invasive venous angiography and radiocontrast agents
an acceleration time >70 ms (tardus phenomenon) sug- have made this method almost obsolete. Renal scintigra-
gests a stenosis >65% [66]. phy without captopril also has only low diagnostic accu-
There is general agreement [67] that a diameter stenosis racy, but captopril scintigraphy is an established method
of <50% causes neither high blood pressure nor impair- to predict a treatment effect. Captopril causes a further
ment of renal function. For this reason, intervention should fall in GFR in the stenosed kidney. The positive predictive
not be performed for these low-degree stenoses. Stenoses value for blood pressure improvement of a positive capto-
<60% progress slowly and almost never proceed directly to pril scintigraphy is reported to be 92%; however, sensitiv-
occlusion [68]. A policy of watchful waiting (e.g. ultrasono- ity drops to 80% in patients with impaired renal
graphic follow-up of stenoses) is reasonable. Stenoses of function [50]. This could be due to dependence of renal
>65–80% are considered to be hemodynamically rele- filtration on vasoconstriction of the efferent arteriole in
vant [69, 70]. Stenoses in most studies have been graded by patients with advanced renoparenchymatous disease,
angiography. Simon et al. found hypersecretion of renin, which is impaired after ACEi treatment. A further draw-
suggesting functional relevance, in stenoses of >80% diam- back of captopril scintigraphy is the requirement for
eter reduction [70]. Unfortunately, data on the improve- meticulous patient preparation in order to obtain reliable
ment of blood pressure or renal function after correction of results. Patient preparation includes controlled hydra-
stenosis are lacking. Giroux retrospectively evaluated the tion, controlled sodium diet, and removal of diuretics and
degree of stenosis in patients who did or did not have ACEi 4–14 days before the investigation [50, 82]. van
improved blood pressure after correction of RAS. They Jaarsveld et al. did not find a predictive value of captopril
found tighter stenoses in patients that did benefit (79 ± 9% scintigraphy for improvement of blood pressure in their
vs. 74 ± 9%, P < 0.05), but the differences were small [69]. study [15].
Future studies should quantify stenoses using noninvasive Gadolinium-assisted magnetic resonance imaging after
Doppler ultrasound, intravascular ultrasound, or a pres- ACEi treatment has been used to quantify renal filtration
sure gradient measured by pressure wire (an expensive and has provided predictive results comparable to captopril
method) or catheter. scintigraphy [83]. The study protocol was later refined [84].
Table 23.3 Stenosis criteria and sensitivity and specificity for detection of RAS with Doppler ultrasound compared to selective angiographya.
Criterion and study [reference] No. of patients Criterion % technical failureb Degree (%) of stenosisc % sensitivity/specificity
Direct stenosis criteria

Hansen et al. 1990 [52] 74 RAR > 3.5 8 60 93/98
Karasch et al. 1993 [53] 53 Vmax > 180 cm/s 15 50 92/92
Olin et al. 1995 [5] 102 Vmax > 200 cm/s or RAR > 3.5 10 60 98/98
Postma et al. 1992 [54] 61 Doppler freq. >4 kHz and broadened 25 50 63/86
Doppler spectrum
Schäberle et al. 1992 [55] 76 Vmax > 140 cm/s NA 50 86/83
AbuRahma et al. 2012 [56] 313 Vmax > 285 cm/s 4 60 67/90
RAR > 3.7 69/91
Indirect stenosis criteria (paravus, tardus)
Baxter et al. 1996 [57] 73 AT > 70 ms 16 70 89/97
Kliewer et al. 1993 [58] 57 AT 70 ms 0 50 82/20
Riehl et al. 1997 [59] 214 RI < 0.45 or ΔRI 8% 0 70 93/96
Schwerk et al. 1994 [60] 72 ΔRI 5% 0 50 82/92
0 60 100/94
Speckamp et al. 1995 [61] 123 ΔAI 80% NA 70 100/94
Stavros et al. 1992 [62] 56 Loss of ESP 0 60 95/97
Strunk et al. 1995 [63] 50 AT 70 ms 4 50 77/46
Combination of direct and indirect stenosis
criteria
Krumme et al. 1996 [64] 135 Vmax > 180 cm/s and/or ΔRI 5% 0 50 89/92
Radermacher et al. 2000 [51] 226 Vmax > 180 cm/s and RRR > 4 and/or 0 50 97/98
AT 70 ms
AI, acceleration index; AT, acceleration time; ESP, early systolic peak; RAR, renal aortic ratio; RI, resistance index (Pourcelot index); RRR, renal renal ratio.
a
Only prospective studies with more than 50 patients and comparison with intra-arterial angiography as the gold standard were considered.
b
Technical failure: the renal artery or intrarenal arteries could not be visualized by color Doppler ultrasonography.
c
The lowest degree of stenosis (% diameter stenosis) detected by the respective ultrasound method. NA, data not available.
0005152411.INDD 359 09-12-2022 15:41:47

Table 23.4 Predictors of a lack of effect of successful correction Ostial stenoses should always be treated with angioplasty
of RAS related to blood pressure and/or renal function. plus stenting; stenoses located >1 cm distally to the aorta or
stenoses due to fibromuscular dysplasia can be treated with
Studies [reference] that angioplasty alone [88]. A randomized controlled study
found the parameter to be:
reported better initial patency rates (88% vs. 57%) and lower
restenosis rates (14% vs. 48%) 6 months after stent-assisted
Not angioplasty compared to angioplasty alone in ostial sten-
Patient characteristic Predictive predictive
oses [89]. However, the same study failed to show superiority
Advanced age, >65 years [15, 16, 75, 76] [69, 77, 78] of stenting regarding blood pressure or renal function
Male gender [75] [69, 77, 78] improvement. Newer studies report primary patency rates in
the range from 94 to 100% [69, 77, 90], but long-term results
Severe atherosclerosis [16, 69, 78]
are less favorable. Restenosis, de novo stenosis, or thrombosis
Proteinuria >1 g/day [16, 79]
rates in the range of 10% per year have been reported both in
Severely impaired renal [16, 78] [69, 76, 77]
stent-treated patients and patients who had angioplasty
function (GFR <40 ml/min)
alone [16, 69, 77, 81, 90–93] Revascularization success should
No sudden appearance of [15] [16]
hypertension or sudden
therefore be monitored by ultrasonography at 3 months,
worsening of previously 6 months, and yearly intervals thereafter. Newer technologies
well-controlled hypertension using sirolimus-coated stents are being evaluated [94].
Duration of hypertension [69, 75] [76]
>10 years
Diastolic blood pressure [69, 75, 76] [77] holesterol Embolism
C
<80 mmHg and Radiocontrast Toxicity
Systolic blood pressure [77] [76]
<160 mmHg Cholesterol embolism in the renal arterial bed or the aorta is
Diabetes mellitus [69] [16, 78] always a dramatic event. It can lead to progressive loss of
Nonsmoker [16, 69] renal function despite patency of the renal artery. At present,
Degree of stenosis <70% [69] [15, 77] there is no reliable method to detect cholesterol embolism.
Resistance index >0.80 [16] [80] Renal biopsy is invasive and may miss the site of embolism
due to sampling error. Blood eosinophilia, hypocomple-
mentemia or the “blue toe” suggest cholesterol embolism
Prospective studies on this fascinating technique as a but are not sufficiently sensitive or specific to allow the diag-
predictor for renovascular hypertension or azotemia are nosis of renal cholesterol embolism. Dejani et al. found
lacking at present, however. blood eosinophilia of >5% in seven of 20 patients older than
In two studies with low patient numbers, an increased 55 years with impaired renal function (serum creatinine
intrarenal resistance index value measured by Doppler >2 mg/dl) treated by renal angioplasty [95]. The feasibility of
ultrasonography was associated with a rapid loss of renal renal protection devices catching cholesterol crystals and
function [85] and a lack of blood pressure improve- other debris has been shown but has not been tested in ran-
ment [86], despite correction of RAS. A prospective study domized controlled trials [96].
found that a resistance index of >0.8 in the intrarenal seg- A further frequent complication associated with renal
mental arteries had a sensitivity and specificity of >90% for angiography and renal arterial stenting is radiocontrast
predicting renal function deterioration and a lack of blood toxicity. Treatment of RAS may be of greatest benefit in
pressure improvement [16] (Figure 23.1). patients with impaired renal function, and these patients are
most sensitive to radiocontrast toxicity. Controlled hydration
Improving Short-and Long-term Results with NaCl at 1 ml/kg/h for 12 hours prior to and after angio-
of Correction of RAS plasty is considered standard prophylactic therapy. A rand-
Ten-year patency rates are still only available for surgically omized controlled trial has shown the superiority of this
corrected renal artery stenoses, although outcomes may be regimen, compared with additional mannitol or furosemide
better after surgery compared to angioplasty [87]. However, treatment [97]. Prehydration with sodium bicarbonate
angioplasty has become the method of choice because it is (154 mmol plus 900 ml 5% glucose solution) instead of saline
less invasive and has a lower mortality rate. Only when an alone may further lessen the risk. However, these data have not
aortic aneurysm or other severe aortic disease requires simul- been confirmed in subsequent studies [98]. Acetylcysteine
taneous treatment should surgery be the preferred method. (600–1200 mg twice a day) has been shown to have an additive
Reference 361
RI > = 80
negative captopril scintigraphy
proteinuria > 1 g/day
GFR < 40 ml/min
CHD, AOD, carotid stenosis
age > 65 Jahre
pulse pressure > 70 mmHg
uric acid > 430 μmol/l
Diabetes mellitus
male sex
renal length < 9 cm
Renin < 5,7 ng/ml/h
No sudden increase in blood pressure
No nocturnal blood pressure drop
Smoking
0,1 1 10 100 1000 10000 100000
Odds ratio
Figure 23.1 Odds ratios for various factors to predict deterioration of renal function. Squares depict odds ratios; lines depict 95%
confidence intervals. RI, resistance index; CHD, coronary heart disease; AOD, atheroocclusive disease of the legs.
protective effect compared with hydration alone for preser- Conclusions

vation of renal function and prevention of acute kidney fail-
ure in several randomized controlled trials [99, 100], but a Patients should only be screened for the presence of RAS
recent meta-analysis including 19 randomized trials failed to when the likelihood that they have RAS is sufficiently high.
show such a protective effect [101]. A good screening method is Doppler ultrasonography. If
The newer nonionic radiocontrast agents have lessened local experience favors CT or MRI these methods can also
the risk of toxicity but have not abolished it [102]. be applied. Only patients with a high likelihood for
Furthermore, the isoosmolar radiocontrast agent iodixanol improvement should be treated with angioplasty.
has been associated with significantly less contrast Randomized trials have failed to show a significant superi-
nephropathy than the low-osmolar agent iohexol [103]. ority of angioplasty compared to OMT alone. However,
These results have not been confirmed in other studies. these studies included patients because they had anatomi-
Hemodialysis during or directly after application of radio- cal RAS and not because they had RAS and a significant
contrast agents does not prevent radiocontrast toxicity, but associated clinical problem. Future technical improve-
hemofiltration may have a beneficial effect [104]. Here ments regarding prevention of cholesterol embolism and
also, confirmatory studies have not been performed, and radiocontrast toxicity may increase the proportion of
this method of treatment is invasive and expensive. patients who benefit from stenosis correction. Based on
Calcium channel blockers, ANP [105], dopamine [106], current data, stenoses of >60% can be treated with angio-
and endothelin receptor antagonist [107] have been proven plasty if there is both uncontrolled hypertension and
useless to prevent radiocontrast toxicity. Theophylline may declining renal function, or if the patient suffers from flash
have a protective effect, but its effects are not additive com- pulmonary edema or recurring congestive heart failure.
pared to hydration treatment alone [108].
References
van Ampting, J.M., Penne, E.L., Beek, F.J. et al. (2003).

1 incidence, clinical correlates, and outcomes: a 20-year
Prevalence of atherosclerotic renal artery stenosis in clinical experience. Am. J. Kidney Dis. 24: 622–629.
patients starting dialysis. Nephrol. Dial. Transplant. 18: Fatica, R.A., Port, F.K., and Young, E.W. (2001). Incidence
3
1147–1151. trends and mortality in end-stage renal disease attributed
2 Mailloux, L.U., Napolitano, B., Bellucci, A.G. et al. (1994). to renovascular disease in the United States. Am. J. Kidney
Renal vascular disease causing end-stage renal disease, Dis. 37: 1184–1190.
4 Safian, R.D. and Textor, S.C. (2001). Renal-artery stenosis. 18 Meyrier, A., Hill, G.S., and Simon, P. (1998). Ischemic
N. Engl. J. Med. 344: 431–442. renal diseases: new insights into old entities. Kidney Int.
5 Olin, J.W., Piedmonte, M.R., Young, J.R. et al. (1995). The 54: 2–13.
utility of duplex ultrasound scanning of the renal arteries 19 Hilgers, K.F., Hartner, A., Porst, M. et al. (2000).
for diagnosing significant renal artery stenosis. Ann. Monocyte chemoattractant protein-1 and macrophage
Intern. Med. 122: 833–838. infiltration in hypertensive kidney injury. Kidney Int. 58:
6 Zoccali, C., Mallamaci, F., and Finocchiaro, P. (2002). 2408–2419.
Atherosclerotic renal artery stenosis: epidemiology, 20 Favreau, F., Zhu, X.Y., Krier, J.D. et al. (2010).
cardiovascular outcomes, and clinical prediction rules. J. Revascularization of swine renal artery stenosis improves
Am. Soc. Nephrol. 13 (Suppl 3): S179–S183. renal function but not the changes in vascular structure.
7 Harding, M.B., Smith, L.R., Himmelstein, S.I. et al. Kidney Int. 78: 1110–1118.
(1992). Renal artery stenosis: prevalence and associated 21 Plouin, P.F., Chatellier, G., Darne, B., and Raynaud, A.
risk factors in patients undergoing routine cardiac (1998). Blood pressure outcome of angioplasty in
catheterization. J. Am. Soc. Nephrol. 2: 1608–1616. atherosclerotic renal artery stenosis: a randomized
8 Olin, J.W., Melia, M., Young, J.R. et al. (1990). Prevalence trial. Essai Multicentrique Medicaments vs
of atherosclerotic renal artery stenosis in patients with Angioplastie (EMMA) Study Group. Hypertension 31:
atherosclerosis elsewhere. Am. J. Med. 88: 46N–51N. 823–829.
9 Aboyans, V., Ricco, J.B., Bartelink, M.E.L. et al. (2018). 22 Webster, J., Marshall, F., Abdalla, M. et al. (1998).
Editor’s choice – 2017 ESC guidelines on the diagnosis Randomised comparison of percutaneous angioplasty vs
and treatment of peripheral arterial diseases, in continued medical therapy for hypertensive patients with
collaboration with the European Society for Vascular atheromatous renal artery stenosis. Scottish and
Surgery (ESVS). Eur. J. Vasc. Endovasc. Surg. 55: 305–368. Newcastle Renal Artery Stenosis Collaborative Group. J.
10 Grim, C.E., Luft, F.C., Yune, H.J. et al. (1981). Hum. Hypertens. 12: 329–335.
Percutaneous transluminal dilatation in the treatment of 23 Bax, L., Woittiez, A.J., Kouwenberg, H.J. et al. (2009).
renal vascular hypertension. Ann. Intern. Med. 95: 439–442. Stent placement in patients with atherosclerotic renal
11 Geyskes, G.G., Puylaert, C.B.A.J., Oei, H.Y., and Dorhout artery stenosis and impaired renal function: a
Mees, E.J. (1983). Follow up study of 70 patients with randomized trial. Ann. Intern. Med. 150: 840–848.
renal artery stenosis treated by percutaneous W150–1.
transluminal dilatation. BMJ 287: 333–336. 24 Siddiqui, E.U., Murphy, T.P., Naeem, S.S. et al. (2018).
12 Svetkey, L.P., Kadir, S., Dunnick, N.R. et al. (1991). Interaction between albuminuria and treatment group
Similar prevalence of renovascular hypertension in outcomes for patients with renal artery stenosis: the
selected blacks and whites. Hypertension 17: 678–683. NITER study. J. Vasc. Interv. Radiol. 29: 966–970.
13 Fommei, E., Ghione, S., Hilson, A.J. et al. (1993). 25 Investigators, A., Wheatley, K., Ives, N. et al. (2009).
Captopril radionuclide test in renovascular hypertension: Revascularization versus medical therapy for renal-artery
a European multicentre study. European Multicentre stenosis. N. Engl. J. Med. 361: 1953–1962.
Study Group. Eur. J. Nucl. Med. 20: 617–623. 26 Marcantoni, C., Zanoli, L., Rastelli, S. et al. (2012). Effect
14 Mann, S.J., Pickering, T.G., Sos, T.A. et al. (1991). of renal artery stenting on left ventricular mass: a
Captopril renography in the diagnosis of renal artery randomized clinical trial. Am. J. Kidney Dis. 60: 39–46.
stenosis: accuracy and limitations. Am. J. Med. 90: 30–40. 27 Cooper, C.J., Murphy, T.P., Cutlip, D.E. et al. (2014).
15 van Jaarsveld, B.C., Krijnen, P., Pieterman, H. et al. Stenting and medical therapy for atherosclerotic renal-
(2000). The effect of balloon angioplasty on hypertension artery stenosis. N. Engl. J. Med. 370: 13–22.
in atherosclerotic renal-artery stenosis. Dutch Renal 28 Zeller, T., Krankenberg, H., Erglis, A. et al. (2017). A
Artery Stenosis Intervention Cooperative Study Group. N. randomized, multi-center, prospective study comparing
Engl. J. Med. 342: 1007–1014. best medical treatment versus best medical treatment
16 Radermacher, J., Chavan, A., Bleck, J. et al. (2001). Use of plus renal artery stenting in patients with
Doppler ultrasonography to predict the outcome of hemodynamically relevant atherosclerotic renal artery
therapy for renal-artery stenosis. N. Engl. J. Med. 344: stenosis (RADAR) – one-year results of a pre-maturely
410–417. terminated study. Trials 18: 380.
17 Johansson, M., Jensen, G., Aurell, M. et al. (2000). 29 Rossi, G.P., Seccia, T.M., Miotto, D. et al. (2012). The
Evaluation of duplex ultrasound and captopril medical and endovascular treatment of atherosclerotic
renography for detection of renovascular hypertension. renal artery stenosis (METRAS) study: rationale and
Kidney Int. 58: 774–782. study design. J. Hum. Hypertens. 26: 507–516.
Reference 363
0 Jenks, S., Yeoh, S.E., and Conway, B.R. (2014). Balloon

3 41 van Jaarsveld, B.C., Pieterman, H., van Dijk, L.C. et al.
angioplasty, with and without stenting, versus medical (1999). Inter-observer variability in the angiographic
therapy for hypertensive patients with renal artery assessment of renal artery stenosis. DRASTIC study
stenosis. Cochrane Database Syst. Rev.: CD002944. group. Dutch Renal Artery Stenosis Intervention
31 Bavry, A.A., Kapadia, S.R., Bhatt, D.L., and Kumbhani, Cooperative. J. Hypertens. 17: 1731–1736.
D.J. (2014). Renal artery revascularization: updated 42 Sigmund, G., Hettinger, M., Block, T. et al. (2000).
meta-analysis with the CORAL trial. JAMA Intern. Med. Evaluation of renal artery stenosis: comparison of
174: 1849–1851. angiography and invasive blood pressure measurement
32 Oberhuber, A., Vonend, O., Radermacher, J. et al. (2018). and Doppler ultrasound. Rofo 172: 615–622.
Leitlinie – S2K Erkrankungen der Nierenarterie. 43 Selkurt, E.E. (1951). Effect of pulse pressure and mean
Arbeitsgemeinschaft der Wissenschaftlichen arterial pressure modification on renal hemodynamics
Medizinischen Fachgesellschaften (AWMF) – Ständige and electrolyte and water excretion. Circulation 4:
Kommission Leitlinien. Available: HYPERLINK “https:// 541–551.
www.awmf.org/uploads/tx_szleitlinien/004-008l_S2k_ 44 Dawson, D.L. (1996). Noninvasive assessment of renal
Erkrankungen-der-Nierenarterien_2020-01_01.pdf” artery stenosis. Semin. Vasc. Surg. 9: 172–181.
Endversion Leitlinie - S2K Erkrankungen der 45 Olbricht, C.J., Paul, K., Prokop, M. et al. (1995).
Nierenarterie 004-008 inkl. DGA (awmf.org) (Access Minimally invasive diagnosis of renal artery stenosis by
03.09.2021). spiral computed tomography angiography. Kidney Int. 48:
33 Ritchie, J., Green, D., Chrysochou, C. et al. (2014). 1332–1337.
High-risk clinical presentations in atherosclerotic 46 Elkohen, M., Beregi, J.P., Deklunder, G. et al. (1995).
renovascular disease: prognosis and response to renal Evaluation of spiral computed tomography of the renal
artery revascularization. Am. J. Kidney Dis. 63: 186–197. arteries alone or combined with Doppler ultrasonography
34 Vassallo, D., Ritchie, J., Green, D. et al. (2018). The effect in the detection of renal artery stenosis. Prospective study
of revascularization in patients with anatomically of 114 renal arteries. Arch. Mal. Coeur Vaiss. 88: 1159–1164.
significant atherosclerotic renovascular disease 47 Vasbinder, G.B., Nelemans, P.J., Kessels, A.G. et al.
presenting with high-risk clinical features. Nephrol. Dial. (2004). Accuracy of computed tomographic angiography
Transplant. 33: 497–506. and magnetic resonance angiography for diagnosing
35 Gray, B.H., Olin, J.W., Childs, M.B. et al. (2002). Clinical renal artery stenosis. Ann. Intern. Med. 141: 674–682.
benefit of renal artery angioplasty with stenting for the discussion 82.
control of recurrent and refractory congestive heart 48 Lenz, T., Kia, T., Rupprecht, G. et al. (1999). Captopril
failure. Vasc. Med. 7: 275–279. test: time over? J. Hum. Hypertens. 13: 431–435.
36 Mancia, G., Fagard, R., Narkiewicz, K. et al. (2014). 2013 49 Radermacher, J., Weinkove, R., and Haller, H. (2001).
ESH/ESC practice guidelines for the Management of Techniques for predicting a favourable response to renal
Arterial Hypertension. Blood Press. 23: 3–16. angioplasty in patients with renovascular disease. Curr.
37 Cheung, C.M., Patel, A., Shaheen, N. et al. (2007). The Opin. Nephrol. Hypertens. 10: 799–805.
effects of statins on the progression of atherosclerotic 50 Taylor, A. (2000). Functional testing: ACEI renography.
renovascular disease. Nephron Clin. Pract. 107: c35–c42. Semin. Nephrol. 20: 437–444.
38 Silva, V.S., Martin, L.C., Franco, R.J. et al. (2008). 51 Radermacher, J., Schäffer, J., Stoess, B. et al. (2000).
Pleiotropic effects of statins may improve outcomes in Detection of significant renal artery stenosis with color
atherosclerotic renovascular disease. Am. J. Hypertens. 21: Doppler sonography:combining extrarenal and intrarenal
1163–1168. approaches to minimise technical failure. Clin. Nephrol.
39 European Stroke, O., Tendera, M., Aboyans, V. et al. 53: 333–343.
(2011). ESC guidelines on the diagnosis and treatment of 52 Hansen, K.J., Tribble, R.W., Reavis, S.W. et al. (1990).
peripheral artery diseases: document covering Renal duplex sonography: evaluation of clinical utility. J.
atherosclerotic disease of extracranial carotid and Vasc. Surg. 12: 227–236.
vertebral, mesenteric, renal, upper and lower extremity 53 Karasch, T., Strauss, A.L., Grun, B. et al. (1993). Color-
arteries: the task force on the diagnosis and treatment of coded duplex ultrasonography in the diagnosis of renal
peripheral artery diseases of the European Society of artery stenosis. Dtsch. Med. Wochenschr. 118: 1429–1436.
Cardiology (ESC). Eur. Heart J. 32: 2851–2906. 54 Postma, C.T., van Aalen, J., de Boo, T. et al. (1992).
40 Buller, C.E., Nogareda, J.G., Ramanathan, K. et al. (2004). Doppler ultrasound scanning in the detection of renal
The profile of cardiac patients with renal artery stenosis. artery stenosis in hypertensive patients. Br. J. Radiol. 65:
J. Am. Coll. Cardiol. 43: 1606–1613. 857–860.
5 Schäberle, W., Strauss, A., Neuerburg, H.D., and Roth, F.J.

5 69 Giroux, M.F., Soulez, G., Therasse, E. et al. (2000).
(1992). Duplex scan in diagnosis of renal artery stenoses Percutaneous revascularization of the renal arteries:
and in the assessment of patency following transluminal predictors of outcome [in process citation]. J. Vasc. Interv.
angioplasty. Ultraschall. Med. 13: 271–276. Radiol. 11: 713–720.
56 AbuRahma, A.F., Srivastava, M., Mousa, A.Y. et al. 70 Simon, G. (2000). What is critical renal artery stenosis?
(2012). Critical analysis of renal duplex ultrasound Implications for treatment. Am. J. Hypertens. 13:
parameters in detecting significant renal artery stenosis. 1189–1193.
J. Vasc. Surg. 56: 1052–1059. 60 e1; discussion 9–60. 71 Textor, S.C. and Wilcox, C.S. (2001). Renal artery stenosis:
57 Baxter, G.M., Aitchison, F., Sheppard, D. et al. (1996). a common, treatable cause of renal failure? Annu. Rev.
Colour Doppler ultrasound in renal artery stenosis: Med. 52: 421–442.
intrarenal waveform analysis. Br. J. Radiol. 69: 72 Wright, J.R., Duggal, A., Thomas, R. et al. (2001).
810–815. Clinicopathological correlation in biopsy-proven
58 Kliewer, M.A., Tupler, R.H., Carroll, B.A. et al. (1993). atherosclerotic nephropathy: implications for renal
Renal artery stenosis: analysis of Doppler waveform functional outcome in atherosclerotic renovascular
parameters and tardus-parvus pattern. Radiology 189: disease. Nephrol. Dial. Transplant. 16: 765–770.
779–787. 73 Plouin, P.F., Clement, D.L., Boccalon, H. et al. (2003). A
59 Riehl, J., Schmitt, H., Bongartz, D. et al. (1997). Renal clinical approach to the management of a patient with
artery stenosis: evaluation with colour duplex suspected renovascular disease who presents with leg
ultrasonography. Nephrol. Dial. Transplant. 12: ischemia. Int. Angiol. 22: 333–339.
1608–1614. 74 Safian, R.D. (2003). Atherosclerotic renal artery stenosis.
60 Schwerk, W.B., Restrepo, I.K., Stellwaag, M. et al. (1994). Curr. Treat. Options Cardiovasc. Med. 5: 91–101.
Renal artery stenosis: grading with image-directed 75 Barri, Y.M., Davidson, R.A., Senler, S. et al. (1996).
Doppler US evaluation of renal resistive index. Radiology Prediction of cure of hypertension in atherosclerotic
190: 785–790. renal artery stenosis. South. Med. J. 89: 679–683.
61 Speckamp, F., Vorwerk, D., Schurmann, K. et al. (1995). 76 Helin, K.H., Lepantalo, M., Edgren, J. et al. (2000).
Color-coded duplex ultrasonography in the diagnosis of Predicting the outcome of invasive treatment of renal
renal artery stenosis. Rofo. Fortschr. Geb. Rontgenstr. artery disease. J. Intern. Med. 247: 105–110.
Neuen. Bildgeb. Verfahr. 162: 412–419. 77 Burket, M.W., Cooper, C.J., Kennedy, D.J. et al. (2000).
62 Stavros, A.T., Parker, S.H., Yakes, W.F. et al. (1992). Renal artery angioplasty and stent placement: predictors
Segmental stenosis of the renal artery: pattern of a favorable outcome. Am. Heart J. 139: 64–71.
recognition of tardus and parvus abnormalities with 78 Paulsen, D., Klow, N.E., Rogstad, B. et al. (1999).
duplex sonography. Radiology 184: 487–492. Preservation of renal function by percutaneous
63 Strunk, H., Jaeger, U., and Teifke, A. (1995). Intrarenal transluminal angioplasty in ischaemic renal disease.
color Doppler ultrasound for exclusion of renal artery Nephrol. Dial. Transplant. 14: 1454–1461.
stenosis in cases of multiple renal arteries. Analysis of the 79 Makanjuola, A.D., Suresh, M., Laboi, P. et al. (1999).
Doppler spectrum and tardus parvus phenomenon. Proteinuria in atherosclerotic renovascular disease. QJM
Ultraschall. Med. 16: 172–179. 92: 515–518.
64 Krumme, B., Blum, U., Schwertfeger, E. et al. (1996). 80 Zeller, T., Muller, C., Frank, U. et al. (2003). Stent
Diagnosis of renovascular disease by intra-and extrarenal angioplasty of severe atherosclerotic ostial renal artery
Doppler scanning. Kidney Int. 50: 1288–1292. stenosis in patients with diabetes mellitus and
65 Radermacher, J. (2014). Rationales Vorgehen bei nephrosclerosis. Catheter. Cardiovasc. Interv. 58: 510–515.
Nierenarterienstenose. Der Nephrologe 9: 357–363. 81 Muray, S., Martin, M., Amoedo, M.L. et al. (2002). Rapid
66 Saeed, A., Bergstrom, G., Zachrisson, K. et al. (2009). decline in renal function reflects reversibility and predicts
Accuracy of colour duplex sonography for the the outcome after angioplasty in renal artery stenosis.
diagnosis of renal artery stenosis. J. Hypertens. 27: Am. J. Kidney Dis. 39: 60–66.
1690–1696. 82 Pedersen, E.B. (2000). New tools in diagnosing renal
67 Kaplan, N.M. (2006). Renal vascular hypertension. In: F. artery stenosis. Kidney Int. 57: 2657–2677.
D, Jackson A. In: Clinical Hypertension, 6e, 347–368. 83 Grenier, N., Trillaud, H., Combe, C. et al. (1996).
Philadelphia: Williams and Wilkins. Diagnosis of renovascular hypertension: feasibility of
68 Schreiber, M.J., Pohl, M.A., and Novick, A.C. (1984). The captopril-sensitized dynamic MR imaging and
natural history of atherosclerotic and fibrous renal artery comparison with captopril scintigraphy. Am. J.
disease. Urol. Clin. North Am. 11: 383–392. Roentgenol. 166: 835–843.
Reference 365
4 Zhang, J.L., Rusinek, H., Bokacheva, L. et al. (2009).

8 protection in patients with ischemic nephropathy.
Angiotensin-converting enzyme inhibitor-enhanced MR Kidney Int. 70: 948–955.
renography: repeated measures of GFR and RPF in 97 Solomon, R., Werner, C., Mann, D. et al. (1994). Effects
hypertensive patients. Am. J. Physiol. Renal Physiol. 296: of saline, mannitol, and furosemide on acute decreases
F884–F891. in renal function induced by radiocontrast agents. N.
85 Petersen, L.J., Petersen, J.R., Talleruphuus, U. et al. Engl. J. Med. 331: 1416–1420.
(1997). The pulsatility index and the resistive index in 98 Merten, G.J., Burgess, W.P., Gray, L.V. et al. (2004).
renal arteries. Associations with long-term progression in Prevention of contrast-induced nephropathy with
chronic renal failure. Nephrol. Dial. Transplant. 12: sodium bicarbonate: a randomized controlled trial.
1376–1380. JAMA 291: 2328–2334.
86 Frauchiger, B., Zierler, R., Bergelin, R.O. et al. (1996). 99 Tepel, M., van der Giet, M., Schwarzfeld, C. et al. (2000).
Prognostic significance of intrarenal resistance indices in Prevention of radiographic-contrast-agent-induced
patients with renal artery interventions: a preliminary reductions in renal function by acetylcysteine. N. Engl. J.
duplex sonographic study. Cardiovasc. Surg. 4: 324–330. Med. 343: 180–184.
87 Paty, P.S., Darling, R.C. 3rd, Lee, D. et al. (2001). Is 100 Marenzi, G., Assanelli, E., Marana, I. et al. (2006).
prosthetic renal artery reconstruction a durable N-acetylcysteine and contrast-induced nephropathy in
procedure? An analysis of 489 bypass grafts. J. Vasc. Surg. primary angioplasty. N. Engl. J. Med. 354: 2773–2782.
34: 127–132. 101 Bagshaw, S.M., McAlister, F.A., Manns, B.J., and Ghali,
88 Martin, L.G., Rundback, J.H., Sacks, D. et al. (2003). W.A. (2006). Acetylcysteine in the prevention of
Quality improvement guidelines for angiography, contrast-induced nephropathy: a case study of the
angioplasty, and stent placement in the diagnosis and pitfalls in the evolution of evidence. Arch. Intern. Med.
treatment of renal artery stenosis in adults. J. Vasc. Interv. 166: 161–166.
Radiol. 14: S297–S310. 102 Rudnick, M.R., Goldfarb, S., Wexler, L. et al. (1995).
89 van de Ven, P.J., Kaatee, R., Beutler, J.J. et al. (1999). Nephrotoxicity of ionic and nonionic contrast media in
Arterial stenting and balloon angioplasty in ostial 1196 patients: a randomized trial. The Iohexol
atherosclerotic renovascular disease: a randomised trial. Cooperative Study. Kidney Int. 47: 254–261.
Lancet 353: 282–286. 103 Aspelin, P., Aubry, P., Fransson, S.G. et al. (2003).
90 Bush, R.L., Najibi, S., MacDonald, M.J. et al. (2001). Nephrotoxic effects in high-risk patients undergoing
Endovascular revascularization of renal artery stenosis: angiography. N. Engl. J. Med. 348: 491–499.
technical and clinical results. J. Vasc. Surg. 33: 1041–1049. 104 Marenzi, G., Marana, I., Lauri, G. et al. (2003). The
91 Williams, G.J., Macaskill, P., Chan, S.F. et al. (2007). prevention of radiocontrast-agent-induced
Comparative accuracy of renal duplex sonographic nephropathy by hemofiltration. N. Engl. J. Med. 349:
parameters in the diagnosis of renal artery stenosis: 1333–1340.
paired and unpaired analysis. Am. J. Roentgenol. 188: 105 Kurnik, B.R., Allgren, R.L., Genter, F.C. et al. (1998).
798–811. Prospective study of atrial natriuretic peptide for the
92 La Batide-Alanore, A., Azizi, M., Froissart, M. et al. prevention of radiocontrast-induced nephropathy. Am.
(2001). Split renal function outcome after renal J. Kidney Dis. 31: 674–680.
angioplasty in patients with unilateral renal artery 106 Abizaid, A.S., Clark, C.E., Mintz, G.S. et al. (1999).
stenosis. J. Am. Soc. Nephrol. 12: 1235–1241. Effects of dopamine and aminophylline on contrast-
93 Rodriguez-Lopez, J.A., Werner, A., Ray, L.I. et al. (1999). induced acute renal failure after coronary angioplasty in
Renal artery stenosis treated with stent deployment: patients with preexisting renal insufficiency. Am. J.
indications, technique, and outcome for 108 patients. J. Cardiol. 83: 260–263. A5.
Vasc. Surg. 29: 617–624. 107 Wang, A., Holcslaw, T., Bashore, T.M. et al. (2000).
94 Klugherz, B.D., Jones, P.L., Cui, X. et al. (2000). Gene Exacerbation of radiocontrast nephrotoxicity by
delivery from a DNA controlled-release stent in porcine endothelin receptor antagonism. Kidney Int. 57:
coronary arteries. Nat. Biotechnol. 18: 1181–1184. 1675–1680.
95 Dejani, H., Eisen, T.D., and Finkelstein, F.O. (2000). 108 Erley, C.M., Duda, S.H., Rehfuss, D. et al. (1999).
Revascularization of renal artery stenosis in patients Prevention of radiocontrast-media-induced
with renal insufficiency. Am. J. Kidney Dis. 36: nephropathy in patients with pre-existing renal
752–758. insufficiency by hydration in combination with the
96 Holden, A., Hill, A., Jaff, M.R., and Pilmore, H. (2006). adenosine antagonist theophylline. Nephrol. Dial.
Renal artery stent revascularization with embolic Transplant. 14: 1146–1149.
366
24
Secondary Diseases of the Kidney: Diabetic Nephropathy

Patrizia Natale1,2, Muh Geot Wong2,5, Giovanni F.M. Strippoli1,2, David J. Tunnicliffe2,4,
Tess E. Cooper2,4, and Carol Pollock2,3
1
Department of Emergency and Organ Transplantation, University of Bari, Bari, Italy
2
Sydney School of Public Health, The University of Sydney, Sydney, Australia
3
Kolling institute of Medical Research, St Leonards, Australia
4
Centre for Kidney Research, The Children’s Hospital at Westmead, Westmead, Australia
5
Department of Renal Medicine, Concord Repatriation General Hospital, Concord 2139 Australia
iabetic Nephropathy vs. Diabetic

D I ncidence and Prevalence of Diabetic
Kidney Disease Kidney Disease
Diabetic nephropathy classically refers to the renal microvas- Diabetes is recognized as the world’s fastest growing non-
cular pathology that occurs in patients with type 2 diabetes communicable chronic disease. In 2003 the International
mellitus (DM) resulting in glomerular hyperfiltration, moder- Diabetes Federation (IDF) estimated that 194 million adults
ately increased urine protein, and severely increased urine (aged 20–79 years) were living with diabetes and projected
protein, and followed by a reduction in glomerular filtration that this number would increase to 333 million people by
rate (GFR). It is generally accompanied by hypertension and 2025. However, in 2017 the IDF estimated 425 million people
other evidence of microvascular pathology, including retin- were living with diabetes and this number is expected to
opathy and neuropathy. It is characterized histologically increase to 629 million by 2045 [3] (Figure 24.1).
by thickening of the glomerular and tubular basement In the 2017 report from the IDF [3], the highest preva-
membranes, mesangial expansion, and nodular sclerosis lence is in China with 114 million people, followed by India
(Kimmelstiel–Wilson lesions). Although a recognized entity with 73 million, and the USA with 30 million. The esti-
for 50 years [1], a pathological classification, dominantly mated worldwide expenditure on diabetes in 2017 was US
based on glomerular pathology, has only been developed in $ 727 billion. Alarmingly, rates of childhood diabetes are
the last decade [2] (Table 24.1). The classification also notes increasing, being highest in the USA. Contrary to expecta-
the presence of tubulointerstitial lesions, vascular lesions, and tion, approximately one-third of the world population liv-
nondiabetic glomerular lesions which may be present to a ing with diabetes resides in nonurban environments and
variable degree in classic diabetic nephropathy [2] (Table 24.2). thus is arguably less likely to be strongly influenced by a
It is now recognized that individuals, particularly those Western diet. The vast majority of cases are due to type 2
with type 2 DM and chronic kidney disease (CKD), do DM, but the incidence of type 1 DM is also increasing, with
not uniformly present with the clinical features and a marked variation in the incidence and prevalence of type
trajectory of biochemical abnormalities outlined above, 1 DM across the globe that is currently unexplained.
nor with the glomerular histological features listed in The increasing prevalence of type 2 DM is driving the
Table 24.1. Hence modern nomenclature has favored increased prevalence of kidney failure, with kidney failure
the term diabetic kidney disease (DKD) over diabetic 10 times more likely to occur in people with diabetes than in
nephropathy as this is more representative of the coexist- the remainder of the population [4]. The IDF estimate DKD is
ent obesity-induced, vascular, and hypertensive kidney estimated to be responsible for between 12% and 55% of cases
injury that is likely to be over-represented in patients of kidney failure worldwide [3]. The proportion of kidney fail-
with type 2 DM. ure attributable to diabetes, hypertension, or a combination of
Pathogenesis of Diabetic Kidney Diseas 367
Table 24.1 Glomerular classification of DKD. Table 24.2 Interstitial and vascular lesions of DKD* [2].
Class Description Inclusion criteria Lesion Criteria Score
I Mild or nonspecific LM Biopsy does not meet any of Interstitial lesions

changes and EM proven the criteria mentioned below IFTA No IFTA 0
GBM thickening for classes II, III, or IV
<25% 1
GBM >395 nm in female and
>430 nm in male individuals 25–50% 2
9 years of age and older >50% 3
IIa Mild mesangial Biopsy does not meet criteria Interstitial Absent 0
expansion** for class III or IV inflammation
Mild mesangial expansion in Infiltration only in relation 1
>25% of the observed to IFTA
mesangium
Infiltration in areas 2
IIb Severe mesangial Biopsy does not meet criteria without IFTA
expansion for class III or IV
Vascular lesions
Severe mesangial expansion
in >25% of the observed Arteriolar hyalinosis Absent 0
mesangium At least one area of 1
III Nodular sclerosis Biopsy does not meet criteria arteriolar hyalinosis
(Kimmelstiel–Wilson for class IV More than one area of 2
lesion) arteriolar hyalinosis
At least one convincing Presence of large vessels – Yes/no
Kimmelstiel–Wilson lesion Arteriosclerosis No intimal thickening 0
IV Advanced diabetic Global glomerular sclerosis in (score worst artery)
glomerulosclerosis >50% of glomeruli Intimal thickening less 1
Lesions from classes I than thickness of media
through III Intimal thickening greater 2
than thickness of media
DKD: diabetic kidney disease
LM: light microscopy
EM: electron microscopy
IFTA: renal interstitial fibrosis and tubular atrophy
GBM: glomerular baseline membrane
Source: Tervaert et al. [2]. © 2010, American Society of Nephrology.
IFTA: interstitial fibrosis tubular atrophy
**The difference between mild and severe mesangial expansion is
based on whether the expanded mesangial area is smaller or larger
verlap with additional factors more prevalent in patients
o
than the mean area of a capillary lumen
Source: Tervaert et al. [2]. © 2010, American Society of Nephrology. with type 2 DM, including hypertension, macrovascular
disease, and infection to result in a complex pathogenesis of
kidney disease in patients with DM. With some significant
both in the majority of countries where data are available is exceptions, particularly regarding glycemic control, the
approximately 80%. As hypertension and diabetes are inextri- majority of evidence to improve outcomes in DKD is based
cably linked, ascertainment bias as to the cause of kidney fail- on trials in patients with type 2 DM and associated DKD.
ure may contribute to this variation. It is estimated that Hyperglycemia is considered a key driver of DKD [6, 7].
patients with CKD not requiring dialysis have 50% higher It is directly or indirectly responsible for both metabolic
health costs compared to patients with diabetes but no CKD. abnormalities and altered intrarenal hemodynamics that
In Australia, patients with CKD are 20 times more likely to die contribute collectively to DKD. Intracellular hyperglyce-
of a cardiovascular event than to reach kidney failure [5]. mia contributes to activation of the polyol pathway [8],
hyperuricemia [9], dyslipidemia [10], the production of
advanced glycation end products (AGEs), and interaction
Pathogenesis of Diabetic Kidney Disease with receptors such as the receptor for advanced glycation
end products (RAGE) leading to oxidative stress [11]. As a
The pathogenesis of DKD is complex, with multiple inter- result, fibrosis, thrombosis, DNA damage, cellular dys-
acting and redundant pathways contributing to progressive function, vascular leakage, angiogenesis, inflammation,
kidney pathology. Mediators that are responsible for DKD and decreased oxidative stress defense mechanisms are ini-
(classically seen in younger patients with type 1 DM) tiated, resulting in tubulointerstitial fibrosis [12, 13].
368 Secondary Diseases of the Kidney: Diabetic Nephropathy
500 millions
450 415 425
382
400 366
350
285
300 246
250 194
200 151
150
100
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
Figure 24.1 Estimates of the number of adults (aged 20–79 years) living with diabetes worldwide. Source: Based on Ogurtsova
et al. [3].
Somewhat paradoxically DKD is associated with reduced molecules, derived from injured cells and components of
mitochondrial superoxide production. A basal level of the extracellular matrix (ECM), such as hyaluronan, bigly-
mitochondrial superoxide level is required to maintain can, and fibronectin. DAMPs trigger innate immunity by
adenosine monophosphate kinase (AMPK) activation and activating Toll-like receptors, purinergic receptors, or the
phosphorylation of endothelial nitric oxide to promote vas- NLRP3 inflammasome, leading to necroptosis increased
cular dilatation under physiological circumstances. interferon gamma and downstream chemoattractants,
Reduced AMPK activity stimulates inflammatory tran- macrophage chemoattractant protein-1 (MCP-1), and
scription factors, including nuclear factor kappa B (NFκβ), CCl10 [28–34]. Necroinflammation, an autoamplification
and proinflammatory and profibrotic molecules in a parac- loop between tubular cell death and interstitial inflamma-
rine manner from multiple renal cells. Under hyperglyce- tion, leads to the exacerbation of kidney injury [35, 36].
mic conditions there is a reduction in mitochondrial Although not yet specifically reported in DKD it is likely
superoxide generation, oxidative phosphorylation, and that common mechanisms are in play across the spectrum
mitochondrial adenosine triphosphate (ATP) generation in of CKD. Recent evidence suggests DAMPs also contribute
the kidney and abnormalities in cellular bioenergetics to epithelial-mesenchymal transition of tubular cells
occur [14, 15]. to myofibroblast differentiation and proliferation.
Vascular and glomerular lesions contribute to hypoxia, Furthermore, tubular cells also play an active role in pro-
which further exacerbates oxidative stress in both the glo- gressive kidney injury via emerging mechanisms associ-
merulus and tubules [13, 16]. Hypoxia-inducible factors ated with a partial epithelial-mesenchymal transition and
(HIFs) are induced with downstream transcription of cell-cycle arrest at both G1/S and G2/M checkpoints [37].
genes such as vascular endothelial growth factor (VEGF) It is clear that inflammation plays a major role in the
and erythropoietin to improve tissue oxygenation. pathogenesis of CKD. Inflammation occurs as a result of
However, there is experimental evidence that hyperglyce- cellular injury, mediated by macrophages and lympho-
mia may interfere with the stability of HIF and facilitate cytes [38–40]. The transcription factor nuclear factor kappa
tissue fibrosis [17, 18]. B (NF-κB) is responsible for upregulating genes involved in
These diverse mediators collectively lead to downstream inflammatory processes within macrophages and tubular
activation of hormones and proinflammatory media- cells. Amplification of pro-inflammatory and pro-fibrotic
tors [19] and profibrotic cascades. It is increasingly recog- cytokines, such as transforming growth factor β (TGF-β),
nized that inflammation and fibrosis are inextricably platelet derived growth factor (PDGF), and fibroblast
linked in all forms of CKD, including DKD [20–23]. It is growth factor 2 (FGF-2), then occurs that causes migration
also now evident that as well as the adaptive immune sys- and proliferation of resident fibroblasts. Such fibroblasts
tem being activated in CKD, the innate immune system is are differentiated, and along with tubular and glomerular
also upregulated [24–28]. These responses are initiated by cells ECM production is enhanced [41–43]. Resident and
several classes of pattern recognition receptors (PRRs) in invading cells then secrete various chemokines, including
response to immune activators such as pathogen-associated MCP-1, IL-6, and tumor necrosis factor α (TNFα), that lead
molecular patterns (PAMPs) and damage-associated to further macrophage accumulation, perpetuating the
molecular patterns (DAMPs). The latter are heterogenous development of fibrosis [44]. TGF-β, VEGFs, insulin-like
Screening and Diagnosis of Diabetic Kidney Diseas 369
growth factors (IGFs), and lipid mediators also contribute mechanisms to ensure orchestrated fatty acid uptake, oxi-
to inflammation and fibrosis [45]. dation, and synthesis are tightly regulated to avoid exces-
As stated above, multiple transcription factors and sive intracellular lipid accumulation. The transcription
cytokines are involved in expansion of the ECM, but TGFβ1 factors, peroxisome proliferator-activated receptors
is considered the “master regulator” of kidney fibrosis in (PPARs), and PPARδ coactivator-1a (PGC1A) regulate the
DKD. TGF-β1 is a well-characterized key mediator in the expression of proteins involved in fatty acid uptake and oxi-
pathogenesis of tubulointerstitial fibrosis due to its direct dation. In the setting of CKD lower expression of genes
and indirect effect on various cells types [46–49]. responsible for fatty acid oxidation, PPARs, and PGC1A is
The direct action of TGF-β1 can initiate the transition of observed suggesting that lipid metabolism is impaired.
tubular and endothelial cells to a fibroblastic phenotype as Human proximal tubular studies have demonstrated that
described above and proliferation, migration, and synthe- TGF-β is intrinsically involved in blocking fatty acid oxida-
sis of profibrotic proteins, such as collagens and fibronec- tion pathways, mediated by PGC1A [72].
tin [50–56]. TGF-β1 can also induce an indirect fibrotic Genetic and epigenetic modifications of both genes and
response via accelerating apoptosis of resident healthy histones induced by hyperglycemia, hypoxia, and as a con-
cells and promoting resident and infiltrating cells to sequence of proinflammatory and profibrotic pathways are
increase ECM deposition [57–59]. increasingly recognized to influence the phenotype of indi-
Hyperglycemia also causes release of vasoactive media- viduals that directly or indirectly influences the develop-
tors, such as IGF-1, glucagon, nitric oxide (NO), VEGF, and ment of diabetes and the propensity to develop DKD. Gene
prostaglandins, resulting in afferent arteriolar vasodilation. methylation or acetylation induced by hyperglycemia sig-
Additionally, both hyperglycaemia and insulin resistance nificantly influences the likelihood of gene expression or
result in dysregulation of the renin-angiotensin-aldoster- repression involved in the development of DKD. Similarly,
one and endothelin [60] systems that act both systemically hyperglycemia modifies the expression of noncoding
and specifically on the efferent arteriole. This results in mRNA including microRNA, long noncoding RNA, and
peripheral and intraglomerular hypertension, leading to small RNA, which collectively modify the expression of
kidney impairment. [61]. Upregulation of proximal tubular coding genes and therefore the phenotype of the individual
sodium reabsorption occurs due to increased expression of with DM [73–80]. Increasingly it is recognized that epige-
sodium-glucose linked transport-2 and increased sodium- netic modifications, particularly of mitochondrial DNA,
hydrogen exchange leading to a low sodium concentration may be maternally transmitted from mother to offspring
at the macula densa and inhibition of tubuloglomerular and hence predispose to population shifts in the propensity
feedback, thus resulting in single nephron glomerular to chronic disease.
hyperfiltration [62]. A schematic view of the key processes involved in the
Shear stress as well as podocyte apoptosis driven by angi- development of progressive kidney disease is summarized
otensin II and endothelin, reduction in the slit diaphragm in Figure 24.2.
proteins nephrin and podocin, and glycation of the glomer-
ular basement membrane with associated loss of negative
charge results in albuminuria [63–65].
More recently, additional pathways have been identified creening and Diagnosis of Diabetic
S
as contributing to cellular dysfunction that predisposes to Kidney Disease
or accelerates the development of DKD. Both hyperglyce-
mia and hyperinsulinemia trigger abnormalities in cell Moderately increased urine protein is an independent risk
bioenergetics [66] that ultimately lead to inefficient use of factor for the development of CKD and GFR loss [81, 82], as
metabolic substrates that contribute to tissue hypoxia. well as cardiovascular (CV) morbidities and mortality [83–
Additional dysregulated metabolic pathways, include 86]. Most guidelines recommend annual screening for mod-
upregulation of the mammalian target of rapamycin erately increased urine protein in individuals with
(mTOR), which drives suboptimal use of metabolic sub- DM [87–90]. Screening should begin 5 years after the onset of
strates, resulting in both cytosolic and mitochondrial type 1 DM and at the diagnosis of type 2 DM because of the
derived dysregulated oxidative stress and dysfunctional inability to establish the onset of type 2 DM with certainty.
autophagy and mitophagy [66–71]. Annual assessments of both albuminuria and of kidney func-
Fatty acid beta-oxidation is the catabolic process by tion are commonly used screening tools [88]. Although a
which fatty acid molecules are broken down in the cytosol 24 hour urine collection was originally accepted as the gold
and in the mitochondria to generate acetyl-CoA, which standard for urinary albumin measurement, it is inconven-
enters the citric acid cycle to produce ATP. The cellular ient to most patients and accuracy is flawed by inadequate
Hyperglycaemia
Systemic and renal Ischaemia, Epigenetic Podocyte and tubular Hyperuricaemia Dysregulated Mitochondrial
haemodynamic inflammation and regulation cell autophagy nutrient utilisation dysfunction
Changes fibrosis
DNA methylation,
IGF, VEGF, NO NO acetylation, mIRNAs Dysregulated Vascular Ability to optimise Electron chain
leading to afferent Long non-coding RNAs autophagy flux pathology substrate ultisation transport
arteriolar dilatation Histone post-translational and mTOR activity AMPK activity
ROS, AGEs, modification
dyslipidaemia ROS
All and ET1 leading to Inflammatory and fibrotic
Activity of Ras, production
efferent arteriolar transcription factors and
constriction proinflammatory and
cytokines Hypoxia/
profibrotic pathways Oxidative stress and
oxidative DND
Podocyte apoptosis inflammation
stress Oxygen oxidation
Mesangial expansion demand and
HIF stability Tubulointerstitial tissue
Glomerular inflammation hypoxia
hypertrophy Inflammation ATP
leading to
and fibrosis production
Glomerular hypertrophy Proteinuria oxidative
Renal inflammation and stress
fibrosis
Glomerular sclerosis Bioenergesis/
vascular
dysfunction
Tubulointerstitial fibrosis
and tubular atrophy
Figure 24.2 Schematic of the processes leading to tubulointerstitial fibrosis and tubular atrophy in DKD.
IGF: insulin-like growth factor
VEGF: vascular endothelial growth factor
NO: nitric oxide
ET1: endothelin-1
ROS: reactive oxygene species Test for moderately
AGEs: advanced glycation end products increased urine protein
miRNAs: microRNA
HIF: hypoxia inducible factor
ATP: adenosine triphosphate No + for albumin
DNA: deoxyribonucleic acid
RNA: ribonucleic acid Yes
Condition that may invalidate
urine albumin excretion?
collection. Early-morning (or indeed random) urine assess- Yes
ment of the albumin-to-creatinine ratio (uACR) is now com-
Treat and/or wait until
monly used as a screening tool for the detection of DKD [88]. No
resolved. Repeat test. No
The random measurement of uACR has an inherent day-to- + for protein?
day variability (30–40%) [91], and may be elevated by factors
Yes
independent of kidney disease, including strenuous exercise
within 24 hours of measurement, urinary tract infection, Repeat moderately
increased urine protein
menstruation, heart failure, marked hyperglycemia, and test twice within 3–6 month period.
drugs, e.g. nonsteroidal anti-inflammatory drugs (NSAIDs).
Therefore, an elevated uACR should be confirmed in the
absence of urinary tract infection with two additional early- Rescreen No
2 of 3 tests positive?
in one year
void specimens collected 3–6 months apart (Figure 24.3).
Although 24-hour collections are not commonly performed Yes
for the screening of DKD, they are useful when there is doubt moderately increased
about the accuracy of an estimated glomerular filtration rate urine protein, begin treatment
(eGFR) (e.g. in populations where eGFR has not been vali-
Figure 24.3 Screening for moderately increased urine
dated such as where muscle mass is low) or when uACRs vary protein recommended by American Diabetic Association
widely. Occasionally, 24-hour urine collections are also used (ADA) [89].
Prevention, Treatment, and Follow-up of Diabetic Kidney Diseas 371
to estimate daily dietary sodium and protein intake in indi- prescence of other microvascular complications ii) moder-
viduals with DKD and refractory edema or hypertension. ately increase proteinuria (previously known as microalbu-
The evidence for the utility of eGFR alone as a screening minuria is present with either diabetic retinopathy, or in
test for CKD in diabetes is unclear. The eGFR can be patients with type 1 DM of at least 10 years’ duration. Other
calculated using either the Modification of Diet in Renal causes of CKD should be considered if any of the features
Disease (MDRD) equation [92] or the Chronic Kidney listed in Table 24.3 is present. A kidney biopsy may be war-
Disease Epidemiology Collaboration (CKD-EPI) for- ranted in some cases for confirmation of diagnosis.
mula [93]. Both equations perform well when the GFR is
<60 ml/min/1.73 m2 [92], but the CKD-EPI is more accu-
rate at higher levels of kidney function [93]. Many patients revention, Treatment, and Follow-up
P
with diabetes and classical “diabetic nephropathy” may of Diabetic Kidney Disease
have high-normal GFRs, particularly in the early stages of
DKD. Therefore, progressive CKD is best determined by Lifestyle and Multifaceted Approach to
the slope of sequential eGFR measurements, rather than a Prevent and Treat DKD
single estimate. As alluded to above, other markers of kid-
ney damage (e.g. albuminuria) are additionally required to Lifestyle interventions such as reduced salt intake, weight
detect early stages of CKD. The eGFR is useful for assessing loss, physical exercise, and supplementation with polyun-
progressive changes in kidney function but should not be saturated fats have been shown to improve glycemic con-
used in situations where kidney function is changing rap- trol, lower blood pressure (BP), reduce albuminuria, and
idly, such as acute kidney injury (AKI) that can be observed change metabolic risk profiles in the general population as
in infection, hypotension or volume depletion. well as in individuals with diabetes and CKD [98–101].
Up to 52% of Japanese [94] and 55% of Australians [95] However, there is no evidence that lifestyle modifications
with type 2 DM had an eGFR of less than 60 ml/min and no reduce the risk of DKD or slow its progression. Dietary
evidence of moderately increased urine protein. In an sodium restriction has been shown to reduce BP and albu-
Italian multicenter study 57% of patients with type 2 DM minuria, and enhance the response to RAAS inhibi-
had an eGFR of less than 60 ml/min and normoalbuminu- tion [102–104]. However, optimal dietary sodium intake in
ria, and had a significant risk of cardiovascular disease DKD remains controversial, with observational studies
(CVD), although this was lower than if albuminuria and showing a benefit [105] or conversely a detrimental effect
kidney impairment coexisted [96]. Porrini et al. have sum- on wellbeing and mortality [104, 106, 107]. In reality, mul-
marized the incidence of normoalbuminuric DKD to vary tiple risk factors are managed concurrently in patients
between 14% and 57% [97]. Hence CKD in patients with with diabetes and kidney disease. In the STENO study,
DM is likely to have multiple etiologies, including classical 80 Danish patients of European descent with type 2 DM
DKD. Despite this, an increase in urinary albumin, a and moderately increased urine protein were randomly
decrease in eGFR or both in individuals with DM should assigned to receive conventional treatment in accordance
reflect a diagnosis of DKD. In patients with DM, CKD with national guidelines and another 80 to receive inten-
should be attributed to diabetes if: i) severely increased pro- sive treatment, with a stepwise implementation of behav-
teinuria (previously known as macroalbuminuria) in the ior modification and pharmacologic therapy that targeted
hyperglycemia, hypertension (RAAS blockade), dyslipi-
Table 24.3 Features indicative of a potential alternative demia, and moderately increased urine protein, along with
diagnosis other than DKD. secondary prevention of CVD with aspirin. The outcomes
were first reported in 2003 with a mean follow-up of
●● Absence of diabetic retinopathy 7.8 years [108, 109]. Patients receiving intensive therapy
●● Low or rapidly decreasing GFR had a significantly lower risk of CVD (hazard ratio [HR]
●● Rapidly increasing proteinuria or nephrotic syndrome 0.47, 95% confidence interval [CI] 0.24–0.73), nephropathy
●● Refractory hypertension (HR 0.39, 95% CI 0.17–0.87), retinopathy (HR 0.42, 95% CI
●● Presence of active urinary sediment, e.g. persistent 0.21–0.86), and autonomic neuropathy (HR 0.37, 95% CI
microscopic hematuria
0.18–0.79) compared to those receiving a standard
Signs or symptoms of other systemic disease
approach. Patients in the intensive arm showed a mean
●●
>30% reduction in GFR within 2–3 months after initiation of

decrease in albuminuria and lower incidence of progres-
●●
an ACE inhibitor or ARB
sion from moderately to severely increased urine protein
GFR: glomerular filtration rate
(but not for the development of kidney failure) [109] and a
ACE: angiotensin-converting enzyme reduction in the composite outcome of CVD events or
ARB: angiotensin II receptor blockers death by about 50% when compared to the group treated
with standard therapy. After 7.8 years, the study continued importance of addressing multiple risk factors needs to be
as an observational follow-up with all patients receiving considered for renoprotection but also for benefits on mor-
treatment as for the original intensive-therapy group, and tality and cardiovascular events, which are more likely to
recently reported a 21.2 years follow-up (n = 22 conven- occur in patients with DKD than is kidney failure [112–114].
tional group, n = 42 intensive group) that shows a median The Kidney Disease Improving Global Outcomes (KDIGO)
of 7.9 years of gain of life mainly driven by time free from Guideline 2020 [115] recommended moderately intensity
incident CVD [110]. Median time before first CV event exercise in people with diabetes and CKD. Recently, a new
after randomization was 8.1 years longer in the intensive- study by Cohen et al [116] compared bariatric surgery (lapa-
therapy group (P = 0.001). Microvascular complications roscopic surgery) versus standard care (non-surgery inter-
were study predefined tertiary endpoints. Progression of vention) in 100 patients with type 2 DM and CKD stages
retinopathy was decreased by 33%, autonomic neuropathy G1-G3, during a follow-up of 2 years. The study reported
was decreased by 41%, and progression to diabetic nephrop- zero events for mortality (both all-cause and cardiovascular)
athy (severely increased urine protein) was reduced by 48% and no cardiovascular events. However, bariatric surgery
in the intensive-therapy group when compared to the con- improved remission of albuminuria (<30 mg/g) in partici-
ventional groups. Ten patients in the conventional-therapy pants who received bariatric surgery compared to standard
groups versus five patients in the intensive-therapy group of care (Relative risk (RR) 1.44 , 95%CI 1.03, 2.02).
progressed to kidney failure (P = 0.061). The nonfatal end-
points and causes of death were adjudicated by an external Glycemic Control
endpoint committee blinded for treatment allocation.
Tight sugar control is critical, and hyperglycemia can cause
The Look AHEAD (Action for Health in Diabetes) multi-
organ complications. Therefore, optimal glycemic control is
centre trial is the largest randomized trial to date evaluat-
integral to the management of DKD. Regardless of the under-
ing the efficacy of a physical activity and dietary control
lying treatment, glycated hemoglobin HbA1c: glycated hae-
intervention (targeting a 7% weight loss) in older obese
moglobin levels >7.0% (53 mmol/mol) are associated with a
adults with type 2 DM. The study randomly assigned 5145
significantly increased risk of both microvascular and CV
obese participants with type 2 DM to either intensive life-
complications [7, 117,118]. These studies show a curvilinear
style intervention (n = 2570) or standard diabetes support
relationship between HbA1c and diabetes complications,
and education (n = 2575). Intensive lifestyle intervention
with no apparent threshold of benefit, although the absolute
was designed to achieve and maintain weight loss through
reduction in risk was much at lower HbA1c levels. Most of
reduced caloric consumption and increased physical activ-
the evidence comes from studies of intensive glycemic con-
ity (a minimum of 175 min/week of unsupervised exer-
trol in people with type 1 and 2 DM and early CKD. The evi-
cise). Major CV event rates were not significantly different
dence for the risks and benefits of intensive glycemic control
in the two groups, which resulted in early termination of
is scant for more advanced stages of CKD, including individ-
the study. After a median of 8.0 years of follow-up, the
uals receiving dialysis or kidney transplant recipients.
study demonstrated slower progression of CKD in the
The Diabetes Control and Complications Trial (DCCT)
intensive lifestyle intervention group than patients in the
comparator group. The incident CKD rate was 0.91 cases in type 1 DM [6], the Kumamoto trial [119], and the
per 100 person-years in the standard group and 0.63 per United Kingdom Prospective Diabetes Study
100 person-years in the intensive lifestyle group (difference (UKPDS) [117,120] in patients with type 2 DM confirmed
0.27 cases per 100 person-years, HR 0.69, 95% CI 0.55–0.87, that improved glycemic control significantly reduced the
P = 0.0016) [111]. The Intensive lifestyle intervention risk of microvascular complications, but had no signifi-
group achieved significantly greater and more sustained cant effect on CV outcomes. Subsequent observational
improvements secondary outcomes, including weight loss, long-term follow-up after termination of both the DCCT
improved cardiorespiratory fitness, improved glycemic and UKPDS studies showed a persistence of significant
control, BP, and lipids with fewer medications, as well as microvascular benefits and also demonstrated the emer-
decreased rate of sleep apnea, retinopathy, depression, sex- gence of a beneficial effect on CV outcomes attributed to
ual dysfunction, urinary incontinence, and knee pain, plus intensive glycemic control. In the DCCT cohort, there
better physical mobility maintenance and quality of life, was a significant reduction in CV outcomes (42%), nonfa-
with lower overall healthcare costs. There were no safety tal myocardial infarct (MI), stroke, and CV death (57%) as
signals suggestive of kidney-related adverse events. well as all-cause mortality (33%) in previously intensively
These studies are limited by the difficulty in attributing treated participants compared with those who were previ-
which facets of the intervention are associated with reduced ously in the standard arm [121–123]. Similarly, there was
risk. Despite the low-level evidence that studies using a mul- a significant reduction in MI (15–33%) and all-cause mor-
tipronged approach to risk-factor management provide, the tality (13–27%) in the UKPDS cohort in participants who
had been originally randomized to intensive failure. The number of participants needed to treat over
treatment [124]. 5 years to prevent one kidney failure event ranged from
The Action to Control Cardiovascular Risk in Diabetes 410 in the overall study to 41 participants with severely
(ACCORD), Action in Diabetes and Vascular Disease: increased urine protein at baseline. Thus, improved glucose
Preterax and Diamicron MR Controlled Evaluation control improved major kidney outcomes in patients with
(ADVANCE), and Veterans Affairs Diabetes Trial (VADT) type 2 DM [129]. However, the small number of kidney fail-
are three major trials that examined the effect of intensive ure events observed during the trial limited the strength of
glycemic control on people with long-standing type 2 DM. the conclusion [129]. There was no reduction in major CV
The ACCORD trial randomly assigned 10 251 participants events or all-cause mortality in the intensive glucose lower-
with type 2 DM who had either an established CVD history ing either during the in-trial period or observational follow-
or multiple risk factors for CVD to intensive therapy target- up of a median 5.4 years [130]. In the post-trial observational
ing an HbA1c <6.0% (42.1 mmol/mol) or standard therapy follow-up ADVANCE-ON study, there was a significant
targeting an HbA1c level of 7.0–7.9% (53–62.8 mmol/ reduction in the risk of kidney failure with intensive glucose
mol) [125]. The mean age of participants was 62 years and lowering that was observed during the in-trial period (7 vs.
the mean duration of diabetes was 10 years. A difference in 20 events, HR 0.35, 95% CI 0.15–0.83, P = 0.02) that persisted
HbA1c was rapidly obtained and maintained throughout after a total of 9.9 years of follow-up (29 vs. 53, HR 0.54, 95%
the trial at 6.4% (46.4 mmol/mol) and 7.5% (58.5 mmol/ CI 0.34–0.85, P < 0.01) [129, 130]. The patients who appear
mol) in the intensive and standard therapy groups, respec- to benefit most from intensive glycemic control are those
tively. The primary composite major CV outcomes (nonfa- with preserved kidney function (early stages of CKD) and
tal MI, nonfatal stroke, or death from CV causes) were not those with better blood pressure control at baseline (systolic
different between two arms (HR 0.90, P = 0.16). The glyce- BP <140 mmHg). This also shows that CKD stages at base-
mic control portion of the trial was prematurely terminated line did not affect the impact of intensive glucose lowering
after 3.5 years due to higher mortality in the intensively on mortality or major cardiovascular events [131].
treated arm. Intensive glycemic control arm showed a 32% Neither study demonstrated a reduction in CV events or
reduction in the development of incident of severely mortality with intensive glycemic control. This raises the
increased urine protein (2.7% vs. 3.9%) and a 21% reduction question of whether optimal HbA1c may differ for micro-
in the development of incident of moderately increased vascular/renoprotection versus cardiovascular events. In
urine protein (12.5% vs. 15.3%) [125, 126]. However, an addition, the safety of intensive glucose control in the pres-
observational follow-up of the surviving participants in the ence of CKD has been questioned, with the ACCORD
ACCORDION study (n = 8601) over a median of 8.8 years trial [132] reporting that its intensive glucose lowering
showed a neutral long-term effect of intensive glucose con- strategy increased the risk of cardiovascular and all-cause
trol on the composite CV outcome and all-cause mortality death among participants with CKD, but not in those with
(HR 1.01, CI 0.92–1.10) [127]. In the same follow-up study, normal kidney function.
intensive glycemic control significantly reduced the devel- A smaller VADT study randomly assigned 1791 United
opment of severely increased urine protein (HR 0.68, CI States military veterans with a mean duration of diabetes
0.59–0.77) but had no impact on doubling of serum creati- of 12 years and poor glycemic control ( 7.5% or 58.5 mmol/l)
nine or need of dialysis [128]. to either standard or intensive glucose lowering therapy,
The ADVANCE study was a 2 × 2 factorial designed rand- which aimed for an overall reduction in HbA1c levels by
omized control trial randomly assigning 11 140 participants 1.5% [133]. The mean duration of diabetes was 12 years and
to standard (targeting HbA1c based on local guidelines) or the HbA1c levels achieved in the intensive and standard
intensive glucose control therapy aimed at reducing HbA1c therapy groups were 6.9% (51.9 mmol/mol) and 8.4%
to 6.5% (47.5 mmol/mol) [23]. Participants were 55 years (68.3 mmol/mol), respectively. During a median follow-up
of age with a history of major CV or microvascular disease or of 5.6 years, there was a nonsignificant reduction in the pri-
at least one other risk factor for CVD. The mean duration of mary outcome (first occurrence of a major CV event), but
diabetes was 8 years. After a 5-year follow-up, the mean the progression to albuminuria was significantly reduced
achieved HbA1c was 6.5% (47.5%) in the intensive group and in the intensive-treated group 17% vs. 35% standard-treated
7.3% (56.3 mmol/mol) in the standard group. The primary group after 2 years (P = 0.05). During an observational
outcome was a composite of microvascular events (nephrop- median follow-up of 9.8 years, the intensive-treated group
athy and retinopathy) and CV disease defined by major had a significantly lower risk of the primary outcome (MI,
adverse CV events. There was a significant reduction in the stroke, new or worsening congestive heart failure [CHF],
incidence of major microvascular events in the intensive amputation for ischemic gangrene, or CV-related death)
arm, largely driven by a 21% relative reduction in nephropa- than did the standard therapy group (HR 0.83, P = 0.04),
thy, including new or worsening nephropathy and kidney with an absolute reduction in risk of 8.6 major CV events
per 1000 person-years [134]. There were no differences in from adequately powered clinical trials is required to deter-
kidney failure between groups in the VADT [133] or mine whether any of these monitoring tools provide superior
VADT-F [134], its long-term follow-up study. to HbA1c in patients with DKD.
Data from a meta-analysis from 58 160 patients in 13 ran-
domized controlled trials suggest that people with type 2
Pharmacotherapy for Glucose Lowering in DKD
DM who receive intensive glucose-lowering therapy have a
reduced risk of the composite major adverse cardiovascular Slowing the progression of DKD has been the subject of
events (MACE) and MI, with no significant effect on the risk both primary and secondary endpoints in studies of
of total mortality, cardiac death, stroke, and CHF [135]. A patients with DM. Initial studies were primarily targeted to
further meta-analysis examined intensive versus standard lower blood glucose and/or blood pressure. Subsequently,
glycemic control in 27 391 adult patiants with type 2 DM and specific studies were undertaken to assess the benefit of
CKD from four randomized controlled trials (RCTs) and interrupting the renin angiotensin system in patients with
found the pooled odds ratio (OR) for the doubling of serum DKD, with or without tight glycemic control. Collectively
creatinine and need of dialysis were not statistically signifi- these studies have established that current best practice
cant, i.e. 0.98, 95% CI 0.81–1.19, and 0.84, 95% CI 0.69–1.02, includes interruption of the renin angiotensin system to
respectively. The pooled OR for the outcome of death from slow progression of kidney disease [143, 144].
kidney failure was 0.62, 95% CI 0.39–0.98 [136]. The requirement of the Food and Drug Administration
Hypoglycemia is more common as lower HbA1c levels are (FDA) to undertake cardiovascular endpoints in drugs
targeted, and people with CKD are at an increased risk of being brought to market for glycemic control has increased
hypoglycemia [137, 138]. In the ACCORD study, hypoglyce- our knowledge of potential kidney benefits in the diabetic
mia in the intensively treated group was associated with an population when specific interventions are used to improve
increase in mortality [125]. The frequency of severe hypogly- glycemic control. Although currently available data reflect
cemia in these trials was two to three times higher in the secondary endpoints, studies adequately powered to dem-
intensive therapy groups and a higher mortality was reported onstrate kidney benefits of agents primarily developed for
in participants with one or more episodes of severe hypogly- glycemic control are underway. More recently therapies
cemia in the ACCORD [30], ADVANCE [31], and VADT tri- have been trialed to interrupt specific pathways considered
als [25], irrespective of the different treatment arms in which to be of key importance in the development of DKD.
individual participants were allocated. Therefore, it has been
suggested that tight glycemic control with a target HbA1c of
Sodium-Glucose Linked
6.0% may not be ideal for older/frail individuals or those
Co-transport-2 Inhibitors
with longer duration of diabetes, advanced coronary artery
disease (CAD), and a known history of severe hypoglycemia. The combined sodium-glucose linked co-transport-1 and
It also highlights individualized glycemic control targeting, -2 (SGLT1 and 2) inhibitor phlorizin was initially isolated
with patient education, and monitoring for hypoglycemic from the bark of apple trees in 1835. It was recognized
episodes should be an important component of manage- soon after that the compound caused glycosuria, polyu-
ment plans in patients with DKD. ria, and weight loss. However, interest in inhibiting tubu-
Monitoring of glycemic control in DKD can be problem- lar reabsorption of glucose as a potential treatment for
atic. HbA1c is most commonly used but is influenced by DM did not emerge till the 1970s with discovery of the
shortened red blood cells (RBCs) survival in patients with location of SGLT transporters in the kidney. Phlorizin
DKD in the uremic millieu [139, 140], use of erythropoietin, was not pursued as a treatment for hyperglycemia due to
carbamylation of hemoglobin [141], and mechanical destruc- poor oral absorption and concurrent blockade of SGLT1,
tion of RBCs on dialysis. Thus, clinicians may often need to resulting in diarrhea and gastrointestinal side effects due
rely more on random or continuous home blood glucose to inhibition of intestinal glucose and galactose absorp-
monitoring, which raises practical and cost challenges. tion. As SGLT2 is responsible for more than 90% of glu-
Alternatives to HbA1c such as fructosamine, glycated cose reabsorption in the proximal tubule of the kidney,
albumin, and 1,5-anhydroglucitol (1,5-AG) have been pro- specific SGLT2 inhibitors were developed. Clinical trials
posed. These biomarkers have been shown to reflect glycemic initially focused on the efficacy of glycemic control in
control in CKD and also post-prandial glucose fluctuations. patients with type 2 DM. Given the glucose-lowering
Serum 1,5-AG may be useful for estimating within-day effects are dependent on glomerular filtration delivering
glucose variations. However, each has its inherent limita- glucose to the tubular lumen, where the SGLT2 transport-
tions [142] and no relationship to the development or limiting ers are situated, initial studies did not include patients
the progression of DKD has been demonstrated. Evidence with more than mild degrees of kidney impairment. Two
large cardiovascular (CV) endpoint trials have been uACR was 18% lower (95% CI 16–20) in participants treated
reported to date in patients with type 2 DM at high risk for with canagliflozin than in those treated with
future cardiovascular events. Both the EMPA-REG [145] placebo [148].
and CANVAS/CANVAS R trials [146] showed an impres- The DECLARE-TIMI58 study compared dapagliflozin
sive reduction in important cardiovascular endpoints, but with placebo in >17 000 individuals with type 2 DM and
importantly secondary analyses of kidney endpoints in variable risk for CVD. The outcomes were reported in a sci-
both studies showed a significant and remarkably similar entific session at the American Heart Association (AHA)
benefit on kidney-specific outcomes in this high-risk pop- meeting in November 2018 and indicate that dapagliflozin
ulation. Specifically, of the 7020 patients studied in is noninferior but not superior for reducing MACE in
EMPA-Reg, approximately 1800 had eGFR <45 ml/min, patients with this cohort. The composite rate of cardiovas-
750 had severely increased urine protein, and 80% were cular death or hospitalization for heart failure was 17%
on renin-angiotensin system blockers. Overall the study lower for patients receiving dapagliflozin compared with
demonstrated a 39% reduction in new-onset or worsening those who received placebo (4.9% vs. 5.8%). Kidney events
of existing nephropathy (HR 0.61, 95%CI 0.53–0.7), a 46% and predefined composite outcomes (defined as new kid-
reduction in the composite outcome of doubling of serum ney failure, eGFR decrease by at least 40% to less than
creatinine, initiation of kidney replacement therapy, and 60 m/min/1.73 m2, or death from kidney or CVD) were 23%
death due to kidney disease in the empagliflozin group lower in the dapagliflozin group (4.3% vs. 5.6%, P < 0.05,
(HR 0.61, 95% CI 0.53–0.7). The impact of empagliflozin compared to placebo) [149].
on the primary endpoint was not diminished in patients The mechanism of action benefit for this class of drug for
with CKD compared to those without it (MACE HR 0.88, both cardiac and kidney outcomes has been widely dis-
CV death HR 0.78, heart failure HR 0.59, all-cause mor- cussed and potential mechanisms are detailed in
tality HR 0.80, no signal of MI, stroke) [147]. Figure 24.4.
Similarly, in the CANVAS study the composite outcome Clearly these results are derived from secondary endpoints
of sustained doubling of serum creatinine, kidney failure, of studies primarily powered to determine CV outcomes.
and death from kidney causes occurred less frequently in The CREDENCE study was designed to determine the effi-
the canagliflozin group compared with the placebo group cacy and safety of canagliflozin versus placebo at preventing
(HR 0.53, 95% CI 0.33–0.84), with consistent findings clinically important kidney and cardiovascular outcomes in
across prespecified patient subgroups. Annual eGFR patients with diabetes and established kidney disease [151].
decline was slower (slope difference between groups CREDENCE randomized 4401 adults with type 2 DM, eGFR
1·2 ml/min/1.73 m2 per year, 95% CI 1.0–1.4) and mean 30 to <90 ml/min/1.73 m2, and severely increased urine
SGLT2 inhibition
Glycosuria Natriuresis Renal and cardiac metabolism
Negative ↑Tubulo-
↓Blood ↓Plasma ATP dependent FFA as an energy
caloric ↓HbA1c ↑Uricosuria glomerular
pressure volume Na transport substrate
balance feedback
Afferent
↓Total body ↓Inflammation ↓Plasma ↓Arterial ↓Myocardial
arteriole hypoxia
fat mass ↓Glucose toxicity uric acid stiffness stretch
constriction
Improved
↓Epicardial ↓Ventricular Activation of
↓Atherosclerosis O2 consumption autophagy/
fat arrhythmias ACE2 – Ang1/7
mitophagy
↓Intraglomerular
↓Inflammation ↑Cardiac
hypertension IFTA
↓Fibrosis contractility
↓Hyperfiltration
Cardiac & Renal Protection
Figure 24.4 Proposed mechanisms of benefit for cardiovascular and kidney outcomes with the use of SGLT2 inhibitors. Source:
Adapted from Rajasekeran et al. [150].
protein (uACR >300 to 5000 mg/g) with a median follow- [153] reported that SGLT2 inhibitors decrease all-cause mor-
up of 2.62 years. The CREDENCE study was terminated tality (17 studies, 31523 participants; HR 0.88, 95% CI 0.82 to
early on the advice of the Data Safety Monitoring Committee 0.95; I2 = 0%, high certainty evidence), cardiovascular mortal-
because of the magnitude of efficacy in the intervention ity (12 studies, 46442 participants; HR 0.90, 95% CI 0.83 to
arm, which found a 30% lower risk of the primary composite 0.98; I2 = 0%, high certainty evidence), major adverse cardio-
endpoint is kidney failure, doubling of serum creatinine, vascular events 3-point (MACE-3P) (10 studies, 40866 par-
and kidney or CV death in the canagliflozin arm compared ticipants; HR 0.88, 95% CI 0.82 to 0.95; I2 = 28%, high certainty
to standard of care (HR 0.70, 95% CI 0.59, 0.82) [151]. evidence), acute kidney injury (AKI) (8 studies, 24187 partici-
Because the mechanism of benefit of SGLT2 inhibitors pants; HR 0.76, 95% CI 0.66 to 0.89; I2 = 0%, high certainty
appears to be somewhat independent of glycemic control, evidence), and kidney composite outcome (8 studies, 34788
several studies have been developed and undertaken to assess participants; HR 0.60, 95% CI 0.52 to 0.70; I2 = 46%, high cer-
whether these drugs have renoprotective benefits in patients tainty evidence), compared to placebo. SGLT2 inhibitors
with type 1 DM and in patients with non-DKD. DAPA-CKD probably increase diabetic ketoacidosis (9 studies, 31442 par-
study [152], compared dapagliglozin 10 mg with placebo in ticipants; HR 2.22, 95% CI 1.18 to 4.16; I2 = 6%, moderate cer-
4304 participants with CKD with or without diabetes. The tainty evidence) and genital infection (13 studies, 14843
study found a decrease in the primary composite outcome of participants; HR 2.6, 95% CI 2.02 to 3.34; I2 = 0%, moderate
sustained decline in eGFR at least 50%, kidney failure, kid- certainty evidence) compared to placebo (Table 24.4).
ney-related or CV mortality in both participants with type 2
DM (HR 0.64, 95% CI 0.52,0.79) and those without diabetes
Glucagon-like Peptide-1 Receptor Agonists
(HR 0.50, 95% CI 0.35, 0.72). Further studies are also planned
in the CKD population without diabetes (EMPA- Kidney, Glucagon-like peptide-1 (GLP-1) agonists have been shown
http://ClinicalTrials.gov, Identifier NCT03594110). A in experimental models to reduce albuminuria, decrease
Cochrane systematic review which is undergoing an update mesangial expansion and glomerular basement membrane
Table 24.4 SGLT2 versus placebo in people with type 2 diabetes mellitus and chronic kidney disease
No. of
Relative effect Absolute effect Participants(No. of Quality of the
Outcomes (95% CI) (95% CI) studies) evidence(GRADE) Conclusions
All-cause mortality HR 0.88 10 fewer per 31523 (17 studies) High SGLT2 inhibitors
Mean follow-up: 37 months (0.82 to 0.95) 1,000(from 16 decrease all-cause
fewer to 4 fewer) mortality
Cardiovascular mortality HR 0.90 7 fewer per 46442 (12 studies) High SGLT2 inhibitors
Mean follow-up: 26 months (0.83 to 0.98) 1,000(from 12 decrease cardiovascular
fewer to 1 fewer) mortality
MACE-3P HR 0.88 15 fewer per 40866 (10 studies) High SGLT2 inhibitors
Mean follow-up: 31 months (0.82 to 0.95) 1,000(from 23 decrease MACE-3P
fewer to 6 fewer)
Kidney composite HR 0.60 39 fewer per 34788 (8 studies) High SGLT2 inhibitors
outcome (0.52 to 0.70) 1,000(from 48 decrease kidney
Mean follow-up: 33 months fewer to 29 fewer) composite outcome
AKI HR 0.76 11 fewer per 24187 (8 studies) High SGLT2 inhibitors
Mean follow-up: 30 months (0.66 to 0.89) 1,000(from 15 decrease AKI
fewer to 5 fewer)
Diabetic ketoacidosis HR 2.22 37 more per 31442 (9 studies) Moderate SGLT2 inhibitors
Mean follow-up: 28 months (1.18 to 4.16) 1,000(from 5 more probably increases
to 92 more) diabetic ketoacidosis
Genital infection HR 2.6 14 more per 14843 (13 studies) Moderate SGLT2 inhibitors
(2.02 to 3.34) 1,000(from 9 more probably increases
Mean follow-up: 18 months to 21 more) genital infections
AKI = Acute kidney injury; CI = Confidence interval; HR = Hazard ratio; MACE-3P = Major adverse cardiovascular events 3-point;
SGLT2 = Sodium-glucose cotransporter-2
thickness, confer endothelial protection, and restore podo- overall reached the composite endpoint of severely
cyte function. The main factors mediating renoprotection increased urine protein, doubling of serum creatinine to
include inhibition of protein kinase B signaling, inhibition reach an eGFR of less than 45 ml/min/m2 or persistent
of NAD(P)H oxidative signaling, increased cAMP and pro- renal replacement therapy. Overall the HR of developing
tein kinase A, and reduction in inflammatory cell infiltra- new or worsening nephropathy was 0.64 (CI 0.46–0.88,
tion and ICAM-1 expression [154, 155]. GLP-1 receptor P < 0.005) [159].
agonists have also been shown to normalize glomerular The ELIXA trial investigated the effects of lixisenatide
hyperfiltration in diabetes through inhibition of atrial in acute coronary syndrome (ACS) [160]. Although no
natriuretic peptide, which limits afferent arteriolar vasodi- benefit was observed in CV outcomes, secondary analyses
lation, and inhibition of endothelin, which limits efferent demonstrated a reduction in severely increased urine pro-
arteriolar vasoconstriction [156]. These positive effects tein in the lixisenatide-treated patients (P < 0.01) with a
have been replicated in clinical trials with lira- [155], modest reduction in moderately increased urine protein
sema- [156], and dula-glutides and lixisenatide all demon- (P < 0.05) [161]. Despite no benefit in eGFR being observed
strating a reduction in albuminuria; liraglutide and dura- with lixisenatide treatment in the overall population, in
glutide have shown a reduction in the decline of eGFR. the subgroup with an eGFR of between 30 and 59 ml/
Both CV and kidney benefits persist in patients with an min/1.73m2 a slowing in the rate of decline in eGFR was
eGFR of <60 ml/min. observed [157].
The secondary kidney outcomes of the LEADER study AWARD-7 assessed the effect of dulaglutide versus insulin
were reported in 2017 [157]. Of the 9340 patients rand- glargine in a study population enriched with DKD (mean
omized 20.7% had an eGFR of between 30 and 59 ml/min eGFR 38 ml/min/1.73 m2). A slowing in the rate of decline of
(stage 3 CKD), 26.3% had moderately increased urine pro- GFR, measured by cystatin C and a reduction in albuminu-
tein, and 10.5% had macroproteinuria. The kidney outcome, ria, was observed in patients treated with either 1.5 or 1.75 mg
i.e. a composite of new-onset persistent severely increased of dulaglutide/day. The benefit for dulaglutide was more
urine protein, persistent doubling of the serum creatinine pronounced in patients with severely increased urine protein
level, kidney failure, or death due to kidney disease, compared to those with moderately increased urine protein.
occurred in fewer participants in the liraglutide group than These benefits were independent of glycemic control. A
in the placebo group (268 of 4668 patients vs. 337 of 4672, greater weight loss and lower incidence of hyperglycemia
HR 0.78, 95% CI 0.67–0.92, P = 0.003). This result was driven were also observed in the dulaglutide-treated patients [162].
primarily by the new onset of persistent severely increased A Cochrane systematic review which is undergoing an
urine protein, which occurred in fewer participants in the update [153]reported that GLP-1 agonists decrease all-cause
liraglutide group than in the placebo group (161 vs. 215 mortality (6 studies, 3363 participants; HR 0.76, 95% CI 0.62
patients, HR 0.74, 95% CI 0.60– 0.91, P = 0.004). Liraglutide to 0.93; I2 = 0%, high certainty evidence), cardiovascular mor-
administration was associated with a significant reduction tality (4 studies, 3138 participants; RR 0.70, 95% CI 0.54 to
in albuminuria, with nearly 25% lower likelihood of devel- 0.92; I2 = 0%, high certainty evidence), probably decrease
opment of severely increased urine protein, and with the MACE-3P (6 studies, 31245 participants; HR 0.83, 95% CI
uACR approximately 20% lower among treated people, 0.75 to 0.93; I2 = 55%, moderate certainty evidence), decrease
regardless of the baseline level of eGFR. No significant kidney composite outcome (2 studies, 4357 participants; HR
effect on eGFR was found with liraglutide, although both 0.83, 95% CI 0.74 to 0.93; I2 = 0%, high certainty evidence),
those receiving and not receiving the drug showed a decline and probably make little or no difference on AKI (2 studies,
in eGFR from approximately 75 to 65 ml/min/1.73 m2 over 2482 participants; HR 0.83, 95% CI 0.62 to 1.11; I2 = 0%,
moderate certainty evidence) compared to placebo or stand-
the 48-month period of observation. In subgroup analyses
ard care (Table 24.5).
presented at the EASD in October 2018, it was demon-
strated that the cardiovascular benefit was more significant
in patients with an eGFR between 30 and 60 ml/min/1.73m2
Dipeptidyl Peptidase 4 Inhibitors
compared to those with higher levels of eGFR [158].
Similarly, in the Trial to Evaluate Cardiovascular and Dipeptidyl peptidase (DPP) inhibitors are increasingly
Other Long-term Outcomes with Semaglutide in Subjects used in the CKD population to achieve optimal glycemic
with Type 2 Diabetes trial (SUSTAIN-6), persistent control. DPP-4 is involved in the development of kidney
severely increased urine protein developed among 2.7% of fibrosis process via catalytic-dependent or -independent
those receiving semaglutide, but among 4.9% of those mechanisms [163]. DPP-4 exhibits several important bio-
receiving placebo. Fewer subjects treated with semaglutide logical functions that predict that inhibition of DPP-4
Table 24.5 GLP-1 versus placebo/standard care in people with type 2 diabetes mellitus and chronic kidney disease
Relative effect Absolute effect No. of Participants Quality of the

Outcomes (95% CI) (95% CI) (No. of studies) evidence(GRADE) Conclusions
All-cause mortality HR 0.76 28 fewer per 3363 (6 studies) High GLP-1 agonists
Mean follow-up: 5 months (0.62 to 0.93) 1,000(from 45 fewer decrease all-cause
to 8 fewer) mortality
Cardiovascular mortality RR 0.70 25 fewer per 3138 (4 studies) High GLP-1 agonists
Mean follow-up: 20 months (0.54 to 0.92) 1,000(from 38 fewer decrease
to 7 fewer) cardiovascular
mortality
MACE-3P HR 0.83 42 fewer per 31245 (6 studies) Moderate GLP-1 agonists
Mean follow-up: 44 months (0.75 to 0.93) 1,000(from 62 fewer probably decrease
to 17 fewer) MACE-3P
Kidney composite outcome HR 0.83 12 fewer per 4357 (2 studies) High GLP-1 agonists
Mean follow-up: 4.6 years (0.74-0.93) 1,000(from 22 fewer decrease kidney
to 0 fewer) composite outcome
AKI HR 0.83 1 fewer per 2482 (2 studies) Moderate GLP-1 agonists
Mean follow-up: 26 months (0.62 to 1.11) 1,000(from 3 fewer probably make little or
to 1 more) no difference on AKI
AKI = Acute kidney injury; CI = Confidence interval; GLP-1 = glucagon-like peptide 1 receptor; HR = Hazard ratio;
MACE-3P = Major adverse cardiovascular events 3-point; RR = Relative risk
should reduce inflammation and fibrosis. More than 30 (SAVOR-TIMI) [184], sitagliptin (TECOS) [185], aloglip-
potential substrates for DPP-4 have been identified, tin (EXAMINE) [186], or linagliptin. The CARMELINA
including GLP-1, glucose-dependent insulinotropic pep- trial compared DPP-4 inhibitor linagliptin against pla-
tide (GIP), high mobility group protein 1 (HMGB1), brain cebo for cardiovascular safety as specified by the FDA.
natriuretic peptide, neuropeptide Y, and meprin [163– The CARMELINA trial is unique as 60% of the partici-
169]. DPP-4 and integrin-β1 interactions induce profi- pants had CKD, with patients enrolled down to an eGFR
brotic cellular signaling and endothelial mesenchymal of 15 ml/min/1.73 m2. Linagliptin did not increase the
transition [170]. DPP4 inhibitors have experimentally risk of CV events or heart failure and hence was deemed
been shown to be renoprotective, with in vitro and ani- safe in individuals with DKD. Although linagliptin
mal data suggestive of inhibition of inflammatory and showed a significant reduced risk for progression to albu-
fibrotic responses that would otherwise drive kidney minuria (HR 0.86, 95% CI 0.78–0.96, P < 0.0034), it had
injury [171–174]. DPP-4 inhibition induces miR-29, a key no impact on the composite kidney endpoint of a sus-
antifibrotic miR, integrin β1, and IFN-γ. Suppression of tained >40% reduction in eGFR, kidney failure, or death
IFN-γ results in the induction of FGF 1, which induces due to kidney disease (HR 1.04, 95% CI 0.89–1.22) [187].
miR-let-7. Increased levels of miR-let-7 suppress the A Cochrane systematic review which is undergoing an
expression of TGFbR1, resulting in the auto amplifica- update [153] reported that DPP-4 inhibitors make little or
tion of miR-29s [163]. no difference on all-cause mortality (9 studies, 11509
Many preclinical studies have confirmed the potential participants; HR 0.92, 95% CI 0.80 to 1.05; I2 = 0%, high
renoprotective effects of DPP-4 inhibitors in multiple kid- certainty evidence), cardiovascular mortality (4 studies,
ney injury models, including diabetic [175–178] and non- 7569 participants; HR 0.98, 95% CI 0.83 to 1.15; I2 = 0%,
diabetic models of nephropathy [179–182]. Thus far, no high certainty evidence), MACE-3P (4 studies, 14441 par-
clinical trial has revealed beneficial effects of DPP-4 inhibi- ticipants; HR 1.00, 95% CI 0.92 to 1.10; I2 = 0%, high cer-
tors on kidney hard outcomes. tainty evidence), kidney composite outcome (2 studies,
Smaller and retrospective clinical studies demon- 9555 participants; HR 1.08, 95% CI 0.89 to 1.31; I2 = 33%,
strated a benefit in favor of DPP4 inhibitors reducing high certainty evidence), and may make little or no differ-
albuminuria [183], but cardiovascular endpoint studies ence on AKI (2 studies, 7112 participants; RR 0.95, 95%
where kidney outcomes have been reported have shown CI 0.73 to 1.24; I2 = 0%, low certainty evidence) compared
no benefit compared to that of placebo for saxagliptin to placebo (Table 24.6).
Table 24.6 DPP-4 versus placebo in people with type 2 diabetes mellitus and chronic kidney disease

All-cause mortality HR 0.92 8 fewer per 11509 (9 studies) High DPP-4 inhibitors make
Mean follow-up: 15 months (0.8 to 1.05) 1,000(from 20 fewer little or no difference
to 5 more) on all-cause mortality
Cardiovascular mortality HR 0.98 1 fewer per 7569 (4 studies) High DPP-4 inhibitors make
to 11 more) on cardiovascular
mortality
MACE-3P HR 1.0 3 more per 14441 (4 studies) High DPP-4 inhibitors make
to 16 more) on MACE-3P
Kidney composite HR 1.08 8 more per 9555 (2 studies) High DPP-4 inhibitors make
outcome (0.89 to 1.31) 1,000(from 11 fewer little or no difference
Mean follow-up: 40 months to 29 more) on kidney composite
outcome
AKI RR 0.95 12 more per 7112 (2 studies) Low DPP-4 inhibitors may
Mean follow-up: 20 months (0.73 to 1.24) 1,000(from 41 fewer make little or no
to 202 more) difference on AKI
AKI = Acute kidney injury; CI = Confidence interval; DPP-4 = Dipeptidyl peptidase 4; HR = Hazard ratio; MACE-3P = Major adverse
cardiovascular events 3-point; RR = Relative risk
Metformin mence on low dose with gradual dose escalation based on

Metformin is a widely used and effective hypoglycemic tolerability. Metformin is associated with vitamin B12 defi-
drug that remains the first line of treatment for individuals ciency and periodic testing of vitamin B12 levels should be
with type 2 DM. The mechanisms underlying improved considered, especially in those with anemia or peripheral
glycemic control are complex and not fully understood. neuropathy [197]. Combination glucose-lowering therapy
Metformin reduces hepatic gluconeogenesis and increases is often recommended by guidelines with insulin,
peripheral glucose sensitivity. Although not tested in clini- DPP-4 inhibitors, GLP-1 receptor agonists or SGLT2 inhib-
cal trials, experimental evidence suggests it may limit kid- itors over other antihyperglycemic agents as these combi-
ney disease via both AMP-activated protein kinase nations are associated with less hypoglycemia and weight
(AMPK)-dependent and AMPK-independent mechanisms gain provided there are no contraindications or barriers to
and by inhibition of mitochondrial respiration [188]. affordability or access [189, 190].
Current guidelines recommend metformin monotherapy Adding metformin to insulin therapy may reduce insulin
should be started when a diagnosis of type 2 DM is made requirements and improve metabolic control in patients
(in different jurisdictions a failure of lifestyle interventions with type 1 DM. Evidence suggests the use of metformin, in
also needs to be demonstrated), unless there are contrain- addition to insulin therapy in type 1 DM, results in reduced
dications [189, 190]. This recommendation is based on its insulin requirements [198], weight reduction [199],
efficacy in lowering HbA1c, its safety profile, affordability, improvement in lipid profiles [200] but no improvement in
negligible risk of hypoglycemia, and lack of weight gain. glycemic control or CV outcomes [198, 200].
Compared to sulfonylureas, metformin monotherapy has Metformin can lead to rises in serum lactate due to con-
comparable HbA1c-lowering effects, glycemic durabil- version of glucose to lactate by the intestinal mucosa and
ity [191], lower risks of hypoglycemia [192], less weight reduced uptake of lactate in the liver [201–203]. The
gain [192, 193], and in observational studies a lower CV molecular weight (MW) of metformin is 129.16 g/
risk [193]. In the UKPDS, metformin was shown to confer mol [204] (MW of creatinine is 113.12 g/mol). Metformin
CV benefit in newly diagnosed overweight included indi- is nonprotein-bound and is primarily eliminated
viduals with type 2 DM [194]. However, conflicting evi- unchanged by the kidneys and therefore CKD is associ-
dence from a meta-analysis of metformin trials has been ated with higher drug levels that could increase the risk
presented [195, 196]. Metformin has well-documented gas- of lactic acidosis [205, 206]. However, significant lactic
trointestinal side effects. Hence, it is recommended to com- acidosis associated with metformin use is rare. A small
nonplacebo-controlled study involving 35 patients on Sulfonylureas

automated peritoneal dialysis with insulin-dependent
Sulfonylureas bind ATP-dependent K+ channels on cell
type 2 DM, receiving 0.5–1 g daily dose of metforminm,
membranes of the pancreatic beta cells which, through
found no incidence of lactic acidosis, although a higher
increasing intracellular calcium, leads to increased insulin
number of individuals receiving metformin had plasma
secretion [221]. Sulfonylurea use in CKD requires careful
lactate levels >2 mmol/l [204]. The FDA recently revised
attention to dosing to avoid hypoglycemia [222, 223].
the label for metformin to reflect its safety in patients
However, if a sulfonylurea is to be added to metformin, gli-
with eGFR 30 ml/min/1.73 m2 [207].
clazide should be considered as first choice as it is associ-
In a meta-analysis of 347 comparative trials and cohort
ated with a lower risk of hypoglycemia [222, 224]. Glyburide
studies there was no evidence to suggest the use of met-
is extensively metabolized in the liver into several active
formin in CKD stage 3b or higher enhances the risk for
metabolites that are excreted by the kidney and is not rec-
lactic acidosis. In fact, no cases of lactic acidosis were
ommended for use in CKD. Glipizide is metabolized by the
identified in 70 490 patient-years of metformin use [208].
liver into several inactive metabolites, and its clearance and
Individual studies have reported crude incidences of 3.3–
elimination half-life are not affected by a reduction in
10.4 cases per 100 000 patient-years [209–212], which is
eGFR, thus dose adjustments in patients with CKD are not
comparable to rates of lactic acidosis in individuals with
necessary. Although glimepiride is associated with less
type 2 DM taking nonmetformin agents [210, 213]. In the
hypoglycemia when compared with glyburide, it has simi-
only randomized control trial, Rachmani et al. randomly
lar rates of hypoglycemia when compared to other sulfony-
assigned 393 patients with type 2 DM with a serum creati-
lureas. No studies have demonstrated a kidney benefit for
nine of 130–200 μmol/l to continue or stop met-
the use of sulfonylureas, with conflicting data on CV events
formin [214]. The authors demonstrated a close
and mortality. A network meta-analyses of 14 464 patients
correlation between serum creatinine and lactate levels
concluded that gliclazide and glimepiride were associated
(r = 0.78, P < 0.001), which was not shown in other stud-
with a lower risk of all-cause and cardiovascular-related
ies [215, 216]. Hung et al. reported metformin use in 3254
mortality compared with glibenclamide [225]. A prospec-
patients with advanced CKD approaching kidney failure
tive cohort study from the UK Clinical Practice Research
(creatinine >530 μmol/l) was not associated with a sig-
Datalink, using the prevalent new metformin user, matched
nificantly greater risk of metabolic acidosis (adjusted HR
1 : 1 with patients adding or switching to sulfonylureas. The
1.30, 95% CI 0.88–1.93) [217]. Richy et al. examined the
study looked at 77 138 metformin new users of whom 25 699
Datalinkage records of 77 601 patients in UK treated with
added or switched to sulfonylureas between 1998 and 2013.
metformin for type 2 DM between 2007 and 2012 [211],
During a mean follow-up of 1.1 years, sulfonylureas were
and found no difference in the incidence of lactic acidosis
associated with an increased risk of myocardial infarction
in patients with more advanced CKD compared to those
(HR 1.26, 95% CI 1.01–1.56), all-cause mortality (HR 1.28,
with normal kidney function. Eppenga et al. investigated
95% CI 1.15–1.44), and severe hypoglycemia (HR 7.60, 95%
a larger cohort of 223 968 patients with type 2 DM pre-
CI 4.64–12.44) compared with metformin monother-
scribed metformin between 2004 and 2012 [212], and
apy [226]. Given the lack of observational data to support a
reported no difference in incidence of lactic acidosis
differentially positive effect in patients with DKD it is
between metformin or nonmetformin users if eGFR
unlikely that sulfonylureas will be further studied.
60 ml/min, but the risk was significantly higher for met-
formin users with an eGFR <60 ml/min (adjusted HR
6.37, 95% CI 1.48–27.5). Interestingly, the risk of lactic
Thiazolidinediones
acidosis of those with eGFR 30–44 ml/min was compara-
ble to those with an eGFR of 45–59 ml/min. Thiazolidinediones (TZDs) are a class of oral hypoglycemic
The existing evidence is largely extrapolated from studies agents that improve glycemic control primarily by decreasing
involving participants with normal or mildly impaired kid- insulin resistance, thus improving the insulin sensitivity of
ney function. There is a lack of objective outcome data to the liver and peripheral tissues. They have also been impli-
guide the use of metformin in the CKD population itself cated in reducing pancreatic beta cell apoptosis, and are pre-
and further studies are needed. For the CKD population, dominantly metabolized by the liver. Despite the demonstrated
the question is not whether to start metformin but whether anti-inflammatory and antifibrotic effects in experimental
to continue. In patients who have been well maintained on models [227–230] and lack of a need for dosage adjustments
metformin, particularly those who are overweight, the evi- in patients with CKD, TZD use is generally avoided in CKD
dence supporting its routine withdrawal upon reaching an due to side effects such as refractory fluid retention, a reported
eGFR of 30 ml/min is equivocal [218–220]. increase in heart failure, and increased fracture risk.
The PROspective pioglitAzone Clinical Trial (PROactive) presented outcomes and deaths from cardiovascular causes
study randomized 5238 high-risk subjects with type 2 DM in all 4447 subjects. Although there were no statistically sig-
with established macrovascular disease to pioglotazone or nificant differences between the rosiglitazone versus con-
matching placebo [231]. In a mean follow-up of trol groups regarding myocardial infarction and death from
34.5 months, the pioglitazone-treated group had fewer sec- CV causes or all-cause, there were more people with heart
ondary endpoints (composite of all-cause mortality, non- failure in the rosiglitazone group than in the control group
fatal myocardial infarction, and stroke) (301 vs. 358, HR (HR 2.15, 95% CI 1.30– 3.57) [235]. The final analysis of the
0.84, 95% CI 0.72–0.98, P = 0.03) when compared with the RECORD study, with a mean 5.5-year follow-up, showed
placebo. The incidence of hospitalization with heart fail- that addition of rosiglitazone to glucose-lowering therapy
ure or mortality rates from heart failure did not differ in people with type 2 DM increased the risk of heart failure
between groups. In a follow-up analysis of the PROactive and fractures, especially in women. The addition of rosigli-
study, although pioglitazone (5.7%) was associated with tazone to the therapeutic regimen of people with type 2 DM
more serious heart failure events during the study than was not associated with reduced risk of overall cardiovascu-
placebo (4.1%) (P = 0.007), pioglitazone was associated lar morbidity or mortality compared with standard glucose-
with lower fatal/nonfatal MI (28% risk reduction [RR], lowering medications. A meta-analysis that included 20 191
P = 0.045) and ACS (37% RR, P = 0.04) compared to pla- subjects with either prediabetes or type 2 DM who had CHF
cebo in the subgroup of participants with significant heart examined the risk of exacerbation of congestive cardiac fail-
failure [232]. A post hoc analysis investigated the relation- ure (CCF) in those taking TZDs compared with control and
ship between CKD and incident CVD in this population as found the risk of CCF was higher in people given TZDs
well as the effects of pioglitazone treatment on recurrent (RR 1.72, 95% CI 1.21–2.4, P = 0.002). There was no hetero-
CVD. CKD, defined as an eGFR <60 ml/min per 1.73 m2, geneity of effect across the seven studies, suggesting that it
was present in 597 (11.6%) of 5154 patients. More patients is a class effect of TZDs. However, a higher rate of CV death
with CKD reached the primary composite endpoint (all- was not observed [236]. There were no kidney outcomes
cause mortality, acute myocardial infarction (AMI), stroke, reported in these studies. A Cochrane systematic review
ACS, coronary/carotid arterial intervention, leg revascu- which is undergoing an update [153] reported that thiazoli-
larization, or amputation above the ankle) than patients dinediones may make little or no difference on all-cause
without CKD (27.5% vs. 9.6%, P < 0.0001) as well as sec- mortality (2 studies, 147 participants; RR 0.35, 95% CI 0.06
ondary composite endpoint (defined above). A greater to 2.23; I2 = 0%, low certainty evidence) and heart failure (2
decline in eGFR was observed in the pioglitazone-treated studies, 123 participants; RR 0.34, 95% CI 0.01 to 8.13;
group (between-group difference 0.8 ml/min/1.73 m2/ low certainty evidence), compared to placebo or control
year) than with placebo [233]. (Table 24.7).
The RECORD study randomized 4458 individuals with
type 2 DM with inadequate control while on metformin or
Blood Pressure Lowering in DKD
sulfonylurea in combination with rosiglitazone showed at
18 months that HbA1c reduction is similar between rosigli- Observational studies show a strong relationship between
tazone added to metformin or sulfonylureas [234]. In a fur- a higher level of BP and an increased risk of kidney
ther interim analysis (3.75 years follow-up) the same group failure and worsening kidney function in DKD. Optimal
Table 24.7 Thiazolidinediones versus placebo/control in people with type 2 diabetes mellitus and chronic kidney disease

All-cause mortality RR 0.35 36 fewer per 147 (2 studies) Low Thiazolidinediones may
Mean follow-up: (0.06 to 2.23) 1,000(from 52 fewer make little or no difference
25 weeks to 68 more) on all-cause mortality
Heart failure RR 0.34 (0.01 11 fewer per 123 (2 studies) Low Thiazolidinediones may
Mean follow-up: to 8.13) 1,000(from 16 fewer make little or no difference
18 months to 114 more) on heart failure
Kidney failure - - - - No studies were found that
looked at kidney failure
CI = Confidence interval; RR = Relative risk

BP control is accepted as an important therapy in the pre- arm were attenuated but remained significant [130].
vention and progression of CKD in diabetes, although the The Hypertension Optimal Treatment (HOT) trial included
optimal targets for BP in patients DKD have not been patients with and without diabetes and compared DBP tar-
determined [237–241]. gets of 90, 85, and 80 mmHg. Post hoc analyses found
The UKPDS study suggested that a target BP of cardiovascular benefit with more intensive targets in the
<150/85 mmHg was associated with a reduction in micro- subpopulation with diabetes [248].
vascular events, including kidney outcomes [194]. In the A meta-analysis of five RCTs enrolled 7312 adults with
Systolic Hypertension in Europe (Syst-Eur) trial, a target sys- type 2 DM comparing intensive BP targets (upper limit of
tolic BP of <150 mmHg was associated with a lower inci- 130 mmHg systolic and 80 mmHg diastolic) with standard
dence of proteinuria among those with diabetes and in the targets (upper limit of 140–160 mmHg systolic and
overall study group was associated with fewer people devel- 85–100 mmHg diastolic) and found no significant reduc-
oping a creatinine >177 mmol/l [242]. The Appropriate tion in mortality or nonfatal MI. There was a statistically
Blood Pressure Control in Diabetes (ABCD) normotensive significant 35% RR reduction in stroke with intensive tar-
study found that achieving a systolic BP of <130 mmHg was gets, but the absolute risk reduction was only 1% and inten-
associated with fewer people developing moderately sive targets were associated with an increased risk for
increased urine protein and reduced risk of progression adverse events such as hypotension and syncope [249]. The
from moderately to severely increased urine protein [243]. A threshold of BP lowering in diabetics with CKD remains
study in type 1 DM found that a target mean arterial pres- controversial [250].
sure of 92 mmHg (125/75) was associated with a reduction
in proteinuria [244].
Pharmacotherapy for BP Lowering
RAAS Blockade
Intensive Versus Standard BP Lowering in DKD
RAAS blockade using various drugs, including angiotensin
The ACCORD BP trial examined whether an spontaneous converting enzyme (ACE) inhibitors, angiotensin receptor
bacterial peritonitis (SBP) of <120 mmHg in patients with blockers (ARBs), direct renin inhibitors, and mineralocor-
type 2 DM at high CV risk provided greater cardiovascular ticoid receptor antagonists have shown efficacy in animal
protection than an SBP of 130–140 mmHg [245]. The study models of DKD. In humans, RAAS inhibition has proved to
did not find a benefit in the primary endpoint (i.e. nonfatal be the single most effective therapy for slowing the pro-
MI, nonfatal stroke, and cardiovascular death) comparing gression of DKD.
intensive BP treatment (mean BP achieved 119/64 mmHg The CAPTOPRIL study was the first large trial to exam-
on 3.4 medications) with standard treatment (mean BP ine the effect of ACE inhibitors on the progression of
achieved 143/70 mmHg on 2.1 medications). Stroke was sig- advanced DKD. The study randomly assigned 409 patients
nificantly reduced in the intensive BP treatment groups. with type 1 DM, overt proteinuria (protein excretion
The ACCORD BP study found less progression of proteinu- 500 mg/day), and reduced kidney function to captopril
ria in the intensive BP lowering group, but this was associ- 25 mg three times a day or matching placebo. There was a
ated with increased risk of AKI events [245]. A follow-up 48% reduction in risk for doubling of serum creatinine
analysis found the intensive BP/intensive glycemia, inten- concentration and a 50% reduction in the composite end-
sive BP/standard glycemia, and standard BP/intensive gly- point of death, dialysis therapy, or transplantation [251].
cemia groups all showed benefit for reducing the risk of This trial established the efficacy of ACE inhibitors inde-
major CVD when compared to the standard glycemia/ pendent of BP control in slowing the progression of DKD
standard BP control group in the BP trial [246]. in patients with type 1 DM and overt proteinuria. The
In ADVANCE, the intensive BP intervention arm (a other trials using RAAS blockade in patients with type 1
single-pill, fixed-dose combination of perindopril and inda- DM including the (Renin-Angiotensin System Study
pamide) showed a significant reduction in the risk of the [RASS] [252], Diabetic Retinopathy Candesartan Trial
primary composite endpoint (major macrovascular or [DIRECT] -Prevent 1, and DIRECT-Protect 1) [253] failed
microvascular event) and in the risk of death from any to show that RAAS blockade prevented the development
cause and of death from cardiovascular causes [247]. of moderately increased urine protein in those with nor-
Compared with the placebo group, the patients in the inten- moalbuminuria at baseline.
sive BP lowering arm experienced an average reduction of The Effect of Irebesrtan in the Development of Diabetic
5.6 mmHg in SBP and 2.2 mmHg in diastolic blood pressure Nephropathy in Patients With T2DM (IRMA-2) trial ran-
(DBP). The ADVANCE-ON study reported that the reduc- domly assigned 590 patients with type 2 DM and moder-
tions in the risk of all-cause mortality in the intensive ately increased urine protein to irbesartan 150 mg daily,
irbesartan 300 mg daily or matching placebo with follow-up the study was not powered to assess this as an endpoint. In
for 2 years to determine the effect of irbesartan in preven- recent years, new studies have been conducted in people
tion of new onset of overt proteinuria. Irbesartan reduced with DKD, comparing RAAS blockade with placebo or no
the risk for the development of overt proteinuria (defined as treatment.A Cochrane systematic review which is under-
albumin excretion >200 mg/day) in a dose-dependent ben- going an update [258] reported that ACE inhibitors may
efit [254]. The Irbesartan in Diabetic Nephropathy Trial make little or no difference on all-cause mortality (24 stud-
(IDNT) [144] and the Reduction of Endpoints in Non ies, 7413 participants; RR 0.91, 95% CI 0.73 to 1.15; I2 =
Insuline Dependent Diabetes Mellitus (NIDDM) with the 23%, low certainty evidence), fatal or nonfatal stroke (2 stud-
Angiotensin II Antagonist Losartan (RENAAL) [143] study ies, 4944 participants; RR 1.02, 95% CI 0.80 to 1.31; I2 = 0%,
independently showed irbesartan and losartan, respec- low certainty evidence) and fatal or nonfatal myocardial
tively, can delay the progression of DKD in patients with infarction (3 studies, 5100 participants; RR 0.79, 95% CI
type 2 DM, overt proteinuria, and reduced kidney function. 0.57 to 1.09; I2 = 0%, low certainty evidence) compared to
IDNT randomly assigned 1715 participants to irbesartan, placebo or no treatment. The effects of ACE inhibitor on
amlodipine, or placebo with follow-up for a mean of cardiovascular mortality and doubling of serum creatinine
2.6 years. BP was targeted at <135/85 mmHg and was were very uncertain (Table 24.8).
achieved with agents in classes other than those under The same Cochrane review reported that ARB may make
study. Independent of BP control, irbesartan reduced the little or no difference on all-cause mortality (11 studies, 4260
risk for the composite outcome of doubling of serum creati- participants; RR 0.99, 95% CI 0.85 to 1.16; I2 = 0%, low cer-
nine concentration, kidney failure, or death as compared to tainty evidence), and has uncertain effects on cardiovascular
amlodipine or placebo. The RENAAL trial followed up 1513 mortality (6 studies, 878 participants; RR 3.36, 95% CI 0.93 to
patients with type 2 DM and overt proteinuria for a mean of 12.07; low certainty evidence), fatal or nonfatal stroke (2 stud-
3.4 years and demonstrated that losartan 100 mg daily was ies, 619 participants; RR 0.76, 95% CI 0.32 to 1.77; I2 = 0%,
superior to placebo to reduce the risk for the same compos- low certainty evidence) and fatal or nonfatal myocardial
ite endpoint as in IDNT [143]. Taken together, these studies infarction (2 studies, 619 participants; RR 0.43, 95% CI 0.11
provide robust evidence supporting the benefit independ- to 1.65; low certainty evidence), and may prevent doubling of
ent of BP lowering of RAS-blocking medication on slowing serum creatinine (4 studies, 3280 participants; RR 0.84, 95%
the progression of DKD. Despite the dramatic benefit with CI 0.72 to 0.97; I2 = 37%, low certainty evidence) compared to
ARB therapy, many participants on ARB therapy still ended placebo or no treatment (Table 24.9).
up with kidney failure, therefore novel therapy and drug
development is required for prevention of progressive DKD. Combination Therapy
In patients with type 2 DM with early and incipient In the Antihypertensive and Lipid Lowering Treatment to
nephropathy, mixed results were observed. In the Heart Prevent Heart Attack (ALLHAT) trial subgroup, 13 101
Outcomes Prevention Evaluation (HOPE) trial, the use of participants with type 2 DM were randomly assigned to
ramipril was not effective in preventing new onset microal- chlorthalidone, amlodipine (dihydropyridine calcium
buminuria [255]. The Bergamo Nephrologic Diabetes channel blocker [CCB]) or lisinopril. Despite the
Complications Trial (BENEDICT) study randomly assigned chlorthalidone-treated group having a lower systolic BP
patients to one of four arms (placebo, trandolapril, vera- when compared with lisinopril or amlodipine, no differ-
pamil, or trandolapril plus verapamil) for at least 3 years ence in the primary endpoint of combined fatal coronary
with a goal BP <120/80 mmHg. The two arms containing heart disease or nonfatal or fatal MI (HR 0.97, 95% CI
trandolapril showed a benefit in preventing the develop- 0.86–1.10) was seen between amlodipine and chlortha-
ment of albuminuria, with post hoc analyses suggesting lidone. The Avoiding Cardiovascular events through
that the effect was independent of BP reduction [256]. The Combination therapy in Patients Living with Systolic
Randomized Olmesartan and Diabetes Microalbuminuria Hypertension (ACCOMPLISH) trial compared benazepril/
Prevention (ROADMAP) trial followed up 4449 partici- amlodipine combination treatment with benazepril/thi-
pants for a median of 3.2 years. There was a statistically azide therapy [259]. The trial enrolled 6946 high-risk par-
significant difference in BP between the olmesartan and ticipants with type 2 DM and 2842 participants with high
placebo arms. The primary analysis of the trial showed that CV and kidney risk. The primary endpoint was a compos-
olmesartan prevented or delayed the onset of moderately ite of MI, stroke, CV death, hospitalization for angina,
increased urine protein, with moderately increased urine resuscitated cardiac arrest, and coronary revasculariza-
protein developing in 8.2% versus 9.8% of participants tion. Benazepril/amlodipine reduced occurrence of the
(olmesartan vs. placebo). However, the olmesartan group primary event compared to benazepril/thiazide in all sub-
had lower BPs and an increase in CV deaths [257], although jects with diabetes (8.8% vs. 11%; HR 0.79, 95%
Table 24.8 ACEi versus placebo/no treatment in people with diabetic kidney disease.

All-cause mortality RR 0.91 11 fewer per 7413 (24 studies) Low ACEi may make little or no
Mean follow-up: 2.9 years (0.73 to 1.15) 1000 (34 fewer difference on all-cause mortality
to 19 more) compared to placebo/no treatment
Cardiovascular mortality RR 0.67 16 fewer per 5625 (17 studies) Very low It is very uncertain whether ACEi
Mean follow-up: 2.3 years (0.13 to 3.57) 1000 (43 fewer makes any difference on
to 126 more) cardiovascular mortality compared
to placebo/no treatment
Doubling of serum RR 0.69 13 fewer per 6702 (8 studies) Very low It is very uncertain whether ACEi
creatinine (0.47 to 1.01) 1000 (23 fewer makes any difference on doubling
Mean follow-up: 4 years to 0) of serum creatinine compared to
placebo/no treatment
Fatal or nonfatal stroke RR 1.02 1 more per 4944 (2 studies) Low ACEi may make little or no
Mean follow-up: 7 years (0.80 to 1.31) 1000 (9 fewer difference on fatal or nonfatal
to 15 more) stroke compared to placebo/no
treatment
Fatal or nonfatal RR 0.79 6 fewer per 5100 (3 studies) Low ACEi may make little or no
myocardial infarction (0.57 to 1.09) 1000 (13 fewer difference on fatal or nonfatal
Mean follow-up: 3.2 years to 3 more) myocardial infarction compared to
placebo/no treatment
ACEi = Angiotensin converting enzyme inhibitors; CI = Confidence interval; RR = Relative risk
Table 24.9 ARB versus placebo/no treatment in people with diabetic kidney disease.

All-cause mortality RR 0.99 (0.85 1 fewer per 4260 (11 studies) Low ARB may make little or no
Mean follow-up: 2.2 years to 1.16) 1000 (20 fewer difference on all-cause mortality
to 22 more) compared to placebo/no treatment
Cardiovascular mortality RR 3.36 (0.93 17 more per 878 (6 studies) Low ARB have uncertain effects on
Mean follow-up: 1.6 years to 12.07) 1000 (0 to cardiovascular mortality compared
78 more) to placebo/no treatment
Doubling of serum RR 0.84 (0.72 45 fewer per 3280 (4 studies) Low ARB may prevent doubling of
creatinine to 97) 1000 (79 fewer serum creatinine compared to
Mean follow-up: 2.9 years to 8 fewer) placebo/no treatment
Fatal or nonfatal stroke RR 0.76 (0.32 9 fewer per 619 (2 studies) Low ARB have uncertain effects on
Mean follow-up: 2.8 years to 1.77) 1000 (26 fewer fatal or nonfatal stroke compared
to 30 more) to placebo/no treatment
Fatal or nonfatal RR 0.43 (0.11 13 fewer per 619 (2 studies) Low ARB have uncertain effects on
myocardial infarction to 1.65) 1000 (20 fewer fatal or nonfatal myocardial
Mean follow-up: 2.8 years to 15 more) infarction compared to placebo/no
treatment
ARB = Angiotensin receptor blockers; CI = Confidence interval; RR = Relative risk
CI 0.68–0.92). In an intention-to-treat post hoc analysis, more frequent in the benazepril plus amlodipine group
lower events of CKD progression in the benazepril plus than in the benazepril plus hydrochlorothiazide group.
amlodipine group compared to the benazepril plus hydro- Based on this evidence, for persons in whom combination
chlorothiazide group (HR 0·52, 95% CI 0.41–0.65, therapy with an ACE inhibitor is being considered, a dihy-
P < 0·0001) [260]. Although the peripheral oedema is com- dropyridine CCB is preferable to a thiazide/thiazide-like
mon among both groups with CKD, angioedema was diuretic.
Combination RAAS blockers or the addition of a renin DKD and secondary endpoint for FIGARO-DKD study). A
inhibitor such as alliskerin to an ACE inhibitor or angio- Cochrane systematic review which is undergoing an update
tensin 2 receptor blocker has not been a positive strategy to [270] showed that mineralocorticoid receptor antagonist
reduce progressive loss of kidney function. On the con- probably makes little or no effect on all-cause mortality (8
trary, three large RCTs (ONTARGET, VA-NEPHROND, studies, 14293 participants; RR 0.90, 95% CI 0.80 to 1.00; I2 =
and ALTITUDE) have shown such interventions increase 0%, moderate certainty evidence), cardiovascular mortality (4
risk of AKI and hyperkalemia in individuals with studies, 13366 participants; RR 0.88, 95% CI 0.76 to 1.02; I2 =
DKD [261–263]. Although the combination of agents that 0%, moderate certainty evidence), myocardial infarction (3
block the RAAS are generally not recommended in the studies, 13080 participants; RR 0.92, 95% CI 0.75 to 1.13; I2 =
management of DKD, a network meta-analysis has also 0%, moderate certainty evidence), stroke (5 studies, 14259
identified dual RAAS blockade as a promising potential participants; RR 0.99, 95% CI 0.82 to 1.2; I2 = 0%, moderate
therapy if it can be offered safely [264]. certainty evidence) and kidney failure (3 studies, 13080 par-
ticipants; RR 0.86, 95% CI 0.73 to 1.01; I2 = 0%, moderate
certainty evidence) compared to placebo or no treatment
Miniralcorticoid Receptors Inhibitor
(Table 24.10).
Aldosterone blockade reduces CV mortality in patients with
heart failure and reduces albuminuria [265–267]. Potential
Lipid Lowering
mechanisms of action include promotion of salt and water
loss, potentially improving volume and BP management, Dyslipidemia is common in people with diabetes and CKD.
with antifibrotic and antioxidant effects demonstrated in Individuals with diabetes and CKD typically have lower
animal studies. The third-generation mineralocorticoid levels of high-density lipoprotein (HDL) cholesterol, high
receptor antagonist finerenone has been shown to reduce triglyceride, and low-density lipoprotein cholesterol (LDL-
albuminuria DKD at 90 days without significant safety con- C) levels, hence a more atherogenic profile. The evidence
cerns. The phase 2 study involved 821 participants with type that low LDL-C concentration reduces the risk of CV
2 DM, of whom 36.7% had albuminuria (uACR 300 mg/g) events in patients with diabetes and early CKD generally
and 40.0% had an eGFR of 60 ml/min/1.73 m2. Finerenone reflects the evidence reported in the general population.
demonstrated a dose-dependent reduction in uACR. The The Cholesterol Treatment Trialists’ (CTT)
prespecified secondary outcome of hyperkalemia leading to Collaboration meta-analysis of >170 000 subjects receiv-
discontinuation was not observed in the placebo and lowing statin found that for every 1.0 mmol/l reduction in
dose finerenone groups, but showed an increasing incidence LDL-C: low-density lipoprotein-cholesterol there was an
with higher doses [267]. The impact of adding a mineralo- approximately 20% reduction in CVD events, regardless
corticoid receptor antagonist to background standard of care of baseline LDL-C [271]. The effects of statins on all
including an ACE inhibitor or ARB is being evaluated in the fatal and nonfatal CV outcomes were similar for partici-
Efficacy and Safety of Finerenone in Subjects With Type 2 pants with or without diabetes (14 trials, >18 000
Diabetes Mellitus and Diabetic Kidney Disease (FIDELIO- participants with diabetes) [272]. The updated CTT
DKD) [268] and Efficacy and Safety of Finerenone in meta-analysis of 170 000 participants showed that addi-
Subjects With Type 2 Diabetes Mellitus and the Clinical tional reductions in LDL-C (down to approximately 1.0–
Diagnosis of Diabetic Kidney Disease (FIGARO-DKD) trials 2.0 mmol/l) with more intensive therapy further reduced
[269]. The pre-specified meta-analysis FIDELITY, which the incidence of major vascular events and that these
included FIDELIO-DKD and FIGARO-DKD trials pre- reductions could be achieved safely, even in individuals
sented at the European Society of Cardiology 2021 Congress with lower baseline LDL-C levels [273]. The IMproved
demonstrated that the non-steroidal mineralocorticoid Reduction of Outcomes: Vytorin Efficacy International
receptor antagonist finerenone in addition to ACE inhibi- Trial (IMPROVE-IT) showed that the addition of
tion or ARB in patients with CKD and type 2 DM decreased ezetimibe to simvastatin in participants with recent ACS
4-point MACE (HR 0.86, 95% CI 0.78, 0.95) compared to pla- provided CVD event benefit compared to use of simvas-
cebo. Other non-steroidal mineralocorticoid receptor antag- tatin alone. The event reductions were particularly evi-
onists (apararenone, esaxerenone) have been examined in dent in people with type 2 DM [274].
smaller non kidney and CV outcome based trials. The end- The data from more advanced CKD are largely derived
points of this study consist of a composite of first occurrence from the Study of Heart and Renal Protection (SHARP)
of the composite endpoint of onset of kidney failure, a sus- trial [271]. SHARP randomized 9438 participants >40 years
tained decrease of eGFR 40% from baseline over at least old with CKD (mean eGFR of 27 ml/min/1.73 m2) to receive
4 weeks, and kidney death (primary endpoint for FIDELIO- simvastatin 20 mg plus ezetimibe 10 mg daily or placebo and
Table 24.10 Mineralocorticoid versus placebo in people with diabetes mellitus and chronic kidney disease.
Relative effect Absolute effect| No. of Participants Quality of the

All-cause mortality RR 0.90 9 fewer per 14293 (8 studies) Moderate Mineralocorticoid receptor
Mean follow-up: 20 months (0.80 to 1.00) 1,000(from 18 antagonist probably makes
fewer to 0) little or no effect on all-cause
mortality
Cardiovascular mortality RR 0.88 7 fewer per 13366 (4 studies) Moderate Mineralocorticoid receptor
fewer to 1 more) little or no effect on
cardiovascular mortality
Myocardial infarction RR 0.92 2 fewer per 13080 (3 studies) Moderate Mineralocorticoid receptor
Mean follow-up: 12 months (0.75 to 1.13) 1,000(from 7 antagonists probably makes
fewer to 4 more) little or no difference on
myocardial infarction
Stroke RR 0.99 0 per 1,000 14259 (5 studies) Moderate Mineralocorticoid receptor
Mean follow-up: 16 months (0.82 to 1.2) (from 5 fewer antagonist probably makes
to 6 more) little or no effect on stroke
Kidney failure RR 0.86 6 fewer per 13080 (3 studies) Moderate Mineralocorticoid receptor
fewer to 0) little or no effect on kidney
failure
CI = Confidence interval; RR = Relative risk
followed them for 5 years. Thirty-three percent of the tal myocardial infarction, or nonfatal stroke, except that
patients (n = 3023) were receiving maintenance dialysis at there was an increase in the risk of fatal stroke (RR 2.03,
randomization and 23% (n = 2094) of the participants had 95% CI 1.05–3.93) among those receiving atorvastatin.
diabetes, with equal proportions in the simvastatin plus Atorvastatin reduced the rate of all cardiac events com-
ezetimibe and placebo groups. The study showed statin plus bined (RR 0.82, 95% CI 0.68–0.99, P = 0.03, nominally sig-
ezetimibe therapy was associated with a 17% RR in the risk nificant), but not all cerebrovascular events or combined or
of the major atherosclerotic events (coronary death, myocar- total mortality. In another study in individuals receiving
dial infarction, nonhemorrhagic stroke, or any revasculari- hemodialysis, the AURORA study [276], 2776 participants
zation) compared with placebo (HR 0.83, 95% CI 0.74–0.94). (approximately 20% diabetic) were randomized to rosuvas-
This finding was attributable in large part to significant tatin 10 mg or matching placebo daily, with a mean follow-
reductions in nonhemorrhagic stroke and arterial revascu- up of 3.2 years. A similar effect of rosuvastatin was shown
larization procedures. There was no reduction in the risk of on CV primary outcomes. There was a small but statisti-
all-cause mortality, and among the patients with CKD not cally significant increase in the incidence of hemorrhagic
treated by dialysis at randomization (n = 6247), treatment stroke in the diabetic patients who received rosuvastatin,
with simvastatin plus ezetimibe did not reduce the frequency an observation also reported in the 4D study. A post hoc
of doubling of the baseline serum creatinine concentration analysis of the 731 diabetic patients receiving hemodialysis
or progression to kidney failure. Although the study was not reported a 32% risk reduction of composite primary end-
powered to reliably estimate the effect of treatment on pri- points among patient with diabetes on hemodialysis in the
mary outcomes among clinical subgroups, the proportional treated arm (HR 0.68, 95% CI 0.51–0.90) [277].
effect on major atherosclerotic events did not appear to dif- There were three randomized trials that examined the
fer between those with or without diabetes. effect of fibrates relative to placebo in patients with diabetes
The 4D study [275] investigated 1255 subjects with type 2 and CKD. The Veterans Affairs High-density lipoprotein
DM receiving maintenance hemodialysis who were ran- Intervention Trial (VA-HIT) [278] found evidence that gem-
domly assigned to receive 20 mg of atorvastatin or match- fibrozil reduces risk of major cardiovascular events (fatal
ing placebo daily. During a median follow-up period of coronary heart disease, nonfatal MI, and stroke) by 42%
4 years, atorvastatin had no significant effect on the pri- compared with placebo (RR 0.58, 95% CI 0.38–0.89) in a post
mary endpoint (RR 0.92, 95% CI 0.77–1.10), defined as hoc analysis of 297 individuals with low eGFR (GFR <75 ml/
composite of death from cardiac causes, fatal stroke, nonfa- min/1.73 m2) and diabetes. The Diabetes Atherosclerosis
Intervention Study (DAIS) [279] and the Fenofibrate participants with CKD had mild CKD. There are potential
Intervention and Event Lowering in Diabetes (FIELD) harms associated with antiplatelet/antithrombotic agents.
study [280, 281] reported that fenofibrate treatment signifi- Given the very high rates of thrombotic and embolic events
cantly lowered the risk of developing new onset of moder- (venous and arterial), particularly in advanced CKD, more
ately increased urine protein compared with placebo (RR studies are needed to investigate the longer-term benefits of
0.87 in patients with type 2 DM, 95% CI 0.77–0.97). these therapies in diabetes with CKD.
Fenofibrate also promoted regression from moderately
increased urine protein to normoalbuminuria (RR 1.15, 95%
Promising Biomarkers in Predicting Risk of DKD
CI 1.04–1.28, n = 2260, two trials), but did not change the
risk of progression from moderately to severely increased Not all patients with albuminuria progress to kidney fail-
urine protein (RR 1.07, 95% CI 0.43–2.63, one trial, n = 97). ure or CVD, and the same is true of many patients with
Placebo-controlled trials evaluating the addition of the impaired kidney function (eGFR <60 ml/min/1.73 m2),
PCSK9 inhibitors evolocumab and alirocumab to maxi- therefore an intensive search for novel biomarkers in blood
mally tolerated doses of statin therapy in participants who or urine that could improve diagnostic and prognostic pre-
were at high risk for atheroscletoric CVD demonstrated an cision in early or later stages of DKD has been ongoing dur-
average reduction in LDL cholesterol of 36–59%. These ing the past decades.
agents can be considered as adjunctive therapy for patients In a longitudinal cohort study of 410 patients with type 2
with diabetes at high risk for or who require but are intoler- DM with 12 years of follow-up, 59 patients developed kidney
ant to high-intensity statin therapy [282, 283]. The effects failure and 84 patients died without kidney failure. Baseline
of this class of agents on CVD outcomes in patients with circulating tumor necrosis factor receptor (TNFR1) and
CKD remain unknown. TNFR2 were associated with risk for kidney failure. The
cumulative incidence of kidney failure for patients in the
highest TNFR1 quartile was 54% after 12 years but only 3%
Antiplatelet Agents
for the other quartiles (P < 0.001). In Cox proportional haz-
The use of antiplatelet and antithrombotic agents in patients ard analyses, TNFR1 predicted risk for kidney failure in both
with DKD or CKD for prevention of CVD has not been proteinuric and nonproteinuric groups [289]. Similar find-
robustly studied. The Antithrombotic Trialists’ (ATT) col- ings were reported in a cohort of American Indians with
laborators conducted a patient-level meta-analysis of the six type 2 DM in which elevated serum concentrations of
large trials of aspirin for primary prevention in the general TNFR1 or TNFR2 were associated with increased risk of
population. These trials included 95 000 participants, includ- kidney failure after accounting for traditional risk factors
ing almost 4000 with diabetes. The study found aspirin including uACR and eGFR [290].
reduced the risk of serious vascular events by 12% (RR 0.88, Using propensity scoring methodology the ADVANCE
95% CI 0.82–0.94), and the largest reduction was for nonfatal cohort was divided into those who developed major kidney
MI, with little effect on CVD death (RR 0.95, 95% CI 0.78– endpoints (n = 102) compared to the control (n = 179). This
1.15) or total stroke [284]. A post hoc analysis of the HOT subsidiary analysis showed the baseline total and active
trial indicated net benefit for prevention of CV events for TGF-β1 were substantially increased, and bone morphoge-
aspirin in patients with CKD and high BP, but predomi- netic protein 7 (bone morphogenetic factor (BMP)-7) levels
nantly included patients with relatively mild CKD [248]. decreased in people with type 2 DM who subsequently
Evidence supports use of either ticagrelor or clopidogrel in developed kidney endpoints compared to matched con-
combination with aspirin for at least 1 year in patients fol- trols, despite relatively normal baseline kidney function in
lowing an ACS if no percutaneous coronary intervention both groups. The c-statistic was 0.94 for circulating BMP-7
was performed, and clopidogrel, ticagrelor, or prasugrel if a combined with total and active TGF-β1 compared to that of
percutaneous coronary intervention was performed [285]. eGFR and uACR combined (c-statistic of 0.73) [291]. Hara
In patients with diabetes and prior myocardial infarction, et al. have shown that anti-erythropoietin receptor (EPOR)
adding ticagrelor to aspirin significantly reduces the risk of antibodies were found to be a promising marker in
recurrent ischemic events, including cardiovascular and 112 Japanese patients with type 2 DM with CKD with a
coronary heart disease death [286]. Post hoc analyses of tri- mean follow-up period of 45 months. The event-free rate of
als of other antiplatelet agents have raised questions of kidney events (defined as kidney failure requiring dialysis)
whether the balance of benefit and harm is the same in in the patients with antibodies was significantly lower than
patients with diabetes and/or CKD [287] and whether some in those without. The presence of EPOR antibodies confers
agents may be more effective in CKD than others [288]. an HR of 2.78 of kidney failure in individuals with
These results need to be interpreted with caution, as most DKD [292]. Promising, large, prospective validation cohort
studies conducted in conjunction with clinical trial to con- Table 24.11 Novel therapeutics being trialed in DKD.
firm the potential clinical use of these circulating biomark-
ers is required before routine testing can be recommended Aldosterone receptor antagonist Phase 2 study completed
in clinical practice.
Bardoxolone TSUBAKI Study Phase 2 study recruiting
methyl BEACON Phase 3 study terminated
Novel Interventions Under Current or Recent due to safety concerns
Investigation in DKD C–C chemokine receptor type 2 Phase 2 study completed
(CCR2) antagonism
Given the escalating numbers of patients with DKD, new
therapies are constantly in development. A snapshot of Dapagliflozin Phase 4 study recruiting
treatments in early and late phase trials is shown in Endothelin-A antagonist Phase 3 ceased due to futility
atrasentan
Table 24.11; however, results are not yet available.
Exenatide Phase 4 study active but not
recruiting
Conclusion Mineralocorticoid receptor Phase 2 study completed and
antagonist/finerenone Phase 3 underway
The growing global burden of DKD needs urgent attention
Phase 3 study recruiting
to identify novel treatment strategies to prevent progressive
kidney failure and its complications. Several proven treat- Nox1/4 inhibitor (oral Phase 2 study completed:
GKT137831) negative results in type 2
ments supported by evidence are available, but the gap in DM@ with DKD#
delaying the progression of DKD warrants discovery of Now being studied in type 1
novel therapies. The diagnosis of DKD, which traditionally DM with DKD
relies on measuring and monitoring of uACR and kidney Phosphodiesterase 5 inhibitor Phase 2 study completed
function in combination with clinical assessment, may not Pirfenidone Phase 3 study recruiting
identify individuals at risk of progressive kidney disease.
Pyridorin Phase 3 study terminated
New markers are constantly being evaluated to improve due to lack of funding
diagnostic and prognostic precision, but validated evidence
Nox1/4 inhibitor: NADPH oxidase Nox 1 and Nox 4 inhibitor
for their use is lacking. The discovery, development of an
DM: diabetes mellitus
evidence base, and then implementation of new therapeutic DKD: diabetic kidney disease
agents is necessary for patients with DKD, and will benefit
affected individuals, their families, and the community.
References
1 Gellman, D.D., Pirani, C.L., Soothill, J.F. et al. (1959). cms:uploads/docs/state-of-the-nation-2015-web.pdf

Diabetic nephropathy: a clinical and pathologic study based (accessed 22 May 2021).
on renal biopsies. Medicine (Baltimore) 38: 321–367. 6 Diabetes Control and Complications Trial Research
2 Tervaert, T.W., Mooyaart, A.L., Amann, K. et al. (2010). Group, Nathan, D.M., Genuth, S. et al. (1993). The effect
Pathologic classification of diabetic nephropathy. J. Am. of intensive treatment of diabetes on the development
Soc. Nephrol. 21 (4): 556–563. and progression of long-term complications in insulin-
3 Ogurtsova, K., da Rocha Fernandes, J.D., Huang, Y. et al. dependent diabetes mellitus. N. Engl. J. Med. 329 (14):
(2017). IDF diabetes atlas: global estimates for the 977–986.
prevalence of diabetes for 2015 and 2040. Diabetes Res. 7 Stratton, I.M., Adler, A.I., Neil, H.A. et al. (2000).
Clin. Pract. 128: 40–50. Association of glycaemia with macrovascular and
4 DeFronzo, R.A., Ferrannini, E., Zimmet, P., and Alberti, microvascular complications of type 2 diabetes (UKPDS
G. (2015). International Textbook of Diabetes Mellitus, 4e 35): prospective observational study. BMJ 321 (7258):
(eds. R.A. DeFronzo, E. Ferrannini, P. Zimmet and G. 405–412.
Alberti). Wiley Blackwell 1228 p. 8 Dunlop, M. (2000). Aldose reductase and the role of the
5 Kidney Health Australia. State of the nation. Chronic polyol pathway in diabetic nephropathy. Kidney Int.
kidney disease in Australia 2015. https://kidney.org.au/ Suppl. 77: S3–S12.
Reference 389
9 Bjornstad, P., Lanaspa, M.A., Ishimoto, T. et al. (2015). 24 Anders, H.J. (2013). Innate versus adaptive immunity in
Fructose and uric acid in diabetic nephropathy. kidney immunopathology. BMC Nephrol. 14: 138.
Diabetologia 58 (9): 1993–2002. 25 Navarro-Gonzalez, J.F., Mora-Fernandez, C., Muros de
10 Glastras, S.J., Tsang, M., Teh, R. et al. (2016). Maternal Fuentes, M., and Garcia-Perez, J. (2011). Inflammatory
obesity promotes diabetic nephropathy in rodent molecules and pathways in the pathogenesis of
offspring. Sci. Rep. 6: 27769. diabetic nephropathy. Nat. Rev. Nephrol. 7 (6):
11 Tan, A.L., Forbes, J.M., and Cooper, M.E. (2007). AGE, 327–340.
RAGE, and ROS in diabetic nephropathy. Semin. Nephrol. 26 Shao, B.Z., Xu, Z.Q., Han, B.Z. et al. (2015).
27 (2): 130–143. NLRP3 inflammasome and its inhibitors: a review. Front.
12 Du, X., Matsumura, T., Edelstein, D. et al. (2003). Pharmacol. 6: 262.
Inhibition of GAPDH activity by poly(ADP-ribose) 27 Wang, W., Wang, X., Chun, J. et al. (2013).
polymerase activates three major pathways of Inflammasome-independent NLRP3 augments TGF-beta
hyperglycemic damage in endothelial cells. J. Clin. Invest. signaling in kidney epithelium. J. Immunol. 190 (3):
112 (7): 1049–1057. 1239–1249.
13 Giacco, F. and Brownlee, M. (2010). Oxidative stress and 28 Anders, H.J. and Schaefer, L. (2014). Beyond tissue
diabetic complications. Circ. Res. 107 (9): 1058–1070. injury-damage-associated molecular patterns, toll-like
14 Gordon, J.W., Dolinsky, V.W., Mughal, W. et al. (2015). receptors, and inflammasomes also drive regeneration
Targeting skeletal muscle mitochondria to prevent type 2 and fibrosis. J. Am. Soc. Nephrol. 25 (7): 1387–1400.
diabetes in youth. Biochem. Cell Biol. 93 (5): 452–465. 29 Sheedy, F.J. and O’Neill, L.A. (2007). The Troll in Toll:
15 Sharma, K. (2015). Mitochondrial hormesis and diabetic Mal and Tram as bridges for TLR2 and TLR4 signaling. J.
complications. Diabetes 64 (3): 663–672. Leukocyte Biol. 82 (2): 196–203.
16 Gill, P.S. and Wilcox, C.S. (2006). NADPH oxidases in 30 Beutler, B. (2004). Inferences, questions and possibilities
the kidney. Antioxid. Redox Signaling 8 (9–10): in Toll-like receptor signalling. Nature 430 (6996):
1597–1607. 257–263.
17 Bernhardt, W.M., Schmitt, R., Rosenberger, C. et al. 31 Kawai, T. and Akira, S. (2010). The role of pattern-
(2006). Expression of hypoxia-inducible transcription recognition receptors in innate immunity: update on
factors in developing human and rat kidneys. Kidney Int. Toll-like receptors. Nat. Immunol. 11 (5): 373–384.
69 (1): 114–122. 32 Li, F., Yang, N., Zhang, L. et al. (2010). Increased
18 Haase, V.H. (2012). Hypoxia-inducible factor signaling in expression of toll-like receptor 2 in rat diabetic
the development of kidney fibrosis. Fibrogenesis Tissue nephropathy. Am. J. Nephrol. 32 (2): 179–186.
Repair 5 (Suppl 1): S16. 33 Devaraj, S., Tobias, P., Kasinath, B.S. et al. (2011).
19 Wada, J. and Makino, H. (2013). Inflammation and the Knockout of toll-like receptor-2 attenuates both the
pathogenesis of diabetic nephropathy. Clin. Sci. (Lond) proinflammatory state of diabetes and incipient diabetic
124 (3): 139–152. nephropathy. Arterioscler. Thromb. Vasc. Biol. 31 (8):
20 Vanhove, T., Kinashi, H., Nguyen, T.Q. et al. (2017). 1796–1804.
Tubulointerstitial expression and urinary excretion of 34 Lin, M. and Tang, S.C. (2014). Toll-like receptors: sensing
connective tissue growth factor 3 months after renal and reacting to diabetic injury in the kidney. Nephrol.
transplantation predict interstitial fibrosis and tubular Dial. Transplant. 29 (4): 746–754.
atrophy at 5 years in a retrospective cohort analysis. 35 Xiao, X., Du, C., Yan, Z. et al. (2017). Inhibition of
Transplant. Int. 30 (7): 695–705. necroptosis attenuates kidney inflammation and
21 Martini, S., Nair, V., Keller, B.J. et al. (2014). Integrative interstitial fibrosis induced by unilateral ureteral
biology identifies shared transcriptional networks in obstruction. Am. J. Nephrol. 46 (2): 131–138.
CKD. J. Am. Soc. Nephrol. 25 (11): 2559–2572. 36 Zhu, Y., Cui, H., Xia, Y., and Gan, H. (2016). RIPK3-
22 Nair, V., Komorowsky, C.V., Weil, E.J. et al. (2018). mediated necroptosis and apoptosis contributes to renal
A molecular morphometric approach to diabetic kidney tubular cell progressive loss and chronic kidney disease
disease can link structure to function and outcome. progression in rats. PLoS One 11 (6): e0156729.
Kidney Int. 93 (2): 439–449. 37 Liu, B.C., Tang, T.T., Lv, L.L., and Lan, H.Y. (2018). Renal
23 Modena, B.D., Kurian, S.M., Gaber, L.W. et al. (2016). tubule injury: a driving force toward chronic kidney
Gene expression in biopsies of acute rejection and disease. Kidney Int. 93 (3): 568–579.
interstitial fibrosis/tubular atrophy reveals highly shared 38 Duffield, J.S. (2010). Macrophages and immunologic
mechanisms that correlate with worse long-term inflammation of the kidney. Semin. Nephrol. 30 (3):
outcomes. Am. J. Transplant. 16 (7): 1982–1998. 234–254.
9 Liu, Y. (2006). Renal fibrosis: new insights into the

3 and inflammatory cytokine expression in kidney tubular
pathogenesis and therapeutics. Kidney Int. 69 (2): 213–217. cells. Am. J. Physiol. Renal Physiol. 291 (5): F1070–F1077.
40 Cao, Q., Harris, D.C., and Wang, Y. (2015). Macrophages 56 Qi, W., Chen, X., Poronnik, P., and Pollock, C.A. (2008).
in kidney injury, inflammation, and fibrosis. Physiology Transforming growth factor-beta/connective tissue
(Bethesda) 30 (3): 183–194. growth factor axis in the kidney. Int. J. Biochem. Cell Biol.
41 Kanasaki, K., Taduri, G., and Koya, D. (2013). Diabetic 40 (1): 9–13.
nephropathy: the role of inflammation in fibroblast 57 Lebrin, F., Deckers, M., Bertolino, P., and Ten Dijke, P.
activation and kidney fibrosis. Front. Endocrinol. (2005). TGF-beta receptor function in the endothelium.
(Lausanne) 4: 7. Cardiovasc. Res. 65 (3): 599–608.
42 Strutz, F. and Zeisberg, M. (2006). Renal fibroblasts and 58 Das, R., Xu, S., Quan, X. et al. (2014). Upregulation of
myofibroblasts in chronic kidney disease. J. Am. Soc. mitochondrial Nox4 mediates TGF-beta-induced
Nephrol. 17 (11): 2992–2998. apoptosis in cultured mouse podocytes. Am. J. Physiol.
43 Klingberg, F., Hinz, B., and White, E.S. (2013). The Renal Physiol. 306 (2): F155–F167.
myofibroblast matrix: implications for tissue repair and 59 Mack, M. and Yanagita, M. (2015). Origin of
fibrosis. J. Pathol. 229 (2): 298–309. myofibroblasts and cellular events triggering fibrosis.
44 Anders, H.J., Vielhauer, V., and Schlondorff, D. (2003). Kidney Int. 87 (2): 297–307.
Chemokines and chemokine receptors are involved in the 60 Zeravica, R., Cabarkapa, V., Ilincic, B. et al. (2015).
resolution or progression of renal disease. Kidney Int. 63 (2): Plasma endothelin-1 level, measured glomerular
401–415. filtration rate and effective renal plasma flow in diabetic
45 Schlondorff, D.O. (2008). Overview of factors contributing nephropathy. Ren. Fail. 37 (4): 681–686.
to the pathophysiology of progressive renal disease. 61 Tuttle, K.R. (2017). Back to the future: glomerular
Kidney Int. 74 (7): 860–866. hyperfiltration and the diabetic kidney. Diabetes 66 (1):
46 Wang, W., Koka, V., and Lan, H.Y. (2005). Transforming 14–16.
growth factor-beta and Smad signalling in kidney 62 Vallon, V., Richter, K., Blantz, R.C. et al. (1999).
diseases. Nephrology 10 (1): 48–56. Glomerular hyperfiltration in experimental diabetes
47 Bottinger, E.P. (2007). TGF-beta in renal injury and mellitus: potential role of tubular reabsorption. J. Am.
disease. Semin. Nephrol. 27 (3): 309–320. Soc. Nephrol. 10 (12): 2569–2576.
48 Roberts, A.B. (1998). Molecular and cell biology of 63 Eid, S., Boutary, S., Braych, K. et al. (2016). mTORC2
TGF-beta. Miner. Electrolyte Metab. 24 (2–3): 111–119. signaling regulates Nox4-induced podocyte depletion
49 Voelker, J., Berg, P.H., Sheetz, M. et al. (2017). Anti-TGF- in diabetes. Antioxid. Redox Signaling 25 (13): 703–719.
beta1 antibody therapy in patients with diabetic 64 Gagliardini, E., Perico, N., Rizzo, P. et al. (2013).
nephropathy. J. Am. Soc. Nephrol. 28 (3): 953–962. Angiotensin II contributes to diabetic renal dysfunction
50 Border, W.A., Okuda, S., Languino, L.R., and Ruoslahti, in rodents and humans via Notch1/Snail pathway. Am. J.
E. (1990). Transforming growth factor-beta regulates Pathol. 183 (1): 119–130.
production of proteoglycans by mesangial cells. Kidney 65 de Zeeuw, D., Coll, B., Andress, D. et al. (2014). The
Int. 37 (2): 689–695. endothelin antagonist atrasentan lowers residual
51 Haberstroh, U., Zahner, G., Disser, M. et al. (1993). albuminuria in patients with type 2 diabetic nephropathy.
TGF-beta stimulates rat mesangial cell proliferation in J. Am. Soc. Nephrol. 25 (5): 1083–1093.
culture: role of PDGF beta-receptor expression. Am. J. 66 Kume, S., Koya, D., Uzu, T., and Maegawa, H. (2014).
Physiol. 264 (2 Pt 2): F199–F205. Role of nutrient-sensing signals in the pathogenesis of
52 Wilson, H.M., Reid, F.J., Brown, P.A. et al. (1993). Effect diabetic nephropathy. Biomed Res. Int. 2014: 315494.
of transforming growth factor-beta 1 on plasminogen 67 Huang, C., Lin, M.Z., Cheng, D. et al. (2016).
activators and plasminogen activator inhibitor-1 in renal KCa3.1 mediates dysfunction of tubular autophagy in
glomerular cells. Exp. Nephrol. 1 (6): 343–350. diabetic kidneys via PI3k/Akt/mTOR signaling pathways.
53 Qi, W., Chen, X., Poronnik, P., and Pollock, C.A. (2006). Sci. Rep. 6: 23884.
The renal cortical fibroblast in renal tubulointerstitial 68 Esterline, R.L., Vaag, A., Oscarsson, J., and Vora, J. (2018).
fibrosis. Int. J. Biochem. Cell Biol. 38 (1): 1–5. Mechanisms in Endocrinology: SGLT2 inhibitors: clinical
54 Qi, W., Chen, X., Polhill, T.S. et al. (2006). TGF- benefits by restoration of normal diurnal metabolism?
beta1 induces IL-8 and MCP-1 through a connective Eur. J. Endocrinol. 178 (4): R113–R125.
tissue growth factor-independent pathway. Am. J. Physiol. 69 Huang, C., Zhang, L., Shi, Y. et al. (2018). The KCa3.1
Renal Physiol. 290 (3): F703–F709. blocker TRAM34 reverses renal damage in a mouse
55 Qi, W., Chen, X., Holian, J. et al. (2006). Transforming model of established diabetic nephropathy. PLoS One 13
growth factor-beta1 differentially mediates fibronectin (2): e0192800.
Reference 391
0 Panchapakesan, U. and Pollock, C. (2018). The role of

7 85 MacLeod, J.M., Lutale, J., and Marshall, S.M. (1995).
toll-like receptors in diabetic kidney disease. Curr. Opin. Albumin excretion and vascular deaths in NIDDM.
Nephrol. Hypertens 27 (1): 30–34. Diabetologia 38 (5): 610–616.
71 Gray, S.P. and Jandeleit-Dahm, K. (2014). The pathobiology 86 Yudkin, J.S., Forrest, R.D., and Jackson, C.A. (1988).
of diabetic vascular complications--cardiovascular and Microalbuminuria as predictor of vascular disease in
kidney disease. J. Mol. Med. (Berl) 92 (5): 441–452. non-diabetic subjects. Islington diabetes survey. Lancet 2
72 Kang, H.M., Ahn, S.H., Choi, P. et al. (2015). Defective fatty (8610): 530–533.
acid oxidation in renal tubular epithelial cells has a key 87 KDOQI (2007). KDOQI clinical practice guidelines and
role in kidney fibrosis development. Nat. Med. 21 (1): 37–46. clinical practice recommendations for diabetes and chronic
73 Portela, A. and Esteller, M. (2010). Epigenetic kidney disease. Am. J. Kidney Dis. 49 (2 Suppl 2): S12–S154.
modifications and human disease. Nat. Biotechnol. 28 88 National Kidney Foundation (2012). KDOQI clinical
(10): 1057–1068. practice guideline for diabetes and CKD: 2012 update.
74 Horsburgh, S., Robson-Ansley, P., Adams, R., and Smith, Am. J. Kidney Dis. 60 (5): 850–886.
C. (2015). Exercise and inflammation-related epigenetic 89 Zitkus, B.S. (2014). Update on the American Diabetes
modifications: focus on DNA methylation. Exerc. Association standards of medical care. Nurse Pract. 39
Immunol. Rev. 21: 26–41. (8): 22–32. quiz -3.
75 Bell, C.G., Teschendorff, A.E., Rakyan, V.K. et al. (2010). 90 Diabetes Canada Clinical Practice Guidelines Expert
Genome-wide DNA methylation analysis for diabetic Committee, McFarlane, P., Cherney, D. et al. (2018).
nephropathy in type 1 diabetes mellitus. BMC Med. Genet. Chronic kidney disease in diabetes. Can. J. Diabetes 42
3: 33. (Suppl 1): S201–S209.
76 Bechtel, W., McGoohan, S., Zeisberg, E.M. et al. (2010). 91 Lambers Heerspink, H.J., Gansevoort, R.T., Brenner, B.M.
Methylation determines fibroblast activation and et al. (2010). Comparison of different measures of urinary
fibrogenesis in the kidney. Nat. Med. 16 (5): 544–550. protein excretion for prediction of renal events. J. Am.
77 Bartel, D.P. (2009). MicroRNAs: target recognition and Soc. Nephrol. 21 (8): 1355–1360.
regulatory functions. Cell 136 (2): 215–233. 92 Poggio, E.D., Nef, P.C., Wang, X. et al. (2005).
78 Kato, M., Arce, L., Wang, M. et al. (2011). A microRNA Performance of the Cockcroft-Gault and modification of
circuit mediates transforming growth factor-beta1 diet in renal disease equations in estimating GFR in ill
autoregulation in renal glomerular mesangial cells. hospitalized patients. Am. J. Kidney Dis. 46 (2): 242–252.
Kidney Int. 80 (4): 358–368. 93 Levey, A.S., Stevens, L.A., Schmid, C.H. et al. (2009). A
79 Kato, M., Zhang, J., Wang, M. et al. (2007). new equation to estimate glomerular filtration rate. Ann.
MicroRNA-192 in diabetic kidney glomeruli and its Intern. Med. 150 (9): 604–612.
function in TGF-beta-induced collagen expression via 94 Yokoyama, H., Sone, H., Oishi, M. et al. (2009).
inhibition of E-box repressors. Proc. Natl. Acad. Sci. USA. Prevalence of albuminuria and renal insufficiency
104 (9): 3432–3437. and associated clinical factors in type 2 diabetes: the
80 Panchapakesan, U. and Pollock, C. (2016). Long non- Japan Diabetes Clinical Data Management study
coding RNAs-towards precision medicine in diabetic (JDDM15). Nephrol. Dial. Transplant. 24 (4): 1212–1219.
kidney disease? Clin. Sci. (Lond) 130 (18): 1599–1602. 95 Thomas, M.C., Macisaac, R.J., Jerums, G. et al. (2009).
81 Adler, A.I., Stevens, R.J., Manley, S.E. et al. (2003). Nonalbuminuric renal impairment in type 2 diabetic
Development and progression of nephropathy in type 2 patients and in the general population (national
diabetes: the United Kingdom Prospective Diabetes Study evaluation of the frequency of renal impairment
(UKPDS 64). Kidney Int. 63 (1): 225–232. cO-existing with NIDDM [NEFRON] 11). Diabetes Care
82 Messent, J.W., Elliott, T.G., Hill, R.D. et al. (1992). 32 (8): 1497–1502.
Prognostic significance of microalbuminuria in 96 Penno, G., Solini, A., Bonora, E. et al. (2011). Clinical
insulin-dependent diabetes mellitus: a twenty-three year significance of nonalbuminuric renal impairment in
follow-up study. Kidney Int. 41 (4): 836–839. type 2 diabetes. J. Hypertens. 29 (9): 1802–1809.
83 Ninomiya, T., Perkovic, V., de Galan, B.E. et al. (2009). 97 Porrini, E., Ruggenenti, P., Mogensen, C.E. et al. (2015).
Albuminuria and kidney function independently predict Non-proteinuric pathways in loss of renal function in
cardiovascular and renal outcomes in diabetes. J. Am. Soc. patients with type 2 diabetes. Lancet Diabetes
Nephrol. 20 (8): 1813–1821. Endocrinol. 3 (5): 382–391.
84 Jun, M., Ohkuma, T., Zoungas, S. et al. (2018). Changes 98 Ndanuko, R.N., Tapsell, L.C., Charlton, K.E. et al.
in albuminuria and the risk of major clinical outcomes in (2016). Dietary patterns and blood pressure in adults: a
diabetes: results from ADVANCE-ON. Diabetes Care 41 systematic review and meta-analysis of randomized
(1): 163–170. controlled trials. Adv. Nutr. 7 (1): 76–89.
99 Navaneethan, S.D., Yehnert, H., Moustarah, F. et al. nephropathy in overweight or obese adults with type 2
(2009). Weight loss interventions in chronic kidney diabetes: a secondary analysis of the Look AHEAD
disease: a systematic review and meta-analysis. Clin. J. randomised clinical trial. Lancet Diabetes Endocrinol. 2
Am. Soc. Nephrol. 4 (10): 1565–1574. (10): 801–809.
100 Schwingshackl, L., Dias, S., and Hoffmann, G. (2014). 112 Tonelli, M., Muntner, P., Lloyd, A. et al. (2012). Risk of
Impact of long-term lifestyle programmes on weight loss coronary events in people with chronic kidney disease
and cardiovascular risk factors in overweight/obese compared with those with diabetes: a population-level
participants: a systematic review and network meta- cohort study. Lancet 380 (9844): 807–814.
analysis. Syst. Rev. 3: 130. 113 Fox, C.S., Matsushita, K., Woodward, M. et al. (2012).
101 Schwingshackl, L., Strasser, B., and Hoffmann, G. Associations of kidney disease measures with mortality
(2011). Effects of monounsaturated fatty acids on and end-stage renal disease in individuals with and
glycaemic control in patients with abnormal glucose without diabetes: a meta-analysis. Lancet 380 (9854):
metabolism: a systematic review and meta-analysis. 1662–1673.
Ann. Nutr. Metab. 58 (4): 290–296. 114 de Boer, I.H., Katz, R., Cao, J.J. et al. (2009). Cystatin C,
102 Ezzati, M. and Riboli, E. (2013). Behavioral and dietary albuminuria, and mortality among older adults with
risk factors for noncommunicable diseases. N. Engl. J. diabetes. Diabetes Care 32 (10): 1833–1838.
Med. 369 (10): 954–964. 115 KDIGO 2020 Clinical Practice Guideline for Diabetes
103 He, F.J., Marciniak, M., Visagie, E. et al. (2009). Effect of Management in Chronic Kidney Disease. Kidney Int
modest salt reduction on blood pressure, urinary Suppl 2020; 98 (4S).
albumin, and pulse wave velocity in white, black, and 116 Cohen, R. V.; Pereira, T. V.; Aboud, C. M.; Petry, T. B. Z.;
Asian mild hypertensives. Hypertension 54 (3): 482–488. Lopes Correa, J. L.; Schiavon, C. A.; Pompilio, C. E.;
104 Ekinci, E.I., Thomas, G., Thomas, D. et al. (2009). Pechy, F. N. Q.; da Costa Silva, A. C. C.; de Melo, F. L.
Effects of salt supplementation on the albuminuric G.; Cunha da Silveira, L. P.; de Paris Caravatto, P. P.;
response to telmisartan with or without Halpern, H.; Monteiro, F. L. J.; da Costa Martins, B.;
hydrochlorothiazide therapy in hypertensive patients Kuga, R.; Palumbo, T. M. S.; Docherty, N. G.; le Roux, C.
with type 2 diabetes are modulated by habitual dietary W. Effect of gastric bypass vs best medical treatment on
salt intake. Diabetes Care 32 (8): 1398–1403. early-stage chronic kidney disease in patients with type
105 McMahon, E.J., Bauer, J.D., Hawley, C.M. et al. (2013). 2 diabetes and obesity: a randomized clinical trial.
A randomized trial of dietary sodium restriction in JAMA Surgery 2020; 155 (8): e200420.
CKD. J. Am. Soc. Nephrol. 24 (12): 2096–2103. 117 (1998). Intensive blood-glucose control with
106 Maillot, M. and Drewnowski, A. (2012). A conflict sulphonylureas or insulin compared with
between nutritionally adequate diets and meeting the conventional treatment and risk of complications in
2010 dietary guidelines for sodium. Am. J. Prev. Med. 42 patients with type 2 diabetes (UKPDS 33). UK
(2): 174–179. Prospective Diabetes Study (UKPDS) Group. Lancet
107 Thomas, M.C., Moran, J., Forsblom, C. et al. (2011). The 352 (9131): 837–853.
association between dietary sodium intake, ESRD, and 118 Anonymous. (1995) The relationship of glycemic
all-cause mortality in patients with type 1 diabetes. exposure (HbA1c) to the risk of development and
Diabetes Care 34 (4): 861–866. progression of retinopathy in the diabetes control and
108 Gaede, P., Vedel, P., Larsen, N. et al. (2003). Multifactorial complications trial. Diabetes 44 (8): 968–983.
intervention and cardiovascular disease in patients with 119 Ohkubo, Y., Kishikawa, H., Araki, E. et al. (1995).
type 2 diabetes. N. Engl. J. Med. 348 (5): 383–393. Intensive insulin therapy prevents the progression of
109 Gaede, P., Vedel, P., Parving, H.H., and Pedersen, O. diabetic microvascular complications in Japanese
(1999). Intensified multifactorial intervention in patients with non-insulin-dependent diabetes mellitus:
patients with type 2 diabetes mellitus and a randomized prospective 6-year study. Diabetes Res.
microalbuminuria: the Steno type 2 randomised study. Clin. Pract. 28 (2): 103–117.
Lancet 353 (9153): 617–622. 120 (1998). Effect of intensive blood-glucose control with
110 Gaede, P., Oellgaard, J., Carstensen, B. et al. (2016). metformin on complications in overweight patients with
Years of life gained by multifactorial intervention in type 2 diabetes (UKPDS 34). UK Prospective Diabetes
patients with type 2 diabetes mellitus and Study (UKPDS) Group. Lancet 352 (9131): 854–865.
microalbuminuria: 21 years follow-up on the Steno-2 121 Nathan, D.M., Cleary, P.A., Backlund, J.Y. et al. (2005).
randomised trial. Diabetologia 59 (11): 2298–2307. Intensive diabetes treatment and cardiovascular disease
111 The Look AHEAD Research Group (2014). Effect of a in patients with type 1 diabetes. N. Engl. J. Med. 353
long-term behavioural weight loss intervention on (25): 2643–2653.
Reference 393
122 DCCT/EDIC Research Group, Orchard, T.J., Nathan, 135 Fang, H.J., Zhou, Y.H., Tian, Y.J. et al. (2016). Effects of
D.M. et al. (2015). Association between 7 years of intensive glucose lowering in treatment of type 2
intensive treatment of type 1 diabetes and long-term diabetes mellitus on cardiovascular outcomes: a
mortality. JAMA 313 (1): 45–53. meta-analysis of data from 58,160 patients in 13
123 Diabetes Control and Complications Trial/Epidemiology randomized controlled trials. Int. J. Cardiol. 218: 50–58.
of Diabetes Interventions and Complications Research 136 Herrera-Gomez, F., Asensio-Gonzalez, M., Gonzalez-
Group (2016). Intensive Diabetes Treatment and Lopez, A., and Alvarez, F.J. (2017). Effects of intensive
cardiovascular outcomes in type 1 Diabetes: the DCCT/ control of glycemia on clinical kidney outcomes in type
EDIC study 30-year follow-up. Diabetes Care 39 (5): 2 diabetes patients compared with standard control: a
686–693. meta-analysis. Front. Pharmacol. 8: 845.
124 Holman, R.R., Paul, S.K., Bethel, M.A. et al. (2008). 137 Pathak, R.D., Schroeder, E.B., Seaquist, E.R. et al.
10-year follow-up of intensive glucose control in type 2 (2016). Severe hypoglycemia requiring medical
diabetes. N. Engl. J. Med. 359 (15): 1577–1589. intervention in a large cohort of adults with diabetes
125 Action to Control Cardiovascular Risk in Diabetes Study receiving care in U.S. integrated health care delivery
Group, Gerstein, H.C., Miller, M.E. et al. (2008). Effects systems: 2005-2011. Diabetes Care 39 (3): 363–370.
of intensive glucose lowering in type 2 diabetes. N. Engl. 138 Yun, J.S., Ko, S.H., Ko, S.H. et al. (2013). Presence of
J. Med. 358 (24): 2545–2559. macroalbuminuria predicts severe hypoglycemia in
126 Ismail-Beigi, F., Craven, T., Banerji, M.A. et al. (2010). Effect patients with type 2 diabetes: a 10-year follow-up study.
of intensive treatment of hyperglycaemia on microvascular Diabetes Care 36 (5): 1283–1289.
outcomes in type 2 diabetes: an analysis of the ACCORD 139 Ly, J., Marticorena, R., and Donnelly, S. (2004). Red
randomised trial. Lancet 376 (9739): 419–430. blood cell survival in chronic renal failure. Am. J. Kidney
127 Action to Control Cardiovascular Risk in Diabetes Dis. 44 (4): 715–719.
Follow-On Eye Study Group (2016). Persistent effects of 140 Brown, J.N., Kemp, D.W., and Brice, K.R. (2009). Class
intensive glycemic control on retinopathy in type 2 effect of erythropoietin therapy on hemoglobin A(1c) in
diabetes in the Action to Control Cardiovascular Risk in a patient with diabetes mellitus and chronic kidney
Diabetes (ACCORD) follow-on study. Diabetes Care 39 disease not undergoing hemodialysis. Pharmacotherapy
(7): 1089–1100. 29 (4): 468–472.
128 Mottl, A.K., Buse, J.B., Ismail-Beigi, F. et al. (2018). 141 Chachou, A., Randoux, C., Millart, H. et al. (2000).
Long-term effects of intensive glycemic and blood Influence of in vivo hemoglobin carbamylation on
pressure control and fenofibrate use on kidney HbA1c measurements by various methods. Clin. Chem.
outcomes. Clin. J. Am. Soc. Nephrol. 13 (11): 1693–1702 Lab. Med. 38 (4): 321–326.
129 Perkovic, V., Heerspink, H.L., Chalmers, J. et al. (2013). 142 Lee, J.E. (2015). Alternative biomarkers for assessing
Intensive glucose control improves kidney outcomes in glycemic control in diabetes: fructosamine, glycated
patients with type 2 diabetes. Kidney Int. 83 (3): 517–523. albumin, and 1,5-anhydroglucitol. Ann. Pediatr.
130 Zoungas, S., Chalmers, J., Neal, B. et al. (2014). Endocrinol. Metab. 20 (2): 74–78.
Follow-up of blood-pressure lowering and glucose 143 Brenner, B.M., Cooper, M.E., de Zeeuw, D. et al. (2001).
control in type 2 diabetes. N. Engl. J. Med. 371 (15): Effects of losartan on renal and cardiovascular
1392–1406. outcomes in patients with type 2 diabetes and
131 Wong, M.G., Perkovic, V., Chalmers, J. et al. (2016). nephropathy. N. Engl. J. Med. 345 (12): 861–869.
Long-term benefits of intensive glucose control for 144 Lewis, E.J., Hunsicker, L.G., Clarke, W.R. et al. (2001).
preventing end-stage kidney disease: ADVANCE-ON. Renoprotective effect of the angiotensin-receptor
Diabetes Care 39 (5): 694–700. antagonist irbesartan in patients with nephropathy due
132 Papademetriou, V., Lovato, L., Doumas, M. et al. (2015). to type 2 diabetes. N. Engl. J. Med. 345 (12): 851–860.
Chronic kidney disease and intensive glycemic control 145 Zinman, B., Wanner, C., Lachin, J.M. et al. (2015).
increase cardiovascular risk in patients with type 2 Empagliflozin, cardiovascular outcomes, and mortality
diabetes. Kidney Int. 87 (3): 649–659. in type 2 diabetes. N. Engl. J. Med. 373 (22): 2117–2128.
133 Duckworth, W., Abraira, C., Moritz, T. et al. (2009). 146 Neal, B., Perkovic, V., and Matthews, D.R. (2017).
Glucose control and vascular complications in veterans Canagliflozin and cardiovascular and renal events in
with type 2 diabetes. N. Engl. J. Med. 360 (2): 129–139. type 2 diabetes. N. Engl. J. Med. Vol 377, issue 7, page
134 Hayward, R.A., Reaven, P.D., Emanuele, N.V., and number 644-657.
Investigators, V. (2015). Follow-up of glycemic control 147 Wanner, C., Inzucchi, S.E., and Zinman, B. (2016).
and cardiovascular outcomes in type 2 diabetes. N. Engl. Empagliflozin and progression of kidney disease in type
J. Med. 373 (10): 978. 2 diabetes. N. Engl. J. Med. 375 (18): 1801–1802.
148 Perkovic, V., de Zeeuw, D., Mahaffey, K.W. et al. (2018). Diabetology volume 18, Article number: 55 (2019)
Canagliflozin and renal outcomes in type 2 diabetes: effects-of-liraglutide-versus-placebo-on-cardiovascular.
results from the CANVAS program randomised clinical 159 Marso, S.P., Bain, S.C., Consoli, A. et al. (2016).
trials. Lancet Diabetes Endocrinol. 6 (9): 691–704. Semaglutide and cardiovascular outcomes in patients
149 Wiviott, S.D., Raz, I., Bonaca, M.P. et al. (2018). with type 2 diabetes. N. Engl. J. Med. 375 (19): 1834–1844.
Dapagliflozin and cardiovascular outcomes in type 2 160 Pfeffer, M.A., Claggett, B., Diaz, R. et al. (2015).
diabetes. N. Engl. J. Med. 380 (4): 347–357. Lixisenatide in patients with type 2 diabetes and acute
150 Rajasekeran, H., Lytvyn, Y., and Cherney, D.Z. (2016). coronary syndrome. N. Engl. J. Med. 373 (23): 2247–2257.
Sodium-glucose cotransporter 2 inhibition and 161 Muskiet, M.H.A., Tonneijck, L., Huang, Y. et al. (2018).
cardiovascular risk reduction in patients with type 2 Lixisenatide and renal outcomes in patients with type 2
diabetes: the emerging role of natriuresis. Kidney Int. 89 diabetes and acute coronary syndrome: an exploratory
(3): 524–526. analysis of the ELIXA randomised, placebo-controlled
151 Perkovic V, Jardine MJ, Neal B, Bompoint S, Heerspink trial. Lancet Diabetes Endocrinol. 6 (11): 859–869.
HJL, Charytan DM, Edwards R, Agarwal R, Bakris G, 162 Tuttle, K.R., Lakshmanan, M.C., Rayner, B. et al. (2018).
Bull S, Cannon CP, Capuano G, Chu PL, de Zeeuw D, Dulaglutide versus insulin glargine in patients with type
Greene T, Levin A, Pollock C, Wheeler DC, Yavin Y, 2 diabetes and moderate-to-severe chronic kidney disease
Zhang H, Zinman B, Meininger G, Brenner BM, (AWARD-7): a multicentre, open-label, randomised trial.
Mahaffey KW; CREDENCE Trial Investigators. Lancet Diabetes Endocrinol. 6 (8): 605–617.
Canagliflozin and Renal Outcomes in Type 2 Diabetes 163 Takagaki, Y., Koya, D., and Kanasaki, K. (2017).
and Nephropathy. N Engl J Med. 2019 Jun Dipeptidyl peptidase-4 inhibition and renoprotection:
13;380(24):2295–2306. the role of antifibrotic effects. Curr. Opin. Nephrol.
152 Wheeler DC, Stefánsson BV, Jongs N, Chertow GM, Hypertens 26 (1): 56–66.
Greene T, Hou FF, McMurray JJV, Correa-Rotter R, 164 Mulvihill, E.E. and Drucker, D.J. (2014). Pharmacology,
Rossing P, Toto RD, Sjöström CD, Langkilde AM, physiology, and mechanisms of action of dipeptidyl
Heerspink HJL; DAPA-CKD Trial Committees and peptidase-4 inhibitors. Endocr. Rev. 35 (6): 992–1019.
Investigators. Effects of dapagliflozin on major adverse 165 Marchetti, C., Di Carlo, A., Facchiano, F. et al. (2012).
kidney and cardiovascular events in patients with High mobility group box 1 is a novel substrate of
diabetic and non-diabetic chronic kidney disease: a dipeptidyl peptidase-IV. Diabetologia 55 (1): 236–244.
prespecified analysis from the DAPA-CKD trial. Lancet 166 Brandt, I., Lambeir, A.M., Ketelslegers, J.M. et al. (2006).
Diabetes Endocrinol. 2021 Jan; 9 (1): 22–31. doi: Dipeptidyl-peptidase IV converts intact B-type
10.1016/S2213-8587(20)30369–7. PMID: 33338413. natriuretic peptide into its des-SerPro form. Clin. Chem.
153 Lo C, Toyama T, Lin J, et al. Insulin and glucose 52 (1): 82–87.
lowering agents for treating people with diabetes and 167 Mentlein, R. (1999). Dipeptidyl-peptidase IV
chronic kidney disease. Cochrane Database Syst Rev. (CD26) – role in the inactivation of regulatory peptides.
2018; 9: CD011798. Regul. Pept. 85 (1): 9–24.
154 Tanaka, T., Higashijima, Y., Wada, T., and Nangaku, M. 168 Kirby, M., Yu, D.M., O’Connor, S., and Gorrell, M.D.
(2014). The potential for renoprotection with incretin- (2009). Inhibitor selectivity in the clinical application of
based drugs. Kidney Int. 86 (4): 701–711. dipeptidyl peptidase-4 inhibition. Clin. Sci. (Lond) 118
155 Fujita, H., Morii, T., Fujishima, H. et al. (2014). The (1): 31–41.
protective roles of GLP-1R signaling in diabetic 169 Gorrell, M.D., Gysbers, V., and McCaughan, G.W.
nephropathy: possible mechanism and therapeutic (2001). CD26: a multifunctional integral membrane and
potential. Kidney Int. 85 (3): 579–589. secreted protein of activated lymphocytes. Scand. J.
156 Muskiet, M.H.A., Tonneijck, L., Smits, M.M. et al. Immunol. 54 (3): 249–264.
(2017). GLP-1 and the kidney: from physiology to 170 Shi, S., Srivastava, S.P., Kanasaki, M. et al. (2015).
pharmacology and outcomes in diabetes. Nat. Rev. Interactions of DPP-4 and integrin beta1 influences
Nephrol. 13 (10): 605–628. endothelial-to-mesenchymal transition. Kidney Int. 88
157 Mann, J.F.E., Orsted, D.D., and Buse, J.B. (2017). (3): 479–489.
Liraglutide and renal outcomes in type 2 diabetes. N. 171 Panchapakesan, U., Mather, A., and Pollock, C. (2013).
Engl. J. Med. 377 (22): 2197–2198. Role of GLP-1 and DPP-4 in diabetic nephropathy and
158 Maurice B. Bizino, Ingrid M. Jazet, Jos J. M. Westenberg, cardiovascular disease. Clin. Sci. (Lond) 124 (1): 17–26.
Huub J. van Eyk, Elisabeth H. M. Paiman, Jan W. A. 172 Panchapakesan, U. and Pollock, C. (2014). Invited
Smit & Hildebrandus J. Lamb Cardiovascular commentary on linagliptin-mediated DPP-4 inhibition
Reference 395
ameliorates kidney fibrosis in streptozotocin-induced 185 Green, J.B., Bethel, M.A., Armstrong, P.W. et al. (2015).
diabetic mice by inhibiting endothelial-to-mesenchymal Effect of sitagliptin on cardiovascular outcomes in type
transition in a therapeutic regimen. Diabetes 63 (6): 2 diabetes. N. Engl. J. Med. 373 (3): 232–242.
1829–1830. 186 White, W.B., Heller, S.R., Cannon, C.P. et al. (2018).
173 Gangadharan Komala, M., Gross, S., Zaky, A. et al. Alogliptin in patients with type 2 diabetes receiving
(2015). Linagliptin limits high glucose induced metformin and sulfonylurea therapies in the EXAMINE
conversion of latent to active TGFss through interaction trial. Am. J. Med. 131 (7): 813–819. e5.
with CIM6PR and limits renal tubulointerstitial 187 V P. ASN (2018). https://www.eurekalert.org/pub_
fibronectin. PLoS One 10 (10): e0141143. releases/2018-10/ason-hct102418.php (accessed 22 May
174 Panchapakesan, U. and Pollock, C. (2015). The role of 2021).
DPP4 inhibitors in diabetic kidney disease. Front. 188 Rena, G., Hardie, D.G., and Pearson, E.R. (2017). The
Immunol. 6: 443 Page 1-5. mechanisms of action of metformin. Diabetologia 60 (9):
175 Alter, M.L., Ott, I.M., von Websky, K. et al. (2012). 1577–1585.
DPP-4 inhibition on top of angiotensin receptor 189 Keely, E., Berger, H., and Feig, D.S. (2018). New diabetes
blockade offers a new therapeutic approach for diabetic Canada clinical practice guidelines for diabetes and
nephropathy. Kidney Blood Press. Res. 36 (1): 119–130. pregnancy-what’s changed? J. Obstet Gynaecol. 40 (11):
176 Matsui, T., Nakashima, S., Nishino, Y. et al. (2015). 1484–1489. Can. Diabetes Canada Clinical Practice
Dipeptidyl peptidase-4 deficiency protects against Guidelines Diabetes in Pregnancy Expert Committee
experimental diabetic nephropathy partly by blocking 190 American Diabetes Association (2018). 8.
the advanced glycation end products-receptor axis. Lab. Pharmacologic approaches to glycemic treatment:
Invest. 95 (5): 525–533. standards of medical care in diabetes-2018. Diabetes
177 Kanasaki, K., Shi, S., Kanasaki, M. et al. (2014). Care 41 (Suppl 1): S73–S85.
Linagliptin-mediated DPP-4 inhibition ameliorates 191 Kahn, S.E., Haffner, S.M., Heise, M.A. et al. (2006).
kidney fibrosis in streptozotocin-induced diabetic mice Glycemic durability of rosiglitazone, metformin, or
by inhibiting endothelial-to-mesenchymal transition in glyburide monotherapy. N. Engl. J. Med. 355 (23):
a therapeutic regimen. Diabetes 63 (6): 2120–2131. 2427–2443.
178 Mega, C., de Lemos, E.T., Vala, H. et al. (2011). Diabetic 192 Maruthur, N.M., Tseng, E., Hutfless, S. et al. (2016).
nephropathy amelioration by a low-dose sitagliptin in Diabetes medications as monotherapy or metformin-
an animal model of type 2 diabetes (Zucker diabetic based combination therapy for type 2 diabetes: a
fatty rat). Exp. Diabetes Res. 2011: 162092. systematic review and meta-analysis. Ann. Intern. Med.
179 Nistala, R., Habibi, J., Lastra, G. et al. (2014). Prevention 164 (11): 740–751.
of obesity-induced renal injury in male mice by 193 Hong, J., Zhang, Y., Lai, S. et al. (2013). Effects of
DPP4 inhibition. Endocrinology 155 (6): 2266–2276. metformin versus glipizide on cardiovascular outcomes
180 Tsuprykov, O., Ando, R., Reichetzeder, C. et al. (2016). in patients with type 2 diabetes and coronary artery
The dipeptidyl peptidase inhibitor linagliptin and the disease. Diabetes Care 36 (5): 1304–1311.
angiotensin II receptor blocker telmisartan show renal 194 (1998). Tight blood pressure control and risk of
benefit by different pathways in rats with macrovascular and microvascular complications in type
5/6 nephrectomy. Kidney Int. 89 (5): 1049–1061. 2 diabetes: UKPDS 38. UK Prospective Diabetes Study
181 Joo, K.W., Kim, S., Ahn, S.Y. et al. (2013). Dipeptidyl Group. BMJ 317 (7160): 703–713.
peptidase IV inhibitor attenuates kidney injury in rat 195 Palmer, S.C., Mavridis, D., Nicolucci, A. et al. (2016).
remnant kidney. BMC Nephrol. 14: 98. Comparison of clinical outcomes and adverse events
182 Min, H.S., Kim, J.E., Lee, M.H. et al. (2014). Dipeptidyl associated with glucose-lowering drugs in patients with
peptidase IV inhibitor protects against renal interstitial type 2 diabetes: a meta-analysis. JAMA 316 (3): 313–324.
fibrosis in a mouse model of ureteral obstruction. Lab. 196 Boussageon, R., Supper, I., Bejan-Angoulvant, T. et al.
Invest. 94 (6): 598–607. (2012). Reappraisal of metformin efficacy in the
183 Kim, Y.G., Byun, J., Yoon, D. et al. (2016). Renal protective treatment of type 2 diabetes: a meta-analysis of
effect of DPP-4 inhibitors in type 2 diabetes mellitus randomised controlled trials. PLoS Med. 9 (4):
patients: a cohort study. J. Diabetes Res. 2016: 1423191. e1001204.
184 Scirica, B.M., Bhatt, D.L., Braunwald, E. et al. (2013). 197 Aroda, V.R., Edelstein, S.L., Goldberg, R.B. et al. (2016).
Saxagliptin and cardiovascular outcomes in patients Long-term metformin use and vitamin B12 deficiency in
with type 2 diabetes mellitus. N. Engl. J. Med. 369 (14): the diabetes prevention program outcomes study. J. Clin.
1317–1326. Endocrinol. Metab. 101 (4): 1754–1761.
198 Vella, S., Buetow, L., Royle, P. et al. (2010). The use of diabetes with and without renal impairment treated
metformin in type 1 diabetes: a systematic review of with metformin: a retrospective cohort study. Diabetes
efficacy. Diabetologia 53 (5): 809–820. Care 37 (8): 2291–2295.
199 Libman, I.M., Miller, K.M., DiMeglio, L.A. et al. (2015). 212 Eppenga, W.L., Lalmohamed, A., Geerts, A.F. et al.
Effect of metformin added to insulin on glycemic control (2014). Risk of lactic acidosis or elevated lactate
among overweight/obese adolescents with type 1 diabetes: concentrations in metformin users with renal
a randomized clinical trial. JAMA 314 (21): 2241–2250. impairment: a population-based cohort study. Diabetes
200 Petrie, J.R., Chaturvedi, N., Ford, I. et al. (2017). Care 37 (8): 2218–2224.
Cardiovascular and metabolic effects of metformin in 213 Bodmer, M., Meier, C., Krahenbuhl, S. et al. (2008).
patients with type 1 diabetes (REMOVAL): a double- Metformin, sulfonylureas, or other antidiabetes drugs
blind, randomised, placebo-controlled trial. Lancet and the risk of lactic acidosis or hypoglycemia: a nested
Diabetes Endocrinol. 5 (8): 597–609. case-control analysis. Diabetes Care 31 (11): 2086–2091.
201 Lalau, J.D., Arnouts, P., Sharif, A., and De Broe, M.E. 214 Rachmani, R., Slavachevski, I., Levi, Z. et al. (2002).
(2015). Metformin and other antidiabetic agents in renal Metformin in patients with type 2 diabetes mellitus:
failure patients. Kidney Int. 87 (2): 308–322. reconsideration of traditional contraindications. Eur. J.
202 Liu, F., Lu, J.X., Tang, J.L. et al. (2009). Relationship of Intern. Med. 13 (7): 428.
plasma creatinine and lactic acid in type 2 diabetic 215 Lim, V.C., Sum, C.F., Chan, E.S. et al. (2007). Lactate
patients without renal dysfunction. Chin. Med. J. (Engl) levels in Asian patients with type 2 diabetes mellitus on
122 (21): 2547–2553. metformin and its association with dose of metformin
203 Davis, T.M., Jackson, D., Davis, W.A. et al. (2001). The and renal function. Int. J. Clin. Pract. 61 (11): 1829–1833.
relationship between metformin therapy and the fasting 216 Connolly, V. and Kesson, C.M. (1996). Metformin
plasma lactate in type 2 diabetes: The Fremantle treatment in NIDDM patients with mild renal
Diabetes Study. Br. J. Clin. Pharmacol. 52 (2): 137–144. impairment. Postgrad. Med. J. 72 (848): 352–354.
204 Al-Hwiesh, A.K., Abdul-Rahman, I.S., El-Deen, M.A. 217 Hung, S.C., Chang, Y.K., Liu, J.S. et al. (2015). Metformin
et al. (2014). Metformin in peritoneal dialysis: a pilot use and mortality in patients with advanced chronic
experience. Perit. Dial. Int. 34 (4): 368–375. kidney disease: national, retrospective, observational,
205 Frid, A., Sterner, G.N., Londahl, M. et al. (2010). Novel cohort study. Lancet Diabetes Endocrinol. 3 (8): 605–614.
assay of metformin levels in patients with type 2 218 Christiansen, C.F., Ehrenstein, V., Heide-Jorgensen, U.
diabetes and varying levels of renal function: clinical et al. (2015). Metformin initiation and renal
recommendations. Diabetes Care 33 (6): 1291–1293. impairment: a cohort study in Denmark and the UK.
206 Sambol, N.C., Chiang, J., Lin, E.T. et al. (1995). Kidney BMJ Open 5 (9): e008531.
function and age are both predictors of 219 Chowdhury, T.A., Wright, R., and Yaqoob, M.M. (2015).
pharmacokinetics of metformin. J. Clin. Pharmacol. 35 Using metformin in the presence of renal disease. BMJ
(11): 1094–1102. 350: h1758.
207 Food and Drug Administration. Metformin-containing 220 Nye, H.J. and Herrington, W.G. (2011). Metformin: the
drugs: drug safety communication -revised warnings safest hypoglycaemic agent in chronic kidney disease?
for certain patients with reduced kidney function Nephron. Clin. Pract. 118 (4): c380–c383.
2016. https://www.fda.gov/drugs/drug-safety-and- 221 Ashcroft, F.M. (1996). Mechanisms of the glycaemic
availability/fda-drug-safety-communication-fda-revises- effects of sulfonylureas. Horm. Metab. Res. 28 (9):
warnings-regarding-use-diabetes-medicine-metformin- 456–463.
certain 222 Schopman, J.E., Simon, A.C., Hoefnagel, S.J. et al.
208 Salpeter, S.R., Greyber, E., Pasternak, G.A., and Salpeter, (2014). The incidence of mild and severe hypoglycaemia
E.E. (2010). Risk of fatal and nonfatal lactic acidosis in patients with type 2 diabetes mellitus treated with
with metformin use in type 2 diabetes mellitus. sulfonylureas: a systematic review and meta-analysis.
Cochrane Database Syst. Rev. (4): CD002967. Diabetes Metab. Res. Rev. 30 (1): 11–22.
209 Stang, M., Wysowski, D.K., and Butler-Jones, D. (1999). 223 Yu, O., Azoulay, L., Yin, H. et al. (2018). Sulfonylureas
Incidence of lactic acidosis in metformin users. Diabetes as initial treatment for type 2 diabetes and the risk of
Care 22 (6): 925–927. severe hypoglycemia. Am. J. Med. 131 (3): 317
210 Brown, J.B., Pedula, K., Barzilay, J. et al. (1998). Lactic e11–e22.
acidosis rates in type 2 diabetes. Diabetes Care 21 (10): 224 Andersen, S.E. and Christensen, M. (2016).
1659–1663. Hypoglycaemia when adding sulphonylurea to
211 Richy, F.F., Sabido-Espin, M., Guedes, S. et al. (2014). metformin: a systematic review and network meta-
Incidence of lactic acidosis in patients with type 2 analysis. Br. J. Clin. Pharmacol. 82 (5): 1291–1302.
Reference 397
25 Simpson, S.H., Lee, J., Choi, S. et al. (2015). Mortality

2 and diabetes: a consensus approach. National Kidney
risk among sulfonylureas: a systematic review and Foundation Hypertension and Diabetes Executive
network meta-analysis. Lancet Diabetes Endocrinol. 3 Committees Working Group. Am. J. Kidney Dis. 36 (3):
(1): 43–51. 646–661.
226 Douros, A., Dell’Aniello, S., Yu, O.H.Y. et al. (2018). 238 Diabetes Control and Complications Trial/Epidemiology
Sulfonylureas as second line drugs in type 2 diabetes of Diabetes Interventions and Complications Research
and the risk of cardiovascular and hypoglycaemic Group, Lachin, J.M., Genuth, S. et al. (2000).
events: population based cohort study. BMJ 362: k2693. Retinopathy and nephropathy in patients with type 1
227 Zafiriou, S., Stanners, S.R., Polhill, T.S. et al. (2004). diabetes four years after a trial of intensive therapy. N.
Pioglitazone increases renal tubular cell albumin uptake Engl. J. Med. 342 (6): 381–389.
but limits proinflammatory and fibrotic responses. 239 Schrier, R.W., Estacio, R.O., Mehler, P.S., and Hiatt, W.R.
Kidney Int. 65 (5): 1647–1653. (2007). Appropriate blood pressure control in
228 Panchapakesan, U., Pollock, C.A., and Chen, X.M. (2004). hypertensive and normotensive type 2 diabetes mellitus:
The effect of high glucose and PPAR-gamma agonists on a summary of the ABCD trial. Nat. Clin. Pract. Nephrol.
PPAR-gamma expression and function in HK-2 cells. Am. 3 (8): 428–438.
J. Physiol. Renal Physiol. 287 (3): F528–F534. 240 de Galan, B.E., Perkovic, V., Ninomiya, T. et al. (2009).
229 Zafiriou, S., Stanners, S.R., Saad, S. et al. (2005). Lowering blood pressure reduces renal events in type 2
Pioglitazone inhibits cell growth and reduces matrix diabetes. J. Am. Soc. Nephrol. 20 (4): 883–892.
production in human kidney fibroblasts. J. Am. Soc. 241 Maki, D.D., Ma, J.Z., Louis, T.A., and Kasiske, B.L.
Nephrol. 16 (3): 638–645. (1995). Long-term effects of antihypertensive agents on
230 Panchapakesan, U., Sumual, S., Pollock, C.A., and Chen, proteinuria and renal function. Arch. Intern. Med. 155
X. (2005). PPARgamma agonists exert antifibrotic effects (10): 1073–1080.
in renal tubular cells exposed to high glucose. Am. J. 242 Voyaki, S.M., Staessen, J.A., Thijs, L. et al. (2001).
Physiol. Renal Physiol. 289 (5): F1153–F1158. Follow-up of renal function in treated and untreated
231 Dormandy, J.A., Charbonnel, B., Eckland, D.J. et al. older patients with isolated systolic hypertension.
(2005). Secondary prevention of macrovascular events Systolic Hypertension in Europe (Syst-Eur) Trial
in patients with type 2 diabetes in the PROactive Study Investigators. J. Hypertens. 19 (3): 511–519.
(PROspective pioglitAzone Clinical Trial In 243 Schrier, R.W., Estacio, R.O., Esler, A., and Mehler, P.
macroVascular Events): a randomised controlled trial. (2002). Effects of aggressive blood pressure control in
Lancet 366 (9493): 1279–1289. normotensive type 2 diabetic patients on albuminuria,
232 Erdmann, E., Charbonnel, B., Wilcox, R.G. et al. (2007). retinopathy and strokes. Kidney Int. 61 (3): 1086–1097.
Pioglitazone use and heart failure in patients with type 2 244 Lewis, J.B., Berl, T., Bain, R.P. et al. (1999). Effect of
diabetes and preexisting cardiovascular disease: data intensive blood pressure control on the course of type 1
from the PROactive study (PROactive 08). Diabetes Care diabetic nephropathy. Collaborative Study Group. Am. J.
30 (11): 2773–2778. Kidney Dis. 34 (5): 809–817.
233 Schneider, C.A., Ferrannini, E., Defronzo, R. et al. 245 ACCORD Study Group, Cushman, W.C., Evans, G.W.
(2008). Effect of pioglitazone on cardiovascular outcome et al. (2010). Effects of intensive blood-pressure control
in diabetes and chronic kidney disease. J. Am. Soc. in type 2 diabetes mellitus. N. Engl. J. Med. 362 (17):
Nephrol. 19 (1): 182–187. 1575–1585.
234 Home, P.D., Jones, N.P., Pocock, S.J. et al. (2007). 246 Margolis, K.L., O’Connor, P.J., Morgan, T.M. et al.
Rosiglitazone RECORD study: glucose control outcomes (2014). Outcomes of combined cardiovascular risk
at 18 months. Diabetes Med. 24 (6): 626–634. factor management strategies in type 2 diabetes: the
235 Home, P.D., Pocock, S.J., Beck-Nielsen, H. et al. ACCORD randomized trial. Diabetes Care 37 (6):
(2007). Rosiglitazone evaluated for cardiovascular 1721–1728.
outcomes – an interim analysis. N. Engl. J. Med. 357 247 Patel, A., ADVANCE Collaborative Group, MacMahon,
(1): 28–38. S. et al. (2007). Effects of a fixed combination of
236 Lago, R.M., Singh, P.P., and Nesto, R.W. (2007). perindopril and indapamide on macrovascular and
Congestive heart failure and cardiovascular death in microvascular outcomes in patients with type 2 diabetes
patients with prediabetes and type 2 diabetes given mellitus (the ADVANCE trial): a randomised controlled
thiazolidinediones: a meta-analysis of randomised trial. Lancet 370 (9590): 829–840.
clinical trials. Lancet 370 (9593): 1129–1136. 248 Hansson, L., Zanchetti, A., Carruthers, S.G. et al. (1998).
237 Bakris, G.L., Williams, M., Dworkin, L. et al. (2000). Effects of intensive blood-pressure lowering and
Preserving renal function in adults with hypertension low-dose aspirin in patients with hypertension:
principal results of the Hypertension Optimal 260 Bakris, G.L., Sarafidis, P.A., Weir, M.R. et al. (2010).
Treatment (HOT) randomised trial. HOT Study Group. Renal outcomes with different fixed-dose combination
Lancet 351 (9118): 1755–1762. therapies in patients with hypertension at high risk for
249 McBrien, K., Rabi, D.M., Campbell, N. et al. (2012). cardiovascular events (ACCOMPLISH): a prespecified
Intensive and standard blood pressure targets in secondary analysis of a randomised controlled trial.
patients with type 2 diabetes mellitus: systematic review Lancet 375 (9721): 1173–1181.
and meta-analysis. Arch. Intern. Med. 172 (17): 261 Parving, H.H., Brenner, B.M., McMurray, J.J. et al.
1296–1303. (2012). Cardiorenal end points in a trial of aliskiren for
250 Abaterusso, C., Lupo, A., Ortalda, V. et al. (2008). type 2 diabetes. N. Engl. J. Med. 367 (23): 2204–2213.
Treating elderly people with diabetes and stages 3 and 4 262 Mann, J.F., Schmieder, R.E., McQueen, M. et al. (2008).
chronic kidney disease. Clin. J. Am. Soc. Nephrol. 3 (4): Renal outcomes with telmisartan, ramipril, or both, in
1185–1194. people at high vascular risk (the ONTARGET study): a
251 Lewis, E.J., Hunsicker, L.G., Bain, R.P., and Rohde, R.D. multicentre, randomised, double-blind, controlled trial.
(1993). The effect of angiotensin-converting-enzyme Lancet 372 (9638): 547–553.
inhibition on diabetic nephropathy. The Collaborative 263 Fried, L.F., Emanuele, N., Zhang, J.H. et al. (2013).
Study Group. N. Engl. J. Med. 329 (20): 1456–1462. Combined angiotensin inhibition for the treatment of
252 Mauer, M., Zinman, B., Gardiner, R. et al. (2009). Renal diabetic nephropathy. N. Engl. J. Med. 369 (20):
and retinal effects of enalapril and losartan in type 1 1892–1903.
diabetes. N. Engl. J. Med. 361 (1): 40–51. 264 Palmer, S.C., Mavridis, D., Navarese, E. et al. (2015).
253 Chaturvedi, N., Porta, M., Klein, R. et al. (2008). Effect Comparative efficacy and safety of blood pressure-
of candesartan on prevention (DIRECT-Prevent 1) and lowering agents in adults with diabetes and kidney
progression (DIRECT-Protect 1) of retinopathy in type 1 disease: a network meta-analysis. Lancet 385 (9982):
diabetes: randomised, placebo-controlled trials. Lancet 2047–2056.
372 (9647): 1394–1402. 265 Bolignano, D., Palmer, S.C., Navaneethan, S.D., and
254 Parving, H.H., Lehnert, H., Brochner-Mortensen, J. et al. Strippoli, G.F. (2014). Aldosterone antagonists for
(2001). The effect of irbesartan on the development of preventing the progression of chronic kidney disease.
diabetic nephropathy in patients with type 2 diabetes. N. Cochrane Database Syst. Rev. 4: CD007004.
Engl. J. Med. 345 (12): 870–878. 266 Pitt, B., Zannad, F., Remme, W.J. et al. (1999). The effect of
255 Heart Outcomes Prevention Evaluation Study spironolactone on morbidity and mortality in patients
Investigators, Yusuf, S., Sleight, P. et al. (2000). Effects of with severe heart failure. Randomized Aldactone
an angiotensin-converting-enzyme inhibitor, ramipril, Evaluation Study Investigators. N. Engl. J. Med. 341 (10):
on cardiovascular events in high-risk patients. N. Engl. J. 709–717.
Med. 342 (3): 145–153. 267 Bakris, G.L., Agarwal, R., Chan, J.C. et al. (2015). Effect
256 Ruggenenti, P., Perna, A., Ganeva, M. et al. (2006). of finerenone on albuminuria in patients with diabetic
Impact of blood pressure control and angiotensin- nephropathy: a randomized clinical trial. JAMA 314 (9):
converting enzyme inhibitor therapy on new-onset 884–894.
microalbuminuria in type 2 diabetes: a post hoc analysis 268 Filippatos G, Anker SD, Agarwal R, Pitt B, Ruilope LM,
of the BENEDICT trial. J. Am. Soc. Nephrol. 17 (12): Rossing P, Kolkhof P, Schloemer P, Tornus I, Joseph A,
3472–3481.BENEDICT Study Group Bakris GL; FIDELIO-DKD Investigators. Finerenone
257 Haller, H., Ito, S., Izzo, J.L. Jr. et al. (2011). Olmesartan and Cardiovascular Outcomes in Patients With Chronic
for the delay or prevention of microalbuminuria in type Kidney Disease and Type 2 Diabetes. Circulation. 2021
2 diabetes. N. Engl. J. Med. 364 (10): 907–917. Feb 9; 143 (6): 540–552. doi: 10.1161/
258 Strippoli GF, Bonifati C, Craig M, et al. Angiotensin CIRCULATIONAHA.120.051898. Epub 2020 Nov 16.
converting enzyme inhibitors and angiotensin II PMID: 33198491; PMCID: PMC7864612.
receptor antagonists for preventing the progression of 269 Pitt B, Filippatos G, Agarwal R, Anker SD, Bakris GL,
diabetic kidney disease. Cochrane Database Syst Rev. Rossing P, Joseph A, Kolkhof P, Nowack C, Schloemer P,
2006 Oct 18; 2006 (4): CD006257. doi: Ruilope LM; FIGARO-DKD Investigators.
10.1002/14651858.CD006257. PMID: 17054288; Cardiovascular Events with Finerenone in Kidney
PMCID: PMC6956646. Disease and Type 2 Diabetes. N Engl J Med. 2021 Aug
259 Weber, M.A., Bakris, G.L., Jamerson, K. et al. (2010). 28. doi: 10.1056/NEJMoa2110956. Epub ahead of print.
Cardiovascular events during differing hypertension PMID: 34449181.
therapies in patients with diabetes. J. Am. Coll. Cardiol. 270 Chung EY, Ruospo M, Natale P, Bolignano D,
56 (1): 77–85. Navaneethan SD, Palmer SC, Strippoli GF. Aldosterone
Reference 399
antagonists in addition to renin angiotensin system in 9795 people with type 2 diabetes mellitus (the FIELD
antagonists for preventing the progression of chronic study): randomised controlled trial. Lancet 366 (9500):
kidney disease. Cochrane Database Syst Rev. 2020 Oct 1849–1861.
27; 10 (10): CD007004. doi: 10.1002/14651858.CD007004. 282 Moriarty, P.M., Jacobson, T.A., Bruckert, E. et al.
pub4. PMID: 33107592; PMCID: PMC8094274. (2014). Efficacy and safety of alirocumab, a
271 Baigent, C., Landray, M.J., Reith, C. et al. (2011). The monoclonal antibody to PCSK9, in statin-intolerant
effects of lowering LDL cholesterol with simvastatin patients: design and rationale of ODYSSEY
plus ezetimibe in patients with chronic kidney disease ALTERNATIVE, a randomized phase 3 trial. J. Clin.
(Study of Heart and Renal Protection): a randomised Lipidol. 8 (6): 554–561.
placebo-controlled trial. Lancet 377 (9784): 2181–2192. 283 Zhang, X.L., Zhu, Q.Q., Zhu, L. et al. (2015). Safety and
272 Cholesterol Treatment Trialists Collaboration, Kearney, efficacy of anti-PCSK9 antibodies: a meta-analysis of 25
P.M., Blackwell, L. et al. (2008). Efficacy of cholesterol- randomized, controlled trials. BMC Med. 13: 123.
lowering therapy in 18,686 people with diabetes in 14 284 Antithrombotic Trialists Collaboration, Baigent, C.,
randomised trials of statins: a meta-analysis. Lancet 371 Blackwell, L. et al. (2009). Aspirin in the primary and
(9607): 117–125. secondary prevention of vascular disease:
273 Cholesterol Treatment Trialists Collaboration, Baigent, collaborative meta-analysis of individual participant
C., Blackwell, L. et al. (2010). Efficacy and safety of data from randomised trials. Lancet 373 (9678):
more intensive lowering of LDL cholesterol: a meta- 1849–1860.
analysis of data from 170,000 participants in 26 285 Vandvik, P.O., Lincoff, A.M., Gore, J.M. et al. (2012).
randomised trials. Lancet 376 (9753): 1670–1681. Primary and secondary prevention of cardiovascular
274 Cannon, C.P., Blazing, M.A., Giugliano, R.P. et al. disease: Antithrombotic Therapy and Prevention of
(2015). Ezetimibe added to statin therapy after acute Thrombosis, 9th ed: American College of Chest
coronary syndromes. N. Engl. J. Med. 372 (25): Physicians Evidence-Based Clinical Practice Guidelines.
2387–2397. Chest 141 (2 Suppl): e637S–e668S.
275 Wanner, C., Krane, V., Marz, W. et al. (2005). Atorvastatin 286 Bhatt, D.L., Bonaca, M.P., Bansilal, S. et al. (2016).
in patients with type 2 diabetes mellitus undergoing Reduction in ischemic events with Ticagrelor in
hemodialysis. N. Engl. J. Med. 353 (3): 238–248. diabetic patients with prior myocardial infarction in
276 Fellstrom, B.C., Jardine, A.G., Schmieder, R.E. et al. PEGASUS-TIMI 54. J. Am. Coll. Cardiol. 67 (23):
(2009). Rosuvastatin and cardiovascular events in 2732–2740.
patients undergoing hemodialysis. N. Engl. J. Med. 360 287 Montalescot, G. and Silvain, J. (2010). Ticagrelor in the
(14): 1395–1407. renal dysfunction subgroup: subjugated or
277 Holdaas, H., Holme, I., Schmieder, R.E. et al. (2011). substantiated? Circulation 122 (11): 1049–1052.
Rosuvastatin in diabetic hemodialysis patients. J. Am. 288 James, S., Budaj, A., Aylward, P. et al. (2010). Ticagrelor
Soc. Nephrol. 22 (7): 1335–1341. versus clopidogrel in acute coronary syndromes in
278 Tonelli, M., Collins, D., Robins, S. et al. (2004). relation to renal function: results from the Platelet
Gemfibrozil for secondary prevention of cardiovascular Inhibition and Patient Outcomes (PLATO) trial.
events in mild to moderate chronic renal insufficiency. Circulation 122 (11): 1056–1067.
Kidney Int. 66 (3): 1123–1130.Veterans’ Affairs 289 Niewczas, M.A., Gohda, T., Skupien, J. et al. (2012).
High-Density Lipoprotein Intervention Trial Circulating TNF receptors 1 and 2 predict ESRD in type
Investigators 2 diabetes. J. Am. Soc. Nephrol. 23 (3): 507–515.
279 Ansquer, J.C., Foucher, C., Rattier, S. et al. (2005). 290 Pavkov, M.E., Nelson, R.G., Knowler, W.C. et al. (2015).
Fenofibrate reduces progression to microalbuminuria Elevation of circulating TNF receptors 1 and 2 increases
over 3 years in a placebo-controlled study in type 2 the risk of end-stage renal disease in American Indians
diabetes: results from the Diabetes Atherosclerosis with type 2 diabetes. Kidney Int. 87 (4): 812–819.
Intervention Study (DAIS). Am. J. Kidney Dis. 45 (3): 283 Wong, M.G., Perkovic, V., Woodward, M. et al. (2013).
485–493.DAIS Investigators Circulating bone morphogenetic protein-7 and
280 Davis, T.M., Ting, R., Best, J.D. et al. (2011). Effects of transforming growth factor-beta1 are better predictors
fenofibrate on renal function in patients with type 2 of renal end points in patients with type 2 diabetes
diabetes mellitus: the Fenofibrate Intervention and mellitus. Kidney Int. 83 (2): 278–284.
Event Lowering in Diabetes (FIELD) study. Diabetologia 292 Hara, A., Furuichi, K., Koshino, A. et al. (2018). Clinical
54 (2): 280–290. and pathological significance of autoantibodies to
281 Keech, A., Simes, R.J., Barter, P. et al. (2005). Effects of erythropoietin receptor in type 2 diabetic patients with
long-term fenofibrate therapy on cardiovascular events CKD. Kidney Int. Rep. 3 (1): 133–141.
400
25
Lupus Nephritis
Matthew A. Roberts1 and David J Tunnicliffe2,3
1
Eastern Health Clinical School, Monash University, Melbourne, VIC, Australia
2
Sydney School of Public Health, The University of Sydney, Sydney, NSW, Australia
3
Centre for Kidney Research, The Children’s Hospital at Westmead, Westmead, NSW, Australia
I ntroduction D
efinition
Lupus nephritis is an important manifestation of the mul- Lupus nephritis is the term for kidney involvement in
tisystem autoimmune disease systemic lupus erythemato- patients with SLE and is defined by clinical features, labo-
sus (SLE) with clinical features ranging from asymptomatic ratory abnormalities, and the histological findings on kid-
abnormalities on urine microscopy, to the nephrotic syn- ney biopsy. Because SLE involves multiple systems, the
drome, to rapid loss of kidney function. It is important to disease is best defined by diagnostic criteria such as the
frame the current evidence base for the treatment of lupus American College of Rheumatology (ACR) [1,2] or the
nephritis into some historical context (Figure 25.1). It is Systemic Lupus International Collaborative Clinics
nearly half a century since the first randomized controlled (SLICC) [3] criteria (Table 25.2). Lupus nephritis is also
trial (RCT) in lupus nephritis was published, and this heterogeneous and is best defined by a specific histological
period can be divided into three eras based on the standard classification, the International Society of Nephrology
of care at the time: steroid alone, steroid plus cytotoxic (ISN) Renal Pathology Society (RPS) Classification
then steroid plus mycophenolate. The first histological (Table 25.3) [4,5].
classifications developed because percutaneous kidney
biopsy enabled analysis of histology in life rather than at
autopsy. Every decade since has seen developments and E
pidemiology
refinements to the way the disease is classified histologi-
cally and thus reported to clinicians by pathologists. Before SLE predominantly affects young females from non-
end-stage kidney disease (ESKD) programs became widely Caucasian populations, particularly African Americans.
available, many patients with lupus nephritis died from The female: male ratio is 9 : 1 [6]. Three SLE surveillance
renal failure. With improved immunosuppressive thera- programs in different United States populations reported
pies, fewer patients now progress to ESKD and the lower almost 4000 prevalent cases of SLE and 752 incident cases
mortality in the modern era may also be attributed to between them over a 3-year period [7–9]. The prevalence
improvements in supportive care with renin-angiotensin per 100 000 population using the ACR criteria in African
system inhibitors to reduce blood pressure and proteinuria, American, Hispanic, Asian, and Caucasian females
increased use of hydroxychloroquine, better management ranged from 182 to 477, 58 to 179, 50 to 165, and 73 to 122,
of complications of treatment such as infection, and better respectively. The corresponding figures in males were
management of cardiovascular risk factors. This chapter 18–59, 12–22, 5–25, and 5–11, respectively. Prevalence
presents the current treatment for lupus nephritis and the data are lacking in African, subcontinental, and South
evidence base behind it (Table 25.1). Asian cohorts [10].
Pathophysiolog 401
Pollak 1964 WHO Buffalo NY WHO Paris, 1980, WHO, Churg 1995: ISN/RPS 2004: ISN/RPS 2018:
Normal 1974, Geneva Churg 1982: Kept Classes I-VI I Minimal (i.e. not Clarifies definitions
Lupus glomerulitis 1975, Appel 1978: I Normal but suggested that “normal”) of terms i.e. term
Histological classification
Active lupus GN I Normal II Mesangial separation into II Mesangial “endocapillary

Membranous II Pure mesangial III Focal & Class III or IV based III Focal LN proliferation”
Iupus GN III Focal segmental on number of IV Diffuse LN replaced by
proliferative GN proliferative GN glomeruli (<50% “Segmental” if “endocapillary
(<50% of gloms) Active necrotizing versus ≥50%) was segmental lesions hypercellularity”
IV Diffuse Active sclerosing not clear-cut in ≥50% of gloms Segmental and
proliferative GN Sclerosing Also suggested “Global” if global Global subdivisions
Baldwin 1970 (≥50% of gloms) IV Diffuse Class III be “focal & lesions in ≥50% of of Class IV removed
Focal proliferative V Membranous GN proliferative GN necrotizing lesions” gloms Proposes use of
LN (as for III) V Membranous LN activity and
Diffuse V Membranous GN VI Advanced chronicity scoring
proliferative LN (+/–II, III or IV) sclerosing LN systems
Membranous LN VI Advanced
sclerosing
Year: 1960 1970 1980 1990 2000 2010 2020

Steroid alone CYC/AZA + Steroid Mycophenolate + steroid
Pollak 1964: Steinberg 1971: Felson& Multiple NIH (& Chan 2000: KDIGO 2012:
Treatment
Better survival in First RCTof Anderson 1984: other) RCTs RCTshowing Major renal
active lupus GN with treatment in LN First meta- refining IV CYC renal remission guideline
high dose steroid analysis of regimens with recommends
than low dose treatment in LN MMF+steroids steroid + either
similar to CYC MMF or CYC for
Austin 1986:
1954: Percutaneous Lewis 1993: ACEi then AZA + Class III or IV
1972: US RCT showing steroids
kidney biopsy first “renoprotective” disease
Other factors
Medicare better renal

described in diabetic
amendment to survival with
(First successful nephropathy
include ESKD in cytotoxic versus
kidney transplant) GISEN Group Improvement of general supportive care
Medicare steroid alone
1957: anti-dsDNA 1997: ACEi beyond blood pressure: infection treatment &
Abs discovered coverage– “renoprotective” prophylaxis, prevention of thrombosis,
1960: First chronic dialysis more in non-diabetic treatment of cardiovascular risk factors
dialysis program available nephropathies
Figure 25.1 Historical perspective of lupus nephritis. CYC, cyclophosphamide; AZA, azathioprine; GN, glomerulonephritis; LN, lupus
nephritis; WHO, World Health Organization; ISN/RPS, International Society of Nephrology/Renal Pathology Society; RCT, randomized
controlled trial; ESKD, end-stage kidney disease; NIH, US National Institutes of Health; ACEi, angiotensin converting enzyme inhibitor;
GISEN, Gruppo Italiano di Studi Epidemiologici in Nefrologia; MMF, mycophenolate mofetil; KDIGO, Kidney Disease Improving Global
Outcomes Guideline.
An international inception cohort of 1826 patients with immune complexes either deposit in, or form within, the
SLE reported that 38.3% of people with SLE have a “renal kidney. This leads to complement activation, inflamma-
disorder” [11]. In the United States surveillance studies [7–9], tion, and ultimately the various histopathological lesions
the proportion with “renal disorder,” defined as proteinuria and clinical manifestations of lupus nephritis.
or urinary casts in the ACR criteria, ranged from 32 to 45% Preceding these events is genetic susceptibility. Genetic
and, like SLE itself, was consistently highest in African factors play an important role in SLE and lupus nephritis,
Americans and lowest in Caucasian Americans. Lupus and the wide variety of gene loci that increase the risk of
nephritis may also differ by sex. In a meta-analysis of clinical SLE and lupus nephritis go some way to explaining the
features by sex, males had greater odds of renal involvement clinical heterogeneity of the disease. Concordance of SLE
(OR 1.51, 95% CI 1.31–1.75), whereas females were more in monozygotic twins is 24% compared to 2% in dizygotic
likely to have skin, joint, and serosal involvement [6]. twins and some specific alleles of the major histocompati-
bility complex on chromosome 6 are strongly associ-
P
athophysiology ated [10]. For example, having both HLA-DRB1*03 : 01 and
DRB1*15 : 01 alleles conferred greater risk for SLE than
In SLE there is disordered immune regulation leading to being homozygous for one allele in a large genotype asso-
loss of self-tolerance, production of autoantibodies, and ciation study [11]. In this study, the greater the number of
formation of immune complexes. In lupus nephritis, these risk polymorphisms, the greater the odds of SLE. In African
Table 25.1 Treatment for lupus nephritis.
Strength of
Treatment Recommendation Certainty of evidence (relevant outcome measure) recommendation
All patient with lupus nephritis

We recommend that all patients with lupus nephritis be treated Low Strong [13–16]
with hydroxycholoroquine.
ISN/RPS Class I disease

We suggest treatment only for extrarenal manifestations of SLE. Very low Weak
ISN/RPS Class II disease
We suggest treatment only for extrarenal manifestations of SLE. Very low Weak
Consider immunosuppressive treatment if substantial
proteinuria or evidence of podocytopathy.
ISN/RPS Class III or IV disease, with or without Class V
Induction
Mycophenolate We suggest mycophenolate (i.e. 3 g per day of mycophenolate Low (complete remission, partial remission) Weak [17–18]
mofetil in divided doses) be used for initial treatment of lupus Moderate (alopecia with cyclophosphamide,
nephritis Class III or IV because it is equivalent in inducing diarrhea with mycophenolate)
remission to cyclophosphamide but has an overall more
favorable toxicity profile.
Cyclophsophamide We suggest that intravenous cyclophosphamide (either 0.5–1.0 g/ Low (complete remission, partial remission) Weak [17–18]
m2 monthly or 0.5 g fortnightly for six doses) be used as an
alternative to mycophenolate to induce remission in lupus
nephritis because it is equivalent to mycophenolate. This could
be for patients who do not tolerate mycophenolate or whose
disease is resistant to mycophenolate.
Mycophenolate plus tacrolimus We suggest that mycophenolate mofetil plus tacrolimus be Low (complete remission, stable kidney function) Weak [17–18]
considered as a third alternative as although it was better than
intravenous cyclophosphamide at inducing complete remission,
results are from a single center with participants largely of Asian
ethnicity.
c25.indd 402 09-12-2022 15:46:46

Corticosteroid We recommend that corticosteroids be included with the above Moderate (major infection) Strong [17]
regimens as they were a component of these regimens in all
RCTs. The dose (prednisolone 0.6–1.0 mg/kg/day) and use of
intravenous methyprednisolone should be dictated by severity of
renal and extrarenal disease and reduced as quickly as possible.
It should be noted that major infections were similar in
comparison with prednisolone alone with prednisolone plus
cytotoxic.
Maintenance
Mycophenolate We recommend mycophenolate be continued as maintenance Moderate (renal relapse, doubling of serum Strong [17–18]
therapy because treatment results in less renal relapse and creatinine)
doubling of serum creatinine than azathioprine and with a more Low (leucopenia)
favorable toxicity profile than azathioprine. Whether and when
to withdraw maintenance therapy is not known.
ISN/RPS Class V disease with no proliferative (Class III or IV) disease

Nonimmunosuppressive We suggest that all patients with Class V lupus nephritis receive Low Strong
“anti-proteinuria” therapy with blockade of the renin-
angiotensin system.
Immunosuppressive We suggest that patients with pure Class V lupus nephritis who Low Weak [17]
have nephrotic syndrome or persistent nephrotic range
proteinuria can be treated with similar immunosuppression to
Class III or IV disease.
Low Weak [19]
We suggest that a calcineurin inhibitor could be considered as it
was effective in inducing remission in one RCT.
We suggest that for patients with Class V lupus nephritis and Low Weak
proteinuria below 3 g/day, the decision to treat with
immunosuppression and the choice of immunosuppression be
guided by individual patient risk factors.
ISN/RPS Class VI disease

We recommend immunosuppressive therapy not be given to Very low Strong
patients with Class VI disease.
GRADE assessment of the certainty of the evidence [20]: High: This research provides a very good indication of the likely effect. The likelihood that the effect will be substantially different* is
low. Moderate: This research provides a good indication of the likely effect. The likelihood that the effect will be substantially different* is moderate. Low: This research provides some
indication of the likely effect. However, the likelihood that it will be substantially different* is high. Very low: This research does not provide a reliable indication of the likely effect. The
likelihood that the effect will be substantially different* is very high. *Substantially different = a large enough difference that it might affect decision-making.
ISN/RPS, International Society of Nephrology/Renal Pathology Society; RCT, randomized controlled trial; SLE, systemic lupus erythematosus.
c25.indd 403 09-12-2022 15:46:46

404 Lupus Nephritis
Table 25.2 Major diagnostic criteria for the diagnosis of SLE.
Organ ACR criteria [1,2] SLICC criteria [3]
Skin Malar rash Acute cutaneous lupus

Discoid rash Chronic cutaneous lupus
Photosensitivity
Mucosa Oral ulcers Oral ulcers
Hair Alopecia
Joint Nonerosive arthritis in 2 joints Synovitis in 2 joints
Serosa Pleuritis or pericarditis Typical pleurisy or pericardial pain lasting 1 day
Kidney Renal disorder = proteinuria>0.5 g/day or > 3+ on Renal = proteinuria>0.5 g/day or equivalent
dipstick OR cellular casts of any type protein:creatinine ratio OR red blood cell casts
Nervous system Seizures OR psychosis Seizures, psychosis, mononeuritis multiplex, myelitis,
peripheral or cranial neuropathy, or acute confusion
Hematological Haemolytic anemia OR leukopenia (<4000/ml twice) Haemolytic anemia
OR lympnopenia (<1500/ml twice) OR Leukopenia (<4000/ml once) OR lympnopenia (<1000/
thrombocytopenia (<100 000/ml) ml once)
Thrombocytopenia (<100 000/ml)
Immunological Immunological: positive lupus erythrmatosus cell test Positive ANA
OR anti-dsDNA OR anti-Sm antibody OR false-
positive syphilis serology
Abnormal ANA titer Elevated anti-dsDNA
Positive anti-Sm
Positive anti-phospholipid antibody
Low complement
Direct coombs test
ACR, American College of Rheumatology; ANA, antibodies directed against nuclear components; anti-dsDNA, anti-double-stranded DNA; SLE,
systemic lupus erythematosus; SLICC Systemic Lupus International Collaborative Clinics.
American patients, the coding alleles G1 and G2 of the “BAFF”), A Proliferating Inducing Ligand (APRIL) and
APOL1 gene are strongly associated with nondiabetic kid- co-stimulatory molecules are targets of novel therapies
ney disease. In lupus nephritis, the presence of two APOL1 (Figure 25.2) [22].
gene risk alleles increased the odds of developing ESKD 2- This immune activation and proliferation of autoreactive
to 3-fold, and development of ESKD occurred 2 years ear- T and B cell populations leads to production of various
lier than in people without two risk alleles [21]. antibodies directed against nuclear material, including
After genetic susceptibility, the next event in developing anti-dsDNA. Although clearly important to measure clini-
SLE is the loss of immune tolerance to endogenous nuclear cally and pathogenic in experimental models, the precise
material with autovaccination leading to production of role anti-dsDNA antibodies play in lupus nephritis is still
antinuclear antibodies [22]. When cells die, the presence being elucidated. Binding directly to cross-reactive anti-
of nuclear material in the extracellular space is minimized gens such as annexin II or indirectly to chromatin material
by apoptotic cell death or phagocytic clearance. If the that deposits in the kidney leads to effects such as cell pro-
methylation state of nuclear material is altered, a failure to liferation and recruitment of inflammatory cells [24].
recognize the nuclear material as “self” and, worse, inter- The variable pattern of immune complex deposition con-
preting the material to be viral nucleic acid by Toll-like tributes to the diversity of histological lesions in lupus
receptors on antigen presenting cells, leads to immune nephritis. Immune complex deposition or formation
activation by various mechanisms and production of auto- restricted to the mesangium generally leads to milder dis-
antibodies such as anti-double-stranded DNA (anti- ease (ISPN/RPS Class I and II), immune complexes in the
dsDNA) [23]. Some of these mechanisms, such as B subendothelial space causes the most kidney-threatening
lymphocyte stimulator (Blys or B-cell activating factor forms of lupus nephritis (ISPN/RPS Class III and IV), and
Diagnosi 405
Table 25.3 The ISN/RPS classification of lupus nephritis with 2018 clarifications.
Class Description Detailed description ISN/RPS 2018 Clarifications
I Normal Normal by light microscopy (LM), mesangial

immune deposits may be present on
immunofluorescence (IF) or electron
microscopy (EM)
II Mesangial disease Mesangial hypercellularity; subepithelial or Mesangial hypercellularity defined as
subendothelial immune deposits not visible 4 mesangial cell nuclei surrounded by matrix
by LM, but may be evident on IF or EM and away from the hilum
III Focal lupus nephritis Focal (<50% of glomeruli) segmental or The term “endocapillary hypercellularity”
global disease; lesions may include: replaces the term “endocapillary proliferation”
●● endocapillary hypercellularity as cells other than endothelial cells may be
●● subendothelial immune deposits visible
present.
on LM as “wire loop” lesions Adhesions, fibrinoid necrosis specifically
●● adhesions
defined
●● fibrinoid necrosis
Definition of crescent changed to extracapillary
hypercellularity involving 10% (not 25%) of the
●● crescents
circumference of Bowman’s capsule; cellular,
fibrocellular and fibrous crescents defined by
proportion of fibrous and cellular elements
New classification recommends not using the
categories of global (G) and segmental (S), and
not using terms for active (A), chronic (C) and
active and chronic (A/C) lesions; rather, a
semiquantitative scoring system is suggested
IV Diffuse lupus nephritis Diffuse ( 50% of glomeruli) segmental or Clarifications as per Class III
(sometimes referred to as global disease; lesions as per Class III but
“diffuse proliferative more widespread; features of
lupus nephritis”) membranoproliferative (mesangiocapillary)
pattern may be present
V Membranous lupus Subepithelial immune deposits in a typical
nephritis membranous pattern; there may be
mesangial deposits and hypercellularity
May be present with focal (Class V + III) or
diffuse (Class V + IV) disease
VI Advanced sclerosing Global glomerulosclerosis in 90% of Likely to be removed in future classifications as
lupus nephritis glomeruli rarely seen
ISN/RPS, International Society of Nephrology/Renal Pathology Society.
subepithelial immune complexes cause membranous dis- Histological Classification of Lupus Nephritis
ease (ISPN/RPS Class V). This binding to intrarenal autoan-
The current histological classification of lupus nephritis has
tigens results in complement activation, activation of Fc
been developed over many decades and continues to evolve
receptors, recruitment of inflammatory cells, and tissue
(Figure 25.1 and Table 25.3) [4,5]. These classifications
inflammation [23]. Untreated, this inflammation leads to
formed an important component of the inclusion criteria
tissue damage, fibrosis, and loss of kidney function.
for the RCTs that provide the evidence base for treatment of
lupus nephritis and an appreciation of their evolution over
time is important. The earliest classification of the glomeru-
D
iagnosis lar lesions in SLE was published in 1964 by Pollak et al. [25],
who analyzed kidney tissue from 176 biopsy and necropsy
The diagnosis of lupus nephritis ultimately rests on the specimens collected between 1953 and 1962 to study the
kidney biopsy and histological classification of the lesion. natural history of the disease. They classified what they saw
Because the histological classification is pertinent to subse- under the microscope as normal, lupus glomerulits with
quent discussion, this will be described first. “mild local and focal changes in the glomerular tuft,” active
406 Lupus Nephritis
Dysregulated cell death and dead cell clearance
Opsonization Apoptosis
Phagocytosis
DNA digestion
Dying
Circulating particles containing lymphocyte
Necrosis
(a) immunostimulatory nucleic acids
Autovaccination with nuclear antigens Obinutuzumab

Rituximab
Milatuzumab
CD74 Ocrelizumab
MHC-II TCR CD28 CD80/86 CD20
T cell
B cell
CD80/86 CD28
Dendritic cell CLTA4-B7 CD22
CD40L CD40
IFN-α Anifrolumab Abatacept Anti-CD40 ligand Epratuzumab
Cyclophosphamide Chloroquine
Belimumab
Cell proliferation Mycophenolate mofetil Leflunomide Blys
Tabalumab
Azathioprine Tacrolimus
Proteasome
Immune complexes Autoantibodies Bortezomib

Plasma cell
(b)
In situ immune complex formation and glomerulonephritis
Macrophages TWEAK Anti-tweak

IFN-α Infliximab
IL-6
Sirukumab
Complement
factor Tissue
Cytokines inflammation
Laquinimod
Prednisolone
Lupus nephritis Glomerulus Neutrophils
(c)
Figure 25.2 Pathogenesis of lupus nephritis. Dysregulated cell death (a) leads to development of autoantibodies (b) and ultimately
immune complexes that result in kidney injury. The site of action of established and novel treatments is indicated. result in kidney injury ( c ).
Source: Courtesy of Anders & Rovin [23]. © 2016, Elsevier.
Diagnosi 407
lupus glomerulonephritis with “a more severe proliferative multiple systems. Two diagnostic criteria (Table 25.2) are
and membranous process, frequently associated with widely used to help clinicians make a diagnosis of SLE and
necrotizing features,” or membranous lupus glomerulone- report sensitivity and specificity for a diagnosis of SLE to be
phritis with “diffuse thickening of the glomerular basement greater than 90% [3].
membrane.” Baldwin et al. published a similar study in The “renal disorder” criteria in both classifications are
1970 and used the terms “focal proliferative lupus nephri- based on urinary findings: a measure of proteinuria and
tis” and “diffuse proliferative lupus nephritis” in place of the presence of red blood cells or casts in the urine. In a
“lupus glomerulitis” and “active lupus glomerulonephri- predominantly Caucasian population of 499 patients with
tis,” respectively [26]. The first World Health Organization lupus nephritis, 35% presented with urinary abnormalities
(WHO) classification was formulated at conferences in such as hematuria or proteinuria, 35% presented with
Buffalo, New York in 1974 and Geneva, Switzerland in 1975 nephrotic syndrome, 18% presented with nephritic syn-
but did not appear in peer-reviewed literature until 1978, drome, and 6% had rapidly progressive renal failure [30].
when it was used “with minor changes” in a study reporting Over time, the proportion with nephrotic syndrome has
the clinical course of 56 patients over time [27]. This classi- remained stable and the more severe clinical presentations
fication introduced mesangial lesions as a distinct entity. declined (Figure 25.3).
Advanced sclerosing glomerulonephritis was added at the The work-up of a patient with suspected lupus nephri-
International Study of Kidney Diseases in Children Meeting tis, whether or not they have known SLE, takes into
in Paris in 1980 and this classification was published in account clinical symptoms, assessment of glomerular dis-
Churg’s Renal Disease: Classification and Atlas of ease activity, and serological abnormalities. Glomerular
Glomerular Diseases [28]. Further revisions were made in disease activity is assessed with urine microscopy for evi-
1995 and published in the second edition of that text- dence of glomerular bleeding such as dysmorphic red
book [29]. Along the way, debate about the distinction blood cells and casts, assessment of urine protein excre-
between Class III and Class IV disease, and definitions of tion, and measurement of kidney function. Serological
terms such as “proliferative” has continued. A Consensus testing in lupus includes measurement of complement
Conference at which the ISN and RPS combined to refine activity, with C3 and C4 being the most widely available
this classification took place at Columbia University, tests, antibodies directed against nuclear components
New York in 2002 [4] and this ISN/RPS classification, which (ANA), which are sensitive but not specific for lupus, and
is now in widespread use (Table 25.3) continues to be anti-dsDNA antibodies, which are more specific for lupus.
refined in order to improve clarity and reproducibility [5]. Other evaluations might include general measures of
Patients who have been recruited into RCTs of therapy inflammation such as erythrocyte sedimentation rate
have generally had a kidney biopsy demonstrating ISN/ (ESR) or C-reactive protein (CRP) and evaluation of
RPS Class III, IV, and/or V lupus nephritis, or equivalent in extracted nuclear antigens (ENA), which include various
older classifications, and had to meet the ACR or SLICC antibodies associated with specific SLE manifestations. A
criteria for a diagnosis of SLE. In clinical practice it is pos- person with lupus nephritis will typically have positive
sible to have lupus nephritis diagnosed on kidney biopsy ANA, depressed levels of complement components C3
without strictly meeting all of the clinical criteria. and C4, and raised anti-dsDNA.
Because of the high prevalence of kidney disease in SLE,
careful periodic assessment of kidney function and glo-
Clinical Presentations of Lupus Nephritis
merular disease activity as described above is also part of
The kidney is but one organ involved in SLE, a chronic con- the ongoing management of patients with SLE. Patients
dition that relapses and remits over time. Thus, lupus who have been diagnosed with or treated for lupus nephri-
nephritis can be the first presentation of SLE, can occur in tis require close monitoring of these parameters and a fall
someone with a known diagnosis of SLE, or can recur in in C3 and C4 and a rise in anti-dsDNA may precede onset
someone with SLE who has previously had lupus nephritis. of significant kidney disease.
Given the epidemiology of lupus nephritis, young non- Importantly, definitions of renal remission and renal
Caucasian females presenting with features of glomerular relapse in clinical trials are based on the glomerular dis-
disease such as nephrotic syndrome should have lupus ease features, particularly proteinuria. Although there is
nephritis high on the list of differential diagnoses, even in no consensus for the definition of remission in lupus
the absence of known SLE. The presence of systemic symp- nephritis [14], a common definition of complete remission
toms such as fever or weight loss, musculoskeletal, serosal following induction therapy is the return of serum creati-
or skin features should raise that clinical suspicion even nine to normal, urinary protein excretion to less than
higher. Clinical features of SLE are variable and involve 0.5 g/day, and an inactive urinary sediment. This was used
408 Lupus Nephritis
100 3.9% (7) 3.4% (8) Figure 25.3 Distribution of (a) clinical

14.2% (15) features of lupus nephritis and (b) histological
12.4% (29)
20.3% (32) classification of lupus nephritis a single center
80 across three eras. Source: Courtesy of Moroni
29.2% (31)
et al. [30]. © 2018, BMJ Publishing Group Ltd.
Proportion of patients
35.4% (83)
60
37.3% (59)
30.2% (32)
40
48.9% (115)
20 38% (60)
26.4% (28)
0
1970–1985 1986–2001 2002–2016
Urinary abnormalities Acute nephritic syndrome

Nephrotic syndrome Rapidly progressive renal failure
(a)
100
16.2% (38)
20% (21) 21.7% (34)
80
Proportion of patients
60 51.1% (120)
53.3% (54)
58% (91)
40
20 27.2% (64)
21.9% (23)
17.8% (28)
4.8% (5) 2.5% (4) 5.5% (13)
0
1970–1985 1986–2001 2002–2016
Class II Class III Class IV Class V

(b)
in the 2018 Cochrane review update [17]. A common defi- Role and Interpretation of the Kidney Biopsy
nition of partial remission is a fall in proteinuria to less in Lupus Nephritis
than 3.0 g/day if baseline proteinuria was greater than
A kidney biopsy is critical to making a diagnosis of lupus
3.0 g/day, or a greater than 50% reduction if baseline pro-
teinuria was less than 3.0 g/day in addition to stabilization nephritis and determining treatment. The indications for
of serum creatinine to within 25% of baseline, as used in performing a kidney biopsy in SLE are not dissimilar to
the induction phase of the Aspreva Lupus Management non-SLE presentations. Presentations with the nephrotic
Study (ALMS) Group study [31]. Renal relapse, or “flare,” syndrome, nephritic syndrome, and/or rapidly progressive
in clinical trials is defined variously as an increase in pro- kidney failure warrant a kidney biopsy. Hematuria and/or
teinuria and/or creatinine. For example, in the mainte- non-nephrotic proteinuria in patients with SLE are also
nance phase of the ALMS study, this required a doubling strong reasons to consider a kidney biopsy, particularly
of urinary protein:creatinine ratio from the end of induc- proteinuria of 0.5 g/day or greater. Because clinical fea-
tion with proteinuria above either 1 or 2 g/day depending tures don’t always match histological features, a lower
on whether the postinduction value was above or below threshold should be applied to the decision to perform a
0.5 g/day [32]. kidney biopsy in patients with SLE. The potential risks of
Treatmen 409
the procedure should be weighed against the potential RPS report [5]. This report also recommended against the
benefits of a kidney biopsy in each clinical situation. The terms “global” and “segmental” that were introduced with
benefits of a kidney biopsy in patients with suspected the 2002 Classification as their clinical significance was
lupus nephritis include (i) the appropriate classification of uncertain.
the renal lesion to enable an appropriate level of immuno-
suppressive therapy (i.e. pure membranous versus mem-
branous with proliferative lesions), (ii) identification of P
rognosis
alternative diagnoses or other renal pathology, including
thrombotic microangiopathy in antiphospholipid syn- The natural history of lupus nephritis was described in the
drome, and (iii) assessment of tubule-interstitial damage first description of the different histological appearances
that is important for prognosis [33]. On occasions, one by Pollak [25]. After up to 8 years of follow-up, eight of 10
might consider treating a patient with a severe clinical patients with normal kidney biopsy survived, 17 of 23 with
presentation (i.e. overt nephrotic syndrome) and clinical lupus glomerulitis survived, 15 of 47 with active lupus glo-
and laboratory features of SLE without a kidney biopsy. merulonephritis survived, and all seven patients with
However, the arguments in favor of kidney biopsy are suf- membranous glomerulonephritis survived. Twenty-six of
ficiently strong in most cases. Furthermore, because lupus the deaths in the active lupus glomerulonephritis group
nephritis is a disease that remits with treatment but can were attributed to renal failure. Similarly, in one of the ear-
relapse over time, many patients with lupus nephritis have lier RCTs for lupus nephritis [35], 50 patients were enrolled,
multiple kidney biopsies over the course of their disease. 23 died, and 15 deaths were attributed as “renal death.”
After the biopsy is performed, the clinician should Only one of the 15 “renal deaths” received renal replace-
review the histology face to face with the renal pathologist ment therapy, the rest dying of renal failure at an average
to marry the clinical, laboratory, and histological features age of just over 20 years.
of each patient. It is insufficient to just determine which The efficacy of modern therapy, including modern sup-
“Class” of lupus nephritis one is dealing with. Face-to-face portive care, is demonstrated by the improvements in
review enables discussion of the adequacy of the biopsy 5-year survival of Class IV lupus nephritis over the decades
upon which therapeutic decisions with important conse- from 17% in the 1950s and 1960s to 55%, 80%, and 82% in
quences are being made. Given the classification relies on the decades that followed [36]. In 499 patients with lupus
the proportions of glomeruli involved, at least 10 glomer- nephritis referred to Italian referral centers from 1970 to
uli should be available for light microscopic examina- 2016, 19.8%, 5.9%, and 3.6% died in the periods 1970–1985,
tion [4], but preferably more. Immunofluorescence for 1986–2001, and 2002–2016, respectively. Corresponding
immunoglobulin and complement components is also figures for ESKD were 24.8%, 9.1%, and 1.3% [30]. In the
essential as lupus nephritis usually has extensive immune international inception cohort of the SLICC investigators
deposition, particularly of IgG, C3, and C1q, and lack of recruited between 1999 and 2012, 700 of 1827 were consid-
immune deposition should prompt consideration of other ered to have lupus nephritis though only 395 had a kidney
diagnoses. Activity and chronicity have been designated biopsy [11]. This showed predominantly Class III (26.8%),
“A,” “C” or “A/C” in older histological classifications of Class IV (43.2%), and Class V (31.8%) disease. In patients
lupus nephritis but the most recent report recommends an with lupus nephritis, the 10 year cumulative incidence of
activity and chronicity scoring system rather than these ESKD was 10.1% (95% CI 6.6–13.6%) compared to 0.5% in
designations [5]. Such a scoring system is based on that those without lupus nephritis. The 10-year cumulative
derived by National Institutes of Health (NIH) investiga- incidence of death was 5.0% (95% CI 2.3–7.6%) with lupus
tors where a score of 0, 1, 2, or 3 was applied to histological nephritis and 3.6% (95% CI 0.9–6.2%) without.
features of activity such as glomerular cell proliferation,
leukocyte exudation, karryorrhexis, fibrinoid necrosis, cel-
lular crescents, hyaline deposits, and interstitial inflam- T
reatment
mation [34]. A similar score was applied to features of
chronicity: glomerular sclerosis, fibrous crescents, tubular Lupus nephritis is a classic autoimmune condition that
atrophy, and interstitial fibrosis. This report assessed this requires giving enough immunosuppression to treat the
semiquantitative scoring against outcome and reported disease adequately, but not enough to cause harm. The
that the combination of tubular atrophy, glomerular scle- goals of treatment are (i) to induce remission and relieve
rosis, and cellular crescents portended the poorest renal symptoms of nephrotic syndrome and (ii) to preserve kid-
outcomes. A modified version of this scoring system has ney function in the long term. The first goal is relatively
been proposed for further assessment in the latest ISN/ straightforward to assess in RCTs (although definitions of
410 Lupus Nephritis
remission may vary) but the second goal requires very cussed in the following sections with treatment of pure
long-term follow-up of relatively infrequent outcomes. The membranous Class V disease considered separately.
challenge is to simultaneously achieve both of these goals Class VI disease is essentially near-ESKD and so immu-
and minimize toxicity of therapy. nosuppression is not recommended unless it is required for
extrarenal disease. These patients generally require prepa-
ration for renal replacement therapy.
Who to Treat
As stated, treatment decisions should be made after careful
Therapies Recommended for all Patients
review of the clinical data and kidney biopsy. In general,
ISN/RPS Class I lesions have a good prognosis and do not Nonimmunosuppressive Therapy
require specific therapy apart from what is required to treat As in other forms of glomerulonephritis, non-immunosup-
extrarenal manifestations of SLE. For Class II disease, the pressive therapy for lupus nephritis is intended to reduce
minority in biopsy studies, the risk of progressing to more proteinuria and control hypertension. Although RCTs
kidney-threatening lesions is difficult to determine. assessing the kidney protective effects of renin angiotensin
Published reports have few patients, and those that are system inhibitors have not been performed specifically in
biopsied again are biopsied for a worsening clinical presen- lupus nephritis, these agents delay progression in non-dia-
tation. Of 19 patients from Hong Kong whose initial biopsy betic proteinuric nephropathies [41]. Given that hyperten-
demonstrated Class II lupus nephritis (5% of lupus nephri- sion often requires treatment in lupus nephritis, it is
tis biopsies at that institution), 10 required a second biopsy reasonable to extrapolate such data to lupus nephritis and
a median of 33 months (range 7–109) after the first [37]. Of recommend the use of renin-angiotensin system inhibitors.
these, nine demonstrated more severe disease on the sec- In addition to hypertension, other cardiovascular risk
ond biopsy (Class III 2, Class IV 5, Class V 2). Similarly, of factors warrant attention as people with SLE are at 2–3-fold
41 Argentinian patients who had an initial biopsy with greater risk of cardiovascular disease [42] and this risk is
Class II lupus nephritis, 18 had a second biopsy a median greater again in patients with lupus nephritis [43].
of 40 months (range 11–120) later for clinical flares and 17 Traditional cardiovascular risk factors likely combine with
demonstrated more severe disease (Class III 4, Class IV 10, inflammation (particularly interferon pathways) and ther-
Class V 3) [38]. The Kidney Disease Improving Global apies such as corticosteroids to augment this risk [44, 45].
Outcomes (KDIGO) Guidelines [39] suggest treatment Again, in the absence of RCTs specifically in people with
directed only at extrarenal manifestations of SLE if pro- lupus nephritis, one is left to extrapolate the benefits of
teinuria is less than 1 g/day and immunosuppressive ther- lipid-lowering therapy in non-lupus studies. It is notewor-
apy as for minimal change disease if proteinuria is more thy that systematic reviews have demonstrated the benefit
than 3 g/day (graded 2D in the GRADE system, the lowest of lipid lowering in women, who make up the majority of
level of evidence). The Joint European League Against SLE patients, as well as men [46], and in people with
Rheumatism and European Renal Association–European chronic kidney disease (CKD) [47]. Thus, controlling blood
Dialysis and Transplant Association (EULAR/ERA-EDTA) pressure, smoking cessation, and lipid-lowering therapy
Guidelines also suggest no treatment in Class II disease if are important considerations for prevention of cardiovas-
proteinuria is less than 1 g/day; however, this guideline cular disease in the long term.
suggests 0.25–0.5 mg/kg/day of prednisolone if proteinuria Supportive therapy also includes prevention of venous
is above 1 g/day (with other immunosuppressive therapy if thromboembolism in patients who are nephrotic, prophy-
required to taper steroid) [40]. Thus, immunosuppressive laxis for opportunistic infections such as Pneumocystis
therapy in Class II disease may be used where a podocy- jirovecii, prevention and management of corticosteroid-
topathy is considered possible, which emphasizes the induced osteoporosis, and active management of fertility
importance of a thorough and thoughtful review of the issues in females and males who may be exposed to large
kidney biopsy with the renal pathologist. However, there is cumulative doses of cyclophosphamide.
no RCT evidence that treating Class II lupus with immuno-
suppression prevents progression to more severe lupus Antimalarial Therapy
nephritis (Classes III–V). Antimalarial therapy with hydroxychloroquine is recom-
The majority of RCTs in lupus nephritis have enrolled mended for all patients with lupus nephritis. Although this
patients with Class III, IV, and/or V disease, or equivalent is not based on high-quality RCTs and systematic reviews,
in older classifications [25]. Thus, there is a good evidence analysis of a mixture of randomized and non-randomized
base for treating these lesions, which are known to have a studies suggests that antimalarials are effective at prevent-
much worse prognosis than other lesions. This will be dis- ing lupus flares and prolonging survival [13,15].
Treatmen 411
Antimalarial therapy is widely recommended for the treat- high-dose corticosteroids survived compared to one of 16
ment of constitutional symptoms in SLE guidelines [14] patients receiving low-dose corticosteroids. Although cor-
and may delay progression to an estimated glomerular fil- ticosteroids have not been compared to placebo in an RCT,
tration rate (eGFR) <60 ml/min [16]. The most concerning they have remained an integral component of treatment
adverse effect is retinal toxicity, which is greater with chlo- regimens. A common starting dose is 1 mg/kg daily (up to
roquine than hydroxychloroquine, and the Royal College 60 mg/day) with varying rates of tapering. Intravenous
of Ophthalmologists recommends annual screening for methylprednisolone is included in some regimens and may
retinopathy after taking hydroxychloroquine for more than be considered for more severe clinical presentations, such
5 years, or earlier if receiving a dose of more than 5 mg/kg/ as with crescents on biopsy. Although effective in gaining
day, renal impairment, or concurrent tamoxifen use [48]. control of active disease, corticosteroids have substantial
The risk of retinal toxicity is considered low on a dose of metabolic, cosmetic, and other toxicities that influence tol-
200 mg daily, which is sufficient for most people. erability and adherence. There is therefore much merit in
finding treatment regimens that use minimal corticoster-
oid doses [49].
Immunosuppressive Therapy for Class III and
IV Disease, Alone or Combined with Class V
Cyclophosphamide or Azathioprine plus
Induction Versus Maintenance Therapy Corticosteroids
Immunosuppressive therapy aims to induce and maintain The benefit of adding cytotoxic therapy to corticosteroids
disease remission to maximize patient and renal survival. was demonstrated in a pooled analysis of eight trials that
Induction therapy may last for 6 months or more. included between 13 and 50 participants each [50]. Patients
Important clinical outcomes such as death and ESKD are receiving steroid combined with cytotoxic therapy (either
more difficult to study in RCTs because they require either cyclophosphamide or azathioprine) were less likely to
a very large sample size or very long follow-up. RCTs of experience renal deterioration, ESKD or die as a result of
induction therapy have been of insufficient follow-up their nephritis compared to patients receiving steroids
(median 12 months, range 2.5–48 months) and sample size alone. All-cause mortality did not differ. The first large
(only three trials have recruited more than 350 partici- RCT randomized 107 patients between 1969 and 1981 to
pants) to assess these outcomes [17]. Complete remission prednisolone either alone, with azathioprine, with oral
at 24 weeks is commonly reported and generally defined as cyclophosphamide, with azathioprine plus oral cyclophos-
proteinuria less than 0.5 g/day with normal (or near nor- phamide, or with intravenous cyclophosphamide [51]. The
mal) serum creatinine and normal urinalysis. Maintenance last two groups were introduced part-way through the trial
therapy follows induction therapy and cannot be consid- in 1973. Oral and intravenous cyclophosphamide in combi-
ered as entirely separate. Studies of maintenance therapy nation with corticosteroids was superior to corticosteroids
require much longer follow-up and outcomes include dou- alone in preventing doubling of serum creatinine and
bling of serum creatinine, ESKD, death, and renal flare. ESKD, and oral and intravenous cyclophosphamide exhib-
The definition of renal flare is dependent on how com- ited similar efficacy and safety. This was the first of a series
plete a remission was achieved after induction. Measures of trials performed through the National Institutes of
comprising the definition of renal flare in one RCT [32] Health in the United States, the so-called “NIH” studies.
included a doubling of a measure of proteinuria with pro- The NIH regimen consisted of six 0.5–1.0 g/m2 doses of
teinuria above 1 g/day (or 2 g/day depending on earlier intravenous cyclophosphamide every month adjusted
response), a 25% increase in creatinine, and new or according to the white cell count, along with prednisolone
increased red blood cells or casts on urine microscopy. 0.5 mg/kg/day. This regimen was then compared to a simi-
Regimens for induction and maintenance therapy are lar dose of prednisolone plus six infusions of methylpred-
described according to the three major eras of treatment nisolone 1.0 g/m2 every month (steroid alone), and a third
for lupus nephritis (Figure 25.1). arm in which patients continued to have intravenous cyclo-
phosphamide every 3 months for 2 years after the initial six
Corticosteroids Alone doses of cyclophosphamide (cyclophosphamide “mainte-
The earliest reports of lupus nephritis demonstrated bene- nance”) [52]. This RCT recruited 65 patients between 1981
fit in treating patients with high doses of corticosteroids. and 1991. Continuing cyclophosphamide as maintenance
For example, the report by Pollack with the first histologi- resulted in fewer exacerbations after the first 6 months, and
cal classification reported better renal outcomes in active fewer patients experienced a sustained doubling of serum
lupus glomerulonephritis with high-dose versus low-dose creatinine in this group than the short course (6 months) of
corticosteroid [25]. Fourteen of 31 patients receiving cyclophosphamide group and the steroid alone group (15%
412 Lupus Nephritis
vs. 35% vs. 48%, respectively). A subsequent study recruited Azathioprine was used for induction of remission in
82 patients between 1986 and 1990 and modified the steroid- lupus nephritis less frequently than cyclophosphamide.
alone regimen, continuing monthly doses of methylpredni- However, azathioprine was commonly used to maintain
solone for at least 12 months (maximum 36 doses) [53]. Two remission after induction.
intravenous cyclophosphamide arms were similar to the The Cochrane Kidney and Transplant group’s first sys-
previous study except that one was combined with monthly tematic review of lupus nephritis included most of the
methylprednisolone. This latter combination therapy group above studies and demonstrated that cyclophosphamide-
had more renal remission (proteinuria <1 g/day) than the based regimens reduced the risk of doubling of serum cre-
cyclophosphamide alone and methylprednisolone alone atinine (relative risk [RR] 0.59, 95% confidence interval
groups: 17 of 28 vs. 13 of 27 vs. 7 of 27, respectively. After a [CI] 0.40–0.88) compared to steroid alone, but did not
median of 11 years of follow-up, the probability of no treat- reduce mortality or ESKD [57]. However, the risk of ovar-
ment failure (first occurrence of death, doubling of serum ian toxicity was increased with cyclophosphamide, an
creatinine, or need for off-protocol immunosuppression) important toxicity in a patient group that consists predomi-
was ~70% in the combination group, ~55% in the cyclophos- nantly of women of childbearing age. Major infections and
phamide alone group, and ~40% in the methylprednisolone herpes zoster infections did not occur more frequently with
alone group [54]. During follow-up, 17 of 27 patients rand- cytotoxic therapy than with steroids alone.
omized to methylprednisolone alone received cyclophos-
phamide and the benefits of these regimens came at a cost: Mycophenolate Mofetil plus Corticosteroids
36% of the combination therapy patients had avascular In 2000, Chan et al. published a RCT of 42 patients from
necrosis, 60% developed amenorrhoea at a median age of Hong Kong with WHO Class IV lupus nephritis [58]. They
33 years, and 26% experienced herpes zoster infection. compared mycophenolate mofetil 1 g twice daily and pred-
Given the concerns of cumulative cyclophosphamide nisolone 0.8 mg/kg/day for 12 months to oral cyclophos-
toxicity, the use of lower dose cyclophosphamide (six phamide 2.5 mg/kg/day and prednisolone 0.8 mg/kg/day
fortnightly intravenous cyclophosphamide pulses at a for 6 months then azathioprine 1.5 mg/kg/day and predni-
fixed dose of 500 mg, followed by azathioprine as mainte- solone for 6 months. Complete remission by 12 months was
nance therapy) was compared to the NIH cyclophospha- seen in 17 out of 21 and 16 of 21 patients, respectively.
mide regimen described above in the Euro-Lupus Mycophenolate mofetil thus had similar efficacy, as well as
Nephritis Trial (ELNT) [55]. This trial showed low-dose similar infectious complications but no alopecia or amen-
cyclophosphamide was comparable to high-dose cyclo- orrhoea. Given that nephrologists had developed signifi-
phosphamide in inducing disease remission but no sta- cant experience with mycophenolate in kidney transplant
tistically different reduction in adverse effects of therapy by this stage, it became a viable alternative to cyclophos-
could be demonstrated. phamide in lupus nephritis.
Azathioprine as induction therapy was tested in the Chan’s findings were confirmed in a 24-week non-inferi-
very early trials and although it was compared directly to ority RCT comparing mycophenolate mofetil to the NIH
intravenous cyclophosphamide in Austin’s study [51], cyclophosphamide regimen in 140 patients from centers in
there was insufficient power because this trial included the United States [59]. Just over half of these participants
five treatment groups. A study addressing this issue was were African American. Complete and partial remissions
performed in The Netherlands and randomized 87 people occurred in 16 and 21 of 71 patients randomized to
with lupus nephritis to either intravenous cyclophospha- mycophenolate mofetil, respectively, and 4 and 17 of 69
mide as per the NIH regimen for 2 years (but at a fixed randomized to cyclophosphamide, respectively. The differ-
dose of 750 mg/m2) or azathioprine 2 mg/kg/day plus ence in complete remissions was significant for “superior-
three daily pulses of methylprednisolone 1 g at baseline ity,” even though the trial was designed as a noninferiority
then 2 and 6 weeks later [56]. Patients randomized to trial. There were fewer severe infections with mycopheno-
cyclophosphamide received oral prednisolone 1 mg/kg/ late mofetil but more patients experienced diarrhea, a
day initially and patients randomized to azathioprine known side effect. Subsequently, an international collabo-
received oral prednisolone 20 mg/day initially. After a ration called the ALMS Study Group (including many
median follow-up of 5.7 years, relapses were significantly United States authors from the aforementioned study)
more frequent in the azathioprine group, and doubling of demonstrated in the largest RCT to that time (n = 370) that
serum creatinine was also more frequent but did not mycophenolate mofetil exhibited similar efficacy in induc-
reach statistical significance. There was more herpes zos- ing disease remission to intravenous cyclophosphamide
ter infection with azathioprine, but other parameters (56.2% vs. 53.0%, respectively) [31]. Overall rates of adverse
such as kidney function and proteinuria were similar. events were similar, although alopecia, nausea, and vomiting
Treatmen 413
occurred more often with cyclophosphamide and diarrhea The updated Cochrane review did identify a pilot study
with mycophenolate mofetil. Because it was an interna- (n = 40) [63] and larger RCT (n = 362) [64] from the same
tional study, there were fewer African American patients group in China that compared a regimen combining the
but subgroup and post hoc analyses demonstrated that calcineurin inhibitor tacrolimus 2 mg twice daily with
both African American (plus “other”) and Hispanic mycophenolate mofetil 500 mg twice daily to intravenous
patients had better response rates with mycophenolate cyclophosphamide 0.75 g/m2 every month for six doses.
mofetil than cyclophosphamide (~60% vs. ~40%). The pre- Both groups initially received methylprednisolone 500 mg
specified interaction between treatment and race was sig- daily for 3 days then prednisolone 0.6 mg/kg. This large
nificant (P = 0.047). RCT demonstrated more complete remission at 24 weeks in
Importantly, the ALMS Group then randomized the 227 the mycophenolate mofetil plus tacrolimus group (45.9%)
patients who met response criteria at 24 weeks to mycophe- than the intravenous cyclophosphamide group (25.6%).
nolate mofetil 2 g/day or azathioprine 2 mg/kg/day for This promising result has not been replicated and must be
36 months [32]. Treatment failure was defined as death, interpreted in light of the limited generalizability, being
ESKD, doubling of serum creatinine, renal flare or rescue from a homogenous Chinese population, and the influence
therapy, and occurred in 16.4% of patients receiving of the non-immune proteinuria-lowering effect of tacroli-
mycophenolate mofetil and 32.4% receiving azathioprine. mus where a proteinuria-based definition of complete
Overall adverse events were common and similar between remission was used.
groups. Prior to this, a RCT comparing quarterly intrave-
nous cyclophosphamide, oral azathioprine, and oral myco- The Current Era: Biologic Therapies
henolate mofetil as maintenance therapy after induction Conventional immunosuppressive therapy with corticos-
with up to seven monthly doses of intravenous cyclophos- teroids plus either mycophenolate or intravenous cyclo-
phamide demonstrated that the oral azathioprine and phosphamide has considerably improved the outlook for
mycophenolate mofetil groups had superior event-free sur- patients with Classes III, IV, or V lupus nephritis, with
vival (events being death or doubling of serum creatinine) responses at 24 weeks in one large study of more than 50%
to the quarterly intravenous cyclophosphamide [60]. A with either therapy [31]. However, there remains consider-
European study, the MAINTAIN Nephritis Trial, rand- able treatment-related toxicity and room to improve effi-
omized 105 patients to either mycophenolate mofetil or cacy. As our understanding of the pathophysiology of
azathioprine at week 12 after all patients received induc- lupus nephritis evolves, so-called biologic therapies tar-
tion with six 500 mg doses of intravenous cyclophospha- geted at specific immunological processes (Figure 25.2)
mide each a fortnight apart [61]. Renal flares occurred in have been developed [65]. Results with these agents to date
13 (25%) patients receiving azathioprine and 10 (19%) have been suggestive but inconclusive.
patients receiving mycophenolate mofetil but there were Rituximab is a monoclonal antibody that binds specifi-
no significant differences between the groups in any major cally to CD20 and depletes autoreactive B cells (but not
parameters measured. plasma cells) and reduces autoantibody production [65].
The Cochrane Kidney and Transplant Group updated The Lupus Nephritis Assessment with Rituximab
their systematic review in 2012 and this update included (LUNAR) study recruited 144 participants and compared
the RCTs examining the role of mycophenolate discussed rituximab to placebo when added to standard therapy of
in this section [62]. Mycophenolate mofetil was as effective mycophenolate mofetil and corticosteroids and demon-
as intravenous cyclophosphamide for complete remission, strated no difference in clinical outcomes between groups
stable kidney function, major infection, and mortality, with after 12 months [66]. In a post hoc analysis of patients
less ovarian failure (RR 0.15, 0.03–0.80). Used as mainte- who received rituximab, achieving complete B cell deple-
nance therapy, the risk of renal relapse was increased tion (zero B cells per μl), shorter time to complete B cell
2-fold with azathioprine compared to mycophenolate depletion, and longer duration of complete B cell deple-
mofetil (RR 1.83, 95% CI 1.24–2.71). This systematic review tion were all associated with increased odds of complete
was updated again in 2018 with a total of nine studies (868 response [67]. A rituximab-based regimen has been used
participants) comparing mycophenolate mofetil to intrave- at Imperial College London and non-randomized results
nous cyclophosphamide for induction [17]. The two treat- reported [68]. The purpose of this regimen is to reduce
ments were equivalent for most outcomes, and findings oral corticosteroid use and it consists of rituximab 1 g plus
from this review are used for the Summary of Findings methylprednisolone 500 mg given at days 1 and 15, with
table (Table 25.4). Interestingly, the expected association of mycophenolate mofetil commencing at 500 mg twice
more ovarian failure with cyclophosphamide was weak- daily and titrated up to 1500 mg twice daily (or the maxi-
ened in this most recent systematic review. mum tolerated dose). In this series of 50 consecutive
414 Lupus Nephritis
Table 25.4 Summary of findings: treatment of lupus nephritis.
No. of Absolute number of

participants patients affected per Certainty of
(no. of 1000 patients treated Relative risk the evidence
Outcomes studies) for 1 year (95% CI) (95% CI) (GRADE) Conclusion
Intravenous cyclophosphamide versus corticosteroids alone

All-cause 226 (5) 3 fewer 0.98 Low Intravenous cyclophosphamide compared to
mortality (80 fewer to (0.53–1.82) ●●○○ corticosteroids alone may make little or no
139 more) difference to mortality
End-stage 452 (4) 90 fewer 0.63 Moderate Intravenous cyclophosphamide compared to
kidney disease (148 fewer–7 more) (0.39, 1.03) ●●○○ corticosteroids alone probably decreases
end-stage kidney disease
Complete 13 (1) 0 2.63 Very low It is uncertain whether mycophenolate
remission in (no events in control (0.13–54.64) ●○○○ mofetil plus tacrolimus compared to
proteinuria group) intravenous cyclophosphamide makes any
difference to complete remission in
proteinuria
Doubling of 228 (4) 162 fewer 0.59 Moderate Intravenous cyclophosphamide compared to
serum (237–47 fewer) (0.40–0.88) ●●○○ corticosteroids alone probably decreases
creatinine doubling serum creatinine
Renal relapse 84 (2) 337 fewer 0.23 Moderate Intravenous cyclophosphamide compared to
(403–166 fewer) (0.08–0.62) ●●○○ corticosteroids alone probably decreases
renal relapse
Stable kidney 278 (5) 118 more 1.20 Moderate Intravenous cyclophosphamide compared to
function (0–265 fewer) (1.00–1.45) ●●○○ corticosteroids alone probably increases
stable kidney function
Ovarian failure 147 (3) 222 more 2.18 Low Intravenous cyclophosphamide compared to
(19 fewer–628 more) (1.10–4.34) ●●○○ corticosteroids alone may increase ovarian
failure
Major infection 291 (6) 19 fewer 0.87 Moderate Intravenous cyclophosphamide compared to
(75 fewer–76 more) (0.50–1.51) ●●○○ corticosteroids alone probably decreases
major infection
Mycophenolate mofetil versus intravenous cyclophosphamide for initial treatment
All-cause 826 (8) 6 more 1.12 Very low It is uncertain whether mycophenolate
mortality (19 fewer–51 more) (0.61–2.06) ●○○○ mofetil compared to intravenous
cyclophosphamide makes any difference to
all-cause mortality
End-stage 231 (3) 25 fewer 0.71 Very low It is uncertain whether mycophenolate
kidney disease (62 fewer–71 more) (0.27–1.84) ●○○○ mofetil compared to intravenous
cyclophosphamide makes any difference to
end-stage kidney disease
Complete 868 (8) 38 more 1.17 Low Mycophenolate mofetil compared to
remission (7 fewer–93 more) (0.97–1.42) ●●○○ intravenous cyclophosphamide may have
little or no difference on complete remission
Partial 868 (8) 9 more 1.02 Low Mycophenolate mofetil compared to
remission (47 fewer–77 more) (0.89–1.18) ●●○○ intravenous cyclophosphamide may have
little or no difference on partial remission
Ovarian failure 539 (3) 26 more 0.36 Very low It is uncertain whether mycophenolate
(39 fewer–48 more) (0.02–2.18) ●○○○ mofetil compared to intravenous
cyclophosphamide makes any difference on
ovarian failure
Major infection 699 (5) 2 more 1.02 Low Mycophenolate mofetil compared to
(38 fewer–62 more) (0.67–1.54) ●●○○ intravenous cyclophosphamide may have
little or no difference on major infection
Treatmen 415

Leucopenia 753 (6) 81 fewer 0.59 Moderate Mycophenolate mofetil compared to

(133 fewer–16 more) (0.33–1.08) ●●○○ intravenous cyclophosphamide probably
decreases leucopenia
Alopecia 622 (3) 170 fewer 0.29 Moderate Mycophenolate mofetil compared to
(194–129 fewer) (0.19–0.46) ●●○○ intravenous cyclophosphamide probably
decreases alopecia
Diarrhea 609 (4) 142 more 2.42 Moderate Mycophenolate mofetil compared to
(64–258 more) (1.64–3.58) ●●○○ intravenous cyclophosphamide probably
increases diarrhea
Mycophenolate mofetil plus tacrolimus versus intravenous cyclophosphamide for initial treatment
All-cause 402 (2) Not estimable No events Absent No events of all-cause mortality occurred in
mortality the studies that evaluated mycophenolate
mofetil plus tacrolimus compared to
intravenous cyclophosphamide
End-stage Not estimable Absent No studies comparing low-dose
kidney disease mycophenolate mofetile plus tacrolimus
compared to intravenous cyclophosphamide
examined end-stage kidney disease
Complete 402 (2) 337 more 2.38 Low Mycophenolate mofetil plus tacrolimus
remission (17–1049 more) (1.07–5.3) ●●○○ compared to intravenous cyclophosphamide
may increase complete remission
Partial 402 (2) 0 1.00 Low Mycophenolate mofetil plus tacrolimus
remission (83 fewer–106 more) (0.78–1.28) ●●○○ compared to intravenous cyclophosphamide
may make little to no difference to partial
remission
Stable kidney 402 (2) 222 more 1.78 Low Mycophenolate mofetil plus tacrolimus
function (114–358 more) (1.40–2.26) ●●○○ compared to intravenous cyclophosphamide
may increase achievement of stable kidney
function
Ovarian failure 34 (1) Not estimable No events- Absent No events of ovarian failure occurred in the
studies that evaluated mycophenolate
mofetil plus tacrolimus compared to
intravenous cyclophosphamide
Major infection 402 (2) 23 more 1.65 Very low It is uncertain whether mycophenolate
(31 fewer–820 more) (0.11–24.44) ●○○○ mofetil plus tacrolimus compared to
intravenous cyclophosphamide makes any
difference to major infection
Leucopenia 402 (2) 62 fewer 0.23 Very low It is uncertain whether mycophenolate
difference to leucopenia
Alopecia 402 (2) 14 fewer 0.78 Very low It is uncertain whether mycophenolate
difference to alopecia
Diarrhea 362 (1) 44 more 2.33 Very low It is uncertain whether mycophenolate
difference to diarrhea
(Continued)
416 Lupus Nephritis

Mycophenolate versus azathioprine to maintain remission after initial treatment
All-cause 451 (4) 3 fewer 0.87 Very low It is uncertain whether mycophenolate
mortality (17 fewer–44 more) (0.26–2.92) ●○○○ mofetil compared to azathioprine makes any
difference to all-cause mortality
End-stage 452 (4) 15 fewer 0.59 Very low It is uncertain if mycophenolate mofetil
kidney disease (30 fewer–33 more) (0.18, 1.92) ●○○○ compared to azathioprine makes any
difference to end-stage kidney disease
Renal relapse 452 (4) 114 fewer 0.57 Moderate Mycophenolate mofetil compared to
(162–45 fewer) (0.39–0.83) ●●●○ azathioprine probably decreases renal relapse
Doubling of 452 (4) 46 fewer 0.46 Moderate Mycophenolate mofetil compared to
serum (68–3 fewer) (0.21–0.97) ●●●○ azathioprine compared to mycophenolate
creatinine mofetil probably decreases doubling of
serum creatinine
Ovarian failure 177 (2) 10 more 1.30 Very low It is uncertain if mycophenolate mofetil
(24 fewer–161 more) (0.29–5.73) ●○○○ compared to azathioprine makes any
difference to ovarian failure
Major infection 412 (3) 8 fewer 0.92 Low Mycophenolate mofetil compared to
(49 fewer–66 more) (0.51–1.67) ●●○○ azathioprine may make little or no difference
to major infection
Leucopenia 412 (3) 65 fewer 0.18 Low Mycophenolate mofetil compared to
(75–32 fewer) (0.05–0.6) ●●○○ azathioprine may decrease leucopenia
Alopecia 412 (3) 3 more 1.05 Low Mycophenolate mofetil compared to
(31 fewer–74 more) (0.51–2.15) ●●○○ azathioprine may make little to no difference
to alopecia
Diarrhea 307 (2) 53 more 1.36 Low Mycophenolate mofetil compared to
(61 fewer–318 more) (0.58–3.18) ●●○○ azathioprine may make little or no difference
to diarrhea
CI, confidence interval. Treatment estimates are drawn from a systematic review with network meta-analysis of randomized controlled
trials [17, 18]. GRADE assessment of the certainty of the evidence [20]: High: This research provides a very good indication of the likely effect.
The likelihood that the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect.
The likelihood that the effect will be substantially different* is moderate. Low: This research provides some indication of the likely effect.
However, the likelihood that it will be substantially different* is high. Very low: This research does not provide a reliable indication of the likely
effect. The likelihood that the effect will be substantially different* is very high. *Substantially different = a large enough difference that it might
affect decision-making.
Sources: Tunnicliffe et al. [17, 18] and Hultcrantz et al. [20].
patients, 72% achieved a complete remission at a median Ocrelizumab also selectively depletes CD20+ B cells. In
of 36 weeks and 90% achieved either complete or partial the largest RCT studying lupus nephritis to date, 378 par-
remission at a median of 37 weeks. Eleven patients ticipants were recruited from multiple countries and rand-
relapsed after achieving remission. Oral corticosteroids omized to placebo, ocrelizumab 400 mg, or ocrelizumab
for more than 2 weeks were only required in two patients 1000 mg on top of standard care for induction or remis-
who responded and two who did not. Interestingly, 22 of sion [70]. Standard care, at the discretion of the treating
the 50 patients in this report had Class V lupus nephritis. physician, was either mycophenolate mofetil or the ELNT
This is relevant because rituximab is increasingly being regimen of six fortnightly doses of intravenous cyclophos-
used for primary membranous glomerulonephritis [69]. phamide 500 mg followed by azathioprine 2 mg/kg/day.
This RITUXILUP regimen is the subject of a current RCT The primary outcome was response at 48 weeks assessed
(NCT01773616). as either complete, partial, or no response. The trial was
Treatmen 417
terminated early because of concerns about an increased patients with renal involvement assessed by the Safety of
frequency of serious opportunistic infections, coupled with Estrogens in Lupus Erythematosus National Assessment–
the lack of benefit demonstrated in the LUNAR study SLE Disease Activity Index (SELENA-SLEDAI).
reported at that time. Opportunistic infections occurred in Improvement in the renal domain of the SELENA-SLEDAI
0.8% receiving placebo, 3.2% receiving ocrelizumab 400 mg, occurred in 44.6%, 51.1%, and 54.1% of patients receiving
and 0.8% receiving ocrelizumab 1000 mg. Serious infec- placebo, belimumab 1 mg/kg, and belimumab 10 mg/kg,
tions were also more frequent in the ocrelizumab 400 mg respectively. In the whole cohort, renal flares occurred in
group and, when stratified by background therapy, more 3.0%, 2.5%, and 1.4%, respectively, and the median percent-
frequent in the patients treated with mycophenolate. This age reduction in proteinuria over 52 weeks was greater in
latter subgroup also received more corticosteroid. Because the belimumab groups. However, this agent was not used
of the early termination, the study outcomes are reported to treat patients presenting with lupus nephritis and thus
for a modified intention-to-treat group of 223 patients and should not be extrapolated as a treatment option for such.
thus had reduced power. The overall (complete plus par-
tial) renal response rate at 48 weeks was above 50% in all
Pure Membranous (Class V) Lupus Nephritis
three groups, and the difference in response rate between
placebo and the two ocrelizumab groups combined almost Pure membranous (Class V) lupus nephritis differs from pri-
reached statistical significance (12.7%, 95% CI −0.8–26.1). mary membranous glomerulonephritis in that it is associ-
The effect appeared greater in the subgroup with ocreli- ated with an inflammatory disease that relapses and remits,
zumab added to the ELNT regimen than when added to and is less likely than primary membranous glomerulone-
mycophenolate mofetil. phritis to remit spontaneously [39]. On the other hand,
Abatacept modulates the CD28-CD80/86 signaling path- guidelines suggest similar strategies for treatment [40] such
way that leads to co-stimulation and T cell activation, and as controlling proteinuria with renin-angiotensin system
has also been the subject of a large RCT in lupus nephri- inhibition if non-nephrotic and reserving immune suppres-
tis [71]. This study compared placebo to two different dos- sion for more severe (nephrotic) disease. Most of the trials of
ing regimens of abatacept with all patients receiving therapies in lupus nephritis enrolled some patients with
mycophenolate mofetil and corticosteroids. This trial dem- Class III, IV, and V disease. In Austin’s early trial of 107
onstrated no safety concerns but did not demonstrate a dif- patients [51], 16 had pure membranous lupus nephritis and
ference in the primary outcome of time to complete in the more recent ALMS Maintenance Study [32], 35 out of
response, defined as stable eGFR compared to baseline 227 had pure membranous lupus nephritis (more had Class
(within 10%), urine protein:creatinine ratio <30 mg/mmol, V combined with either Class III or IV). To some degree, it is
and normal urine microscopy on two successive visits (the reasonable to apply the results of these trials to Class V lupus
most stringent definition use in RCTs to date). Using the nephritis. There has been little evidence exclusively examin-
dataset from this trial, different definitions of complete ing the treatment of class V lupus nephritis. Twenty-three
response from a number of trials as well as that recom- years after his first lupus nephritis RCT, Austin et al. com-
mended by the ACR [72] were applied [73]. With alterna- pared the addition of the NIH cyclophosphamide regimen
tive complete response definitions, the abatacept groups or cyclosporine to prednisone alone in 42 patients with Class
had an estimated difference in complete response of about V lupus nephritis and nephrotic range proteinuria [19].
15% that was statistically significant. For example, using Complete remission of proteinuria (to <0.3 g/day) at 1 year
the ALMS criteria the complete response rates in the pla- occurred in 27% of patients receiving prednisone alone, 60%
cebo, abatacept 10/10 regimen, and abatacept 30/10 regi- receiving prednisone plus the NIH cyclophosphamide regi-
men were 14, 25, and 30%, respectively, compared to 3, 3, men, and 83% receiving prednisone plus cyclosporine. This
and 5% using the original trial criteria. supports the use of immunosuppressive agents in patients
Another biologic agent with appealing but not conclu- with nephrotic proteinuria. Treatment for Class V (membra-
sive evidence for benefit in lupus nephritis is belimumab, a nous) lupus nephritis therefore depends on the degree of
monoclonal antibody that inhibits soluble B-lymphocyte proteinuria. Patients with proteinuria less than 3.0–3.5 g/day
stimulator. This agent demonstrated a good response in can be treated with renin-angiotensin system blockade to try
patients with SLE and thus a post hoc pooled analysis of to reduce proteinuria, just as in primary membranous glo-
two trials was performed to assess the renal outcomes [74]. merulonephritis. Other supportive care such as anticoagula-
Participants were randomized to two doses of belimumab tion for patients considered at high risk of venous
or placebo on top of standard therapy for active SLE (corti- thromboembolism is also important. For patients with
costeroids, hydroxychloroquine). Mycophenolate was used “nephrotic” proteinuria above this level, immunosuppres-
in 10% of the whole cohort (n = 1684) and 20% of the 227 sion is recommended [39] and this could be any of the
418 Lupus Nephritis
regimens used for Class III and IV lupus nephritis. B-cell failure to respond and over what time frame. Both the
targeted therapies such as rituximab are gaining favor in idi- KDIGO [39] and EULAR/ERA-EDTA [40] Guidelines
opathic membranous glomerulonephritis [69] but what this recommend switching from one regimen to the other (i.e.
means for pure membranous (Class V) lupus nephritis to mycophenolate if cyclophosphamide used first or to
remains to be seen. cyclophosphamide if mycophenolate used first). One
must also be confident that the optimal dose has been
given. These guidelines don’t prefer one regimen over the
Summary of Therapies for Class III, IV, and V
other. A systematic review specifically addressing severe
Disease
lupus nephritis reported that mycophenolate and cyclo-
The Summary of Findings table (Table 25.4) presents vari- phosphamide were equivalent for short-term outcomes
ous comparisons of treatment and the treatment effects but that there was a suggestion from the data that cyclo-
from the most recent Cochrane Kidney and Transplant phosphamide had better long-term outcomes such as
meta-analysis [17]. A problem with the lupus nephritis preservation of kidney function [75]. However, this was
RCT literature is that a wide range of treatments has been not a pooled analysis of individual patient-level data to
studied in RCTs that have mostly been too small to assess enable stratification based on kidney function. The
important clinical outcomes. Most of the trials cited in this ALMS trial reported outcomes in 32 (out of 370) partici-
chapter are the largest and least susceptible to bias. To pants who had an eGFR less than 30 ml/min/1.73 m2 at
explore comparisons and rank therapies that have not been baseline [76]. Four of 20 patients randomized to
made directly in RCTs, a network meta-analysis was per- mycophenolate mofetil responded, compared to two of
formed [18]. This allowed both direct and indirect compar- 12 patients randomized to cyclophosphamide (P = 0.9)
isons of treatments. Using intravenous cyclophosphamide but ultimately this study was underpowered to reach any
as the reference treatment, the odds ratio for complete conclusion regarding patients with eGFR less than 30 ml/
remission was 1.44 (95% CI 1.00–2.06) with mycophenolate min/1.73 m2.
mofetil, and 2.69 (95% CI 1.74–4.16) for mycophenolate Other therapies with lower levels of evidence that
mofetil combined with tacrolimus. The limitations of this might be “added on” in very unwell patients not respond-
finding have been discussed previously. These treatments ing to mycophenolate or cyclophosphamide include
were no different with respect to all-cause mortality or rituximab (discussed above), intravenous immunoglobu-
ESKD. Overall, this network meta-analysis confirmed that lin, or a calcineurin inhibitor. Intravenous immunoglob-
mycophenolate mofetil is similar to intravenous cyclophos- ulin has been reported to be beneficial in case reports [77]
phamide, with a better adverse effect profile for both induc- but the only RCT in lupus nephritis was small (n = 14)
tion and maintenance therapy. and compared intravenous immunoglobulin (IVIg) to
cyclophosphamide after 6 months of the NIH regimen of
intravenous cyclophosphamide [78]. This was really a
Special Situations
trial of maintenance therapy and so is difficult to extrap-
Very Severe or Nonresponsive Lupus Nephritis olate as “add-on” therapy in patients with severe disease.
Patients with very severe lupus nephritis are often excluded Addition of a calcineurin inhibitor such as tacrolimus as
from RCTs or are a subset of participants. There is no uni- in the RCT of “multitarget” therapy with prednisolone,
versally accepted definition of “severe” and many people tacrolimus, and mycophenolate mofetil [64] might be
would consider any Class III or IV lupus nephritis to be a considered. However, in addition to the limitations of
“severe” kidney-threatening manifestation of SLE. One this trial that have been discussed, kidney function at
arbitrary classification of severe that has been advanced baseline was relatively well preserved in this trial (80%
defines “severe” as (i) Class IV lupus nephritis with more had eGFR >60 ml/min). Plasma exchange is also sug-
than 15% crescents and/or glomerular capillary necrosis, gested as an option in the EULAR ERA-EDTA guide-
(ii) persistence or relapse despite cyclophosphamide, or line [40], although an RCT carried out between 1981 and
(iii) Class III or IV lupus nephritis with elevated serum cre- 1986 was stopped early because there was no benefit of
atinine [75]. This definition has not been used in RCTs but plasma exchange added to prednisolone and oral cyclo-
the common theme is that patients with these features usu- phosphamide [79]. Patients had quite “severe” disease
ally have elevated serum creatinine. with segmental and proliferative lesions being required
Patients with very severe lupus nephritis can be treated for inclusion and a mean serum creatinine at baseline of
with the same regimens described above for Class III or ~180 umol/l. It remains to be seen whether rituximab or
IV disease. There is again no accepted definition of other biologicals will find their place in very severe lupus
“refractory” or “resistant” disease or what constitutes a nephritis.
Treatmen 419
Pregnancy and Lupus Nephritis The ALMS maintenance trial lasted for 36 months and
Most patients with lupus nephritis are women of child- over this time there were fewer treatment failures or flares
bearing age and pregnancy should therefore be discussed with mycophenolate mofetil [32]. Thus, there is evidence
early in their course to determine the management to be to continue mycophenolate as maintenance therapy for
implemented before conception. Management of pregnan- 3 years. Beyond this time, evidence for what to do is lack-
cies in women with lupus nephritis requires a multidisci- ing. In an Italian cohort (n = 161) in whom patients in
plinary approach and has been recently reviewed [80, 81]. complete remission (proteinuria <0.5 g/day) for 12 months
Most women with a history of lupus nephritis have good underwent withdrawal of immunosuppression (n = 73), 21
pregnancy outcomes provided the disease is in remission, (29%) patients had flares during reduction in therapy and
and they are lupus anticoagulant negative, not hyperten- 20 of the remaining 52 patients (38%) experienced a flare
sive, and not non-Caucasian or Hispanic. after withdrawal of maintenance therapy [83]. Overall,
Before pregnancy, women with lupus nephritis should be these patients received no therapy for 41% of their total
in remission for at least 6 months. Ideally, women will have follow-up time. These authors recommended that with-
ceased potentially teratogenic medications such as renin drawal of therapy was possible but only after 5 years of
angiotensin system inhibitors, cyclophosphamide, and maintenance therapy with 3 years in complete remission,
mycophenolate, and switched to appropriate alternatives. and only if done very slowly.
Discontinuation or switching of immunosuppressive medi- In a group of Argentinian lupus nephritis patients who
cations runs the risk of a flare so should be done before con- received maintenance mycophenolate mofetil for 3 years
ception and with close monitoring. Hydroxychloroquine is after induction with an NIH cyclophosphamide regimen
safe in pregnancy and should be continued before, during, for 6 months, and who were in complete remission for
and after pregnancy. 12 months (proteinuria<0.5 g/day), a kidney biopsy was
Once pregnant, aspirin should be started before 12 weeks performed before mycophenolate mofetil was weaned and
gestation and consideration given to anticoagulation in ceased over a 6-month period regardless of the kidney
women at risk, such as those with positive lupus anticoag- biopsy findings [84]. Of the 36 patients who completed the
ulant or a history of venous thromboembolism. study, 20 were in histological remission at the time of this
Close monitoring during pregnancy to detect flares of biopsy (NIH activity index = 0), nine had an activity index
lupus nephritis or later in pregnancy to detect preeclamp- 1–2, and seven had an activity index of 3–5. Over the next
sia (and indeed to differentiate one from the other) is 24 months 11 patients had a flare of lupus nephritis after
essential on top of monitoring for obstetric complications. ceasing maintenance mycophenolate and 10 of these had
Management of flares is complicated and depends upon an activity index above zero (mainly due to finding endo-
the severity and gestation. Corticosteroids, azathioprine, capillary proliferation and subendothelial deposits). Six of
and calcineurin inhibitors can be used safely at any gesta- 25 patients who did not flare over 24 months had an activ-
tion. Severe or resistant flares bring about very difficult ity index of 1 or 2. Although kidney biopsy appears helpful,
decisions: one must weigh up the risks to the mother and it must be noted that this study did not demonstrate
fetus of continuing the pregnancy and using therapies that whether continuation of maintenance immunosuppres-
are more efficacious for the lupus nephritis but less safe for sion if the kidney biopsy was unfavorable would prevent
the fetus, against the risks to mother and fetus of delivery renal flares.
of the immediate delivery of the fetus to enable effective Thus, maintenance therapy should be prolonged and
therapies. withdrawal of immunosuppression is possible but the ben-
efits (less immunosuppression) and risks (30–40% risk of
Withdrawal of Maintenance Therapy renal flare) must be carefully weighed up. A kidney biopsy
Maintenance therapy can be challenging in lupus nephri- may help to guide the decision.
tis. Once patients are feeling better following induction of
remission, it can be hard to convince them that they need Renal Replacement Therapy
ongoing therapy to prevent relapse, particularly when they A major goal of therapy in lupus nephritis is avoidance of
are experiencing cosmetic and other side effects of corti- ESKD. In the earliest days this meant death for most
costeroids. Duration of and withdrawal of maintenance patients, but now renal replacement therapy is widely
therapy are the subject of debate [82], with the arguments available in most, but not all, countries [85]. Although
based on whether it is better to minimize treatment toxicity ESKD is an important outcome in the management of
by giving patients time off immunosuppressive therapy or lupus nephritis, lupus nephritis is the cause of ESKD in
to prevent flares of lupus nephritis, which add to kidney less than 2% of prevalent patients with ESKD [86, 87]. The
function loss. incidence rate of ESKD is three to four times higher in
420 Lupus Nephritis
African American patients, who also have a higher mortal- receiving dialysis (P < 0.01) and the distribution of the 55
ity [88]. Compared to other patients with ESKD, the patients with no (SLEDAI = 0), low (SLEDAI = 1–10), or
patients with lupus nephritis are much younger, more high (SLEDAI > 10) non-renal disease activity was 2, 36, and
likely to be female, and have less comorbidity from diabe- 17 before dialysis and 19, 28, and 8, respectively, after dialy-
tes and cardiovascular disease [87]. Although in the United sis [90]. The same authors reported that in 28 patients who
States mortality rates of patients with ESKD with SLE were received a kidney transplant for lupus nephritis, the number
similar to people with ESKD who did not have SLE [88], in with a non-renal SLEDAI > 10 was 17 before dialysis, three
Australia and New Zealand this was not the case [87]. The during dialysis and none after transplant [91]. In contrast, a
adjusted hazard ratio for patient death was 1.33 (1.12–1.58) Brazilian study compared 57 patients on dialysis with SLE to
for patients on dialysis due to lupus nephritis, and 1.87 57 controls and defined lupus activity by the SLEDAI score
(1.18–2.98) for kidney transplant recipients, compared to or if patients required prednisolone 20 mg daily or
ESKD controls who did not have lupus nephritis. greater [92]. The proportion with SLEDAI scores of 0, 1–3,
Despite the possibility of higher mortality, kidney trans- 4–7, and 8 were 19%, 44%, 19%, and 18% for the SLE patients
plant outcomes are still favorable in patients who develop and 63%, 33%, 4%, and 0% for the non-SLE patients, respec-
ESKD from lupus nephritis. Five-year renal allograft and tively. Thus, SLE disease activity should be actively consid-
death-censored renal allograft survival were 88 and 93%, ered even after renal replacement therapy becomes the
respectively [87]. Lupus nephritis recurred in 2.3% of kid- primary focus of management.
ney transplant in patients in Australia and New
Zealand [87], and in 2.44% of kidney transplant recipients
in the United States [89]. In the US patients, African C
onclusion
American, female and younger patients were more likely to
experience recurrence of lupus nephritis, and allograft Lupus nephritis is a kidney-threatening manifestation of
median survival after recurrence was between 4 and SLE for which a number of treatments have been studied
5 years [89]. The survival of kidneys with recurrent lupus in RCTs over many years. However, there is still a long way
nephritis was similar to those that had rejection and the to go to increase the complete remission rates with these
latter was a far more important cause of graft loss in this regimens or newer therapies while at the same time reduc-
study (156 patients lost their graft from recurrent lupus ing the toxicity of treatment. Acquiring good evidence
nephritis compared to 1517 from rejection). remains a challenge due to the heterogeneity of the disease
Another issue in patients with SLE who develop ESKD itself and difficulties with recruiting large numbers of
from lupus nephritis is whether the SLE becomes “quies- patients into treatment trials, the need to consider both
cent” or not. Early studies indicate that this is true. In a induction and maintenance therapy, and the long follow-
Dutch population, the mean non-renal SLEDAI was up required to measure outcomes such as renal relapse
8.22 ± 6.3 before starting dialysis and 4.41 ± 5.94 when after remission, loss of kidney function, and ESKD.
R
eferences
Hochberg, M.C. (1997). Updating the American College of

1 systemic lupus erythematosus revisited. Kidney Int. 65
Rheumatology revised criteria for the classification of (2): 521–530.
systemic lupus erythematosus. Arthritis Rheum. 40 (9): 5 Bajema, I.M., Wilhelmus, S., Alpers, C.E. et al. (2018).
1725. Revision of the International Society of Nephrology/Renal
2 Tan, E.M., Cohen, A.S., Fries, J.F. et al. (1982). The Pathology Society classification for lupus nephritis:
1982 revised criteria for the classification of systemic clarification of definitions, and modified National
lupus erythematosus. Arthritis Rheum. 25 (11): Institutes of Health activity and chronicity indices. Kidney
1271–1277. Int. 93 (4): 789–796.
3 Petri, M., Orbai, A.M., Alarcon, G.S. et al. (2012). Boodhoo, K.D., Liu, S., and Zuo, X. (2016). Impact of sex
6
Derivation and validation of the systemic lupus disparities on the clinical manifestations in patients with
international collaborating clinics classification criteria for systemic lupus erythematosus: a systematic review and
systemic lupus erythematosus. Arthritis Rheum. 64 (8): meta-analysis. Medicine (Baltimore) 95 (29): e4272.
2677–2686. 7 Dall’Era, M., Cisternas, M.G., Snipes, K. et al. (2017). The
4 Weening, J.J., D’Agati, V.D., Schwartz, M.M. et al. incidence and prevalence of systemic lupus Erythematosus
(2004). The classification of glomerulonephritis in in San Francisco County, California: the California lupus
Reference 421
surveillance project. Arthritis Rheumatol. 69 (10): membranous nephropathy. J. Am. Soc. Nephrol. 20 (4):
1996–2005. 901–911.
8 Somers, E.C., Marder, W., Cagnoli, P. et al. (2014). 20 Hultcrantz, M., Rind, D., Akl, E.A. et al. (2017). The
Population-based incidence and prevalence of systemic GRADE Working Group clarifies the construct of
lupus erythematosus: the Michigan lupus epidemiology certainty of evidence. J. Clin. Epidemiol. 87: 4–13.
and surveillance program. Arthritis Rheumatol. 66 (2): 21 Freedman, B.I. and Skorecki, K. (2014). Gene-gene and
369–378. gene-environment interactions in apolipoprotein L1
9 Izmirly, P.M., Wan, I., Sahl, S. et al. (2017). The incidence gene-associated nephropathy. Clin. J. Am. Soc. Nephrol. 9
and prevalence of systemic lupus Erythematosus in (11): 2006–2013.
New York County (Manhattan), New York: the 22 Anders, H.J. and Rovin, B. (2016). A pathophysiology-
Manhattan lupus surveillance program. Arthritis based approach to the diagnosis and treatment of lupus
Rheumatol. 69 (10): 2006–2017. nephritis. Kidney Int. 90 (3): 493–501.
10 Lewis, M.J. and Jawad, A.S. (2017). The effect of ethnicity 23 Anders, H.J. and Fogo, A.B. (2014). Immunopathology
and genetic ancestry on the epidemiology, clinical of lupus nephritis. Semin. Immunopathol. 36 (4):
features and outcome of systemic lupus erythematosus. 443–459.
Rheumatology (Oxford) 56 (suppl_1): i67–i77. 24 Yung, S. and Chan, T.M. (2008). Anti-DNA antibodies in
11 Hanly, J.G., O’Keeffe, A.G., Su, L. et al. (2016). The the pathogenesis of lupus nephritis--the emerging
frequency and outcome of lupus nephritis: results from mechanisms. Autoimmun. Rev. 7 (4): 317–321.
an international inception cohort study. Rheumatology 25 Pollak, V.E., Pirani, C.L., and Schwartz, F.D. (1997). The
(Oxford) 55 (2): 252–262. natural history of the renal manifestations of systemic
12 Langefeld, C.D., Ainsworth, H.C., Cunninghame lupus erythematosus. 1964. J. Am. Soc. Nephrol. 8 (7):
Graham, D.S. et al. (2017). Transancestral mapping and 1189–1198.
genetic load in systemic lupus erythematosus. Nat. 26 Baldwin, D.S., Lowenstein, J., Rothfield, N.F. et al. (1970).
Commun. 8: 16021. The clinical course of the proliferative and membranous
13 Zheng, Z.H., Zhang, L.J., Liu, W.X. et al. (2012). forms of lupus nephritis. Ann. Intern. Med. 73 (6):
Predictors of survival in Chinese patients with lupus 929–942.
nephritis. Lupus 21 (10): 1049–1056. 27 Appel, G.B., Silva, F.G., Pirani, C.L. et al. (1978). Renal
14 Tunnicliffe, D.J., Singh-Grewal, D., Kim, S. et al. (2015). involvement in systemic lupud erythematosus (SLE): a
Diagnosis, monitoring, and treatment of systemic lupus study of 56 patients emphasizing histologic classification.
Erythematosus: a systematic review of clinical practice Medicine (Baltimore) 57 (5): 371–410.
guidelines. Arthritis Care Res. (Hoboken) 67 (10): 28 Churg, J. and Sobin, L.H. (1982). Renal disease:
1440–1452. Classification and atlas of glomerular diseases. Igaku-
15 Ruiz-Irastorza, G., Ramos-Casals, M., Brito-Zeron, P. shoin, Tokyo.
et al. (2010). Clinical efficacy and side effects of 29 Churg, J., Bernstein, J., and Glassock, R. (1995). Renal
antimalarials in systemic lupus erythematosus: a Disease. Classification and Atlas of Glomerular Diseases,
systematic review. Ann. Rheum. Dis. 69 (1): 20–28. 2e. Igaky-Shoin, Tokyo.
16 Pokroy-Shapira, E., Gelernter, I., and Molad, Y. (2014). 30 Moroni, G., Vercelloni, P.G., Quaglini, S. et al. (2018).
Evolution of chronic kidney disease in patients with Changing patterns in clinical-histological presentation
systemic lupus erythematosus over a long-period and renal outcome over the last five decades in a cohort
follow-up: a single-center inception cohort study. Clin. of 499 patients with lupus nephritis. Ann. Rheum. Dis. 77
Rheumatol. 33 (5): 649–657. (9): 1318–1325.
17 Tunnicliffe, D.J., Palmer, S.C., Henderson, L. et al. (2018). 31 Appel, G.B., Contreras, G., Dooley, M.A. et al. (2009).
Immunosuppressive treatment for proliferative lupus Mycophenolate mofetil versus cyclophosphamide for
nephritis. Cochrane Database Syst. Rev. 6: Cd002922. induction treatment of lupus nephritis. J. Am. Soc.
18 Palmer, S.C., Tunnicliffe, D.J., Singh-Grewal, D. et al. Nephrol. 20 (5): 1103–1112.
(2017). Induction and maintenance immunosuppression 32 Dooley, M.A., Jayne, D., Ginzler, E.M. et al. (2011).
treatment of proliferative lupus nephritis: a network Mycophenolate versus azathioprine as maintenance
meta-analysis of randomized trials. Am. J. Kidney Dis. 70 therapy for lupus nephritis. N. Engl. J. Med. 365 (20):
(3): 324–336. 1886–1895.
19 Austin, H.A. 3rd, Illei, G.G., Braun, M.J. et al. (2009). 33 Parikh, S.V., Alvarado, A., Malvar, A. et al. (2015). The
Randomized, controlled trial of prednisone, kidney biopsy in lupus nephritis: past, present, and
cyclophosphamide, and cyclosporine in lupus future. Semin. Nephrol. 35 (5): 465–477.
422 Lupus Nephritis
34 Austin, H.A. 3rd, Muenz, L.R., Joyce, K.M. et al. (1984). 46 Fulcher, J., O’Connell, R., Voysey, M. et al. (2015).
Diffuse proliferative lupus nephritis: identification of Efficacy and safety of LDL-lowering therapy among men
specific pathologic features affecting renal outcome. and women: meta-analysis of individual data from
Kidney Int. 25 (4): 689–695. 174,000 participants in 27 randomised trials. Lancet 385
35 Cade, R., Spooner, G., Schlein, E. et al. (1973). (9976): 1397–1405.
Comparison of azathioprine, prednisone, and heparin 47 Palmer, S.C., Navaneethan, S.D., Craig, J.C. et al. (2014).
alone or combined in treating lupus nephritis. Nephron HMG CoA reductase inhibitors (statins) for people with
10 (1): 37–56. chronic kidney disease not requiring dialysis. Cochrane
36 Cameron, J.S. (1999). Lupus nephritis. J. Am. Soc. Database Syst. Rev. (5): Cd007784.
Nephrol. 10 (2): 413–424. 48 Hughes, G. (2018). Hydroxychloroquine: an update.
37 Tam, L.S., Li, E.K., Lai, F.M. et al. (2003). Mesangial Lupus 27 (9): 1402–1403.
lupus nephritis in Chinese is associated with a high rate 49 Lightstone, L., Doria, A., Wilson, H. et al. (2018). Can we
of transformation to higher grade nephritis. Lupus 12 (9): manage lupus nephritis without chronic corticosteroids
665–671. administration? Autoimmun. Rev. 17 (1): 4–10.
38 Collado, M.V., Dorado, E., Rausch, S. et al. (2016). 50 Felson, D.T. and Anderson, J. (1984). Evidence for the
Long-term outcome of lupus nephritis class II in superiority of immunosuppressive drugs and prednisone
argentine patients: an open retrospective analysis. J. Clin. over prednisone alone in lupus nephritis. Results of a
Rheumatol. 22 (6): 299–306. pooled analysis. N. Engl. J. Med. 311 (24): 1528–1533.
39 Group KDIGOKGW (2012). KDIGO Clinical Practice 51 Austin, H.A. 3rd, Klippel, J.H., Balow, J.E. et al. (1986).
Guideline for Glomerulonephritis. Kidney Int. Suppl. 2: Therapy of lupus nephritis. Controlled trial of
139–274. prednisone and cytotoxic drugs. N. Engl. J. Med. 314
40 Bertsias, G.K., Tektonidou, M., Amoura, Z. et al. (2012). (10): 614–619.
Joint European League Against Rheumatism and 52 Boumpas, D.T., Austin, H.A. 3rd, Vaughn, E.M. et al.
European Renal Association-European Dialysis and (1992). Controlled trial of pulse methylprednisolone
Transplant Association (EULAR/ERA-EDTA) versus two regimens of pulse cyclophosphamide in severe
recommendations for the management of adult and lupus nephritis. Lancet 340 (8822): 741–745.
paediatric lupus nephritis. Ann. Rheum. Dis. 71 (11): 53 Gourley, M.F., Austin, H.A. 3rd, Scott, D. et al. (1996).
1771–1782. Methylprednisolone and cyclophosphamide, alone or in
41 The GISEN Group (1997). Randomised placebo- combination, in patients with lupus nephritis. A randomized,
controlled trial of effect of ramipril on decline in controlled trial. Ann. Intern. Med. 125 (7): 549–557.
glomerular filtration rate and risk of terminal renal 54 Illei, G.G., Austin, H.A., Crane, M. et al. (2001).
failure in proteinuric, non-diabetic nephropathy. The Combination therapy with pulse cyclophosphamide plus
GISEN Group (Gruppo Italiano di Studi Epidemiologici pulse methylprednisolone improves long-term renal
in Nefrologia). Lancet 349 (9069): 1857–1863. outcome without adding toxicity in patients with lupus
42 Avina-Zubieta, J.A., To, F., Vostretsova, K. et al. (2017). nephritis. Ann. Intern. Med. 135 (4): 248–257.
Risk of myocardial infarction and stroke in newly 55 Houssiau, F.A., Vasconcelos, C., D’Cruz, D. et al. (2002).
diagnosed systemic lupus Erythematosus: a general Immunosuppressive therapy in lupus nephritis: the
population-based study. Arthritis Care Res. (Hoboken) 69 Euro-Lupus Nephritis Trial, a randomized trial of
(6): 849–856. low-dose versus high-dose intravenous
43 Hermansen, M.L., Lindhardsen, J., Torp-Pedersen, C. cyclophosphamide. Arthritis Rheum. 46 (8): 2121–2131.
et al. (2017). The risk of cardiovascular morbidity and 56 Grootscholten, C., Ligtenberg, G., Hagen, E.C. et al.
cardiovascular mortality in systemic lupus erythematosus (2006). Azathioprine/methylprednisolone versus
and lupus nephritis: a Danish nationwide population- cyclophosphamide in proliferative lupus nephritis. A
based cohort study. Rheumatology (Oxford) 56 (5): randomized controlled trial. Kidney Int. 70 (4): 732–742.
709–715. 57 Flanc, R.S., Roberts, M.A., Strippoli, G.F. et al. (2004).
44 Liu, Y. and Kaplan, M.J. (2018). Cardiovascular disease in Treatment for lupus nephritis. Cochrane Database Syst.
systemic lupus erythematosus: an update. Curr. Opin. Rev. (1): Cd002922.
Rheumatol. 30 (5): 441–448. 58 Chan, T.M., Li, F.K., Tang, C.S. et al. (2000). Efficacy of
45 Skaggs, B.J., Hahn, B.H., and McMahon, M. (2012). mycophenolate mofetil in patients with diffuse
Accelerated atherosclerosis in patients with SLE-- proliferative lupus nephritis. Hong Kong-Guangzhou
mechanisms and management. Nat. Rev. Rheumatol. 8 Nephrology Study Group. N. Engl. J. Med. 343 (16):
(4): 214–223. 1156–1162.
Reference 423
9 Ginzler, E.M., Dooley, M.A., Aranow, C. et al. (2005).

5 College of Rheumatology response criteria for
Mycophenolate mofetil or intravenous cyclophosphamide proliferative and membranous renal disease in systemic
for lupus nephritis. N. Engl. J. Med. 353 (21): 2219–2228. lupus erythematosus clinical trials. Arthritis Rheum. 54
60 Contreras, G., Pardo, V., Leclercq, B. et al. (2004). (2): 421–432.
Sequential therapies for proliferative lupus nephritis. N. 73 Wofsy, D., Hillson, J.L., and Diamond, B. (2012).
Engl. J. Med. 350 (10): 971–980. Abatacept for lupus nephritis: alternative definitions of
61 Houssiau, F.A., D’Cruz, D., Sangle, S. et al. (2010). complete response support conflicting conclusions.
Azathioprine versus mycophenolate mofetil for long-term Arthritis Rheum. 64 (11): 3660–3665.
immunosuppression in lupus nephritis: results from the 74 Dooley, M.A., Houssiau, F., Aranow, C. et al. (2013).
MAINTAIN Nephritis Trial. Ann. Rheum. Dis. 69 (12): Effect of belimumab treatment on renal outcomes: results
2083–2089. from the phase 3 belimumab clinical trials in patients
62 Henderson, L., Masson, P., Craig, J.C. et al. (2012). with SLE. Lupus 22 (1): 63–72.
Treatment for lupus nephritis. Cochrane Database Syst. 75 Rovin, B.H., Parikh, S.V., Hebert, L.A. et al. (2013). Lupus
Rev. 12: Cd002922. nephritis: induction therapy in severe lupus nephritis –
63 Bao, H., Liu, Z.H., Xie, H.L. et al. (2008). Successful should MMF be considered the drug of choice? Clin. J.
treatment of class V+IV lupus nephritis with multitarget Am. Soc. Nephrol. 8 (1): 147–153.
therapy. J. Am. Soc. Nephrol. 19 (10): 2001–2010. 76 Walsh, M., Solomons, N., Lisk, L. et al. (2013).
64 Liu, Z., Zhang, H., Liu, Z. et al. (2015). Multitarget Mycophenolate mofetil or intravenous cyclophosphamide
therapy for induction treatment of lupus nephritis: a for lupus nephritis with poor kidney function: a subgroup
randomized trial. Ann. Intern. Med. 162 (1): 18–26. analysis of the Aspreva Lupus Management Study. Am. J.
65 Rovin, B.H. and Parikh, S.V. (2014). Lupus nephritis: the Kidney Dis. 61 (5): 710–715.
evolving role of novel therapeutics. Am. J. Kidney Dis. 63 77 Rauova, L., Lukac, J., Levy, Y. et al. (2001). High-dose
(4): 677–690. intravenous immunoglobulins for lupus nephritis--a
66 Rovin, B.H., Furie, R., Latinis, K. et al. (2012). Efficacy salvage immunomodulation. Lupus 10 (3):
and safety of rituximab in patients with active 209–213.
proliferative lupus nephritis: the lupus nephritis 78 Boletis, J.N., Ioannidis, J.P., Boki, K.A. et al. (1999).
assessment with rituximab study. Arthritis Rheum. 64 (4): Intravenous immunoglobulin compared with
1215–1226. cyclophosphamide for proliferative lupus nephritis.
67 Gomez Mendez, L.M., Cascino, M.D., Garg, J. et al. Lancet 354 (9178): 569–570.
(2018). Peripheral blood B cell depletion after rituximab 79 Lewis, E.J., Hunsicker, L.G., Lan, S.P. et al. (1992). A
and complete response in lupus nephritis. Clin. J. Am. controlled trial of plasmapheresis therapy in severe lupus
Soc. Nephrol. 13 (10): 1502–1509. nephritis. The Lupus Nephritis Collaborative Study
68 Condon, M.B., Ashby, D., Pepper, R.J. et al. (2013). Group. N. Engl. J. Med. 326 (21): 1373–1379.
Prospective observational single-centre cohort study to 80 Kattah, A.G. and Garovic, V.D. (2015). Pregnancy and
evaluate the effectiveness of treating lupus nephritis with lupus nephritis. Semin. Nephrol. 35 (5):
rituximab and mycophenolate mofetil but no oral 487–499.
steroids. Ann. Rheum. Dis. 72 (8): 1280–1286. 81 Lightstone, L. and Hladunewich, M.A. (2017). Lupus
69 Ruggenenti, P., Fervenza, F.C., and Remuzzi, G. (2017). nephritis and pregnancy: concerns and management.
Treatment of membranous nephropathy: time for a Semin. Nephrol. 37 (4): 347–353.
paradigm shift. Nat. Rev. Nephrol. 13 (9): 563–579. 82 Moroni, G., Gatto, M., Raffiotta, F. et al. (2018). Can we
70 Mysler, E.F., Spindler, A.J., Guzman, R. et al. (2013). withdraw immunosuppressants in patients with lupus
Efficacy and safety of ocrelizumab in active proliferative nephritis in remission? An expert debate. Autoimmun.
lupus nephritis: results from a randomized, double-blind, Rev. 17 (1): 11–18.
phase III study. Arthritis Rheum. 65 (9): 2368–2379. 83 Moroni, G., Longhi, S., Giglio, E. et al. (2013). What
71 Furie, R., Nicholls, K., Cheng, T.T. et al. (2014). Efficacy happens after complete withdrawal of therapy in patients
and safety of abatacept in lupus nephritis: a twelve- with lupus nephritis. Clin. Exp. Rheumatol. 31 (4 Suppl
month, randomized, double-blind study. Arthritis 78): S75–S81.
Rheumatol. 66 (2): 379–389. 84 De Rosa, M., Azzato, F., Toblli, J.E. et al. (2018). A
72 Liang, M.H., Schur, P.H., Fortin, P. et al. for the Renal prospective observational cohort study highlights kidney
Disease Subcommittee of the American College of biopsy findings of lupus nephritis patients in remission
Rheumatology Ad Hoc Committee on Systemic Lupus who flare following withdrawal of maintenance therapy.
Erythematosus Response Criteria (2006). The American Kidney Int. 94 (4): 788–794.
424 Lupus Nephritis
5 Liyanage, T., Ninomiya, T., Jha, V. et al. (2015). Worldwide

8 90 Nossent, H.C., Swaak, T.J., and Berden, J.H. (1990).
access to treatment for end-stage kidney disease: a Systemic lupus erythematosus: analysis of disease activity
systematic review. Lancet 385 (9981): 1975–1982. in 55 patients with end-stage renal failure treated with
86 Sabucedo, A.J. and Contreras, G. (2015). ESKD, hemodialysis or continuous ambulatory peritoneal dialysis.
transplantation, and dialysis in lupus nephritis. Semin. Dutch working party on SLE. Am. J. Med. 89 (2):
Nephrol. 35 (5): 500–508. 169–174.
87 Zhang, L., Lee, G., Liu, X. et al. (2016). Long-term 91 Nossent, H.C., Swaak, T.J., and Berden, J.H. (1991).
outcomes of end-stage kidney disease for patients with Systemic lupus erythematosus after renal transplantation:
lupus nephritis. Kidney Int. 89 (6): 1337–1345. patient and graft survival and disease activity. The Dutch
88 Sexton, D.J., Reule, S., Solid, C. et al. (2015). ESRD from working party on systemic lupus Erythematosus. Ann.
lupus nephritis in the United States, 1995-2010. Clin. J. Intern. Med. 114 (3): 183–188.
Am. Soc. Nephrol. 10 (2): 251–259. 92 Ribeiro, F.M., Leite, M.A., Velarde, G.C. et al. (2005).
89 Contreras, G., Mattiazzi, A., Guerra, G. et al. (2010). Activity of systemic lupus erythematosus in end-stage
Recurrence of lupus nephritis after kidney renal disease patients: study in a Brazilian cohort. Am. J.
transplantation. J. Am. Soc. Nephrol. 21 (7): 1200–1207. Nephrol. 25 (6): 596–603.
425
26
Hemolytic Uremic Syndrome

Asaf Lebel1, Amrit Kirpalani1, and Christoph Licht1,2
1
Division of Nephrology and Cell Biology Program, Research Institute, The Hospital for Sick Children, Toronto, ON, Canada
2
Department of Paediatrics, University of Toronto, Toronto, ON, Canada
I ntroduction to the different etiologies which cause this syndrome, as

shown in Figure 26.1. As aHUS is known to be caused by
The Nomenclature of TMA complement dysregulation the term complement-
mediated TMA is preferred. Rare primary TMA are not
Thrombotic microangiopathy (TMA) describes a group of related to complement may be termed primary noncom-
highly diverse disorders in children and adults that is plement TMA. All other forms of TMA that are have been
defined by microvascular thrombosis with characteristic included in the aHUS group are classified as secondary
abnormalities of the vascular endothelial lining. Clinically, TMA, which should be termed based on the underlying
TMA is characterized by a triad including microangio- disorder, i.e. infection-related TMA, autoimmune disorder-
pathic hemolytic anemia (MAHA), thrombocytopenia, and related TMA, drug-related TMA, etc.
organ dysfunction, primarily of the kidneys [1–3]. The pathophysiology-based classification used in this
TMA can be hereditary (either familial or sporadic) or chapter provides a practical guide for clinicians facing a
acquired, and it can also be primary or secondary to patient with an undiagnosed TMA. However, the classifi-
numerous different conditions including infections, cation of TMA is still evolving, and it is likely that with
immunological diseases, glomerular diseases, stem cell time new classifications will be suggested as despite great
and solid organ transplantation, drugs, cancer, malignant advancement in the revelation of molecular basis of TMA
hypertension, and pregnancy [2–4]. Thus, the classifica- over the last 20 years, a considerable number of cases (up
tion of TMA is challenging and debatable, and today there to 30%) are still classified as “unknown” or “idiopathic”
is no uniform accepted nomenclature in the literature or TMA, and further pathophysiological insights can be
in clinical practice. Historical classifications were based expected [2–4, 7, 8] Table 26.1.
on clinical features, mainly distinguishing between throm-
botic thrombocytopenic purpura (TTP), which is clinically
The TMA Spectrum
characterized by predominant neurologic involvement
and is more common in adults, and hemolytic uremic syn- The first report of TMA, presumably TTP, dates from 1924,
drome (HUS), which involves primarily the kidneys and is when Moschcowitz described a 16-year-old girl with pallor,
more common in children. HUS has been further divided purpura, and hemiparesis who died with cardiac failure.
into two diseases, Shiga toxin-producing Escherichia coli On autopsy, thrombi were seen in terminal arterioles and
(E. coli)-related HUS (STEC-HUS, typical HUS) and atypi- capillaries in many organs [1, 3]. It was not until 1982,
cal hemolytic uremic syndrome (aHUS), which has been however, that unusually large multimers of von Willebrand
used for any TMA type that is not related to STEC-HUS or factor (vWF) were found to accumulate in the plasma of
TTP [2–6]. More recent classifications which are based on patients with chronic TTP, leading to occlusive microvas-
pathophysiology have evolved due to greater understating cular platelet thrombi. Subsequently, a deficiency (enzyme
of the underlying mechanism of the different types of activity <10%) of a disintegrin and metalloproteinase with
TMA [2, 3, 5]. We, among other authors, opt to further a thrombospondin type 1 motif member 13 (ADAMTS13),
classify the highly heterogeneous group aHUS according a vWF cleaving-protease, was found to be the cause of the
Microangiopathic Prolonged PPT and INR, low fibrinogen
hemolytic anemia, DIC
thrombocytopenia,
organ failure
TTP ADAMTS13 < 10%
TMA
Positive STEC in stool culture/PCR
STEC-HUS
Secondary TMA Primary TMA
Hematopoietic Non-
Glomerular Solid organ Malignant Complement
Infection Autoimmunity stem cell Drugs Cancer Pregnancy complement
diseases transplantation hypertension mediated
transplantation mediated
Bacteria: SLE IgA nephropathy Calcineurin inhibitors: Gastric MMACHC (Col-C) Congenital:
Pneumococcus APLA syndrome ANCA associated vasculitis Tacrolimus Ovarian DGKE CFH
Bordetella pertussis Systemic Membranous nephropathy Cyclosporine Breast INF2 MCP/CD46
Snigella dysenteriae Sclerosis FSGS Lymphoma PLG CFHRs
Dermatomyositis C3 glomerulopathy Anti VEGF: Colon VTN CFI
Viruses: Bevacizumab Pancreas vWF C3
Influenza Sunitinib F12 CFB
HIV THBD/CD141
CMV Chemotherahy:
BBV Mitomycin Acquired:
HHVS Cisplatin Anti-CFH
HCV Vincristine autoantibodies
HAV Gemoitabine
Parvovirus B19 Others:
Quinine
Parasites:
Sirolimus
Malaria Cocaine
Interferon α/β
Figure 26.1 Classification and diagnostic workup of TMA. ADAMTS13, a disintegrin and metalloproteinase with a thrombospondin type 1 motif member 13; ANCA, antineutrophil
cytoplasmic antibody; APLA, antiphospholipid antibody; CFB, complement factor B; CFH, complement factor H; CFHR, complement factor H related protein; CFI, complement factor
I; CMV, cytomegalovirus; DGKE, diacylglycerol kinase ε; DIC, disseminated intravascular coagulation; EBV, Epstein–Barr virus; F12, factor XII; FSGS, focal segmental
glomerulosclerosis; HAV, hepatitis A virus; HCV, hepatitis C virus; HHV6, human herpes virus 6; HIV, human immunodeficiency virus; HUS, hemolytic uremic syndrome; IgA,
immunoglobulin A; INF2, inverted formin-2; INR, international normalized ratio; MCP/CD46, membrane cofactor protein; MMACHC,– methylmalonic aciduria and homocystinuria
type C protein; PCR, polymerase chain reaction; PLG, plasminogen; PTT, partial thromboplastin time; SLE, systemic lupus erythematosus; STEC, Shiga toxin producing E. coli; THBD/
CD141, thrombomodulin; TMA, thrombotic microangiopathy; TTP, thrombotic thrombocytopenic purpura; VEGF, vascular endothelial growth factor; VTN, vitronectin.
Introductio 427
Table 26.1 Summary of diagnostic tests in TMA.
TMA diagnosis
Creatinine, urea ●● Kidney function tests ●● Both are elevated in TMA and signify AKI
Hemoglobin, haptoglobin, ●● MAHA markers ●● Low hemoglobin and haptoglobin
LDH, direct antiglobulin ●● High LDH
(coombs) test, blood film ●● Negative coombs test (except in pneumococcal TMA)
●● Fragmented red blood cells (schistocytes) on blood film
Platelets ●● Platelet count ●● Thrombocytopenia is usually the first sign to appear and increase
in platelets count is usually the first sign of recovery
Further assessment of kidney involvement

Urine analysis, urine ●● Degree of hematuria and ●● Proteinuria is often nephrotic-ranged
protein, or albumin to proteinuria ●● Hematuria is often the first sign of relapse of complement-
creatinine ratio mediated TMA
Kidney US + Doppler ●● Structure of the kidneys, ●● Findings in TMA are not specific, however kidney US + Doppler
patency of kidney vessels may aid in excluding other clinically relevant kidney diseases
such as renal vein thrombosis
Assessment of extra-renal involvement

Brain MRI ●● Structure of the brain ●● For assessment of neurologic complications; findings include
posterior reversible encephalopathy syndrome (PRES),
hemorrhage, infarcts, edema, and others
Chest x-ray ●● Structure of the lungs and ●● For assessment of pulmonary complications; findings include
heart pulmonary edema, pulmonary hemorrhage, and others
ECG, echocardiography ●● Heart structure and function, ●● For assessment of cardiac complications; findings include left
arrhythmias ventricular hypertrophy, cardiomyopathy, valvular disease,
myocardial infarction, and others
Abdominal US/CT ●● Structure of abdominal ●● For assessment of gastrointestinal complications; findings include
viscera cholelithiasis, pancreatitis, hepatitis, and others
AST, ALT, GGT ●● Liver enzymes tests ●● For assessment of hepatitis as a complication of TMA or as a
cause of TMA (i.e. in HAV or HCV)
Amylase, lipase ●● Pancreatic enzymes tests ●● For assessment pancreatitis as a complication of TMA
Exclusion of DIC
PTT, INR, fibrinogen ●● Basic coagulation tests PTT and INR are prolonged and fibrinogen is low in DIC, while in
TMA these tests are usually normal
Diagnosis/exclusion of TTP
ADAMTS13 activity ●● Activity of the vWF ●● Activity level >10% rules out TTP
cleaving-protease
Diagnosis/exclusion of STEC-HUS
Stool culture and/or PCR ●● Microbiologic tests for STEC ●● In STEC-HUS these tests are positive for E. coli O157:H7 or other
identification less common STEC
Diagnosis/exclusion of secondary TMA

Blood/stool cultures, viral ●● Microbiologic tests for ●● For diagnosis/exclusion of infection-related TMA in the relevant
serology/PCR non-STEC pathogens which clinical context
may cause TMA (i.e. blood
culture for pneumococcus,
stool culture for Shigella
dysenteriae, serology for HIV,
PCR for influenza)
(Continued)
428 Hemolytic Uremic Syndrome
ANA, anti dsDNA (SLE) ●● Autoimmune antibodies in ●● For diagnosis/exclusion of autoimmune disease-related TMA in
Lupus anticoagulant, SLE the relevant clinical context
anticardiolipin antibodies ●● Autoimmune antibodies in
(APLA) APLA
Antitopoisomerase I, ●● Autoimmune antibodies in
anti-RNA polymerase III, systemic sclerosis
anticentromere antibody
(systemic sclerosis)
ANCA, C3, IgA ●● Markers of different types of ●● For diagnosis/exclusion of GN-related TMA in the relevant
GN clinical context
●● Positive ANCA may signify ANCA-associated glomerulonephritis
(GN)
●● Low C3 is not specific and can be seen in many GN as well as in
complement-mediated TMA
●● Elevated IgA in the right clinical context may suggest IgA
nephropathy, but normal IgA is common in this entity
Kidney biopsy ●● Histopathology of the kidney ●● For diagnosis/exclusion of GN-related TMA and of solid organ
transplantation-related TMA in the relevant clinical context
●● Kidney biopsy is often needed for the diagnosis of the specific
type of GN, and of TMA associated with solid organ
transplantation, but it is not required for the diagnosis of TMA
Urine/serum β-HCG ●● Pregnancy test ●● For diagnosis/exclusion of pregnancy-related TMA
●● Should be part of the diagnostic work-up of any case of TMA in a
woman of childbearing age
Blood homocysteine, blood ●● Molecules related to ●● For diagnosis/exclusion of cobalamin C-related TMA
methionine, urine cobalamin metabolism ●● Blood homocysteine and urine methylmalonic acid are elevated
methylmalonic acid in cobalamin C-related TMA, while blood methionine is reduced
Diagnosis/exclusion of primary noncomplement-mediated TMA

Genetic tests of DGKE, ●● Screening for pathogenic ●● For diagnosis/exclusion of primary, noncomplement-mediated
MMACHC, INF2, PLG, variants in non-complement TMA
VTN, vWF, F12 related genes ●● DGKE gene should be tested in infantile onset TMA
●● MMACHC gene should be tested to confirm the diagnosis of
cobalamin C-related TMA if the screening biochemical tests
(blood homocysteine and methionine, urine methylmalonic acid)
suggest this disorder
Diagnosis of primary, complement-mediated TMA

Genetic tests of CFH, MCP/ ●● Screening for pathogenic ●● For molecular confirmation of the clinical diagnosis of the
CD46, CFHRs, CFI, C3, variants in complement- congenital form of complement-mediated TMA
CFB, THBD/CD141 related genes
Anti CFH-autoantibodies ●● Screening for anti-CFH ●● For diagnosis/exclusion of the acquired form of complement-
autoantibodies mediated TMA
Further investigation in primary complement-mediated TMA

CH50, AP50, CFH activity ●● Complement functional ●● Low CH50 may result from consumption of complement factors
assays in complement-mediated TMA but also from other causes such as
hepatic disease. It may also be marker of eculizumab therapy
●● Low AP50 is common during the active phase of complement-
mediated TMA but may be low also in STEC-HUS and other
disorders
●● Decreased CFH activity implies a CFH mutation or anti-CFH
autoantibodies
Introductio 429
C3, C4, CFH, CFI, CFB, ●● Levels of complement ●● C3 is low in only 30–50% of patients with complement-mediated
MCP/CD46 components and inhibitors TMA (nonspecific)
●● Low C4 is associated with complement classical pathway
disorders and not with complement-mediated TMA
●● Low CFH, CFI, or MCP/CD46 may signify mutations in these
genes
C5b-C9, C3d, Bb ●● Degree of complement ●● High C5b-9 levels (membrane attack complex) expected in active
activation complement mediated TMA; may be also seen in STEC-HUS
●● High C3d in combination with low C3 levels associated with C3
consumption rather than deficiency
●● High Bb and low CFB level found in both, complement mediated
TMA and STEC-HUS
disease. In hereditary TTP, recessive mutations in the gene symptoms. However, TMA can also occur in these patients
encoding for ADAMTS13 are responsible for the deficiency, as abnormalities in cobalamin (vitamin B12) metabolism
while autoantibodies against ADAMTS13 cause the more may lead to endothelial dysfunction. This disorder is
common acquired TTP [1–3, 9, 10]. TTP was historically uncommon but relevant, since both diagnostic tests (serum
almost universally fatal, but the systematic use of plasma homocysteine and urine methyl-malonic acid levels) and
exchange (PE) therapy decreased mortality to <10% [2]. efficient therapies (with hydroxycobalamin, folinic acid,
With a frequency of 1–3/100 000, STEC-HUS is the most and betaine) are available, simple, and inexpensive [3, 15,
common type of TMA in children [2, 7, 11]. The term 16] Table 26.1. Another disorder in this group is diacylglyc-
“hemolytic uremic syndrome” was first proposed in 1955 erol kinase ε (DGKE)-related TMA. Homozygous muta-
for children with renal failure, anemia, and thrombocyto- tions in DGKE result in infantile aHUS, but may also cause
penia, and subsequently a preceding hemorrhagic colitis membranoproliferative glomerulonephritis (MPGN) or a
was described [12, 13]. In 1982, during an investigation of clinical and pathological overlap between these two disor-
outbreaks of hemorrhagic colitis, E. coli O157:H7 was iden- ders. The mechanism is unknown, but may be related to
tified as a pathogen and in the same year the association protein kinase C-mediated upregulation of prothrombotic
between HUS and STEC was described [1, 3, 11]. The terms factors. Proteinuria, often nephrotic, is a major component
“D+ HUS” (diarrhea positive HUS) for HUS caused by of this disorder and most patients slowly progress to end-
STEC or other rarer enteric bacteria and “D−HUS” (diar- stage renal disease (ESRD) [2–4, 6, 17–19]. Recently,
rhea negative HUS) for other TMA should be deferred, as patients from two different families were found to have
some (albeit rare) patients with STEC-HUS do not have TMA presumably caused by mutations in inverted formin-
diarrhea, while up to 30% of patients with complement- 2 (INF2) gene, which is also known to be involved in focal
mediated TMA may have diarrhea at presentation [2, 5]. and segmental glomerulosclerosis (FSGS). Interestingly,
Complement-mediated TMA is primarily caused by two patients from one family were also diagnosed with the
mutations in complement factors (regulators and/or acti- hereditary peripheral neuropathy disorder Charcot–Marie–
vators), leading to uncontrolled activity of the complement Tooth [20]. It is likely that TMA-causing mutations in novel
alternative pathway (AP), but it can be also caused by genes will be added to this group of primary noncomple-
acquired abnormalities such as antibodies against complement TMA in the future.
ment factor H (CFH) [2, 4, 5, 14]. Many diseases and conditions may lead to secondary
Primary noncomplement-mediated TMA is caused by TMA. Infectious agents include bacteria other than STEC,
mutations in genes that are typically not related to the which can produce Shiga toxin (such as Shigella dysente-
complement system. The first of these is cobalamin riae), Streptococcus pneumoniae, and viruses such as HIV
C-related TMA, which is a hereditary metabolic cause of and influenza A and B [2, 7, 21–24]. TMA may complicate
TMA caused by recessive mutations in methylmalonic aci- autoimmune diseases such as systemic lupus erythematosus
duria and the homocystinuria type C protein (MMACHC) (SLE), catastrophic antiphospholipid syndrome (CAPLS),
gene. Patients usually present in infancy with diverse and systemic sclerosis [2, 25–28]. It can be seen in association
developmental abnormalities and primarily neurologic with other glomerular disorders, including immunoglobulin
A (IgA) nephropathy, anti-neutrophil cytoplasmic antibody However, outbreaks of O157:H7 or other serotypes are also
(ANCA)-associated vasculitis, and others, although this may common, and the most significant recent outbreaks
be a histopathologic finding only without the typical clinical occurred in Germany in 2011 (E. coli O104/H4), in
findings [2, 29, 30]. TMA is a known and multifactorial com- Oklahoma, USA, in 2008 (E. coli O111/NM), and in Ontario,
plication of hematopoietic stem cell transplantation [31–34]. Canada, in 2000 (E. coli O157:H7) [51–54]. Most E. coli
It can also complicate solid organ transplantations, includ- O157:H7 infections and HUS occur in summer and
ing kidney transplants, either as recurrence of known autumn [5, 9]. The primary reservoir of E. coli O157:H7 is
complement-mediated TMA post-transplant or, more com- the intestinal tract of cattle, and it is more common in rural
monly, de novo [7, 35, 36]. Many different drugs have been areas and in countries with high-density cattle raising,
suspected as a cause of TMA, though a clear evidence for such as Scotland and Argentina. Ground beef, lettuce,
causal association has been shown with relatively few. sprouts, municipal water, and many other sources have
Quinine is an example of a drug which may cause TMA by been all recognized as major transmission vehicles of
immune mechanism, while calcineurin inhibitors (CNI) STEC, while airborne and person-to-person transmission
may represent drugs which cause TMA via direct toxicity [2, have also been described [11, 55, 56]. The risk of develop-
3, 37, 38]. Cancer has long been associated with risk of TMA, ing HUS after STEC-colitis is around 10% in sporadic infec-
either caused by malignancy itself or more often resulting tions, and up to 20% in epidemic outbreaks [4, 7, 11, 55].
from anticancer therapy [2, 39, 40]. Malignant hypertension Complement-mediated TMA accounts for 5–10% of all
with concomitant TMA places a clinical dilemma as any HUS cases, with incidence of 0.5–2 cases per million per
TMA can cause hypertension which may be severe, there- year [5, 7, 8, 57]. It can occur in children and adults, and
fore it may not be clear which entity is the cause and which may be sporadic (most often) or familial. Genetic muta-
is the consequence [2, 41–44]. Finally, pregnancy and the tions affecting genes related to the complement alternative
postpartum period are associated with various TMA syn- pathway or autoantibodies against CFH are found in
dromes: complement-mediated TMA, TTP, and the TMA- 50–60% of cases of complement-mediated TMA [4–6, 8,
like syndrome HELLP (hemolysis, elevated liver enzymes, 57–59]. In 40–50% of patients clinically diagnosed with
and low platelets), which is related to eclampsia [45–48]. complement-mediated TMA, no specific cause, genetic or
Importantly, in nearly all of these secondary forms of TMA autoimmune, is identified. CFH mutations were the first
abnormalities in complement-related genes have been to be described and are the most common abnormality
described in some patients, many of whom were treated suc- responsible for complement-mediated TMA, accounting for
cessfully with eculizumab, the complement C5-inhibitor [25, 24–27% of cases [6, 58, 60, 61]. The second most common
26, 39, 41, 45, 49, 50]. This implies that a proportion of mutations (7–9%) are found in the gene encoding for mem-
patients with secondary forms of TMA actually have pri- brane cofactor protein (MCP or CD46), followed by com-
mary complement-mediated TMA and the secondary condi- plement factor I (CFI) and complement component 3 (C3)
tion (i.e. pregnancy, drug, infection, etc.) constitutes a trigger mutations, each of which accounts for 4–8% of cases [6, 58,
or a “second hit” which is required for developing the full 60–63]. Mutations in complement factor B (CFB), throm-
picture of TMA, rather than the primary cause of it [2, 4, 21, bomodulin (THBD or CD141), and complement factor
25, 26, 34, 41, 42, 45, 46, 48, 49]. H-related proteins (CFHRs) are responsible together for
The following will focus on the epidemiology, patho- 5–10% of cases [6, 58, 60, 61, 64]. Anti-CFH autoantibodies
physiology, clinical manifestations, diagnosis, prognosis, account for 10–15% of cases and are frequently associated
and treatment of two entities of the TMA spectrum: STEC- in deletions of CFHR1 and 3 [6, 57, 58, 60, 61]. Five to ten
HUS and complement-mediated TMA. percent of complement-mediated TMA patients carry more
than one mutation [57, 58, 65]. The presence of one or
more mutation and/or the presence of specific genetic risk
E
pidemiology variants, as discussed in the next section, may increase the
risk of disease manifestation [3, 4, 6, 7, 56–58, 60, 65].
In children, 85–90% of TMA cases are caused by infection
with STEC. STEC-HUS is more frequent in children than
in adults, with median age around 2 years, although adults P
athophysiology
have more severe disease with higher mortality [3, 4]. The
annual incidence in Europe and North America is 1.9–2.9 The common pathologic basis of all TMA is injury to the
cases per 100 000 children <3–5 years, and it decreases with vascular endothelium with subsequent microvascular
age [4]. The predominant cause in most of the world is E. thrombosis, which leads to consumptive thrombocytopenia,
coli O157:H7 and most occurrences are sporadic [3, 7, 11]. MAHA, and kidney and other organs injury [2–4, 56, 66].
Pathophysiolog 431
STEC-HUS endothelial cell, thereby markedly reducing the regulatory

activity of CFH on complement activation [74, 76].
There has been great advancement in understanding the
mechanism of the disease since the first description of STEC-
HUS. However, one of the main questions regarding STEC- Complement-mediated TMA
HUS pathophysiology – why only 10–20% of patients with
Dysregulation of the complement alternative pathway is the
STEC infection develop HUS –remains unanswered [2–4].
cause of complement-mediated TMA [2–4, 6–8, 78]. The
Enteric STEC infections are almost never accompanied by
complement system is part of the innate immune system,
bacteremia. Rather, the systemic complications, such as
which has an important role in targeting and killing of path-
HUS, and at least partially also the intestinal manifestations,
ogens, and induction of an inflammatory response. It con-
are caused by the (Shiga) toxin, which is the key factor in the
sists of >30 plasma and membrane-bound proteins and has
pathogenesis of STEC-HUS [7, 11]. There are two kinds of three activation pathways: classical, MBL, and alternative.
toxin, Shiga toxins 1 and 2 (Stx1 and Stx2), with the latter Activation of each of these pathways leads to production of
being more virulent. Both toxins are made of two subunits: protease complexes termed C3 and C5 convertases that
subunit A is an N-glycosidase which inhibits protein synthe- cleave C3 and C5, respectively, eventually resulting in the
sis and subunit B binds to globotriaosylceramide (Gb3, also formation of the membrane-attack complex (MAC; C5b-9),
known as CD77) on the surface endothelial, mesangial, and which causes cell lysis. In addition, activation of C3 and C5
epithelial (podocytes and tubular) cells, as well as blood cells results in the generation of the highly reactive anaphylatox-
including monocytes, polymorphonuclear cells, and plate- ins C3a and C5a, respectively [2, 3, 6, 8, 78, 79].
lets [3, 4, 7, 11, 55, 56]. The strong expression of Gb3 on glo- Unlike the other two pathways, the alternative pathway
merular endothelial cells, as well as the apparent propensity is constitutively active as a result of spontaneous,
of the glomerular circulation to endothelial damage and hydrolysis-mediated generation of C3b on both pathogen
occlusion, is thought to be the explanation of why the kid- and normal host cells, therefore requiring tight regulation
ney is the main organ involved in STEC-HUS [2, 7, 11, 56]. to limit activation both spatially and temporally [2, 3, 6, 79,
After inducing hemorrhagic colitis, Stx binds to these blood 80]. The plasma regulators are CFH and CFI, while MCP
cells and circulates in the bloodstream. Stx is released from (CD46), decay accelerating factor (DAF/CD55), protectin
blood cells within microvesicles, a process which both facili- (CD59), and complement receptor 1 (CR1/CD35) are
tates its transport to the target cells in the kidney and ena- membrane-bound regulators [4, 7, 78, 79]. Thrombomodulin
bles the toxin to evade the host immune system. These (CD141), encoded by THBD gene, is another membrane-
microvesicles were also shown to be prothrombotic and thus bound protein which is an antithrombotic factor, but also
may contribute to the pathogenesis of HUS in several mech- has complement inhibitory activity [6, 8, 81, 82]. Loss of
anisms [67, 68]. The toxin is subsequently internalized to function mutations in these regulatory genes or gain of
endothelial and epithelial kidney cells by endocytosis, splits function mutations in effector genes such as CFB and
into its A and B subunits, and then – via its B subunit – C3 leads to complement dysregulation and overactivation,
inhibits the 28S unit of ribosomal ribonucleic acid (rRNA) eventually causing injury of the vascular endothelium and
and thus protein synthesis. This step induces a stress TMA [3, 4, 6, 7, 60, 62, 64, 78, 79, 83–88].
response that can lead to apoptotic cell death [2, 3, 11, 55, Pathogenic mutations identified in patients with
56]. Another effect of Stx is induction of thrombosis by complement-mediated TMA include nonsense and mis-
enhancement of tissue factor expression, increasing throm- sense mutations, deletions, splice site changes, and com-
bin formation and vWF secretion, inhibition of fibrinolysis, plex rearrangements. Most of these changes are
and direct platelets activation [3, 55, 56, 66, 69]. In addition heterozygous, even though many family members with
to its apoptotic and prothrombotic effects, Stx induces heterozygous mutations are asymptomatic. Thus, the pen-
inflammation by upregulation of cytokines, chemokines, etrance of these mutations is incomplete and is known to
and adhesion molecules [2, 3, 7, 70]. increase with age and with the presence of additional risk
Complement activation, which is responsible for factors [2–4, 6, 7, 57, 61]. Genetic risk factors include a sec-
complement-mediated TMA, has been noted to be involved ond mutation or genetic variants such as specific single
in STEC-HUS as well over the last 10 years [71–77]. There nucleotide polymorphisms in CFH or MCP genes, copy-
is evidence of Stx-induced complement activation via both number variations in CFHR 1 and 3, and fusions genes of
the mannose-binding lectin (MBL) and the alternative the CFHRs region [3, 4, 6, 7, 57, 59, 64, 65, 89]. Additionally,
complement pathways [71, 72, 74, 77]. Moreover, Stx can there are known environmental triggers for complement-
bind CFH and significantly delay its activity as a cofactor mediated TMA episodes or relapses, including infections,
for CFI-mediated cleavage of C3b on the surface of the drugs, and pregnancy [2, 6, 7, 45, 46, 48, 79, 90].
Acquired complement-mediated TMA occurs in the nal manifestations include pulmonary edema, hemorrhage
form of autoantibodies against CFH [2, 6, 8, 91–94]. CFH and embolism, ocular complications, and gangrenous fin-
autoantibodies identified in complement-mediated TMA gertips [4, 11, 92, 97–99]. In complement-mediated TMA
patients exclusively bind to epitopes located to the extrarenal manifestations are more common during the
C-terminus of CFH, thus preventing binding of CFH to initial 6 months post presentation, but can also occur dur-
endothelial surfaces and C3 [6, 8, 93, 95]. There is a strong ing the chronic phase [61].
(but so far unexplained) association between the homozy-
gous deletion of complement factor H-related proteins 1
and 3 (CFHR1 and CFHR3) and the presence of anti-CFH D
iagnosis
autoantibodies [2, 6, 8, 92–94, 96].
Figure 26.1 presents a general approach to diagnosis of
TMA. The defining laboratory features in TMA include
C
linical Manifestation elevated creatinine and urea, thrombocytopenia, and signs
of MAHA. The latter include low hemoglobin, high lactate
The clinical symptoms of TMA reflect anemia, thrombocyto- dehydrogenase (LDH), low haptoglobin, negative direct
penia, and renal or other organ involvement, and include pal- antiglobulin (coombs) test (except in pneumococcal TMA),
lor, fatigue, shortness of breath, oliguria, and edema [4]. In and evidence of fragmented red blood cells (schistocytes)
almost all cases of STEC-HUS, manifestations are preceded on peripheral blood film. Urinalysis usually shows hema-
by severe abdominal pain and diarrhea, which start 2–12 days turia and proteinuria [2, 4, 5, 101].
(average 3) after ingestion of contaminated vehicle. The diar- The typical findings on kidney biopsy are thickening of
rhea becomes bloody in up to 90% of patients, and 30–60% of arterioles and capillaries, endothelial swelling and detach-
patients have fever and vomiting. TMA manifestations in ment, subendothelial accumulation of proteins and cell
those patients who develop the syndrome occurs after the debris, and fibrin and platelet-rich thrombi obstructing
colitis improves, approximately a week after the onset of diar- vessel lumina. However, biopsy is not requisite for the diag-
rhea [3, 4, 7, 11]. Half of complement-mediated TMA epi- nosis of TMA, and it cannot distinguish between the differ-
sodes are triggered by intercurrent events, including upper ent etiologies of TMA [2, 3, 5, 6, 8].
respiratory tract infection, gastroenteritis, vaccination, and The first step in the diagnostic process of a patient with
childbirth [4, 57, 58]. Renal involvement includes hematuria, the triad of AKI, MAHA, and thrombocytopenia, who pre-
proteinuria (commonly nephrotic ranged), oliguria, and sumably has TMA, is to exclude the other entity which can
hypertension, which may be severe, especially in complement- present with these manifestations, disseminated intravas-
mediated TMA. The severity of renal involvement is greater cular coagulation (DIC). DIC and TMA can be easily distin-
in adults, and in general varies between mild nephritis with- guished by sending coagulation tests as in the former
out acute kidney injury (AKI) to severe anuric AKI, and partial thromboplastin time (PTT) and prothrombin time
many patients with both STEC-HUS and complement- (PT) or international normalized ratio (INR) are prolonged
mediated TMA require acute dialysis [3, 4, 7, 97]. and fibrinogen level is low, while in the latter these tests
Extrarenal manifestations occur in 19–38% of patients should be normal [2, 102, 103].
with STEC-HUS and complement-mediated TMA [2, 4–6, The next step is investigation for the cause of TMA,
61, 98]. Possible mechanisms include vascular injury which is crucial because different forms of TMA have com-
caused by Stx in STEC-HUS and by complement activation pletely different treatments. Since TTP is a common cause
in complement-mediated TMA, as well as hypertension of TMA, especially in adults, and it can efficiently be
and uremia caused by renal failure [4, 5, 7, 11, 98, 99]. treated via plasma exchange (PE), one of the first priority
Neurologic involvement is the most common extrarenal tests should be the detection of the activity level of
manifestation in both STEC-HUS and complement- ADAMTS13. Activity level >10% rules out TTP [2, 4, 5,
mediated TMA and includes seizures, alteration in con- 104]. The other common cause of TMA, especially in chil-
sciousness, hemiparesis, vision loss, and other rarer dren, is STEC-HUS, and therefore one should always
symptoms [4, 5, 11, 58, 97–100]. Gastrointestinal involve- exclude this cause before diagnosing rarer types of TMA. In
ment is also common and includes diarrhea, vomiting, gas- STEC-HUS, stool cultures should be positive for E. coli
trointestinal bleeding, cholelithiasis, pancreatitis, and O157:H7 or other less common STEC, and another com-
hepatitis [2, 4, 61, 98, 99]. Cardiac complications include monly used diagnostic method is polymerase chain reac-
left ventricular hypertrophy, dilated or hypertrophic cardi- tion (PCR) assay for the bacteria or Stx [11, 55, 105].
omyopathy, valve insufficiency, myocardial infarction, and Exclusion of a secondary TMA (related to infection, auto-
intracardiac thrombus [4, 7, 11, 57, 98, 99]. Other extrare- immunity, glomerular disease, stem cell and solid organ
Prognosi 433
transplantation, drugs, cancer, malignant hypertension, or dropped to 1–4% [106, 107]. CNS involvement is the most
pregnancy) follows the clinical context, history and physical common cause of death, occurring in <50% of children
examination, and additional ancillary tests, depending on with STEC-HUS [108], with cardiac failure, pulmonary
the specific clinical presentation. Screening for cobalamin hemorrhage, bowel perforation, and electrolyte imbal-
C-related TMA, with serum homocysteine and urine meth- ances following [106].
ylmalonic acid levels, which are typically high, is recom- Early literature identified the following risk factors pre-
mended [2, 4, 6]. dicting a more severe course of postdiarrheal HUS: oligoa-
In only 30–50% of patients with complement-mediated nuria, younger age (<2–3 years), high initial leukocyte
TMA are C3 levels low, and they may also be low in other count, hypertension, and severe colitis [106, 109–115].
types of TMA [5, 6, 104]. More specific assays such as CH50 More recent studies have found that CNS involvement,
(measuring total complement activity), AP50 (measuring high hemoglobin, and low sodium levels were the best pre-
alternative pathway activity), CFH functional assay, and dictors of death in STEC-HUS patients [108].
levels of complement activation markers such as C3d and CNS involvement may include seizures, hemiplegia or
C5b-9 allow for insights into functional aspects of the com- hemiparesis, extrapyramidal syndrome, and coma. MRI
plement system. Additionally, comprehensive screening of findings in patients showed that all parts of the CNS were
all genes currently known to be involved in complement- susceptible to injury, and that there was no correlation
mediated TMA is recommended, including CFH, MCP/ between localization of injury and final prognosis. The
CD46, CFI, C3, CFB, CFHR 1, and 5, and THBD/CD141 [2– overall prognosis in this patient group was poor with nine
6, 57]. Infants should also be tested for mutations in DGKE, patient deaths and 13 with severe sequelae [100]. Of note,
which is responsible for a form of primary noncomplement the exact mechanism resulting in CNS injury may differ
TMA [2, 4, 17, 18, 57]. Finally, anti-CFH autoantibodies from that of renal involvement and treatment may require
should be tested for to diagnose the known acquired form a multifaceted approach [116].
of complement-mediated TMA [5, 6, 57, 104]. As mortality has improved markedly in the past few dec-
However, complement functional assays are available only ades, long-term sequelae may still require management in
in few laboratories and, similarly to genetic testing, require STEC-HUS patients. Persistent hypertension is seen in
long time to being processed. Moreover, only 50–60% of 5–15%, progression to CKD in up to 18% of children, and
patients with presumed complement-mediated TMA are ESRD in approximately 3% of patients. Given the multior-
eventually found to have a disease causing mutation in one of gan involvement, complications of the nervous system,
the above-mentioned genes. Thus, complement functional heart, gut, biliary system, and pancreas may need to be fol-
and genetic tests are important to confirm the diagnosis, lowed over time [106].
inform outcome, and guide treatment, but they do not play a
role in the acute diagnosis of complement-mediated TMA,
Complement-mediated TMA
which remains clinical and is established after excluding all
other forms of TMA. In the era of an efficacious anticomple- The prognosis of complement-mediated TMA has classi-
ment therapy, which may completely change outcome, the cally been poor, with most patients developing ESRD
diagnosis of complement-mediated TMA should be made in within 2 years of presentation [5]. However, in the past dec-
a timely manner. Therefore, every patient with no obvious ade the management of complement-mediated TMA has
clinical picture of STEC-HUS, TTP, or a secondary TMA, undergone a paradigm shift and the associated outcomes
with ADAMTS13 activity >10% and with STEC negative have evolved as well.
stool culture or PCR should be diagnosed with complement- Complement-mediated TMA carried a mortality rate of up
mediated TMA, at least until proven otherwise, and treated to 70% before the advent of terminal complement blockade [7].
with anticomplement therapy [2, 4, 6, 101, 103]. Since the monoclonal C5 antibody eculizumab has widely
become available, short-term mortality has fallen and out-
comes have improved significantly [117–119]. At this time,
many long-term outcomes of eculizumab-treated complement-
P
rognosis
mediated TMA patients have yet to become available.
As a disorder of complement dysregulation, identifying
STEC-HUS
specific mutations can offer prognostic information in
When first described, the mortality of STEC-HUS was complement-mediated TMA. Mutations in CFH [61], CFI,
>30% for all cases. In more recent years, with an improved and C3 have been associated with a worse prognosis, while
understanding of pathogenesis, recognition, and early dial- MCP/CD46 mutation carriers with childhood disease onset
ysis for oligoanuric patients, the acute mortality rate has have shown improved outcomes. Adult patients with no
identifiable mutations in a study of 214 patients with potentially require kidney transplantation, clinicians must
complement-mediated TMA were found to have similarly be judicious regarding blood transfusions [121, 122].
poor prognoses as the above-mentioned subgroups, where A patient whose serum hemoglobin concentration is
children without identifiable mutations had similar out- below the center-specific transfusion threshold, but who is
comes to the MCP/CD46 mutation group [120]. well-perfused, hemodynamically stable, and whose other
Interestingly, outcomes from this study show discrepan- hemolytic markers (LDH, haptoglobin, reticulocyte count)
cies based on the age of the patient: children with suggest declining hemolysis may be managed expectantly,
complement-mediated TMA have a higher mortality com- whereas a patient with a low hemoglobin, ongoing hemol-
pared to adults (6–7% vs. <1% at 1 year), whereas progres- ysis, and any signs of cardiovascular instability would
sion to ESRD after the first episode was more frequent in likely necessitate a blood transfusion.
adults (46% vs. 16%). Approximately 56% of adult patients Thrombocytopenia can be significant in all TMAs.
progress to ESRD within the first year of illness [120]. Sex, However, as previously detailed, platelets may contribute
race, time from initial presentation to diagnosis, and family to microthrombi formation and may aggravate the underly-
history of complement-mediated TMA do not appear to be ing microangiopathic cascade, worsening symptoms [121].
associated with risk of progression to ESRD [61]. The safety of platelet transfusion in STEC-HUS is still
A substantial number of patients may also develop extra- uncertain, with mixed data in the literature, but it may be
renal complications in both the chronic and acute phases. reasonable in patients with evidence of active bleeding or
Gastrointestinal, cardiovascular, CNS, and pulmonary those who are undergoing a surgical procedure, and risks
manifestations may impact adults and children and may and benefits must be weighed before any decision to pro-
require long-term follow-up [61]. ceed with transfusion.
The advent of the terminal complement blockader eculi- Intravascular volume must be considered with adminis-
zumab has fundamentally shifted the trajectory of illness tration of blood products as well, and patients may benefit
in complement-mediated TMA, but the disease process still from diuretics during or following transfusion, though not
carries considerable morbidity and mortality risk both in all will respond [121].
the acute and chronic phases.
Disease-specific Therapy
STEC-HUS
T
reatment Prevention
STEC-HUS, unlike other TMAs, has a characteristic ante-
General Supportive Therapy for all TMA
cedent illness of bloody diarrhea. Recognition of this pro-
Acute Kidney Injury dromal phase offers clinicians the opportunity to
Once established, TMA care requires management of olig- implement preventative measures to reduce the risk of pro-
uric AKI. Fluid and electrolyte intake and output should be gression to HUS.
carefully calculated and reassessed as patients may be anu-
ric and require fluid restriction. Dietary and other intake IV Hydration
restrictions on electrolytes such as potassium or phosphate In patients presenting during the diarrheal phase of enter-
may be required as the GFR declines. Patients may also ohemorrhagic Escherichia coli (EHEC) illness, provision of
have intolerance to oral food intake and may require assis- IV fluids may reduce the risk of subsequent HUS [123]. A
tance with enteral feeds or parenteral nutrition. recent meta-analysis established an association between
Hypertension management may also be necessary in these lack of fluid administration and higher hematocrit (>23%)
patients [121]. at onset of diarrheal illness with poor outcomes including
oligoanuria, need for renal replacement therapy, and
Anemia and Thrombocytopenia death [124].
Anemia and thrombocytopenia may be profound, and pro- The role of volume expansion even after onset of HUS
viders are often faced with a dilemma regarding the need has also been explored. In a multicentre surveillance pro-
for transfusion of blood products. In addition to the typical gram of children with STEC-HUS, early IV volume expan-
risks of blood transfusion (e.g. allergic reaction, infection, sion was associated with significantly better short-term
electrolyte derangement), one must be cognizant of the and long-term outcomes compared to historical controls,
exposure of the patient to donor antigens and the subse- including a lower rate of CNS involvement. Restoration of
quent production of donor-specific antibodies. In a patient the circulating volume was hypothesized to reduce the for-
with a potential risk of developing CKD and who may mation of thrombi and limit hypoxic ischemic injury [125].
Treatmen 435
IV volume expansion should therefore be undertaken for below), the use of the monoclonal antibody eculizumab in
patients presenting with STEC-HUS, and should be consid- STEC-HUS is being further elucidated. As detailed previ-
ered as well for patients presenting with established HUS ously, greater understanding of the mechanism of STEC-
early in their illness. However, careful monitoring of urine HUS has implicated complement dysregulation in the
output and total fluid balance is paramount as fluid ther- pathogenesis.
apy will need frequent reassessment, particularly in Support for use of eculizumab in STEC-HUS was pro-
patients who progress to oligoanuria [126]. vided in a case series of three children with severe STEC-
HUS with neurologic involvement and dialysis-dependent
Antibiotics AKI. Administration of eculizumab was associated with
There have been mixed results in the literature as to a rapid clinical improvement in all three patients [134].
whether use of antibiotics in STEC illness increases the In 2011, during the E. coli O104 outbreak in Germany
risk of progression to HUS. The available evidence would and France, eculizumab was used in more than 250
suggest that providers should limit the use of antibiotics to patients. The patients treated with the drug had a more
patients in whom there is a suspicion of sepsis or severe severe renal injury and more frequent neurological com-
bacterial illness wherein the risks of potentially increasing plications at baseline compared to those who were not
likelihood of progression to HUS are outweighed by the treated, and outcomes were similar in both groups. This
benefits of infection control [126, 127]. may suggest a lack of drug effect or may be interpreted
as treatment benefit as the group treated with eculi-
zumab may have progressed toward worse out-
Management
comes [135]. Since then additional case series reported a
In addition to the supportive management of AKI detailed significant improvement in neurological function in
previously, other modalities have been investigated for children with STEC-HUS and neurological impairment
their therapeutic potential in STEC-HUS. treated with eculizumab [136, 137]. Of note, renal
impairment persisted in most of these patients [138].
Erythropoietin While data for eculizumab in STEC-HUS is limited to
Erythropoietin (EPO) has been investigated as a potential observational studies, it may be reasonable to consider
option to reduce the need for red blood cell transfusion in this drug for patients with severe disease, particularly
patients with HUS in small pediatric studies [128]. At this with neurologic involvement and when standard ther-
time there is insufficient evidence to recommend EPO as a apy has failed.
standard therapy for patients with HUS.
Shiga Toxin Neutralizing Agents
Plasma Infusion and Plasmapheresis In an attempt to reduce the pathogenicity of the Stx, oral
The American Society for Apheresis (ASFA) lists PE as a Stx-binding agents have been developed [139], though to
recommended therapy for STEC-HUS [129], though this date no such agents (either oral or intravenous) have been
treatment plan may not be any more efficacious than sup- proven efficacious in human studies and therefore toxin
portive care [130–132]. binding agents are not a standard component of STEC
diarrheal illness management [121, 140].
Dialysis
With progression to anuric renal failure, approximately
Controversy
two-thirds of patients with STEC-HUS ultimately require
renal replacement therapy [121]. Both hemodialysis and While supportive management and AKI therapy are well
peritoneal dialysis may be used, and evidence for superior established in STEC-HUS, the questions surrounding early
outcomes with one modality versus the other is limited. therapy remain. It appears reasonable to trial volume
Renal transplantation is rarely required in STEC HUS, expansion early in the illness course, with careful monitor-
though when necessary the risk of disease recurrence is ing. The role of antibiotics has conflicting evidence, though
minimal [133]. weighing risks and benefits, avoidance in patients with a
low risk of sepsis is favored. Eculizumab as a therapy in
STEC-HUS is currently under investigation in larger stud-
Emerging Therapy and Future Perspectives
ies. Until these and future analyses can provide more infor-
Eculizumab mation, we would consider the use of eculizumab in
While terminal complement blockade is the mainstay of patients with severe STEC-HUS who are not responding to
therapy for complement-mediated TMA (as described conventional therapy.
Complement-Mediated TMA (aHUS) Dialysis

Plasma Therapy As with STEC-HUS, many patients with complement-
Therapeutic PE has previously been recommended as early mediated TMA may require dialysis [142]. Patients may be
therapy for complement-mediated TMA [141]. Further treated with either hemodialysis or peritoneal dialysis, and
analysis of outcomes has revealed that there is limited effi- to date there is insufficient evidence to recommend on
cacy only on renal prognosis in patients treated with modality over the others. Prospective studies of treatment
plasma therapy compared to those who were not [142]. In with eculizumab suggest that it may decrease dialysis
the era of complement blockade, the importance of plasma dependence [118].
therapy has diminished.
Kidney Transplantation
Eculizumab Before the availability of C5-blockade, post-transplant
For patients with complement-mediated TMA, terminal recurrence of complement-mediated TMA was 60%, with
complement blockade is the therapeutic option of choice. 5-year graft survival of 30% in those patients [147].
Eculizumab is a humanized monoclonal antibody that For individuals with complement-mediated TMA sec-
binds the complement protein C5 and prevents production ondary to a mutation in the gene for a circulating comple-
of the proinflammatory anaphylatoxin C5a and the prod- ment regulators (such as CFH, CFI, CFB, or C3), liver
uct of the terminal complement cascade C5b-9 [118]. transplantation has been used as a means of providing a
In two prospective phase two trials, eculizumab inhib- new source of circulating factor production. Liver-kidney
ited TMA in patients with complement-mediated TMA and transplant has been used as a viable option to maintain
resulted in a significant improvement in renal func- graft function before the advent of terminal complement
tion [118]. A follow-up analysis demonstrated that these blockade therapy [147–151].
benefits were maintained after 2 years, with no new safety In patients with complement-mediated TMA at high
concerns surrounding the therapy [117, 143]. The safety of risk of disease recurrence post-transplant and for those
eculizumab has been demonstrated in further analyses, who experience recurrence, eculizumab has also been
wherein the most common adverse effects attributed to the shown to induce remission and graft function recov-
drug were fever, and mild gastrointestinal and respiratory ery [147, 152–154].
symptoms. No deaths of meningococcal infections were In an observational study of 13 renal transplant recipi-
noted as patients were systematically vaccinated (Neisseria ents who were given eculizumab for post-transplant
meningitidis strains A, C, W, Y, and B, S. pneumoniae, recurrence of complement-mediated TMA, complete
Haemophilus influenzae) and subjected to long-term anti- reversal of the disease was seen in all patients. Of note, a
biotic prophylaxis [14, 144]. delay in initiation of anti-C5 therapy was associated with
Eculizumab, when available, should be initiated within a low degree of renal functional improvement [152].
24–48 hours of onset of disease or admission. In children Eculizumab therapy is promising for both prevention and
who have been initiated on plasma therapy as a first-line cure of recurrent complement-mediated TMA in renal
treatment, a switch to eculizumab should be done as soon transplant recipients [155].
as the diagnosis of complement-mediated TMA is made
[14, 114]. In autoimmune complement-mediated TMA (e.g. Treatment Duration
CFH) eculizumab is successful in stabilizing patients in the The optimal length of therapy for complement-mediated
acute phase, though chronic management should target TMA has yet to be determined. Published literature sug-
suppression of disease-causing antibodies. When success- gests that different underlying mutations may warrant
ful, eculizumab may be discontinued in these patients, who individualized regimens, as the recurrence risk of the dis-
may require long-term immunosuppression [5, 145, 146]. ease may vary with the genetic defect [120]. Tools for moni-
Monitoring of complement activity or eculizumab drug toring complement activity to help guide eculizumab
levels may be helpful between doses of the drug, as dose dosing frequency have had some success in smaller
adjustments may be considered if there are signs of insuf- cohorts [156] though no large-scale validation of such tools
ficient blockade. In the long-term management of patients has yet been completed. Strategies toward restrictive use of
on eculizumab, routine complement activity monitoring eculizumab continue to be studied. Whereas lifelong ther-
may be unnecessary unless there are concerns for clinical apy had previously been advocated, it does appear that
status changes (such as infection, trauma, or other triggers) withdrawal of the drug, followed by close monitoring for
or relapse [125]. Total complement activity (CH50) may be signs of disease recurrence (e.g. hematuria, proteinuria,
followed. Of note, the normal range of CH50 may vary and hypertension) may be appropriate [157], and prospec-
depending on individual laboratory assays [5]. tive studies are needed to provide more insight in this area.
Reference 437
Controversy independent of the rest of this spectrum of diseases as it

has a defined trigger and known secondary cause of micro-
While the management of complement-mediated TMA has
angiopathy. However, it has more recently been discov-
been revolutionized with terminal complement blockade,
ered that much like in other TMA, complement
there are many questions that remain. Most notable is the
dysregulation may play a prominent role in the pathophys-
appropriate duration of therapy with eculizumab. At this
iology as well.
time more data are needed on individual patient risk pro-
In complement-mediated TMA the importance of the
files, complete with genotype-phenotype correlation.
complement cascade in disease progression has led to a
Efforts are ongoing in this field to establish optimal treat-
revolution in therapy with the advent of eculizumab, a
ment regimens.
form of terminal pathway blockade. Since the introduc-
tion of this drug, disease remission has become attainable
C
onclusion and mortality has improved considerably. Understanding
the role of complement in TMA has led to the use of ecu-
TMAs represent a disease spectrum for which our under- lizumab in select cases of STEC-HUS as well, and its
standing has increased dramatically over the past dec- role in the management of this entity continues to be
ade. What was once defined as a TTP/HUS spectrum has elucidated.
now expanded into multiple disease processes with many While complement pathway blockade has offered the
clinically similarities and distinct features. An increasing prospect of improved outcomes, questions remain to be
body of literature supports the role of complement dys- answered regarding the optimal dosing or duration of ther-
regulation as one of the disease-causing mechanisms in apy. In STEC HUS, the notion of preventative management
the TMA spectrum. Common to the TMA disease entities or neutralizing toxins has thus far not shown any benefit in
are the triad of MAHA, thrombocytopenia, and AKI, and clinical studies, though investigations are ongoing. In
the severity of illness can range from mild disease complement-mediated TMA, although many mutations
to severe disease with prolonged sequelae and high have been defined, others continue to be discovered. The
mortality. significance and therapeutic potential of these mutations is
HUS is an entity under the umbrella of TMA. The most still unknown and further study in genetics will hopefully
common cause, STEC-HUS, has long been considered as help to inform future clinical trials in this area.
R
eferences
Moake, J.L. (2002). Thrombotic microangiopathies. N.

1 8 Noris, M. and Remuzzi, G. (2009). Atypical hemolytic-
Engl. J. Med. 347 (8): 589–600. uremic syndrome. N. Engl. J. Med. 361 (17):
2 Brocklebank, V., Wood, K.M., and Kavanagh, D. (2018). 1676–1687.
Thrombotic microangiopathy and the kidney. Clin. J. Am. 9 Joly, B.S., Coppo, P., and Veyradier, A. (2018). Pediatric
Soc. Nephrol. 13 (2): 300–317. thrombotic thrombocytopenic purpura. Eur. J. Haematol.
3 George, J.N. and Nester, C.M. (2014). Syndromes of 101 (4): 425–434.
thrombotic microangiopathy. N. Engl. J. Med. 371 (7): 10 Tsai, H.-M. (2003). Advances in the pathogenesis,
654–666. diagnosis, and treatment of thrombotic thrombocytopenic
4 Fakhouri, F., Zuber, J., Frémeaux-Bacchi, V., and Loirat, C. purpura. J. Am. Soc. Nephrol. 14 (4): 1072–1081.
(2017). Haemolytic uraemic syndrome. Lancet 390 (10095): 11 Tarr, P.I., Gordon, C.A., and Chandler, W.L. (2005).
681–696. Shiga-toxin-producing Escherichia coli and haemolytic
5 Goodship, T.H.J., Cook, H.T., Fakhouri, F. et al. (2017). uraemic syndrome. Lancet 365 (9464): 1073–1086.
Atypical hemolytic uremic syndrome and C3 12 Gasser, C., Gautier, E., Steck, A. et al. (1955). Hemolytic-
glomerulopathy: conclusions from a “Kidney Disease: uremic syndrome: bilateral necrosis of the renal cortex in
Improving Global Outcomes” (KDIGO) controversies acute acquired hemolytic anemia. Schweiz. Med.
conference. Kidney Int. 91 (3): 539–551. Wochenschr. 85 (38–39): 905–909.
6 Nester, C.M., Barbour, T., de Cordoba, S.R. et al. (2015). 13 Barnard, P.J. and Kibel, M. (1965). The haemolytic-
Atypical aHUS: state of the art. Mol. Immunol. 67 (1): uraemic syndrome of infancy and childhood. A report of
31–42. eleven cases. Cent. Afr. J. Med. 11: 31–34.
7 Noris, M. and Remuzzi, G. (2005). Hemolytic uremic 14 Loirat, C., Fakhouri, F., Ariceta, G. et al. (2016). An
syndrome. J. Am. Soc. Nephrol. 16 (4): 1035–1050. international consensus approach to the management of
atypical hemolytic uremic syndrome in children. Pediatr. 29 El Karoui, K., Hill, G.S., Karras, A. et al. (2012). A
Nephrol. 31 (1): 15–39. clinicopathologic study of thrombotic microangiopathy in
15 Chen, M., Zhuang, J., Yang, J. et al. (2017). Atypical IgA nephropathy. J. Am. Soc. Nephrol. 23 (1): 137–148.
hemolytic uremic syndrome induced by CblC subtype of 30 Chen, S.-F., Wang, H., Huang, Y.-M. et al. (2015).
methylmalonic academia: a case report and literature Clinicopathologic characteristics and outcomes of renal
review. Medicine 96 (43): e8284. thrombotic microangiopathy in anti-neutrophil
16 George, J.N. (2015). Cobalamin C deficiency-associated cytoplasmic autoantibody-associated glomerulonephritis.
thrombotic microangiopathy: uncommon or Clin. J. Am. Soc. Nephrol. 10 (5): 750–758.
unrecognised? Lancet 386 (9997): 1012. 31 Wanchoo, R., Bayer, R.L., Bassil, C., and Jhaveri, K.D.
17 Lemaire, M., Frémeaux-Bacchi, V., Schaefer, F. et al. (2018). Emerging concepts in hematopoietic stem cell
(2013). Recessive mutations in DGKE cause atypical transplantation-associated renal thrombotic
hemolytic-uremic syndrome. Nat. Genet. 45 (5): 531–536. microangiopathy and prospects for new treatments. Am.
18 Westland, R., Bodria, M., Carrea, A. et al. (2014). J. Kidney Dis. 72 (6): 857–865.
Phenotypic expansion of DGKE-associated diseases. J. 32 Khosla, J., Yeh, A.C., Spitzer, T.R., and Dey, B.R. (2018).
Am. Soc. Nephrol. 25 (7): 1408–1414. Hematopoietic stem cell transplant-associated thrombotic
19 Azukaitis, K., Simkova, E., Majid, M.A. et al. (2017). The microangiopathy: current paradigm and novel therapies.
phenotypic spectrum of nephropathies associated with Bone Marrow Transplant. 53 (2): 129–137.
mutations in diacylglycerol kinase ε. J. Am. Soc. Nephrol. 33 Changsirikulchai, S., Myerson, D., Guthrie, K.A. et al.
28 (10): 3066–3075. (2009). Renal thrombotic microangiopathy after
20 Challis, R.C., Ring, T., Xu, Y. et al. (2017). Thrombotic hematopoietic cell transplant: role of GVHD in
microangiopathy in inverted Formin 2-mediated renal pathogenesis. Clin. J. Am. Soc. Nephrol. 4 (2): 345–353.
disease. J. Am. Soc. Nephrol. 28 (4): 1084–1091. 34 Jodele, S., Licht, C., Goebel, J. et al. (2013). Abnormalities
21 Bitzan, M. and Zieg, J. (2017). Influenza-associated in the alternative pathway of complement in children
thrombotic microangiopathies. Pediatr. Nephrol. with hematopoietic stem cell transplant-associated
22 Saab, K.R., Elhadad, S., Copertino, D., and Laurence, J. thrombotic microangiopathy. Blood 122 (12): 2003–2007.
(2016). Thrombotic microangiopathy in the setting of 35 Devresse, A., de Meyer, M., Aydin, S. et al. (2018). De
HIV infection: a case report and review of the differential novo atypical haemolytic uremic syndrome after kidney
diagnosis and therapy. AIDS Patient Care STDs 30 (8): transplantation. Case Rep. Nephrol. 2018: 1727986.
359–364. 36 Java, A., Edwards, A., Rossi, A. et al. (2015).
23 Copelovitch, L. and Kaplan, B.S. (2008). Streptococcus Cytomegalovirus-induced thrombotic microangiopathy
pneumoniae-associated hemolytic uremic syndrome. after renal transplant successfully treated with
Pediatr. Nephrol. 23 (11): 1951–1956. eculizumab: case report and review of the literature.
24 Spinale, J.M., Ruebner, R.L., Kaplan, B.S., and Transpl. Int. 28 (9): 1121–1125.
Copelovitch, L. (2013). Update on Streptococcus 37 George, J.N., Morton, J.M., Liles, N.W., and Nester, C.M.
pneumoniae associated hemolytic uremic syndrome. (2017). After the party’s over. N. Engl. J. Med. 376 (1):
Curr. Opin. Pediatr. 25 (2): 203–208. 74–80.
25 de Holanda, M.I., Pôrto, L.C., Wagner, T. et al. (2017). Use 38 Al-Nouri, Z.L., Reese, J.A., Terrell, D.R. et al. (2015).
of eculizumab in a systemic lupus erythemathosus patient Drug-induced thrombotic microangiopathy: a systematic
presenting thrombotic microangiopathy and heterozygous review of published reports. Blood 125 (4): 616–618.
deletion in CFHR1-CFHR3. A case report and systematic 39 Weitz, I.C. (2018). Thrombotic microangiopathy in
review. Clin. Rheumatol. 36 (12): 2859–2867. cancer. Thromb. Res. 164 (Suppl 1): S103–S105.
26 Park, M.H., Caselman, N., Ulmer, S., and Weitz, I.C. 40 Izzedine, H. and Perazella, M.A. (2015). Thrombotic
(2018). Complement-mediated thrombotic microangiopathy, cancer, and cancer drugs. Am. J. Kidney
microangiopathy associated with lupus nephritis. Blood Dis. 66 (5): 857–868.
Adv. 2 (16): 2090–2094. 41 Timmermans, S.A.M.E.G., Abdul-Hamid, M.A.,
27 Woodworth, T.G., Suliman, Y.A., Li, W. et al. (2016). Vanderlocht, J. et al. (2017). Patients with hypertension-
Scleroderma renal crisis and renal involvement in associated thrombotic microangiopathy may present with
systemic sclerosis. Nat. Rev. Nephrol. 12 (11): 678–691. complement abnormalities. Kidney Int. 91 (6): 1420–1425.
28 Rodríguez-Pintó, I., Moitinho, M., Santacreu, I. et al. 42 Van Laecke, S. and Van Biesen, W. (2017). Severe
(2016). Catastrophic antiphospholipid syndrome (CAPS): hypertension with renal thrombotic microangiopathy:
descriptive analysis of 500 patients from the international what happened to the usual suspect? Kidney Int. 91 (6):
CAPS registry. Autoimmun. Rev. 15 (12): 1120–1124. 1271–1274.
Reference 439
3 Shavit, L., Reinus, C., and Slotki, I. (2010). Severe renal

4 58 Fremeaux-Bacchi, V., Fakhouri, F., Garnier, A. et al.
failure and microangiopathic hemolysis induced by (2013). Genetics and outcome of atypical hemolytic
malignant hypertension – case series and review of uremic syndrome: a nationwide French series comparing
literature. Clin. Nephrol. 73 (2): 147–152. children and adults. Clin. J. Am. Soc. Nephrol. 8 (4):
44 Timmermans, S.A.M.E.G., Abdul-Hamid, M.A., 554–562.
Potjewijd, J. et al. (2018). C5b9 formation on endothelial 59 Bu, F., Maga, T., Meyer, N.C. et al. (2014). Comprehensive
cells reflects complement defects among patients with genetic analysis of complement and coagulation genes in
renal thrombotic microangiopathy and severe atypical hemolytic uremic syndrome. J. Am. Soc. Nephrol.
hypertension. J. Am. Soc. Nephrol. 29 (8): 2234–2243. 25 (1): 55–64.
45 Huerta, A., Arjona, E., Portoles, J. et al. (2018). A 60 Noris, M., Caprioli, J., Bresin, E. et al. (2010). Relative
retrospective study of pregnancy-associated atypical role of genetic complement abnormalities in sporadic and
hemolytic uremic syndrome. Kidney Int. 93 (2): 450–459. familial aHUS and their impact on clinical phenotype.
46 Bruel, A., Kavanagh, D., Noris, M. et al. (2017). Hemolytic Clin. J. Am. Soc. Nephrol. 5 (10): 1844–1859.
uremic syndrome in pregnancy and postpartum. Clin. J. 61 Schaefer, F., Ardissino, G., Ariceta, G. et al. (2018).
Am. Soc. Nephrol. 12 (8): 1237–1247. Clinical and genetic predictors of atypical hemolytic
47 Fakhouri, F., Vercel, C., and Frémeaux-Bacchi, V. (2012). uremic syndrome phenotype and outcome. Kidney Int. 94
Obstetric nephrology: AKI and thrombotic (2): 408–418.
microangiopathies in pregnancy. Clin. J. Am. Soc. 62 Fremeaux-Bacchi, V., Moulton, E.A., Kavanagh, D. et al.
Nephrol. 7 (12): 2100–2106. (2006). Genetic and functional analyses of membrane
48 Fakhouri, F., Roumenina, L., Provot, F. et al. (2010). cofactor protein (CD46) mutations in atypical hemolytic
Pregnancy-associated hemolytic uremic syndrome uremic syndrome. J. Am. Soc. Nephrol. 17 (7): 2017–2025.
revisited in the era of complement gene mutations. J. Am. 63 Bu, F., Borsa, N.G., Jones, M.B. et al. (2016). High-
Soc. Nephrol. 21 (5): 859–867. throughput genetic testing for thrombotic
49 Jodele, S., Fukuda, T., Vinks, A. et al. (2014). Eculizumab microangiopathies and C3 glomerulopathies. J. Am. Soc.
therapy in children with severe hematopoietic stem cell Nephrol. 27 (4): 1245–1253.
transplantation-associated thrombotic microangiopathy. 64 Valoti, E., Alberti, M., Tortajada, A. et al. (2015). A novel
Biol. Blood Marrow Transplant. 20 (4): 518–525. atypical hemolytic uremic syndrome-associated hybrid
50 Garg, N., Rennke, H.G., Pavlakis, M., and Zandi-Nejad, K. CFHR1/CFH gene encoding a fusion protein that
(2018). De novo thrombotic microangiopathy after kidney antagonizes factor H-dependent complement regulation.
transplantation. Transplant. Rev. 32 (1): 58–68. J. Am. Soc. Nephrol. 26 (1): 209–219.
51 Matsell, D.G. and White, C.T. (2009). An outbreak of 65 Bresin, E., Rurali, E., Caprioli, J. et al. (2013). Combined
diarrhea-associated childhood hemolytic uremic complement gene mutations in atypical hemolytic uremic
syndrome: the Walkerton epidemic. Kidney Int. Suppl. syndrome influence clinical phenotype. J. Am. Soc.
112: S35–S37. Nephrol. 24 (3): 475–486.
52 Frank, C., Werber, D., Cramer, J.P. et al. (2011). Epidemic 66 Chandler, W.L., Jelacic, S., Boster, D.R. et al. (2002).
profile of Shiga-toxin-producing Escherichia coli O104:H4 Prothrombotic coagulation abnormalities preceding the
outbreak in Germany. N. Engl. J. Med. 365 (19): 1771–1780. hemolytic-uremic syndrome. N. Engl. J. Med. 346 (1):
53 Bradley, K.K., Williams, J.M., Burnsed, L.J. et al. (2012). 23–32.
Epidemiology of a large restaurant-associated outbreak of 67 Villysson, A., Tontanahal, A., and Karpman, D. (2017).
Shiga toxin-producing Escherichia coli O111:NM. Microvesicle involvement in Shiga toxin-associated
Epidemiol. Infect. 140 (9): 1644–1654. infection. Toxins 9 (11): 19.
54 Kunzendorf, U., Karch, H., Werber, D., and Haller, H. 68 Ståhl, A., Arvidsson, I., Johansson, K.E. et al. (2015). A
(2011). Recent outbreak of hemolytic uremic syndrome in novel mechanism of bacterial toxin transfer within host
Germany. Kidney Int. 80 (9): 900–902. blood cell-derived microvesicles. PLoS Pathog. 11 (2):
55 Pennington, H. (2010). Escherichia coli O157. Lancet 376 e1004619.
(9750): 1428–1435. 69 Grabowski, E.F. (2002). The hemolytic-uremic
56 Jokiranta, T.S. (2017). HUS and atypical HUS. Blood syndrome – toxin, thrombin, and thrombosis. N. Engl. J.
25;129 (21): 2847–2856. Med. 346 (1): 58–61.
57 Sridharan, M., Go, R.S., and Willrich, M.A.V. (2018). 70 Desch, K. and Motto, D. (2007). Is there a shared
Atypical hemolytic uremic syndrome: review of clinical pathophysiology for thrombotic thrombocytopenic
presentation, diagnosis and management. J. Immunol. purpura and hemolytic-uremic syndrome? J. Am. Soc.
Methods 461: 15–22. Nephrol. 18 (9): 2457–2460.
71 Thurman, J.M., Marians, R., Emlen, W. et al. (2009). 84 Zipfel, P.F., Edey, M., Heinen, S. et al. (2007). Deletion of
Alternative pathway of complement in children with complement factor H-related genes CFHR1 and CFHR3
diarrhea-associated hemolytic uremic syndrome. Clin. J. is associated with atypical hemolytic uremic syndrome.
Am. Soc. Nephrol. 4 (12): 1920–1924. PLoS Genet. 3 (3): e41.
72 Karpman, D. and Tati, R. (2016). Complement contributes 85 Rougier, N., Kazatchkine, M.D., Rougier, J.P. et al. (1998).
to the pathogenesis of Shiga toxin-associated hemolytic Human complement factor H deficiency associated with
uremic syndrome. Kidney Int. 90 (4): 726–729. hemolytic uremic syndrome. J. Am. Soc. Nephrol. 9 (12):
73 Locatelli, M., Buelli, S., Pezzotta, A. et al. (2014). Shiga 2318–2326.
toxin promotes podocyte injury in experimental 86 Kavanagh, D., Pappworth, I.Y., Anderson, H. et al. (2012).
hemolytic uremic syndrome via activation of the Factor I autoantibodies in patients with atypical
alternative pathway of complement. J. Am. Soc. Nephrol. hemolytic uremic syndrome: disease-associated or an
25 (8): 1786–1798. epiphenomenon? Clin. J. Am. Soc. Nephrol. 7 (3):
74 Orth, D. and Würzner, R. (2010). Complement in typical 417–426.
hemolytic uremic syndrome. Semin. Thromb. Hemost. 36 87 Lhotta, K., Janecke, A.R., Scheiring, J. et al. (2009). A
(6): 620–624. large family with a gain-of-function mutation of
75 Ståhl, A., Sartz, L., and Karpman, D. (2011). Complement complement C3 predisposing to atypical hemolytic
activation on platelet-leukocyte complexes and uremic syndrome, microhematuria, hypertension and
microparticles in enterohemorrhagic Escherichia chronic renal failure. Clin. J. Am. Soc. Nephrol. 4 (8):
coli-induced hemolytic uremic syndrome. Blood 117 (20): 1356–1362.
5503–5513. 88 Kavanagh, D., Kemp, E.J., Mayland, E. et al. (2005).
76 Poolpol, K., Orth-Höller, D., Speth, C. et al. (2014). Mutations in complement factor I predispose to
Interaction of Shiga toxin 2 with complement regulators development of atypical hemolytic uremic syndrome. J.
of the factor H protein family. Mol. Immunol. 58 (1): Am. Soc. Nephrol. 16 (7): 2150–2155.
77–84. 89 Goicoechea de Jorge, E., Tortajada, A., García, S.P. et al.
77 Ozaki, M., Kang, Y., Tan, Y.S. et al. (2016). Human (2018). Factor H competitor generated by gene conversion
mannose-binding lectin inhibitor prevents shiga toxin- events associates with atypical hemolytic uremic
induced renal injury. Kidney Int. 90 (4): 774–782. syndrome. J. Am. Soc. Nephrol. 29 (1): 240–249.
78 Zipfel, P.F., Misselwitz, J., Licht, C., and Skerka, C. 90 Allen, U. and Licht, C. (2011). Pandemic H1N1 influenza
(2006). The role of defective complement control in a infection and (atypical) HUS – more than just another
hemolytic uremic syndrome. Semin. Thromb. Hemost. 32 trigger? Pediatr. Nephrol. 26 (1): 3–5.
(2): 146–154. 91 Dragon-Durey, M.-A., Loirat, C., Cloarec, S. et al. (2005).
79 Teoh, C.W., Riedl, M., and Licht, C. (2016). The Anti-factor H autoantibodies associated with atypical
alternative pathway of complement and the thrombotic hemolytic uremic syndrome. J. Am. Soc. Nephrol. 16 (2):
microangiopathies. Transfus. Apher. Sci. 54 (2): 555–563.
220–231. 92 Brocklebank, V., Johnson, S., Sheerin, T.P. et al. (2017).
80 Riedl, M., Fakhouri, F., Le Quintrec, M. et al. (2014). Factor H autoantibody is associated with atypical
Spectrum of complement-mediated thrombotic hemolytic uremic syndrome in children in the United
microangiopathies: pathogenetic insights identifying Kingdom and Ireland. Kidney Int. 92 (5): 1261–1271.
novel treatment approaches. Semin. Thromb. Hemost. 40 93 Zipfel, P.F., Mache, C., Müller, D. et al. (2010). DEAP-
(4): 444–464. HUS: deficiency of CFHR plasma proteins and
81 Demeulenaere, M., Devreese, K., Vanbelleghem, H. et al. autoantibody-positive form of hemolytic uremic
(2018). Thrombomodulin and endothelial dysfunction: a syndrome. Pediatr. Nephrol. 25 (10): 2009–2019.
disease-modifier shared between malignant hypertension 94 Hofer, J., Janecke, A.R., Zimmerhackl, L.B. et al. (2013).
and atypical hemolytic uremic syndrome. Nephron 140 Complement factor H-related protein 1 deficiency and
(1): 63–73. factor H antibodies in pediatric patients with atypical
82 Delvaeye, M., Noris, M., De Vriese, A. et al. (2009). hemolytic uremic syndrome. Clin. J. Am. Soc. Nephrol. 8
Thrombomodulin mutations in atypical hemolytic- (3): 407–415.
uremic syndrome. N. Engl. J. Med. 361 (4): 345–357. 95 Józsi, M., Heinen, S., Hartmann, A. et al. (2006). Factor H
83 Marinozzi, M.C., Vergoz, L., Rybkine, T. et al. (2014). and atypical hemolytic uremic syndrome: mutations in
Complement factor B mutations in atypical hemolytic the C-terminus cause structural changes and defective
uremic syndrome-disease-relevant or benign? J. Am. Soc. recognition functions. J. Am. Soc. Nephrol. 17 (1):
Nephrol. 25 (9): 2053–2065. 170–177.
Reference 441
96 Józsi, M., Licht, C., Strobel, S. et al. (2008). Factor H 109 Spizzirri, F.D., Rahman, R.C., Bibiloni, N. et al. (1997).
autoantibodies in atypical hemolytic uremic syndrome Childhood hemolytic uremic syndrome in Argentina:
correlate with CFHR1/CFHR3 deficiency. Blood 111 (3): long-term follow-up and prognostic features. Pediatr.
1512–1514. Nephrol. 11 (2): 156–160.
97 Lee, J.M., Park, Y.S., Lee, J.H. et al. (2015). Atypical 110 Renaud, C., Niaudet, P., Gagnadoux, M.F. et al. (1995).
hemolytic uremic syndrome: Korean pediatric series. Haemolytic uraemic syndrome: prognostic factors in
Pediatr. Int. 57 (3): 431–438. children over 3 years of age. Pediatr. Nephrol. 9 (1): 24–29.
98 Formeck, C. and Swiatecka-Urban, A. (2018). Extra- 111 Robson, L.M., Leung, A.K.C., and Brant, R. (1993).
renal manifestations of atypical hemolytic uremic Relationship of the recovery in the glomerular filtration
syndrome. Pediatr. Nephrol. rate to the duration of anuria in diarrhea-associated
99 Fidan, K., Göknar, N., Gülhan, B. et al. (2018). Extra- hemolytic uremic syndrome. Am. J. Nephrol. 13 (3):
renal manifestations of atypical hemolytic uremic 194–197.
syndrome in children. Pediatr. Nephrol. 33 (8): 112 Lopez, E.L., Devoto, S., Fayad, A. et al. (1992).
1395–1403. Association between severity of gastrointestinal
100 Nathanson, S., Kwon, T., Elmaleh, M. et al. (2010). prodrome and long-term prognosis in classic hemolytic-
Acute neurological involvement in diarrhea-associated uremic syndrome. J. Pediatr. 120 (2): 210–215.
hemolytic uremic syndrome. Clin. J. Am. Soc. Nephrol. 5 113 Walters, M.D.S., Matthei, I.U., Kay, R. et al. (1989). The
(7): 1218–1228. polymorphonuclear leucocyte count in childhood
101 Sawai, T., Nangaku, M., Ashida, A. et al. (2014). haemolytic uraemic syndrome. Pediatr. Nephrol. 3 (2):
Diagnostic criteria for atypical hemolytic uremic 130–134.
syndrome proposed by the Joint Committee of the 114 Loirat, C. and Frémeaux-Bacchi, V. (2011). Atypical
Japanese Society of Nephrology and the Japan Pediatric hemolytic uremic syndrome. Orphanet J. Rare Dis. 6
Society. Pediatr. Int. 56 (1): 1–5. (1): 60.
102 Wada, H., Matsumoto, T., Suzuki, K. et al. (2018). 115 Gianviti, A., Tozzi, A.E., De Petris, L. et al. (2003). Risk
Differences and similarities between disseminated factors for poor renal prognosis in children with
intravascular coagulation and thrombotic hemolytic uremic syndrome. Pediatr. Nephrol. 18 (12):
microangiopathy. Thromb. J. 16: 14. 1229–1235.
103 Clark, W.F., Patriquin, C., Licht, C. et al. (2017). Simple 116 Trachtman, H., Austin, C., Lewinski, M., and Stahl,
diagnosis and treatment algorithm for adult R.A.K. (2012). Renal and neurological involvement in
thrombotic microangiopathy. Transfus. Apher. Sci. 56 typical Shiga toxin-associated HUS. Nat. Rev. Nephrol. 8
(1): 50–51. (11): 658–669.
104 Angioi, A., Fervenza, F.C., Sethi, S. et al. (2016). 117 Licht, C., Greenbaum, L.A., Muus, P. et al. (2015).
Diagnosis of complement alternative pathway disorders. Efficacy and safety of eculizumab in atypical hemolytic
Kidney Int. 89 (2): 278–288. uremic syndrome from 2-year extensions of phase 2
105 Newell, D.G. and La Ragione, R.M. (2018). studies. Kidney Int. 87 (5): 1061–1073.
Enterohaemorrhagic and other Shiga toxin-producing 118 Legendre, C.M., Licht, C., Muus, P. et al. (2013). Terminal
Escherichia coli (STEC): where are we now regarding complement inhibitor eculizumab in atypical hemolytic–
diagnostics and control strategies? Transbound. Emerg. uremic syndrome. N. Engl. J. Med. 368 (23): 2169–2181.
Dis. 65 (Suppl 1): 49–71. 119 Licht, C., Muus, P., Legendre, C.M. et al. (2012).
106 Spinale, J.M., Ruebner, R.L., Copelovitch, L., and Eculizumab (ECU) safety and efficacy in atypical
Kaplan, B.S. (2013). Long-term outcomes of Shiga toxin hemolytic uremic syndrome (aHUS) patients with long
hemolytic uremic syndrome. Pediatr. Nephrol. 28 (11): disease duration and chronic kidney disease (CKD):
2097–2105. 2-year results. Blood 120 (21): 985–985.
107 Grisaru, S., Midgley, J.P., Hamiwka, L.A. et al. (2011). 120 Fremeaux-Bacchi, V., Fakhouri, F., Garnier, A. et al.
Diarrhea-associated hemolytic uremic syndrome in (2013). Genetics and outcome of atypical hemolytic
southern Alberta: a long-term single-centre experience. uremic syndrome: a Nationwide French series
Paediatr. Child Health 16 (6): 337–340. comparing children and adults. CJASN https://doi.
108 Alconcher, L.F., Coccia, P.A., del Suarez, A.C. et al. org/10.2215/CJN.047605120.
(2018). Hyponatremia: a new predictor of mortality in 121 Scheiring, J., Andreoli, S.P., and Zimmerhackl, L.B.
patients with Shiga toxin-producing Escherichia coli (2008). Treatment and outcome of Shiga-toxin-
hemolytic uremic syndrome. Pediatr. Nephrol. 33 (10): associated hemolytic uremic syndrome (HUS). Pediatr.
1791–1798. Nephrol. 23 (10): 1749.
122 Walsh, P.R. and Johnson, S. (2017). Treatment and 133 Loirat, C. and Niaudet, P. (2003). The risk of recurrence
management of children with haemolytic uraemic of hemolytic uremic syndrome after renal transplantation
syndrome. Arch. Dis. Child. https://doi.org/10.1136/ in children. Pediatr. Nephrol. 18 (11): 1095–1101.
archdischild-2016-311377. 134 Lapeyraque, A.-L., Malina, M., Fremeaux-Bacchi, V.
123 Hickey, C.A., Beattie, T.J., Cowieson, J. et al. (2011). et al. (2011). Eculizumab in severe Shiga-toxin–
Early volume expansion during diarrhea and relative associated HUS. N. Engl. J. Med. 364 (26): 2561–2563.
nephroprotection during subsequent hemolytic uremic 135 Fakhouri, F. and Loirat, C. (2018). Anticomplement
syndrome. Arch. Pediatr. Adolesc. Med. 165 (10): treatment in atypical and typical hemolytic uremic
884–889. syndrome. Semin. Hematol. 55 (3): 150–158.
124 Grisaru, S., Xie, J., Samuel, S. et al. (2017). Associations 136 Percheron, L., Gramada, R., Tellier, S. et al. (2018).
between hydration status, intravenous fluid Eculizumab treatment in severe pediatric STEC-HUS: a
administration, and outcomes of patients infected with multicenter retrospective study. Pediatr. Nephrol. 33 (8):
Shiga toxin–producing Escherichia coli: a systematic 1385–1394.
review and meta-analysis. JAMA Pediatr. 171 (1): 68–76. 137 Goldwater, P.N. and Bettelheim, K.A. (2012). Treatment
125 Ardissino, G., Tel, F., Possenti, I. et al. (2017). Early of enterohemorrhagic Escherichia coli (EHEC) infection
volume expansion improves the outcome of Shigatoxin- and hemolytic uremic syndrome (HUS). BMC Med. 10
associated hemolytic uremic syndrome. Data from the (1): 12.
North Italian HUS network. Pediatrics 140 138 Gitiaux, C., Krug, P., Grevent, D. et al. (2013). Brain
(1 MeetingAbstract): 16. magnetic resonance imaging pattern and outcome in
126 Freedman, S.B., Xie, J., Neufeld, M.S. et al. (2016). Shiga children with haemolytic-uraemic syndrome and
toxin–producing Escherichia coli infection, antibiotics, neurological impairment treated with eculizumab. Dev.
and risk of developing hemolytic uremic syndrome: a Med. Child Neurol. 55 (8): 758–765.
meta-analysis. Clin. Infect. Dis. 62 (10): 1251–1258. 139 Tremblay, J.M., Mukherjee, J., Leysath, C.E. et al.
127 Bielaszewska, M., Idelevich, E.A., Zhang, W. et al. (2013). A single VHH-based toxin neutralizing agent
(2012). Effects of antibiotics on Shiga toxin 2 production and an effector antibody protects mice against challenge
and bacteriophage induction by epidemic Escherichia with Shiga toxins 1 and 2. Infect. Immun. https://doi.
coli O104:H4 strain. Antimicrob. Agents Chemother. 56 org/10.1128/IAI.01033-13.
(6): 3277–3282. 140 Trachtman, H., Cnaan, A., Christen, E. et al. (2003).
128 Balestracci, A., Martin, S.M., Toledo, I. et al. (2015). Effect of an oral Shiga toxin–binding agent on diarrhea-
Early erythropoietin in post-diarrheal hemolytic uremic associated hemolytic uremic syndrome in children: a
syndrome: a case–control study. Pediatr. Nephrol. 30 (2): randomized controlled trial. JAMA 290 (10): 1337–1344.
339–344. 141 The European Paediatric Study Group for HUS, Ariceta,
129 Szczepiorkowski, Z.M., Winters, J.L., Bandarenko, N. G., Besbas, N. et al. (2009). Guideline for the investigation
et al. (2010). Guidelines on the use of therapeutic and initial therapy of diarrhea-negative hemolytic uremic
apheresis in clinical practice – evidence-based approach syndrome. Pediatr. Nephrol. 24 (4): 687–696.
from the apheresis applications committee of the 142 Johnson, S., Stojanovic, J., Ariceta, G. et al. (2014). An
American Society for Apheresis. J. Clin. Apher. 25 (3): audit analysis of a guideline for the investigation and
83–177. initial therapy of diarrhea negative (atypical) hemolytic
130 Kielstein, J.T., Beutel, G., Fleig, S. et al. (2012). Best uremic syndrome. Pediatr. Nephrol. 29 (10): 1967–1978.
supportive care and therapeutic plasma exchange with 143 Rathbone, J., Kaltenthaler, E., Richards, A. et al. (2013).
or without eculizumab in Shiga-toxin-producing E. coli A systematic review of eculizumab for atypical
O104:H4 induced haemolytic–uraemic syndrome: an haemolytic uraemic syndrome (aHUS). BMJ Open 3
analysis of the German STEC-HUS registry. Nephrol. (11): e003573.
Dial. Transplant. 27 (10): 3807–3815. 144 Greenbaum, L.A., Fila, M., Ardissino, G. et al. (2016).
131 Dundas, S., Murphy, J., Soutar, R. et al. (1999). Eculizumab is a safe and effective treatment in pediatric
Effectiveness of therapeutic plasma exchange in the patients with atypical hemolytic uremic syndrome.
1996 Lanarkshire Escherichia coli O157:H7 outbreak. Kidney Int. 89 (3): 701–711.
Lancet 354 (9187): 1327–1330. 145 Sinha, A., Gulati, A., Saini, S. et al. (2014). Prompt
132 Rizzoni, G., Claris-Appiani, A., Edefonti, A. et al. (1988). plasma exchanges and immunosuppressive treatment
Plasma infusion for hemolytic-uremic syndrome in improves the outcomes of anti-factor H autoantibody-
children: results of a multicenter controlled trial. J. associated hemolytic uremic syndrome in children.
Pediatr. 112 (2): 284–290. Kidney Int. 85 (5): 1151–1160.
Reference 443
146 Noone, D., Waters, A., Pluthero, F.G. et al. (2014). 152 Zuber, J., Quintrec, M.L., Krid, S. et al. (2012).
Successful treatment of DEAP-HUS with eculizumab. Eculizumab for atypical hemolytic uremic syndrome
Pediatr. Nephrol. 29 (5): 841–851. recurrence in renal transplantation. Am. J. Transplant.
147 Zuber, J., Le Quintrec, M., Morris, H. et al. (2013). 12 (12): 3337–3354.
Targeted strategies in the prevention and management 153 Durán, C.E., Blasco, M., Maduell, F., and Campistol,
of atypical HUS recurrence after kidney transplantation. J.M. (2012). Rescue therapy with eculizumab in a
Transplant. Rev. 27 (4): 117–125. transplant recipient with atypical haemolytic-uraemic
148 Saland, J. (2014). Liver–kidney transplantation to cure syndrome. Clin. Kidney J. 5 (1): 28–30.
atypical HUS: still an option post-eculizumab? Pediatr. 154 Larrea, C.F., Cofan, F., Oppenheimer, F. et al. (2010).
Nephrol. 29 (3): 329–332. Efficacy of eculizumab in the treatment of recurrent
149 Weitz, M., Amon, O., Bassler, D. et al. (2011). atypical hemolytic-uremic syndrome after renal
Prophylactic eculizumab prior to kidney transplantation transplantation. Transplantation 89 (7): 903.
for atypical hemolytic uremic syndrome. Pediatr. 155 Zuber, J., Quintrec, M.L., Sberro-Soussan, R. et al.
Nephrol. 26 (8): 1325. (2011). New insights into postrenal transplant hemolytic
150 Xie, L., Nester, C.M., Reed, A.I. et al. (2012). Tailored uremic syndrome. Nat. Rev. Nephrol. 7 (1): 23–35.
eculizumab therapy in the management of complement 156 Ardissino, G., Tel, F., Sgarbanti, M. et al. (2018).
factor H–mediated atypical hemolytic uremic syndrome Complement functional tests for monitoring eculizumab
in an adult kidney transplant recipient: a case report. treatment in patients with atypical hemolytic uremic
Transplant. Proc. 44 (10): 3037–3040. syndrome: an update. Pediatr. Nephrol. 33 (3): 457–461.
151 Nester, C., Stewart, Z., Myers, D. et al. (2011). 157 Wijnsma, K.L., Duineveld, C., Wetzels, J.F.M., and van
Pre-emptive eculizumab and plasmapheresis for de Kar, N.C.A.J. (2018). Eculizumab in atypical
renal transplant in atypical hemolytic uremic hemolytic uremic syndrome: strategies toward
syndrome. CJASN https://doi.org/10.2215/ restrictive use. Pebdiatr. Nephrol. Available from: http://
CJN.10181110. link.springer.com/10.1007/s00467-018-4091-3.
444
27
Pregnancy
Georgina Irish1,2 and Shilpanjali Jesudason1,2
1
Central and Northern Adelaide Renal and Transplantation Service, Royal Adelaide Hospital, Adelaide, South Australia, Australia
2
Department of Medicine, University of Adelaide, Adelaide, South Australia, Australia
I ntroduction chronic hypertension. Over time, there has been growing

confidence in supporting pregnancies in women with kid-
Major physiological adaptive changes to renal hemody- ney disease, even those receiving dialysis, and for women
namics, and glomerular and tubular function occurs with after kidney transplantation. It is no longer routine practice
pregnancy. Normal physiological adaptations leading to to advise against pregnancy in these women.
hyperfiltration in pregnancy are not pathogenic in women Parenthood remains an important goal, but the desire for
with normal kidneys, but may unmask underlying chronic motherhood is balanced with fears of compromising their
kidney disease (CKD) for the first time or contribute to own and their child’s health. Approaches to care involve
worsening of existing CKD. This may manifest as tempo- understanding the patient’s experiences and values whilst
rary or permanent loss of renal function, and adverse preg- acknowledging medical risks. Key management points
nancy outcomes. Therefore pregnancy is an important relate to meticulous preconception planning, informed
challenge to the maternal kidney. shared decision-making about the options, and timing and
This chapter primarily discusses pregnancy in women management of pregnancy based on individualized assess-
with CKD at all stages. Clinical decision-making in this ment of risks. The involvement of multidisciplinary expert
high-risk cohort is focused mainly on risk assessment for teams, and careful maternal and fetal surveillance through-
maternal and fetal outcomes, with very limited evidence out pregnancy are essential.
regarding interventions. Appraisal of the literature is
hampered by the quality of existing data, and there are
large knowledge gaps regarding best clinical practice.
E
pidemiology
Observational studies of small, often heterogeneous, and
Chronic Kidney Disease
usually retrospectively studied cohorts may not be readily
compared. Definitions of CKD differ between studies and Defining CKD in pregnant women is challenging, especially
over time, with the widespread introduction of CKD stag- where preconception kidney function is unknown – this is
ing highlighting the outcomes of pregnancy with less common given that assessment of kidney function is not rou-
severe CKD. Registry and population cohort studies have tine in pregnancy. Robust prevalence data is lacking for many
illuminated outcomes but often lack detailed clinical data populations, with significant heterogeneity between studies
for more complex analysis of risks and outcomes. that often have small numbers. Definitions vary, data is often
Maternal (obstetric and renal) and perinatal outcomes historical, and clinical care practices may differ between cent-
worsen as preconception estimated glomerular filtration ers and change over time. In population pregnancy datasets,
rate (eGFR) declines. Hypertensive disorders of pregnancy kidney data is poorly captured, whereas renal registry datasets
often drive premature delivery and thereby define fetal out- lack detailed pregnancy data. In all cohorts, early pregnancy
comes. Management approaches are usually extrapolated loss is difficult to capture and likely underestimated. These
from other cohorts at risk of preeclampsia or those with factors limit comparison and risk quantification [1], and
Pathophysiolog 445
should be considered when interpreting studies and when and the literature is highly variable. AKI diagnosis based
synthesizing data from different studies. on rise in serum creatinine (absolute value >90 μmol/l,
The largest study identifying CKD in pregnancy was a doubling of serum creatinine) is likely more common but
Norwegian population-based data linkage study of 5655 poorly captured in studies. In the developed world, AKI is
births from 3405 women, assessing eGFR by the Modification uncommon and predominantly due to volume loss or hyper-
of Diet in Renal Disease (MDRD) formula. This study found tensive disorders of pregnancy, and less commonly de novo
CKD in 3.3% of pregnancies, with 2.4% with stage 1, 0.8% renal disorders. AKI requiring acute dialysis in pregnancy is
with stage 2, and 0.1% with stage 3 [2]. This data may be rare, occurring in 1 per 10 000 (95% confidence interval [CI]
extrapolated to similar countries, but the true prevalence of 0.8–1.1) [15]. By contrast, in developing nations AKI is
CKD in most pregnant cohorts remains unknown. mainly due to sepsis and rates are higher but variable [16].
Dialysis
P
athophysiology
The number of pregnancies in patients receiving dialysis
reported in the literature is increasing over time, although they
Physiological Changes with Pregnancy
remain rare [3, 4]. Absolute pregnancy rates are difficult to
define and compare. The Australian and New Zealand Dialysis The kidney undergoes marked physiological and structural
and Transplant Registry (ANZDATA) has reported the preg- changes to adapt to the demands of pregnancy, occurring
nancy rate in chronic hemodialysis patients was 8.4/1000 as early as 6 weeks from conception and continuing post-
patient-years between 2001 and 2011, with a live birth rate of partum (summarized extensively in the paper by Odutayo
1.26/1000 patient-years [5]. Comprehensive survey data from et al. [17]). These physiological changes are essential adap-
Belgium reported a conception rate of 0.3/1000 patient-years, tions to maintain maternal and fetal health, but previously
and a study from Japan showed 3.4% of women of childbear- unidentified kidney disease may be exposed and underly-
ing age conceived [6–8]. An Italian study found that a live birth ing renal dysfunction may worsen. Systemic hemodynamic
rate was 0.7–1.1 per 1000 women per year compared to the changes include vasodilatation mediated by hormones
national rate of 72.5 per 1000 women per year [9]. These data including relaxin and nitric oxide, decreased systemic vas-
suggest that the birth rate in patients on peritoneal dialysis cular resistance and increased cardiac output [18]. A nadir
(PD) is lower than that in hemodialysis recipients [3–5]. The in the mean arterial pressure occurs by 18–24 weeks, with
reasons for this are uncertain, but may be due to hypertonic subsequent rise in later gestation. Concomitantly, the
fluid disrupting implantation or causing adhesions [10], or renin-angiotensin system is activated in pregnancy. Renal
change in dialysis modality to prior to conception [5]. tubular changes occur due to complex shifts in hormone
responsiveness, with a net increase in total body sodium
Transplantation and potassium, resetting of receptor thresholds for antidiu-
retic hormone release, and fall in plasma osmolality [17].
Since the first successful birth post renal transplantation in All of these factors contribute to expanded plasma vol-
1958, the rate of conception and numbers of live births post ume, increased renal plasma flow (by 50–80%), and increased
kidney transplantation have continued to increase [11]. The glomerular hyperfiltration (by 50%) but without glomerular
ANZDATA registry has reported a rate of 20/1000 patient years hypertension [19]. It is important to note that many physio-
from 1996 to 2005 with a live birth rate of 16/1000 patient- logical studies of pregnancy are historical and lacking con-
years [4, 11]. The United States Renal Data System registry temporaneous data exploring hyperfiltration in mothers
reported a pregnancy rate of 33/1000 patient-years [12]. The who are older or have comorbidities of diabetes and obesity.
UK Obstetric Surveillance Study recognized all pregnancies in Structural changes of the renal tract also occur, with
renal transplant recipients, with a rate of 7.6/1000 patient- increased kidney size due to expansion of intravascular
years [13]. Italian data revealed a birth rate of 5.5–8.3 per and interstitial volume. Ureteral and pelvic dilatation is
1000 women [14]. The impact of widespread use of mycophe- commonly seen, may mimic obstruction, and predisposes
nolate mofetil immunosuppression on pregnancy rates in to urinary stasis and infection.
patients with transplant or glomerulonephritis in the recent
era of care is yet to be determined.
Fertility and CKD
Acute Kidney Injury Fertility is affected by CKD and declines as renal function
The epidemiology of acute kidney injury (AKI) in pregnancy worsens. Mechanisms include impaired ovulation due to hor-
depends on the geographical location and definition used, monal dysregulation with hyperprolactinaemia and decrease
446 Pregnancy
in gonadotropin-releasing and luteinizing hormone pulsatil- the common features of hypertension, proteinuria, and
ity, decreased ovarian reserve and decrease in libido [20–22]. progressive renal dysfunction [34]. Superimposed preec-
Women may also be discouraged from conceiving due to con- lampsia with inpatients with underlying CKD contributes
cerns for their health [23]. Intensive dialysis may increase the to worse outcomes of preterm delivery and small for gesta-
ability for conception [24], but transplantation remains the tional age. Parameters such as liver function tests and
main mechanism for restoring fertility. platelet count can assist with differentiation. Systemic
lupus erythematosus (SLE) nephritis flares can be difficult
to separate from preeclampsia; systemic features of SLE
D
iagnosis
and low complement levels, and elevated double-stranded
DNA can aid differentiation [35, 36]. Doppler ultrasound
Markers of Kidney Function in Pregnancy
of the umbilical artery to assess changes in waveform and
In normal pregnancy, GFR increment results in greater cre- vascular resistance can be useful. Whilst only small studies
atinine clearance and fall in serum creatinine by the second have been performed, results suggest a high specificity for
trimester. An accurate estimation of GFR is consequently dif- predicting preeclampsia [34, 37].
ficult, with reliance on serum creatinine alone. The gold Novel biomarkers including soluble fms-like tyrosine
standard 24 hour urine collection for creatinine or inulin is kinase (sFlt), placental growth factor (PIGF), and serum
impractical for clinical use. Commonly used formulae to esti- placental protein 12 may help in differentiating and pre-
mate GFR are not validated in pregnancy. The Cockcroft– dicting preeclampsia in women with CKD [34, 38]. A meta-
Gault formula underestimates GFR compared to inulin analysis reviewed use of the sFLT:PIGF ratio as a predictive
clearance and overestimates compared to 24-hour urine cre- and diagnostic marker of preeclampsia, and found in both
atinine clearance. The MDRD and CKD – EPI formulae low and high risk patients this could predict development
underestimate GFR [25–27]. The international classifications of preeclampsia with a sensitivity of 80%, specificity of
of AKI are also not validated in the pregnant population. 92%, positive likelihood ratio of 10.5, and negative likeli-
Urinary protein excretion increases during pregnancy and hood ratio of 0.22 [39]. Larger cohort studies are needed to
can rise until delivery; proteinuria up to 300 mg/24 h may be clarify the diagnostic benefit of these markers.
observed in normal pregnancy [28]. There are multiple con-
tributory mechanisms, including physiological hyperfiltra-
tion, reduced tubular protein reabsorption, and increase in Kidney Biopsy
glomerular basement membrane pore size. Whilst 24-hour The safety of kidney biopsy in pregnancy is contentious. A
collection is still considered gold standard, the spot systematic review of 197 biopsies showed a complication rate
protein:creatinine ratio can be used instead for assessing sig- of 7% with the highest risk period being after 25 weeks gesta-
nificant proteinuria above 30 mg/mmol and is also validated tion [40]. It is rarely indicated for differentiation between
in hypertensive women [29]. preeclampsia and primary renal pathology. Biopsy should be
undertaken cautiously, preferably in early pregnancy, and
Diagnosing Pregnancy in Patients with CKD only when clinical management is likely to substantially alter
Renal excretion of β-human chorionic gonadotropin (β- based on biopsy results. Otherwise it should be deferred until
HCG) is reduced in renal impairment and thus urine after delivery. Postpartum kidney biopsy to investigate abnor-
pregnancy tests can be falsely positive [30]. Irregular mal renal function or proteinuria commonly identifies path-
menstruation is often observed in advanced CKD, and ological glomerular disease [41]. Preconception kidney
thus serial β-HCG and ultrasound is a better indicator of biopsy may be very informative in assisting with pregnancy
gestational age than the date of the last menstrual planning and risk stratification, especially for transplanted
cycle [30]. β-HCG is used in antenatal screening for ane- women and those with SLE nephritis.
uploidies, which can increase false positives and should
be interpreted with caution [31, 32]. Pregnancy-associated
plasma protein A (PAPP-A) used in first trimester screen- P
rognosis
ing panels is elevated in patients receiving heparin and
unpredictable in late-stage CKD [32, 33]. Impact of Maternal Kidney Disease
on Pregnancy Outcomes
Diagnosing Preeclampsia in Patients with CKD
There is a substantial body of evidence from cohort and
Differentiating superimposed preeclampsia from worsen- population studies that both maternal and fetal outcomes
ing CKD due to pregnancy is a diagnostic challenge due to are worse with any degree of renal dysfunction compared
Prognosi 447
to women without CKD, and deteriorate along the con- Primary Disease and Pregnancy Outcomes
tinuum of CKD severity [14, 42–46]. The key adverse
The impact of primary renal disease varies depending on
maternal pregnancy complication is worsening hyperten-
the specific disease and the severity of kidney dysfunction
sion or preeclampsia, which likely drives preterm delivery
(which may not always be reported or comparable between
and high cesarean section rates. There is increased risk of
studies). The most robust data exists for IgA nephropathy,
premature birth, low birth weight, small for gestational
where the preconception kidney function, proteinuria, and
age, neonatal intensive care unit admissions, and infant
hypertension are the key prognostic factors [53, 57]. There
death [3]. Risks are compounded by comorbidities, par-
is little data on other glomerular diseases and pregnancy.
ticularly chronic hypertension and proteinuria (>1 g/day),
An exception is SLE nephritis, where clear evidence sug-
and some primary diseases [47]. Chronic hypertension is
gests maternal and fetal outcomes are worse irrespective of
highly prevalent in women with CKD, and alone confers a
renal function, particularly with antiphospholipid syn-
7–8-fold relative risk of preeclampsia, which drives other
drome or active renal disease [35, 47]. Active disease pre
adverse perinatal outcomes [48]. Hypertension is an addi-
conception and during pregnancy is associated with
tive risk for adverse outcome when combined with any
increased maternal and fetal morbidity [35, 36, 47, 58]. A
stage of CKD [2, 14]. Early-stage CKD may be under-
meta-analysis of outcomes of pregnancy with SLE nephri-
recognized in pregnancy, but also confers an increased
tis (Table 27.1) found increased risk of SLE flares, hyper-
risk of pregnancy complications and adverse perinatal
tension, preeclampsia, premature birth, and low birth
outcome compared to women with no CKD, and is not
weight, particularly in women with active SLE nephri-
entirely benign [2, 14, 43]. Reassuringly maternal mortal-
tis [47]. Diabetic nephropathy is also an additive risk factor
ity is very rare [3].
for worse outcomes for both type 1 and type 2 diabetic
women who already have higher rates of pregnancy com-
Impact of Pregnancy on Kidney Function plications [54, 59, 60]. There may be an increased risk of
congenital malformations with diabetic nephropathy
The influence of pregnancy on renal outcomes is not
above that of poorly controlled diabetes alone [61].
always predictable or guaranteed despite best efforts at risk
stratification, which makes counseling a challenge. Our
understanding of the risk of decline in kidney function Pregnancy and Dialysis
with pregnancy in women with CKD has been informed by
several cohort studies [14, 46, 49]. A large systematic Pregnancies in the hemodialysis population have histori-
review showed no difference in renal outcomes between cally had poor outcomes. Since the 1990s, with better
women with or without CKD in pregnancy (Table 27.1). perinatal care and understanding of the importance of
However, this review included few studies in women with intensive dialysis, outcomes have improved, with
advanced CKD who are at greatest risk of declining renal increased live birth rates of 70–80% [4]. However, preg-
function during and after pregnancy, and evaluated sub- nancies in chronically dialysed women remain compli-
stantial renal decline only [43]. A long-standing cohort cated by maternal hypertension (50–70%), preeclampsia
study indicated that across all stages of CKD, pregnancy (18–67%), polyhydramnios (up to 40%), intrauterine
was associated with an increase in hypertension, increase growth restriction (17–77%), and prematurity and low
in proteinuria, and worsening of CKD stage postpartum, birth weight (50–100%) [3–5, 62]. Compared to trans-
and the risk increased based on preconception CKD planted patients, patients receiving hemodialysis have a
stage [14]. A smaller longitudinal cohort study in later higher rate of growth restriction, placental abruption,
stages of CKD showed proteinuria >1 g/24 h and an eGFR blood transfusion, and intrauterine death [63]. Enhanced
under 40 ml/min/1.73 m2 predicted decline in eGFR clearance of urea and other molecules through intensi-
postpartum [46]. fied dialysis regimens (at minimum >20 hours per week)
Overall, preconception kidney function is an important has beneficial effects on fetal mortality, birth weight, and
predictor of risk of decline in kidney function during and gestational age, and is especially relevant in women with
after pregnancy, but it must be considered along with other minimal residual renal function [62]. The presence of
factors such as hypertension, proteinuria as an indicator of residual renal function, particularly being nondialysis
chronic damage or primary disease activity and renal dependent at time of conception, leads to higher live birth
biopsy information. In addition, antenatal complications rates (up to 90% for pregnancies reaching 20 weeks of ges-
such as infection, preeclampsia and hypovolaemia may tation) compared to those who conceive once established
cause an AKI, which can also have an impact on longer- on hemodialysis with less residual function [5, 64]. No
term outcomes. studies have assessed the level of renal function at which
Table 27.1 Summary of selected systematic reviews relevant to kidney function assessment, pregnancy in women with kidney disease, and transplants.
Search Patients Analysis Data quality (as per

Author (year) Title period Studies (pregnancies) Inclusion criteria method reviewers) Key findings
Diagnosis and management

Cote, et al. Diagnostic accuracy 1980–2007 13 1241 Studies that compared Quality Minimal study Spot PCR can exclude proteinuria of
(2008) [29] of urinary spot urinary protein Assessment of heterogeneity. Studies >0.3 g/day in hypertensive women.
protein:creatinine creatinine ratio or Diagnostic inadequately described Sensitivity 83.6%; specificity 76.3%;
ratio for proteinuria albumin creatinine ratio Accuracy selection criteria, positive likelihood ratio 3.53; negative
in hypertensive with urine protein Studies tool for disease spectrum and likelihood ratio 0.21.
pregnant women: excretion over 24 h in assessing test execution.
systematic review pregnant women with validity;
hypertension diagnostic
meta-analysis
Piccoli et al. Kidney biopsy in 1980–2010 39 243 biopsies Studies on kidney biopsy Narrative Heterogeneous cohorts; Higher complication rate for renal
(2013) [40] pregnancy: evidence in pregnancy; in pregnancy with >5 synthesis only narrative review biopsies in pregnancy compared to
for counseling? A 1236 post patients. Excluded renal possible postpartum (7% vs.1%). Biopsy led to
systematic narrative transplant therapeutic changes in 66% of cases.
review
Tong et al. Perspectives on To April 15 257 Qualitative studies on The The review was limited Seven themes identified: pursuing
(2015) [50] pregnancy in women 2014 experiences of adult Enhancing by study cohort motherhood, failure to fulfill social
with chronic kidney women with pregnancy transparency heterogeneity. norms, fear of birth defects,
disease: systematic and CKD in reporting Pregnancy was not exacerbating disease, decisional
review of qualitative the synthesis always the major focus insecurity and conflict, control and
studies of qualitative of the study. determination, withholding emotional
research investment.
framework;
thematic
synthesis
Transplantation
Bar et al. Pregnancy outcome NA 15 410 Studies reporting Meta-analysis Definition of Overall prevalence of malformations
(2001) [51] after cyclosporine congenital malformations malformation varied. with cyclosporine exposure was similar
therapy during with babies exposure to Studies were at risk of to the background rate in the general
pregnancy: a cyclosporine, with >10 publication and population.
meta-analysis patients selection bias.
Deshpande Pregnancy outcomes 2000–2010 50 3570 (4706) Studies on pregnancy, Meta-analysis Potential patient overlap Live birth rate was higher and
et al. in kidney transplant obstetric, delivery between studies miscarriage rate was lower than the
(2011) [52] recipients: a complications in kidney (repeated inclusion). general population (possibly due to
systematic review and transplant recipients Definitions and bias). Preeclampsia, gestational diabetes,
meta-analysis Case reports, abstracts classification of key preterm delivery, and cesarean section
and articles not in English outcomes varied were higher than the general
were excluded between studies. population. Hypertension, proteinuria
Reporting and selection and elevated serum creatinine were
bias was likely due to associated with adverse outcomes.
voluntary reporting.
c27.indd 448 09-12-2022 15:47:22

Dialysis
Piccoli et al. Pregnancy in dialysis January 126 647 (681) Studies on pregnancy in Meta-analysis Study cohorts were Fewer hours of dialysis per week was
(2016) [3] patients in the new 2000– chronic dialysis patients heterogeneous. associated with preterm deliveries, small
millennium: a December who started dialysis in the Historical studies were for gestational age births. Lower
systematic review and 2014 first 20 weeks of gestation; excluded. incidence of small for gestational age
meta-regression gestation >24 weeks; with babies in hemodialysis compared to
analysis correlating >4 patients peritoneal dialysis.
dialysis schedules and
pregnancy outcomes
Piccoli et al. Pregnancy in dialysis 2000–2008 10 78 (90) Studies of pregnancy in Descriptive Studies had varied Intrauterine death, polyhydramnios,
(2010) [40] patients: is the dialysis with >5 patients narrative definition of outcomes preterm delivery and small for
evidence strong review and study period with gestational age were prevalent in
enough to lead us to high heterogeneity and dialysed women. Possibility of a healthy
change our low numbers. child was 50–100%.
counseling policy
Chronic kidney disease
Bramham Chronic hypertension To June 55 (795 221 Studies of pregnant Preferred Study cohorts were Women with hypertension in pregnancy
et al. and pregnancy 2013 pregnancies) women with chronic Reporting heterogeneous. had a high incidence of preeclampsia,
(2014) [48] outcomes: systematic hypertension excluding Items for cesarean section, preterm delivery, low
review and women with Systematic birth weight, intensive care unit,
meta-analysis superimposed Reviews and admission and perinatal death.
preeclampsia with Meta-
>20 women Analyses,
guidelines;
Newcastle-
Ottawa scale
for grading;
meta-analysis
Nevis et al. Pregnancy outcomes To June 13 Cases 2682, Studies that reported fetal Descriptive Confounding factors The risk of adverse fetal outcomes of
(2011) [45] in women with 2010 controls or maternal outcomes in review and were adjusted in premature birth, small for gestational
chronic kidney 26 149 women with CKD with meta-analysis analyses in only six age, intrauterine growth restriction, low
disease: a systematic small size >5, dialysis and studies. Studies were birth weight and stillbirth were two times
review transplant patients were often small and higher in women with any CKD. Due to
excluded retrospective, with a study limitations, definitive risks for
risk of selection, women with CKD are not able to be
ascertainment, quoted on the basis of this literature.
surveillance and
misclassification bias.
Zhang et al. A systematic review 1946–2014 23 (506 340) Cohort and case control GRADE The outcomes of Pregnancy in women with CKD had a
(2015) [43] and meta-analysis of studies of women with system for preeclampsia and greater risk of preeclampsia, premature
outcomes of CKD and maternal or assessing premature delivery had delivery, small for gestational age and
pregnancy in CKD fetal outcomes, and with study quality; high-grade evidence cesarean section. Proteinuria was a risk
and CKD outcomes in a non-CKD control group. meta-analysis quality. The other factor for preeclampsia and preterm
pregnancy Studies of women with a outcomes were very low delivery. Pregnancy was not a risk factor
history of autoimmune grade due to for progression of kidney disease, but
disease, on maintenance heterogeneity. Studies included patients had mainly early-stage
HD, kidney transplant, or included had few CKD.
acute kidney injury were women with renal
excluded. impairment.
(Continued)
c27.indd 449 09-12-2022 15:47:22

Search Patients Analysis Data quality (as per

Author (year) Title period Studies (pregnancies) Inclusion criteria method reviewers) Key findings
Blom et al. Pregnancy and 1980– 18 887 (1414) Studies on frequency of Narrative Significant study Regardless of etiology of glomerular
(2017) [53] glomerular disease: a February live births of women with review heterogeneity which disease, increased proteinuria,
systematic review of 2016 biopsy proven precluded pooling of hypertension or renal dysfunction were
the literature with glomerulonephritis with data thus review was associated with worse pregnancy
management >10 cases narrative. outcomes
guidelines
Piccoli et al. Type 1 diabetes, 1980–2012 35 (1 511) Studies on pregnancy Narrative Studies were highly The evidence on pregnancy outcomes in
(2013) [54] diabetic nephropathy, outcomes in women with review heterogeneous and only diabetic nephropathy in scarce. The
and pregnancy: a type 1 diabetes and narrative review was risks of adverse pregnancy events are
systematic review and nephropathy with <5 possible. high and have not changed over time.
meta-study sample size Stillbirth and fetal death are in increased
in diabetic nephropathy compared to
other primary disease.
Piccoli et al. A systematic review 2000–April 27 581 (729) Case reports or case series Meta-analysis There was No increased risk in progression of kidney
(2018) [12] on the materno-fetal 2017 control 562 that reported on heterogeneity in terms disease in women with IgA who had
outcomes in pregnant pregnancy outcome or of definitions, pregnancy vs. no pregnancy. Adverse
women with IgA renal function in women populations, outcomes pregnancy outcomes were increased
nephropathy: A case with IgA nephropathy and follow up duration. compared to healthy controls – 10×
of “late maternal” Meta-analysis was increased risk of preeclampsia and
preeclampsia conducted. Most studies hypertension. Preterm birth and risk of low
had women with birth weight babies was increased to a
normal kidney lesser degree. Authors hypothesized that
function. late preeclampsia may have less impact on
fetal outcomes.
Liu et al. A systematic review 1966–2015 4 273 (376) vs. Cohort or case control GRADE Moderate quality No increase in the risk of adverse renal
(2016) [55] and meta-analysis of 241 IgA studies of patients with criteria to evidence. No evidence outcomes compared to IgA nephropathy
kidney and pregnancy without biopsy proven IgA on assess study of publication bias. without pregnancy. There were high
outcomes in IgA pregnancy pregnancy outcomes quality; rates of preeclampsia, infant loss,
nephropathy meta-analysis preterm delivery and low birth weight.
Smyth et al. A systematic review 1966–2009 37 1 842 (2 751) Studies of pregnancy Cochrane Studies are mainly Controlling SLE activity prior to
(2010) [47] and meta-analysis of outcomes with at least search observational and conception is important as active SLE
pregnancy outcomes 1 month of follow-up strategy; subject to reporting and nephritis and antiphospholipid antibodies
in patients with SLE meta-analysis selection bias. increase the risk of maternal hypertension
and SLE nephritis and poor fetal outcomes.
Wu et al. Management and 2000–June 16 1760 Cohort or cross sectional Newcastle– Statistical heterogeneity Worse maternal and fetal outcomes
(2018) [56] outcomes of 2017 studies comparing Ottawa Scale was seen across the were seen with SLE nephritis. Anti-Ro
pregnancy with or Pregnant women with and Strength studies. antibodies had worse outcome. More
without lupus SLE without nephritis vs. ening the immunosuppression was used in SLE
nephritis: a without nephritis Reporting of nephritis.
systematic review and Observational
meta-analysis Studies in
Epidemiology
checklist to
assess study
quality;
meta-analysis
Reviews relating to pregnancy in non-CKD or mixed cohorts are not included due to space constraints. Analysis methodology describes systems used for grading quality of studies (if done),
and broad method of data synthesis (narrative, thematic or meta-analysis).
c27.indd 450 09-12-2022 15:47:22

Treatmen 451
dialysis should be commenced to optimize pregnancy it is worth noting that 30–40% of women will have graft fail-
outcomes or the impact of commencement of dialysis for ure by their child’s 10th birthday and 10–20% will not live to
fetal indications on maternal renal survival. see their child into adulthood [11, 69, 70, 73]. Pregnancy is
There is less data on pregnancy in peritoneal dialysis less likely to adversely impact graft outcomes in women with
recipients, which remains very rare [3, 5]. Successful preg- normal and stable allograft function. There is growing data
nancy outcomes have been observed in women receiving demonstrating that preconception graft function predicts
PD throughout pregnancy, but peritonitis remains the long-term graft function after pregnancy [67, 68, 70–72]. In
main concern [65]. Most women are switched to hemodi- one cohort study where detailed kidney function data was
alysis [5], and there is some data suggesting fetal outcomes available at multiple time-points, the risk of decline in graft
are better with peritoneal dialysis than haemodialysis [3]. function during pregnancy and at 12 months postpartum
was increased as preconception creatinine declined. Women
with serum creatinine >140 μmol/l had worse renal trajec-
Pregnancy Outcomes after Kidney Transplantation
tory and risk of graft loss [68]. Preeclampsia can increase the
Transplantation restores fertility and offers an opportunity risk of temporary decline in kidney function but has not
to achieve parenthood with less risk than in dialysed women. been shown to impact adversely on longer term graft out-
In the recent era, elective termination rates have fallen and come [72]. Time from transplant to pregnancy may also
transplanted women are very likely to have a live birth (over- influence outcomes, with one large US study suggesting
all live birth rates of 76–98% in pregnancies reaching pregnancies within 3 years post transplantation were associ-
20 weeks gestation) [66, 67]. Compared to patients receiving ated with poorer survival [74] but this was not consistently
dialysis, the maternal and fetal outcomes post transplant are demonstrated in other cohorts [11, 66].
superior, though the rate of conception remains lower than The longer-term impact of immunosuppression modi-
in the general population [9, 63]. However, obstetric compli- fication, particularly cessation of mycophenolic acid
cations remain high, with increased risk of gestational derivatives, for facilitation of pregnancy remains
hypertension, preeclampsia (seen in 25–35% of pregnan- unknown. Episodes of acute rejection in pregnancy are
cies), growth restriction, and preterm birth (in 30–50%) [52, reassuringly low (2–14.5%) and no higher than in the
63, 68, 69]. Australian registry analysis demonstrated a mean nonpregnant population [52, 69, 73, 75–77]. This could
difference in gestational age of 3.5 weeks, gestational weight be under-reported as biopsy may selectively be avoided
of 873 g, and lower perinatal survival for babies born to during pregnancy. De novo allo-sensitization is frequent
transplanted mothers compared to the general popula- after nontransplant pregnancy, but only reported in 6% of
tion [66]. Higher gestational diabetes rates have been transplanted women, presumably due to immunosup-
reported in some studies [52], but not consistently [66, 68]. A pression use [78].
transplantation-to-conception interval of under 2 years has
been associated with higher rates of gestation diabetes,
preeclampsia, and cesarean section [52]. Similar to the CKD
cohort, preconception kidney function and chronic hyper-
T
reatment
tension are predictive of adverse obstetric outcomes, par-
Prepregnancy Counseling and Planning
ticularly preeclampsia [52, 68–71]. Majak et al. reported the
risk of preeclampsia increased in women with chronic Parenthood should be discussed early to determine the
hypertension, previous preeclampsia or preconception optimal “window” for conception based on an individual’s
serum creatinine 125 μmol/l. The effect of each risk factor life plan and clinical situation. There is increasing support
was cumulative, and those with all three factors had a 96% for women with CKD or transplants in pursuing mother-
risk of preeclampsia [72]. Immunosuppression exposure, hood, and clinical practice guidelines no longer advise
infections, and diabetes may influence outcomes but the against pregnancy but encourage counseling with individ-
data defining this are limited. Despite longer immunosup- ual risk stratification based on relevant comorbidities.
pression exposure, women who are transplanted in child- Given the morbidity associated with pregnancy, careful
hood have similar outcomes in pregnancy to those planning is essential with effective contraception in place.
transplanted in adulthood [67]. This provides the opportunity to optimize health and risk
factors for adverse outcomes and adjust medications. In
addition, clinicians can obtain multidisciplinary assess-
Impact of Pregnancy on Transplant Outcomes
ment, engage additional expertise, for example preimplan-
Several registry and cohort studies have reported that graft tation genetic testing for inherited diseases, assess fertility
and patient failure is not accelerated by pregnancy, although and confirm the best timing of pregnancy [79, 80].
452 Pregnancy
Navigating pregnancy can be emotionally daunting for Optimize Maternal Health Pre Conception
women with CKD or transplants. Counseling should incor-
Women planning pregnancy should cease smoking, com-
porate shared decision-making, which involves understand-
mence folic acid and vitamins, exchange teratogenic med-
ing patient values and perspectives, balancing the emotional
vulnerabilities, while sensitively acknowledging the medical ications for pregnancy-safe options, and address relevant
risks [23]. Women with CKD have described the need to bal- comorbidities including infection, anemia, and diabetes.
ance social expectations and their strong desire for mother- Primary renal disease should be controlled or in remis-
hood with fears regarding disease exacerbation, their own sion, and transplant function and immunological param-
survival, and fetal harm. Explanation of medical risk should eters should be stable. Assessment of renal status should
avoid catastrophizing and be balanced with a women’s val- include abiopsy if necessary to clarify diagnosis, disease
ues of autonomy and family, focusing on shared decision- activity or exclude subclinical rejection. Optimization of
making and provision of hope [23, 81]. In a retrospective glycaemic control in patients with diabetes can reduce
study of women managed in a specialized multidisciplinary maternal and fetal adverse outcomes [59]. Vesicoureteric
preconception counseling clinic for CKD, 92% found expert reflux and polycystic kidney disease confer increased risk
counseling useful in assisting with decision-making [80]. of urinary tract infections [90]. Symptomatic urinary tract
Risk stratification based on key factors associated with infections and asymptomatic bacteriuria are associated
adverse pregnancy outcome can identify areas where mater- with maternal sepsis as well as preterm birth and low
nal health may be optimized (Table 27.2). birth weight and should be treated aggressively [91].
Table 27.2 Recommendations for treatment and management of women with CKD in pregnancy.
Strength of
Preconception We recommend expert pre conception counseling Moderate Strong [23, 50, 80, 81]
counseling be undertaken early to enable shared decision
making about parenthood options for women with
CKD
Risk assessment We recommend using pre-conception kidney Moderate Strong [43, 45]
function as the primary parameter for assessing
risk of adverse pregnancy and renal outcomes
Cessation of We suggest MMF or MPA be ceased at least Moderate Strong [82, 83]
mycophenolate mofetil 3 months prior to conception due to
and Mycophenolic acid embryotoxicity
Cessation of ACEI or We suggest in continuing ACEI/ARB until Very low Weak [48, 84, 85]
ARB conception in selected women, where the benefit
of controlling proteinuria may outweigh fetal risks
from early first trimester exposure
Controlling hypertension We suggest controlling hypertension prior to Moderate Strong [2, 46, 86]
conception and during pregnancy
Aiming for a target blood pressure of 140/90 may
reduce episodes of severe hypertension in
pregnancy and does not adversely affect the fetus
Prophylactic aspirin We suggest women at high risk of preeclampsia Very low (in CKD Strong [87, 88]
with CKD, kidney transplant, hypertension or SLE population);
as their primary disease take prophylactic moderate-strong in
low-dose aspirin from the first trimester for general pregnancy
preeclampsia prevention cohorts
Intensive dialysis We suggest that dialysis delivery be intensified in Low–moderate Strong [40, 46, 89]
pregnant women receiving renal replacement
therapy, tailored to residual renal function
More than 20 and up to 36 h per week should be
considered in women with no residual function,
but must be balanced with what can be tolerated
and accommodated
Treatmen 453
SLE nephritis should be quiescent 6 months prior to con- Women should commence low-dose aspirin (75–100 mg
ception to minimize risk of disease flare and improve odds daily) before 12 weeks and continue until late in the third
of pregnancy success [35, 47]. Hydroxychloroquine should trimester. Reduction in preeclampsia risk due to aspirin
be continued during pregnancy as cessation is a risk factor has been demonstrated in low-risk [101] and high-risk
for disease flare and continuation improves fetal out- populations [87, 104]. A meta-analysis that included mod-
come [58, 92, 93]. A meta-analysis of hydroxychloroquine erate-to high-risk patients, including those with CKD and
use found no increase in rate of congenital malformation hypertension, confirmed a risk reduction of 0.47 (95% CI
but observed more spontaneous abortion, possibly from 0.36–0.62) vs. 0.78 (95% CI 0.61–0.99; P < 0.01), if com-
indication bias related to underlying disease activity [94]. menced prior to 16 weeks gestation [88]. These conclusions
All patients with SLE should be screened for antiphospho- have been extrapolated to patients with CKD and incorpo-
lipid antibodies as they portend a worse prognosis [35, 36, rated into guidelines [72, 99, 105].
47]. Antiphospholipid syndrome with thrombosis should Given pregnancy is a prothrombotic state, women with
be treated with anticoagulation. Screening for anti-Ro and nephrotic range proteinuria should be considered for pro-
anti-La antibodies should occur to identify those at risk of phylactic anticoagulation [79], although there are negligi-
neonatal cardiac SLE, particularly given hydroxychloro- ble outcomes data. Nutrition should be evaluated and
quine can reduce this risk [35, 95–97]. supplementation provided as required. Gestational diabe-
tes screening should occur earlier than routine screening
in women with risk factors for diabetes including prodiabe-
Hypertension
togenic medications such as glucocorticoids or calcineurin
Controlling hypertension before and during pregnancy is inhibitors. Managing anemia in pregnancy with concur-
essential. There is insufficient evidence for blood pressure rent CKD can be challenging due to increased iron require-
targets in pregnancy with CKD, though consensus guide- ments with dose increases of 30–50% required in
lines recommend a target of <140/90 [98–100]. The CHIPS pregnancy [106]. Erythropoietin (EPO)-stimulating agent
trial showed tighter control of blood pressure reduced epi- dose may increase due to EPO resistance from increase
sodes of maternal hypertension over 160/110 mmHg with- cytokines production in pregnancy.
out adverse fetal outcomes, but this has not been studied in There are no data available on optimal timing of delivery
CKD populations [86, 101]. Antihypertensive therapy in a CKD population and thus decisions should be based
should be changed to pregnancy-safe drugs such as on local practice, and maternal and fetal health. Cesarean
labetalol, methyldopa, and nifedipine, but there are no section is not specifically indicated other than for obstetric
data to guide choice of one agent over another [13, 79]. reasons, but rates are substantially higher in women with
Angiotensin receptor blockers (ARB) and angiotensin con- CKD or post transplant, likely due to need for emergency
verting enzymes inhibitors (ACEI) are contraindicated delivery [14, 45]. In patients with renal transplantation,
beyond the first trimester due to risk of renal dysplasia, oli- cesarean sections should be performed with input from the
gohydramnios, neonatal renal failure, and fetal death [102]. transplant surgical team regarding anatomy. Breastfeeding
More recent population studies suggest hypertension and while receiving medications considered safe in pregnancy
underlying comorbidities may be teratogenic rather than should not be discouraged, as fetal drug exposure is mini-
ACEI or ARB themselves [48, 84, 85]. Continued stabiliza- mal compared with intrauterine exposure and benefits
tion of proteinuria with these drugs, with careful surveil- may outweigh risks [107]. Contraceptive options should be
lance for pregnancy, may be recommended in selected discussed and implemented postpartum, and may require
women where the risk of prolonged cessation of renin- careful consideration based on patient preferences and
angiotensin blockade outweighs the potential fetal risk guidelines for contraception in women with medical
from exposure at conception [103]. comorbidity [13].
Pregnancy Management: General Principles Immunosuppression

Once pregnant, multidisciplinary antenatal care and a tai- Immunosuppression may require manipulation prior to
lored plan for maternal and fetal surveillance should be conception for both transplanted women and those with
established based on CKD stage, comorbidity, and trans- glomerulonephritis [13], although the impact this has on
plant status (Figure 27.1) [79]. Careful vigilance should renal outcomes is unknown. Observation for 3–6 months to
occur for declining maternal renal health, superimposed ensure renal condition remains stable is recommended.
preeclampsia, fetal growth restriction, polyhydramnios, and Calcineurin inhibitors, glucocorticoids, and azathioprine
infection, as these events will define timing of delivery. are most commonly used and have no increased risk of
454 Pregnancy
CKD with Potential for Pregnancy

And Sexually Active
• Highly-effective Contraception
• Precautions for STD
Deferring Pregnancy
• Cervical Cancer screening
• Breast Cancer Screening
Contemplating Pregnancy Not Contemplating Pregnancy
• Consider tubal ligation
Pre-Pregnancy Planning
Going Ahead–Prepare for Pregnancy
Informed, Shared Decision-making
Risk Stratification Optimise Maternal Health

• CKD stage • BMI • Control of Primary Disease
• BP • Comorbidities • BP Control
• Proteinuria • Infection risk • Adjust Medications
• Anaemia • Smoking • Timing of ACEI/ARB cessation
• Thrombosis risk • Nutrition • Commence Folic Acid
• DM/GDM risk • Obstetric history
Monitor Closely while Trying to
Additional Advice Conceive
• Maternal-fetal Medicine / Obstetrics
• Geneticist, Anaesthetics, Other
• Reproductive / Fertility Medicine
Pregnancy Management
Discuss Timing
• Consider age and fertility
• Defer pregnancy
• Proceed with pregnancy planning • Confirm Pregnancy
• Establish Multi-disciplinary Team
• Low-Dose Aspirin
• Continue Folic Acid
Potential for Pregnancy • Plan Delivery
and Sexually Active
Maternal Surveillance
• Renal Function – monthly*
• Proteinuria – monthly*
• Repeat Algorithm • Blood Pressure – weekly*
• Bacteriuria – each trimester*
• GDM Screening – possibly early
• Immunosuppression levels – monthly*
• Immune Parameters (e.g. in SLE)
Post-Partum Management
Fetal Surveillance
• 1st and 2nd Trimester Screening
• Breast-feeding – don’t discourage • Growth scans after 28–30 weeks
• Monitor and Manage Renal Disease • Doppler Studies
• Adjust Medications • Cardiotocographs
• Contraception • Cardiac monitoring in Ro+/La+ women
Figure 27.1 Algorithm for pregnancy management in women with CKD and transplant. *Or more frequently if required. ACEI,
angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; BMI, body mass index; BP, blood pressure; CKD, chronic
kidney disease; DM, diabetes mellitus; GDM, gestational diabetes mellitus; SLE, systemic lupus erythematosus; STD, sexually
transmitted disease. Source: Adapted with permission from Fitzpatrick et al. [79]. © 2016, Dove Medical Press Limited.
Conclusio 455
congenital malformation [51, 108–110]. Cyclosporine lev- is important to outcomes [5, 64]. Less dialysis may be
els decline in pregnancy and require dose alteration [108]. required if residual renal function is present, substan-
Similarly, the pharmokinetics of tacrolimus are altered in tially reducing the dialysis burden [89].
pregnancy-related shifts in albumin and hematocrit, with a Intensive dialysis offers better control maternal of hyper-
decrease in the inactive, bound fraction that is measured in tension and fluid status, although fetal and placental
assays. Tacrolimus trough levels reach a nadir in the sec- hypoperfusion from excessive fluid removal should be
ond trimester [68, 108]. Subsequent dose alterations by up avoided, with regular clinical assessment as women gain
to 25% can increase the unbound active fraction that is not weight [62, 89, 116]. Other sequelae of intense dialysis
currently routinely measured, increasing toxicity risk [111, include electrolyte imbalance with risk of hypokalaemia,
112]. Mycophenolate mofetil causes birth defects and mis- hypocalcaemia, and hypophosphatemia requiring dialysate
carriage [82, 83] so should be ceased 3 months prior to con- changes [62, 116, 117]. Increased protein loss requires care-
ception, usually with conversion to azathioprine, which is ful dietetic management of dietary protein intake [62, 118].
transferred to the fetus but not activated due to absence of Increased folate supplementation to 5 mg/day is advised to
inosinate pyrophosphorylase. Safety data on mammalian reduce risk of neural tube defects [116].
target of rapamycin in pregnancy is limited though it may There are few data on the use of peritoneal dialysis in
affect fetal growth therefore cessation is advised [113]. pregnancy. Outcomes may be inferior to pregnancies man-
Treatment of transplant rejection in pregnancy should be aged with hemodialysis, but direct comparison cannot be
as per local practice. Options for treating acute glomerulo- made with the existing evidence base [3]. Increasing dialy-
nephritis, including SLE nephritis in pregnancy, are also sis delivery through PD can be a successful management
limited, with reports of use of steroids, myophenolate strategy, but a growing uterus may limit exchange volumes
mofetil cyclophosphamide, and rituximab, all of which and clearance, and cause catheter migration [65].
have adverse safety profiles in pregnancy [13]. Tacrolimus
may be a useful therapy for SLE nephritis in pregnancy but
Management of Transplanted Women
further data are required [114]. Women with SLE should
receive hydroxychloroquine pre conception and through- The management of women post kidney transplantation
out pregnancy due to fetal benefits [47]. should incorporate the general principles as outlined
above. Determining the appropriate timing of conception
post transplant is a balance between stabilization of graft
Dialysis Management
function and fitting into the window of fertility. Guidelines
There is no evidence base regarding when to initiate dialy- recommend conception only 1–2 years post transplanta-
sis in pregnant patients with CKD [67]. Women with tion [39, 119], although the optimal time is unknown and
advanced CKD may require commencement of dialysis early pregnancy post transplant (within 3 years) may
during pregnancy for maternal symptoms, hypertension or impact on graft function [74]. Registry data suggest most
fluid management, or reducing urea levels, which improves women wait 3–5 years before conceiving [66]. Prior to con-
fetal outcomes [115]. ception, graft function should be stable on pregnancy-safe
In women who conceive while receiving chronic hemo- immunosuppression, with controlled hypertension and
dialysis, intensive dialysis has been demonstrated to low risk of rejection or infection [39, 119]. Other specific
improve fetal outcomes [3, 62] due to increased urea management includes adjustment of immunosuppression
clearance causing less fetal solute diuresis and polyhy- and other relevant medication such as antivirals and anti-
dramnios [14]. Historically a target urea of under hypertensives. Surveillance should be undertaken of allo-
17.9 mmol/l (50 mg/dl) was recommended [46], with fur- graft dysfunction, gestational diabetes, bacterial infection,
ther studies confirming better outcomes with low urea particularly urinary tract, viral infection, including cyto-
targets [3]. A large meta-analysis of 681 pregnancies megalovirus and herpes simplex virus, and vigilant moni-
from 126 studies reviewed dialysis schedule and preg- toring for preeclampsia and rejection.
nancy outcomes, demonstrating a relationship between
greater number of dialysis hours and reduced risk of
small gestational age and preterm delivery [3]. Intensive C
onclusion
dialysis (>36 h/week) compared to <20 h/week was asso-
ciated with higher live birth rates (85% vs. 48%), greater Many physiological changes occur during pregnancy
birth weight, and older gestational age [89]. Logistically, affecting the kidney, and management of kidney disease in
this is challenging but can be achieved through daily or pregnancy presents a number of challenges to clinicians,
nocturnal dialysis. The degree of residual renal function particularly as the evidence base that underpins complex
456 Pregnancy
clinical decision-making is weak. Risk of adverse outcomes to cautious optimism. The need to incorporate shared
for mother and baby are present in even mild CKD and decision-making between patients, families, and a multi-
escalate with disease severity and added comorbidity. disciplinary team, with an understanding of patients’ moti-
Experience in managing these complex pregnancies has vations and values, is an essential part of embarking on
increased over recent decades, with a change in approach this journey with this group of high-risk women.
R
eferences
1 Piccoli, G.B., Conijn, A., Attini, R. et al. (2011). Pregnancy 12 Gill, J.S., Zalunardo, N., Rose, C., and Tonelli, M. (2009).
in chronic kidney disease: need for a common language. The pregnancy rate and live birth rate in kidney
J. Nephrol. 24 (3): 282–299. transplant recipients. Am. J. Transplant. 9 (7): 1541–1549.
2 Munkhaugen, J., Lydersen, S., Romundstad, P.R. et al. 13 Wiles, K.S., Nelson-Piercy, C., and Bramham, K. (2018).
(2009). Kidney function and future risk for adverse Reproductive health and pregnancy in women with
pregnancy outcomes: a population-based study from chronic kidney disease. Nat. Rev. Nephrol. 14 (3):
HUNT II, Norway. Nephrol. Dial. Transplant. 24 (12): 165–184.
3744–3750. 14 Piccoli, G.B., Cabiddu, G., Attini, R. et al. (2015). Risk of
3 Piccoli, G.B., Minelli, F., Versino, E. et al. (2016). adverse pregnancy outcomes in women with CKD. J. Am.
Pregnancy in dialysis patients in the new millennium: a Soc. Nephrol. 26 (8): 2011–2022.
systematic review and meta-regression analysis 15 Hildebrand, A.M., Liu, K., Shariff, S.Z. et al. (2015).
correlating dialysis schedules and pregnancy outcomes. Characteristics and outcomes of AKI treated with dialysis
Nephrol. Dial. Transplant. 31 (11): 1915–1934. during pregnancy and the postpartum period. J. Am. Soc.
4 Shahir, A.K., Briggs, N., Katsoulis, J., and Levidiotis, V. Nephrol. 26 (12): 3085–3091.
(2013). An observational outcomes study from 1966-2008, 16 Prakash, J., Niwas, S.S., Parekh, A. et al. (2010). Acute
examining pregnancy and neonatal outcomes from kidney injury in late pregnancy in developing countries.
dialysed women using data from the ANZDATA Registry. Ren. Fail. 32 (3): 309–313.
Nephrology (Carlton) 18 (4): 276–284. 17 Odutayo, A. and Hladunewich, M. (2012). Obstetric
5 Jesudason, S., Grace, B.S., and McDonald, S.P. (2014). nephrology: renal hemodynamic and metabolic
Pregnancy outcomes according to dialysis commencing physiology in normal pregnancy. Clin. J. Am. Soc.
before or after conception in women with ESRD. Clin. J. Nephrol. 7 (12): 2073–2080.
Am. Soc. Nephrol. 9 (1): 143–149. 18 Davison, J.M. and Dunlop, W. (1980). Renal
6 Toma, H., Tanabe, K., Tokumoto, T. et al. (1999). hemodynamics and tubular function normal human
Pregnancy in women receiving renal dialysis or pregnancy. Kidney Int. 18 (2): 152–161.
transplantation in Japan: a nationwide survey. Nephrol. 19 Kristensen, K., Lindstrom, V., Schmidt, C. et al. (2007).
Dial. Transplant. 14 (6): 1511–1516. Temporal changes of the plasma levels of cystatin C,
7 Bagon, J.A., Vernaeve, H., De Muylder, X. et al. (1998). beta-trace protein, beta2-microglobulin, urate and
Pregnancy and dialysis. Am. J. Kidney Dis. 31 (5): creatinine during pregnancy indicate continuous
756–765. alterations in the renal filtration process. Scand. J. Clin.
8 Okundaye, I., Abrinko, P., and Hou, S. (1998). Registry of Lab. Invest. 67 (6): 612–618.
pregnancy in dialysis patients. Am. J. Kidney Dis. 31 (5): 20 Sikora-Grabka, E., Adamczak, M., Kuczera, P. et al.
766–773. (2016). Serum anti-Mullerian hormone concentration in
9 Piccoli, G.B., Cabiddu, G., Daidone, G. et al. (2014). The young women with chronic kidney disease on
children of dialysis: live-born babies from on-dialysis hemodialysis, and after successful kidney transplantation.
mothers in Italy–an epidemiological perspective Kidney Blood Press. Res. 41 (5): 552–560.
comparing dialysis, kidney transplantation and the 21 Palmer, B.F. and Clegg, D.J. (2017). Gonadal dysfunction
overall population. Nephrol. Dial. Transplant. 29 (8): in chronic kidney disease. Rev. Endocr. Metab. Disord. 18
1578–1586. (1): 117–130.
10 Hou, S. (2001). Conception and pregnancy in peritoneal 22 Holley, J.L. (2004). The hypothalamic-pituitary axis in
dialysis patients. Perit. Dial. Int. 21 (Suppl 3): S290–S294. men and women with chronic kidney disease. Adv.
11 Levidiotis, V., Chang, S., and McDonald, S. (2009). Chronic Kidney Dis. 11 (4): 337–341.
Pregnancy and maternal outcomes among kidney transplant 23 Tong, A., Brown, M.A., Winkelmayer, W.C. et al. (2015).
recipients. J. Am. Soc. Nephrol. 20 (11): 2433–2440. Perspectives on pregnancy in women with CKD: a
Reference 457
semistructured interview study. Am. J. Kidney Dis. 66 38 Rolfo, A., Attini, R., Tavassoli, E. et al. (2015). Is it
(6): 951–961. possible to differentiate chronic kidney disease and
24 Barua, M., Hladunewich, M., Keunen, J. et al. (2008). preeclampsia by means of new and old biomarkers? A
Successful pregnancies on nocturnal home hemodialysis. prospective study. Dis. Markers 2015: 127083.
Clin. J. Am. Soc. Nephrol. 3 (2): 392–396. 39 Agrawal, S., Cerdeira, A.S., Redman, C., and Vatish, M.
25 Alper, A.B., Yi, Y., Webber, L.S. et al. (2007). Estimation (2018). Meta-analysis and systematic review to assess the
of glomerular filtration rate in preeclamptic patients. Am. role of soluble FMS-like tyrosine kinase-1 and placenta
J. Perinatol. 24 (10): 569–574. growth factor ratio in prediction of preeclampsia: the
26 Smith, M.C., Moran, P., Ward, M.K., and Davison, J.M. SaPPPhirE study. Hypertension 71 (2): 306–316.
(2008). Assessment of glomerular filtration rate during 40 Piccoli, G.B., Daidola, G., Attini, R. et al. (2013). Kidney
pregnancy using the MDRD formula. BJOG 115 (1): biopsy in pregnancy: evidence for counselling? A
109–112. systematic narrative review. BJOG 120 (4): 412–427.
27 Koetje, P.M.J.L., Spaan, J.J., Kooman, J.P. et al. (2011). 41 Webster, P., Webster, L.M., Cook, H.T. et al. (2017). A
Pregnancy reduces the accuracy of the estimated multicenter cohort study of histologic findings and
glomerular filtration rate based on Cockroft-Gault and long-term outcomes of kidney disease in women who
MDRD formulas. Reprod. Sci. 18 (5): 456–462. have been pregnant. Clin. J. Am. Soc. Nephrol. 12 (3):
28 Higby, K., Suiter, C.R., Phelps, J.Y. et al. (1994). Normal 408–416.
values of urinary albumin and total protein excretion 42 Bramham, K., Briley, A.L., Seed, P.T. et al. (2011).
during pregnancy. Am. J. Obstet. Gynecol. 171 (4): Pregnancy outcome in women with chronic kidney
984–989. disease: a prospective cohort study. Reprod. Sci. 18 (7):
29 Cote, A.-M., Brown, M.A., Lam, E. et al. (2008). 623–630.
Diagnostic accuracy of urinary spot protein:creatinine 43 Zhang, J.-J., Ma, X.-X., Hao, L. et al. (2015). A systematic
ratio for proteinuria in hypertensive pregnant women: review and meta-analysis of outcomes of pregnancy in
systematic review. BMJ 336 (7651): 1003–1006. CKD and CKD outcomes in pregnancy. Clin. J. Am. Soc.
30 Hou, S. (1999). Pregnancy in chronic renal insufficiency Nephrol. 10 (11): 1964–1978.
and end-stage renal disease. Am. J. Kidney Dis. 33 (2): 44 Kendrick, J., Sharma, S., Holmen, J. et al. (2015). Kidney
235–252. disease and maternal and fetal outcomes in pregnancy.
31 Benachi, A., Dreux, S., Kaddioui-Maalej, S. et al. (2010). Am. J. Kidney Dis. 66 (1): 55–59.
Down syndrome maternal serum screening in patients 45 Nevis, I.F., Reitsma, A., Dominic, A. et al. (2011).
with renal disease. Am. J. Obstet. Gynecol. 203 (1): 60. Pregnancy outcomes in women with chronic kidney
e1–60.e4. disease: a systematic review. Clin. J. Am. Soc. Nephrol. 6
32 Valentin, M., Muller, F., Beaujard, M.P. et al. (2015). (11): 2587–2598.
First-trimester combined screening for trisomy 21 in 46 Imbasciati, E., Gregorini, G., Cabiddu, G. et al. (2007).
women with renal disease. Prenat. Diagn. 35 (3): 244–248. Pregnancy in CKD stages 3 to 5: fetal and maternal
33 Wittfooth, S., Tertti, R., Lepantalo, M. et al. (2011). outcomes. Am. J. Kidney Dis. 49 (6): 753–762.
Studies on the effects of heparin products on pregnancy- 47 Smyth, A., Oliveira, G.H.M., Lahr, B.D. et al. (2010). A
associated plasma protein A. Clin. Chim. Acta 412 (3–4): systematic review and meta-analysis of pregnancy
376–381. outcomes in patients with systemic lupus erythematosus
34 Bramham, K., Seed, P.T., Lightstone, L. et al. (2016). and lupus nephritis. Clin. J. Am. Soc. Nephrol. 5 (11):
Diagnostic and predictive biomarkers for pre-eclampsia 2060–2068.
in patients with established hypertension and chronic 48 Bramham, K., Parnell, B., Nelson-Piercy, C. et al. (2014).
kidney disease. Kidney Int. 89 (4): 874–885. Chronic hypertension and pregnancy outcomes:
35 Buyon, J.P., Kim, M.Y., Guerra, M.M. et al. (2015). systematic review and meta-analysis. BMJ 348: g2301.
Predictors of pregnancy outcomes in patients with lupus: 49 Jones, D.C. and Hayslett, J.P. (1996). Outcome of
a cohort study. Ann. Intern. Med. 163 (3): 153–163. pregnancy in women with moderate or severe renal
36 Moroni, G., Doria, A., Giglio, E. et al. (2016). Maternal insufficiency. N. Engl. J. Med. 335 (4): 226–232.
outcome in pregnant women with lupus nephritis. A 50 Tong, A., Jesudason, S., Craig, J.C., and Winkelmayer,
prospective multicenter study. J. Autoimmun. 74: W.C. (2015). Perspectives on pregnancy in women with
194–200. chronic kidney disease: systematic review of qualitative
37 Piccoli, G.B., Gaglioti, P., Attini, R. et al. (2013). Pre- studies. Nephrol. Dial. Transplant. 30 (4): 652–661.
eclampsia or chronic kidney disease? The flow 51 Bar Oz, B., Hackman, R., Einarson, T., and Koren, G.
hypothesis. Nephrol. Dial. Transplant. 28 (5): 1199–1206. (2001). Pregnancy outcome after cyclosporine therapy
458 Pregnancy
during pregnancy: a meta-analysis. Transplantation 71 65 Jefferys, A., Wyburn, K., Chow, J. et al. (2008). Peritoneal
(8): 1051–1055. dialysis in pregnancy: a case series. Nephrology (Carlton)
52 Deshpande, N.A., James, N.T., Kucirka, L.M. et al. (2011). 13 (5): 380–383.
Pregnancy outcomes in kidney transplant recipients: a 66 Wyld, M.L., Clayton, P.A., Jesudason, S. et al. (2013).
systematic review and meta-analysis. Am. J. Transplant. Pregnancy outcomes for kidney transplant recipients.
11 (11): 2388–2404. Am. J. Transplant. 13 (12): 3173–3182.
53 Blom, K., Odutayo, A., Bramham, K., and Hladunewich, 67 Wyld, M.L., Clayton, P.A., Kennedy, S.E. et al. (2015).
M.A. (2017). Pregnancy and glomerular disease: a Pregnancy outcomes for kidney transplant recipients
systematic review of the literature with management with transplantation as a child. JAMA Pediatr. 169 (2):
guidelines. Clin. J. Am. Soc. Nephrol. 12 (11): 1862–1872. e143626.
54 Piccoli, G.B., Clari, R., Ghiotto, S. et al. (2013). Type 1 68 Mohammadi FA, Borg M, Gulyani A, McDonald SP,
diabetes, diabetic nephropathy, and pregnancy: a systematic Jesudason S. Pregnancy outcomes and impact of
review and meta-study. Rev. Diabetic Stud. 10 (1): 6–26. pregnancy on graft function in women after kidney
55 Liu, Y., Ma, X., Zheng, J. et al. (2016). A systematic review transplantation. Clin Transplant. 2017;31:e13089
and meta-analysis of kidney and pregnancy outcomes in Mohammadi FA, Borg M, Gulyani A, McDonald SP,
IgA nephropathy. Am. J. Nephrol. 44 (3): 187–193. Jesudason S. Pregnancy outcomes and impact of
56 Wu, J., Ma, J., Zhang, W.-H., and Di, W. (2018). pregnancy on graft function in women after kidney
Management and outcomes of pregnancy with or without transplantation. Clin Transplant. 2017;31:e13089
lupus nephritis: a systematic review and meta-analysis. 69 Bramham, K., Nelson-Piercy, C., Gao, H. et al. (2013).
Ther. Clin. Risk Manage. 14: 885–901. Pregnancy in renal transplant recipients: a UK national
57 Limardo, M., Imbasciati, E., Ravani, P. et al. (2010). cohort study. Clin. J. Am. Soc. Nephrol. 8 (2): 290–298.
Pregnancy and progression of IgA nephropathy: results of 70 Vannevel, V., Claes, K., Baud, D. et al. (2018).
an Italian multicenter study. Am. J. Kidney Dis. 56 (3): Preeclampsia and long-term renal function in women
506–512. who underwent kidney transplantation. Obstet. Gynecol.
58 Moroni, G., Doria, A., Giglio, E. et al. (2016). Fetal 131 (1): 57–62.
outcome and recommendations of pregnancies in lupus 71 Sibanda, N., Briggs, J.D., Davison, J.M. et al. (2007).
nephritis in the 21st century. A prospective multicenter Pregnancy after organ transplantation: a report from the
study. J. Autoimmun. 74: 6–12. UK transplant pregnancy registry. Transplantation 83
59 Bramham, K. and Rajasingham, D. (2012). Pregnancy in (10): 1301–1307.
diabetes and kidney disease. J. Ren. Care 38 (Suppl 1): 72 Majak, G.B., Reisaeter, A.V., Zucknick, M. et al. (2017).
78–89. Preeclampsia in kidney transplanted women; outcomes
60 Klemetti, M.M., Laivuori, H., Tikkanen, M. et al. (2015). and a simple prognostic risk score system. PLoS One 12
Obstetric and perinatal outcome in type 1 diabetes (3): e0173420.
patients with diabetic nephropathy during 1988-2011. 73 Stoumpos, S., McNeill, S.H., Gorrie, M. et al. (2016).
Diabetologia 58 (4): 678–686. Obstetric and long-term kidney outcomes in renal
61 Bell, R., Glinianaia, S.V., Tennant, P.W.G. et al. (2012). transplant recipients: a 40-yr single-center study. Clin.
Peri-conception hyperglycaemia and nephropathy are Transplant. 30 (6): 673–681.
associated with risk of congenital anomaly in women 74 Rose, C., Gill, J., Zalunardo, N. et al. (2016). Timing of
with pre-existing diabetes: a population-based cohort pregnancy after kidney transplantation and risk of
study. Diabetologia. 936–947 (2012). allograft failure. Am. J. Transplant. 16 (8): 2360–2367.
62 Hladunewich, M.A., Hou, S., Odutayo, A. et al. (2014). 75 McKay, D.B. and Josephson, M.A. (2006). Pregnancy in
Intensive hemodialysis associates with improved recipients of solid organs–effects on mother and child. N.
pregnancy outcomes: a Canadian and United States Engl. J. Med. 354 (12): 1281–1293.
cohort comparison. J. Am. Soc. Nephrol. 25 (5): 76 Rahamimov, R., Ben-Haroush, A., Wittenberg, C. et al.
1103–1109. (2006). Pregnancy in renal transplant recipients: long-
63 Saliem, S., Patenaude, V., and Abenhaim, H.A. (2016). term effect on patient and graft survival. A single-center
Pregnancy outcomes among renal transplant recipients experience. Transplantation 81 (5): 660–664.
and patients with end-stage renal disease on dialysis. J. 77 Kwek, J.L., Tey, V., Yang, L. et al. (2015). Renal and
Perinat. Med. 44 (3): 321–327. obstetric outcomes in pregnancy after kidney
64 Luders, C., Castro, M.C.M., Titan, S.M. et al. (2010). transplantation: twelve-year experience in a Singapore
Obstetric outcome in pregnant women on long-term transplant center. J. Obstet. Gynaecol. Res. 41 (9):
dialysis: a case series. Am. J. Kidney Dis. 56 (1): 77–85. 1337–1344.
Reference 459
8 Hebral, A.L., Cointault, O., Connan, L. et al. (2014).

7 92 Levy, R.A., Vilela, V.S., Cataldo, M.J. et al. (2001).
Pregnancy after kidney transplantation: outcome and Hydroxychloroquine (HCQ) in lupus pregnancy:
anti-human leucocyte antigen alloimmunization risk. double-blind and placebo-controlled study. Lupus 10 (6):
Nephrol. Dial. Transplant. 29 (9): 1786–1793. 401–404.
79 Fitzpatrick, A., Mohammadi, F., and Jesudason, S. (2016). 93 Leroux, M., Desveaux, C., Parcevaux, M. et al. (2015).
Managing pregnancy in chronic kidney disease: Impact of hydroxychloroquine on preterm delivery and
improving outcomes for mother and baby. Int. J. Womens intrauterine growth restriction in pregnant women with
Health 8: 273–285. systemic lupus erythematosus: a descriptive cohort
80 Wiles, K.S., Bramham, K., Vais, A. et al. (2015). Pre- study. Lupus 24 (13): 1384–1391.
pregnancy counselling for women with chronic kidney 94 Kaplan, Y.C., Ozsarfati, J., Nickel, C., and Koren, G.
disease: a retrospective analysis of nine years’ experience. (2016). Reproductive outcomes following
BMC Nephrol. 16: 28. hydroxychloroquine use for autoimmune diseases: a
81 Bramham, K. and Lightstone, L. (2012). Pre-pregnancy systematic review and meta-analysis. Br. J. Clin.
counseling for women with chronic kidney disease. J. Pharmacol. 81 (5): 835–848.
Nephrol. 25 (4): 450–459. 95 Izmirly, P.M., Saxena, A., Kim, M.Y. et al. (2011).
82 Kim, M., Rostas, S., and Gabardi, S. (2013). Maternal and fetal factors associated with mortality and
Mycophenolate fetal toxicity and risk evaluation and morbidity in a multi-racial/ethnic registry of anti-SSA/
mitigation strategies. Am. J. Transplant. 13 (6): Ro-associated cardiac neonatal lupus. Circulation 124
1383–1389. (18): 1927–1935.
83 Hoeltzenbein, M., Elefant, E., Vial, T. et al. (2012). 96 Izmirly, P.M., Costedoat-Chalumeau, N., Pisoni, C.N.
Teratogenicity of mycophenolate confirmed in a prospective et al. (2012). Maternal use of hydroxychloroquine is
study of the European Network of Teratology Information associated with a reduced risk of recurrent
Services. Am. J. Med. Genet. A 158A (3): 588–596. anti-SSA/Ro-antibody-associated cardiac
84 Bateman, B.T., Patorno, E., Desai, R.J. et al. (2017). manifestations of neonatal lupus. Circulation
Angiotensin-converting enzyme inhibitors and the risk of 126 (1): 76–82.
congenital malformations. Obstet. Gynecol. 129 (1): 97 Jaeggi, E., Laskin, C., Hamilton, R. et al. (2010). The
174–184. importance of the level of maternal anti-Ro/SSA
85 Li, D.-K., Yang, C., Andrade, S. et al. (2011). Maternal antibodies as a prognostic marker of the development of
exposure to angiotensin converting enzyme inhibitors in cardiac neonatal lupus erythematosus a prospective
the first trimester and risk of malformations in offspring: study of 186 antibody-exposed fetuses and infants. J.
a retrospective cohort study. BMJ 343: d5931. Am. Coll. Cardiol. 55 (24): 2778–2784.
86 Magee, L.A., von Dadelszen, P., Rey, E. et al. (2015). 98 Abalos, E., Duley, L., and Steyn, D. (2014).
Less-tight versus tight control of hypertension in Antihypertensive drug therapy for mild to moderate
pregnancy. N. Engl. J. Med. 372 (5): 407–417. hypertension during pregnancy. Cochrane Database
87 Duley, L., Henderson-Smart, D.J., Meher, S., and King, Syst. Rev. (2): CD002252.
J.F. (2007). Antiplatelet agents for preventing pre- 99 National Institute for Health, Excellence, Care (2011).
eclampsia and its complications. Cochrane Database Syst. Hypertension in pregnancy: the management of
Rev. (2): CD004659. hypertensive disorders in pregnancy – NICE clinical
88 Roberge, S., Nicolaides, K.H., Demers, S. et al. (2013). guidelines.
Prevention of perinatal death and adverse perinatal 100 The American College of Obstetricians and
outcome using low-dose aspirin: a meta-analysis. Gynaecologists (2013). Hypertension in pregnancy.
Ultrasound Obstet. Gynecol. 41 (5): 491–499. Report of the American College of Obstetricians and
89 Cabiddu, G., Castellino, S., Gernone, G. et al. (2015). Best Gynecologists’ Task Force on Hypertension in
practices on pregnancy on dialysis: the Italian Study Group Pregnancy. Obstet. Gynecol. 122 (5): 1122–1131. doi:
on Kidney and Pregnancy. J. Nephrol. 28 (3): 279–288. 10.1097/01.AOG.0000437382.03963.88. PMID: 24150027.
90 Wu, M., Wang, D., Zand, L. et al. (2016). Pregnancy 101 Henderson, J.T., Whitlock, E.P., O’Connor, E. et al.
outcomes in autosomal dominant polycystic kidney (2014). Low-dose aspirin for prevention of morbidity
disease: a case-control study. J. Matern. Fetal Neonatal and mortality from preeclampsia: a systematic evidence
Med. 29 (5): 807–812. review for the U.S. Preventive Services Task Force. Ann.
91 Vazquez, J.C. and Abalos, E. (2011). Treatments for Intern. Med. 160 (10): 695–703.
symptomatic urinary tract infections during pregnancy. 102 Cooper, W.O., Hernandez-Diaz, S., Arbogast, P.G. et al.
Cochrane Database Syst. Rev. (1): CD002256. (2006). Major congenital malformations after first-
460 Pregnancy
trimester exposure to ACE inhibitors. N. Engl. J. Med. 111 Zheng, S., Easterling, T.R., Umans, J.G. et al. (2012).
354 (23): 2443–2451. Pharmacokinetics of tacrolimus during pregnancy. Ther.
103 Piccoli, G.B., Cabiddu, G., Attini, R. et al. (2015). Drug Monit. 34 (6): 660–670.
Pregnancy in chronic kidney disease: questions and 112 Hebert, M.F., Zheng, S., Hays, K. et al. (2013).
answers in a changing panorama. Best Pract. Res. Clin. Interpreting tacrolimus concentrations during
Obstet. Gynaecol. 29 (5): 625–642. pregnancy and postpartum. Transplantation 95 (7):
104 Rolnik, D.L., Wright, D., Poon, L.C. et al. (2017). Aspirin 908–915.
versus placebo in pregnancies at high risk for preterm 113 Framarino-dei-Malatesta, M., Derme, M., Napoli, A.
preeclampsia. N. Engl. J. Med. 377 (7): 613–622. et al. (2014). Placental, lipid, and glucidic effects of
105 LeFevre, M.L. (2014). Low-dose aspirin use for the mammalian target of rapamycin inhibitors: impact on
prevention of morbidity and mortality from fetal growth and metabolic disorders during pregnancy
preeclampsia: U.S. Preventive Services Task Force after solid organ transplantation. Transplant. Proc. 46
recommendation statement. Ann. Intern. Med. 161 (11): (7): 2254–2258.
819–826. 114 Webster, P., Wardle, A., Bramham, K. et al. (2014).
106 Bothwell, T.H. (2000). Iron requirements in pregnancy Tacrolimus is an effective treatment for lupus nephritis
and strategies to meet them. Am. J. Clin. Nutr. 72 (1 in pregnancy. Lupus 23 (11): 1192–1196.
Suppl): 257S–264S. 115 Asamiya, Y., Otsubo, S., Matsuda, Y. et al. (2009). The
107 Bramham, K., Chusney, G., Lee, J. et al. (2013). importance of low blood urea nitrogen levels in
Breastfeeding and tacrolimus: serial monitoring in pregnant patients undergoing hemodialysis to optimize
breast-fed and bottle-fed infants. Clin. J. Am. Soc. birth weight and gestational age. Kidney Int. 75 (11):
Nephrol. 8 (4): 563–567. 1217–1222.
108 Kim, H., Jeong, J.C., Yang, J. et al. (2015). The optimal 116 Holley, J.L. and Reddy, S.S. (2003). Pregnancy in dialysis
therapy of calcineurin inhibitors for pregnancy in patients: a review of outcomes, complications, and
kidney transplantation. Clin. Transplant. 29 (2): management. Semin. Dial. 16 (5): 384–388.
142–148. 117 Hladunewich, M. and Schatell, D. (2016). Intensive
109 Blume, C., Pischke, S., von Versen-Hoynck, F. et al. dialysis and pregnancy. Hemodial. Int. 20 (3): 339–348.
(2014). Pregnancies in liver and kidney transplant 118 Alkhunaizi, A., Melamed, N., and Hladunewich, M.A.
recipients: a review of the current literature and (2015). Pregnancy in advanced chronic kidney disease
recommendation. Best Pract. Res. Clin. Obstet. Gynaecol. and end-stage renal disease. Curr. Opin. Nephrol.
28 (8): 1123–1136. Hypertens 24 (3): 252–259.
110 Deshpande, N.A., Coscia, L.A., Gomez-Lobo, V. et al. 119 McKay, D.B., Josephson, M.A., Armenti, V.T. et al.
(2013). Pregnancy after solid organ transplantation: a (2005). Reproduction and transplantation: report on the
guide for obstetric management. Rev. Obstet. Gynecol. 6 AST consensus conference on reproductive issues and
(3–4): 116–125. transplantation. Am. J. Transplant. 5 (7): 1592–1599.
461
28
ANCA-associated Vasculitis
Maria Prendecki and Stephen McAdoo
Imperial College London, London, UK
Introduction and Definitions immune, and in 1982 the first report of circulating ANCA
associated with a pauci-immune glomerulonephritis (GN)
The antineutrophil cytoplasmic antibody (ANCA)-associated was published [9, 10]. The two major targets of ANCA have
vasculitides (AAVs) are a group of systemic necrotizing vas- been shown to be myeloperoxidase (MPO) and proteinase 3
culitides which affect small to medium-sized blood vessels. (PR3), enzymes expressed in the granules and lysosomes of
There is a spectrum of organ involvement, including the kid- neutrophils and monocytes [11–13]. GPA is more com-
ney, lungs, nerves, gut, and ear, nose, and throat (ENT). AAVs monly associated with PR3-ANCA positivity and MPA with
are classified on the basis of clinical and pathological features MPO-ANCA, although there is considerable overlap and
into granulomatosis with polyangitis (GPA), microscopic around 10% of patients with MPA or GPA will be ANCA
polyangitis (MPA), eosinophilc granulomatosis with polyan- negative. In EGPA around 40% of patients are reported to be
gitis (EGPA), and renal limited vasculitis (RLV). The 2012 ANCA positive, and vasculitic features such as GN are more
Chapel Hill Consensus Conference (CHCC) nomenclature of
vasculitis defines AAVs as necrotising vasculitis which pre-
dominately affects small vessels such as capillaries, venules, Table 28.1 Definitions for AAV.
arterioles, and small arteries. There must be few or no
immune deposits present [5]. The CHCC definitions of GPA, Microscopic Necrotizing vasculitis, with few or no
polyangiitis (MPA) immune deposits, predominantly
MPA, and EGPA are summarized in Table 28.1. The CHCC is affecting small vessels. Necrotizing
a nomenclature system rather than a diagnostic classifica- glomerulonephritis and pulmonary
tion. The 1990 American College of Rheumatology (ACR) capillaritis are very common. No
criteria provides diagnostic criteria for GPA and EGPA but not granulomatous inflammation is seen.
MPA; but ANCAs were not included as part of the diagnostic Granulomatosis Necrotizing granulomatous
with polyangiitis inflammation usually involving the
criteria. Based on the results of the Diagnostic and Classification
(Wegener’s) (GPA) upper and lower respiratory tract, and
Criteria for Vasculitis (DCVAS) study, new criteria for classi- necrotizing vasculitis affecting
fication of GPA, EGPA, and MPA to standardize the classifi- predominantly small to medium
cation of these diseases to support clinical research were vessels. Glomerulonephritis is
common.
co-jointly endorsed by the American College of Rheumatology
(ACR)/European Alliance of Associations for Rheumatology Eosinophilic Eosinophillic necrotizing
granulomatosis with granulomatous inflammation, often
(EULAR) have recently been published [6,7,8]. These polyangiitis involving the respiratory tract, and
classification criteria are shown in Table 28.2. (Churg-Strauss) necrotizing vasculitis predominantly
Prior to 1979 it was assumed that all cases of rapidly pro- (EGPA) affecting small to medium vessels.
gressive glomerulonephritis (RPGN) were caused by Often associated with asthma and
eosinophilia.
immune complex deposition or circulating antiglomerular
basement membrane (anti-GBM) antibodies. Stilmant et al. Source: Adapted from the Chapel Hill 2012 consensus on
recognized that in fact many of these cases were pauci- nomenclature of vasculitis [5].
462 ANCA-associated Vasculitis
Table 28.2 ACR/EULAR Classification Criteria for ANCA-Associated Vasculitides (AAVs).
Eosinophilic granulomatosis with

Granulomatosis with polyangiitis (GPA) Microscopic polyangiitis (MPA) polyangiitis (EGPA)
Bloody nasal discharge, nasal crusting, or +3 Perinuclear ANCA or +6 Maximum eosinophil Count +5
sino-nasal congestion anti-myeloperoxidase ANCA >1 x 109/liter
positivity
Cartilaginous involvement (Any of inflamed +2 Pauci-immune +3 Obstructive airway disease +3
ear or nose cartilage, hoarse voice/stridor, glomerulonephritis
endobronchial involvement or saddle nose)
Conductive or sensorineural hearing loss +1 Lung fibrosis or interstitial +3 Nasal polyps +3
lung disease
Cytoplasmic antineutrophilic antibody +5 Sino-nasal symptoms or signs +3 Cytoplasmic ANCA or –3
(ANCA) or anti-proteinase 3 ANCA anti-proteinase 3 ANCA
positivity positivity
Pulmonary nodules, mass, or cavitation of +2 Cytoplasmic ANCA or –1 Mononeuritis multiplex/ +1
chest on imaging anti-proteinase 3 ANCA motor neuropathy due to
positivity radiculopathy
Granuloma or giant cells on biopsy +2 Eosinophil Count >1 x 109/ –4 Extravascular eosinophilic +2
liter predominant inflammation
Inflammation or consolidation of the nasal/ +1 Hematuria –1
paranasal sinuses on imaging
Pauci-immune glomerulonephritis +1
Perinuclear ANCA or antimyeloperoxidase –1
ANCA positivity
Eosinophil Count >1 x 109/liter –4
Total score of >5 for diagnosis of GPA Total score of >5 for diagnosis of MPA Total score of >6 for diagnosis of EGPA
Sensitivity 93% (95%-CI 87-96%) Sensitivity 91% (95%-CI 85-95%) Sensitivity 85% (95%-CI 77-91%)
Specificity 94% (95%-CI 89-97%) Specificity 94% (95%-CI 92-96%) Specificity 99% (95%-CI 98-100%)
Adapted from reference 6-8.
common in the patients with ANCA positivity [14]. This and Japan showed a similar incidence of AAV in the two
chapter will focus on the management of MPA and GPA as populations, but the majority of the Japanese population
there are several differences in the pathogenesis, features, had MPA and GPA was extremely rare [20]. Within France,
and management of EGPA. A summary of the quality of the the prevalence of AAV was twice as high in those with a
evidence and strength of the recommendation for treat- European background than in those of non-European
ment of the AAVs is shown in Table 28.3. ethnicity [21]. Studies in Sweden and the UK have both
reported increasing incidence of AAV since the early 1980s
although it is not clear whether this represents a true
Epidemiology
increase in disease or reflects greater awareness of the con-
AAVs are rare conditions with reported annual incidence dition and the more widespread availability of ANCA test-
between 13 and 20 cases per million individuals in ing [22, 23].
Europe [15]. There is a slight male preponderance and There is a genetic basis for AAV, and this may underlie
peak age of onset of AAV has been reported as age 55–64, some of the reported population differences. There are two
65–74, and more than 75 years in different studies [16–18]. published genome-wide association studies (GWAS), both
AAV is very rare in childhood. Within Europe there may be carried out in patients of Northern European ancestry,
a geographical variation in AAV subtype, with reported which identified both MHC and non-MHC genetic associa-
higher incidence of GPA in Northern Europe and MPA in tions with AAV [24, 25]. There is a stronger genetic associa-
the south [19]. This trend to geographic variation was also tion related to the antigenic specificity of ANCA than there
present in studies conducted in the United States, where is to clinical phenotype, and MPO-ANCA and PR3-ANCA
there is higher prevalence of GPA in northern states and may define differing diseases. In PR3-ANCA related disease
more MPA in southern states. Studies comparing the UK there are strong associations with PRTN3, the gene which
Introduction and Definition 463
Table 28.3 Treatment for AAV.
Remission induction
Cyclophosphamide and Oral or IV cyclophosphamide should be used for induction therapy in High (IV vs. 1A [1, 2]
corticosteroids patients with severe or organ-threatening AAV. Pulsed treatment oral-low)
with IV cyclophosphamide is equivalent to oral treatment for
remission induction but may result in increased risk of relapse.
Rituximab and Rituximab is an alternative to cyclophosphamide, particularly in High 1B [1, 2]
corticosteroids those who are unable or unwilling to receive cyclophosphamide such
as patients wishing to preserve fertility. Some guidelines restrict the
use of rituximab to those unable to receive cyclophosphamide.
Rituximab may be superior to cyclophosphamide for remission
induction in patients presenting with relapsing disease.
Adjunctive plasma There is insufficient evidence for routine use of adjunctive plasma Low 2D [1, 2]
exchange in renal vasculitis exchange but current guidelines recommend it for patients with
RPGN, creatinine>500 μmol/l at diagnosis or those with pulmonary
hemorrhage.
Remission maintenance
Azathioprine Azathioprine is as effective as oral cyclophosphamide for remission Moderate 1B [1]
maintenance, but is safer with fewer adverse events.
Rituximab Rituximab is as effective as azathioprine in preventing major relapse Moderate 1B [1]
with similar levels of adverse events. Some guidelines include
rituximab as an alternative therapy for remission maintenance.
MMF MMF is less effective than azathioprine in preventing relapse but may Moderate 1B [1]
be used as an alternative in patients who are intolerant or allergic to
azathioprine. Guidelines vary as to whether MMF or methotrexate
should be second line.
Methotrexate Methotrexate can be used for remission maintenance in patients Moderate 1B/1C [1, 2]
unable to tolerate azathioprine. There are similar relapse rates in
patients treated with methotrexate and those treated with
azathioprine but more severe adverse events with methotrextate.
Methotrexate should not be used in those with eGFR<30 ml/min and
dose should be adjusted in patients with eGFR<60 ml/min.
Guidelines vary as to whether MMF or methotrexate should be
second line.
Co-trimoxazole There is no evidence to support the use of co-trimoxazole for Low 2B [2]
remission maintenance.
GRADE assessment of the evidence [3, 4]: High: This research provides a very good indication of the likely effect. The likelihood that the effect
will be substantially different is low. Moderate: This research provides a good indication of the likely effect. The likelihood that the effect will be
substantially different is moderate. Low: This research provides some indication of the likely effect. However, the likelihood that it will be
substantially different is high. Very low: This research does not provide a reliable indication of the likely effect. The likelihood that the effect will
be substantially different is very high.
AAVs, antineutrophil cytoplasmic antibody-associated vasculitides; eGFR (estimated GFR), ; IV, intravenous; MMF (Mycophenolate mofetil);
RPGN, rapidly progressive glomerulonephritis.
encodes PR3, SERPINA1, the gene encoding α1-antitrypsin, correlates with relapse [27]. This may be due to molecular
and HLA-DP. MPO-ANCA-associated disease is linked to mimicry whereby antibodies against microbial antigens
HLA-DQ polymorphisms. There are several other HLA and cross-react with self antigen [28]. Drugs, including propylth-
non-HLA genes which have been associated with AAV [26]. iouracil, hydralazine, and penicillamine, are associated
There are also several reported environmental associa- with positive ANCA serology, and less commonly may be
tions with AAV such as toxins, drugs, and infections. associated with clinical vasculitis [29, 30]. An association
In patients with ENT disease and anti-PR3 ANCA, infection with levamisole-contaminated cocaine, positive ANCA
may precede disease relapse and nasal carriage of Staphylococci serology, and clinical disease resembling GPA has also been
reported [31]. Environmental triggers such as silica and C4-deficient mice were not protected, suggesting a role
heavy metal exposure have also been implicated as triggers. for the alternative and not the classical pathway [48].
After the 1995 Kobe earthquake in Japan there was a regional Mice with knock-in of the human C5a receptor (C5aR)
threefold increase in disease incidence, suggesting that air and MPO-AAV respond to treatment with an antagonist
pollutants can play a role in the etiology of disease [32]. In of human C5aR (CCX168; avacopan) and have decreased
some studies, but not others, an association with farming disease severity. A phase II trial has shown avacopan was
has been reported [33, 34]. noninferior to high-dose glucocorticoids when used in
combination with rituximab or cyclophosphamide for
induction of remission [49].
Pathogenesis
The mechanisms underlying the development of ANCA
There are several clinical observations which support a role are less well understood. One suggested mechanism is
for ANCA being directly pathogenic. Most patients with that of molecular mimicry. An atypical ANCA, antihu-
AAV have circulating ANCA and in some studies ANCA man lysosome-associated membrane protein-2 (anti-
titer has been shown to predict relapse [35]. There is a LAMP-2) antibody, has been identified in patients with
reported case of neonatal AAV following placental transfer pauci-immune GN. It has 100% sequence homology with
of MPO-ANCA, resulting in renal disease and pulmonary FimH, a protein present on gram negative bacteria, and
hemorrhage shortly after birth [36]. Another case report when rats are immunized with FimH they develop GN
describes no clinical disease despite maternal-fetal transfer and antibodies which cross-react with human and rat
of MPO-ANCA [37]. Removal of antibodies using plasma- LAMP-2 [28]. Another potential mechanism is of anti-
pheresis and use of B cell targeted therapies have both been idiotype interactions to peptides complementary to the
shown to be effective treatments for AAV [38–40]. These auto-antigen, such as complementary PR3 peptides which
findings are supported by experimental work from both are homologous to some Staphylococcus aureus pep-
in vitro experiments and in vivo models of AAV. Passive tides [50]. There is also a role for disordered T cell
transfer models in mice, where anti-MPO antibodies are responses: patients with AAV have defective Treg sup-
raised in MPO-deficient mice and then transferred into pressor function and increased frequency of a T cell sub-
naïve recipients, results in the recipient mice developing a set which is resistant to the suppressive effects of
severe vasculitis with crescentic GN and pulmonary hem- Tregs [51, 52]. Regulatory B cells have also been shown to
orrhage [41]. in vitro, neutrophils which have been primed be decreased in patients with AAV [53, 54].
with TNF(Tumor necrosis factor)α, lipopolysaccharide
(LPS), or complement (C5a) can be activated by ANCA,
Clinical Features
causing degranulation, ROS production, and endothelial
cell damage [42–44]. ANCA are also mediators of NETosis, AAV is a multisystem disease and clinical presentation var-
a form of neutrophil cell death which occurs with release ies widely in both severity and organ involvement. Patients
of neutrophil extracellular traps (NETs). Patients with AAV often have a nonspecific prodromal illness which may pre-
have increased levels of NETs in their circulation and NETs cede diagnosis by several weeks or months, with features
have been shown at sites of tissue damage in AAV [45]. such as fever, anorexia lethargy, arthralgia, or weight loss.
Although there is much evidence for the pathogenicity of Patients may have asymptomatic urinary abnormalities
ANCA and ANCA-induced activation of neutrophils, early in disease course (or in those with milder disease) but
ANCA are not always pathogenic. Ten percent of patients renal involvement typically presents as a RPGN. In MPA
with AAV are ANCA negative, and ANCA can either per- 90% of patients will have evidence of renal disease, com-
sist in clinical remission or recur without evidence of pared to 80% of those with GPA and around 40% of those
relapse. Natural anti-MPO antibodies have been identified with EGPA [55]. In GPA extrarenal features of disease are
in healthy individuals and these are of lower avidity and common, with 90% of patients having lung or ENT involve-
titer than antibodies in patients with AAV [46]. One study ment. Patients with MPA often have more severe renal
of MPO-ANCA has shown that anti-MPO antibodies from involvement. Other organ systems involved include the
healthy individuals had different target epitopes to anti- skin, nervous system, eyes, musculoskeletal system, and
MPO antibodies from patients with AAV [47]. gastrointestinal tract [55]. In EGPA patients typically have
Complement is increasingly recognized to play a role in features of asthma and nasal polyps with peripheral blood
the pathogenesis of AAV. Mice deficient in C5 or depleted eosinophilia. In patients with RPGN, as disease progresses,
of complement did not develop disease in the passive features of overt renal failure will develop and patients may
antibody transfer model of AAV described above. have symptoms related to salt and water overload such as
Table 28.4 Summary of diagnostic tests in patients with suspected AAV.
Investigation Results/comments
FBC Leucocytosis
Thrombocytosis
Normochromic normocytic anemia
CRP, ESR Often raised but nonspecific
Renal function Raised urea and creatinine in patients with RPGN
Complement C3, C4 Normal or elevated in AAV. Low C3/C4 may indicate bacterial infection such as
endocarditis or immune complex GN. Isolated low C3 is seen in postinfectious
GN or C3 glomerulopathy.
Immunoglobulins Polyclonal hypergammaglobulinemia may be present. Prior to commencing
cytotoxic therapies, hypogammaglobulinemia should be excluded.
Viral serology: HBsAg, HIV, HCV Ab Prior to initiating immunosuppressive therapies
Blood cultures To investigate for the presence of bacterial endocarditis in patients with ANCA
positivity and systemic symptoms
ANCA IIF or immunoassay
anti-GBM, ANA, dsDNA To investigate for other causes of RPGN
Urine dipstick Hematuria and proteinuria. Leukocyturia often represents the presence of
nonspecific inflammatory cells.
Urine microscopy Needs to be performed by a skilled operator on fresh urine. Red cell casts are
seen in glomerulonephritis.
Urine PCR Low-grade proteinuria is common. In RPGN decreased glomerular blood flow
limits protein filtration.
Renal biopsy Pauci-immune, necrotizing, creacsentic GN
Skin biopsy Leukocytoclastic vasculitis
ENT or lung biopsy Granuloma often seen
Nerve biopsy Necrotizing vasculitis of associated small and medium blood vessels
Pulmonary function testing Raised KCO in 25% of patients with pulmonary hemorrhage
AAV, ANCA-associated vasculitis; ANA, anti-nuclear antibody; ANCA, Anti-neutrophil cytoplasmic antibody; anti-GBM, anti-glomerular
basement antibody; C3, complement component 3; C4, complement component 4; CRP, C-Reactive Protein; dsDNA, double stranded DNA;
ENT, Ear-Nose-Throat; ESR, Erythrocyte Sedimentation Ratio; FBC, Full blood count; GN, Glomerulonephritis; HBsAg, Hepatitis B surface
antigen ; HCV Ab, Hepatitis C Virus antibody; HIV, Human immunodeficiency virus; IIF, indirect immunofluorescence; RPGN, rapidly
progressive glomerulonephritis.
breathlessness, cough, orthopnea, and peripheral oedema performed. The diagnostic tests used to diagnose and eval-
or to uremia such as itch, anorexia, and fatigue. Relapse of uate AAV are summarized below and in Table 28.4.
disease is common and occurs in 30–50% of patients.
Hematological and Biochemical Tests
Diagnostic Tests
Full blood count may show leucocytosis, thrombocytosis,
The usual screening test for ANCA is to perform indirect or normochromic normocytic anemia. Raised C reactive
immunofluorescence (IIF) on neutrophils; positive sam- protein and erythrocyte sedimentation rate are often pre-
ples are then used in antigen-specific immunoassays to sent. Raised urea and creatinine will be present in those
determine ANCA specificity. Due to the increasing availa- with RPGN. Urine dipstick may show hematuria, proteinu-
bility of high-quality immunoassays for MPO- and PR3- ria or leukocyturia, and urine microscopy should be used
ANCA, the current consensus is that these tests can be to confirm the presence of hematuria and to identify red
used for screening without absolute need for IIF [56]. In cell casts. Urinary protein leak should be quantified using
patients with renal involvement a renal biopsy is usually a spot urine protein:creatinine ratio (uPCR); proteinuria is
often low grade (<100 mg/mmol) and usually will be sub- Antigen Specific Tests
nephrotic (<300 mg/mmol) [57].
Several immunoassays are available, including enzyme
immune-assays (EIAs), bead-based multiplex assays,
Imaging fluorescent enzyme immune assays (FEIA), and immu-
noblot assays. These assays usually detect IgG and
Chest imaging (radiograph or computed tomography)
provide quantitative results. It was previously recom-
should be undertaken in all patients with respiratory symp-
mended that a positive IIF should be followed by an
toms such as cough, dyspnea, or hemoptysis, and may show
antigen-specific test, however recent studies have
pulmonary infiltrates suggestive of hemorrhage, nodules or
shown that immunoassays are more sensitive and spe-
cavitation. There should be a low threshold for chest screen-
cific than IIF [60]. Additionally, IIF can be labor inten-
ing in patients in whom a diagnosis of AAV is suspected.
sive and automated interpretation is not available
Renal imaging with ultrasound should be performed to
worldwide. Current guidelines recommend immunoas-
exclude structural kidney diseases prior to renal biopsy, but
say testing for PR3 and MPO-ANCA for diagnosis of
in patients with AAV will usually be normal or show non-
AAV without need for IIF [56]. In patients with a high
specific features of acute kidney disease such as changes in
index of suspicion for AAV, IIF should be carried out
echogenicity. Other imaging should be directed toward
despite negative immunoassay results. Anti-MPO or
symptoms such as imaging of the sinuses, gut, retro-orbital
anti-PR3 antibodies have good positive predictive val-
space or nervous system.
ues for the diagnosis of AAV but have been reported in
other inflammatory conditions and are not diagnostic.
Physiological Testing A histological diagnosis is required for definitive diag-
nosis where possible [61].
Lung function testing may show an increased diffusing
capacity for CO (KCO) in pulmonary hemorrhage,
whereas it should be low in other causes of pulmonary Histological Tests
infiltrates such as pulmonary oedema. However, raised
KCO is only present in 25% of patients with pulmonary In patients with clinical evidence of renal disease such as
hemorrhage and normal or low KCO does not have a good urinary abnormalities or deranged renal function, renal
negative predictive value [58]. Patients with pulmonary biopsy is the gold standard for diagnosis. The typical pattern
hemorrhage may also be too unwell to undergo testing. of glomerular injury seen on light microscopy is of necrotiz-
Other physiological tests will be guided by the patient’s ing and crescentic GN: immunofluorescence will show
clinical presentation, such as nerve conduction studies or minimal or negative staining for immunoglobulin or com-
echocardiography. plement, and electron microscopy may show microthrombi
or degranulation of neutrophils but no or minimal immune
deposits [62, 63]. A histopathological classification system
ANCA Testing has been developed for AAV with four categories of disease:
Indirect Immunofluorescence focal, where <50% of glomeruli are affected, crescentic,
IIF is usually performed on ethanol-fixed neutrophils, with where >50% of glomeruli contain cellular crescents, scle-
two patterns of staining seen in positive samples: cytoplas- rotic, where >50% of glomeruli are globally sclerosed, and
mic (c-ANCA), which is associated with anti-PR3 antibod- mixed, where normal, crescentic, and sclerotic glomeruli are
ies, and perinuclear (p-ANCA), which is typically seen all present and >50% of glomeruli are affected [63]. The clas-
with anti-MPO antibodies. An atypical pattern is seen sification system has been shown to predict renal outcomes
rarely and comprises cytoplasmic, perinuclear, and nuclear in AAV with patients with sclerotic lesions least likely to
staining; it is associated with several antigen specifici- recover renal function. [63] Biopsy of other sites of disease
ties [59]. The perinuclear pattern seen with MPO-ANCA may also be undertaken, such as skin, ENT, lung, or nerve,
arises due to the process of ethanol fixation, which leads to particularly if there is diagnostic uncertainty.
MPO dissolving from the neutrophil granules and attach-
ing to the nucleus. When formalin-fixed neutrophils are
Treatment
used this does not occur and a cytoplasmic pattern is seen
with both MPO and PR3-ANCA. Atypical ANCA IIF can be Modern immunosuppression regimes have transformed
seen in other diseases such as inflammatory bowel disease, patient outcomes in patients with AAV. Prior to the 1960s,
HIV, bacterial endocarditis, or related to drug-induced mortality at 1 year was as high as 80%, whereas current
autoimmunity. A range of atypical target antigens have studies show 1 year survival of around 90% [64]. Usually an
been described, including lactoferrin or elastase. initial period of intense immunosuppression is used to
induce remission followed by a period of sustained low- Systemic Vasculitis (CYCLOPS) trial, which included 149
dose treatment to prevent relapse. newly diagnosed patients with either GPA or MPA and renal
involvement, but not life-threatening disease Table 28.7 [65].
It showed no difference in time to remission between the two
Remission Induction
groups, but lower cumulative cyclophosphamide doses in the
The most widely used first-line treatment regimen for IV arm with lower rates of leukopenia [65]. Long-term fol-
induction of remission in patients with AAV and severe low-up of this study showed higher relapse rates in the group
disease is cyclophosphamide Table 28.1 and high-dose glu- treated with IV cyclophosphamide but no difference in
cocorticoids with the addition of plasma exchange for ESRD, morbidity, or mortality [66]. Of the available pub-
those with severe renal impairment or RPGN. An example lished guidelines, European League Against Rheumatism
induction treatment regimen is shown in Table 28.5. (EULAR)/European Renal Association – European Dialysis
and Transplant Association (ERA-EDTA) and British Society
of Rheumatology (BSR) guidelines favor IV over oral cyclo-
Cyclophosphamide
phosphamide, and Kidney Disease Improving Global
Several clinical trials have studied the optimum route, dose, Outcomes (KDIGO) recommend either oral or pulsed IV
and duration of treatment with cyclophosphamide. There are cyclophosphamide at clinician discretion [1, 2, 67] Table 28.6.
four published studies of oral versus pulsed intravenous (IV) Only one study has investigated the duration of treatment
cyclophosphamide, the largest and most recent of which with IV cyclophosphamide (without subsequent mainte-
is the Randomized Trial of Daily Oral vs Pulse nance therapy) and it showed no difference in mortality,
Cyclophosphamide as Therapy for ANCA-Associated duration of remission, or infection rate between six and 12
Table 28.5 Initial treatment of severe AAV [1, 2].
Agent Route Initial dose Comments
Cyclophosphamide IV, oral IV: 15 mg/kg every 2 weeks for three Reduce dose in severe renal
pulses then every 3 weeks impairment or in older patients
PO: 1.5–2 mg/kg/day (>60 years)
Glucocorticoids IV, oral IV methylprednisolone for 3 days
PO prednisolone 1 mg/kg/day (max.
60 mg) for 4 weeks then gradual taper
3–4 months
Rituximab IV 375 mg/m2 weekly for 4 weeks or 2 × 1 g Alternative to cyclophosphamide in
infusions 2 weeks apart some guidelines [1]
Plasma exchange 60 ml/kg volume replacement to a Monitor platelet count (aim
maximum of 4 l with human albumin >70 × 109/l); fibrinogen (aim >1 g/l
solution or fresh frozen plasma if and replace with cryoprecipitate if
bleeding risk (pulmonary hemorrhage necessary); corrected calcium (aim
or recent kidney biopsy) normal range)
IV, intravenous; PO (Per os)
Table 28.6 Summary of evidence for pulsed vs. oral cyclophosphamide induction therapy for AAV.
Comparative risks
Number of participants Quality of
Outcome (number of studies) Continuous CYC Pulsed CYC Relative effect evidence (GRADE)
Death 278 (4) 206/1000 159/1000 (91–272) RR 0.77 (0.44–1.32) Low

RRT 245 (4) 74/1000 141/1000 (68–289) RR 1.9 (0.92–3.91) Low
Relapse 235 (4) 181/1000 324/1000 (201–519) RR 1.79 (1.11–2.87) Low
Serious infections 278 (4) 348/1000 247/1000 (132–463) RR 0.71 (0.38–1.33) Very low
AAV, antineutrophil cytoplasmic antibody-associated vasculitides; CYC, cyclophosphamide; RR, ;RRT, renal replacement therapy.
Source: Data drawn from Cochrane review of the evidence [3].
Table 28.7 Summary of clinical trials investigating pulsed vs. oral cyclophosphamide induction therapy for AAV.
Study Design Primary end point Results Secondary end points Comments
CYCLOPS [65] Open label, Time to No difference: Lower cumulative CYC Lower relapse rates in
parallel RCT remission median 3 months in does in pulsed IV group daily oral CYC group on
Daily oral vs. both groups and lower incidence of retrospective long-term
pulsed CYC No difference in leukopenia follow-up (median
proportion in No difference in other 4.3 years)
remission adverse events or death Low risk of bias
Adu 1997 [70] Parallel RCT Drug toxicity More leukopenia in No difference in number Patients in pulsed group
Daily oral CYC CCAZP group in complete remission at had poorer renal function
followed by final follow-up, relapse at baseline than oral group
AZA (at median rates, death or
3 months) vs. requirement for RRT
pulsed CYC
Guillevin Parallel RCT Remission No difference in Greater relapse rates in Recruitment terminated
1997 [71] Daily oral vs. induction and proportion of pulsed CYC group early due to significantly
pulsed CYC death patients in remission Lower cumulative CYC differing rates of severe
at 6 months dose in pulsed group and side effects and relapse
No difference in fewer severe side effects rates between the groups
5-year survival (infection, PCP)
Haubitz Parallel RCT Disease Patient survival, Four patients excluded due
1998 [72] Daily oral vs. progression remission rate, to protocol violations
pulsed CYC Nonremission at relapse rate similar Study terminated early
12 months between two groups when significant
Relapse up to Decreased differences between the
1 year after end cumulative CYC two groups found
of CYC dose, and leukopenia
Severe adverse and severe infections
events in IV group
AAV(ANCA-associated vasculitis), AZA, azathioprine; CCAZP(oral cyclophosphamide followed by azothioprine); CYC, cyclophosphamide;
PCP, Pneumocystis jirovecci pneumonia; RCT(Randomized control led trial).
pulses. There were fewer relapses in the group of patients (PEXIVAS) trial is a large multicenter RCT including patients
treated with 12 pulses [68]. Published guidelines recommend with moderate to severe renal impairment (eGFR(estimated
3–6 months treatment with oral cyclophosphamide based on GFR)<50 ml/min) or pulmonary hemorrhage, which was
time to remission in studies of maintenance therapy [69]. designed to provide definitive answers regarding PEX in
AAV but also included a comparison of high-and low-dose
glucocorticoid in both arms [74]. It showed that a reduced
Glucocorticoids
dose of oral glucocorticoid (following pulsed IV methyl-
High-dose glucocorticoids should be used as part of initial prednisolone in both groups) was noninferior on the com-
treatment, but there is limited evidence for the benefit of IV posite primary end point of ESRD or death and was
pulsed methylprednisolone other than in the MEPEX trial, associated with fewer serious infections [75].
where it was shown to be less efficacious than plasma
exchange (PEX) when added to cyclophosphamide and oral
Plasma Exchange
steroids for severe disease [38]. Dose and duration of oral
steroids varies between published guidelines, and duration Current guidelines advocate the addition of PEX as adjunc-
of therapy is often guided by patient response and relapse. tive therapy in patients with severe pulmonary or renal dis-
There are no published studies comparing glucocorticoid ease, particularly those who are refractory to initial treatment
regimens; one meta-analysis of 13 studies concluded that with cyclophosphamide and high-dose glucocorticoids. The
studies with longer courses of glucocorticoids were associ- largest published study to date, MEPEX, included 137
ated with fewer relapses although there was considerable patients with severe renal disease (creatinine >500 μmol/l or
heterogeneity between other immunosuppressive agents dialysis requiring at presentation) and randomized them to
used between studies [73]. The Plasma Exchange and either receive seven plasma exchanges or IV methylpredni-
Glucocorticoids in Severe ANCA- Associated Vasculitis solone. There was improved renal survival in the PEX group
at 3 months and decreased requirement for dialysis at There are two published RCTs comparing rituximab to
12 months [38]. Long-term follow-up of this group (median cyclophosphamide in AAV for first-line induction ther-
follow-up of 3.95 years) showed no difference in a primary apy [39, 78]. Both studies used a dose of rituximab of
composite end point of death or ESRD [76]. There are seven 375 mg/m2 body area weekly for 4 weeks, and patients
other published studies investigating plasma exchange as an received high-dose glucocorticoids initially. In the rituxi-
adjunctive therapy. A Cochrane review and a meta-analysis mab versus cyclophosphamide in ANCA-associated renal
have both reviewed the current evidence for PEX in AAV vasculitis (RITUXIVAS) trial, patients in the rituximab
(summarized in Table 28.8) [3, 77]. Both concluded that PEX arm also received two doses of IV cyclophosphamide with
may be of benefit in preventing ESRD and death at 1 year, the aim of ensuring early control of severe disease [39].
however this was not thought to be conclusive as the overall Rituximab was noninferior to cyclophosphamide in both
quality of the evidence is low and there is heterogeneity trials, and in the Rituximab for the Treatment of Wegener’s
between included studies [77]. The PEXIVAS trial (described Granulomatosis and Microscopic Polyangiitis (RAVE)
above) has been reported in abstract form and has shown no trial, rituximab seemed to be more effective than cyclo-
benefit of PEX on the primary composite outcome of death phosphamide for those with disease relapse rather than
from any cause or ESRD (28% of PEX group vs. 31% in the de novo presentation. Subgroup analysis of patient in the
no-PEX group, HR 0.86, 95% CI 0.65–1.13, P = 0.27). Further RAVE trial who had evidence of renal impairment showed
subgroup analysis and publication of the full trial is that rituximab was also noninferior to cyclophosphamide
awaited [75]. in these patients, although the trial did not include a large
number of patients with severe renal involvement
(median eGFR at trial entry was 41 ml/min in the rituxi-
Rituximab mab group and 50 ml/min in the cyclophosphamide
group) [79]. These trials and a summary of the evidence
Rituximab is recommended in most guidelines as an alter-
for rituximab induction therapy are shown in Table 28.9
native induction agent in patients who are unable to receive
and Table 28.10. Remission induction using a combination
or who are resistant to cyclophosphamide; the EULAR/
of low-dose cyclophosphamide and rituximab in combi-
ERA-EDTA guidelines recommend treatment with either
nation with a steroid-sparing approach has been described
agent with no requirement for intolerance or resistance to
in an open-label cohort study. Long-term outcomes in this
cyclophosphamide [1, 2].
Table 28.8 Summary of evidence for adjunctive PEX in AAV.
Comparative risks
Outcome (number of studies) Control Plasma exchange Relative effect evidence (GRADE)
Death at 1 year 267 (5) 189/1000 195/1000 (117–310) RR 1.03 (0.62–1.64) Low
Dialysis at 1 year 235 (6) 376/1000 169/1000 (109–271) RR 0.45 (0.29–0.72) Low
Serious infections 252 (4) 220/1000 262/1000 (167–405) RR 1.19 (0.76–1.84) Low
AAV,(ANCA-Associated Vasculitis); PEX,(Plasma Exchange); RR(relative risk),

The authors of the Cochrane review comment on the heterogeneity of the available studies with some being of high quality but others with
significant problems and risk of bias.
Source: Data drawn from Cochrane review of the evidence [3, 77].
Table 28.9 Summary of evidence for rituximab as induction therapy in AAV.
Comparative risks
Outcome (number of studies) CYC RTX Relative effect evidence (GRADE)
Remission at 6 months 236 (2) 661/1000 720/1000 (608–853) RR 1.09 (0.92–1.29) Moderate
Serious infections 241 (2) 92/1000 84/1000 (39–180) RR 0.91 (0.42–1.96) High
AAV,(ANCA-Associated Vasculitis); CYC, cyclophosphamide; RR(relative risk); RTX, rituximab.

Table 28.10 Summary of clinical trials investigating rituximab vs. cyclophosphamide as induction therapy for AAV.
Secondary end
Study Design Primary end point Results points Comments
RITUXIVAS [37] Parallel RCT Sustained remission No difference in No difference Patients in the RTX arm
Rituximab vs. at 12 months remission rates at in mortality also received two pulses of
CYC in a 3 : 1 Severe adverse events 12 months (76% RTX vs. IV CYC with doses 1 and 3
ratio 82% CYC) of RTX
No difference in rates of
severe adverse events
RAVE [76] Parallel RCT Complete remission No difference in No difference Rituximab superior in
without prednisolone remission rates (64% in in adverse those with relapsing
during the first RTX arm, 53% in CYC events disease
6 months arm)
AAV(ANCA-Associated Vasculitis); CYC, cyclophosphamide; IV, intravenous; RCT(Randomized control led trial); RTX, rituximab.
group were favorable when compared to historical cohorts disease, with a preference for MMF in those who have evi-
and a prospective trial of this treatment regimen may be dence of renal involvement.
warranted [80]. Other studies have investigated intravenous immuno-
globulin, infliximab, and alemtuzumab and these could be
considered for resistant disease [86–89].
Other Induction Agents
Several studies have compared other induction agents to
Remission Maintenance
cyclophosphamide in AAV and these could be considered
for resistant disease but are not currently recommended for Current guidelines recommend that patients with severe
severe disease in published guidelines. AAV who are in remission after induction therapy with
There are three trials which have compared high-dose glucocorticoids plus cyclophosphamide/rituxi-
MMF(Mycophenolate mofetil) to cyclophosphamide for mab should be treated with low-dose glucocorticoids plus
induction of remission. The Mycophenolate Mofetil versus another immunosuppressive agent. Most commonly, aza-
Cyclophosphamide for Remission Induction of ANCA- thioprine is recommended with methotrexate or MMF as
Associated Vasculitis (MYCYC) trial was a noninferiority an alternative [2, 69]. Introduction of azathioprine at
trial comparing 6 months of induction with MMF to 6–10 6 months after induction therapy has been shown to be
pulses of IV cyclophosphamide, which to date has only been noninferior to continuing cyclophosphamide to 12 months
presented in abstract form, and it failed to demonstrate non- on rates of relapse at 18 months, and is associated with
inferiority of MMF [81]. Cochrane review of the three trials fewer adverse events [90]. Long-term follow-up of this
showed remission at 6 months was better with MMF com- study suggested there may be greater frequency of relapse
pared to cyclophosphamide (RR(relative risk ratio) 1.17, CI in the group treated with azathioprine although this was
1.02–1.35), and there were no difference in adverse event not statistically significant [91]. Azathioprine has also been
rates between the groups [3, 81–83]. Patients with RPGN or shown to be superior to MMF for remission maintenance
severe renal failure were excluded from these studies. but with similar rates of adverse events in MMF and aza-
One trial has compared methotrexate to cyclophospha- thioprine groups [92]. One trial has shown similar rates of
mide, showing it to be noninferior with no difference in relapse in patients treated with methotrexate or azathio-
mortality or rates of remission at 6 months. Again, patients prine for remission maintenance; however, the primary
with severe renal impairment were not included in this end point of this trial was safety of methotrexate rather
trial and there was longer time to remission and higher than efficacy [93]. There were similar numbers of adverse
relapse rates at 18 months in patients treated with metho- events between the two groups but severity of adverse
trexate. Median time to relapse was 13 months in those events was greater in those treated with methotrexate [93].
treated with methotrexate and 15 months in the cyclophos- These studies are summarized in Table 28.11.
phamide group [84, 85]. One RCT has investigated the duration of azathioprine
Based on the studies using MMF and methotrexate maintenance therapy comparing tapering after 1 year or
published guidelines recommend the use of either of these 4 years in patients with anti-PR3 AAV. This showed no
agents in patients with mild and nonorgan-threatening significant difference in relapse-free survival at 4 years
Table 28.11 Summary of evidence for azathioprine maintenance therapy in AAV.
Comparative risks Quality of

Number of Primary end Secondary end Relative evidence
Study Design participants point Results points Outcomes Comparator AZA effect (GRADE)
CYCAZAREM [90] Parallel RCT 144 Relapse rate at No difference No difference Relapse 137/1000 155/1000 RR 1.13 Moderate
AZA vs. oral CYC 18 months in relapse rate in mortality or (70–342) (0.51–2.50)
following CYC/ (15.5% AZA, adverse events Infections 178/1000 183/1000 RR 1.03 Moderate
steroid induction 13.7% CYC) (91–361) (0.51–2.03)
WEGENT [93] Parallel RCT 126 Adverse No difference No significant Adverse 190/1000 110/1000 RR 0.58 Moderate
AZA vs. MTX after reaction (11.1% AZA, difference in event (48–262) (0.25–1.38)
pulsed CYC/steroid causing death 19% MTX) relapse rate Relapse 333/1000 366/1000 RR 1.1 Moderate
induction or withdrawal (226–589) (0.68–1.77)
of study drug
IMPROVE [92] Parallel open-label 156 Relapse-free More relapses No difference Relapse 552/1000 375/1000 RR 1.47 Moderate
RCT survival in MMF group in adverse (265–531) (1.04–2.09)
AZA vs. MMF after (55% vs. 37.5% events
CYC/steroid in AZA group)
induction
AAV(ANCA-Associated Vasculitis); AZA; azathioprine; CYC, cyclophosphamide; MMF(Mycophenolate mofetil); MTX; methotrexate; RCT(Randomized control led trial); RR(relative risk
ratio); RTX, rituximab.
Table 28.12 Summary of evidence for rituximab maintenance therapy in AAV.
Comparative risks
Number of
Study Design participants Outcome AZA RTX Relative effect
MAINRITSAN [40] Parallel RCT 118 Major relapse within 293/1000 52/1000 (18–170) RR 0.17 (0.06–0.58)
RTX or AZA after CYC/ 28 months
steroid induction
Severe adverse events 138/1000 193/1000 (42–228) RR 0.71 (0.31–1.65)
Fixed Tailored
MAINRITSAN 2(98) RCT 162 Relapse within 28 months 99/1000 173/1000 (77–389) RR 1.75 (0.78–3.94)
Tailored or fixed schedule
rituximab
AZA, azathioprine; CYC, cyclophosphamide; RCT(Randomized control led trial); RR(relative risk ratio); RTX, rituximab.
0005152416.INDD 471 09-12-2022 15:54:55

between the two groups [94]. The prolonged Remission- exposure should be taken into account when planning to
Maintenance Therapy in Systemic Vasculitis (REMAIN) re-dose patients.
study also investigated duration of maintenance therapy
and included 117 patients with stable remission following
Co-trimoxazole
induction with cyclophosphamide and prednisolone. It
showed that more prolonged maintenance treatment with Prophylaxis against pneumocystis jirovecci is recommended
azathioprine and prednisolone (48 months vs. 24 months) in all patients receiving induction therapy with cyclophos-
was associated with significantly lower risk of relapse. The phamide or rituximab. First-line treatment is with co-
prolonged treatment group had better renal outcomes but trimoxazole with pentamidine nebulizers as second line.
there was no difference in patient survival [95]. There are also two studies which have investigated the role of
Rituximab is increasingly being used as an agent for co-trimoxazole in relapse prevention in patients with PR3-
remission maintenance in some centers. The Maintenance ANCA disease [98, 99]. There was a suggestion that the use of
of Remission using Rituximab in Systemic ANCA- co-trimoxazole may reduce the risk of relapse at 1 year, par-
associated Vasculitis (MAINRITSAN) trial included 115 ticularly pulmonary relapse, but Cochrane review of the two
patients who were randomized to either rituximab or aza- studies found this result to be nonstatistically significant and
thioprine for remission maintenance after induction ther- so a firm conclusion regarding the use of antibiotics for
apy with cyclophosphamide and glucocorticoids Table 28.12 remission maintenance was not possible (Table 28.13) [3].
[40]. Rituximab was superior to azathioprine in preventing
relapses (5% vs. 29% at 28 months) and adverse event rates
Prognosis
were similar between the two groups. Long-term follow-up
of this study shows that rituximab remains superior at Current studies estimate survival in patients with AAV to be
60 months (relapse free survival 37.2% vs. 57.9%). Again, around 90% at 1 year and around 70% at 5 years, a significant
the adverse event profile was similar between the two improvement following the introduction of modern immuno-
groups [96]. The MAINRITSAN 2 trial went on to compare suppression regimens [64, 100]. A substantial cause of mortal-
fixed rituximab dosing every 6 months to a patient-tailored ity within the first year is infection rather than active vasculitis,
approach with rituximab re-dosing when CD19+ B cells or emphasizing the need for further research into better targeted
ANCA returned. Relapse rates were similar between the therapies with better side-effect profiles [101, 102]. One year
two arms (17.3% relapse rate in the tailored group vs. 9.9% survival free of ESRD is approximately 75% and is 50% at
in the scheduled treatment group); the tailored treatment 5 years [64]. The increased mortality in patients with AAV

group received fewer rituximab infusions over the study persists at long-term follow-up with increased rates of throm-
period [97]. bosis, cardiovascular disease, malignancy, and infections
when compared to the general population [102, 103]. In one
study, predictors of mortality included advanced renal failure
Treatment of Relapse
and older age [101]. Despite this, it is not the case that older
Relapse is usually treated in the same way as patients pre- patients should not be considered for aggressive treatment.
senting with de novo disease. Patients with minor relapses One multicenter retrospective analysis of patients 75 years
may respond to an increase or reintroduction of their oral showed that patients who were not treated with cyclophos-
immunosuppression or steroids. Those with severe phamide or rituximab had significantly lower survival at
relapse should be treated with high-dose glucocorticoids 2 years (72.9% with cyclophosphamide, 81.3% with rituximab,
with cyclophosphamide or rituximab. Long-term follow- and 45.0% with no/other treatments) [104].
up of the RAVE study suggested that relapsing patients Despite the improvements in mortality, in patients who
may respond better to rituximab than cyclophospha- have renal disease as part of their AAV the risk of relapse has
mide [79]. Additionally, cumulative cyclophosphamide remained relatively constant over recent years. One large
Table 28.13 Summary of evidence for antibiotics as maintenance therapy in AAV.
Comparative risks
Number of participants
Outcome (number of studies) Placebo Co-trimoxazole Relative effect Quality of evidence (GRADE)
Remission at 1 year 111 (2) 796/1000 907/1000 (780–1000) 1.14 (0.98–1.33) Low
The authors comment that one of the two studies has multiple limitations in the study design.
Reference 473
study has shown a cumulative relapse rate of 11% at 1 year modern studies, possibly due to the current transplant
and 35% at 5 years with no change in relapse rates over the immunosuppression regimens used [108, 110].
25 year study period [64]. Patients with PR3-ANCA disease
are more likely to experience relapse than those who are
MPO-ANCA positive or ANCA negative; presence of lung Conclusions
disease and older patient age have also been shown to predict
relapse in some studies [105, 106]. The utility of ANCA titer AAV are a group of rare, multisystem, small and medium
for predicting relapse is debatable although it may correlate vessel vasculitides. Modern immunosuppression regimens
better in patients with renal disease where some studies have using more effective and targeted drugs have significantly
shown rising ANCA titer may predict relapse [35, 107]. improved morbidity and mortality, and mortality in the
Patients with AAV who require renal transplant have first year is more likely to be due to complications of treat-
clinical outcomes which are comparable or better to those ment than active vasculitis. There are several available
with other causes of ESRD [108]. Transplant should be evidence-based guidelines for treatment which are based
deferred until patients are 1 year from remission as one on high-quality RCTs. Most authorities recommend induc-
study has shown that risk of death post transplant was tion treatment with high-dose glucocorticoids plus cyclo-
greatest in those transplanted less than a year after remis- phosphamide or rituximab and maintenance therapy with
sion [109]. The risk of recurrent disease is low after trans- low-dose glucocorticoids and azathioprine.
plantation and relapse rates seem to be lower in more
References
Yates, M., Watts, R.A., Bajema, I.M. et al. (2016). EULAR/

1 9 Stilmant, M.M., Bolton, W.K., Sturgill, B.C. et al. (1979).
ERA-EDTA recommendations for the management of ANCA- Crescentic glomerulonephritis without immune deposits:
associated vasculitis. Ann. Rheum. Dis. 75 (9): 1583–1594. clinicopathologic features. Kidney Int. 15 (2): 184–195.
2 Cattran, D. C., Feehally, J., Cook, H. T., et al. (2012). 10 Davies, D.J., Moran, J.E., Niall, J.F., and Ryan, G.B.
Kidney disease: Improving global outcomes (KDIGO) (1982). Segmental necrotising glomerulonephritis with
glomerulonephritis work group. KDIGO clinical practice antineutrophil antibody: possible arbovirus aetiology?
guideline for glomerulonephritis. Kidney International Br. Med. J. (Clinical research edition) 285 (6342): 606.
Supplements, 2(2), 139–274. 11 Falk, R.J. and Jennette, J.C. (1988). Anti-neutrophil
3 Walters GD, Willis NS, Cooper TE, Craig JC. Interventions for cytoplasmic autoantibodies with specificity for
renal vasculitis in adults. Cochrane Database of Systematic myeloperoxidase in patients with systemic vasculitis and
Reviews 2020, Issue 1. Art. No.: CD003232. Rev. 9. idiopathic necrotizing and crescentic glomerulonephritis.
4 Hultcrantz, M., Rind, D., Akl, E.A. et al. (2017). The N. Engl. J. Med. 318 (25): 1651–1657.
GRADE Working Group clarifies the construct of certainty 12 Niles, J.L., McCluskey, R.T., Ahmad, M.F., and Arnaout,
of evidence. J. Clin. Epidemiol. 87: 4–13. M.A. (1989). Wegener’s granulomatosis autoantigen is a
5 Jennette, J.C., Falk, R.J., Bacon, P.A. et al. (2013). 2012 novel neutrophil serine proteinase. Blood 74 (6):
Revised International Chapel Hill Consensus Conference: 1888–1893.
Nomenclature of vasculitides. Arthritis Rheum. 65 (1): 1–11. 13 van der Woude, F.J., Rasmussen, N., Lobatto, S. et al.
6 Grayson, P.C., Ponte, C., Suppiah, R., et.al. (2022). 2022 (1985). Autoantibodies against neutrophils and monocytes:
American College of Rheumatology/European Alliance of tool for diagnosis and marker of disease activity in
Associations for Rheumatology classification criteria for Wegener’s granulomatosis. Lancet 1 (8426): 425–429.
eosinophilic granulomatosis with polyangiitis. Arthritis 14 Sable-Fourtassou, R., Cohen, P., Mahr, A. et al. (2005).
Rheum 74 (3): 386–392. Antineutrophil cytoplasmic antibodies and the Churg-
7 Robson J.C., Grayson, P.C., Ponte, C., et.al. (2022). 2022 Strauss syndrome. Ann. Intern. Med. 143 (9): 632–638.
American College of Rheumatology/European Alliance of 15 Watts, R.A., Mahr, A., Mohammad, A.J. et al. (2015).
Associations for Rheumatology classification criteria for Classification, epidemiology and clinical subgrouping of
microscopic polyangiitis. Arthritis Rheum 74 (3): 393–399. antineutrophil cytoplasmic antibody (ANCA)-associated
8 Suppiah, R., Robson J.C., Grayson, P.C., et.al. (2022). 2022 vasculitis. Nephrol. Dial. Transplant. 30 (Suppl 1): i14–i22.
American College of Rheumatology/ European Alliance of 16 Watts, R.A., Lane, S.E., Bentham, G., and Scott, D.G.
Associations for Rheumatology classification criteria for (2000). Epidemiology of systemic vasculitis: a ten-year
granulomatosis with polyangiitis. Arthritis Rheum 74 (3): study in the United Kingdom. Arthritis Rheum. 43 (2):
400–406. 414–419.
7 Gonzalez-Gay, M.A., Garcia-Porrua, C., Guerrero, J. et al.

1 30 de Lind van Wijngaarden, R.A., van Rijn, L., Hagen, E.C.
(2003). The epidemiology of the primary systemic et al. (2008). Hypotheses on the etiology of antineutrophil
vasculitides in Northwest Spain: implications of the cytoplasmic autoantibody associated vasculitis: the cause
Chapel Hill consensus conference definitions. Arthritis is hidden, but the result is known. Clini. J. Am. Soci.
Rheum. 49 (3): 388–393. Nephrol. 3 (1): 237–252.
18 Mohammad, A.J., Jacobsson, L.T., Westman, K.W. et al. 31 Pendergraft, W.F. 3rd, Herlitz, L.C., Thornley-Brown, D.
(2009). Incidence and survival rates in Wegener’s et al. (2014). Nephrotoxic effects of common and
granulomatosis, microscopic polyangiitis, Churg-Strauss emerging drugs of abuse. Clin. J. Am. Soc. Nephrol. 9 (11):
syndrome and polyarteritis nodosa. Rheumatology 48 1996–2005.
(12): 1560–1565. 32 Yashiro, M., Muso, E., Itoh-Ihara, T. et al. (2000).
19 Watts RA, Gonzalez-Gay MA, Lane SE, Garcia-Porrua C, Significantly high regional morbidity of MPO-ANCA-
Bentham G, Scott DG. Geoepidemiology of systemic related angitis and/or nephritis with respiratory tract
vasculitis: comparison of the incidence in two regions of involvement after the 1995 great earthquake in Kobe
Europe. Ann. Rheum. Dis. 2001;60(2):170–172. (Japan). Am. J. Kidney Dis. 35 (5): 889–895.
20 Fujimoto, S., Watts, R.A., Kobayashi, S. et al. (2011). 33 Knight, A., Sandin, S., and Askling, J. (2008). Risks and
Comparison of the epidemiology of anti-neutrophil relative risks of Wegener’s granulomatosis among close
cytoplasmic antibody-associated vasculitis between Japan relatives of patients with the disease. Arthritis Rheum.
and the U.K. Rheumatology 50 (10): 1916–1920. 58 (1): 302–307.
21 Mahr, A., Guillevin, L., Poissonnet, M., and Ayme, S. 34 Lane, S.E., Watts, R.A., Bentham, G. et al. (2003). Are
(2004). Prevalences of polyarteritis nodosa, microscopic environmental factors important in primary systemic
polyangiitis, Wegener’s granulomatosis, and Churg- vasculitis? A case-control study. Arthritis Rheum. 48 (3):
Strauss syndrome in a French urban multiethnic 814–823.
population in 2000: a capture-recapture estimate. 35 Fussner, L.A., Hummel, A.M., Schroeder, D.R. et al.
Arthritis Rheum. 51 (1): 92–99. (2016). Factors determining the clinical utility of serial
22 Andrews, M., Edmunds, M., Campbell, A. et al. (1990). measurements of antineutrophil cytoplasmic antibodies
Systemic vasculitis in the 1980s: Is there an increasing targeting proteinase 3. Arthritis Rheum. 68 (7):
incidence of Wegener’s granulomatosis and microscopic 1700–1710.
polyarteritis? J. R. Coll. Physicians Lond. 24 (4): 284–288. 36 Bansal, P.J. and Tobin, M.C. (2004). Neonatal microscopic
23 Knight, A., Ekbom, A., Brandt, L., and Askling, J. (2006). polyangiitis secondary to transfer of maternal
Increasing incidence of Wegener’s granulomatosis in myeloperoxidase-antineutrophil cytoplasmic antibody
Sweden, 1975-2001. J. Rheumatol. 33 (10): 2060–2063. resulting in neonatal pulmonary hemorrhage and renal
24 Lyons, P.A., Rayner, T.F., Trivedi, S. et al. (2012). involvement. Ann. Allergy Asthma Immunol. 93 (4):
Genetically distinct subsets within ANCA-associated 398–401.
vasculitis. N. Engl. J. Med. 367 (3): 214–223. 37 Silva, F., Specks, U., Sethi, S. et al. (2009). Successful
25 Merkel, P.A., Xie, G., Monach, P.A. et al. (2017). pregnancy and delivery of a healthy newborn despite
Identification of functional and expression transplacental transfer of antimyeloperoxidase antibodies
polymorphisms associated with risk for antineutrophil from a mother with microscopic polyangiitis. Am. J.
cytoplasmic autoantibody-associated vasculitis. Arthritis Kidney Dis. 54 (3): 542–545.
Rheum. 69 (5): 1054–1066. 38 Jayne, D.R., Gaskin, G., Rasmussen, N. et al. (2007).
26 Rahmattulla, C., Mooyaart, A.L., van Hooven, D. et al. Randomized trial of plasma exchange or high-dosage
(2016). Genetic variants in ANCA-associated vasculitis: a methylprednisolone as adjunctive therapy for severe
meta-analysis. Ann. Rheum. Dis. 75 (9): 1687. renal vasculitis. J. Am. Soci. Nephrol 18 (7): 2180–2188.
27 Popa, E.R., Stegeman, C.A., Abdulahad, W.H. et al. 39 Jones, R.B., Tervaert, J.W., Hauser, T. et al. (2010).
(2007). Staphylococcal toxic-shock-syndrome-toxin-1 as a Rituximab versus cyclophosphamide in ANCA-associated
risk factor for disease relapse in Wegener’s renal vasculitis. N. Engl. J. Med. 363 (3): 211–220.
granulomatosis. Rheumatology. 46 (6): 1029–1033. 40 Guillevin, L., Pagnoux, C., Karras, A. et al. (2014).
28 Kain, R., Exner, M., Brandes, R. et al. (2008). Molecular Rituximab versus azathioprine for maintenance in
mimicry in pauci-immune focal necrotizing ANCA-associated vasculitis. N. Engl. J. Med. 371 (19):
glomerulonephritis. Nat. Med. 14 (10): 1088–1096. 1771–1780.
29 Yu, F., Chen, M., Gao, Y. et al. (2007). Clinical and 41 Xiao, H., Heeringa, P., Hu, P. et al. (2002). Antineutrophil
pathological features of renal involvement in cytoplasmic autoantibodies specific for myeloperoxidase
propylthiouracil-associated ANCA-positive vasculitis. cause glomerulonephritis and vasculitis in mice. J. Clin.
Am. J.Kidney Dis. 49 (5): 607–614. Invest. 110 (7): 955–963.
Reference 475
2 Falk, R.J., Terrell, R.S., Charles, L.A., and Jennette, J.C.

4 55 Jennette, J.C. and Falk, R.J. (1997). Small-vessel
(1990). Anti-neutrophil cytoplasmic autoantibodies induce vasculitis. N. Engl. J. Med. 337 (21): 1512–1523.
neutrophils to degranulate and produce oxygen radicals 56 Bossuyt, X., Cohen Tervaert, J.W., Arimura, Y. et al.
in vitro. Proc. Natl. Acad. Sci. USA. 87 (11): 4115–4119. (2017). Position paper: revised 2017 international
43 Savage, C.O., Gaskin, G., Pusey, C.D., and Pearson, J.D. consensus on testing of ANCAs in granulomatosis with
(1993). Anti-neutrophil cytoplasm antibodies can polyangiitis and microscopic polyangiitis. Nat. Rev.
recognize vascular endothelial cell-bound anti-neutrophil Rheumatol. 13 (11): 683–692.
cytoplasm antibody-associated autoantigens. Exp. 57 Hoffman, G.S., Kerr, G.S., Leavitt, R.Y. et al. (1992).
Nephrol. 1 (3): 190–195. Wegener granulomatosis: an analysis of 158 patients.
44 Schreiber, A., Xiao, H., Jennette, J.C. et al. (2009). C5a Ann. Intern. Med. 116 (6): 488–498.
receptor mediates neutrophil activation and ANCA- 58 Lazor, R., Bigay-Game, L., Cottin, V. et al. (2007). Alveolar
induced glomerulonephritis. J. Am. Soci. Nephrol. 20 (2): hemorrhage in anti-basement membrane antibody
289–298. disease: a series of 28 cases. Medicine 86 (3): 181–193.
45 Kessenbrock, K., Krumbholz, M., Schonermarck, U. et al. 59 Csernok, E. and Moosig, F. (2014). Current and emerging
(2009). Netting neutrophils in autoimmune small-vessel techniques for ANCA detection in vasculitis. Nat. Rev.
vasculitis. Nat. Med. 15 (6): 623–625. Rheumatol. 10 (8): 494–501.
46 Xu, P.C., Cui, Z., Chen, M. et al. (2011). Comparison of 60 Csernok, E., Damoiseaux, J., Rasmussen, N. et al. (2016).
characteristics of natural autoantibodies against Evaluation of automated multi-parametric indirect
myeloperoxidase and anti-myeloperoxidase immunofluorescence assays to detect anti-neutrophil
autoantibodies from patients with microscopic cytoplasmic antibodies (ANCA) in granulomatosis with
polyangiitis. Rheumatology 50 (7): 1236–1243. polyangiitis (GPA) and microscopic polyangiitis (MPA).
47 Roth, A.J., Ooi, J.D., Hess, J.J. et al. (2013). Epitope Autoimmun. Rev. 15 (7): 736–741.
specificity determines pathogenicity and detectability in 61 Csernok, E., Mahrhold, J., and Hellmich, B. (2018).
ANCA-associated vasculitis. J. Clin. Invest. 123 (4): Anti-neutrophil cytoplasm antibodies (ANCA): recent
1773–1783. methodological advances – Lead to new consensus
48 Xiao, H., Schreiber, A., Heeringa, P. et al. (2007). recommendations for ANCA detection. J. Immunol.
Alternative complement pathway in the pathogenesis of Methods 456: 1–6.
disease mediated by anti-neutrophil cytoplasmic 62 Joh, K., Muso, E., Shigematsu, H. et al. (2008). Renal
autoantibodies. Am. J. Pathol. 170 (1): 52–64. pathology of ANCA-related vasculitis: proposal for
49 Jayne, D.R.W., Bruchfeld, A.N., Harper, L. et al. (2017). standardization of pathological diagnosis in Japan. Clin.
Randomized trial of C5a receptor inhibitor avacopan in Exp. Nephrol. 12 (4): 277–291.
ANCA-associated vasculitis. J. Am. Soc. Nephrol. 28 (9): 63 Berden, A.E., Ferrario, F., Hagen, E.C. et al. (2010).
2756–2767. Histopathologic classification of ANCA-associated
50 Pendergraft Iii, W.F., Preston, G.A., Shah, R.R. et al. glomerulonephritis. J. Am. Soc. Nephrol. 21 (10): 1628.
(2003). Autoimmunity is triggered by cPR-3(105–201), a 64 Rhee, R.L., Hogan, S.L., Poulton, C.J. et al. (2016).
protein complementary to human autoantigen Trends in long-term outcomes among patients with
proteinase-3. Nat. Med. 10: 72. Antineutrophil cytoplasmic antibody-associated Vasculitis
51 Morgan, M.D., Day, C.J., Piper, K.P. et al. (2010). Patients with renal disease. Arthritis Rheum 68 (7): 1711–1720.
with Wegener’s granulomatosis demonstrate a relative 65 de Groot, K., Harper, L., Jayne, D.R. et al. (2009). Pulse
deficiency and functional impairment of T-regulatory versus daily oral cyclophosphamide for induction of
cells. Immunology 130 (1): 64–73. remission in antineutrophil cytoplasmic antibody-
52 Free, M.E., Bunch, D.O., McGregor, J.A. et al. (2013). associated vasculitis: a randomized trial. Ann. Intern.
Patients with antineutrophil cytoplasmic antibody- Med. 150 (10): 670–680.
associated vasculitis have defective Treg cell function 66 Harper, L., Morgan, M.D., Walsh, M. et al. (2012). Pulse
exacerbated by the presence of a suppression-resistant versus daily oral cyclophosphamide for induction of
effector cell population. Arthritis Rheum. 65 (7): 1922–1933. remission in ANCA-associated vasculitis: long-term
53 Todd, S.K., Pepper, R.J., Draibe, J. et al. (2014). Regulatory follow-up. Ann. Rheum. Dis. 71 (6): 955–960.
B cells are numerically but not functionally deficient in 67 Ntatsaki, E., Carruthers, D., Chakravarty, K. et al. (2014).
anti-neutrophil cytoplasm antibody-associated vasculitis. BSR and BHPR guideline for the management of adults
Rheumatology 53 (9): 1693–1703. with ANCA-associated vasculitis. Rheumatology 53 (12):
54 Wilde, B., Thewissen, M., Damoiseaux, J. et al. (2013). 2306–2309.
Regulatory B cells in ANCA-associated vasculitis. Ann. 68 Guillevin, L., Cohen, P., Mahr, A. et al. (2003). Treatment
Rheum. Dis. 72 (8): 1416. of polyarteritis nodosa and microscopic polyangiitis with
poor prognosis factors: a prospective trial comparing 80 McAdoo, S.P., Medjeral-Thomas, N., Gopaluni, S. et al.
glucocorticoids and six or twelve cyclophosphamide pulses (2018). Long-term follow-up of a combined rituximab
in sixty-five patients. Arthritis Rheum. 49 (1): 93–100. and cyclophosphamide regimen in renal anti-neutrophil
69 Geetha, D., Jin, Q., Scott, J. et al. (2018). Comparisons of cytoplasm antibody-associated vasculitis. Nephrol. Dial.
guidelines and recommendations on managing Transplant. 33 (5): 899.
antineutrophil cytoplasmic antibody-associated vasculitis. 81 Jones, R., Harper, L., Ballarin, J., et al. (2013).
Kidney Int. Rep. 3 (5): 1039–1049. A randomized trial of mycophenolate mofetil versus
70 Adu, D., Pall, A., Luqmani, R.A. et al. (1997). Controlled cyclophosphamide for remission of ANCA-associated
trial of pulse versus continuous prednisolone and vasculitis. “MYCYA.” On behalf of the European
cyclophosphamide in the treatment of systemic vasculitis. vasculitis study group. La Presse Medicale 42 (4):
QJM 90 (6): 401–409. 678–679.
71 Guillevin, L., Cordier, J.F., Lhote, F. et al. (1997). 82 Han, F., Liu, G., Zhang, X. et al. (2011). Effects of
A prospective, multicenter, randomized trial comparing mycophenolate mofetil combined with corticosteroids for
steroids and pulse cyclophosphamide versus steroids and induction therapy of microscopic polyangiitis. Am. J.
oral cyclophosphamide in the treatment of generalized Nephrol. 33 (2): 185–192.
Wegener’s granulomatosis. Arthritis Rheum. 40 (12): 83 Hu, W., Liu, C., Xie, H. et al. (2008). Mycophenolate
2187–2198. mofetil versus cyclophosphamide for inducing remission
72 Haubitz, M., Schellong, S., Gobel, U. et al. (1998). of ANCA vasculitis with moderate renal involvement.
Intravenous pulse administration of cyclophosphamide Nephrol. Dial. Transplant. 23 (4): 1307–1312.
versus daily oral treatment in patients with 84 Faurschou, M., Westman, K., Rasmussen, N. et al. (2012).
antineutrophil cytoplasmic antibody-associated vasculitis Brief report: long-term outcome of a randomized clinical
and renal involvement: a prospective, randomized study. trial comparing methotrexate to cyclophosphamide for
Arthritis Rheum. 41 (10): 1835–1844. remission induction in early systemic antineutrophil
73 Walsh, M., Merkel, P.A., Mahr, A., and Jayne, D. (2010). cytoplasmic antibody-associated vasculitis. Arthritis
Effects of duration of glucocorticoid therapy on relapse Rheum. 64 (10): 3472–3477.
rate in antineutrophil cytoplasmic antibody-associated 85 De Groot, K., Rasmussen, N., Bacon, P.A. et al. (2005).
vasculitis: a meta-analysis. Arthritis Care Res. 62 (8): Randomized trial of cyclophosphamide versus
1166–1173. methotrexate for induction of remission in early systemic
74 Walsh, M., Merkel, P.A., Peh, C.A. et al. (2013). Plasma antineutrophil cytoplasmic antibody-associated vasculitis.
exchange and glucocorticoid dosing in the treatment of Arthritis Rheum. 52 (8): 2461–2469.
anti-neutrophil cytoplasm antibody associated vasculitis 86 Jayne, D.R., Chapel, H., Adu, D. et al. (2000). Intravenous
(PEXIVAS): protocol for a randomized controlled trial. immunoglobulin for ANCA-associated systemic vasculitis
Trials 14 (1): 73. with persistent disease activity. QJM 93 (7): 433–439.
75 Walsh, M., Merkel, P., Peh, C.A. et al. (2018). LB01 the 87 Stone, J.H., Hoffman, G.S., and The Wegener’s
effect of plasma exchange on end stage renal disease and Granulomatosis Etanercept Trial (WGET) Research
death in patients with severe ANCA associated vasculitis. Group. (2005). Etanercept plus standard therapy for
Nephrol. Dial. Transplant. 33 (suppl_1): i636–i. Wegener’s granulomatosis. N Engl J Med 352 (4):
76 Walsh, M., Casian, A., Flossmann, O. et al. (2013). 351–361.
Long-term follow-up of patients with severe ANCA- 88 Booth, A., Harper, L., Hammad, T. et al. (2004).
associated vasculitis comparing plasma exchange to Prospective study of TNFalpha blockade with infliximab
intravenous methylprednisolone treatment is unclear. in anti-neutrophil cytoplasmic antibody-associated
Kidney Int. 84 (2): 397–402. systemic vasculitis. J. Am. Soc. Nephrol. 15 (3): 717–721.
77 Walsh, M., Catapano, F., Szpirt, W. et al. (2011). Plasma 89 Walsh, M., Chaudhry, A., and Jayne, D. (2008). Long-term
exchange for renal vasculitis and idiopathic rapidly follow-up of relapsing/refractory anti-neutrophil
progressive glomerulonephritis: a meta-analysis. Am. J. cytoplasm antibody associated vasculitis treated with the
Kidney Dis. 57 (4): 566–574. lymphocyte depleting antibody alemtuzumab
78 Stone, J.H., Merkel, P.A., Spiera, R. et al. (2010). (CAMPATH-1H). Ann. Rheum. Dis. 67 (9): 1322–1327.
Rituximab versus cyclophosphamide for ANCA- 90 Jayne, D., Rasmussen, N., Andrassy, K. et al. (2003).
associated vasculitis. N. Engl. J. Med. 363 (3): 221–232. A randomized trial of maintenance therapy for vasculitis
79 Geetha, D., Specks, U., Stone, J.H. et al. (2015). Rituximab associated with antineutrophil cytoplasmic
versus cyclophosphamide for ANCA-associated vasculitis autoantibodies. N. Engl. J. Med. 349 (1): 36–44.
with renal involvement. J. Am. Soci. Nephrol. 26 (4): 91 Walsh, M., Faurschou, M., Berden, A. et al. (2014).
976–985. Long-term follow-up of cyclophosphamide compared
Reference 477
with azathioprine for initial maintenance therapy in 100 Hilhorst, M., Wilde, B., van Paassen, P. et al. (2013).
ANCA-associated vasculitis. Clin. J. Am. Soci. Nephrol. 9 Improved outcome in anti-neutrophil cytoplasmic
(9): 1571–1576. antibody (ANCA)-associated glomerulonephritis: a
92 Hiemstra, T.F., Walsh, M., Mahr, A. et al. (2010). 30-year follow-up study. Nephrol. Dial. Transplant. 28
Mycophenolate mofetil vs azathioprine for remission (2): 373–379.
maintenance in antineutrophil cytoplasmic antibody- 101 Little, M.A., Nightingale, P., Verburgh, C.A. et al. (2010).
associated vasculitis: a randomized controlled trial. Early mortality in systemic vasculitis: relative
JAMA 304 (21): 2381–2388. contribution of adverse events and active vasculitis.
93 Pagnoux, C., Mahr, A., Hamidou, M.A. et al. (2008). Ann. Rheum. Dis. 69 (6): 1036–1043.
Azathioprine or methotrexate maintenance for 102 Flossmann, O., Berden, A., de Groot, K. et al. (2011).
ANCA-associated vasculitis. N. Engl. J. Med. 359 (26): Long-term patient survival in ANCA-associated
2790–2803. vasculitis. Ann. Rheum. Dis. 70 (3): 488.
94 Sanders, J.-S.F., de Joode, A.A.E., DeSevaux, R.G. et al. 103 Englund, M., Merkel, P.A., Tomasson, G. et al. (2016).
(2016). Extended versus standard azathioprine Comorbidities in patients with antineutrophil
maintenance therapy in newly diagnosed proteinase-3 cytoplasmic antibody-associated vasculitis versus
anti-neutrophil cytoplasmic antibody-associated the general population. J. Rheumatol. 43 (8): 1553–1558.
vasculitis patients who remain cytoplasmic anti- 104 Weiner, M., Goh, S.M., Mohammad, A.J. et al. (2015).
neutrophil cytoplasmic antibody-positive after induction Outcome and treatment of elderly patients with ANCA-
of remission: a randomized clinical trial. Nephrol. Dial. associated vasculitis. Clin. J. Am. Soc. Nephrol. 10 (7): 1128.
Transplant. 31 (9): 1453–1459. 105 Pagnoux, C., Hogan, S.L., Chin, H. et al. (2008).
95 Karras, A., Pagnoux, C., Haubitz, M. et al. (2017). Predictors of treatment resistance and relapse in
Randomised controlled trial of prolonged treatment in antineutrophil cytoplasmic antibody-associated
the remission phase of ANCA-associated vasculitis. Ann. small-vessel vasculitis: comparison of two independent
Rheum. Dis. 76 (10): 1662–1668. cohorts. Arthritis Rheum. 58 (9): 2908–2918.
96 Terrier, B., Pagnoux, C., Perrodeau, É. et al. (2018). 106 Hogan, S.L., Falk, R.J., Chin, H. et al. (2005). Predictors
Long-term efficacy of remission-maintenance regimens of relapse and treatment resistance in antineutrophil
for ANCA-associated vasculitides. Ann. Rheum. Dis. 77 cytoplasmic antibody-associated small-vessel vasculitis.
(8): 1150. Ann. Intern. Med. 143 (9): 621–631.
97 Charles, P., Terrier, B., Perrodeau, E. et al. (2018). 107 Kemna, M.J., Damoiseaux, J., Austen, J. et al. (2015).
Comparison of individually tailored versus fixed-schedule ANCA as a predictor of relapse: useful in patients with
rituximab regimen to maintain ANCA-associated renal involvement but not in patients with nonrenal
vasculitis remission: results of a multicentre, randomised disease. J. Am. Soc. Nephrol. 26 (3): 537.
controlled, phase III trial (MAINRITSAN2). Ann. Rheum. 108 Hruskova, Z., Geetha, D., and Tesar, V. (2015). Renal
Dis. 77 (8): 1143–1149. transplantation in anti-neutrophil cytoplasmic antibody-
98 Stegeman, C.A., Cohen Tervaert, J.W., de Jong, P.E., and associated vasculitis. Nephrol. Dial. Transplant. 30
Kallenberg, C.G.M. (1996). Trimethoprim– (Suppl 1): i159–i163.
sulfamethoxazole (co-trimoxazole) for the prevention of 109 Little, M.A., Hassan, B., Jacques, S. et al. (2009). Renal
relapses of Wegener’s granulomatosis. N. Engl. J. Med. 335 transplantation in systemic vasculitis: when is it safe?
(1): 16–20. Nephrol. Dial. Transplant 24 (10): 3219–3225.
99 Zycinska, K., Wardyn, K.A., Zielonka, T.M. et al. (2009). 110 Allen, A., Pusey, C., and Gaskin, G. (1998). Outcome of
Co-trimoxazole and prevention of relapses of PR3-ANCA renal replacement therapy in antineutrophil
positive vasculitis with pulmonary involvement. Eur. J. cytoplasmic antibody-associated systemic vasculitis.
Med. Res. 14 (Suppl 4): 265–267. J. Am. Soci. Nephrol. 9 (7): 1258–1263.
478
29
Paraprotein-associated Kidney Disorders

Vincent Javaugue1,2,3, Frank Bridoux1,2,3, and Nelson Leung4
1
Department of Nephrology, Centre Hospitalier Universitaire et Université de Poitiers, Poitiers, France
2
Department of Immunology CNRS UMR7276, Université de Limoges, Limoges, France
3
Centre de Référence Amylose AL et Autres Maladies par Dépôt d’Immunoglobulines Monoclonales, Université de Poitiers, Poitiers, France
4
Division of Nephrology, Hematology, Department of Internal Medicine Mayo Clinic, Rochester, MN, USA
I ntroduction treatment for the tumor burden. These are currently

referred to as monoclonal gammopathy of renal signifi-
Monoclonal gammopathies consist of a heterogenous cance (MGRS) [6, 7]. In this setting, structural peculiari-
group of disorders defined by the presence in serum and/or ties of the paraprotein, particularly the variable domain,
urine of a paraprotein produced by an abnormal B-cell and not the rate of production are the main determinants
clone. The clone usually derives from plasma cells when of renal lesions [6, 8]. The spectrum of renal diseases in
the paraprotein is IgG, IgA, IgD, or light chain only and MGRS is vast and can be classified according to the locali-
CD20+ lymphocytic clones including lymphoplasmacytic zation of renal lesions (glomerular, tubular, vascular,
clones when it involves an IgM. In its advanced state, mon- etc.), and to the composition and the pattern of ultras-
oclonal gammopathy may indicate symptomatic malignant tructural organization of deposits. Tubular lesions include
disorder such as multiple myeloma, Waldenström mac- light-chain proximal tubulopathy related to endolysoso-
roglobulinemia, or other non-Hodgkin lymphomas. More mal engorgement of proximal tubular cells by light chains
frequently, the clone can remain quiescent for a prolonged and crystal-storing histiocytosis defined by crystalline
period of time defining monoclonal gammopathy of unde- inclusions within interstitial histiocytes. Glomerulopathies
termined significance (MGUS) [1]. are featured either by organized deposits which can be
Renal complications associated with monoclonal gam- fibrillar (Ig-related amyloidosis), microtubular (type I and
mopathy are frequent. They can be separated into two cat- type II cryoglobulinemic glomerulonephritis, immuno-
egories according to the characteristics of the underlying tactoid glomerulopathy), or crystalline deposits (crystal-
B-cell clone. The first group of renal diseases always globulinemic glomerulonephritis), or by nonorganized
occurs in the setting of high tumor mass with production deposits (monoclonal immunoglobulin deposition dis-
of large amounts of monoclonal immunoglobulins. The ease [MIDD] and proliferative glomerulonephritis with
prime example is light-chain cast nephropathy, which monoclonal immunoglobulin deposits [PGNMID]). The
almost invariably complicates high tumor mass B-cell dis- third subgroup includes glomerular lesions not induced
order, particularly symptomatic multiple myeloma [2–4]. by the deposition of the monoclonal immunoglobulin but
Although rare, another example is glomerular deposition by its functional activity. The most common is C3 glomer-
of monoclonal IgM during high tumor mass lymphoplas- ulopathy associated with monoclonal gammopathy where
macytic lymphoma (Waldenström macroglobuline- deposits are composed of C3 complement fraction only,
mia) [5]. The second group includes all renal disorders consecutive to local or systemic activation of the comple-
caused by a monoclonal immunoglobulin secreted by a ment alternative pathway (CAP) by the monoclonal
nonmalignant B-cell clonal disorder that does not require immunoglobulin [9, 10].
Pathophysiolog 479
D
iagnostic Evaluation and efficient chemotherapy targeting the causal B-cell
clone are mandatory to improve renal prognosis and
The diagnosis of monoclonal immunoglobulin-related kidney patient survival [22].
diseases almost always requires a kidney biopsy. The only
exception is Ig-related amyloidosis, where the diagnosis could
be made from other tissue [11]. Renal pathological diagnosis P
athophysiology
relies on careful examination of kidney biopsy samples by
light microscopy, immunofluorescence studies, and electron The kidney disease may complicate an already diagnosed
microscopy [9, 12]. At a minimum, immunofluorescence clonal B-cell disorder or reveal a previously unknown
studies must include antibodies against kappa and lambda hematological disease. Renal disease in B-cell
light chains to determine monoclonality. Determination of lymphoproliferative disorders may result from two main
IgG subclasses would help support the monoclonal nature of pathophysiological mechanisms that can even coexist in
the immunoglobulin deposits, especially if a monoclonal the same patient, i.e. the intrinsic nephrotoxicity of the
immunoglobulin could not be identified in the blood or urine. secreted monoclonal immunoglobulin and malignant
However, IgG subclass restriction without light-chain restric- infiltration of the renal parenchyma by malignant B-cells,
tion is not considered to be monoclonal as it can occur in poly- which will not be addressed further in this chapter. The
clonal immunoglobulin deposition disease [7]. Antibodies physicochemical properties of monoclonal
against heavy chain constant domains may be needed in cases immunoglobulin appear to be the main determinant of
of heavy chain deposition disease (HCDD) to look for heavy their nephrotoxicity as well as influencing the topography
chain truncation [13]. In some cases, additional studies are and pattern of kidney lesions. Deposition of all or part of
required to identify the nature of deposits, such as immuno- the monoclonal immunoglobulin is the most common
electron microscopy or laser microdissection followed by mechanism and this can occur in various ultrastructural
mass spectrometry [14–17]. patterns (amorphous, crystalline, fibrillar, or microtubular)
Hematological workup is crucial for clonal identification of deposits or inclusions. Other mechanisms include
because the same kidney lesion can occur in different B-cell complement activation, autoantibody activity or cytokine
disorders, and treatment differs according to the plasma- secretion induced by the monoclonal immunoglobulin
cytic or lymphocytic nature of the clone. Bone marrow aspi- (Table 29.1) [10, 39].
ration and biopsy are required, including the use of sensitive
techniques (flow cytometry, molecular analysis) to better Nephrotoxic Monoclonal Free Light Chains
identify small clones [7]. Peripheral blood flow cytometry is
helpful in identifying clones of chronic lymphocytic leuke- Several mechanisms of nephrotoxicity of free light chains
mia or monoclonal B-cell lymphocytosis [18]. (FLCs) have been described. These include extracellular
Extrarenal manifestations can occur in some monoclo- deposition in AL amyloidosis (organized deposits) and light-
nal immunoglobulin-related kidney diseases and may be chain deposition disease (LCDD) (nonorganized deposits),
clinically significant. It is important to identify the involve- and precipitation/crystallization in cast nephropathy and
ment and impairment of the extrarenal organs so these can light-chain proximal tubulopathy [8, 40]. Under physiologi-
be managed appropriately. Cardiac evaluation with cal conditions, polyclonal FLCs, as other low molecular
echocardiogram and cardiac biomarkers such as N-terminal weight proteins, are freely filtered by the glomerulus and
pro-brain natriuretic peptide (NT-proBNP) and troponin T reabsorbed in proximal tubular cells by the megalin/cubilin
are standard of care in patients with Ig-related amyloido- multiligand receptor complex [41]. FLCs are then degraded
sis [19, 20]. This is also recommended in patients with in endolysosomal vesicles and only a minute amount is pre-
MIDD that have cardiac-related symptoms [21]. Liver sent in the final urine [42]. During symptomatic multiple
involvement can be detected by liver function test and myeloma large amounts of monoclonal FLCs are filtered by
imaging studies. Nerve involvement is not uncommon in the glomerulus, overwhelming the reabsorption capacity of
amyloidosis and can involve both peripheral nerves and the endocytic receptors megalin/cubilin in the proximal
the autonomic nervous system [20]. Assessment of these tubule. This allows the monoclonal FLCs to flow into the
impairments is important as it can affect therapeutic strate- loop of Henle. Cast nephropathy results from the interaction
gies by altering drug elimination and tolerability of adverse of these FLCs with uromodulin (Tamm–Horsfall protein) in
effects [22]. lumen of the thick ascending limp of the loop of Henle,
Because paraprotein-associated renal disorders are het- leading to tubular obstruction and acute kidney injury
erogeneous, management requires collaborative work (AKI). Uromodulin is the most abundant protein excreted in
between hematologists and nephrologists. Early diagnosis the urine under physiological conditions, but its complete
Table 29.1 Summary of the main pathological and hematological findings in paraprotein associated renal disorders.
Localization of
renal lesions Kidney disease Mechanism of renal toxicity Ultrastructural appearance Characteristics of monoclonal gammopathy Reference
Glomerular Ig-related amyloidosis Deposition Fibrils λ > κ, IgM<10% in AL amyloidosis [19, 23]
Truncated γ in AH amyloidosis
LC + truncated HC in AHL amyloidosis
MIDD Deposition Nonorganized κ > λ in LCDD (usually Vκ1 or Vκ4) [13, 21, 24]
Truncated HC (mostly γ) in HCDD
LC + truncated HC in LHCDD
PGNMID Deposition Nonorganized Mostly IgG3 (κ > λ) [25]
Immunotactoid GN Deposition Microtubules Mostly IgG1 (κ > λ) [26, 27]
Type I cryoglobulinemia Deposition/precipitation Microtubules/crystals Mostly IgG or IgM [28]
Type II cryoglobulinemia Deposition Curved short tubular IgM acting as a rheumatoid factor [12]
C3 glomerulopathy CAP activation structures Nonorganized IgG [29–31]
Tubular Cast nephropathy Precipitation Crystals κ > λ [4, 32]

Proximal tubulopathy Precipitation Mostly crystals κ (>90%, mostly Vκ1) [33, 34]
Cristal-storing histiocytosis Precipitation Crystals κ [35, 36]
Vascular Thrombotic microangiopathy CAP activation or cytokine Microvascular endothelial IgG in atypical hemolytic-uremic syndrome [37, 38]
mediated or unknown cell injury IgA = 50% in POEMS syndrome (λ ~ 100%)
AL, amyloid light chain; AH, amyloid heavy chain; AHL, amyloid light and heavy chain; CAP, complement alternative pathway; HC, heavy chain; HCDD, heavy chain deposition disease; Ig,
immunoglobulin; LC, light chain; LCDD, light chain deposition disease; LHCDD, light and heavy chain deposition disease; MIDD, monoclonal immunoglobulin deposition disease; PGNMID,
proliferative glomerulonephritis with monoclonal immunoglobulin deposits; POEMS, polyneuropathy-organomegaly-endocrinopathy-monoclonal protein-skin changes; Vκ, variability
subgroup of kappa light chain.
0005152417.INDD 480 09-12-2022 15:56:13

Pathophysiolog 481
function remains incompletely understood [43]. The bind- of the first constant (CH1) domain [13, 62, 63]. The
ing affinity of FLCs for uromodulin is influenced by the nephrotoxic heavy chain is often undetectable by elec-
molecular structure of the third complementarity- trophoresis and immunofixation. In this context, immu-
determining region (CDR3) that specifically recognizes a noblotting is a useful technique to detect small amounts
nine amino-acid domain of uromodulin [44–47]. Facilitating of truncated heavy chains. However, this technique is
factors that increase light-chain precipitation and/or their not widely available. In HCDD, cationic charges of the
interaction with uromodulin have been well described in heavy chain variable domain probably favor the forma-
conditions such as dehydration, hypercalcemia, nonsteroi- tion of amorphous deposits along renal basement
dal drug use, furosemide, and acidic urine [48–50]. Kidney membranes [13].
injury in cast nephropathy is also perpetuated by interstitial
inflammation and fibrosis induced by the monoclonal FLCs.
In vitro studies and animal experiments have shown that Nephrotoxic Monoclonal Entire
excessive endocytosis of myeloma light chains triggers oxi- Immunoglobulins
dative stress, activation of NF-κB, and release of inflamma-
Mechanisms of toxicity of entire immunoglobulins in
tory mediators (IL-6, IL-8, MCP-1, and TGF-β) leading to
PGNMID, immunotactoid glomerulonephritis, and cry-
tubulo-interstitial inflammation, fibrosis, and morphologic
oglobulinemic glomerulonephritis are poorly under-
alteration of the tubular epithelium [48, 51, 52].
stood. These disorders are mostly induced by monoclonal
Most other types of kidney injury related to pathogenic
IgG1 or IgG3 known to induce complement activation
light chains occur in patients with small B-cell or plasma cell
through the classical pathway [9, 39]. Immunoglobulin
clones with low amounts of monoclonal immunoglobulin [6,
deposits usually coexist with C3 and sometimes C1q and
10]. In this context, the kidney toxicity is not driven by the
C4 deposits [64, 65]. Complement activation through
tumor burden but is related to structural peculiarities of the
the classical, lectin or alternative pathway is probably a
monoclonal FLC [8]. For example, light-chain proximal tub-
main determinant of glomerular hypercellularity and
ulopathy is mostly related to kappa light chains that are
inflammation in these disorders. Experimental studies
restricted to the Vκ1 subgroup and derive from only two ger-
in mice have shown that in cryoglobulinemic glomeru-
mline genes (Vκ1-33 and Vκ1-39). Some mutations in the
lonephritis the pattern of kidney lesions depends on
variable domain confer a resistance to proteolysis, leading to
peculiarities of the molecular structure of pathogenic
the formation of crystals within the endolysosomal compart-
IgG [66, 67].
ment of proximal tubular cells [53]. Accumulation of these
light chains in proximal tubules inhibits receptor-mediated
endocytosis and provokes urinary leak of phosphate, uric
Complement and Cytokine Activation
acid, glucose, amino acids, and low-molecular-weight pro-
Associated with Monoclonal Gammopathy
teins, with proximal (type 2) renal tubular acidosis [54]. On
the other hand, most light chains involved in renal AL- Recent studies have shown that the incidence of mono-
amyloidosis are derived from two germline Vλ genes (Vλ6-57 clonal gammopathy in patients >50 years old with C3
and Vλ3-1), but no common structural peculiarities have glomerulopathy is high [29, 68–70]. This association is
been identified. However, germline gene usage influences not fortuitous as suggested by the presence in some cases
the organ tropism with predominant kidney involvement in of autoantibody activity against CAP regulatory proteins
Vλ6-57 patients and more common cardiac involvement in (anti-factor H antibody and C3Nef) and by the beneficial
Vλ1-44 patients [55–57]. Structural peculiarities of light effect of clone-directed therapy [30]. However, the mech-
chains have been also reported in LCDD with overrepresen- anism of complement activation is different in the major-
tation of the Vκ4 and Vκ1 subgroup of variability, abnormal ity of patients and may involve bystander polyclonal
glycosylation, and high isoelectric point values of variable antibodies directed against CAP regulatory proteins or
domain complementarity determining regions, possibly direct activation of the CAP by the monoclonal immuno-
accounting for tissue deposition along the anionic proteogly- globulin [71]. Cytokine activation is probably involved in
cans of the basement membranes [21, 58–61]. thrombotic microangiopathy-like lesions in polyneurop-
athy, endocrinopathy, organomegaly, monoclonal gam-
mopathy, and skin changes (POEMS) syndrome. This
Nephrotoxic Monoclonal Free Heavy Chains
disease is characterized by striking elevation of serum
The deposition of free monoclonal heavy chains is a rare vascular endothelial growth factor (VEGF) levels that
condition seen in AH amyloidosis and HCDD. The most probably account for glomerular endothelial cell
striking feature of deposited heavy chains is the deletion injury [37, 72].
482 Paraprotein-associated Kidney Disorders
D
iagnosis (a) (b)
Tubulointerstitial Lesions
Light-chain Cast Nephropathy
AKI is present at diagnosis in approximately 30% of patients
with symptomatic multiple myeloma. AKI is mostly related
to light-chain cast nephropathy, a myeloma defining event
invariably associated with high tumor mass [73]. Apart
from multiple myeloma, light-chain cast nephropathy has
been also described in Waldenström macroglobulinemia,
chronic lymphocytic leukemia, and other lymphomas [64,
74–77]. Patients with cast nephropathy typically present
with severe AKI, sometimes requiring dialysis. AKI is iso- Figure 29.1 Myeloma cast nephropathy by light microscopy. (a)
lated or associated with general symptoms related to high Masson’s trichrome staining (×200). Representative example of
tumor mass such as skeletal pains related to lytic bone cast nephropathy with numerous typical polychromatophilic and
fractured casts (arrows) associated with interstitial fibrosis and
lesions. Urine protein electrophoresis typically shows inflammation. (b) Hematoxylin and eosin staining (×400). Typical
prominent light-chain proteinuria with albuminuria cast (arrow) with giant cell (asterisk) and tubulorrhexis.
<10% [78, 79]. In around 50% of patients, cast nephropathy
is triggered by facilitating factors, namely dehydration, kappa isotype. The presence of hypouricemia and
hypercalcemia, infection, or nephrotoxic drugs (nonsteroi- hypophosphatemia must encourage further explorations to
dal anti-inflammatory agents, diuretics, angiotensin- detect proximal tubule dysfunction. However, signs of
converting enzyme inhibitors or angiotensin receptor tubular dysfunction may be absent in patients with more
blockers) [32, 78, 80]. The use of modern contrast media advanced kidney impairment. The complete form of proxi-
has not been associated with cast nephropathy in recent mal tubulopathy, referred to as light-chain associated
large series [78]. The risk of cast nephropathy is influenced Fanconi syndrome, is defined by the urinary leak of low
by the amount of monoclonal light chains excreted in the molecular weight proteins, phosphate, uric acid, glucose,
urine and is particularly high when light-chain proteinuria and amino acids. Patients who exhibit only some of these
exceeds 2 g/day [81]. features are considered to have a partial Fanconi syndrome.
Histologically, the cast nephropathy is defined by the Patients may also present with prominent bone symptoms,
presence of periodic acid Schiff (PAS) stain negative and including stress fractures, secondary to hypophosphatemic
fractured casts in distal tubule lumina, often with crystal- osteomalacia [33, 83]. Thus, searching for monoclonal
line appearance and usually accompanied by giant cell gammopathy is mandatory in adults with symptoms of
reaction and tubulorrhexis (Figure 29.1). Severe interstitial proximal tubule dysfunction.
inflammation is often present and plays a role in the devel- The kidney biopsy shows typically rhomboid-shaped
opment of irreversible fibrosis and chronic kidney injury. hypereosinophilic and PAS-negative crystals within proxi-
The presence of numerous casts has been found to be pre- mal tubular cells. Their detection by light microscopy is
dictive of a poor renal prognosis [32]. On immunofluores- sometimes difficult, highlighting the diagnostic value of
cence, the casts stain brightly with anti-kappa or immunofluorescence and ultrastructural examination,
anti-lambda antibody. Ultrastructural examination reveals which should be systematically considered in the presence
the crystalline organization of the casts [12]. of suggestive lesions such as focal proximal tubule dedif-
ferentiation and atrophy (Figure 29.2) [33, 84, 85]. In some
Light-chain Proximal Tubulopathy cases, antigenic epitopes normally recognized by antibod-
Light-chain proximal tubulopathy rarely occurs in the con- ies routinely used may be sequestered by the crystalline lat-
text of symptomatic multiple myeloma or overt lymphoid tice, resulting in false negative results. Ancillary techniques
malignancy, and most cases fall into the spectrum of are then needed, including immunofluorescence on
MGRS [33, 82]. The frequency is low and light-chain proxi- paraffin-embedded tissue after protease digestion or immu-
mal tubulopathy represents 5% of Ig-related renal diseases. noelectron microscopy [85, 86]. Light-chain proximal tub-
The median age at diagnosis is around 60 years, with a ulopathy may also manifest with accumulation of light
slight male predominance [33, 34]. Patients typically pre- chains within proximal tubular cells without crystal forma-
sent with slowly progressive chronic kidney failure with tion. These patients often present with evidence of tubular
tubular proteinuria and Bence Jones proteinuria usually of injury, but Fanconi syndrome is rare [34, 87, 88].
Glomerular Lesion 483
deposits with dichroism and birefringence under polar-

(a) (b) ized light [93]. Since more than 35 proteins have been
described as amyloidogenic precursors, characterization
of the native protein is the cornerstone of management.
Immunofluorescence analysis of frozen tissues, using
anti-light and anti-heavy chain antibodies should be pre-
ferred to immunohistochemistry, which has lower diag-
nosis sensibility and specificity [94]. Minimally invasive
biopsies (salivary glands, abdominal fat) should be per-
formed as first-line. When tissue biopsies fail to demon-
strate amyloid deposition or are insufficient for amyloid
typing, biopsy of a clinically affected organ (kidney, liver,
Figure 29.2 Kappa light chain proximal tubulopathy. (a) Light
gastrointestinal tract, or endomyocardial tissue) should
microscopy (toluidine blue staining, ×1000). Patient with massive be considered. When a kidney biopsy is performed in
cytoplasmic crystalline accumulation in proximal tubular cells. patients with renal involvement, it allows characteriza-
(b) Electron microscopy (×50 000). Abundant rhomboid crystals tion of amyloid deposits in most cases (Figure 29.3). The
(arrows) surrounded by a single membrane within the cytoplasm
of proximal tubular cells.
incidence of bleeding complications in systemic AL-
amyloidosis after kidney biopsy appears acceptable when
performed by experience physicians [95]. Deposits that
Crystal-storing Histiocytosis
stain positive for lambda light chain in most cases, pre-
Crystal-storing histiocytosis is a rare complication of
dominately in the mesangium and along the glomerular
symptomatic myeloma or MGRS, defined by the accumu-
basement membranes, are often associated with intersti-
lation of light-chain crystals in histiocytes. These histio-
tial and vascular deposits. Electron microscopy may be
cytes are commonly seen in the bone marrow but can be
useful to confirm the presence of amyloid deposits, par-
found in other sites such as kidney, perirenal fat, lungs,
ticularly in earlier disease stages, and typically shows ran-
cornea, and joints. Crystal-storing histiocytosis may coex-
domly arranged fibrils of 7–10 nm in external diameter
ist with light-chain proximal tubulopathy and/or cast
that initially predominate in the external aspect of the
nephropathy [35, 36].
glomerular basement membrane. When the nature of the
deposits remains indeterminate with routine technics,
additional explorations must be carried out, such as
Glomerular Lesions immunoelectron microscopy or mass spectrometry after
laser microdissection. [9]
Glomerular Disorders with Organized
All organs can be affected in immunoglobulin-related
Immunoglobulin Deposits
amyloidosis. Kidney involvement is the most frequent,
Immunoglobulin-related Amyloidosis found in two-thirds of patients at diagnosis. The heart is
Immunoglobulin-related amyloidosis is the most com- the most common extrarenal organ involved followed by
mon glomerular lesion associated with monoclonal gam- peripheral nerves and liver. Macroglossia, periorbital pur-
mopathy [23]. The average age at diagnosis is 65 years, pura or skeletal muscle pseudohypertrophy are particu-
with a slight male predominance (approximatively 60% of larly suggestive symptoms but are rarely present [19].
cases). Among this group, AL-amyloidosis is the most fre- Renal manifestations include glomerular symptoms in
quent subtype where amyloid is composed of monoclonal most cases with abundant proteinuria predominantly
light chains (usually lambda). In rare cases, composed of albumin, usually without hematuria or
immunoglobulin-related amyloidosis is derived from hypertension. Nephrotic syndrome and chronic kidney
fragments of heavy chains and light chains (AHL) or disease (CKD) are observed in approximately 50% of cases
heavy chains only (AH). AL-amyloidosis is the prototype at diagnosis. When the deposits are predominantly tubu-
of MGRS, since most patients present at diagnosis with an lointerstitial and vascular, chronic interstitial nephritis
isolated monoclonal gammopathy or indolent B-cell with polyuria due to defective urine concentration ability
clone [9]. Progression to symptomatic myeloma is unu- and slowly progressive CKD with little proteinuria is pos-
sual [89, 90]. AL-amyloidosis complicates lymphoplasma- sible [96]. Organ involvement in systemic AL amyloidosis
cytic proliferation with secretion of monoclonal IgM in is now defined by consensus criteria, updated at the
less than 10% of patients [91, 92]. The diagnosis of amy- 2010 meeting of the International Society of Amyloidosis
loidosis is based on the demonstration of Congo-red in Rome [20].
(a) (b)
(c) (d)
Figure 29.3 AL-amyloidosis. (a and b) Light microscopy (Congo red staining, ×200). Glomerular mesangial and capillary wall Congo
red positive deposits (a) with characteristic dichroism and apple green birefringence under polarized light (b). On immunofluorescence
(×200) glomerular amyloid deposits stain brightly for lambda light chain (c) with negative staining for kappa light chain (not shown).
(d) By electron microscopy (×12 000, insert ×50 000), amyloid deposits are composed of haphazardly oriented fibrils ranging from 7 to
14 nm in diameter.
Immunotactoid Glomerulonephritis Cryoglobulinemic Glomerulonephritis

Immunotactoid glomerulonephritis, also referred to as glo- Cryoglobulins are immunoglobulins that precipitate
merulonephritis with organized microtubular monoclonal under cold exposure. Three types exist but only two con-
immunoglobulin deposits (GOMMID), is a very rare entity tain monoclonal immunoglobulin. Type 1 is composed
(fewer than 100 cases in the literature) characterized by of a single type of monoclonal immunoglobulin (usually
Congo red negative deposits that show typical ultrastruc- IgG or IgM) and type 2 (mixed) contains classically a
tural organization into microtubules of 10–60 nm in exter- monoclonal IgM acting as a rheumatoid factor against
nal diameter, with a distinct hollow core at magnification polyclonal IgG [98]. The underlying condition is plasma
of less than 30 000. By definition, cryoglobulin testing cell dyscrasia or lymphoproliferative disorder, particu-
should be negative. Deposits are monotypic by immuno- larly lymphoplasmacytic lymphoma. Type II mixed cryo-
fluorescence, mostly composed of IgG1κ, usually associ- globulinemia is observed in around 10% of patients with
ated with C3. Light microscopic appearance is not specific, chronic virus C infection. Glomerulonephritis is more
but membranous nephropathy or membranoproliferative common with type II cryoglobulinemia. Median age is
glomerulonephritis patterns are commonly observed 62 years old and it is more common in women. [28]
(Figure 29.4). Most patients present with nephrotic range Renal presentation includes classically proteinuria and
proteinuria, hematuria, kidney insufficiency, and hyper- microscopic hematuria, sometimes with moderate
tension. Immunotactoid glomerulonephritis has been chronic kidney insufficiency. Severe hypertension is
mostly reported in patients with MGRS or chronic lympho- commonly observed. Some patients may present an
cytic leukemia with an average age of 60 years [26, 27, 97]. acute nephritic syndrome characterized by hematuria,
Glomerular Lesion 485
in the subendothelial space and intracapillary lumen,

(a) (b)
and appear focally organized (microtubular or “finger-
print” substructure) [9].
Glomerular Disorders with Nonorganized

Monoclonal Immunoglobulin Deposition Disease
Randall-type MIDD is the most common glomerulopathy
with nonorganized deposits. MIDD commonly occurs in the
context of a plasma cell disorder with criteria for sympto-
matic multiple myeloma in nearly 25% of cases and for
Figure 29.4 Immunotactoid glomerulonephritis. (a) Light MGRS (including smoldering myeloma in around 25% of
microscopy (Marinozzi’s silver staining, ×400). Membranous
pattern of injury associated with focal duplication of glomerular patients) in the remaining patients [21, 24, 99]. The diagno-
basement membranes (arrow). (b) By electron microscopy sis is made on approximately 0.5% of kidney biopsies on
(×50 000), the deposits are composed of microtubules with native kidneys. It is usually discovered around the age of
hollow centers and organized in parallel arrays (arrows). 60 years, with a slight male predominance [21, 24, 100].
According to the composition of deposits, three categories
usually macroscopic, severe proteinuria and hyperten- are distinguished: LCDD, HCDD, and light-and heavy-chain
sion, and acute kidney failure. Extrarenal manifesta- deposition disease (LHCDD). MIDD is characterized by lin-
tions are common and often present as purpura, ulcers, ear nonorganized Congo red-negative monoclonal immuno-
peripheral neuropathy, and arthralgia [28]. On light globulin deposits along basement membranes [9]. As
microscopy (Figure 29.5a,b), cryoglobulinemic glomeru- immunoglobulin-related amyloidosis, MIDD is a systemic
lonephritis shows membranoproliferative or endocapil- disease with prominent kidney manifestations, but heart,
lary proliferative glomerulonephritis with intracapillary liver, or peripheral nerve involvement may be encoun-
infiltrating monocytes and large PAS-positive intralumi- tered [21]. In the kidney, deposits are invariably observed
nal immune deposits (thrombi). Careful immunofluo- along tubular basement membranes, and in most cases in
rescence analysis is crucial to determine the type of the mesangium, glomerular basement membranes, and
cryoglobulinemia, as both IgM and IgG are present in around arteriolar myocytes. The most common histologic
type 2 cryoglobulinemia without clear light-chain feature is glomerulosclerosis mimicking diabetic nephropa-
restriction due to the presence of polyclonal IgG. By thy (Figure 29.6). The usual presentation is progressive kid-
electron microscopy (Figure 29.5c), deposits predominate ney failure and glomerular symptoms, including nephrotic
(a) (b) (c)
Figure 29.5 Cryoglobulinemic glomerulonephritis. (a and b) Light microscopy. (a) Hematoxylin and eosin staining, ×400.
Membranoproliferative glomerulonephritis pattern of injury characterized by mesangial expansion and hypercellularity and
duplication of glomerular basement membranes. (b) Masson’s trichrome staining, ×400. Intracapillary immune deposits suggestive of
thrombi (arrow), the hallmark histologic finding of cryoglobulinemic glomerulonephritis. (c) By electron microscopy (×50 000), the
deposits predominate in the subendothelial space with focal microtubular organization (arrow).
(a) (b)
(c) (d)
Figure 29.6 Light-chain deposition disease. (a and b) Light microscopy (periodic acid Schiff staining, ×400). Section of renal cortex
showing nodular glomerulosclerosis with nodular mesangial deposits and aneurysmal dilatation of the capillary lumens (a). Vascular
deposits around arteriolar myocytes (arrow) (b). (c) Immunofluorescence microscopy (anti-κ fluorescein isothiocyanate conjugate,
×200). Linear deposits along glomerular and tubular basement membranes, and around vascular myocytes. (d) Electron microscopy
(×15 000). Enlarged multilayered tubular basement membrane with electron-dense (arrows) powdery punctuate deposits
predominating in the outer aspect.
syndrome in half of patients. Uncommonly, MIDD may component by standard techniques [25, 103]. PGNMID
manifest with slowly progressive kidney disease without sig- appears as a renal limited disorder with prominent glomer-
nificant proteinuria. In this context, glomerulosclerosis is ular symptoms and frequent CKD. The risk of recurrence
uncommon and deposits predominate in the renal vascular after kidney transplantation has been reported to be as
and tubular compartments [101, 102]. LCDD occurring in high as 90% [104].
the context of symptomatic multiple myeloma usually coex-
ists with cast nephropathy and has similar presentation and
Glomerular Disorders without
outcome as pure cast nephropathy [21].
Proliferative Glomerulonephritis with Monoclonal C3 Glomerulopathy Associated with Monoclonal
Immunoglobulin Deposits Gammopathy
PGNMID is a rare glomerular disorder defined by nonor- C3 glomerulopathy regroups two entities characterized by
ganized immunoglobulin (most commonly IgG3) and com- glomerular C3 deposition without immunoglobulin: C3
plement component deposits in the glomerular capillary glomerulonephritis and dense deposit disease. Both enti-
walls and mesangium, mimicking an immune complex- ties are associated with dysregulation of the CAP. By elec-
mediated glomerulonephritis (Figure 29.7). Around 100 tron microscopy, dense deposit disease is distinguished
cases are reported in the literature. The average age at diag- from C3 glomerulonephritis by the presence of character-
nosis is 55 years old. Multiple myeloma is rare and around istic “sausage-shaped” intramembranous electron dense
two-thirds of patients have no detectable monoclonal deposits [105]. Patients typically present with glomerular
Treatmen 487
(a) (b)
(c) (d) (e)
Figure 29.7 Proliferative glomerulonephritis with monoclonal immunoglobulin deposits. (a) Light microscopy (hematoxylin and
eosin staining, ×400). Membranoproliferative glomerulonephritis with global mesangial and segmental endocapillary hypercellularity.
(b–d) Immunofluorescence microscopy. Mesangial and glomerular capillary wall staining for IgG3 (b) and kappa (c). Negative staining
for lambda (d). (e) Electron microscopy (×15 000). Granular nonorganized electron-dense deposits (arrows) predominant in the
subendothelial space of glomerular basement membrane.
symptoms (proteinuria and hematuria) and progressive multiple myeloma, and Waldenström macroglobuline-
CKD. In patients over 50 years of age, most cases of C3 mia, and (ii) glomerular microangiopathy associated with
glomerulopathy are associated with a circulating parapro- POEMS syndrome where hemolysis and thrombocytope-
tein [29, 31, 68, 70]. Although direct evidence of a link nia are absent [7].
between monoclonal gammopathy and C3 glomerulopa-
thy is lacking in most patients, the implication is sug-
gested by the presence in some cases of autoantibody
T
reatment
activity of the monoclonal immunoglobulin against CAP
regulatory proteins and by the beneficial effect on renal
Indication of Therapy
outcome of a clone-directed therapy [30, 70].
Indication of therapy in MGRS is driven by renal damage
Thrombotic Microangiopathy Associated due to the nephrotoxic immunoglobulin produced by the
with Monoclonal Gammopathy clone. Therapy in MGRS aims to improve renal survival
The term “thrombotic microangiopathy” refers to a and in selected patients to reduce the risk of recurrence
diverse group of disorders linked by a common histologic after renal transplantation. Treatment decisions should
finding of endothelial cell injury. Two categories associ- also take into account the presence of extrarenal manifes-
ated with monoclonal gammopathy have to be distinguish: tations and the patient’s status [22]. Light-chain cast
(i) microangiopathy with hemolytic anemia (MAHA) nephropathy seen only in malignant disease is not MGRS
which can occur in patients with MGRS, symptomatic and treatment is a medical emergency [106, 107].
Clone-directed Therapy nephrotoxicity. Hydration is essential with administra-

tion of hypotonic (half normal) saline to reduce aggrega-
The current therapeutic approach in MGRS is to eradi-
tion of Tamm–Horsfall protein and its interaction with
cate the underlying B-cell clone with chemotherapy. If
urine FLC [106, 107]. The use of adjuvant supportive
the clone is plasmacytic, usually secreting a monoclonal
measures to rapidly reduce the level of FLCs is still
IgG, IgA, or light chain, treatment should be based on
debated. Two randomized trials using high cutoff (HCO)
antimyeloma agents. The strategy mainly relies on bort-
dialyzer for the removal of serum FLC produced different
ezomib, usually combined with an alkylating agent and
results in terms of renal recovery. The European trial
dexamethasone (CyBorD regimen) that produces rapid
(EuLITE) had randomized patients to a standard dialysis
and deep hematological response. Bortezomib is easy to
with a high-flux dialyzer versus extended dialysis using a
use in patients with impaired renal function since it
HCO dialyzer in addition to bortezomib-based triplet
does not require dose adaptation. High-dose melphalan
chemotherapy. In this study a similar hemodialysis inde-
followed by autologous stem cell transplantation (ASCT)
pendence rate (60%) was observed in both groups [110].
may be a therapeutic option in some patients with
The French randomized clinical trial with biopsy-proven
plasma cell clone. However, mortality and morbidity
cast nephropathy reported that intensive HCO hemodial-
related to ASCT increase in patients with renal impair-
ysis does not significantly increase early hemodialysis-
ment, and some patients may worse their renal function
independence rate but results in significantly higher 6
during the procedure. Since melphalan is eliminated by
and 12 month-renal recovery rates compared to intensive
renal excretion, dose adaptation is needed in patients
hemodialysis with conventional high-flux dialyzers [78].
with CKD stage 3 or above. In case of relapsing or refrac-
Further larger studies are needed to confirm the benefice
tory disease, immunomodulatory drugs should be con-
of extracorporeal removal of FLCs in light-chain cast
sidered. Lenalidomide must be used with caution in
nephropathy.
patients with renal dysfunction since this drug is pre-
dominantly eliminated through renal excretion and may
sometimes worsen renal function with unclear mecha- Ig-related Amyloidosis
nism, particularly in patients with AL amyloidosis [108]. Symptomatic measures, adapted to organ failures, are
Thalidomide and pomalidomide are probably more essential in the management of Ig-related amyloidosis. In
appropriate since they do not require dose adaptation. patients with nephrotic syndrome, fluid retention and
Anti-CD38 monoclonal antibodies, which are highly edema require treatment with loop diuretics often given
effective in symptomatic multiple myeloma and are cur- at a high dose or combined with other diuretics. Caution
rently being evaluated in AL amyloidosis, are likely to be is required in the use of standard treatment of cardiac
increasingly used in the treatment of various MGRS [10, failure in patients with amyloidosis. Calcium inhibitors,
22]. If the clone is lymphoplasmacytic (usually produc- β-blockers, and digitalis are to be avoided due to excess
ing a monoclonal IgM) or lymphocytic, the treatment toxicity. Amiodarone should be considered as first-line
should be based on rituximab-containing regimens [22]. therapy for arrhythmia. Angiotensin converting enzyme
The choice and modality of chemotherapy in light-chain inhibitors can promote hypotension and are often poorly
cast nephropathy remains to be defined. The goal of tolerated. In patients with severe, life-threatening cardiac
therapy is to rapidly reduce the rate of production of involvement, pacemaker or defibrillator implantation
FLCs, a key factor to induce renal response [106, 107]. may be required [19, 90].
The use of high-dose steroids seems mandatory because
of their anti-inflammatory properties. Bortezomib-based
therapy is associated with high rate of renal response in Other Glomerular Disorders
light-chain cast nephropathy[109]. As in other glomerulonephritis, the use of angiotensin con-
verting enzyme inhibitors and/or angiotensin receptor
blockers is essential to control blood pressure and to reduce
Symptomatic Measures proteinuria.
Cast Nephropathy
Symptomatic treatment is a key step in the management Proximal Tubulopathy
of cast nephropathy. It is based on the rapid correction of Symptomatic measures to prevent osteomalacia are man-
precipitating factors such as treatment of infection with datory, including phosphate, calcium, and vitamin D sup-
non-nephrotoxic antibiotics and correction of hypercal- plementation. Prophylaxis against hypokalemia is often
cemia using bisphosphonates despite their potential necessary [33].
Reference 489
Kidney Transplantation C
onclusion
Recent advances in chemotherapy result in improved patient
survival. Selected patients with CKD stage V are now eligible The spectrum of renal complications associated with para-
for renal transplantation. Due to the risk of recurrence, proteins is heterogeneous and the diagnosis relies on care-
hematological remission seems mandatory before consider- ful examination of biopsy samples by light microscopy,
ing a renal transplant [104, 111–113]. The long-term results immunofluorescence studies, and in most cases electron
of renal transplantation in MGRS, particularly the risk of microscopy. Identification of the underlying B-cell clone is
infection and neoplasia, remain to be evaluated. crucial for treatment strategy.
R
eferences
1 Kyle, R.A., Larson, D.R., Therneau, T.M. et al. (2018). 12 Touchard, G. (2003). Ultrastructural pattern and
Long-term follow-up of monoclonal gammopathy of classification of renal monoclonal immunoglobulin
undetermined significance. N. Engl. J. Med. 378 (3): deposits. In: Monoclonal Gammopathies and the Kidney
241–249. (eds. G. Touchard, P. Aucouturier, O. Hermine and P.
2 Alexanian, R., Barlogie, B., and Dixon, D. (1990). Renal Ronco), 95–120. Dordrecht; Boston: Kluwer Academic
Failure in Multiple Myeloma. Arch. Intern. Med. 150 (8): Publishers. In.
1693–1695. 13 Bridoux, F., Javaugue, V., Bender, S. et al. (2017).
3 Hutchison, C.A., Batuman, V., Behrens, J. et al. (2012). Unravelling the immunopathological mechanisms of
The pathogenesis and diagnosis of acute kidney injury in heavy chain deposition disease with implications for
multiple myeloma. Nat. Rev. Nephrol. 8 (1): 43–51. clinical management. Kidney Int. 91 (2): 423–434.
4 Gonsalves, W.I., Leung, N., Rajkumar, S.V. et al. (2015). 14 Herrera, G.A., Paul, R., Turbat-Herrera, E.A. et al. (1986).
Improvement in renal function and its impact on survival Ultrastructural immunolabeling in the diagnosis of
in patients with newly diagnosed multiple myeloma. light-chain-related renal disease. Pathol. Immunopathol.
Blood Cancer J. 5: e296. Res. 5 (2): 170–187.
5 Morel-Maroger, L., Basch, A., Danon, F. et al. (1970). 15 Leung, N., Nasr, S.H., and Sethi, S. (2012). How I treat
Pathology of the kidney in Waldenström’s amyloidosis: the importance of accurate diagnosis and
Macroglobulinemia. N. Engl. J. Med. 283 (3): 123–129. amyloid typing. Blood 120 (16): 3206–3213.
6 Leung, N., Bridoux, F., Hutchison, C.A. et al. (2012). 16 Loo, D., Mollee, P.N., Renaut, P., and Hill, M.M. (2011).
Monoclonal gammopathy of renal significance: when Proteomics in molecular diagnosis: typing of amyloidosis.
MGUS is no longer undetermined or insignificant. Blood J. Biomed. Biotechnol. 2011: 1–9.
120 (22): 4292–4295. 17 Manabe, S., Hatano, M., Yazaki, M. et al. (2015). Renal
7 Leung, N., Bridoux, F., Batuman, V. et al. (2019). The AH amyloidosis associated with a truncated
evaluation of monoclonal gammopathy of renal immunoglobulin heavy chain undetectable by
significance: a consensus report of the international Immunostaining. Am. J. Kidney Dis. 66 (6): 1095–1100.
kidney and monoclonal Gammopathy research group. 18 Shanafelt, T.D., Ghia, P., Lanasa, M.C. et al. (2010).
Nat. Rev. Nephrol. 15 (1): 45. Monoclonal B-cell lymphocytosis (MBL): biology, natural
8 Sirac, C., Herrera, G.A., Sanders, P.W. et al. (2018). history and clinical management. Leukemia 24 (3):
Animal models of monoclonal immunoglobulin-related 512–520.
renal diseases. Nat. Rev. Nephrol. 14 (4): 246–264. 19 Desport, E., Bridoux, F., Sirac, C. et al. (2012). Al
9 Bridoux, F., Leung, N., Hutchison, C.A. et al. (2015). amyloidosis. Orphanet J. Rare Dis. 7: 54.
Diagnosis of monoclonal gammopathy of renal 20 Gertz, M.A., Comenzo, R., Falk, R.H. et al. (2005).
significance. Kidney Int. 87 (4): 698–711. Definition of organ involvement and treatment
10 Fermand, J.-P., Bridoux, F., Dispenzieri, A. et al. (2018). response in immunoglobulin light chain amyloidosis
Monoclonal gammopathy of clinical significance: a novel (AL): a consensus opinion from the 10th international
concept with therapeutic implications. Blood 132 (14): symposium on amyloid and amyloidosis, Tours, France,
1478–1485. 18-22 April 2004. Am. J. Hematol. 79 (4): 319–328.
11 Muchtar, E., Dispenzieri, A., Lacy, M.Q. et al. (2017). 21 Joly, F., Cohen, C., Javaugue, V. et al. (2019). Randall-type
Overuse of organ biopsies in immunoglobulin light chain monoclonal immunoglobulin deposition disease: novel
amyloidosis (AL): the consequence of failure of early insights from a nationwide cohort study. Blood.
recognition. Ann. Med. 49 (7): 545–551. 133(6):576–587.
2 Fermand, J.-P., Bridoux, F., Kyle, R.A. et al. (2013). How I

2 36 Stokes, M.B., Aronoff, B., Siegel, D., and D’Agati, V.D.
treat monoclonal gammopathy of renal significance (2006). Dysproteinemia-related nephropathy associated
(MGRS). Blood 122 (22): 3583–3590. with crystal-storing histiocytosis. Kidney Int. 70 (3):
23 Said, S.M., Sethi, S., Valeri, A.M. et al. (2013). Renal 597–602.
amyloidosis: origin and clinicopathologic correlations of 474 37 Ye, W., Wang, C., Cai, Q.-Q. et al. (2016). Renal
recent cases. Clin. J. Am. Soc. Nephrol. 8 (9): 1515–1523. impairment in patients with polyneuropathy,
24 Nasr, S.H., Valeri, A.M., Cornell, L.D. et al. (2012). Renal organomegaly, endocrinopathy, monoclonal
monoclonal immunoglobulin deposition disease: a report gammopathy and skin changes syndrome: incidence,
of 64 patients from a single institution. CJASN 7 (2): treatment and outcome. Nephrol. Dial. Transplant. 31
231–239. (2): 275–283.
25 Nasr, S.H., Satoskar, A., Markowitz, G.S. et al. (2009). 38 Ravindran, A., Go, R.S., Fervenza, F.C., and Sethi, S.
Proliferative glomerulonephritis with monoclonal IgG (2017). Thrombotic microangiopathy associated with
deposits. J. Am. Soc. Nephrol. 20 (9): 2055–2064. monoclonal gammopathy. Kidney Int. 91 (3): 691–698.
26 Bridoux, F., Hugue, V., Coldefy, O. et al. (2002). Fibrillary 39 Leung, N., Drosou, M.E., and Nasr, S.H. (2018).
glomerulonephritis and immunotactoid (microtubular) Dysproteinemias and glomerular disease. Clin. J. Am. Soc.
glomerulopathy are associated with istinct immunologic Nephrol. 13 (1): 128–139.
features. Kidney Int. 62 (5): 1764–1775. 40 Sanders, P.W. (2012). Mechanisms of light chain injury
27 Nasr, S.H., Fidler, M.E., Cornell, L.D. et al. (2012). along the tubular nephron. JASN 23 (11): 1777–1781.
Immunotactoid glomerulopathy: clinicopathologic and 41 Christensen, E.I., Verroust, P.J., and Nielsen, R. (2009).
proteomic study. Nephrol. Dial. Transplant. 27 (11): Receptor-mediated endocytosis in renal proximal tubule.
4137–4146. Pflugers Arch. -Eur. J. Physiol. 458 (6): 1039–1048.
28 Zaidan, M., Plasse, F., Rabant, M. et al. (2016). Renal 42 Batuman, V., Verroust, P.J., Navar, G.L. et al. (1998).
involvement during type 1 cryoglobulinemia. Nephrol. Myeloma light chains are ligands for cubilin (gp280). Am.
Ther. 12 (Suppl 1): S71–S81. J. Phys. 275 (2 Pt 2): F246–F254.
29 Zand, L., Kattah, A., Fervenza, F.C. et al. (2013). C3 43 Rampoldi, L., Scolari, F., Amoroso, A. et al. (2011). The
glomerulonephritis associated with monoclonal rediscovery of uromodulin (Tamm–Horsfall protein):
Gammopathy. Am. J. Kidney Dis. 62 (3): 506–514. from tubulointerstitial nephropathy to chronic kidney
30 Chauvet, S., Frémeaux-Bacchi, V., Petitprez, F. et al. disease. Kidney Int. 80 (4): 338–347.
(2017). Treatment of B-cell disorder improves renal 44 Ying, W.-Z. and Sanders, P.W. (2001). Mapping the
outcome of patients with monoclonal gammopathy– binding domain of immunoglobulin light chains for
associated C3 glomerulopathy. Blood 129 (11): 1437–1447. Tamm-Horsfall protein. Am. J. Pathol. 158 (5): 1859–1866.
31 Ravindran, A., Fervenza, F.C., Smith, R.J.H., and Sethi, S. 45 Ying, W.-Z., Allen, C.E., Curtis, L.M. et al. (2012).
(2018). C3 glomerulopathy associated with monoclonal Ig Mechanism and prevention of acute kidney injury from
is a distinct subtype. Kidney Int. 94 (1): 178–186. cast nephropathy in a rodent model. J. Clin. Investig. 122
32 Ecotière, L., Thierry, A., Debiais-Delpech, C. et al. (2016). (5): 1777–1785.
Prognostic value of kidney biopsy in myeloma cast 46 Huang, Z.Q., Kirk, K.A., Connelly, K.G., and Sanders,
nephropathy: a retrospective study of 70 patients. P.W. (1993). Bence Jones proteins bind to a common
Nephrol. Dial. Transplant. 31 (1): 64–72. peptide segment of Tamm-Horsfall glycoprotein to
33 Vignon, M., Javaugue, V., Alexander, M.P. et al. (2017). promote heterotypic aggregation. J. Clin. Invest. 92 (6):
Current anti-myeloma therapies in renal manifestations of 2975–2983.
monoclonal light chain-associated Fanconi syndrome: a 47 Huang, Z.Q. and Sanders, P.W. (1997). Localization of a
retrospective series of 49 patients. Leukemia 31 (1): single binding site for immunoglobulin light chains on
123–129. human Tamm-Horsfall glycoprotein. J. Clin. Invest. 99 (4):
34 Stokes, M.B., Valeri, A.M., Herlitz, L. et al. (2016). Light 732–736.
chain proximal Tubulopathy: clinical and pathologic 48 Sanders, P.W. and Booker, B.B. (1992). Pathobiology of
characteristics in the modern treatment era. J. Am. Soc. cast nephropathy from human Bence Jones proteins. J.
Nephrol. 27 (5): 1555–1565. Clin. Investig. 89 (2): 630–639.
35 El Hamel, C., Thierry, A., Trouillas, P. et al. (2010). 49 Sanders, P.W. (1994). Pathogenesis and treatment of
Crystal-storing histiocytosis with renal Fanconi myeloma kidney. J. Lab. Clin. Med. 124 (4): 484–488.
syndrome: pathological and molecular characteristics 50 Pesce, A.J., Clyne, D.H., Pollak, V.E. et al. (1980). Renal
compared with classical myeloma-associated Fanconi tubular interactions of proteins. Clin. Biochem. 13 (5):
syndrome. Nephrol. Dial. Transplant. 25 (9): 2982–2990. 209–215.
Reference 491
1 Guan, S., el-Dahr, S., Dipp, S., and Batuman, V. (1999).

5 64 Chauvet, S., Bridoux, F., Ecotière, L. et al. (2015). Kidney
Inhibition of Na-K-ATPase activity and gene expression diseases associated with monoclonal immunoglobulin
by a myeloma light chain in proximal tubule cells. J. M-secreting B-cell lymphoproliferative disorders: a case
Investig. Med. 47 (9): 496–501. series of 35 patients. Am. J. Kidney Dis. 66 (5): 756–767.
52 Sengul, S., Zwizinski, C., and Batuman, V. (2003). Role of 65 Sethi, S., Zand, L., Leung, N. et al. (2010).
MAPK pathways in light chain-induced cytokine Membranoproliferative glomerulonephritis secondary to
production in human proximal tubule cells. Am. J. monoclonal gammopathy. Clin. J. Am. Soc. Nephrol. 5 (5):
Physiol. Renal Physiol. 284 (6): F1245–F1254. 770–782.
53 Decourt, C., Rocca, A., Bridoux, F. et al. (1999). 66 Rengers, J.U., Touchard, G., Decourt, C. et al. (2000).
Mutational analysis in murine models for myeloma- Heavy and light chain primary structures control
associated Fanconi’s syndrome or cast myeloma IgG3 nephritogenicity in an experimental model for
nephropathy. Blood 94 (10): 3559–3566. cryocrystalglobulinemia. Blood 95 (11): 3467–3472.
54 Luciani, A., Sirac, C., Terryn, S. et al. (2016). Impaired 67 Galea, H.R., Bridoux, F., Aldigier, J.-C. et al. (2002).
lysosomal function underlies monoclonal light chain– Molecular study of an IgG1kappa cryoglobulin yielding
associated renal Fanconi syndrome. J. Am. Soc. Nephrol. organized microtubular deposits and glomerulonephritis
27 (7): 2049–2061. in the course of chronic lymphocytic leukaemia. Clin.
55 Perfetti, V., Casarini, S., Palladini, G. et al. (2002). Exp. Immunol. 129 (1): 113–118.
Analysis of Vλ-Jλ expression in plasma cells from 68 Lloyd, I.E., Gallan, A., Huston, H.K. et al. (2016). C3
primary (AL) amyloidosis and normal bone marrow glomerulopathy in adults: a distinct patient subset
identifies 3r(λIII) as a new amyloid-associated germline showing frequent association with monoclonal
gene segment. Blood 100 (3): 948–953. gammopathy and poor renal outcome. Clin. Kidney J. 9
56 Kourelis, T.V., Dasari, S., Theis, J.D. et al. (2017). (6): 794–799.
Clarifying immunoglobulin gene usage in systemic and 69 Sethi, S., Sukov, W.R., Zhang, Y. et al. (2010). Dense
localized immunoglobulin light-chain amyloidosis by deposit disease associated with monoclonal gammopathy
mass spectrometry. Blood 129 (3): 299–306. of undetermined significance. Am. J. Kidney Dis. 56 (5):
57 Comenzo, R.L., Zhang, Y., Martinez, C. et al. (2001). The 977–982.
tropism of organ involvement in primary systemic 70 Bridoux, F., Desport, E., Frémeaux-Bacchi, V. et al.
amyloidosis: contributions of Ig V L germ line gene use (2011). Glomerulonephritis with isolated C3 deposits and
and clonal plasma cell burden. Blood 98 (3): 714–720. monoclonal gammopathy: a fortuitous association? Clin.
58 Decourt, C., Cogné, M., and Rocca, A. (1996). Structural J. Am. Soc. Nephrol. 6 (9): 2165–2174.
peculiarities of a truncated V kappa III immunoglobulin 71 Chauvet, S., Roumenina, L.T., Aucouturier, P. et al.
light chain in myeloma with light chain deposition (2018). Both monoclonal and polyclonal immunoglobulin
disease. Clin. Exp. Immunol. 106 (2): 357–361. contingents mediate complement activation in
59 Denoroy, L., Déret, S., and Aucouturier, P. (1994). monoclonal gammopathy associated-C3 Glomerulopathy.
Overrepresentation of the V kappa IV subgroup in light Front. Immunol. 9: 2260.
chain deposition disease. Immunol. Lett. 42 (1–2): 63–66. 72 Soubrier, M., Sauron, C., Souweine, B. et al. (1999).
60 Preud’homme, J.L., Aucouturier, P., Touchard, G. et al. Growth factors and proinflammatory cytokines in the
(1994). Monoclonal immunoglobulin deposition disease renal involvement of POEMS syndrome. Am. J. Kidney
(Randall type). Relationship with structural Dis. 34 (4): 633–638.
abnormalities of immunoglobulin chains. Kidney Int. 46 73 Rajkumar, S.V., Dimopoulos, M.A., Palumbo, A. et al.
(4): 965–972. (2014). International myeloma working group updated
61 Kaplan, B., Livneh, A., and Gallo, G. (2007). Charge criteria for the diagnosis of multiple myeloma. Lancet
differences between in vivo deposits in immunoglobulin Oncol. 15 (12): e538–e548.
light chain amyloidosis and non-amyloid light chain 74 Vos, J.M., Gustine, J., Rennke, H.G. et al. (2016). Renal
deposition disease. Br. J. Haematol. 136 (5): 723–728. disease related to Waldenström macroglobulinaemia:
62 Aucouturier, P., Khamlichi, A.A., Touchard, G. et al. incidence, pathology and clinical outcomes. Br. J.
(1993). Brief report: heavy-chain deposition disease. N Haematol. 175 (4): 623–630.
Engl J Med. 329(19):1389–93. 75 Burke, J.R., Flis, R., Lasker, N., and Simenhoff, M. (1976).
63 Khamlichi, A.A., Aucouturier, P., Preud’Homme, J.-L., Malignant lymphoma with “myeloma kidney” acute
and CognÉ, M. (1995). Structure of abnormal heavy renal failure. Am. J. Med. 60 (7): 1055–1060.
chains in human heavy-chain-deposition disease. Eur. J. 76 Strati, P., Nasr, S.H., Leung, N. et al. (2015). Renal
Biochem. 229 (1): 54–60. complications in chronic lymphocytic leukemia and
monoclonal B-cell lymphocytosis: the Mayo Clinic chain proximal tubulopathy with and without crystal
experience. Haematologica 100 (9): 1180–1188. formation. Mod. Pathol. 24 (11): 1462–1469.
77 Sathick, I.J., Drosou, M.E., and Leung, N. (2018). 89 Merlini, G. and Stone, M.J. (2006). Dangerous small
Myeloma light chain cast nephropathy, a review. J. B-cell clones. Blood 108 (8): 2520–2530.
Nephrolo. [Internet] May 5 [cited 2019 Jan 8]; Available 90 Merlini, G., Seldin, D.C., and Gertz, M.A. (2011).
from: http://link.springer.com/10.1007/ Amyloidosis: pathogenesis and new therapeutic options.
s40620-018-0492-4. J. Clin. Oncol. 29 (14): 1924–1933.
78 Bridoux, F., Carron, P.-L., Pegourie, B. et al. (2017). Effect 91 Milani, P. and Merlini, G. (2016). Monoclonal IgM-
of high-cutoff hemodialysis vs conventional hemodialysis related AL amyloidosis. Best Pract. Res. Clin. Haematol.
on hemodialysis independence among patients with 29 (2): 241–248.
myeloma cast nephropathy: a randomized clinical trial. 92 Sachchithanantham, S., Roussel, M., Palladini, G. et al.
JAMA 318 (21): 2099–2110. (2016). European collaborative study defining clinical
79 Leung, N., Gertz, M., Kyle, R.A. et al. (2012). Urinary profile outcomes and novel prognostic criteria in
albumin excretion patterns of patients with cast monoclonal immunoglobulin M-related light chain
nephropathy and other monoclonal gammopathy-related amyloidosis. J. Clin. Oncol. 34 (17): 2037–2045.
kidney diseases. Clin. J. Am. Soc. Nephrol. 7 (12): 1964–1968. 93 Howie, A.J. and Brewer, D.B. (2009). Optical properties
80 Haynes, R.J., Read, S., Collins, G.P. et al. (2010). of amyloid stained by Congo red: history and
Presentation and survival of patients with severe acute mechanisms. Micron 40 (3): 285–301.
kidney injury and multiple myeloma: a 20-year 94 Picken, M.M. (2007). New insights into systemic
experience from a single Centre. Nephrol. Dial. amyloidosis: the importance of diagnosis of specific
Transplant. 25 (2): 419–426. type. Curr. Opin. Nephrol. Hypertens. 16 (3): 196–203.
81 Drayson, M., Begum, G., Basu, S. et al. (2006). Effects of 95 Fish, R., Pinney, J., Jain, P. et al. (2010). The incidence
paraprotein heavy and light chain types and free light of major hemorrhagic complications after renal biopsies
chain load on survival in myeloma: an analysis of in patients with monoclonal gammopathies. Clin. J. Am.
patients receiving conventional-dose chemotherapy in Soc. Nephrol. 5 (11): 1977–1980.
Medical Research Council UK multiple myeloma trials. 96 Eirin, A., Irazabal, M.V., Gertz, M.A. et al. (2012).
Blood 108 (6): 2013–2019. Clinical features of patients with immunoglobulin light
82 Messiaen, T., Deret, S., Mougenot, B. et al. (2000). Adult chain amyloidosis (AL) with vascular-limited deposition
Fanconi syndrome secondary to light chain gammopathy. in the kidney. Nephrol. Dial. Transplant. 27 (3):
Clinicopathologic heterogeneity and unusual features in 1097–1101.
11 patients. Medicine (Baltimore) 79 (3): 135–154. 97 Rosenstock, J.L., Markowitz, G.S., Valeri, A.M. et al.
83 Narvaez, J., Domingo-Domenech, E., Narvaez, J.A. et al. (2003). Fibrillary and immunotactoid
(2005). Acquired hypophosphatemic osteomalacia associated glomerulonephritis: distinct entities with different
with multiple myeloma. Joint Bone Spine 72 (5): 424–426. clinical and pathologic features. Kidney Int. 63 (4):
84 Herrera, G.A. (2014). Proximal tubulopathies associated 1450–1461.
with monoclonal light chains: the spectrum of 98 Brouet, J.C., Clauvel, J.P., Danon, F. et al. (1974).
clinicopathologic manifestations and molecular Biologic and clinical significance of cryoglobulins. A
pathogenesis. Arch. Pathol. Lab. Med. 138 (10): 1365–1380. report of 86 cases. Am. J. Med. 57 (5): 775–788.
85 Gu, X., Barrios, R., Cartwright, J. et al. (2003). Light chain 99 Lin, J., Markowitz, G.S., Valeri, A.M. et al. (2001). Renal
crystal deposition as a manifestation of plasma cell monoclonal immunoglobulin deposition disease: the
dyscrasias: the role of immunoelectron microscopy. Hum. disease spectrum. J. Am. Soc. Nephrol. 12 (7): 1482–1492.
Pathol. 34 (3): 270–277. 100 Sayed, R.H., Wechalekar, A.D., Gilbertson, J.A. et al.
86 Nasr, S.H., Galgano, S.J., Markowitz, G.S. et al. (2006). (2015). Natural history and outcome of light chain
Immunofluorescence on pronase-digested paraffin deposition disease. Blood 126 (26): 2805–2810.
sections: a valuable salvage technique for renal biopsies. 101 Sicard, A., Karras, A., Goujon, J.-M. et al. (2014). Light
Kidney Int. 70 (12): 2148–2151. chain deposition disease without glomerular
87 Kapur, U., Barton, K., Fresco, R. et al. (2007). Expanding proteinuria: a diagnostic challenge for the nephrologist.
the pathologic spectrum of immunoglobulin light chain Nephrol. Dial. Transplant. 29 (10): 1894–1902.
proximal tubulopathy. Arch. Pathol. Lab. Med. 131 (9): 102 Cohen, C., Royer, B., Javaugue, V. et al. (2015). Bortezomib
1368–1372. produces high hematological response rates with
88 Larsen, C.P., Bell, J.M., Harris, A.A. et al. (2011). The prolonged renal survival in monoclonal immunoglobulin
morphologic spectrum and clinical significance of light deposition disease. Kidney Int. 88 (5): 1135–1143.
Reference 493
03 Nasr, S.H., Markowitz, G.S., Stokes, M.B. et al. (2004).

1 AL amyloidosis. Nephrol. Dial. Transplant. 26 (3):
Proliferative glomerulonephritis with monoclonal IgG 881–886.
deposits: a distinct entity mimicking immune-complex 109 Kastritis, E., Anagnostopoulos, A., Roussou, M. et al.
glomerulonephritis. Kidney Int. 65 (1): 85–96. (2007). Reversibility of renal failure in newly diagnosed
104 Said, S.M., Cosio, F.G., Valeri, A.M. et al. (2018). multiple myeloma patients treated with high dose
Proliferative glomerulonephritis with monoclonal dexamethasone-containing regimens and the impact of
immunoglobulin G deposits is associated with high rate novel agents. Haematologica 92 (4): 546–549.
of early recurrence in the allograft. Kidney Int. 94 (1): 110 Hutchison CA, Cockwell P, Moroz V. et al. (2019). High
159–169. cutoff versus high-flux haemodialysis for myeloma cast
105 Sethi, S., Nester, C.M., and Smith, R.J.H. (2012). nephropathy in patients receiving bortezomib-based
Membranoproliferative glomerulonephritis and C3 chemotherapy (EuLITE): a phase 2 randomised
glomerulopathy: resolving the confusion. Kidney Int. 81 controlled trial. Lancet Haematol. 6(4):e217–e228.
(5): 434–441. 111 Angel-Korman, A., Stern, L., Sarosiek, S. et al. (2019).
106 Bridoux, F. and Fermand, J.-P. (2012). Optimizing Long-term outcome of kidney transplantation in AL
treatment strategies in myeloma cast nephropathy: amyloidosis. Kidney Int. 95 (2): 405–411.
rationale for a randomized prospective trial. Adv. 112 Leung, N., Lager, D.J., Gertz, M.A. et al. (2004).
Chronic Kidney Dis. 19 (5): 333–341. Long-term outcome of renal transplantation in light-
107 Manohar, S., Nasr, S.H., and Leung, N. (2018). Light chain deposition disease. Am. J. Kidney Dis. 43 (1):
chain cast nephropathy: practical considerations in the 147–153.
Management of Myeloma Kidney—What we Know and 113 Herrmann, S.M.S., Gertz, M.A., Stegall, M.D. et al.
What the future may hold. Curr. Hematol. Malig. Rep. 13 (2011). Long-term outcomes of patients with light chain
(3): 220–226. amyloidosis (AL) after renal transplantation with or
108 Specter, R., Sanchorawala, V., Seldin, D.C. et al. (2011). without stem cell transplantation. Nephrol. Dial.
Kidney dysfunction during lenalidomide treatment for Transplant. 26 (6): 2032–2036.
494
30
Primary Immune Complex-mediated Membranoproliferative Glomerulonephritis

and C3 Glomerulopathies: A Clinical Approach
Fadi Fakhouri
Department of Nephrology and Immunology, Centre Hospitalier Universitaire de Nantes, Nantes, France
I ntroduction deposits in the glomerular (and tubular) basement mem-

brane, called “dense deposits disease” (DDD) [4], and C3G
Primary immune complex-mediated membranoprolifera- without such deposits, called C3 glomerulonephritis
tive glomerulonephritis (IC-MPGN) and C3 glomerulopa- (C3GN) [5]. The precise composition of dense deposits
thies (C3G) have attracted a substantial body of basic and remains elusive. They probably contain complement acti-
clinical research in the last decade. This has been mainly vation products (C9) but not predominantly C3 [2].
due to the recent dynamics in the field of complement On light microscopy study, C3G display variable patterns
study and the development of an increasing number of of renal pathological findings: a typical aspect of MPGN in
complement modulators. These nephropathies have been half of the cases [6], a “mesangial” pattern (mild expansion
recently reclassified with the individualization of a new of the mesangial matrix, no “double contour”) [5], a cres-
subtype, C3 glomerulonephritis (C3GN), and experimental centic pattern [7, 8], and in some instances a pathological
and clinical studies have dissected complement pathway presentation very similar to the one noted in acute post-
involvement in their pathogenesis. Nevertheless, the rarity infectious glomerulonephritis (exsudative endocapillary
of theses renal diseases and the absence to date of prospec- proliferation) [6].
tive controlled trials are clear limitations to an “evidence- These definitions have several limitations. First, the patho-
based” approach to these glomerulopathies. logical MPGN pattern is not specific of primary MPGN. It is
also encountered in a wide variety of secondary MPGN [9],
ranging from systemic disease-associated MPGN (lupus and
D
efinitions cryoglobulinemic nephritis, Sjögren’s disease nephropathy)
to infection-related glomerulopathy (hepatitis C), but also
Primary IC-MPGN is classically defined by peculiar pathol- chronic thrombotic microangiopathy or DGK (Diacyl-
ogy findings in kidney biopsy [1]: (i) on light microscopy glycerol kinase) epsilon-associated nephropathy [10].
study, a mesangial matrix expansion and “double contour” Second, the pathological patterns of IC-MPGN and C3G vary
aspects due to a duplication of the glomerular basement from one patient to another but also over time in a given
membrane, and (ii) on immunofluorescence study, a coexist- patient in terms of the relative abundance of C3 and Ig depos-
ence, in equal intensity, of immunoglobulins (mainly IgG), its [11], but also and more importantly in terms of mesangial,
and complement components (mainly C3) deposits. IC- endocapillary, and/or extracapillary proliferation.
MPGN are subdivided in two types: MPGN type I with mesan-
gial and endomembranous immune deposits and MPGN type
III with endo and extramembranous immune deposits. E
pidemiology
C3G are characterized by exclusive or predominant ( 2+
intensity than any other type of deposit) glomerular C3 Primary IC-MPGN and C3G are rare types of nephropathy.
deposits [2, 3]. C3G are divided into two subtypes based on In a recent series from the United States [12], MPGN types I
electron microscopy study findings: C3G with electron-dense and III represented between 2.5% and 4.5% of all glomerular
Clinical Presentation and Outcom 495
diseases diagnosed in kidney biopsies. DDD accounted for way and the generation of the membrane attack complex.
less than 0.5% of all glomerulopathies. The incidence of The alternative complement pathway is in a state of low-
these diseases was stable between 1986 and 2015. In grade continuous activation (rapidly amplified upon infec-
another study from the United States [12], less than 0.3% of tions) and thus requires tight control. This control targets
all patients reaching end-stage renal disease (ESRD) predominantly the C3 convertase with three main inhibi-
between 1996 and 2013 had MPGN (probably combining tors that prevents the formation of the C3 convertase
IC-MPGN and C3G) as a primary renal diagnosis. C3GN is through the binding and inactivation of the C3b. These
a more recently (2010) [2, 5] individualized entity and no inhibitors include complement factor H (CFH), factor I
specific incidence is available. (FI), and membrane-cofactor protein (MCP). The loss of
the regulation of this C3 convertase in the circulation leads
to an increased release of the C3 degradation products
P
athophysiology (C3c, Cb, C3d) and their deposition in the kidney. This
pathophysiological hypothesis is supported by animal
IC-MPGN and C3G are prototypic complement-mediated models. Piglets of the Norwegian Yorkshire breed as well as
diseases, with an activation of the complement cascade ini- mice deficient in CFH, develop a C3GN reminiscent of the
tiated by distinct mechanisms. On one hand, it is assumed human disease [15, 16]. The CFH−/− mice model has pro-
that IC-MPGN results from the deposition in the glomeru- vided valuable insight into the pathogenesis of C3G and its
lus of immune complexes [9] that triggers the activation of dynamics, particularly regarding the potential clearance of
the classical complement pathway, starting with activation renal complement deposits following the restoration of
of C1, C4, and C2 leading to the formation of the classical tight control of the C3 alternative convertase [17].
pathway C3 convertase (C4b2a) which cleaves C3, gener- In patients with C3G dysregulation of the alternative C3
ates a classical C5 convertase (C4b2a3b), and ultimately convertase arises rarely (<10% [18]) from inactivating
leads to the activation of C5 and of the complement termi- mutations in the genes coding for CFH, FI, or MCP or acti-
nal pathway. C5b generation leads to the assembly of the vating mutations in the genes coding for the two main
membrane attack complex (C5b, 6, 7, 8, 9), which induces components of the C3 convertase, C3 and factor B.
cellular injury. IC deposition triggers mesangial cells pro- Nevertheless, C3G and to a lesser extent MPGN patients
liferation, and in some cases tissular injury leads to inflam- have a range of autoantibodies that target the alternative
matory cells influx, release of proteases and reactive oxygen complement pathway, particularly the C3 convertase. The
species, capillary wall necrosis, and extracapillary prolifer- most frequent and extensively documented type of autoan-
ation. “Double contour” aspects have been long been tibody is the C3 nephritic factor (C3Nef), an IgG that stabi-
attributed to an expansion of the mesangial matrix that dis- lizes and prolongs the half-life of the C3 convertase. More
sociates the two leaflets of the glomerular basement mem- recently, a C5nephrtic factor has been identified that stabi-
brane. The pathogenesis of “double contour” may, however, lizes the C5 convertase leading to an excessive activation of
be more complex and is not fully understood. the C5 and of the complement terminal pathway [19].
The assumption that IC deposition is the primary trigger Anti-CFH, anti-C3, and anti-factor B antibodies have also
of IC-MPGN is challenged by pathological findings in been reported in patients with C3G and more rarely IC-
repeat biopsies in patients with this type of nephropathy. Ig MPGN [18, 20, 21].
deposits vary widely from one biopsy to another in a given
patient and may even disappear following treatment,
whereas C3 deposits usually persist [11, 13]. Besides, com- Clinical Presentation and Outcome
plement classical pathway activation, exemplified by C4d
deposition, has been documented in only few cases of pri- The available data regarding the clinical presentation and
mary IC-MPGN [14]. Finally, patients with IC-MPGN fre- outcome of C3G and IC-MPGN derive from retrospective
quently display features of alternative pathway activation series which included a substantial number of patients in
(see infra). regard of the rarity of the disease. The four largest recent
On the other hand, C3G is believed to derive from a dys- series from France [5], the United States [22], the United
regulation of the alternative C3 convertase in the circula- Kingdom [23], and Italy [24] included between 80 and 173
tion. C3 convertase (C3bBb) is a key enzyme of the patients. Two series included only C3G patients [22, 23]
complement alternative pathway (CAP) that cleaves C3, and the remaining two C3G and IC-MPGN cases [5, 24].
leading to the formation of the C5 convertase and ulti- What conclusions can be drawn from the published
mately to the activation of the complement terminal path- studies?
496 Primary Immune Complex-mediated Membranoproliferative Glomerulonephritis and C3 Glomerulopathies: A Clinical Approach
First, C3G is a severe renal disease. The median age of Renal Pathological Findings
affected patients ranges from 12–15 (DDD) to 28 years [5,
22, 24] and 38–40% of affected patients have a disease onset Renal pathological features are obviously different in each
in childhood [18, 23]. It is associated with a nephrotic syn- subtype of MPGN. They are also variable from one patient
drome in 16–41% of cases and progresses to ESRD in to another within the same subgroup of MPGNs and in a
20–41% of cases after a median follow-up of 28 months to given patient from one kidney biopsy to another [11]. The
11 years [18, 20, 22, 24]. In addition, the disease recurs in paradox of these nephropathies is that a renal disease clas-
up to 40–60% of the kidney grafts [18], but some recur- sified as MPGN may not disclose on light microscopy study
rences are subclinical and do not affect the survival of the typical features of MPGN. This is particularly true for C3G.
renal graft. The same severity of disease has been reported In a recent series from the United States [22], 61% of C3GN
in patients with IC-MPGN [24]: 43–65% present with a and only 35% of DDD cases had an MPGN pattern.
nephrotic syndrome, 44% progress to chronic kidney dis- Seventeen percent of C3GN and 35% of DDD cases had a
ease and up to 40% reach ESRD. Forty-two percent of mesangial proliferative pattern, and 13% and 15% had a dif-
patients have a recurrence of the nephropathy in the renal fuse endocapillary proliferative pattern reminiscent of the
graft [18]. one found in acute post-infectious glomerulonephritis.
The available data indicate that DDD, C3GN as well as Besides, cellular or fibro-cellular crescents were found in
IC-MPGN share the same severe renal presentation and 17% of C3GN cases and 10% of DDD cases. This wide spec-
outcome. DDD pattern was associated with an increased trum of pathological findings in C3G has also been out-
risk of ESRD in two studies [18, 23] but not in the three lined by previous studies [23].
remaining ones [22, 24]. Thus, the clinical relevance of the IC-MPGN cases also disclose variable patterns of
current classification of C3G and IC-MPGN remains pathological findings. One of the most striking findings
debated. This is in keeping with the results of a retrospec- is the variability of the relative abundance of C3, C1q,
tive study performed by Little and colleagues [25]. The and immunoglobulin (IgM, IgG, and IgA) deposits in
authors retrospectively analyzed 70 cases of DDD and IC- repeat biopsies in patients with IC-MPGN [11]. In some
MPGN diagnosed between 1972 and 1995, and thus before instances, IgG deposits may even disappear while C3
the individualization of C3GN. After a mean follow-up of deposits persist [31], a finding that may question the
13 years, 59% of patients progressed to ESRD. The median pathogenic relevance of immunoglobulin deposits in
time to ESRF was 8.3 years. In multivariate analysis, the IC-MPGN. In addition, data documenting an activation
severity of interstitial fibrosis, crescent formation, and of the classical complement pathway by IC within the
mesangial proliferation was independently associated with kidney in patients with IC-MPGN remain scarce. In a
ESRD. Age and MPGN type did not predict renal survival. study [14], C4d (a marker of alternative pathway activa-
Regarding renal transplantation, younger age at initial tion) staining was assessed in 18 biopsy specimens of
diagnosis and the presence of crescents on the biopsy of immune-complex GN (including three primary IC-
native kidneys were independently associated with recur- MPGN), 30 biopsy specimens of C3GN, and 13 biopsy
rence of the disease in the renal graft in multivariate analy- specimens of post-infectious GN. In 15 out of 18 biop-
sis. MPGN type was not associated with recurrence. Thus, sies of IC-MPGN, bright (2–3+) C4d staining was
intensity of inflammatory (crescents) and fibrotic changes detected, whereas only six (20%) cases of C3GN had
in the kidney biopsy predict better the renal outcome of only trace/1+ C4d staining. C4d staining was negative
MPGN and C3G than the usual pathological classification in six (46%) of 13 biopsies of acute postinfectious
of C3G and IC-MPGN [11]. glomerulonephritis.
Second, the clinical course of the disease is variable
from one patient to another and most strikingly in a
given patient over time. Most C3G patients present with
chronic subacute forms of the disease characterized
L
aboratory Work-up
mainly by persistent nephrotic syndrome and mildly
Complement Assays
altered renal function, with 40–50% of patients progress-
ing gradually toward ERSD. Some patients experience C3G is a prototypic complement-mediated disease.
acute deterioration of their nephropathy (worsening pro- Similarly, it is assumed that renal injury in IC-MPGN is
teinuria and/or renal function) that may be triggered by mediated by complement activation triggered by immune
infections [24, 26] and a minority (8% of C3G cases in complex deposition [9]. However, features of systemic
one series [22]) present with a rapidly progressing form activation of complement cascade are invariably found in
of the disease [27–30]. patients with MPGN. For instance, a low C3 plasma level
Laboratory Work-u 497
is noted in only 36–74% of patients with C3GN [18, 22–24] of the CAP. It is present in 78–86% of DDD patients,
and in 15–84% of DDD patients [18, 22–24]. In IC-MPGN 38–45% C3GN, and 40–53% of IC-MPGN [18, 24]. The
patients, C3 levels are low in 46–76% of cases [18, 24]. more recently characterized C5 nephritic factor
This is probably related to the fact that assays of circulat- (C5Nef) [19], an autoantibody that stabilizes the C5
ing components of CAP do not necessarily reflect activa- convertase, is present in 49% of C3G patients. C5Nef is
tion of complement within the renal tissue. Thus, normal more frequently associated with C3GN than DDD (88%
complement component levels do not exclude the diagno- vs. 33% [19]) and its presence correlates with the level
sis of C3G and IC-MPGN. Interestingly, a low C4 plasm of soluble C5b-9. Other types of autoantibodies have
level (a marker of classical pathway activation) is present been reported in MPGN (mostly C3G) patients: anti-
in a minority of patients with IC-MPGN (2% [18]) and the CFH antibodies (7–12% of patients [19, 20]), anti-C3
majority (70% [24]) of these patients have normal C4 and antibodies (2–3% [19, 34]), and anti-Factor B antibodies
low C3, and thus features of alternative pathway activa- (2–3% [19, 34]).
tion rather than classical pathway activation. This is in Thus, MPGN appears to be related to an acquired rather
sharp contrast to secondary forms of IC-MPGN (lupus than constitutional dysregulation of the complement sys-
and cryoglobulinemia-associated nephritis) that are char- tem (mostly the CAP). However, autoantibodies detected
acterized by decreased C4 serum levels. in these patients interfere with the CAP regulation in vitro
Another activation product of the complement path- but their clinical relevance is still debated. In addition, the
way, soluble C5b-9 (a marker of the terminal complement presence of autoantibodies that target the CAP in patients
pathway), has been assessed in patients with MPGN, with IC-MPGN suggests that these nephropathies are
especially since the availability of the C5 blocker, eculi- driven not only by an activation of the classical comple-
zumab. In a large series of 127 cases of C3G [19], soluble ment pathway (triggered by immune complexes) but also
C5b-9 levels were elevated in 60% of patients. However, by a dysregulation of the CAP.
the clinical relevance of this marker, particularly in the
prediction of a clinical response to C5 blockade, remains
uncertain (see infra).
Monoclonal Gammopathy
The incidence of complement gene novel or rare vari-
ants is relatively low in patients with MPGN [18, 20, 22, Several studies have outlined an association between
24], 16% in IC-MPGN, 16–33% in DDD, and 12–25% in monoclonal gammopathy and C3GN [35–37]. In the larg-
C3GN, especially if compared to the incidence of such est series of 201 patients with C3G, 29% had monoclonal
variants in another complement-mediated renal disease, gammopathy (excluding cryoglobulinemia). As expected,
hemolytic uremic syndrome (40–70%) [32]. The most fre- the prevalence of monoclonal gammopathy in patients
quently reported variants in MPGN patients are in the with C3G increases with age: 36%, 59%, and 94% for the
CFH gene (19–17%). Some of these variants lead to a total age ranges of 50–59, 60–69, and 70–79 years, respec-
absence or a decrease in CFH plasma level while others tively [37]. Even though this study may carry a recruit-
are believed to affect the regulatory function of the pro- ment bias, the association between C3GN and monoclonal
tein (cluster of variants in the SCR1-3 mediating the gammopathy is increasingly recognized. However, there
interaction of CFH with C3b) [18, 20, 24]. Variants in the is to date no definite proof that monoclonal gammopathy
FI (0–6%), C3 (0–8%), factor B (0–2%), and MCP (0–2%) per se interferes with complement dysregulation and
and thrombomodulin (0–1%) coding gens have been more hence is involved in the pathogenesis of the disease.
rarely reported in IC-MPGN and C3G patients [18, 20, 22, Patients with monoclonal-associated C3GN have features
24]. The pathogenicity and thus clinical relevance have of CAP activation (low C3 plasma level in 44% of patients)
not been documented for the majority of these detected but also carry polyclonal autoantibodies interfering with
variants A peculiar type of C3G, highly prevalent in the the CAP regulation: C3Nef in 7% of cases (vs. 45% in all
Chypriote population, has been linked to internal dupli- C3GN), anti-FH antibodies in 17% of cases (vs. 11%), and
cation of the gene coding for complement factor H-related even anticomplement factor I and anticomplement recep-
protein 5 (CFHR-5) leading to a decreased binding of the tor 1 antibodies [38]. In the majority of tested cases
mutant protein to C3b [33]. (77%) [38], these autoantibodies do not share heavy and
In contrast, autoantibodies that interfere with the light chain characteristics with the monoclonal gammop-
regulation of the CAP are frequently detected in C3G athy. Thus direct proof of a link between monoclonal
and IC-MPGN patients. The most documented type of gammopathy and C3GN is still lacking. The induction of
autoantibodies is the C3Nef, which stabilizes the alter- the remission of C3GN with chemotherapy is only indi-
native C3 convertase and thus amplifies the activation rect evidence with potential bias [37].
oward a Global Approach for C3G

T median age of 27 years. Median serum creatinine (SCr) was
and IC-MPGN 1.4 mg/dl, median estimated glomerular filtration rate
(eGFR) was 66 ml/min/1.73 m2, median proteinuria was
A global approach that takes into account several clinical 3.8 g/24 h, and 52% of patients had a nephrotic syndrome at
and biological parameters is probably the most appropriate baseline. The patients were divided into three groups:
for C3G and IC-MPGN. Such approach has been recently a) 22 patients received MMF and steroids. The median ini-
developed by Iatropoulos and colleagues [24] using the tial dose of MMF was 1 g/day and the median treatment
Italian registry of C3G and IC-MPGN. Their study aimed to duration was 18 months.
classify patients based on 24 variables: clinical (gender, b) 18 patients received immunosuppressors other than
age, triggering event at onset, familiarity for nephropathy), MMF: steroids alone (n = 11), steroids, and cyclophos-
biological (hematuria, proteinuria, renal impairment), phamide (n = 7).
pathological (degree of renal inflammation and fibrosis, c) 20 patients received no immunosuppressors.
type and intensity of immune deposits), and those related
to complement work-up (markers of complement activa- All patients were treated with conventional antihyper-
tion, presence of complement gene variants). They identi- tensive and antiproteinuric drugs. The median follow-up
fied four clusters of patients each including patients with was 47 months. Partial or complete clinical remission was
similar presentation and renal outcome. Patients with C3G achieved in 86% of patients treated with MMF and steroids,
and IC-MPGN were distributed across the four subtypes. 50% of patients were treated with the other immunosup-
One cluster had a worse renal outcome as compared to the pressors and 25% of patients with no specific treatment
other clusters (renal survival at 10-year follow-up of 55% (P = 0.018). A doubling of SCr occurred in 0%, 39%, and
versus 85–90%). This cluster was characterized by (i) a low 35% of patients in the three groups, respectively (P = 0.002).
frequency of decreased C3 level, increased C5b-9 levels, The percentages of patients reaching ESRD (16-month fol-
complement genes variants, and inflammatory changes in low-up) were 0%, 16%, and 35% in the three groups respec-
the kidney and (ii) marked renal fibrotic changes. However, tively (P = 0.08). Finally, six (27%) patients treated with
this difference in terms of renal survival may be related MMF + Cs showed a proteinuria relapse after MMF dose
mainly to the intensity of fibrosis in the kidney and the tapering. This study, the first to suggest the efficacy of
absence of renal inflammatory changes that could be MMF and steroids in the treatment of C3G, carries several
amended by treatment with steroids and immunosuppres- limitations. It was retrospective and included a limited
sors. Nevertheless, such an elegant global approach may be number of patients. Pathological assessment of the kidney
the key for the design of targeted individualized treatment biopsies performed in included patients took into account
of C3G and IC-MPGN. the extent of fibrotic changes but not the inflammatory
changes that are a major predictor of response to treatment
and outcome. Furthermore, patients with significant renal
fibrosis and limited inflammatory changes and thus a less
Treatment of C3G and IC-MPGN active disease were probably more likely to not receive
immunosuppressors. For instance only 20% of patients
There is to date no specific treatment for C3G and IC-MPGN who received no immunosuppressor had a nephrotic syn-
as no available commercialized drug targets the initial and drome at baseline as compared to 77% and 55% of patients
main driver of the disease, C3 and IC deposits. The currently treated with MMF or another immunosuppressor. Finally,
used drugs aim mostly to control the renal inflammation the analysis of the individual cases of the 22 C3G patients
(crescents, mesangial/endocapillary proliferation) that may treated with mycophenolate and steroids suggests that
arise in some patients. In addition, the clinical boundaries some of these patients had rapidly progressing glomerulo-
between C3G and IC-MPGN are unclear and thus it is nephritis, a subtype of MPGN highly likely to respond to an
unclear whether treatments of these two types of MPGN anti-inflammatory treatment.
should differ. Finally, there is to date no large prospective The second study from the United States [40] was also
controlled trials assessing treatments options in C3G and IC- retrospective but was based on a single registry. It included
MPGN. One can only rely on open-label small prospective 30 patients with C3G and treated with MMF among 120
studies or retrospective series (Table 30.1). C3G patients diagnosed between 1997 and 2015. The
The most assessed treatment in MPGN patients (mostly median age of MMF-treated patients was 25 years, median
C3G) is mycophenolate mofetil (MMF) [39, 40], with two SCr was 1 mg/dl, and median proteinuria/creatininuria
published series. The first, from Spain, was retrospective ratio was 3.2 g/g. The median dose of mycophenolate was
and multicentric. It included 60 patients with C3G with a 2 g/day and the median time on therapy was 24 months.
Table 30.1 Main series and prospective trials that assessed treatment options in patients with C3G and IC-MPGN.
Level of
References Type of study Drug assessed Number of patients Main results Limitations evidence
MMF in C3G
Rabasco Retrospective, Cs + MMF Total: n = 60 Partial or complete clinical remissiona achieved in Retrospective and multicentric. Low
et al [39] national, Other IS Cs + MMM (n = 22) 86% of MMF + Cs patients, 50% of patients with Limited number of included patients.
multicentric. (Cs ± CYP) Other IS (n = 18) other IS, and 25% of patients with no IS (P = 0.018).
Patients with significant renal fibrosis
No IS (n = 20) A doubling of SCr occurred in 0%, 39%, and 35% of and limited inflammatory changes
patients in the three groups, respectively (0.002). were probably more likely to not
receive IS.
The percentages of patients reaching ESRD
(16 months follow-up) were 0%, 16%, and 35% in the Some patients treated with Cs + MMF
three groups, respectively (P = 0.08). had RPGN, a subtype of MPGN highly
likely to respond to an anti-
Six (27%) patients treated with MMF + Cs showed a inflammatory treatment.
proteinuria relapse after MMF dose tapering.
Avarese Retrospective, Cs + MMF Total: n = 73 67% of patients treated with MMF + Cs were Retrospective study. Limited number of Low
et al [40] single register. Cs + various IS Cs + MMF (n = 30) classified as respondersb to treatment as compared patients in each treatment group.
Cs alone (n = 23) to 39%, 33%, 29%, and 29% of those treated with Cs
alone, Cs + CYP, CNI, and RTX. Patients with significant renal fibrosis
Cs + CNI (n = 7)
and limited inflammatory changes
Cs + RTX (n = 7)
Patients who responded to MMF + Cs tended to have were probably more likely to not
Cs + CYP (n = 6) lower initial proteinuria and higher soluble receive IS.
membrane attack complex levels as compared to
nonresponders (not statistically significant). Comparison between responders and
nonresponders to MMF (and between
Fifty percent of patients who responded to MMF and other IS) did not take into
eculizumab had a proteinuria relapse after account the presence or absence of
treatment tapering/discontinuation. inflammatory changes in the kidney.
MMF in IC-MPGN
Jones Retrospective Cs + MMF Cs + MMF (n = 5) In the MMF + Cs group, proteinuria decreased from Retrospective with very limited Absent
et al [41] Compared to seven 5 g/day at baseline to 1.9 g/24 h (P = 0.003) at number of patients. Relatively low dose
untreated patients 6 months, 1.96 g/24 h (P = 0.003) at 12 months, and of MMF (mean dose of 1 g/day) and
(six with IC-MPGN 2.59 g/24 h (P = 0.015) at 18 months. high dose of Cs (60 mg/day initially).
and one with DDD) No change in proteinuria over 18 months in
occurred in the control group (MMF + Cs vs.
controls, P = 0.03). In the MMF + Cs group, serum
albumin concentration increased from 29 g/l at
baseline to 36.6 g/l at 6 months, 38.2g/l at 12 months,
and 37 g/l after 18 months.
(Continued)
0005152418.INDD 499 09-12-2022 16:01:20

Level of
References Type of study Drug assessed Number of patients Main results Limitations evidence
Eculizumab in C3G
Le Quintrec Retrospective, Eculizumab 26 patients (13 23% of patients had a global clinical response, 23% a Retrospective design without a control Low
et al [27] multicentric children/ partial clinical response, and 54% no responsec. group, relatively small number of
adolescents) with Compared with those who had a partial clinical or cases. Inclusion of pediatric and adult
C3G no response, patients who had a global clinical cases.
response had lower estimated glomerular filtration
rates, a more rapidly progressive course, and more
extracapillary proliferation in kidney biopsy.
Bomback Prospective, Eculizumab. Six patients: three After 12 months of treatment, two patients had a Noncontrolled. Very limited number of Low
et al [42] open-label, with DDD (one decrease in serum creatinine, one patient had a patients.
noncontrolled, with a recurrence in significant reduction in proteinuria. Three patients
and the renal graft), had no clinical response, including one with an
monocentric three with C3GN improvement in inflammatory changes in the
(including two with kidney.
a recurrence in the Changes in soluble C5b–9 serum levels paralleled
renal allograft). the improvement in serum creatinine and
proteinuria.
C3GN, C3 glomerulonephritis; CNI, calcineurin inhibitor; Cs, corticosteroids; CYP, cyclophosphamide; DDD, dense deposits disease; IS, immunosuppressors; MMF, mycophenolate mofetil;
RPGN, rapidly progressing glomerulonephritis; RTX, rituximab; SCr, serum creatinine.
a
Complete remission was defined by eGFR of 460 ml/min/1.73 m2 (or a return to ±15% of baseline values in those with eGFR < 60 ml/min/1.73 m2) and proteinuria <0.5 g/24 h. Partial
remission was defined by a proteinuria reduction of >50% (and a proteinuria value of o3.5 g/day in patients with nephrotic range proteinuria at baseline), plus stabilization (±25%) or
improvement in renal function.
b
Complete remission of the disease was defined by a stable or improved eGFR with decline in proteinuria/creatininuria ratio to <0.5 g/g creatinine. Partial remission was defined by stable or
improved eGFR with >50% decline in proteinuria/creatinuria ratio to between 0.5 and 3.5 g/g Cr by urine protein-to-Cr ratio at last follow-up. Stable eGFR was defined as within 15% of
baseline. Patients were categorized as “responders” if they had either complete or partial remission concordant with MMF treatment.
c
Response to eculizumab was characterized as: (1) global clinical response, defined as (a) >50% decrease in serum creatinine concentration and urinary protein-creatinine ratio (UPCR) and
serum albumin concentration >3 g/dl or (b) the persistence of normal serum creatinine concentration (±10%) and albumin concentration >3 g/dl and normalization of UPCR (<150 mg/g), or
(2) partial clinical response, defined as (a) <50% decrease in serum creatinine concentration, albumin concentration > 3 g/dl, and >50% decrease in UPCR or (b) the persistence of normal
serum creatinine concentration (±10%), albumin concentration >3 g/dl, and >50% decrease in UPCR.
0005152418.INDD 500 09-12-2022 16:01:20

Conclusio 501
Mycophenolate mofetil was associated with steroids in 93% Compared with those who had a partial clinical or no
of cases. Twenty patients (67%) had a response to MMF and response, patients who had a global clinical response had
achieved partial or complete response to treatment. The lower eGFRs, a more rapidly progressive course, and more
median time to response was 10 months. Four out of eight extracapillary proliferation on kidney biopsy. Age, extent
patients who discontinued MMF had a proteinuria relapse. of renal fibrosis, frequency of nephrotic syndrome, low
The remission rate with MMF in combination with steroids serum C3 and C3Nef, and elevated soluble C5b-9 concen-
(67%) was higher than the one achieved with steroids alone trations or complement gene variants did not differ
(39%), steroids in combination with cyclophosphamide between responders and nonresponders.
(33%), calcineurin inhibitor (29%), and rituximab (29%). The second smaller series from the United States [42]
Mycophenolate mofetil has been assessed in a single included six patients: three patients with DDD (one patient
small retrospective series of IC-MPGN patients [41]. Five had a recurrence of the disease in the renal allograft) and
patients with IC-MPGN were treated with MMF (mean three patients with C3GN (two patients had a recurrence of
dose 1.1 g/day) and high-dose steroids. At baseline, their the disease in the renal allograft). The eculizumab regimen
median SCr was 1.1 mg/dl (mean creatinine clearance of was similar to the one used in atypical HUS. After
108 ml/min) and their mean proteinuria was 6 g/day. Their 12 months of treatment, two patients had a decrease in SCr
evolution was compared to that of seven untreated patients, and one patient had a significant reduction in proteinuria.
six with IC-MPGN and one with DDD. In the mycopheno- Three patients had no clinical response, including one with
late and steroids group, proteinuria decreased from 5 g/day an improvement in inflammatory changes in the kidney.
at baseline to 1.9 g/24 h (P = 0.003) at 6 months, 1.96 g/24 h Changes in soluble C5b-9 serum levels paralleled the
(P = 0.003) at 12 months, and 2.59 g/24 h (P = 0.015) at improvement in SCr and proteinuria. No series or trials
18 months. No change in proteinuria over 18 months in have to date assessed the use of eculizumab in IC-MPGN
occurred in the control group (MMF + steroids vs. controls, but one case report has suggested some efficacy of the
P = 0.03). In the MMF + steroids group, serum albumin drug [43].
concentration increased from 29 g/l at baseline to 36.6 g/l at No definitive conclusions can be drawn from this series
6 months, 38.2 g/l at 12 months, and 37 g/l after 18 months. regarding the use of eculizumab in C3G and IC-MPGN.
In all, the level of evidence of the usefulness of mycophen- Nevertheless, eculizumab does not appear to be the specific
ate mofetil in C3G is low and absent in IC-MPGN. long-awaited treatment for MPGN. The lack of a significant
Nevertheless, mycophenolate associated to a variable dose of impact of eculiuzmab on the course of MPGN may be due
steroids is increasingly used in patients with chronic/suba- at least in part to the fact that these nephropathies have
cute nephrotic patients with C3G and by analogy in patients two aspects. All patients with MPGN display immune
with IC-MPGN. deposits (C3 with or without immunoglobulins) and a vari-
The availability of the first complement inhibitor, the C5 able minority of patients have additional features of glo-
blocker, eculizumab, has initially raised a lot of interest in merular inflammation (mesangial and endocapillary
this drug as a potential treatment for MPGN. Eculizumab proliferation, crescents) that probably depend on the
has been assessed in two series of C3G patients, with very release of C5a and/or C5b. Eculizumab most probably
discrepant results. The largest one from France was retro- treats only one aspect of the disease (inflammatory
spective [27], noncontrolled, and included 26 patients, of changes) with no or very limited impact on C3 and immu-
whom 13 were children or adolescents. At the initiation of noglobulins deposits. Eculizumab may be a potent inflam-
eculizumab therapy, 42% patients had chronic kidney dis- matory drug in patients with inflammatory forms of C3G,
ease, 73% had nephrotic syndrome, 27% had rapidly pro- as suggested by the retrospective French series [27] where
gressive disease, and 12% required dialysis. Median time only patients with crescentic and rapidly progressing forms
from diagnosis to eculizumab treatment initiation was of C3G had a significant and cost-effective response to
16 months. The eculizumab regimen was similar to the one eculizumab.
used in atypical hemolytic uremic syndrome (HUS) [32]:
900 mg intravenously per week for 4 weeks and then
1200 mg intravenously on week 5 and every other week C
onclusion
afterward. At eculizumab treatment initiation, median SCr
concentration was 0.6 mg/dl in children/adolescents and Experimental and clinical studies have transformed the
2.3 mg/dl in adults, corresponding to a median eGFRs of perspective in the management of C3G and IC-MPGN. It
110 and 28 ml/min/1.73 m2, respectively. Twenty-three per- remains unknown, however, whether the distinction
cent of patients had a global clinical response, 23% had a between these two types of nephropathies within the group
partial clinical response; and 54% had no response. of MPGN is clinically relevant. There is to date no specific
treatment of C3G and IC-MPGN, which directly targets the lators that inhibit the activation of the alternative C3
initial driver of the disease, C3, and possibly immunoglob- convertase are currently being developed [44] and specific
ulins deposition. Nevertheless, several complement modu- treatment for C3G and IC-MPGN may be within our reach.
R
eferences
1 Fogo, A.B., Lusco, M.A., Najafian, B. et al. (2015). AJKD 15 Pickering, M.C., Cook, H.T., Warren, J. et al. (2002).
atlas of renal pathology: membranoproliferative Uncontrolled C3 activation causes membranoproliferative
glomerulonephritis. Am. J. Kidney Dis. 66: e19–e20. glomerulonephritis in mice deficient in complement
2 Fakhouri, F., Fremeaux-Bacchi, V., Noel, L.H. et al. factor H. Nat. Genet. 31: 424–428.
(2010). C3 glomerulopathy: a new classification. Nat. Rev. 16 Hogasen, K., Jansen, J.H., Mollnes, T.E. et al. (1995).
Nephrol. 6: 494–499. Hereditary porcine membranoproliferative
3 Pickering, M.C., D’Agati, V.D., Nester, C.M. et al. (2013). C3 glomerulonephritis type II is caused by factor H
glomerulopathy: consensus report. Kidney Int. 84(6):1079–89. deficiency. J. Clin. Invest. 95: 1054–1061.
4 Fogo, A.B., Lusco, M.A., Najafian, B. et al. (2015). AJKD 17 Fakhouri, F., de Jorge, E.G., Brune, F. et al. (2010).
atlas of renal pathology: dense deposit disease. Am. J. Treatment with human complement factor H rapidly
Kidney Dis. 66: e21–e22. reverses renal complement deposition in factor
5 Servais, A., Fremeaux-Bacchi, V., Lequintrec, M. et al. H-deficient mice. Kidney Int. 78: 279–286.
(2007). Primary glomerulonephritis with isolated C3 18 Servais, A., Noel, L.H., Roumenina, L.T. et al. (2012).
deposits: a new entity which shares common genetic risk Acquired and genetic complement abnormalities play a
factors with haemolytic uraemic syndrome. J. Med. Genet. critical role in dense deposit disease and other C3
44: 193–199. glomerulopathies. Kidney Int. 82: 454–464.
6 Nasr, S.H., Valeri, A.M., Appel, G.B. et al. (2009). Dense 19 Marinozzi, M.C., Chauvet, S., Le Quintrec, M. et al.
deposit disease: clinicopathologic study of 32 pediatric (2017). C5 nephritic factors drive the biological
and adult patients. Clin. J. Am. Soc. Nephrol. 4: 22–32. phenotype of C3 glomerulopathies. Kidney Int. 92:
7 Le Quintrec, M., Lionet, A., Kandel, C. et al. (2015). 1232–1241.
Eculizumab for treatment of rapidly progressive C3 20 Sethi, S., Fervenza, F.C., Zhang, Y. et al. (2012). C3
glomerulopathy. Am. J. Kidney Dis. 65: 484–489. glomerulonephritis: clinicopathological findings,
8 Ravindran, A., Fervenza, F.C., Smith, R.J.H. et al. (2017). complement abnormalities, glomerular proteomic profile,
C3 glomerulonephritis with a severe crescentic treatment, and follow-up. Kidney Int. 82: 465–473.
phenotype. Pediatr. Nephrol. 32: 1625–1633. 21 Chen, Q., Muller, D., Rudolph, B. et al. (2011). Combined
9 Sethi, S. and Fervenza, F.C. (2012). C3b and factor B autoantibodies and MPGN type II. N.
Membranoproliferative glomerulonephritis – a new look Engl. J. Med. 365: 2340–2342.
at an old entity. N. Engl. J. Med. 366: 1119–1131. 22 Bomback, A.S., Santoriello, D., Avasare, R.S. et al.
10 Lemaire, M., Fremeaux-Bacchi, V., Schaefer, F. et al. (2018). C3 glomerulonephritis and dense deposit disease
(2013). Recessive mutations in DGKE cause atypical share a similar disease course in a large United States
hemolytic-uremic syndrome. Nat. Genet. 45: 531–536. cohort of patients with C3 glomerulopathy. Kidney Int.
11 Hou, J., Markowitz, G.S., Bomback, A.S. et al. (2014). 93: 977–985.
Toward a working definition of C3 glomerulopathy by 23 Medjeral-Thomas, N.R., O’Shaughnessy, M.M., O’Regan,
immunofluorescence. Kidney Int. 85: 450–456. J.A. et al. (2014). C3 glomerulopathy: clinicopathologic
12 O’Shaughnessy, M.M., Hogan, S.L., Poulton, C.J. et al. features and predictors of outcome. Clin. J. Am. Soc.
(2017). Temporal and demographic trends in glomerular Nephrol. 9: 46–53.
disease epidemiology in the Southeastern United States, 24 Iatropoulos, P., Daina, E., Curreri, M. et al. (2018).
1986–2015. Clin. J. Am. Soc. Nephrol. 12: 614–623. Cluster analysis identifies distinct pathogenetic patterns
13 Kerns, E., Rozansky, D., and Troxell, M.L. (2013). in C3 glomerulopathies/immune complex-mediated
Evolution of immunoglobulin deposition in C3-dominant membranoproliferative GN. J. Am. Soc. Nephrol. 29:
membranoproliferative glomerulopathy. Pediatr. Nephrol. 283–294.
28: 2227–2231. 25 Little, M.A., Dupont, P., Campbell, E. et al. (2006).
14 Sethi, S., Nasr, S.H., De Vriese, A.S. et al. (2015). C4d as a Severity of primary MPGN, rather than MPGN type,
diagnostic tool in proliferative GN. J. Am. Soc. Nephrol. determines renal survival and post-transplantation
26: 2852–2859. recurrence risk. Kidney Int. 69: 504–511.
Reference 503
6 Vernon, K.A., Goicoechea de Jorge, E., Hall, A.E. et al.

2 36 Bridoux, F., Desport, E., Fremeaux-Bacchi, V. et al.
(2012). Acute presentation and persistent (2011). Glomerulonephritis with isolated C3 deposits and
glomerulonephritis following streptococcal infection in a monoclonal gammopathy: a fortuitous association? Clin.
patient with heterozygous complement factor H-related J. Am. Soc. Nephrol. 6: 2165–2174.
protein 5 deficiency. Am. J. Kidney Dis. 60: 121–125. 37 Chauvet, S., Fremeaux-Bacchi, V., Petitprez, F. et al.
27 Le Quintrec, M., Lapeyraque, A.L., Lionet, A. et al. (2016). Treatment of B-cell disorder improves renal
(2018). Patterns of clinical response to eculizumab in outcome of patients with monoclonal gammopathy-
patients with C3 glomerulopathy. Am. J. Kidney Dis. 72: associated C3 glomerulopathy. Blood 129:
84–92. 1437–1447.
28 Oosterveld, M.J., Garrelfs, M.R., Hoppe, B. et al. (2015). 38 Chauvet, S., Roumenina, L.T., Aucouturier, P. et al.
Eculizumab in pediatric dense deposit disease. Clin. J. (2018). Both monoclonal and polyclonal immunoglobulin
Am. Soc. Nephrol. 10: 1773–1782. contingents mediate complement activation in
29 Inman, M., Prater, G., Fatima, H. et al. (2015). monoclonal gammopathy associated-C3 glomerulopathy.
Eculizumab-induced reversal of dialysis-dependent Front. Immunol. 9: 2260.
kidney failure from C3 glomerulonephritis. Clin. Kidney J. 39 Rabasco, C., Cavero, T., Roman, E. et al. (2015).
8: 445–448. Effectiveness of mycophenolate mofetil in C3
30 Rousset-Rouviere, C., Cailliez, M., Garaix, F. et al. (2014). glomerulonephritis. Kidney Int. 88: 1153–1160.
Rituximab fails where eculizumab restores renal function 40 Avasare, R.S., Canetta, P.A., Bomback, A.S. et al. (2018).
in C3nef-related DDD. Pediatr. Nephrol. 29: 1107–1111. Mycophenolate mofetil in combination with steroids for
31 Kerns, E., Rozansky, D., and Troxell, M.L. (2014). treatment of C3 glomerulopathy: a case series. Clin. J.
Evolution of immunoglobulin deposition in C3-dominant Am. Soc. Nephrol. 13: 406–413.
membranoproliferative glomerulopathy. Pediatr. Nephrol. 41 Jones, G., Juszczak, M., Kingdon, E. et al. (2004).
28: 2227–2231. Treatment of idiopathic membranoproliferative
32 Fakhouri, F., Zuber, J., Fremeaux-Bacchi, V. et al. (2017). glomerulonephritis with mycophenolate mofetil and
Haemolytic uraemic syndrome. Lancet. 390(10095):681–696. steroids. Nephrol. Dial. Transplant. 19: 3160–3164.
33 Gale, D.P., de Jorge, E.G., Cook, H.T. et al. (2010). 42 Bomback, A.S., Smith, R.J., Barile, G.R. et al. (2012).
Identification of a mutation in complement factor Eculizumab for dense deposit disease and C3
H-related protein 5 in patients of Cypriot origin with glomerulonephritis. Clin. J. Am. Soc. Nephrol. 7:
glomerulonephritis. Lancet 376: 794–801. 748–756.
34 Marinozzi, M.C., Roumenina, L.T., Chauvet, S. et al. 43 Radhakrishnan, S., Lunn, A., Kirschfink, M. et al.
(2016). Anti-factor B and anti-C3b autoantibodies in C3 (2012). Eculizumab and refractory
glomerulopathy and Ig-associated membranoproliferative membranoproliferative glomerulonephritis. N. Engl. J.
GN. J. Am. Soc. Nephrol. 28(5):1603–1613. Med. 366: 1165–1166.
35 Zand, L., Kattah, A., Fervenza, F.C. et al. (2013). C3 44 Ricklin, D., Mastellos, D.C., Reis, E.S. et al. (2018). The
glomerulonephritis associated with monoclonal renaissance of complement therapeutics. Nat. Rev.
gammopathy: a case series. Am. J. Kidney Dis. 62: 506–514. Nephrol. 14: 26–47.
505
Chronic Kidney Disease and Complications

Part 5
507
31
Cardiovascular Disease and CKD

Tomoko Usui, Masaomi Nangaku, and Tetsuhiro Tanaka
Division of Nephrology and Endocrinology, The University of Tokyo Hospital, Bunkyo-ku, Tokyo, Japan
I ntroduction Definition
CVD is a name for the group of disorders of the heart and
Chronic kidney disease (CKD) is an important risk factor blood vessels. This includes hypertension, coronary heart
for cardiovascular morbidity and mortality. Cardiovascular disease, cerebrovascular disease, peripheral vascular dis-
disease (CVD) is the name for a group of disorders of the ease, and heart failure. CVD may progress over years and
heart and blood vessels, including hypertension, coronary CKD patients may develop cardiomyopathy, valve dysfunc-
heart disease, cerebrovascular disease, peripheral vascu- tion, and arrhythmia (e.g. atrial fibrillation) or sudden car-
lar disease, and heart failure. Reduced kidney function diac death.
and raised albuminuria are both independent risk factors
for CVD [1, 2]. By 2013, cardiovascular death attributable
to reduced glomerular filtration rate (GFR) outnumbered E
pidemiology
end-stage renal disease (ESRD) death in both developed
and developing countries [3]. In a cohort of 3033 patients with CKD stages 3–4 (mean
The increased risk of CVD in CKD patients is partly age 66.8 years, 65% men), the prevalence of CVD was
attributed to high prevalence of comorbidities, such as examined [19]. CVD was divided into atherosclerotic (cor-
hypertension, diabetes mellitus, dyslipidemia, and obesity. onary artery disease [CAD], peripheral artery disease, and
There is not high-quality evidence for the selection of treat- stroke) and nonatherosclerotic (heart failure, arrhythmia,
ment of CKD patients to prevent CVD. Many randomized and valvular heart disease). In age groups <65, 65–74,
controlled trials (RCTs) have investigated treatment of dis- 75–84, and 85 years the prevalence of atherosclerotic CVD
ease associated with CVD, but CKD patients are often was 18.7%, 35.5%, 42.9%, and 37.8%, respectively, and that
excluded from these RCTs. It is not clear whether the results of nonatherosclerotic CVD, 14.9%, 28.4%, 38.1%, and 56.4%,
of clinical studies in patients with non-CKD may be extrap- respectively. The prevalence of both atherosclerotic and
olated to CKD patients. The increased risk of CVD in CKD nonatherosclerotic CVD increased with older age.
patients is not fully explained by the traditional risk factors. The prevalence of heart failure in CKD patients is high
The nontraditional risk factors derived from CKD itself may and it was similar among studies. In the European Heart
also be important. Systematic reviews and meta-analysis of Failure Pilot Survey, comorbidities of a total of 3226 outpa-
post hoc analysis of the CKD subgroup and observational tients with chronic heart failure were examined; the high-
studies are performed to produce evidence for CKD patients. est comorbidity was CKD (41%) [20]. In the Swedish Heart
This chapter examines the evidence for the epidemiol- Failure Registry and the Asian Sudden Cardiac Death in
ogy, pathophysiology, diagnosis, prognosis, and manage- Heart Failure registry, prevalence of CKD in patients with
ment of CVD in people with CKD (Table 31.1). We excluded heart failure was 43.9% and 45.4%, respectively [21, 22].
kidney transplant or dialysis patients that will be covered In the Atherosclerosis Risk in Communities (ARIC)
in other places (Parts 6, 7, and 9). study, 10 328 subjects were followed for median of 10.1 years
508 Cardiovascular Disease and CKD
Table 31.1 Treatment for cardiovascular disease (CVD) with chronic kidney disease (CKD).
Strength of
Blood pressure We suggest blood pressure lowering to prevent High (cardiovascular events) Moderate [4–6]
lowering cardiovascular events for patients with CKD. There was
little evidence to support the preferential choice of
particular drug classes for the prevention of cardiovascular
events in CKD. The specific target blood pressure is not
confirmed for patients with CKD.
Statin therapy We suggest statins to decrease cardiovascular mortality and High (cardiovascular mortality Weak [7–9]
events in early stages of CKD. in CKD stages 3–5)
Moderate (major cardiovascular
events in CKD stages 3–5)
Antiplatelet We suggest antiplatelet agents can be used to lower risk of Low Weak [10]
agents myocardial infarction. Patients with CKD (any stage) may
reasonably choose not to receive antiplatelet agents based
on the very low certainty that treatment makes any
difference to mortality and stroke, and there is increased
risk of major or minor bleeding.
Glycemic We suggest sodium glucose co-transporter-2 (SGLT2) Moderate (kidney failure and Moderate [11, 12]
control inhibitors may be used in patients with diabetes and CKD to cardiovascular events)
lower kidney failure and cardiovascular events risk. Patients
with diabetes and CKD may reasonably choose not to receive
SGLT2 inhibitors based on the very low certainty that the
treatment makes any difference on mortality and may
increase genital infections and creatinine.
Anticoagulants We suggest warfarin to prevent mortality, ischemic stroke, Low (warfarin) Moderate [13, 14]
or thromboembolism in patients with atrial fibrillation and Moderate (direct oral
CKD. We suggest direct oral anticoagulants to prevent anticoagulants, stroke, and
stroke and systemic embolic events without increasing risk systemic embolic events)
of major bleeding events among atrial fibrillation patients Low (direct oral anticoagulants,
with CKD compared to warfarin. major bleeding events)
Coronary We suggest percutaneous coronary intervention (PCI) or Low Moderate [15, 16]
revascularization coronary artery bypass grafting (CABG) for coronary artery
disease with CKD. We suggest drug-eluting stents compared
to bare-metal stents for PCI. PCI is associated with lower early
(30 days) mortality, but the need for repeat revascularization
was significantly higher compared to CABG.
Dietary We suggest that reduced dietary salt may be used to reduce Very low Weak [17, 18]
restrictions blood pressure. Long-term effect of reduced dietary salt on
cardiovascular events is not confirmed.
and incident atrial fibrillation cases was identified [23]. P

athophysiology
Incidence of atrial fibrillation by estimated glomerular fil-
tration rate (eGFR) 90, 60–89, 30–59, and 15–29 ml/ CVD is the primary cause of morbidity and premature
min/1.73 m2 was 6.4, 5.6, 9.5, and 36.2 per 1000 person- mortality in CKD. There may be several reasons for the
years, respectively, and by urinary albumin to creatinine increased risk of CVD in CKD patients [24]. CKD patients
ratio <30, 30–299, and 300 mg/g was 5.8, 14.6, and 26.3 have increased prevalence of traditional and nontradi-
per 1000 person-years, respectively. The reduced kidney tional risk factors for CVD. CKD itself is an independent
function and higher albuminuria were associated with risk factor for CVD. CVD and many risk factors for CVD
higher incident atrial fibrillation. are also risk factors for CKD. CKD and CVD interact to
Pathophysiolog 509
contribute to the progression of both diseases, resulting in subjects with albuminuria had significantly higher intima-
the cardiovascular and kidney disease progression cycle. media thickness than those without albuminuria [35].
Systemic vascular remodeling may cause CVD. The risk Aortic stiffness measured by aortic, brachial artery, and
of stroke and myocardial infarction increased with femoral pulse wave velocity was independently associated
elevating blood pressure in a large cohort with or without with cardiovascular mortality in ESRD patients [36].
CKD [25]. The subjects with CKD had increased risk of In 117 autopsy subjects with CAD, the association
CVD with higher blood pressure compared to subjects between coronary artery calcification (CAC) and CKD was
without CKD. The association of blood pressure and CVD evaluated [37]. CAC was present in the intimal plaque of
may be increased in CKD patients. both nonkidney and kidney patients. Age, smoking,
The long-term pressure and volume overload in CKD diabetes, calcium phosphorus product, inflammation, and
causes concentric left ventricular hypertrophy and left ven- kidney function were associated with intimal calcification.
tricular dilation [26]. Cardiomyopathy progression will In a study of 1541 CKD patients, the association between
cause diastolic and systolic dysfunction. In a cohort of CAC and cardiovascular events during an average follow-up
patients with CKD stages 3–4, 30% had left ventricular of 5.9 years was examined [38]. CAC was independently
hypertrophy [27]. The left ventricular mass increase was and significantly related to risk of CVD, myocardial
confirmed in 36% of a CKD cohort with creatinine clear- infarction, and heart failure in CKD patients.
ance between 25 and 75 ml/min/1.73 m2 [28]. After 1 year, Endothelial dysfunction has been proposed to be an early
left ventricular mass increased in 25% of the subjects. phase of the atherosclerotic process [39]. It is characterized
The independent risk factors for left ventricular growth by a shift in the actions of the endothelium toward reduced
were decrease in hemoglobin level and increase in systolic vasodilation, a proinflammatory state, and prothrombotic
blood pressure. Anemia may also cause left ventricular properties. Reduced nitric oxide generation, oxidative
hypertrophy. excess, and reduced production of hyperpolarizing factor
In a cross-sectional study of 118 hemodialysis patients, are the mechanisms in endothelial dysfunction. Upregulated
systolic blood pressure, pulse pressure, extra-coronary adhesion molecules, generation of chemokines, and pro-
atherosclerosis, and vascular calcification correlated with duction of plasminogen activator inhibitor-1 participate in
left ventricular mass [29]. In a cohort of patients with CKD the inflammatory response. Vasoactive peptides, an endog-
stage 5D, the prevalence of left ventricular hypertrophy enous nitric oxide inhibitor, and dyslipidemia may also
was 16.4% [30]. The left ventricular hypertrophy was asso- interact to contribute to these mechanisms.
ciated with higher mortality during the follow-up period of In addition to traditional risk factors, many nontradi-
2 years. During the 10-year follow-up of 123 patients under- tional risk factors that are directly and indirectly associated
going hemodialysis, 16 died of sudden cardiac death [31]. with CKD contribute to CVD pathophysiology. In a meta-
The worsening of left ventricular mass index (LVMI) was analysis of 21 cohort studies, the association between
highly associated with the sudden cardiac death. routinely collected risk factors and CVD was examined [40].
Dyslipidemia of CKD patients is characterized by dys- Left ventricular hypertrophy, serum albumin, phosphate,
regulated high-density lipoprotein cholesterol (HDL-C) urate, and hemoglobin were associated with the increased
and triglyceride-rich lipoprotein metabolism [32]. Plasma risk of CVD.
triglyceride concentration is frequently elevated in patients Low-grade inflammation is recognized as a common
with CKD. These abnormalities in CKD patients may cause pathway of CKD and CVD [41]. Increased inflammation
dysregulation of lipid metabolism, which contributes to markers, such as high-sensitivity C-reactive protein (hs-
atherogenic change and increases the risk of CVD. CRP), white blood cell count, and fibrinogen, were associ-
Abdominal obesity is known to be associated with ated with peripheral artery disease in CKD patients [42].
increased risk of CVD. In a prospective cohort of 537 sub- After a median follow-up of 6.3 years, the same inflamma-
jects with ESRD, waist circumference was associated with tion markers were associated with increased risk of inci-
all-cause and cardiovascular mortality [33]. Adipose tissue dent peripheral artery disease [43]. Higher levels of
secretes a number of adipokines and cytokines and works hs-CRP and interleukin-6 (IL6) were also associated with
as an important organ [34]. The complex role of adipose higher risk of sudden cardiac death in a cohort undergo-
tissue as not only an energy storage organ but its interaction ing hemodialysis or peritoneal dialysis [44].
with the systemic vascular system is not fully known. Cardiac and kidney diseases interact in both acute and
Atherosclerotic and arteriosclerosis change may be the chronic settings [45]. The bi-directional links between car-
cause of CVD in CKD patients. Many studies report the diac and kidney function are perceived as cardio-renal syn-
existence of subclinical atherosclerosis in CKD patients. drome. The interaction includes activation of the
Even among asymptomatic subjects aged 23–43 years, renin-angiotensin aldosterone system (RAAS), activation
of the sympathetic nervous system, uremic toxins, malnu- Biochemical Abnormalities

trition, bone and mineral disorder, and oxidative stress [46–
Clinical practice guidelines and review of diagnosis of CVD
52]. Acute kidney injury (AKI) was associated with
in CKD recommend biomarkers including lipid profile,
increased risk of heart failure, acute myocardial infarction,
hs-CRP, B-type natriuretic peptides (BNP), and troponins.
and stroke [53]. Anemia was associated with increased risk
of stroke and coronary heart disease [54, 55]. The mecha-
Lipid Profile
nism by which cardio-renal syndrome advances CVD has
It is recommended that CKD patients be evaluated with a
not been fully elucidated.
lipid profile (total cholesterol, low-density lipoprotein cho-
Serum potassium level may also be an important risk
lesterol [LDL-C], high-density lipoprotein cholesterol
factor for CVD in CKD patients. A cohort of nondialysis
[HDL-C], triglycerides, nonhigh-density lipoprotein choles-
CKD patients was followed for a median of 2.8 years [56].
terol [non-HDL-C]) [66, 67]. The lipid profile should be
Mortality, major adverse cardiovascular events, and hos-
measured in the fasting state. Fasting mainly affects triglyc-
pitalization were significantly higher among those with
erides level and does not affect HDL-C level. The appropri-
both low and high serum potassium levels. A predialysis
ate frequency of the measurement of lipid profile is not
serum potassium of 4.6–5.3 mEq/l was associated with
known. Some guidelines recommend lipid-lowering ther-
the greatest survival in a cohort of 81 013 maintenance
apy with treatment target of LDL-C < 70 mg/dl or non-
hemodialysis patients followed for 3 years [57]. In a ret-
HDL-C < 100 mg/dl. However, the Kidney Disease:
rospective study of 228 hemodialysis patient deaths, 80
Improving Global Outcomes (KDIGO) Guideline for Lipid
patients met the criteria of sudden death [58]. Patients
Management does not recommend the treat-to-target strat-
with sudden death had high prevalence of congestive
egy due to the lack of its benefit in any clinical trial [66].
heart failure (CHF) and CAD (CHF or CAD were present
in 71%), and 25% had a serum potassium level less than
Inflammation
4.0 mEq/l. In CKD patients with CAD, hyperkalemia
Inflammation markers such as hs-CRP, IL-6, tumor necro-
(serum K > 5.0 mEq/l) occurred more frequently than
sis factor (TNF), and white blood cell counts are found to
hypokalemia (serum K < 3.5 mEq/l) (16.7% vs. 3%), and
be associated with increased risk of CVD events and mor-
hyperkalemia was associated with risk of sudden cardiac
tality. However, there are no thresholds or criteria to diag-
death [59]. Potassium works as a critical factor to main-
nose the increased risk of CVD for those markers.
tain cardiovascular health [60]. CKD patients may have
The frequency of sudden cardiac death and its associa-
serum potassium abnormalities, which includes both
tion with inflammation and other risk factors were exam-
hypokalemia and hyperkalemia. The factors associated
ined among 1041 incident dialysis patients from the
with abnormal potassium levels are declined kidney
CHOICE (Choices for Healthy Outcomes In Caring for
function, endogenous, and exogenous catecholamine
ESRD) cohort over a median 2.5 years of follow-up [68]. Of
activity, activation of RAAS and anti-RAAS inhibitor
all subjects, 22% of all mortality in this cohort was due to
agent and RAAS blocker agents, and potassium wasting
sudden cardiac death. The highest tertiles of hs-CRP and of
diuretics.
IL-6 were each associated with twice the risk of sudden
cardiac death compared to their lowest tertiles when
adjusted for demographics, comorbidities, and laboratory
D
iagnostic Tests factors. Among 543 patients with stage 5 CKD, IL-6, white
blood cell counts, and TNF independently predicted all-
CVD is highly prevalent in CKD patients. Early detection cause mortality during the follow-up of 28 months [69].
of CVD in CKD patients may lead to modification of risk
factors and pharmacological intervention. However, diag- B-Type Natriuretic Peptides
nosis of CVD in CKD patients may be difficult. The accu- Brain natriuretic peptide (BNP) and N-terminal pro-B-type
racy of diagnostic tests is often reduced. The symptoms of natriuretic peptide (NT-proBNP) are used for diagnosis and
CVD may be difficult to distinguish from volume overload prognosis of heart failure. NT-proBNP has a longer half-life
or uremia in CKD patients. There are several systematic than BNP and it may be less sensitive to acute change. The
reviews for the diagnosis of CVD in CKD patients [61–65]. plasma BNP and NT-proBNP levels are higher in CKD
The diagnostic tests to evaluate CVD in CKD patients are patients [70–72]. The BNP level is influenced by kidney
outlined in Table 31.2. function, anemia, obesity, and heart failure.
Diagnostic Test 511
Table 31.2 Summary of diagnostic tests in cardiovascular disease (CVD) with chronic kidney disease (CKD).
Biochemical
Lipid profile ●● Includes total cholesterol, low-density ●● The current guideline recommend to measure
lipoprotein (LDL) cholesterol, high-density lipid profile at first presentation with CKD
lipoprotein (HDL) cholesterol, non-HDL- ●● No further measurement is required unless the
cholesterol, triglycerides results would change management
High-sensitivity C-reactive ●● Measures circulating high-sensitivity ●● No thresholds of change is specified in CKD

protein C-reactive protein (hs-CRP) patients
●● Reflects increased inflammation status, which
is associated with increased risk of CVD events
and mortality
B-type natriuretic peptides ●● Measures circulating B-type natriuretic ●● CKD patients have elevated BNP and NT-pro-
peptides (BNP) or N-terminal-pro-BNP BNP levels
(NT-pro-BNP) concentration ●● No thresholds of change is specified in CKD
●● Commonly used with patients with symptoms patients
of heart failure
●● Association with abnormal echocardiographic
findings
Troponins ●● Measures circulating cardiac troponin T ●● Patients with kidney dysfunction have elevated
(cTnT) or cardiac troponin I (cTnI) troponins levels
●● Association with cardiac injury such as ●● No thresholds of change is specified in CKD
myocardial infarction patients
Electrocardiogram
12-lead electrocardiogram ●● Measures baseline rhythm, presence of Q ●● Exercise stress electrocardiogram may be
waves, chamber enlargement, left ventricular performed
hypertrophy
24 h Holter monitor ●● Records all beats over 24-h period

●● Identifies ST segment shifts or asymptomatic
arrhythmias
Imaging
Resting echocardiography ●● Evaluates diastolic and systolic function and ●● Noninvasive

valve disease
Exercise stress ●● Evaluates exercise-related functional status ●● No radiation

echocardiography ●● Identifies fixed and reversible ischemia ●● Appropriate for low risk CVD and severe CKD
stages
Exercise stress nuclear ●● Evaluates exercise-related functional status ●● Radiation or drug usage
scintigraphy or ●● Identifies fixed and reversible ischemia ●● Requires specific scanner, not available in many
Pharmacologic stress clinical settings
imaging
Left ventricular ●● Measures left ventricular mass index (LVMI) ●● Noninvasive

hypertrophy or left or left ventricular function by ●● Patients with advanced CKD stages have
ventricular dysfunction echocardiography increased LVMI
●● Association with CVD mortality ●● Uncertain diagnostic utility in CKD
Carotid intima-media ●● Measures the thickness of the intima-media of ●● Noninvasive

thickness the artery on an ultrasound image ●● Uncertain diagnostic utility in CKD
●● Association with cardiovascular events
(Continued)
Cardiac computed ●● Generates calcium score, which can evaluate ●● Appropriate for the evaluation of chest pain with
tomographic angiography degree of lumen stenosis if calcification is not uninterpretable stress test
severe ●● Not appropriate for established coronary artery
disease or ESRD
Coronary artery ●● Measures coronary artery calcification (CAC) ●● Radiation
calcification by computed tomography ●● CKD patients have higher CAC scores
●● Association with the risk of CVD ●● Uncertain diagnostic utility in CKD
●● Not recommended in routine clinical care
Coronary angiography ●● Detects blockages in the coronary arteries ●● Uses contrast dye containing iodinine, which
may cause contrast-associated AKI
●● Radiation
Single-photon emission ●● Measures myocardial function, left ventricular ●● Noninvasive
computed tomography volume, and ejection fraction ●● Requires specific scanner, not available in many
(SPECT) or positron clinical settings
emission tomography (PET)
Cardiac magnetic ●● Measures left ventricular mass, volume, and ●● Noninvasive
resonance or cardiac MRI ejection fraction
Troponin v entricular hypertrophy. The findings of ECG in CKD

Troponins (troponin I and troponin T) are used as a marker patients in acute myocardial infarction may be different
of myocardium damage [73]. Myocardial infarction is from those without CKD [77]. In a study of 1876 elderly
diagnosed a rise and/or fall from the norm of cardiac myocardial infarction patients, the patients with CKD had
biomarkers, preferably cardiac troponin (cTn). The diagnosis lower peak creatinine kinase-MB fraction (CK-MB) level
of myocardial infarction in CKD patients may be difficult and ST-elevated myocardial infarction was less common
because patients with kidney dysfunction can have an compared to those without CKD [78].
increased troponin level, so the diagnosis of myocardial
infarction may be difficult [74]. The cTn cutoff levels in
Imaging
patients with kidney dysfunction were significantly higher
compared to those in patients with normal kidney function. Echocardiography
The optimal cutoff levels to separate AMI from other condi- Echocardiography is useful for the diagnosis of clinical and
tions underlying acute chest pain in the emergency depart- subclinical cardiac dysfunction, for the prediction of
ment in patients with kidney dysfunction were close to the cardiovascular risk, and in the orientation and follow-up of
elevated baseline levels above the 99th percentile for the treatment strategies in CKD patients [79, 80]. Echo-
3 s-cTn assays (1.0 times the 99th percentile for Abbott– cardiogram may be repeated for a patient periodically and
Architect s-cTnI, 1.2 times the 99th percentile for Siemens with changes in symptoms, new clinical event, or a treat-
Ultra s-cTnI, and 0.9 times the 99th percentile for Beckman– ment likely to affect cardiac function. Among 567 subjects
Coulter Accu s-cTnI) and substantially higher for most hs- with CKD and without history of CVD, LVMI evaluated by
cTn assays (2.1 times the 99th percentile for Roche hs-cTnT, echocardiogram was an independent risk factor for inci-
1.1 times the 99th percentile for Abbott–Architect hs-cTnI, dent heart failure during median follow-up period of
3.6 times the 99th percentile for Siemens hs-cTnI, and 2.8 6.6 years [81]. The delay in or absence of renal replacement
times the 99th percentile for Beckman–Coulter hs-cTnI) [75]. therapy results in dyspnea due to fluid overload even if the
The higher cTnT and cTnI levels in patients with CKD stage patient’s heart is structurally normal. The new functional
5D show a two to fivefold increase in mortality [76]. classification system of heart failure in patients with CKD
stage 5 are proposed to exclude patients with volume over-
load and a normal heart [82]. The classification is based on
Electrocardiogram
patient-reported symptoms and standardized echocardio-
The electrocardiogram (ECG) evaluates baseline rhythm, graphic evidence of structural and functional heart abnor-
presence of Q-waves, chamber enlargements, and left malities. When this system was applied to 654 patients who
Managemen 513
began chronic hemodialysis, 87% of patients had one or activities of daily living, kidney disease progression, or
more of echocardiographic criteria of heart failure [83]. mortality after cardiovascular events in CKD patients.
During a median of 2.4 years, both left and right ventricu- A total 3352 CKD patients with CAD diagnosed with
lar systolic dysfunction were independent risk factors for angiography were followed for 2 years [91]. At baseline,
mortality. 31.3% of patients with eGFR 90 ml/min/1.73 m2 had one-
vessel, 34.3% had two-vessel, and 33.9% had three-vessel
Coronary Angiography CAD. In patients with eGFR <30 ml/min/1.73 m2 or on
Coronary angiography uses contrast dye and X-ray to dialysis, the proportion of patients with one-vessel CAD
diagnose CAD. CKD patients need special consideration was 16.8% and 54.5% had three-vessel CAD. The significant
due to potential contrast-associated AKI [84]. The efficacy risk factors for 2-year all-cause mortality were lower eGFR,
of prophylactic intravenous hydration for prevention of current smoking, left ventricular ejection fraction, diabetes
contrast-associated AKI differs among studies [85, 86]. The mellitus, age, and peripheral artery disease. Peripheral
type of contrast media may have different effect on kidney artery disease diagnosed with ankle-brachial index
function. The use of iso-molar and low-osmolar contrast <0.9 was also an independent predictor of major adverse
media in patients with CKD stage 3–5 showed nonsignificant cardiovascular events in 3131 CKD patients hospitalized
difference in incidence of contrast-associated AKI in a for CVD [92].
meta-analysis of randomized trials [87].
Computed Tomography
Management
CAC can be measured by computed tomography. CAC was
independently associated with the risk of CVD and mortal-
As CKD patients are at high risk of cardiovascular morbid-
ity in CKD patients [40]. However, the clinical utility of
ity and mortality, the management of both traditional and
CAC is unknown due to lack of enough data in CKD
nontraditional risk factors is of importance [93, 94]. In
patients. In a cohort study of 3218 CKD subjects followed
randomized trials of treatment of traditional risk factors,
for 12.5 years, elevated BNP, NT-proBNP, and cTnT and left
CKD patients are often excluded. The robust evidence for
ventricular hypertrophy were associated with cardiovascu-
the management of CKD patients is limited (Tables 31.3–
lar events or death [88]. CAC did not improve risk predic-
31.8). The control of blood pressure may have some bene-
tion when added to traditional risk factors in this study.
fits in CKD patients, but the target blood pressure for the
Myocardial Perfusion Imaging prevention of CVD is not confirmed. Lipid-lowering ther-
Myocardial perfusion imaging is nuclear imaging of the apy reduced cardiovascular mortality in CKD patients
function of myocardium. Single-photon emission without dialysis or kidney transplant. The risk reduction
computed tomography (SPECT) or positron emission of CVD by antiplatelet therapy was not confirmed. Some
tomography (PET) can be performed [80]. In a prospective classes of blood glucose agents may reduce cardiovascular
cohort study with 299 CKD patients who underwent events, but there is a lack adequate high-quality studies.
SPECT, the scores calculated as the difference between Lifestyle modification may be recommended based on
stress and rest images were associated with high major observational studies, but few studies have confirmed the
cardiovascular or kidney events [89]. effect on CVD in CKD patients. Anticoagulants therapy
for patients with atrial fibrillation reduced the risk of
Cardiac Magnetic Resonance thromboembolism and stroke. Coronary revascularization
Cardiac magnetic resonance or cardiac magnetic resonance for patients with CAD was evaluated and the short-term
imaging (MRI) is noninvasive imaging that does not use and long-term outcome may differ according to the
contrast administration or radiation. In 72 patients with treatment.
CKD stage 5D, ECG and SPECT had low sensitivity in Patients with CKD and acute CAD or stroke are less
detecting myocardial infarction [90]. Cardiac MRI with likely to receive evidence-based medicine [95–98]. The
myocardial delayed enhancement identified unrecognized reason for this attitude toward CKD patients may be due to
myocardial infarction. atypical symptoms and difficulty in the diagnosis of CVD
in CKD patients, or concerns of AKI after diagnostic
method or treatment. Meta-analysis of 10 observational
Prognosis and randomized clinical studies examined the association
between early invasive approach and all-cause mortality in
CKD patients are known to have high risk of cardiovascu- CKD patients [99]. Early invasive approach was associated
lar morbidity and mortality. Several studies examined the with lower risk of mortality.
c31.indd 513 09-12-2022 16:04:08

Table 31.3 Summary of findings: blood pressure control.
Absolute number of
No. of patients affected per Certainty of
participants 1 000 patients treated Relative risk the evidence
Angiotensin converting enzyme inhibitors versus placebo

Major cardiovascular events 10 044 (13) – HR 0.81 (0.73–0.89) High Angiotensin converting enzyme inhibitors may reduce major
Calcium channel blockers versus placebo
Major cardiovascular events 1 855 (6) – HR 0.74 (0.53–1.03) Moderate Calcium antagonists therapy may reduce major cardiovascular events
More intensive blood pressure lowering versus less intensive blood pressure lowering
Mortality 15 924 (3) – OR 0.86 (0.75–0.97) High Intensive blood pressure lowering may reduce mortality
Cardiovascular mortality 2 646 (4) – HR 0.57 (0.31–1.02) Moderate It is uncertain whether intensive blood pressure lowering may
reduce cardiovascular death
Composite cardiovascular 2 646 (4) – HR 0.81 (0.63–1.05) Moderate It is uncertain whether intensive blood pressure lowering may
outcome reduce cardiovascular events
Acute kidney failure 2 646 (4) – HR 1.46 (1.10–1.95) Moderate Intensive blood pressure lowering may increase acute kidney failure
Hyperkalemia 2 646 (4) – HR 1.36 (1.01–1.82) Moderate Intensive blood pressure lowering may increase hyperkalemia
CI, confidence interval; HR, hazard ratio; OR, odds ratio.

More intensive blood pressure lowering: active blood pressure treatment or intensive blood pressure control.
Less intensive blood pressure lowering: placebo or no treatment or less intensive blood pressure control.
c31.indd 514 09-12-2022 16:04:08

Table 31.4 Summary of findings: statin therapy.
Absolute number of
participants 1 000 patients treated Relative risk the evidence
Chronic kidney disease stages 3–5

All-cause mortality 28 276 (13) 5 fewer (2–8 fewer) 0.79 (0.69–0.91) High Statin therapy leads to slightly lower mortality
Cardiovascular mortality 19 059 (10) 3 fewer (2–5 fewer) 0.77 (0.69–0.87) Moderate Statin therapy leads to slightly lower cardiovascular mortality
Major cardiovascular 36 033 (16) 6 fewer (4–7 fewer) 0.72 (0.66–0.79) High Statin therapy probably leads to slightly lower major cardiovascular events
events
Myocardial infarction 9 018 (11) – 0.55 (0.42–0.72) Moderate Statin therapy may slightly reduce myocardial infarction
Stroke 8 658 (8) – 0.62 (0.35–1.12) Low It is uncertain whether statin therapy prevents stroke
Elevated creatinine kinase 4 514 (10) – 0.84 (0.20–3.48) Low It is uncertain whether statin therapy leads to elevated creatinine kinase
Liver function 7 991 (10) – 0.76 (0.39–1.50) Low It is uncertain whether statin therapy leads to liver function
abnormalities abnormalities
Table 31.5 Summary of findings: antiplatelet therapy.
No. of Absolute number of patients Certainty of

participants affected per 1 000 patients Relative risk the evidence
Outcomes (no. of studies) treated for one year (95% CI) (95% CI) (GRADE) Conclusion
Antiplatelet agents versus placebo or no treatment

All-cause mortality 16 152 (30) – 0.95 (0.83–1.07) Low It is uncertain whether antiplatelet agents decrease mortality
Cardiovascular mortality 9 337 (19) – 0.87 (0.65–1.15) Low It is uncertain whether antiplatelet agents make any difference
to cardiovascular mortality
Myocardial infarction 14 451 (2) – 0.87 (0.76–0.99) Low Antiplatelet agents may prevent myocardial infarction
Stroke 9 544 (14) – 1.00 (0.58–1.72) Low It is uncertain whether antiplatelet agents decrease stroke
Major bleeding 1 592 (27) – 1.35 (1.10–1.65) Low Antiplatelet agents may lead to major bleeding
Minor bleeding 13 106 (3) – 1.54 (1.26–1.90) Low Antiplatelet agents may lead to minor bleeding
c31.indd 515 09-12-2022 16:04:08

Table 31.6 Summary of findings: glucose-lowering therapy.

Outcomes (no. of studies) treated for 1 year (95% CI) (95% CI) (GRADE) Conclusion
Sodium glucose co-transporter-2 inhibitors versus placebo

Cardiovascular mortality 4 401 (4) – HR 0.70 (0.59–0.82) Moderate Sodium glucose co-transporter-2 inhibitors may prevent death
Heart failure 2 519 (6) – 0.59 (0.41–0.87) Moderate Sodium glucose co-transporter-2 inhibitors may reduce heart
failure
Hospitalization for heart 4 401 (4) HR 0.61 (0.47–0.80) Moderate
failure
Hyperkalemia 2 788 (7) – 0.58 (0.42–0.81) Moderate Sodium glucose co-transporter-2 inhibitors may reduce
hyperkalemia
4397 (4) – 0.80 (0.65–1.00) Moderate
Genital infections 3 086 (10) – 2.50 (1.52–4.11) Moderate Sodium glucose co-transporter-2 inhibitors may increase genital
infections
Elevated serum 848 (7) – MD 3.82 ɥmo/l Moderate Sodium glucose co-transporter-2 inhibitors may increase serum
creatinine (1.45–6.19) creatinine
Acute kidney injury 4 397 (4) HR 0.85 Moderate It is uncertain whether sodium glucose co-transporter-2 may
(0.64 to 1.13) increase acute kidney injury
Diabetic ketoacidosis 4 397 (4) HR 10.80 Low It is uncertain whether sodium glucose co-transporter-2 may
(1.39–83.65) increase acute kidney injury since the number of diabetic
ketoacidosis were low (11 versus 1)
Dipeptidyl peptidase-4 inhibitors versus placebo
Cardiovascular mortality 5 897 (5) 0.93 (0.77–1.11) Moderate Dipeptidyl peptidase-4 inhibitors may have little or no effect on
cardiovascular death
Heart failure – – – Low It is uncertain whether dipeptidyl peptidase-4 inhibitors prevent
heart failure
Metformin verses treatment without metformin
All-cause mortality 33 442 (8) 48 fewer (81–6 fewer) HR 0.77 (0.61–0.97) Low Metformin may prevent death
CI, confidence interval; HR, hazard ratio; MD, mean difference.
c31.indd 516 09-12-2022 16:04:08

Table 31.7 Summary of findings: anticoagulant therapy.

Warfarin versus no warfarin

Mortality 34 852 (7) – HR 0.65 (0.59–0.72) Low Warfarin may reduce risk of mortality
Ischemic stroke or 34 226 (8) – HR 0.70 (0.54–0.89) Low Warfarin may reduce ischemic stroke or thromboembolism
thromboembolism
Major bleeding 33 920 (7) – HR 1.15 (0.88–1.49) Low It is uncertain whether warfarin increases risk of major bleeding
Direct oral anticoagulant versus warfarin
Stroke and systemic 12 545 (8) – 0.81 (0.65–1.00) Moderate Direct oral anticoagulant probably reduces the incidence of
embolism events stroke and systemic embolism events compared to warfarin
Major bleeding events 12 521 (8) – 0.79 (0.59–1.04) Moderate Direct oral anticoagulant slightly reduces the incidence of major
bleeding compared to warfarin
CI, confidence interval; HR, hazard ratio.
c31.indd 517 09-12-2022 16:04:09

Table 31.8 Summary of findings: coronary revascularization.

Drug-eluting stents versus bare-metal stents

All-cause mortality 91 046 (30) – 0.77 (1.71–1.84) Low Drug-eluting stents probably lead to lower mortality than
bare-metal stents
Cardiovascular mortality 5 045 (17) – 0.51 (0.38 to 0.70) Low Drug-eluting stents may prevent cardiovascular mortality more
than bare-metal stents
Myocardial infarction 64 389 (24) – 0.90 (0.86–0.95) Low Drug-eluting stents may prevent myocardial infarction more
than bare-metal stents
Target-vessel 10 029 (19) – 0.61 (0.47–0.80) Low Drug-eluting stents may prevent target-vessel revascularization
revascularization more than bare-metal stents
Stent thrombosis 6 714 (2) – 0.75 (0.55–1.01) Low Drug-eluting stents probably lead to lower stent thrombosis than
bare-metal stents
Percutaneous coronary intervention with drug-eluting stents versus coronary artery bypass grafting
Short-term all-cause 25 444 (9) – OR 0.33 (0.29–0.38) Low Percutaneous coronary intervention with drug-eluting stents
mortality may lead to lower short-term mortality than coronary artery
bypass grafting
Long-term all-cause 28 521 (14) – OR 1.22 (1.15–1.29) Low Percutaneous coronary intervention with drug-eluting stents
mortality may lead to higher long-term mortality than coronary artery
bypass grafting
Heart-related mortality 19 104 (9) – OR 1.29 (1.20–1.38) Low Percutaneous coronary intervention with drug-eluting stents
may lead to higher heart-related mortality than coronary artery
bypass grafting
Myocardial infarction 10 579 (12) – OR 1.89 (1.11–3.21) Low Percutaneous coronary intervention with drug-eluting stents
may lead to higher myocardial infarction than coronary artery
bypass grafting
Repeat revascularization 28 530 (14) – OR 3.47 (2.57–4.69) Low Percutaneous coronary intervention with drug-eluting stents
may lead to higher repeat revascularization than coronary artery
bypass grafting
Cerebrovascular 9 319 (9) – OR 0.64 (0.51–0.80) Low Percutaneous coronary intervention with drug-eluting stents
accident may lead to lower cerebrovascular accident than coronary artery
bypass grafting.
Major adverse heart- 3 420 (8) – OR 2.00 (1.39–2.89) Low Percutaneous coronary intervention with drug-eluting stents
related and may lead to higher major adverse heart-related and
cerebrovascular events cerebrovascular events than coronary artery bypass grafting
CI, confidence interval; OR, odds ratio.
c31.indd 518 09-12-2022 16:04:09

Managemen 519
Blood Pressure Control uncertain compared to placebo or no treatment [10]. The

platelet dysfunction in CKD patients may be responsible
CKD patients have high prevalence of hypertension. Blood
for reduced effect of antiplatelet therapy [101].
pressure lowering may be an important strategy to prevent
CVD in CKD patients. Blood pressure lowering reduced
the risk of major cardiovascular events in a meta-analysis Glucose Control
of 26 randomized trials, which included 30 295 CKD sub-
It is not clear whether intensive glucose control is effective
jects [4]. The effect in reducing CVD was similar among
in reducing CVD in CKD patients with diabetes [102]. In a
subjects with or without CKD. There was no significant dif-
meta-analysis of 44 studies, which included 13 036 patients,
ference in the effect on cardiovascular events between
the effect of glucose lowering therapy in CKD patients with
angiotensin converting enzyme inhibitors, calcium chan-
diabetes mellitus was examined [11]. Sodium-glucose co-
nel blockers, or diuretics or β blocker based treatment.
transporter-2 (SGLT2) inhibitors reduced heart failure but
Compared to less intensive blood pressure lowering, inten-
increased genital infection and serum creatinine. SGLT2
sive blood pressure lowering was associated with lower
probably had little or no effect on cardiovascular mortality.
mortality in patients with CKD stages 3–5 in the meta-
Dipeptidyl peptidase-4 (DPP4) inhibitors had little or no
analysis of 18 randomized clinical trials [5]. However,
effect on cardiovascular mortality. There was not enough
there was not enough data to define the target blood pres-
data to examine the effect for other classes of glucose low-
sure for the prevention of CVD in CKD patients. In the sub-
ering therapy. In the double-blind, randomized
group analysis of 2646 CKD patients in the Systolic Blood
Canagliflozin and Renal Events in Diabetes and Established
Pressure Intervention Trial (SPRINT), the target blood
Nephropathy Clinical Evaluation (CREDENCE) trial, oral
pressure <120 mmHg (intensive group) was compared to
SGLT2 inhibitor canagliflozin or placebo were assigned to
<140 mmHg (standard group) for the median follow-up of
4401 patients with type 2 diabetes and albuminuric kidney
3.3 years [6]. The intensive group had a lower rate of all-
disease [12]. After a median follow-up of 2.6 years, the rela-
cause mortality and higher AKI as an adverse clinical
tive risk of composite kidney outcome was lower by 34%.
event. In the analysis of 9361 patients in SPRINT, the risk
The risk of cardiovascular death, myocardial infarction, or
factors for AKI were older age, non-Caucasian, lower base-
stroke and hospitalization for heart failure was also lower.
line eGFR, and presence of CVD at baseline [100]. Intensive
Trials of these newer glucose lowering medications
blood pressure lowering may be suggested to prevent mor-
(SGLT2, DPP4, and glucagon-like peptide-1 receptor ago-
tality, but careful monitoring of kidney function is neces-
nist) on cardiovascular outcomes in patients with diabetes
sary in CKD patients, especially among patients with older
and CKD are underway [102].
age or previous CVD.
Metformin reduced all-cause mortality in CKD patients
with diabetes in the meta-analysis of 33 442 patients in seven
Statins studies [103]. Metformin is not recommended in patients
with low kidney function (eGFR <30 ml/min/1.73 m2)
Meta-analysis of 50 studies, which includes 45 285 because of the risk of metformin-associated lactic acidosis.
nondialysis CKD patients, comparing the effects of statins A systematic review was performed to examine the risk of
on mortality, cardiovascular events, and adverse events, was metformin-associated lactic acidosis, which suggested that
performed [7]. Compared to placebo, statin therapy reduced an increased risk of lactic acidosis in metformin-treated
all-cause mortality, cardiovascular mortality, major cardio- patients with CKD may be limited [104]. Metformin may be
vascular events, and myocardial infarction, but the effect on used in patients with mild to moderate CKD (eGFR
stroke or risk of related adverse events was uncertain in 30–60 ml/min/1.73 m2), but dose adjustment and careful
nondialysis CKD. The statin therapy had little or no effect follow-up of kidney function are recommended (when the
on mortality or cardiovascular events in CKD patients on kidney function declined to eGFR <45 ml/min/1.73 m2).
dialysis [8]. The effect of statin therapy on mortality or CVD
was uncertain in kidney transplant recipients [9].
Anticoagulant
In patients with CKD and atrial fibrillation, anticoagulant
Antiplatelet Therapy
therapy may be suggested to prevent thromboembolism
The effect of antiplatelet therapy in risk reduction of CVD and stroke. The risk of major and minor bleeding must be
in CKD patients is not confirmed. In the meta-analysis of considered. In a meta-analysis of 11 cohort studies, warfa-
30 trials, antiplatelet therapy reduced myocardial infarction rin therapy reduced the risk of ischemic or thromboembo-
but the effect on mortality and increased bleeding was lism and mortality [13]. There was no significant effect on
c31.indd 519 09-12-2022 16:04:09

risk of major bleeding in CKD without ESRD, but there r ecommendations include smoking cessation, exercise,
was increased risk of major bleeding in CKD with ESRD. and weight loss, but the effect on CVD prevention in CKD
Direct oral anticoagulants (DOAC) have been developed patients is not confirmed.
and are used as an alternative therapy to warfarin. In a Physical exercise in CKD patients was associated with
meta-analysis of five studies which included 12 545 CKD improved physical fitness, walking capacity, and cardiovas-
patients with atrial fibrillation, the effect of DOAC was cular dimensions, which include resting systolic and dias-
examined [14]. DOAC therapy reduced incident stroke and tolic blood pressures and heart rates [106]. The effect of
systemic embolism and may slightly reduce major bleeding weight loss intervention was examined in a small number
events compared to warfarin therapy. The use of DOAC in of CKD patients in short duration studies [107]. In a meta-
CKD stage 5D with atrial fibrillation was not determined analysis of 13 studies, nonsurgical weight loss reduced
due to the lack of high-quality data [105]. proteinuria and blood pressure. The long-term effect on
CVD in CKD patients is unknown.
Reduced dietary salt was associated with small reduction
Coronary Revascularization
in systolic and diastolic blood pressure in normotensive
CKD patients with CAD may need myocardial revasculari- subjects, and greater reduction of systolic blood pressure
zation. In percutaneous coronary intervention (PCI), a and no change in diastolic blood pressure in hypertensive
bare metal stent (BMS) or drug-eluting stent (DES) is used. subjects [108]. In CKD patients, salt restriction reduced
Most clinical trials of PCI exclude CKD patients. Using 31 systolic and diastolic blood pressure [17]. The long-term
observational or randomized clinical studies with 91 817 effect of salt restriction on CVD in CKD patients is
patients, the outcome with BMS or DES was evaluated in unknown. The effect of low protein diet in CKD patients
CKD patients [15]. DES implantation was associated with without diabetes was examined in the meta-analysis of 10
lower all-cause mortality, cardiovascular mortality, myo- studies [18]. Low protein diet probably has little or no
cardial infarction, and target-vessel revascularization. effect on mortality compared to normal protein diet.
Coronary artery bypass grafting (CABG) is another strategy
for coronary revascularization. The benefits of PCI or CABG
treatment differs, and the outcome may differ in CKD patients C
onclusion
compared to non-CKD patients with CAD. In a meta-analysis
of 11 studies with 29 246 patients with CKD and multivessel CVD is the major cause of morbidity and mortality in CKD
disease, the outcome of PCI with DES and CABG was exam- patients. Increased risk of CVD in CKD patients is due to
ined [16]. Compared to CABG, PCI with DES had higher traditional and nontraditional risk factors, some of which
long-term all-cause mortality, cardiac mortality, myocardial are specific to CKD.
infarction, repeat revascularization, and major adverse cardi- Many RCTs excluded CKD patients and therefore the
ovascular events, but PCI with DES had lower short-term all- evidence to support the clinical practice of CKD is based on
cause mortality and cerebrovascular events. insufficient and low-quality evidence. Due to the difficulty
in diagnosis of CVD in CKD patients, CKD patients are less
likely to receive evidence-based medicine, and the progno-
Lifestyle Modification
sis tends to be poor compared to non-CKD patients. The
Lifestyle modifications are suggested in CKD patients to adjusted criteria for the diagnosis and treatment of CVD in
prevent CVD, but the evidence is mostly from derived CKD patients based on the best available recommenda-
observational study or small clinical trials [94]. The tions may improve outcomes.
R
eferences
1 Chronic Kidney Disease Prognosis Consortium, Matsushita, are associated with all-cause and cardiovascular mortality.
K., van der Velde, M. et al. (2010). Association of estimated A collaborative meta-analysis of high-risk population
glomerular filtration rate and albuminuria with all-cause cohorts. Kidney Int. 79 (12): 1341–1352.
and cardiovascular mortality in general population cohorts: Thomas, B., Matsushita, K., Abate, K.H. et al.; Global Burden
3
a collaborative meta-analysis. Lancet 375 (9731): 2073–2081. of Disease 2013 GFR Collaborators; CKD Prognosis
2 van der Velde, M., Matsushita, K., Coresh, J. et al., Chronic Consortium; Global Burden of Disease Genitourinary Expert
Kidney Disease Prognosis Consortium (2011). Lower Group (2017). Global cardiovascular and renal outcomes of
estimated glomerular filtration rate and higher albuminuria reduced GFR. J. Am. Soc. Nephrol. 28 (7): 2167–2179.
c31.indd 520 09-12-2022 16:04:09

Reference 521
4 Blood Pressure Lowering Treatment Trialists’ 15 Volodarskiy, A., Kumar, S., Pracon, R. et al. (2018).
Collaboration, Ninomiya, T., Perkovic, V. et al. (2013). Drug-eluting vs bare-metal stents in patients with chronic
Blood pressure lowering and major cardiovascular events kidney disease and coronary artery disease: insights from
in people with and without chronic kidney disease: a systematic review and meta-analysis. J. Invasive Cardiol.
meta-analysis of randomised controlled trials. BMJ 347: 30 (1): 10–17.
f5680. 16 Wang, Y., Zhu, S., Gao, P., and Zhang, Q. (2017).
5 Malhotra, R., Nguyen, H.A., Benavente, O. et al. (2017). Comparison of coronary artery bypass grafting and
Association between more intensive vs less intensive drug-eluting stents in patients with chronic kidney
blood pressure lowering and risk of mortality in chronic disease and multivessel disease: a meta-analysis. Eur. J.
kidney disease stages 3 to 5: a systematic review and Intern. Med. 43: 28–35.
meta-analysis. JAMA Intern. Med. 177 (10): 1498–1505. 17 McMahon, E.J., Campbell, K.L., Bauer, J.D., and Mudge,
6 Cheung, A.K., Rahman, M., Reboussin, D.M. et al.; D.W. (2015). Altered dietary salt intake for people with
SPRINT Research Group (2017). Effects of intensive BP chronic kidney disease. Cochrane Database Syst. Rev. (2)
control in CKD. J. Am. Soc. Nephrol. 28 (9): 2812–2823. (Art. No.: CD010070). DOI: https://doi.
7 Palmer, S.C., Navaneethan, S.D., Craig, J.C. et al. (2014). org/10.1002/14651858.CD010070.pub2.
HMG CoA reductase inhibitors (statins) for people with 18 Hahn, D., Hodson, E.M., and Fouque, D. (2018). Low
chronic kidney disease not requiring dialysis. Cochrane protein diets for chronic kidney disease in non-diabetic
Database Syst. Rev. (5) (Art. No.: CD007784). DOI: https:// adults. Cochrane Database Syst. Rev. (10) (Art. No.:
doi.org/10.1002/14651858.CD007784.pub2. CD001892). DOI: https://doi.org/10.1002/14651858.
8 Palmer, S.C., Navaneethan, S.D., Craig, J.C. et al. (2013). CD001892.pub4.
HMG CoA reductase inhibitors (statins) for dialysis 19 Villain, C., Metzger, M., Combe, C. et al. Chronic Kidney
patients. Cochrane Database Syst. Rev. (9) (Art. No.: Disease-Renal Epidemiology and Information Network
CD004289). DOI: https://doi.org/10.1002/14651858. (CKD-REIN) Study Investigators (2020). Prevalence of
CD004289.pub5 atheromatous and non-atheromatous cardiovascular
9 Palmer, S.C., Navaneethan, S.D., Craig, J.C. et al. (2014). disease by age in chronic kidney disease. Nephrol. Dial.
HMG CoA reductase inhibitors (statins) for kidney Transplant. 35 (5): 827–836.
transplant recipients. Cochrane Database Syst. Rev. (1) 20 van Deursen, V.M., Urso, R., Laroche, C. et al. (2014).
(Art. No.: CD005019). DOI: https://doi. Co-morbidities in patients with heart failure: an analysis
org/10.1002/14651858.CD005019.pub4. of the European Heart Failure Pilot Survey. Eur. J. Heart
10 Razavian, M., Di Micco, L., Palmer, S.C. et al. (2013). Fail. 16 (1): 103–111.
Antiplatelet agents for chronic kidney disease. Cochrane 21 Lawson, C.A., Solis-Trapala, I., Dahlstrom, U. et al.
Database Syst. Rev. (2) (Art. No.: CD008834). DOI: https:// (2018). Comorbidity health pathways in heart failure
doi.org/10.1002/14651858.CD008834.pub2. patients: a sequences-of-regressions analysis using
11 Lo, C., Toyama, T., Wang, Y. et al. (2018). Insulin and cross-sectional data from 10,575 patients in the
glucose-lowering agents for treating people with diabetes Swedish Heart Failure Registry. PLoS Med. 15 (3):
and chronic kidney disease. Cochrane Database Syst. Rev. e1002540.
9: CD011798. https://doi.org/10.1002/14651858. 22 Tromp, J., Tay, W.T., Ouwerkerk, W. et al.; ASIAN-HF
CD011798.pub2. (2018). Multimorbidity in patients with heart failure from
12 Perkovic V, Jardine MJ, Neal B, et al., CREDENCE Trial 11 Asian regions: a prospective cohort study using the
Investigators (2019). Canagliflozin and renal outcomes in ASIAN-HF registry. PLoS Med. 15 (3): e1002541.
type 2 diabetes and nephropathy. N. Engl. J. Med.380(24): 23 Alonso, A., Lopez, F.L., Matsushita, K. et al. (2011).
2295–2306. Chronic kidney disease is associated with the incidence of
13 Dahal, K., Kunwar, S., Rijal, J. et al. (2016). Stroke, major atrial fibrillation: the Atherosclerosis Risk in Communities
bleeding, and mortality outcomes in warfarin users with (ARIC) study. Circulation 123 (25): 2946–2953.
atrial fibrillation and chronic kidney disease: a meta- 24 Menon, V., Gul, A., and Sarnak, M.J. (2005).
analysis of observational studies. Chest 149 (4): 951–959. Cardiovascular risk factors in chronic kidney disease.
14 Kimachi, M., Furukawa, T.A., Kimachi, K. et al. (2017). Kidney Int. 68 (4): 1413–1418.
Direct oral anticoagulants versus warfarin for preventing 25 Kokubo, Y., Nakamura, S., Okamura, T. et al. (2009).
stroke and systemic embolic events among atrial Relationship between blood pressure category and
fibrillation patients with chronic kidney disease. incidence of stroke and myocardial infarction in an urban
Cochrane Database Syst. Rev. (11) (Art. No.: CD011373). Japanese population with and without chronic kidney
DOI: https://doi.org/10.1002/14651858.CD011373.pub2. disease: the Suita Study. Stroke 40 (8): 2674–2679.
c31.indd 521 09-12-2022 16:04:09

26 London, G.M. and Parfrey, P.S. (1997). Cardiac disease in 39 Endemann, D.H. and Schiffrin, E.L. (2004). Endothelial
chronic uremia: pathogenesis. Adv. Ren. Replace. Ther. 4 dysfunction. J. Am. Soc. Nephrol. 15 (8): 1983–1992.
(3): 194–211. 40 Major, R.W., Cheng, M.R.I., Grant, R.A. et al. (2018).
27 McMahon, L.P., Roger, S.D., and Levin, A.; Slimheart Cardiovascular disease risk factors in chronic kidney
Investigators Group (2004). Development, prevention, and disease: a systematic review and meta-analysis. PLoS One
potential reversal of left ventricular hypertrophy in chronic 13 (3): e0192895.
kidney disease. J. Am. Soc. Nephrol. 15 (6): 1640–1647. 41 Golia, E., Limongelli, G., Natale, F. et al. (2014).
28 Levin, A., Thompson, C.R., Ethier, J. et al. (1999). Left Inflammation and cardiovascular disease: from pathogenesis
ventricular mass index increase in early renal disease: to therapeutic target. Curr. Atheroscler. Rep. 16 (9): 435.
impact of decline in hemoglobin. Am. J. Kidney Dis. 34 42 Chen, J., Mohler, E.R. 3rd, Xie, D. et al.; CRIC
(1): 125–134. Investigators (2012). Risk factors for peripheral arterial
29 Yildiz, A., Memisoglu, E., Oflaz, H. et al. (2005). disease among patients with chronic kidney disease. Am.
Atherosclerosis and vascular calcification are J. Cardiol. 110 (1): 136–141.
independent predictors of left ventricular hypertrophy in 43 Chen, J., Mohler, E.R., Xie, D. et al.; CRIC study
chronic haemodialysis patients. Nephrol. Dial. Transplant. investigators (2016). Traditional and non-traditional risk
20 (4): 760–767. factors for incident peripheral arterial disease among
30 Stack, A.G. and Saran, R. (2002). Clinical correlates and patients with chronic kidney disease. Nephrol. Dial.
mortality impact of left ventricular hypertrophy among Transplant. 31 (7): 1145–1151.
new ESRD patients in the United States. Am. J. Kidney 44 Parekh, R.S., Plantinga, L.C., Kao, W.H. et al. (2008). The
Dis. 40 (6): 1202–1210. association of sudden cardiac death with inflammation
31 Paoletti, E., Specchia, C., Di Maio, G. et al. (2004). The and other traditional risk factors. Kidney Int. 74 (10):
worsening of left ventricular hypertrophy is the strongest 1335–1342.
predictor of sudden cardiac death in haemodialysis 45 Schefold, J.C., Filippatos, G., Hasenfuss, G. et al. (2016).
patients: a 10 year survey. Nephrol. Dial. Transplant. 19 Heart failure and kidney dysfunction: epidemiology,
(7): 1829–1834. mechanisms and management. Nat. Rev. Nephrol. 12 (10):
32 Vaziri, N.D. (2006). Dyslipidemia of chronic renal failure: 610–623.
the nature, mechanisms, and potential consequences. 46 Claro, L.M., Moreno-Amaral, A.N., Gadotti, A.C. et al.
Am. J. Physiol. Renal Physiol. 290 (2): F262–F272. (2018). The impact of uremic toxicity induced
33 Postorino, M., Marino, C., Tripepi, G., and Zoccali, C.; inflammatory response on the cardiovascular burden in
CREDIT (Calabria Registry of Dialysis and Transplantation) chronic kidney disease. Toxins 10 (10) pii: E384.
Working Group (2009). Abdominal obesity and all-cause 47 Liu, Y., Coresh, J., Eustace, J.A. et al. (2004). Association
and cardiovascular mortality in end-stage renal disease. J. between cholesterol level and mortality in dialysis
Am. Coll. Cardiol. 53 (15): 1265–1272. patients: role of inflammation and malnutrition. JAMA
34 Axelsson, J. (2008). Obesity in chronic kidney disease: 291 (4): 451–459.
good or bad? Blood Purif. 26 (1): 23–29. 48 Tarrass, F., Benjelloun, M., Zamd, M. et al. (2006). Heart
35 Dasmahapatra, P., Srinivasan, S.R., Mokha, J. et al. valve calcifications in patients with end-stage renal disease:
(2011). Subclinical atherosclerotic changes related to analysis for risk factors. Nephrology 11 (6): 494–496.
chronic kidney disease in asymptomatic black and white 49 Linefsky, J.P., O’Brien, K.D., Katz, R. et al. (2011).
young adults: the Bogalusa heart study. Ann. Epidemiol. Association of serum phosphate levels with aortic valve
21 (5): 311–317. sclerosis and annular calcification: the cardiovascular
36 Pannier, B., Guérin, A.P., Marchais, S.J. et al. (2005). health study. J. Am. Coll. Cardiol. 58 (3): 291–297.
Stiffness of capacitive and conduit arteries: prognostic 50 Scialla, J.J., Xie, H., Rahman, M. et al.; Chronic Renal
significance for end-stage renal disease patients. Insufficiency Cohort (CRIC) Study Investigators (2014).
Hypertension 45 (4): 592–596. Fibroblast growth factor-23 and cardiovascular events in
37 Nakamura, S., Ishibashi-Ueda, H., Niizuma, S. et al. CKD. J. Am. Soc. Nephrol. 25 (2): 349–360.
(2009). Coronary calcification in patients with chronic 51 Ravarotto, V., Simioni, F., Pagnin, E. et al. (2018).
kidney disease and coronary artery disease. Clin. J. Am. Oxidative stress – chronic kidney disease – cardiovascular
Soc. Nephrol. 4 (12): 1892–1900. disease: a vicious circle. Life Sci. 210: 125–131.
38 Chen, J., Budoff, M.J., Reilly, M.P. et al.; CRIC 52 Covic, A., Vervloet, M., Massy, Z.A. et al. (2018). Bone and
Investigators (2017). Coronary artery calcification and risk mineral disorders in chronic kidney disease: implications
of cardiovascular disease and death among patients with for cardiovascular health and ageing in the general
chronic kidney disease. JAMA Cardiol. 2 (6): 635–643. population. Lancet Diabetes Endocrinol. 6 (4): 319–331.
c31.indd 522 09-12-2022 16:04:09

Reference 523
3 Odutayo, A., Wong, C.X., Farkouh, M. et al. (2017). AKI

5 67 Usui, T., Nagata, M., Hata, J. et al. (2017). Serum
and long-term risk for cardiovascular events and non-high-density lipoprotein cholesterol and risk of
mortality. J. Am. Soc. Nephrol. 28 (1): 377–387. cardiovascular disease in community dwellers with
54 Abramson, J.L., Jurkovitz, C.T., Vaccarino, V. et al. (2003). chronic kidney disease: the Hisayama Study. J.
Chronic kidney disease, anemia, and incident stroke in a Atheroscler. Thromb. 24 (7): 706–715.
middle-aged, community-based population: the ARIC 68 Parekh, R.S., Plantinga, L.C., Kao, W.H. et al. (2008). The
Study. Kidney Int. 64 (2): 610–615. association of sudden cardiac death with inflammation
55 Astor, B.C., Coresh, J., Heiss, G. et al. (2006). Kidney and other traditional risk factors. Kidney Int. 74 (10):
function and anemia as risk factors for coronary heart 1335–1342.
disease and mortality: the atherosclerosis risk in 69 Sun, J., Axelsson, J., Machowska, A. et al. (2016). Qureshi
communities (ARIC) study. Am. Heart J. 151 (2): 492–500. AR biomarkers of cardiovascular disease and mortality
56 Luo, J., Brunelli, S.M., Jensen, D.E., and Yang, A. (2016). risk in patients with advanced CKD. Clin. J. Am. Soc.
Association between serum potassium and outcomes in Nephrol. 11 (7): 1163–1172.
patients with reduced kidney function. Clin. J. Am. Soc. 70 Takami, Y., Horio, T., Iwashima, Y. et al. (2004).
Nephrol. 11 (1): 90–100. Diagnostic and prognostic value of plasma brain
57 Kovesdy, C.P., Regidor, D.L., Mehrotra, R. et al. (2007). natriuretic peptide in non-dialysis-dependent CRF. Am. J.
Serum and dialysate potassium concentrations and Kidney Dis. 44 (3): 420–428.
survival in hemodialysis patients. Clin. J. Am. Soc. 71 Codognotto, M., Piccoli, A., Zaninotto, M. et al. (2007).
Nephrol. 2 (5): 999–1007. Renal dysfunction is a confounder for plasma natriuretic
58 Bleyer, A.J., Hartman, J., Brannon, P.C. et al. (2006). peptides in detecting heart dysfunction in uremic and
Characteristics of sudden death in hemodialysis patients. idiopathic dilated cardiomyopathies. Clin. Chem. 53 (12):
Kidney Int. 69 (12): 2268–2273. 2097–2104.
59 Pun, P.H., Goldstein, B.A., Gallis, J.A. et al. (2017). Serum 72 Iwanaga, Y. and Miyazaki, S. (2010). Heart failure,
potassium levels and risk of sudden cardiac death among chronic kidney disease, and biomarkers–an integrated
patients with chronic kidney disease and significant viewpoint–. Circ. J. 74 (7): 1274–1282.
coronary artery disease. Kidney Int. Rep. 2 (6): 1122–1131. 73 Thygesen K, Alpert JS, Jaffe AS, et al.; Joint ESC/
60 Sica, D.A., Struthers, A.D., Cushman, W.C. et al. (2002). ACCF/AHA/WHF Task Force for the Universal
Importance of potassium in cardiovascular disease. J. Definition of Myocardial Infarction, (2012). Third
Clin. Hypertens. 4 (3): 198–206. universal definition of myocardial infarction.
61 Herzog, C.A., Asinger, R.W., Berger, A.K. et al. (2011). Circulation 126(16):2020–3205.
Cardiovascular disease in chronic kidney disease. A 74 Khan, N.A., Hemmelgarn, B.R., Tonelli, M. et al. (2005).
clinical update from Kidney Disease: Improving Global Prognostic value of troponin T and I among
Outcomes (KDIGO). Kidney Int. 80 (6): 572–586. asymptomatic patients with end-stage renal disease: a
62 McCullough, P.A. and Assad, H. (2012). Diagnosis of meta-analysis. Circulation 112 (20): 3088–3096.
cardiovascular disease in patients with chronic kidney 75 Twerenbold, R., Wildi, K., Jaeger, C. et al. (2015).
disease. Blood Purif. 33 (1–3): 112–118. Optimal cutoff levels of more sensitive cardiac troponin
63 Mathew, R.O., Bangalore, S., Lavelle, M.P. et al. (2017). assays for the early diagnosis of myocardial infarction in
Diagnosis and management of atherosclerotic patients with renal dysfunction. Circulation 131 (23):
cardiovascular disease in chronic kidney disease: a 2041–2050.
review. Kidney Int. 91 (4): 797–807. 76 Apple, F.S., Murakami, M.M., Pearce, L.A., and Herzog,
64 Arrigo, M., Cippà, P.E., and Mebazaa, A. (2018). C.A. (2002). Predictive value of cardiac troponin I and T
Cardiorenal interactions revisited: how to improve heart for subsequent death in end-stage renal disease.
failure outcomes in patients with chronic kidney disease. Circulation 106 (23): 2941–2945.
Curr. Heart Fail. Rep. 15 (5): 307–314. 77 Herzog, C.A., Littrell, K., Arko, C. et al. (2007). Clinical
65 Colbert, G., Jain, N., de Lemos, J.A., and Hedayati, S.S. characteristics of dialysis patients with acute myocardial
(2015). Utility of traditional circulating and imaging- infarction in the United States: a collaborative project of
based cardiac biomarkers in patients with predialysis the United States Renal Data System and the National
CKD. Clin. J. Am. Soc. Nephrol. 10 (3): 515–529. Registry of Myocardial Infarction. Circulation 116 (13):
66 Kidney Disease: Improving Global Outcomes (KDIGO) 1465–1472.
Lipid Work Group (2013). KDIGO clinical practice 78 Charytan, D.M., Setoguchi, S., Solomon, D.H. et al.
guideline for lipid management in chronic kidney (2007). Clinical presentation of myocardial infarction
disease. Kidney Int. Suppl. 3: 259–305. contributes to lower use of coronary angiography in
c31.indd 523 09-12-2022 16:04:09

patients with chronic kidney disease. Kidney Int. kidney disease: results from first year of follow-up of the
71 (9): 938–945. Gunma-CKD SPECT multicenter study. Eur. J. Nucl.
79 Sulemane, S., Panoulas, V.F., and Nihoyannopoulos, P. Med. Mol. Imaging 43 (2): 302–311.
(2017). Echocardiographic assessment in patients with 90 Andrade, J.M., Gowdak, L.H., Giorgi, M.C. et al. (2009).
chronic kidney disease: current update. Echocardiography Cardiac MRI for detection of unrecognized myocardial
34 (4): 594–602. infarction in patients with end-stage renal disease:
80 Raggi, P. and Alexopoulos, N. (2017). Cardiac imaging in comparison with ECG and scintigraphy. AJR Am. J.
chronic kidney disease patients. Semin. Dial. 30 (4): Roentgenol. 193 (1): W25–W32.
353–360. 91 Engelbertz, C., Reinecke, H., Breithardt, G. et al. (2017).
81 Dubin, R.F., Deo, R., Bansal, N. et al.; CRIC Study Two-year outcome and risk factors for mortality in
Investigators (2017). Associations of conventional patients with coronary artery disease and renal failure:
echocardiographic measures with incident heart failure the prospective, observational CAD-REF registry. Int. J.
and mortality: the chronic renal insufficiency cohort. Cardiol. 243: 65–72.
Clin. J. Am. Soc. Nephrol. 12 (1): 60–68. 92 Nishimura, H., Miura, T., Minamisawa, M. et al. (2017).
82 Chawla, L.S., Herzog, C.A., Costanzo, M.R. et al.; ADQI Ankle-brachial index for the prognosis of cardiovascular
XI Workgroup (2014). Proposal for a functional disease in patients with mild renal insufficiency. Intern.
classification system of heart failure in patients with Med. 56 (16): 2103–2111.
end-stage renal disease: proceedings of the acute dialysis 93 Sud, M., Tangri, N., Pintilie, M. et al. (2015). ESRD and
quality initiative (ADQI) XI workgroup. J. Am. Coll. death after heart failure in CKD. J Am Soc Nephrol. 26
Cardiol. 63 (13): 1246–1252. (3): 715–722.
83 Hickson, L.J., Negrotto, S.M., Onuigbo, M. et al. (2016). 94 Gregg, L.P. and Hedayati, S.S. (2018). Management of
Echocardiography criteria for structural heart disease in traditional cardiovascular risk factors in CKD: what are
patients with end-stage renal disease initiating the data? Am. J. Kidney Dis. pii: S0272-6386(18)30071-4.
hemodialysis. J. Am. Coll. Cardiol. 67 (10): 1173–1182. DOI: https://doi.org/10.1053/j.ajkd.2017.12.007.
84 Ozkok, S. and Ozkok, A. (2017). Contrast-induced acute 95 Rhee, J.W., Wiviott, S.D., Scirica, B.M. et al. (2014).
kidney injury: a review of practical points. World J. Clinical features, use of evidence-based therapies, and
Nephrol. 6 (3): 86–99. cardiovascular outcomes among patients with chronic
85 Brar, S.S., Aharonian, V., Mansukhani, P. et al. (2014). kidney disease following non-ST-elevation acute
Haemodynamic-guided fluid administration for the coronary syndrome. Clin. Cardiol. 37 (6): 350–356.
prevention of contrast-induced acute kidney injury: the 96 Lau, J.K., Anastasius, M.O., Hyun, K.K. et al. (2015).
POSEIDON randomised controlled trial. Lancet 383 Evidence-based care in a population with chronic kidney
(9931): 1814–1823. disease and acute coronary syndrome. Findings from the
86 Nijssen, E.C., Rennenberg, R.J., Nelemans, P.J. et al. Australian Cooperative National Registry of Acute
(2017). Prophylactic hydration to protect renal function Coronary Care, Guideline Adherence and Clinical Events
from intravascular iodinated contrast material in patients (CONCORDANCE). Am. Heart J. 170 (3): 566–72.e1.
at high risk of contrast-induced nephropathy 97 Ovbiagele, B., Schwamm, L.H., Smith, E.E. et al. (2014).
(AMACING): a prospective, randomised, phase 3, Patterns of care quality and prognosis among
controlled, open-label, non-inferiority trial. Lancet 389 hospitalized ischemic stroke patients with chronic
(10076): 1312–1322. kidney disease. J. Am. Heart Assoc. 3 (3): e000905.
87 Pandya, B., Chalhoub, J.M., Parikh, V. et al. (2017). 98 Shaw, C., Nitsch, D., Lee, J. et al. (2016). Impact of an
Contrast media use in patients with chronic kidney early invasive strategy versus conservative strategy for
disease undergoing coronary angiography: a systematic unstable angina and non-ST elevation acute coronary
review and meta-analysis of randomized trials. Int. J. syndrome in patients with chronic kidney disease: a
Cardiol. 228: 137–144. systematic review. PLoS One 11 (5): e0153478.
88 Gregg, L.P., Adams-Huet, B., Li, X. et al. (2017). Effect 99 House, A.A. (2018). Management of heart failure in
modification of chronic kidney disease on the association advancing CKD: core curriculum 2018. Am. J. Kidney
of circulating and imaging cardiac biomarkers with Dis. 72 (2): 284–295.
outcomes. J. Am. Heart Assoc. 6 (7) pii: e005235. 100 Rocco, M.V., Sink, K.M., Lovato, L.C. et al.; SPRINT
89 Kasama, S., Toyama, T., Sato, M. et al. (2016). Prognostic Research Group (2018). Effects of intensive blood
value of myocardial perfusion single photon emission pressure treatment on acute kidney injury events in the
computed tomography for major adverse cardiac Systolic Blood Pressure Intervention Trial (SPRINT).
cerebrovascular and renal events in patients with chronic Am. J. Kidney Dis. 71 (3): 352–361.
c31.indd 524 09-12-2022 16:04:09

Reference 525
01 Ibrahim, H. and Rao, S.V. (2017). Oral antiplatelet drugs

1 atrial fibrillation in patients with ckd stage 5D: an
in patients with chronic kidney disease (CKD): a review. NKF-KDOQI controversies report. Am. J. Kidney Dis. 70
J. Thromb. Thrombolysis 43 (4): 519–527. (6): 859–868.
102 Neumiller, J.J., Alicic, R.Z., and Tuttle, K.R. (2017). 106 Heiwe, S. and Jacobson, S.H. (2011). Exercise
Therapeutic considerations for antihyperglycemic training for adults with chronic kidney disease.
agents in diabetic kidney disease. J. Am. Soc. Nephrol. 28 Cochrane Database Syst. Rev. (10) (Art. No.:
(8): 2263–2274. CD003236). DOI: https://doi.org/10.1002/14651858.
103 Crowley, M.J., Diamantidis, C.J., McDuffie, J.R. et al. CD003236.pub2.
(2017). Clinical outcomes of metformin use in 107 Navaneethan, S.D., Yehnert, H., Moustarah, F. et al.
populations with chronic kidney disease, congestive (2009). Weight loss interventions in chronic kidney
heart failure, or chronic liver disease: a systematic disease: a systematic review and meta-analysis. Clin. J.
review. Ann. Intern. Med. 166 (3): 191–200. Am. Soc. Nephrol. 4 (10): 1565–1574.
104 Inzucchi, S.E., Lipska, K.J., Mayo, H. et al. (2014). 108 Adler AJ, Taylor F, Martin N, Gottlieb S, Taylor RS,
Metformin in patients with type 2 diabetes and kidney Ebrahim S. (2014). Reduced dietary salt for the
disease: a systematic review. JAMA 312 (24): 2668–2675. prevention of cardiovascular disease. Cochrane
105 Bansal, V.K., Herzog, C.A., Sarnak, M.J. et al. (2017). Database Syst. Rev. Issue 12.(Art. No.: CD009217). DOI:
Oral anticoagulants to prevent stroke in nonvalvular https://doi.org/10.1002/14651858.CD009217.pub3.
c31.indd 525 09-12-2022 16:04:09

526
32
Infection and CKD

Guobin Su1,2,3,4, Masao Iwagami5,6, Helen McDonald7, and Dorothea Nitsch6
1
Department of Nephrology, Guangdong Provincial Hospital of Chinese Medicine, The Second Affiliated Hospital, Guangzhou University of Chinese Medicine, Guangzhou
City, Guangdong Province, China
2
National Clinical Research Center for Kidney Disease, State Key Laboratory of Organ Failure Research, Guangdong Provincial Clinical Research Center for Kidney Disease
3
Department of Nephrology, Nanfang Hospital, Southern Medical University, Guangzhou city, Guangdong Province, China Global Health – Health Systems and Policy,
Department of Global Public Health, Karolinska Institutet, Stockholm, Sweden
4
Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden
5
Department of Health Services Research, University of Tuskuba, Ibaraki, Japan
6
Department of Non-Communicable Disease Epidemiology, London School of Hygiene & Tropical Medicine, London, UK
7
Department of Infectious Disease Epidemiology, London School of Hygiene & Tropical Medicine, London, UK
I ntroduction covered in other chapters: infections as a complication of

hemodialysis (Chapter 43), peritoneal dialysis (Chapter
Nephrologists are familiar with treating dialysis patients 48), and transplantation (Chapter 59), and infections which
with access infections or peritonitis. Nephrologists are also may cause CKD (Chapter 20 and 26).
familiar with the multitude of infectious complications
seen in immunosuppressed patients, such as those with
vasculitis and those who have received a kidney transplant. KD and Risk of Developing
C
However, the vast majority of people with chronic kidney Infections
disease (CKD) are seen by general physicians and other
specialist doctors, and nephrologists will often be asked on A 2014 systematic review of CKD and infection risk found
ways to prevent infections and address complications in 14 studies comparing incidence of infection among people
this broader group of patients, many of whom do not have with CKD to the incidence among people without known
a specified diagnosis as an underlying cause of CKD. history of CKD [2]. Figure 32.1 displays the associations
In this chapter, we discussed the relevant evidence to of CKD with infection risk. Studies that systematically
support management of infectious complications associ- excluded patients receiving dialysis are grouped separately
ated with CKD. The focus is on patients with CKD (not on from studies in which patients receiving dialysis were,
kidney replacement therapy), but frequently evidence to or may have been, included together with patients with
support treatment decisions has been extrapolated from CKD. Seven studies were set among the general popula-
those on dialysis therapy to those on CKD. tion, and the remainder in higher risk populations (e.g.
This chapter covers what is known about the risk of patients seen in kidney clinics, with diabetes, or of older
infection, reasons for the higher infection risk, the risk of age). Although a range of infections were investigated,
infection-associated complications, and how to prevent most required hospital admission or blood culture.
and treat infections among patients with CKD (Table 32.1). Unsurprisingly in view of the range of infection outcomes
Much of the data on infections as a complication of CKD investigated and the diversity of how kidney disease was
derives from high-income countries. Due to limited infor- defined, there was great heterogeneity (I2 = 96.5%) and no
mation on whether CKD predisposes to infections that are meta-analysis was performed. Nevertheless, all point
predominant in low-income settings such as malaria or estimates and most confidence intervals (CIs) suggest a
human immunodeficiency virus (HIV), these conditions strong and graded association of infection risk with CKD
are not discussed. Additional infection-related topics are and its severity.
Table 32.1 Chronic kidney disease and infection risk: KDIGO recommendations [1].
Strength of
Intervention Recommendation Certainty of evidence recommendationa
Antibiotic prescribing Dosing must take the patient’s impaired kidney function into Moderate to high Strong (1A)
consideration by adjusting the dosing interval (if the drug Evidence of drug levels mainly from pharmacokinetic and
effect depends on concentration, e.g. aminoglycosides) or dose pharmacodynamics studies with strong observational post-marketing
level (for drugs which require constant levels such as evidence of drug side effects for older drugs
beta-lactams) or considering a nonrenally cleared alternative Literature on preventing antimicrobial resistance is focused on overall
Consider drug interactions and follow local guidelines to prescribing irrespective of CKD status (unless there is a concern about
minimize antimicrobial resistance toxicity or efficacy)
Annual vaccination KDIGO recommend that all adults with CKD are offered Low Strong (1B)
against influenza annual vaccination with inactivated influenza vaccine, unless Extrapolation from observational data analysis. To date there is no/
contraindicated [1] limited evidence of vaccine efficacy for CKD stages specifically. However,
given that the vaccine is known to be efficacious in older population
overall, amongst whom a sizable proportion have some degree of CKD,
there appears overall benefit, especially considering the high rate of
infection-related hospitalization for those with CKD. Live-attenuated
influenza vaccination has not been tested in patients with CKD.
Vaccination with KDIGO recommend that all adults with eGFR < 30 ml/ Low Strong (1B)
polyvalent min/1.73 m2(GFR categories G4–G5) and those at high risk of There are limited observational data on vaccination efficacy by CKD
pneumococcal pneumococcal infection (e.g. nephrotic syndrome, diabetes, or stages. However, given that the vaccine is known to be efficacious in
vaccine those receiving immunosuppression) receive vaccination with older population overall, amongst whom a sizable proportion have some
polyvalent pneumococcal vaccine unless contraindicated [1] degree of CKD, there appears overall benefit, especially considering the
high rate of infection-related hospitalization for those with CKD.
Repeat vaccination KDIGO recommend that all adults with CKD who have Low Strong (1B)
with polyvalent received pneumococcal vaccination are offered revaccination There are limited data to support a particular revaccination scheme, but
pneumococcal vaccine within 5 years [1] in advanced CKD antibody response is less marked and titres decline
within 5 years faster than for people with normal kidney function
Immunization against KDIGO recommend that all adults who are at high risk of Moderate Strong (1B)
hepatitis B progression of CKD and have GFR > 30 ml/min/1.73 m2 (GFR Those with higher eGFR have shown better seroconversion in response
categories G4–G5) be immunized against hepatitis B and the to hepatitis B vaccine
response confirmed by appropriate serological testing [1]
Use of live vaccines Consideration of live vaccine should include an appreciation Very low Not graded
of the patient’s immune status and should be in line with Due to the fact that people with CKD are often immunocompromised to
recommendations from official or governmental bodies [1] some degree, as evidenced from observational data analysis, there may
be a risk of vaccine-related disease
Vaccination in Pediatric immunization schedules should be followed Very low Not graded
children according to official international and regional As pediatric kidney disease is rare, there are no or very limited data on
recommendations for children with CKD [1] this topic
a
Level of strength of recommendation as per the KDIGO recommendations: Level 1, We recommend; Level 2, We suggest. Letter grading refers to quality of evidence: A, High; B, Moderate.
Source: Adapted with permission from Kidney Disease: Improving Global Outcomes CKD Work Group [1].
c32.indd 527 09-12-2022 16:03:33

528 Infection and CKD
Relative risk
Study
(95% Cl)
ESRD included, or ESRD status unclear
Karunajeewa 2005 Urinary sepsis 1.50 (1.10,1.90)
Caljouw 2011 UTI 0.90 (0.50, 1.70)
Vinogradova 2009 Pneumonia 1.72 (1.43, 2.07)
Loeb 2009 Pneumonia 4.06 (1.98, 8.35)
Schnoor 2007 Pneumonia 1.70 (1.10, 2.80)
Campbell 2011 Influenza A (H1N1) 17.50 (13.40, 22.90)
Watt 2007 Invasive pneumococcal disease 2.60 (0.87, 7.70)
Hackam 2006 Sepsis 1.47 (1.27, 1.72)
Wang 2012 Sepsis 1.99 (1.73, 2.29)
ESRD excluded
Higgins1985 UTI 1.5
Dalrymple 2012 Genitourinary infections 1.17 (0.67, 2.05)
USRDS 2010 UTI 3.15
Dalrymple 2012 Pulmonary infections 1.27 (0.94, 1.71)
James 2009 Pneumonia 18–54 years 3.23 (2.40, 4.36)
James 2009 Pneumonia 55–64years 1.43 (1.11, 1.84)
James 2009 Pneumonia 65–74 years 1.18 (0.99, 1.40)
James 2009 Pneumonia 75+ years 0.95 (0.85, 1.05)
USRDS 2010 Pneumonia 2.76
Dalrymple 2012 Bacteremia/sepsis 1.55 (0.93, 2.57)
James 2008 Bloodstream infection 1.17 (0.92, 1.49)
USRDS 2010 Bacteraemia 3.90
.0437 1 22.9
Effect estimate (log scale)
Figure 32.1 Forest plot of all estimates of the association of chronic kidney disease with infection risk (n = 17) from all 14 studies
identified. The estimates from Higgins 1985 and USRDS 2010 did not include SEs. Dalrymple 2012: presented estimates compare eGFR
45–59 with eGFR ≥90 ml/min/1.73m2; James 2009: presented estimates compare eGFR 45–59 with eGFR 60–104 ml/min/1.73m2;
James 2008: presented estimates compare eGFR 45–59 with eGFR≥60 ml/min/1.73m2. eGFR, estimated glomerular filtration rate; ESRD,
end-stage renal disease; USRDS, US Renal Data System; UTI, urinary tract infection. Source: From McDonald et al. [2].
Since this systematic review, there have been a further risk, adjustment for the comorbidity did not explain this
five studies investigating the association of CKD (defined a ssociation fully.
using estimated glomerular filtration rate [eGFR]) with a As people with CKD are often multimorbid and frail, it
range of infections, from the United States [3], Sweden [4], is perhaps unsurprising that people with CKD are more
and the UK [5–7]. The results were all consistent with the likely to experience severe infections such as sepsis or
previous evidence of a graded association between CKD pneumonia requiring hospitalization. These associations
and increased risk of infection. could be due to increased severity of infection rather than
Most of the available studies of infection in people with increased risk. However, some studies have investigated
CKD have looked at severe infection or hospitalization the association of CKD with community-acquired
with infection, such as those of James and colleagues from infection, including infection diagnosed and treated in the
2008 to 2009 [8, 9]. These studies used laboratory-confirmed community. This approach more fully quantifies the
CKD and investigated severe community-acquired infec- burden of community-acquired infection, an important
tion (sepsis and hospitalization with pneumonia), taking and common problem, but also isolates the impact of CKD
into account underlying comorbidities. Notably, while on increased infection risk, distinct from the impact on
underlying comorbidity did explain some of the underlying infection severity.
CKD and Susceptibility to Infectio 529
Studies of infections in primary care have identified have had previous infections may have more frequent
associations between lower eGFR and increased risk of a assessment for kidney function and repeat UTI. There may
range of infections, including pneumonia, urinary tract also be a particular group of patients in whom the recur-
infections (UTI), sepsis, skin/soft tissue infections, and rent UTIs have caused CKD, which in turn predisposes
gastrointestinal tract infections [4, 5, 10, 11]. Table 32.2 them to more infections. Both these issues introduce a
summarizes two studies of community-acquired infection. problem of reverse causality into the data. While it is
The Swedish study [4] was set among a cohort of all evident that UTIs are common among people with CKD, it
patients from Stockholm Region with a creatinine is not clear whether or to what extent CKD causes UTIs.
laboratory result available. It studied a range of community- In summary, infections are common among people with
acquired infections and showed that there was a graded CKD and the risk of infection increases as kidney function
increase in the risk of community-acquired infections as declines. Infections pose a substantive burden to secondary
eGFR declined. Lower respiratory tract infections, UTIs, care as people with CKD are likely to require hospitaliza-
and sepsis became more common as a percentage of total tion owing to the severity of infection, especially with sepsis
recorded infections as eGFR declined. An association with and severe LRTI. Neither the associations with hospitaliza-
skin/soft tissue infections and gastrointestinal tract infection and severe infections, nor those with infections diag-
tions was also observed. No association was found with nosed and treated in the community, were fully explained
nervous system and upper respiratory tract infections. The by adjustment for comorbidities. Hence, there seems to be a
UK study [5] used a similar design among older people biological association by which the presence of CKD
with diabetes, who have complete kidney function data increases the susceptibility to infection risk as well as sever-
due to incentivized testing for kidney function among ity. How this arises will be discussed in the next section.
these patients in the UK. Both the Swedish and the UK
study adjusted for underlying comorbidities, which again
did not fully explain the associations noted. CKD and Susceptibility to Infection
In addition to a graded association between eGFR and
risk of community-acquired lower respiratory tract infec- As discussed above, patients with CKD are at greater risk of
tions (LRTI) and sepsis, proteinuria was also independently infection, even well before any exposure to kidney replace-
associated with increased risk of both these community- ment therapy and its attendant complications [12–16].
acquired infections – this is in line with what was seen for This may be due to multiple underlying risk factors, or due
severe infections requiring hospitalizations in the athero- to treatment for the kidney disease, or a phenomenon
sclerosis risk in communities (ARIC) study [3]. related to reduced immunity as kidney function declines
The observed association between CKD and UTI may irrespective of the cause of kidney disease. We discuss each
need to be interpreted with caution. Recurrent UTIs are of these in turn.
very common and frequently require hospitalization One risk factor that may be underlying the risk of
among people with CKD. There is a risk that those who observed infection is having had a previous infection. For
Table 32.2 Studies investigating the associations between CKD and risk of community-acquired infections using electronic health
records (since 2014).
Studies of community-acquired infections Crude IR RR for eGFR categories in ml/min/1.73m2 (95% CI)
Per 1000
Population person- eGFR eGFR eGFR
Author Year Country studied Outcome years eGFR 60+ 45–59 30–44 15–30 eGFR <15
Xu [4] 2017 Sweden General all infections 95 1 1.08 1.53

population (1.01–1.14) (1.39–1.69)
McDonald [5] 2015 UK Diabetes LRTI 155.8 1 1.03 1.08 1.17 1.47
(1.01–1.04) (1.05–1.10) (1.13–1.22) (1.34–1.62)
Pneumonia 10.3 1 0.95 1.19 1.73 3.04
(0.89–1.01) (1.11–1.28) (1.57–1.92) (2.42–3.83)
Sepsis 2.5 1 1.11 1.51 2.50 5.56
(0.99–1.25) (1.32–1.73) (2.08–3.00) (3.90–7.94)
RR, risk ratio; IR, incidence rate; 95% CI, 95% confidence interval; LRTI, lower respiratory tract infection.
some patients, CKD is a direct result of infection, from considering anemia as a potential mediator of infection
recurrent UTIs, to HIV and tuberculosis (TB). However, for risk, those patients with severe anemia tend to be most mul-
the majority of patients with CKD, old age, underlying, or timorbid, and anemia may be a marker of underlying
comorbid disease and lifestyle factors, such as diabetes and chronic infection. Moreover, anemia of chronic illness may
smoking, are independent risk factors for infection. With be a result of a physiologic process by which the body tries
regards to old age, the term immunosenescence has been to sequester iron (with resulting anemia) to prevent
used to describe the changes in immune function that make increased infection risk [22]. Again, we have to wait for the
older people more susceptible to infection [17], mainly via results from randomized trials to determine whether
reduced response to vaccination via impaired T-cell func- treating anemia indeed reduces infection risk – it may well
tion and the interplay of T cells with other parts of the be the opposite as intravenous iron can cause an increased
immune system. Diabetes itself is a known risk factor for risk of infection [23].
serious bacterial infections [18], and glycemic control a As CKD worsens, a reduced response of the adaptive
main risk factor for developing infections and requiring immune system has been observed in the context of a
hospitalization with infections [19]. Smoking is a risk factor reduced response to vaccinations [24, 25]. This impairment
common to CKD and bacterial and viral infection. may be in part driven by uremic toxins [26], which impair
Nevertheless, as discussed above, even taking these factors the T-lymphocyte dependent response (required to respond
into account, an increased risk of community-acquired to vaccinations) [27] and also the phagocytosis response of
infections is consistently observed as eGFR declines. polymorphonuclear lymphocytes [28, 29]. This reduced
Treatment for CKD and underlying diseases with steroids response to vaccination itself further increases the risk of
and immunosuppressants can further predispose to vaccine-preventable infections.
infection. However, the vast majority of people with CKD in Figure 32.2 illustrates that the association between CKD
the general population do not have treatment with and susceptibility to infection is likely driven by a
immunosuppressants or severely immunocompromising combination of the patient population’s age and comor-
disease. Instead, the most common underlying conditions bidities, together with a likely impairment of the adaptive
are old age, diabetes, a history of smoking, and exposure to immune system caused by CKD.
infectious agents from frequent healthcare attendance [12].
CKD itself may have a causal role in increased infection
risk through additional disturbance of both innate and omplications of Infections among
C
adaptive immunity. Chronic inflammation, proteinuria, People with CKD
and uremia may each play separate roles. There are multiple
potential mechanisms for CKD to alter cell-mediated and We review below a number of key complications of infec-
humoral immune system function. A number of observation that are relevant to people with CKD, including mor-
tional studies have considered factors such as malnutrition, tality, acute kidney injury (AKI), progression to dialysis,
hypoalbuminemia, anemia, complement loss, disrupted and cardiovascular disease. These associations highlight
calcium regulation and vitamin D insufficiency, and chronic that infections should not be taken lightly, as they may
kidney inflammation [12, 20]. Some of these factors, such have serious consequences.
as anemia, malnutrition, and low vitamin D, may be
modifiable with good clinical care. However, before
Mortality
advocating replacement of vitamin D or treatment of
anemia to prevent infections, clinicians should remember There is a graded association of CKD with mortality risk
that lack of vitamin D and anemia may simply be markers from infection. The rate of infection-related mortality and
of severity of overall illness and that treating these may fail the proportion of deaths due to infection increase as eGFR
to change, or possibly even increase, infection risk. For declines among the general population [3, 30–32]. During
example, vitamin D is a marker of sun exposure, and 13 years of follow-up in the National Health and Nutrition
bed-bound patients may be less likely to be out in the sun, Examination Survey study (NHANES) study, death from
so any observational study is likely confounded by disease infection occurred in 1.2% of the population with normal
severity. A review of existing trials of vitamin D eGFR, increasing to 4.6% in those with eGFR 45–59 ml/
supplementation to prevent acute respiratory infections min/1.73m2 and 10.3% in those with eGFR < 45 ml/
suggests a potential role for patients with very low vitamin min/1.73m2. Compared to people with eGFR 60 ml/
D levels [21]. At present, there are ongoing trials to test min/1.73m2, the corresponding hazards of infection-related
whether vitamin D supplementation improves outcomes, mortality were 1.36 (95% CI 0.81–2.30) and 2.36-fold (1.04–
including infections in the CKD and dialysis population. In 5.38) higher for eGFR of 45–59 and <45 ml/min/1.73m2,
Complications of Infections among People with CK 531
CKD RISK FACTORS AND ASSOCIATED CHARACTERISTICS CHRONIC KIDNEY DISEASE
Population characterised Specific causes (e.g. Infectious Decreased immune

by older age, frailty, Diabetes mellitus, causes of CKD response (e.g. T cells)
history of smoking, Sickle cell disease) (e.g. TB,
multimorbidity recurrent UTI) Decreased vaccine
response
Cardiovascular Exposure to Immunosuppressive

and pathogens from treatments
cerebrovascular healthcare
events attendance
INCREASED INFECTION INCIDENCE, WITH POORER OUTCOMES
Cardiovascular events Death Acute kidney injury Progression of CKD
Figure 32.2 Risk factors and outcomes of infection in chronic kidney disease.
respectively [32]. Influenza/pneumonia and sepsis are the ratio 2.99 (95% CI 1.20–7.51) [40]. In a systematic review of
infections that most commonly cause death in patients with 19 studies of Clostridium difficile infection, a pooled analy-
CKD [31]. The graded association of lower kidney function sis of 116 875 patients found the mortality risk associated
with higher infection-related mortality is repeated in the with CKD was 1.76 (95% CI 1.26–2.47) [44].
older population [8, 9, 33, 34], those with diabetes [35], and For patients with CKD, the implication of an infection for
among older adults resident in nursing homes [34]. longer term prognosis is also considerable. In a Canadian
The increased risk of death is not simply due to increased prospective cohort (CanPREDDICT) of patients with
risk of acquiring infection. If a patient gets admitted with advanced CKD (eGFR 15–45 ml/min/1.73m2) followed for
infection, the presence of CKD worsens their outlook up to 5 years, those who experienced an infection had over
considerably. Several studies show that among those three times increased risk of mortality than those without
hospitalized with any infection, lower kidney function is infection (hazard ratio 3.39, 95% CI 2.65–4.33) [45].
associated with higher in-hospital mortality [36] and
subsequent long-term all-cause mortality [37] (Table 32.3).
AKI and Increased Risk of Progression to ESKD
The crude mortality rates are significant. In those hospital-
ized with sepsis/bloodstream infection, 20–22% of patients It is well established that infection, especially severe infec-
with eGFR < 60 ml/min/1.73m2 died in hospital [36, 43], tion such as sepsis, increases the risks of AKI and of
49–70% died within 28 days [39, 41, 42], 62% died within requiring kidney replacement therapy [46]. Among those
90 days [42], and 82% died within a year [39]. In those hos- with AKI, 86.9% cases were attributed to sepsis or hypov-
pitalized with pneumonia, 9–19% of patients with olemia [47]. When an infection occurs, underlying kidney
eGFR < 60 ml/min/1.73m2 died in hospital [36, 38], 29–47% function impairment increases the risk of AKI. In septic
died within 28 days [42], and 36–53% died within shock, patients with CKD (59.5%) had a higher risk of AKI
90 days [42]. In a study enrolling patients hospitalized with than those without CKD (40.8%) [48]. Among patients
infection at four hospitals in China, heightened risks of infected with influenza A, 67–69% of patients with CKD had
mortality were greatest in the first 7 days following admis- AKI [49, 50]. However, there are other studies that have not
sion and progressively declined thereafter, but remained found associations between CKD and AKI risk in
significantly elevated at a year following admission [37]. infection [51, 52].
This increased mortality risk can be seen also for infections An infection in the context of advanced CKD can propel
other than sepsis and pneumonia. In a retrospective patients onto dialysis. In the CanPREDDICT study of
Japanese cohort study enrolling patients with smear- patients with advanced CKD (eGFR 15–45 ml/min/1.73m2),
positive pulmonary tuberculosis, eGFR of <30 ml/ those with infection had a 58% increased risk of end-
min/1.73m2 was a risk factor for death with adjusted odds stage kidney disease (ESKD) (hazard ratio 1.58, 95%
Table 32.3 Risk of death in people with infection by level of kidney function (baseline eGFR).
Age (mean ± SD, Kidney function

Type of Number of medium (eGFR: ml/ Number with
Identity Region infection Study population subjects [quartile]) min/1.73 m2) Outcome outcome Crude risk (%) Relative risk
For those with infection (sepsis, pneumonia, etc.), how does reduced renal function increase risk of death?
Viasus Spain Pneumonia Hospitalized with 3597 70 [56–79] 60 In-hospital 290 8.30 NA
et al. [38] pneumonia 105 77 [67–84] 30–59 mortality 16 15.20
42 15–29 8 19.00
10 <15 1 10.00
Maizel France Septic shock Patients with septic shock 107 63 [54–76] 60 28-day 54 50 1 (ref)
et al. [39] in the Medical ICU 56 72 [62–79] <60 mortality 39 70 aHR 1.7 (1.08–2.68)
Igari Japan Smear-positive Patients (12–99 years) 499 62.1 (19.4) 60 Death 62 12.40 NA
et al. [40] pulmonary hospitalized for 629 <60 93 14.80 NA
tuberculosis tuberculosis
Mansur Germany Sepsis Adult patients admitted to 426 63 (15) No-CKD Survival up Not reported 27 1 (ref)
et al. [41] surgical ICU 56 CKD (unclear to 90 days Not reported 54 aHR 2.25 (1.46–3.46)
definition)
McDonald England Pneumonia People aged 65 years with 2531 78 [72–83] 60 28-day 687 27.10 1 (ref)
et al. [42] diabetes mellitus and 1162 82 [77–87] 45–60 mortality 332 28.60 aOR 0.93 (0.85–1.04)
community-acquired
pneumonia or sepsis 764 30–44 265 34.70 aOR 0.99 (0.88–1.12)
286 <30 122 42.70 aOR 1.30 (1.12–1.52)
Sepsis 566 76 [71–82] 60 28-day 179 31.60 1 (ref)
232 81 [75–86] 45–60 mortality 76 32.80 aOR 0.91 (0.72–1.14)
190 30–44 88 46.30 aOR 1.24 (0.99–1.54)
70 <30 34 48.60 aOR 1.37 (1.05–1.79)
Neyra USA Sepsis ICU adults with severe 1421 Not reported >60 In-hospital 339 23.90 NA
et al. [43] sepsis or septic shock (around 62) mortality
1211 Not reported 15–59 264 21.80
(around 70)
c32.indd 532 09-12-2022 16:03:37

Su China Overall Adult (>18 years) 4348 66 [53–77] 60 In-hospital 239 5.50 1 (ref)
et al. [36] hospitalized with infection 1935 77 [68–83] <60 mortality 235 12.20 aOR 2.01 (1.34–3.02)
Bloodstream 129 Not reported 60 23 17.80 1 (ref)
infections or 80 <60 16 20.00 aOR 0.93 (0.35–2.48)
sepsis
Pneumonia 1447 60 111 7.67 1 (ref)
774 <60 72 9.30 aOR 1.04 (0.73–1.51)
Urinary tract 875 60 7 0.80 1 (ref)
infections 818 <60 18 2.20 aOR 2.26 (0.58–8.77)
Su China Any infection- Adult (>18 years) 32 388 57 [42–70] 60 All-cause 2964 NA 1 (ref)
et al. [37] related hospitalized with infection 5527 77 [66–82] 30–59 death for a 1088 aHR 1.19 (1.11–1.28)
hospitalization follow-up of
2609 71 [58–80] <30 median 729 aHR 1.73 (1.59–1.88)
2.39 years
c32.indd 533 09-12-2022 16:03:37

CI 1.22–2.05) [45]. In a population-based retrospective Most research into vaccination in kidney disease has
cohort study from Taiwan’s National Health Insurance focused on patients with ESKD [59]. Since immune respon-
Research Database, the overall incidence rate ratio of siveness declines progressively with worsening kidney
ESKD adjusted for competing mortality was 23% higher in function, evidence of vaccine effectiveness and safety
those with pneumococcal pneumonia than in those with- among patients with ESKD is included where necessary to
out pneumococcal pneumonia ((crude ESKD rates 5.26 vs. indicate likely effectiveness or safety for patients with
3.10 per 1000 person-years), with an adjusted sub-hazard CKD. Vaccination guidance for kidney transplant recipi-
ratio of 1.14 (95% CI 1.01–1.29) [53]. ents and patients with end-stage renal disease (ESRD) is
covered in other chapters and should be consulted sepa-
rately, in particular noting advice on live and live-
Cardiovascular Events
attenuated vaccines.
It is well known that for the general population acute infec-
tions can lead to an acute but temporary increase in the Pneumococcal Vaccine
risk of arterial and venous thrombotic events [54, 55]. Streptococcus pneumonia is the most common cause of
Downstream effects of acute infections include dysfunc- community-acquired pneumonia, which causes a high
tion of endothelial, redox, and coagulation systems, all of burden of mortality and morbidity among patients with
which occurs in the face of diminished cellular oxygen CKD [5, 42, 60]. Systematic reviews of the effectiveness of
availability, cytokine storm, sympathetic overload, comple- pneumococcal vaccine in the general population have been
ment activation, and vasoconstriction of some vascular hampered by heterogeneity [61], which may be attributa-
beds (including that of the kidney) [14, 56, 57]. The ble to varying lengths of follow-up as immunity wanes over
CanPREDDICT study of patients with advanced CKD time [62]. An observational study that explored how predi-
(eGFR 15–45 ml/min/1.73m2) found that those with an epi- alysis CKD related to vaccine effectiveness among people
sode of infection had an increased risk of cardiovascular aged 65 years with diabetes found some evidence that vac-
ischemia (hazard ratio 1.80, 95% CI 1.24–2.60) and conges- cine effectiveness was lower among patients with a history
tive heart failure (HR 3.22, 95% CI 2.25–4.61) [45]. of proteinuria (P = 0.04), but no evidence of variation
In summary, kidney dysfunction worsens prognosis in according to eGFR. Vaccine effectiveness was 22% (95% CI
those who present with infection, while infection worsens 11–31%) in the first year after vaccination, but was negligi-
kidney function acutely and may propel people with ble after 5 years [63].
advanced CKD onto dialysis. Most important though is the Kidney Disease: Improving Global Outcomes (KDIGO)
considerable risk of mortality if a person with reduced kid- guidelines recommend that all adults with eGFR < 30 ml/
ney function is hospitalized with infection. In the case of min/1.73 m2 (glomerular filtration rate (GFR) categories
sepsis, which is known to be more common among people G4-G5) and those at high risk of pneumococcal infection
with CKD, there is clear evidence that early recognition (e.g. nephrotic syndrome, diabetes, or those receiving
and treatment can improve outcomes. Patients with CKD immunosuppression) receive vaccination with polyvalent
should be advised to seek help early when they have symp- pneumococcal vaccine unless contraindicated, and that all
toms of infections. adults with CKD who have received pneumococcal vacci-
nation are offered revaccination within 5 years [1]. UK
guidelines recommend vaccination every 5 years with
reventing and Treating Infections
P 23-valent pneumococcal polysaccharide vaccine for chil-
among People with CKD dren and adults with nephrotic syndrome, CKD stages 4
and 5, and those on kidney dialysis or with kidney
Preventing Infections in CKD transplantation [58].
Vaccination is the best strategy for preventing morbidity
and mortality from infection, and is vital among this high- Influenza
risk population. Vaccines particularly relevant for patients Influenza vaccination has the potential to reduce morbidity
with CKD are considered below: pneumococcal and inacti- and mortality from both influenza and pneumonia. Up to
vated influenza vaccinations are recommended for patients one-third of community-acquired pneumonia may be
with CKD, and patients awaiting dialysis or kidney trans- influenza-related, due to bacterial co-infection or secondary
plant are recommended to be vaccinated against hepatitis bacterial pneumonia [64]. A 2014 systematic review of
B and may benefit from varicella or herpes zoster vaccina- influenza vaccination among patients with ESKD found
tion [1, 58]. five observational studies, none of which included safety
Preventing and Treating Infections among People with CK 535
data, and concluded that the evidence on effectiveness was vaccinating at initiation of hemodialysis [71–73]. Early
limited and of low quality [65]. One of the included studies vaccination may be of additional benefit to patients with
is often cited as evidence that influenza vaccination is diabetic nephropathy who may be at risk through blood
effective in ESKD; in this observational study of US glucose monitoring [74]. Most research into hepatitis B
Medicare patients receiving dialysis, influenza vaccination vaccination describes effectiveness at seroconversion as a
was associated with lower risk of hospitalization and surrogate for protection against hepatitis B infection [70].
death [66]. However, McGrath et al. found that US A 2004 Cochrane review identified seven randomized
Medicare patients receiving hemodialysis were less likely clinical trials of hepatitis B vaccine effectiveness among
to be vaccinated if they had frequent or longer patients with kidney failure [70]. The review found that
hospitalizations or required skilled nursing, suggesting an plasma hepatitis B vaccines are safe and effective at gener-
overestimation of the effectiveness of influenza vaccination ating protective antibody levels among patients receiving
in this dataset due to a “healthy vaccinee” effect [67]. When dialysis. The two trials among patients with predialysis
McGrath et al. adjusted the vaccine effectiveness in CKD found no evidence of a difference in seroconversion
well-matched seasons for that observed in mismatched following recombinant or plasma vaccine (risk ratio [RR]
seasons (in which little or no vaccine effect should be seen 0.65, 95% CI 0.28–1.53) [75, 76].
due to mismatch of the influenza type with vaccine), they Seroconversion is lower among patients with CKD than
found little evidence of increased vaccine effectiveness in the general population [25, 59]. Systematic reviews have
well-matched years among US Medicare patients receiving found a short-term improvement in seroconversion with
hemodialysis [68]. intradermal (rather than intramuscular) delivery and no
One study explored whether CKD affects vaccine effec- evidence of additional benefit from “double dose” hepatitis
tiveness among older people with diabetes [63]. The B vaccination compared to standard doses [77, 78]. A rand-
“healthy vaccinee” effect was adjusted for by dividing omized trial of 165 patients with CKD or receiving hemodi-
vaccine effectiveness in winter by that observed in alysis compared 20 μg Fendrix to double doses (2 × 20 μg
summer. It found a modestly protective effect of influenza HBsAg) of Engerix vaccine [79]. Fendrix is adjuvanted by
vaccination against LRTI (7%, 95% CI 3–12) with no monophosophoryl lipid A and adsorbed onto aluminum
evidence of variation in vaccine effectiveness according to phosphate. Patients who received Fendrix had higher levels
CKD stage. This may be an underestimate for patients of seroconversion, with faster onset and longer duration of
with CKD, as vaccine effectiveness diminishes with age protective antibody levels. There was a higher incidence of
and the outcome was not specific to influenza. Given the pain at the injection site with Fendrix, but no evidence of
high morbidity and mortality associated with influenza any difference in serious adverse events. Other adjuvants
and secondary pneumonia among patients with CKD, that have shown promise among patients with kidney dis-
even a low vaccine effectiveness would support ease are levamisole [80], granulocyte macrophage colony-
vaccination. stimulating factor (GM-CSF) [25], and a toll-like receptor 9
KDIGO guidelines recommend that all adults with CKD agonist [81]. There may be scope to improve hepatitis B
are offered annual vaccination with influenza vaccine immunity for patients with CKD through further research
unless contraindicated due to allergy to seasonal influenza into the optimal vaccine dose, type, route, and schedule.
vaccine [1]. Live-attenuated influenza vaccination has not KDIGO guidelines recommend vaccination for adults
been tested in patients with CKD [69]. UK guidelines rec- with eGFR < 30 ml/min/1.73m2 and at high risk of progres-
ommend offering inactivated influenza vaccination to sion of CKD, with serological testing to confirm response [1].
adults with CKD stages 3–5, chronic kidney failure, or UK guidelines recommend vaccinating patients with CKD
nephrotic syndrome, with recommendations for children as soon as it is anticipated they may require dialysis or kid-
varying according to age [58]. ney transplantation, using Fendrix if they are 15 years old,
testing antibody levels annually, and providing a booster
Hepatitis B dose for patients with antibodies below 10 mIU/ml [58].
Guidelines for hepatitis B vaccination among patients with
CKD have aimed to anticipate the risk of infection from Zoster
dialysis [70]. Hepatitis B prevalence among patients receiv- CKD increases the risk of herpes zoster, and the CKD
ing hemodialysis ranges from 1.1% to 6.1%, and patients patient population also has a high prevalence of diabetes
who develop hepatitis B while receiving dialysis are more and older age, both of which are risk factors for herpes zos-
likely to become chronic carriers [70]. Vaccinating patients ter [82–84]. Herpes zoster may worsen kidney outcomes
with CKD stages 3–4 achieves better seroconversion than for patients with CKD [83]. However, older people in
Medicare are less likely to receive herpes zoster vaccination with CKD, of any stage, have sufficient immune function
if they have CKD than if they do not [85]. to receive all live vaccines for which an inactivated vaccine
A large observational study of herpes zoster vaccination is not an alternative [69].
among Medicare recipients aged 65 years found an adjusted
vaccine effectiveness of 49% (95% CI 36–56) among people Vaccination: Summary
with CKD and 46% (95% CI 9–68) among people with both In general, patients with CKD are at greater risk of infec-
CKD and diabetes, both similar to that among the general tion and complications from infection. In addition to
population of older people (vaccine effectiveness 48%, 95% routine vaccinations, influenza, pneumococcal, hepatitis
CI 39–56) [82]. Small studies of varicella zoster vaccination B, and zoster vaccinations should be considered. Immuno-
among children with kidney disease observed high serocon- suppression must be considered before delivering live or
version following a second dose (>98%) and antibodies per- live-attenuated vaccines, but most patients with CKD may
sisting over 3 years [86–88]. safely receive all live vaccines for which an inactivated vac-
Varicella zoster vaccination is part of the routine child- cine is not available [69].
hood vaccination schedule in the United States, while the Vaccination is a vital but under-used tool among patients
UK vaccinates older people against herpes zoster [58]. with CKD [85, 89]. Although it is true that vaccine response
Among patients with CKD, who are at greater risk of severe can be diminished in CKD compared to the general popu-
disease than the general population, vaccination may have lation, vaccines prevent infections and improve outcomes.
greater benefits [82]. As both varicella and herpes zoster vac- Indeed, given the higher risk of vaccine-preventable infec-
cination use live attenuated virus vaccines, care must be tion and poorer outcomes of infection associated with
taken to note contraindicating immunosuppression, and CKD, vaccinations are particularly important among this
susceptibility and eligibility for varicella or herpes zoster high-risk population.
vaccination should be considered in advance of immuno-
suppression for potential kidney transplant candidates [58].
Treating Infections in CKD
Routine Childhood Vaccinations Considering the poor outcomes of infections in those
In general, children with CKD are at increased risk for with CKD, it is imperative that infections are treated
vaccine-preventable diseases and should receive all the promptly but appropriately to achieve maximum effect.
routinely recommended childhood immunizations as far Patients with CKD are not only subject to infection, but,
as possible; however, vaccination is too often delayed or if inappropriately managed, may plausibly become hosts
overlooked [89]. Immunosuppression should be consid- of antibiotic-resistant pathogens, increasing the treat-
ered before delivering live or live-attenuated vaccines ment difficulty. Higher rates of infection in patients
(including rotavirus vaccine, varicella zoster vaccine, mea- with CKD result in more frequent use of antibiotics [90]
sles, mumps, and rubella vaccine, and the live-attenuated and more frequent hospitalization [2, 5, 9], which
influenza vaccine). For children with CKD, KDIGO guide- increases their exposure to microbes including multid-
lines recommend pediatric immunization schedules rug resistant organisms (MDROs) [91]. A recent study
should be followed according to official international and from China indicated the odds of infections by MDROs
regional recommendations [1]. Routine childhood vaccina- was 19% and 41% higher in those with eGFR between 30
tion schedules vary by country and are frequently updated, and 60 ml/min/1.73m2 and eGFR < 30 ml/min/1.73 m2,
and so up-to-date guidance should be consulted. respectively, compared to those with eGFR 60–104 ml/
min/1.73m2 [92].
Live Vaccines It is important that treatment does not cause pathogen
Live or live-attenuated vaccines include rotavirus vaccine, drug resistance and therefore treatment should be rational-
varicella, and herpes zoster vaccines, measles, mumps, and ized whenever possible; this means, for example, that self-
rubella vaccine, yellow fever, and the live-attenuated influ- limiting infections should not be treated with antibiotics.
enza vaccine. Patients with CKD may be receiving immu- For patients with CKD, vigilance is required and they need
nosuppressive medication or have underlying or comorbid to be told at which point they should come back to seek
conditions that result in immunosuppression. KDIGO further help and care should their condition worsen, so
guidelines recommend consideration of live vaccine should there is no delay with treatment should it become neces-
include an appreciation of the patient’s immune status and sary. The way that patients initially presented and what
follow recommendations from official or governmental they were told should be diligently documented so any
bodies [1]. In the United States, the Centers for Disease other healthcare provider will be able to assess whether
Control and Prevention advises that the majority of people patients have improved or worsened.
Preventing and Treating Infections among People with CK 537
When it comes to actual prescribing of antimicrobial will ensure that a sufficient dose of antibiotic is given to
drugs when the infection is deemed to require treatment, it treat the infection effectively, without delay. The dose and
will depend on the health system and the local circum- interval should then be reviewed once more current test
stances as to which drug is chosen. For example, many results are available. Estimation should be using the best
health systems have ongoing monitoring systems of local validated eGFR formula for a given patient, taking into
antibiotic resistance of key pathogens, with resulting anti- account their body surface area (BSA). The standard
biotic treatment handbooks that recommend a particular eGFR is for a BSA of 1.73m2, hence, for the individual the
first-line treatment when a person presents with a particu- GFR is:
lar infection. Antimicrobial stewardship means that when-
GFR eGFR * BSA / 1.73m 2
ever possible there should be cultures taken so that
treatment can be narrowed down once the pathogen has There are further formulas that can be used to adapt the
been identified, to enhance treatment success while drug dose as AKI improves or worsens [93]. Ideally, doses
minimizing further development of antibiotic resistance. should be adapted following measurement of drug levels of
Depending on the health system, some pathogens may nephrotoxic drugs to minimize toxicity and to maximize
need to be notified to central authorities, and infections therapeutic effects, especially if there is prolonged treat-
and treatments given will be monitored. For example, in ment, as, for example, for tuberculosis [94]. At the time of
many hospitals in the UK methicillin-resistant writing, several recent reviews have discussed these issues
staphylococcal infections are part of quality monitoring, in more detail and provide dose adjustment tables for a
with internal targets. If targets are not met, the hospital’s range of antimicrobials [95–97], but this is an area of ongo-
infection control staff will retrain the unit and review ing research and guidance is evolving rapidly.
infection control policies (e.g. decolonization regimes, Many patients with CKD or AKI will not be managed by
swabbing/culturing every patient, etc.) to prevent such nephrologists, and therefore it is important to consider the
infections occurring. Monitoring of infections, treatments surrounding health system to understand how treatment
given, and outcomes are simpler if electronic health records and infection outcomes in CKD can be improved.
are used, as this allows standardized centralized analysis. Unsurprisingly, dosing errors and inappropriate prescrib-
These are all general points of infection control policies ing for CKD stage are common in routine practice.
that have general value irrespective of CKD status, but the Electronic prescribing software has the potential to reduce
stakes of patient management and antimicrobial such errors by identifying CKD through diagnosis or labo-
stewardship are raised by the higher rate of infection with ratory results, suggesting point of care testing, recom-
MDROs among patients with CKD. mending preferred antimicrobials and dose, and alerting
prescribers to drug interactions or kidney dysfunction in
Prescribing Antibiotics in CKD the presence of a renally cleared drug. This has been rolled
If a patient with CKD requires antibiotics with a drug that out in settings from hospitals to community pharmacies
is partially or fully renally cleared, dosing must take the with varying success [98, 99]. Safely implementing
patient’s impaired kidney function into consideration. computer-decision support for medication is not straight-
How this is approached depends on how the drug acts on forward, and outcomes should be monitored. Suboptimal
the infection (i.e. pharmacodynamics). For drugs that act design, alert fatigue, workarounds, and lack of training can
immediately on the pathogen with an effect dependent on result in unintended consequences, while success depends
concentration (e.g. aminoglycosides), the dosing interval on prescribers finding the alerts valuable and acting on the
should be increased. For drugs that require a constant level information received [100].
to be maintained over a prolonged time to eliminate the
pathogen (e.g. beta-lactams), patients should be given Other Considerations
smaller doses in the normally recommended dosing inter- Aside from immediately treating the infection, there also
val. For some infections, an antimicrobial which is not should be consideration given to other medications that a
renally cleared may be an alternative. patient with CKD is prescribed. This may include checking
In practice, it can be challenging to adjust interval or for drug interactions (taking into account the kidney function
dose correctly, as this requires an estimate of the underly- of the patient). For patients receiving long-term immunosup-
ing kidney function. Especially in the context of AKI, it is pressant medications, more serious or atypical infections
very difficult to know what the true kidney function is. may need to be considered as part of a differential diagnosis,
Estimating kidney function using the most recent test and clinicians may need to consider whether there is a need
results before onset of infection will tend to overestimate for preventative antibiotics or antiviral medications (dis-
kidney function during the infection. This is not ideal, but cussed in other chapters, e.g. for transplantation).
C
onclusions risks, to know when to seek help when they are unwell,
and to receive timely and appropriate treatment of their
CKD is a clear risk factor for developing infections, often infections. Whenever possible, patients with CKD should
requiring hospitalization. Hospitalization with an infec- be vaccinated against infections to reduce future risks.
tion may worsen kidney function, and there is a graded Future trials of vaccination for infection should consider
increased risk of adverse outcomes including mortality. To tailored vaccination programs that take into account the
prevent poor outcomes, patients need to be aware of these observed reduced immune response in patients with CKD.
R
eferences
1 Kidney Disease: Improving Global Outcomes (KDIGO) infections among the oldest old in the general population.
CKD Work Group (2013). KDIGO 2012 clinical practice A population-based prospective follow-up study. BMC
guideline for the evaluation and management of chronic Med. 9: 57.
kidney disease. Kidney Int. (Suppl. 3): 1–150. 12 Dalrymple, L.S. and Go, A.S. (2008). Epidemiology of
2 McDonald, H.I., Thomas, S.L., and Nitsch, D. (2014). acute infections among patients with chronic kidney
Chronic kidney disease as a risk factor for acute disease. Clin. J. Am. Soc. Nephrol. 3 (5): 1487–1493.
community-acquired infections in high-income countries: 13 Naqvi, S.B. and Collins, A.J. (2006). Infectious
a systematic review. BMJ Open 4 (4): e004100. complications in chronic kidney disease. Adv. Chronic
3 Ishigami, J., Grams, M.E., Chang, A.R. et al. (2017). CKD Kidney Dis. 13 (3): 199–204.
and risk for hospitalization with infection: the 14 Foley, R.N. (2006). Infections and cardiovascular disease
atherosclerosis risk in communities (ARIC) study. Am. J. in patients with chronic kidney disease. Adv. Chronic
Kidney Dis. 69 (6): 752–761. Kidney Dis. 13 (3): 205–208.
4 Xu, H., Gasparini, A., Ishigami, J. et al. (2017). eGFR and 15 Coresh, J. (2009). CKD prognosis: beyond the traditional
the risk of community-acquired infections. Clin. J. Am. outcomes. Am. J. Kidney Dis. 54 (1): 1–3.
Soc. Nephrol. 12 (9): 1399–1408. 16 Foley, R.N. (2007). Infections in patients with chronic
5 McDonald, H.I., Thomas, S.L., Millett, E.R., and Nitsch, kidney disease. Infect. Dis. Clin. N. Am. 21 (3):
D. (2015). CKD and the risk of acute, community- 659–672. viii.
acquired infections among older people with diabetes 17 Castle, S.C. (2000). Clinical relevance of age-related
mellitus: a retrospective cohort study using electronic immune dysfunction. Clin. Infect. Dis. 31 (2): 578–585.
health records. Am. J. Kidney Dis. 66 (1): 60–68. 18 Shah, B.R. and Hux, J.E. (2003). Quantifying the risk of
6 Vart, P., Bettencourt-Silva, J.H., Metcalf, A.K. et al. (2018). infectious diseases for people with diabetes. Diabetes Care
Low estimated glomerular filtration rate and pneumonia in 26 (2): 510–513.
stroke patients: findings from a prospective stroke registry 19 Kornum, J.B., Thomsen, R.W., Riis, A. et al. (2008).
in the East of England. Clin. Epidemiol. 10: 887–896. Diabetes, glycemic control, and risk of hospitalization
7 Iwagami, M., Caplin, B., Smeeth, L. et al. (2018). Chronic with pneumonia: a population-based case-control study.
kidney disease and cause-specific hospitalisation: a Diabetes Care 31 (8): 1541–1545.
matched cohort study using primary and secondary care 20 Yin, K. and Agrawal, D.K. (2014). Vitamin D and
patient data. Br. J. Gen. Pract. 68 (673): e512–e523. inflammatory diseases. J. Inflamm. Res. 7: 69–87.
8 James, M.T., Laupland, K.B., Tonelli, M. et al. (2008). 21 Martineau, A.R., Jolliffe, D.A., Hooper, R.L. et al. (2017).
Risk of bloodstream infection in patients with chronic Vitamin D supplementation to prevent acute respiratory
kidney disease not treated with dialysis. Arch. Intern. tract infections: systematic review and meta-analysis of
Med. 168 (21): 2333–2339. individual participant data. BMJ 356: i6583.
9 James, M.T., Quan, H., Tonelli, M. et al. (2009). CKD and 22 Babitt, J.L. and Lin, H.Y. (2012). Mechanisms of anemia
risk of hospitalization and death with pneumonia. Am. J. in CKD. J. Am. Soc. Nephrol. 23 (10): 1631–1634.
Kidney Dis. 54 (1): 24–32. 23 Litton, E., Xiao, J., and Ho, K.M. (2013). Safety and efficacy
10 Vinogradova, Y., Hippisley-Cox, J., and Coupland, C. of intravenous iron therapy in reducing requirement for
(2009). Identification of new risk factors for pneumonia: allogeneic blood transfusion: systematic review and
population-based case-control study. Br. J. Gen. Pract. 59 meta-analysis of randomised clinical trials. BMJ 347: f4822.
(567): e329–e338. 24 Kausz, A.T. and Gilbertson, D.T. (2006). Overview of
11 Caljouw, M.A., den Elzen, W.P., Cools, H.J., and vaccination in chronic kidney disease. Adv. Chronic
Gussekloo, J. (2011). Predictive factors of urinary tract Kidney Dis. 13 (3): 209–214.
Reference 539
5 Janus, N., Vacher, L.V., Karie, S. et al. (2008). Vaccination

2 40 Igari, H., Imasawa, T., Noguchi, N. et al. (2015). Advanced
and chronic kidney disease. Nephrol. Dial. Transplant. 23 stage of chronic kidney disease is risk of poor treatment
(3): 800–807. outcome for smear-positive pulmonary tuberculosis. J.
26 Cohen, G. and Horl, W.H. (2012). Immune dysfunction in Infect. Chemother. 21 (8): 559–563.
uremia: an update. Toxins 4 (11): 962–990. 41 Mansur, A., Mulwande, E., Steinau, M. et al. (2015).
27 Eleftheriadis, T., Antoniadi, G., Liakopoulos, V. et al. Chronic kidney disease is associated with a higher 90-day
(2007). Disturbances of acquired immunity in mortality than other chronic medical conditions in
hemodialysis patients. Semin. Dial. 20 (5): 440–451. patients with sepsis. Sci. Rep. 5: 10539.
28 Bosco, A.M., Pereira, P.P., Almeida, B.F. et al. (2016). Free 42 McDonald, H.I., Nitsch, D., Millett, E.R.C. et al. (2015). Are
p-cresol alters neutrophil function in dogs. Artif. Organs pre-existing markers of chronic kidney disease associated
40 (5): 480–488. with short-term mortality following acute community-
29 Cohen, G., Raupachova, J., and Horl, W.H. (2013). The acquired pneumonia and sepsis? A cohort study among
uraemic toxin phenylacetic acid contributes to older people with diabetes using electronic health records.
inflammation by priming polymorphonuclear leucocytes. Nephrol. Dial. Transplant. 30 (6): 1002–1009.
Nephrol. Dial. Transplant. 28 (2): 421–429. 43 Neyra, J.A., Li, X., Canepa-Escaro, F. et al. (2016).
30 Thompson, S., James, M., Wiebe, N. et al. (2015). Cause Cumulative fluid balance and mortality in septic patients
of death in patients with reduced kidney function. J. Am. with or without acute kidney injury and chronic kidney
Soc. Nephrol. 26 (10): 2504–2511. disease. Crit. Care Med. 44 (10): 1891–1900.
31 Navaneethan, S.D., Schold, J.D., Arrigain, S. et al. (2015). 44 Thongprayoon, C., Cheungpasitporn, W.,
Cause-specific deaths in non-dialysis-dependent CKD. J. Phatharacharukul, P. et al. (2015). Chronic kidney disease
Am. Soc. Nephrol. 26 (10): 2512–2520. and end-stage renal disease are risk factors for poor
32 Wang, H.E., Gamboa, C., Warnock, D.G., and Muntner, P. outcomes of Clostridium difficile infection: a systematic
(2011). Chronic kidney disease and risk of death from review and meta-analysis. Int. J. Clin. Pract. 69 (9):
infection. Am. J. Nephrol. 34 (4): 330–336. 998–1006.
33 Fried, L.F., Katz, R., Sarnak, M.J. et al. (2005). Kidney 45 Cheikh Hassan, H.I., Tang, M., Djurdjev, O. et al. (2016).
function as a predictor of noncardiovascular mortality. J. Infection in advanced chronic kidney disease leads to
Am. Soc. Nephrol. 16 (12): 3728–3735. increased risk of cardiovascular events, end-stage kidney
34 Chan, T.C., Yap, D.Y., Shea, Y.F. et al. (2012). Chronic disease and mortality. Kidney Int. 90 (4): 897–904.
kidney disease and its association with mortality and 46 Bagshaw, S.M., Uchino, S., Bellomo, R. et al. (2007).
hospitalization in Chinese nursing home older residents: Septic acute kidney injury in critically ill patients: clinical
a 3-year prospective cohort study. J. Am. Med. Dir. Assoc. characteristics and outcomes. Clin. J. Am. Soc. Nephrol. 2
13 (9): 782–787. (3): 431–439.
35 Charytan, D.M., Lewis, E.F., Desai, A.S. et al. (2015). 47 Evans, R.D., Hemmila, U., Craik, A. et al. (2017).
Cause of death in patients with diabetic CKD Incidence, aetiology and outcome of community-acquired
enrolled in the trial to reduce cardiovascular events with acute kidney injury in medical admissions in Malawi.
Aranesp therapy (TREAT). Am. J. Kidney Dis. 66 (3): BMC Nephrol. 18 (1): 21.
429–440. 48 Oppert, M., Engel, C., Brunkhorst, F.M. et al. (2008).
36 Su, G., Xu, H., Marrone, G. et al. (2017). Chronic kidney Acute renal failure in patients with severe sepsis and
disease is associated with poorer in-hospital outcomes in septic shock – a significant independent risk factor for
patients hospitalized with infections: electronic record mortality: results from the German prevalence study.
analysis from China. Sci. Rep. 7 (1): 11530. Nephrol. Dial. Transplant. 23 (3): 904–909.
37 Su, G., Xu, Y., Xu, X. et al. (2018). Association between 49 Demirjian, S.G., Raina, R., Bhimraj, A. et al. (2011).
reduced renal function and cardiovascular mortality in 2009 Influenza a infection and acute kidney injury:
patients hospitalized with infection: a multi-center cohort incidence, risk factors, and complications. Am. J. Nephrol.
study. Eur. J. Intern. Med. 57: 32–38. 34 (1): 1–8.
38 Viasus, D., Garcia-Vidal, C., Cruzado, J.M. et al. (2011). 50 Pettila, V., Webb, S.A., Bailey, M. et al. (2011). Acute
Epidemiology, clinical features and outcomes of kidney injury in patients with influenza a (H1N1) 2009.
pneumonia in patients with chronic kidney disease. Intensive Care Med. 37 (5): 763–767.
Nephrol. Dial. Transplant. 26 (9): 2899–2906. 51 Yun, S.E., Jeon, D.H., Kim, M.J. et al. (2015). The
39 Maizel, J., Deransy, R., Dehedin, B. et al. (2013). Impact incidence, risk factors, and outcomes of acute kidney
of non-dialysis chronic kidney disease on survival in injury in patients with pyogenic liver abscesses. Clin. Exp.
patients with septic shock. BMC Nephrol. 14: 77. Nephrol. 19 (3): 458–464.
2 Leedahl, D.D., Frazee, E.N., Schramm, G.E. et al. (2014).

5 65 Remschmidt, C., Wichmann, O., and Harder, T. (2014).
Derivation of urine output thresholds that identify a very Influenza vaccination in patients with end-stage renal
high risk of AKI in patients with septic shock. Clin. J. Am. disease: systematic review and assessment of quality of
Soc. Nephrol. 9 (7): 1168–1174. evidence related to vaccine efficacy, effectiveness, and
53 Huang, S., Lin, C., Chang, Y., Sher, Y., Wu, M., Shu, K., safety. BMC Med. 12: 244.
Sung, F., Kao, C. Pneumococcal pneumonia infection is 66 Gilbertson, D.T., Unruh, M., McBean, A.M. et al. (2003).
associated with end-stage renal disease in adult Influenza vaccine delivery and effectiveness in end-stage
hospitalized patients. Kidney Int. 2014; 86(5):1023–30. renal disease. Kidney Int. 63 (2): 738–743.
54 Smeeth, L., Cook, C., Thomas, S. et al. (2006). Risk of 67 McGrath, L.J., Cole, S.R., Kshirsagar, A.V. et al. (2013).
deep vein thrombosis and pulmonary embolism after Hospitalization and skilled nursing care are predictors of
acute infection in a community setting. Lancet 367 influenza vaccination among patients on hemodialysis:
(9516): 1075–1079. evidence of confounding by frailty. Med. Care 51 (12):
55 Smeeth, L., Thomas, S.L., Hall, A.J. et al. (2004). 1106–1113.
Risk of myocardial infarction and stroke after acute 68 McGrath, L., Kshirsagar, A.V., Cole, S.R. et al. (2012).
infection or vaccination. N. Engl. J. Med. 351 (25): Influenza vaccine effectiveness in patients on
2611–2618. hemodialysis. Arch. Intern. Med. 172 (7): 548–554.
56 Hotchkiss, R.S. and Karl, I.E. (2003). The 69 Chi, C., Patel, P., Pilishvili, T. et al. (2012). Guidelines for
pathophysiology and treatment of sepsis. N. Engl. J. Med. vaccinating kidney dialysis patients and patients with
348 (2): 138–150. chronic kidney disease. Atlanta USA Centers for Disease
57 Schrier, R.W. and Wang, W. (2004). Acute renal failure Control and Prevention.
and sepsis. N. Engl. J. Med. 351 (2): 159–169. 70 Schroth, R.J., Hitchon, C.A., Uhanova, J. et al. (2004).
58 Public Health England (2014). Immunisation against Hepatitis B vaccination for patients with chronic renal
infectious disease: the Green Book July 2013 https:// failure. Cochrane Database Syst. Rev. 3 (Art. No.
www.gov.uk/government/collections/immunisation- CD003775).
against-infectious-disease-the-green-book (accesss 71 Ghadiani, M.H., Besharati, S., Mousavinasab, N., and
24 May 2020). Jalalzadeh, M. (2012). Response rates to HB vaccine in
59 Mathew, R., Mason, D., and Kennedy, J.S. (2014). CKD stages 3–4 and hemodialysis patients. J. Res. Med. Sci.
Vaccination issues in patients with chronic kidney 17 (6): 527–533. PMID: 23626628; PMCID: PMC3634289.
disease. Expert Rev. Vaccines 13 (2): 285–298. 72 DaRoza, G., Loewen, A., Djurdjev, O. et al. (2003). Stage
60 Sarnak, M.J. and Jaber, B.L. (2001). Pulmonary infectious of chronic kidney disease predicts seroconversion after
mortality among patients with end-stage renal disease. hepatitis B immunization: earlier is better. Am. J. Kidney
Chest 120 (6): 1883–1887. Dis. 42 (6): 1184–1192.
61 Moberley, S., Holden, J., Tatham, D.P., and Andrews, 73 Dukes, C.S., Street, A.C., Starling, J.F., and Hamilton, J.D.
R.M. (2013). Vaccines for preventing pneumococcal (1993). Hepatitis B vaccination and booster in predialysis
infection in adults. Cochrane Database Syst. Rev. 1 (Art. patients: a 4-year analysis. Vaccine 11 (12): 1229–1232.
No. CD000422). https://doi.org/10.1002/14651858. 74 Sawyer, M.H., Hoerger, T.J., Murphy, T.V. et al. (2011).
CD000422.pub3. Centers for Disease Control and Prevention. Use of
62 Ochoa-Gondar, O., Vila-Corcoles, A., Rodriguez-Blanco, hepatitis B vaccination for adults with diabetes mellitus:
T. et al. (2014). Effectiveness of the 23-valent recommendations of the Advisory Committee on
pneumococcal polysaccharide vaccine against Immunization Practices (ACIP). MMWR Morb. Mortal.
community-acquired pneumonia in the general Wkly Rep. 60 (50): 1709–1711.
population aged >/= 60 years: 3 years of follow-up in the 75 Seaworth, B., Drucker, J., Starling, J. et al. (1988).
CAPAMIS study. Clin. Infect. Dis. 58 (7): 909–917. Hepatitis B vaccines in patients with chronic renal failure
63 McDonald, H.I., Thomas, S.L., Millett, E.R.C. et al. before dialysis. J. Infect. Dis. 157 (2): 332–337.
(2017). Do influenza and pneumococcal vaccines prevent 76 Jungers, P., Chauveau, P., Courouce, A.M. et al. (1994).
community-acquired respiratory infections among older Immunogenicity of the recombinant GenHevac B Pasteur
people with diabetes and does this vary by chronic kidney vaccine against hepatitis B in chronic uremic patients. J.
disease? A cohort study using electronic health records. Infect. Dis. 169 (2): 399–402.
BMJ Open Diabetes Res. Care 5 (1): e000332. 77 Fabrizi, F., Dixit, V., Messa, P., and Martin, P. (2011).
64 Joseph, C., Togawa, Y., and Shindo, N. (2013). Bacterial Intradermal vs intramuscular vaccine against hepatitis B
and viral infections associated with influenza. Influenza infection in dialysis patients: a meta-analysis of
Other Respir. Viruses 7 (Suppl 2): 105–113. randomized trials. J. Viral Hepat. 18 (10): 730–737.
Reference 541
78 Mulley, W.R., Le, S.T., and Ives, K.E. (2017). Primary 89 Neu, A.M. (2012). Immunizations in children with
seroresponses to double-dose compared with standard- chronic kidney disease. Pediatr. Nephrol. 27 (8):
dose hepatitis B vaccination in patients with chronic 1257–1263.
kidney disease: a systematic review and meta-analysis. 90 The lancet infectious D. (2013). Antibiotic resistance:
Nephrol. Dial. Transplant. 32 (1): 136–143. long-term solutions require action now. Lancet Infect.
79 Tong, N.K., Beran, J., Kee, S.A. et al. (2005). Dis. 13 (12): 995. https://doi.org/10.1016/
Immunogenicity and safety of an adjuvanted hepatitis B S1473-3099(13)70290-1.
vaccine in pre-hemodialysis and hemodialysis patients. 91 Calfee, D.P. (2015). Multidrug-resistant organisms
Kidney Int. 68 (5): 2298–2303. within the dialysis population: a potentially preventable
80 Fabrizi, F., Dixit, V., Messa, P., and Martin, P. (2010). perfect storm. Am. J. Kidney Dis. 65 (1): 3–5.
Meta-analysis: levamisole improves the immune response 92 Su, G., Xu, H., Riggi, E. et al. (2018). Association of
to hepatitis B vaccine in dialysis patients. Aliment. Kidney Function with infections by multidrug-resistant
Pharmacol. Ther. 32 (6): 756–762. organisms: an electronic medical record analysis. Sci.
81 Janssen, J.M., Heyward, W.L., Martin, J.T., and Janssen, Rep. 8 (1): 13372.
R.S. (2015). Immunogenicity and safety of an 93 Chen, S. (2013). Retooling the creatinine clearance
investigational hepatitis B vaccine with a toll-like receptor equation to estimate kinetic GFR when the plasma
9 agonist adjuvant (HBsAg-1018) compared with a creatinine is changing acutely. J. Am. Soc. Nephrol.
licensed hepatitis B vaccine in patients with chronic 24 (6): 877–888.
kidney disease and type 2 diabetes mellitus. Vaccine 33 94 British Thoracic Society Standards of Care C, Joint
(7): 833–837. Tuberculosis C, Milburn, H. et al. (2010). Guidelines for
82 Langan, S.M., Thomas, S.L., Smeeth, L. et al. (2016). the prevention and management of Mycobacterium
Zoster vaccination is associated with a reduction of zoster tuberculosis infection and disease in adult patients with
in elderly patients with chronic kidney disease. Nephrol. chronic kidney disease. Thorax 65 (6): 557–570.
Dial. Transplant. 31 (12): 2095–2098. 95 Keller, F., Schroppel, B., and Ludwig, U. (2015).
83 Lin, S.Y., Liu, J.H., Yeh, H.C. et al. (2014). Association Pharmacokinetic and pharmacodynamic
between herpes zoster and end stage renal disease considerations of antimicrobial drug therapy in cancer
entrance in chronic kidney disease patients: a population- patients with kidney dysfunction. World J. Nephrol. 4
based cohort study. Eur. J. Clin. Microbiol. Infect. Dis. 33 (3): 330–344.
(10): 1809–1815. 96 Eyler, R.F. and Mueller, B.A. (2010). Antibiotic
84 Wu, M.Y., Hsu, Y.H., Su, C.L. et al. (2012). Risk of herpes pharmacokinetic and pharmacodynamic considerations
zoster in CKD: a matched-cohort study based on in patients with kidney disease. Adv. Chronic Kidney Dis.
administrative data. Am. J. Kidney Dis. 60 (4): 548–552. 17 (5): 392–403.
85 Langan, S.M., Smeeth, L., Margolis, D.J., and Thomas, S.L. 97 Munar, M.Y. and Singh, H. (2007). Drug dosing
(2013). Herpes zoster vaccine effectiveness against incident adjustments in patients with chronic kidney disease.
herpes zoster and post-herpetic neuralgia in an older US Am. Fam. Physician 75 (10): 1487–1496.
population: a cohort study. PLoS Med. 10 (4): e1001420. 98 Farag, A., Garg, A.X., Li, L., and Jain, A.K. (2014).
86 Webb, N.J., Fitzpatrick, M.M., Hughes, D.A. et al. (2000). Dosing errors in prescribed antibiotics for older persons
Immunisation against varicella in end stage and pre-end with CKD: a retrospective time series analysis. Am. J.
stage renal failure. Trans-Pennine Paediatric Nephrology Kidney Dis. 63 (3): 422–428.
Study Group. Arch. Dis. Child. 82 (2): 141–143. 99 Heringa, M., Floor-Schreudering, A., De Smet, P., and
87 Furth, S.L., Arbus, G.S., Hogg, R. et al. (2003). Varicella Bouvy, M.L. (2017). Clinical decision support and
vaccination in children with nephrotic syndrome: a optional point of care testing of renal function for safe
report of the Southwest Pediatric Nephrology Study use of antibiotics in elderly patients: a retrospective
Group. J. Pediatr. 142 (2): 145–148. study in community pharmacy practice. Drugs Aging 34
88 Furth, S.L., Hogg, R.J., Tarver, J. et al. (2003). Varicella (11): 851–858.
vaccination in children with chronic renal failure. A 100 Tolley, C.L., Slight, S.P., Husband, A.K. et al. (2018).
report of the Southwest Pediatric Nephrology Study Improving medication-related clinical decision support.
Group. Pediatr. Nephrol. 18 (1): 33–38. Am. J. Health Syst. Pharm. 75 (4): 239–246.
542
33
Treatment of Anemia in Chronic Kidney Disease

Patrick S. Parfrey
Faculty of Medicine, Memorial University, St. John’s, NL, Canada
I ntroduction In the early decades of dialysis therapy blood transfu-

sions were the mainstay of anemia therapy, but in the last
The prevalence of anemia increases as chronic kidney three decades large clinical trials have been undertaken to
disease (CKD) progresses by stage. Anemia is caused by determine the efficacy of erythropoiesis-stimulating agents
erythyropoietin deficiency, disordered iron metabolism, (ESAs) [8–12], and more recently peginesatide [13–14] and
blood loss, infection, inflammation, and other less frequent intravenous iron [15]. Currently trials of novel prolyl
causes. hydroxylase inhibitors (PHI) that stabilize hypoxia induci-
Patients with CKD and moderate anemia maintain this ble factor (HIF) are ongoing. Table 33.1 provides recom-
Hb range for years: most patients with moderate anemia mendations on the treatment of anemia in chronic kidney
(hemoglobin [Hb] level < 11 g/dl and > = 9g/dl), nondialysis- disease based on the certainly and strength of the
dependent CKD, and type 2 diabetes are able to maintain a evidence.
Hb level >9 g/dl (i.e. avoid severe anemia) when followed for
an average of 2.3 years [1].
In the Trial to Reduce cardiovascular Events with Aranesp linical Outcomes in Randomized
C
Therapy (TREAT), Hb declined precipitously in the year Controlled Trials
prior to the development of end-stage renal disease (ESRD)
(irrespective of whether assigned to Aranesp or placebo) The choice of interventions to treat or prevent anemia in
without biochemical evidence of iron deficiency [2]. Once CKD should be influenced by the results of randomized
ESRD occurs, severe anemia is more common and the need controlled trials (RCTs). The clinical outcomes studied in
for blood transfusions far higher than in nondialysis CKD [3]. RCTs include change in Hb levels, blood transfusion rates,
Nondialysis CKD and dialysis-dependent ESRD should CV events, and quality of life. In addition, adverse events
probably be studied as two separate phases of kidney need to be considered.
disease because of the following: (i) different trajectories
for Hb levels, (ii) different hemodynamic and metabolic
Change in Hb Levels
milieus, and (iii) changing pathogenesis of cardiac disease
from predominantly atherosclerotic events to an equal This is easy to assess, but creates problems because neph-
prominence of cardiac failure events linked to nonathero- rologists tend to focus on the levels at which therapy should
sclerotic processes [4]. begin and target levels for therapy. It would be preferable to
In cohort studies, anemia has been associated with the individualize therapy by determining whether patients
development of left ventricular hypertrophy (LVH), heart have symptoms attributable to anemia and provide therapy
failure, cardiovascular (CV) hospitalizations, mortality, to treat symptoms, rather than treat everybody based on
and diminished quality of life [5–7]. test values.
Table 33.1 Treatment of anemia in chronic kidney disease.
Strength of
Certainty of evidence: recommendation:
Treatment Recommendation high, weak, low, very low strong or weak
Blood transfusions Avoid red cell transfusions to minimize the general risks related to their use (1B) Weak Strong
In patients eligible for organ transplantation avoid when possible red cell transfusions to minimize the risk Low Strong
of allosensitization (1C)
The benefits of red cell transfusions may outweigh the risk in patients in whom ESAs are ineffective and in Low Weak
patients at risk from ESA therapy (previous or current malignancy or previous stroke) (2C)
The decision to transfuse a CKD patient with nonacute anemia should not be based on any arbitrary Hb Low Weak
threshold, but should be determined by the occurrence of symptoms caused by anemia (2C)
ESAs (A) In initiating and maintaining ESA therapy balance the potential benefits of reducing blood transfusions Weak Strong
and anemia-related symptoms against the harm in individual patients (e.g. stroke, vascular access loss,
hypertension) (1B).
(B) In CKD ND start ESA therapy when Hb falls below 9 g/l. BUT individualize the decision to start ESA Low Weak
therapy based on the rate of decrease of Hb, prior response to iron therapy, the risk of needing a blood
transfusion, the risks of ESA therapy, and the presence of symptoms attributable to anemia (2C).
(C) In dialysis patients start ESA before Hb falls below 9 g/l by starting ESA therapy when the Hb is between Low Weak
9 and 10 g/l (2B).
(D) In CKD ND and D target Hb should be 10–11.5 g/l (2C). Low Weak
(E) ESAs should not be used to intentionally raise Hb above 13 g/l (1A). High Strong
(F) Avoid ESAs in patients with active malignancy, history of solid organ malignancy, or history of stroke Low Weak
(2C).
(G) In patients with ESA hypo-responsiveness avoid repeated escalations in ESA dose beyond double the Low Weak
weight-based dose (2D).
Intravenous iron (A) For anemic adult CKD patients not on iron or ESA therapy, in whom a raise in Hb is desired, and whose Low Weak
TSAT is <30% and ferritin is <500 ng/ml, give a trial of IV iron (2C).
(B) For adult dialysis patients on ESA therapy, in whom a raise in Hb or reduction dose of ESA is desired, Low Weak
and whose TSAT is <30% and ferritin is <500 ng/ml, give a trial of IV iron (2C).
(C) In adult dialysis patients maintain TSAT between 20% and 40% and ferritin between 200 and 700 ng/ml in Weak Strong
a proactive fashion (1B).
ESA: Erythropoiesis–stimulating agents; CKD: Chronic Kidney Disease; Hb: Hemoglobin; ND: Non Dialysis; D: Dialysis; TSAT: Transferrin saturation; IV: Intravenous.
c33.indd 543 09-12-2022 16:04:22

544 Treatment of Anemia in Chronic Kidney Disease
Blood Transfusions treatment was normalization of Hb [16]. As a result of

these trials, the target Hb is between 10 and 12 g/l. A critical
Avoidance of blood transfusions is sufficiently desirable
appraisal of these RCTs follows [17].
that it is currently the major reason for treatment of severe
anemia (Hb < 9 g/dl).
Treatment of Severe Anemia
Cardiovascular Events There are no large RCTs of ESA therapy in patients with
severe anemia (baseline Hb <9 g/dl) in which clinical end-
Events arising from atherosclerosis of coronary, cerebral,
points and safety were assessed. In 1990, the Canadian
and peripheral arteries occur frequently in CKD. As CKD
Erythropoietin Study Group reported an RCT in which ESA
progresses, LVH and cardiac failure events arising from
was utilized in 110 hemodialysis patients with Hb levels
cardiomyopathy occur more frequently [4]. The temptation
below 9 g/dl, and three randomly allocated groups were fol-
is to combine both of these types of CV events, but this
lowed (placebo, low target Hb 9.5–11 g/dl, high target Hb
should be resisted because they have different pathogeneses.
>11 g/dl). Baseline Hb was 7.0 g/dl and transfusion require-
Assessment of CV events requires thousands of patients in
ment was seven transfusions per year. After 8 weeks, 58%
RCTs. In addition, surrogate markers for CV events, such as
(n = 23/40) in the placebo group were transfused and only
echocardiographic measurement of LVH and left ventricular
3% (n = 1/40) in the low Hb group. After 6 months, clini-
(LV) diameter and function, are often undertaken, but they
cally important improvements in fatigue, physical function,
do not substitute for clinical outcomes when applying the
and 6-minute walking tests were reported for the low Hb
results of RCTs to clinical decision making.
group compared with the placebo group, but no improve-
ment was observed when comparing the low and high Hb
Quality of Life groups. A strategy to treat CKD patients with ESAs when
Quality of life has multiple domains and questionnaires Hb falls below 9 g/dl seems justified on the grounds of
exist to measure these domains, such as the SF 36 or Kidney reduction in transfusions and improvement in quality of
Disease Quality of Life (KDQoL). However, anemia life, despite the lack of data on safety.
treatment is unlikely to influence several of the domains.
Consequently, it is advisable to identify a priori prespecified Treatment of Moderate Anemia
outcomes potentially susceptible to anemia treatment,
There are several large RCTs of ESA therapy in which base-
such as tiredness, vitality, and fatigue. A major problem
line Hb was higher than 10 g/dl [8–12]. The intervention
concerns interpretation of results, especially as the
being tested in these trials was correction of anemia with
definition of clinically significant changes in symptom
ESAs, compared with partial correction of anemia with
scores is not based on strong clinical epidemiology science.
ESAs in four RCTs [8–11] and with placebo in one [12].
Harms vs. Benefits Normal Hematocrit Study

Large trials are necessary to provide evidence on the Prevalent hemodialysis patients with symptomatic cardiac
harms:benefits ratio. For example, TREAT was undertaken disease (congestive heart failure or ischemic heart disease)
nearly 20 years after the introduction of ESAs and was big (n = 1233) were enrolled in this RCT [8]. In the normal
enough to identify the increased risk of stroke in anemic hematocrit group treated with epoetin, there were 183
diabetic CKD patients treated to normalize Hb levels with deaths and 19 myocardial infarcts, producing 202 primary
Aranesp [12]. Although some recommend conclusions events, compared with 164 events (150 deaths and
derived from meta-analysis of all types of trials of a particu- 14 myocardial infarcts) in the group in which anemia was
lar agent, sometimes more utility is derived from large RCTs partially corrected with epoetin. Although the primary
designed to answer a clinical management question. event rate was significantly different (P = 0.03), the hazard
ratio (HR) was 1.3, with 95% confidence intervals (CIs)
being 0.9–1.9, following correction for previous interim
Erythropoiesis-stimulating Agents analyses. The trial was stopped early, not because of these
somewhat equivocal results, but because the intervention
Recombinant human erythropoietins have been used to being tested was certainly causing harm in a situation
treat anemia for up to three decades. During the past wherein the primary hypothesis was unlikely to be proven:
decade, ESA doses have declined following publication of 39% had vascular access clotting in the intervention arm
RCTs demonstrating harm when the target Hb for ESA and 29% in the control arm (P = 0.001).
Erythropoiesis-stimulating Agent 545
Canada–Europe Study with the low Hb group (P = 0.03). However, there were
Incident hemodialysis patients without symptomatic car- differences in baseline CV comorbidity such that, when
diac disease (congestive heart failure or ischemic heart dis- taken into account in multivariate models, the hazard
ease) were enrolled in this RCT [9]. A similar number of associated with high Hb was no longer statistically
patients (n = 596) were enrolled in this study as in the significant (P = 0.11) [14]. The conclusion that correction
Cardiovascular Risk reduction in Early Anemia Treatment of anemia with erythropoietin is not associated with
with Epoetin (CREATE) study (see next section) (n = 603), increased CV events is supported by the results of
but the primary outcome was not clinical events, but rather TREAT [12] in which CV outcomes were the same in the
change in LV volume index and LV mass index. The change placebo and high Hb groups.
in both indices was similar in the correction of anemia
group and in the partial correction group. Nonetheless, Trial to Reduce Cardiovascular Events with Aransep
important clinical observations were made. A safety signal Therapy
was observed as 12 neurological events occurred in the This trial in diabetic patients with moderate anemia is by
high Hb group (13.5–14.5 g/dl target) and four in the low far the biggest RCT of ESA therapy performed (n = 4038)
Hb group (9.5–11.5 g/dl target) (P < 0.05), a signal and had the best research design, as it was placebo-
confirmed by TREAT (see below) [8]. As one might expect, controlled and double-blind [12]. In the intervention arm
the high Hb group received significantly fewer transfusions darbepoetin (Aranesp) was used to achieve Hb higher than
than the low Hb group, but the extent of the benefit was 13 g/dl, and in the placebo arm rescue therapy with
modest: although 9% in the high Hb arm received a darbepoetin was undertaken when Hb fell below 9 g/dl and
transfusion compared with 19% in the low Hb arm placebo was restarted when Hb was higher than 9 g/dl.
(P = 0.004) during the 2-year study, the number of Baseline eGFR was 34 ml/min and baseline Hb was 10.4 g/
transfusions per patient per year was 0.3 in the high Hb dl. Median follow-up was 29 months, equivalent to 9941
arm and 0.7 in the low Hb arm (P < 0.0001) [18]. patient-years. Median Hb in the high Hb arm was 12.6 g/dl
and in the low Hb arm it was 10.6 g/dl. There were no
Cardiovascular Risk Reduction in Early Anemia differences in CV or renal outcomes comparing the two
Treatment with Epoetin study groups. The HR for death/composite CV events was 1.05
Predialysis CKD patients (n = 603) with mean estimated (95% CI 0.94–1.17) and for death or ESRD it was 1.06 (95%
glomerular filtration rate (eGFR) equal to 24 ml/min were CI 0.95–1.19). However, there was a substantial increased
enrolled in this RCT [10]. Baseline Hb was 11.0–12.5 g/dl. risk of stroke (HR 1.92, 95% CI 1.38–2.68), although the
Normalization of Hb with epoetin was compared with late risk of stroke attributable to the high Hb intervention was
partial treatment of Hb (i.e. when Hb fell below 10.5 g/dl, relatively small: 5.0% of the high Hb group had a stroke
to a target of 10.5–11.5 g/dl). The trial was underpowered compared with 2.6% in the placebo group. The attributable
for the primary outcome, a composite of eight CV events. risk of stroke in those with a prior history of stroke was
Dialysis was required in significantly more patients in the substantially higher. Venous thromboembolic events
high Hb group than in the low Hb group, but this may have occurred significantly more frequently in the high Hb arm
resulted from the open-label design of the study in which (2%) compared with 1.1% in the placebo arm (P = 0.02). A
symptoms might have been attributed to uremia with the signal that normalization of Hb with darbepoetin may be
knowledge that they were not attributable to anemia. This harmful in patients with a history of malignancy was
interpretation is supported by the fact that the rate of fall of reported following a post hoc analysis: 14 of 188 (7.4%)
eGFR in the two groups during the 3-year study was similar patients with a baseline history of malignancy died from
and by the results of TREAT, in which renal outcomes were cancer in the darbepoetin arm compared with 1 of 160
the same in the placebo and high Hb groups [12]. (0.6%) in the placebo arm (P = 0.002). Red blood cell
transfusions were prescribed surprisingly frequently, and
Correction of Hemoglobin and Outcomes in Renal more often in the placebo arm (25%) compared with the
Insufficiency high Hb arm (15%).
This RCT was undertaken in 1432 predialysis CKD patients,
half of whom had diabetic renal disease; baseline creatinine
Hyporesponsiveness
clearance was 37 ml/min [11]. Withdrawal rate was high:
17% due to renal transplantation and 21% for other reasons. A report from TREAT assessed the initial Hb response to
The trial was stopped prematurely at a time when the darbepoetin (Aranesp) after two weight-based doses at
primary endpoint (death or composite CV event) occurred 2-weekly intervals in 1842 diabetic patients with CKD [19].
significantly more often in the high Hb group compared Patients with a poor response, the lowest quartile who had
less than 2% change in Hb level after 1 month, had higher peginesatide in nondialysis CKD was 328 and in dialysis
rates of the composite CV events (adjusted HR 1.31, 95% CI patients it was 1066. As a result of these RCTs, the FDA
1.09–1.59) or death (adjusted HR 1.41, 95% CI 1.12–1.78) approved peginesatide for use in dialysis patients, but given
compared with those with a better response. Although this the increased risk of CV events in nondialysis CKD patient
differential effect may be related to comorbidity in trials, the FDA did not approve the agent for use in that
hyporesponsive patients, nonetheless it is possible that group of patients.
high ESA doses in hyporesponsive patients may be toxic. On 23 February 2013, shortly after publication of the two
trials, Affymax and Takeda, the producers of peginesatide,
announced a nationwide voluntary recall of all lots of
Meta-analyses
peginesatide injection as a result of new post-marketing
Palmer et al. [20] concluded that higher Hb levels in CKD reports of serious hypersensitivity reactions, including life-
increase risk for stroke (relative risk [RR] 1.51, 95% CI 1.03– threatening or fatal anaphylaxis [27]. The rate of
2.21), hypertension, and vascular access thrombosis (RR hypersensitivity reactions reported was 0.2%, of which
1.33, 95% CI 1.16–1.53), and probably increase risk for death approximately one-third were serious in nature. A safety
(RR 1.09, 95% CI 0.99–1.2), serious CV events (RR 1.15, 95% signal for hypersensitivity reactions was not reported in the
CI 0.98–1.33), or ESRD (RR 1.08, 95% CI 0.97–1.20). two NEJM articles, which supports the belief that these
In 2014, a Cochrane review concluded that darbepoetin noninferiority studies were too small to provide adequate
alfa effectively reduced blood transfusions in adults with data for safety.
CKD stages 3–5, but had little or no effect on mortality [21].
Its effects in adults with CKD stage 5D, kidney transplant
recipients, and children with CKD remain uncertain. I ntravenous Iron
Another Cochrane review concluded there is currently
insufficient evidence to suggest the superiority of any ESA During the past 6 years, use of intravenous iron has
formulation based on available safety and efficacy data [22]. increased following the recommendation by Kidney
A third Cochrane review concluded that continuous Disease Improving Global Outcomes (KDIGO) guidelines
erythropoiesis receptor activator (a new class of third gen- that iron repletion should be a component of anemia
eration ESAs) had little or no effect on patient-centered management in CKD [28].
outcomes compared with placebo, epoetin alfa or beta, or Meta-analysis comparing intravenous versus oral iron
darbepoetin alfa for adults with CKD [23]. supplementation revealed that patients with CKD stages
A meta-analysis of the effect of ESAs on health-related 3–5 were more likely to reach an Hb response >1 g/dl with
quality of life (HRQOL) concluded that ESA treatment intravenous iron (risk ratio 1.61, 95% CI 1.39–1.87) and that
of anemia to obtain higher Hb targets does not result in dialysis patients were even more likely to achieve this
important differences (24), despite significant benefits being response (risk ratio 7.71, 95% CI 1.74–7.94) [29]. Another
reported from the Canadian-Europe study [25], CREATE [10], meta-analysis concluded that significant reductions in ESA
and TREAT [12]. However, there may be a misinterpreta- dosing may be achieved by iron repletion in the
tion of the importance of the results, particularly if viewed hemodialysis population, and suboptimal iron use may
through the lens of requiring a large improvement in symp- require higher ESA dosing to manage anemia [30].
tom scores to be “clinically significant.” In TREAT, linear Nonetheless there were concerns about the safety of
repeated measure models of fatigue incorporated all symp- increased use of intravenous iron.
tom scores through 97 weeks, together with several baseline Recently, in a multicenter, open-label trial with blinded
and interim predictors of long-term changes in HRQOL. end-point evaluation, adults undergoing maintenance
Allocation to the placebo group was similar in magnitude to hemodialysis were randomly assigned to receive either
the effects of having pulmonary disease or CV disease, sug- high-dose iron sucrose administered intravenously in a pro-
gesting that these “modest” changes and fatigue scores active fashion (400 mg monthly, unless the ferritin concen-
were actually clinically important [26]. tration was >700 μg/l or the transferrin saturation was
40%) or low-dose iron sucrose administered intravenously
in a reactive fashion (0–400 mg monthly, with a ferritin con-
P
eginesatide centration of <200 μg/l or a transferrin saturation of <20%
being a trigger for iron administration) [15]. The primary
Peginesatide is a dimeric pegylated peptide erythropoietic end point was the composite of nonfatal myocardial infarc-
agent which was evaluated in two RCTs published in tion, nonfatal stroke, hospitalization for heart failure, or
2013 [13, 14]. The number of patients allocated to death, assessed in a time-to-first-event analysis.
Reference 547
A total of 2141 patients underwent randomization. The ypoxia-sensitive genes for which HIF is a key regulatory
h
median follow-up was 2.1 years. Patients in the high-dose protein, including angiogenic factors. Consequently, long-
group received a median monthly iron dose of 264 mg (inter- term safety is a concern.
quartile range [25th to 75th percentile] 200–336), as com- Prolyl hydroxylase inhibition (PHI) agents are oral drugs,
pared with 145 mg (interquartile range 100–190) in the four of which are already in stage 3 RCTs. They have dem-
low-dose group. The median monthly dose of an ESA was onstrated that they can raise and maintain Hb levels in a
29 757 IU in the high-dose group and 38 805 IU in the low- predictable and controlled manner, while enhancing iron
dose group (median difference −7539 IU, 95% CI −9485 to mobilization in patients with nondialysis CKD [31–33].
−5582). A total of 333 patients (30.5%) in the high-dose group Roxadustat is being studied in four RCTs in nondialysis
had a primary end-point event, as compared with 343 (32.7%) CKD comprising 4750 patients, with one RCT comprising
in the low-dose group (HR 0.88, 95% CI 0.76–1.03, P < 0.001 2700 patients and having a primary outcome for major CV
for noninferiority). In an analysis that used recurrent-events events. The same drug is under investigation in 4508 dialy-
approach, there were 456 events in the high-dose group and sis patients, with one RCT comprising 2100 patients and
538 in the low-dose group (rate ratio 0.78, 95% CI 0.66–0.92). having a primary outcome for major CV events. Vadadustat
The infection rate was the same in the two groups. is under investigation in 3100 nondialysis CKD patients in
These data support an iron repletion high-dose regimen two RCTs and in 2600 dialysis patients in two RCTs.
administered proactively rather than a lower-dose reactive Daprodustat is being studied in an RCT of 4500 nondialysis
regimen. patients, in another of 3000 dialysis patients, and in a third
of 300 incident dialysis patients. The fourth drug is
Molidustat. The results from these trials will be published
Prolyl Hydroxylase Inhibitors during the next 3 years.
HIF is the major transcription factor for the erythropoietin

gene. Molecules were created that inhibit the HIF prolyl C
onclusion
hydroxylase enzyme, preventing degradation of HIF and
thus upregulating the erythropoietin gene. In adults, about Current management of anemia in CKD usually combines
90% of erythropoietin is produced by the kidneys but 10% is use of ESAs and iron repletion, and is driven by the need to
produced by the liver. Liver production can respond to avoid both [1] severe anemia and the higher risk of blood
prolyl hydroxylase inhibition by producing more endoge- transfusions, and [2] normal hemoglobin levels, which
nous erythropoietin. However, there are many other were associated with increased CV risk in ESA trials.
References
Skali, H., Lin, J., Pfeffer, M.A. et al. (2013). Hemoglobin

1 Lower Cardiovascular Events trial. J. Am. Heart Assoc. 3
stability in patients with anemia, CKD and type 2 diabetes: (6): e001363. https://doi.org/10.1161/JAHA.114.001363.
an analysis of the TREAT (Trial to Reduce Cardiovascular 5 Vlagopoulos, P.T., Tighiouart, H., Weiner, D.E. et al. (2005).
Events with Aranesp Therapy) Placebo arm. Am. J. Kidney Anemia as a risk factor for cardiovascular disease and
Dis. 61: 238–246. all-cause mortality in diabetes: the impact of chronic
2 McCausland, F.R., Claggett, B., Pfeffer, M.A. et al. (2017). kidney disease. J. Am. Soc. Nephrol. 16: 3403–3410.
Change in hemoglobin trajectary and darbepoetin dose 6 Weiner, D.E., Tighiouart, H., Vlagopoulos, P.T. et al. (2005).
approaching end-stage renal disease: data from the Trial to Effects of anemia and left ventricularhypertrophy on
Reduce Cardiovascular Events with Aranesp Therapy. Am. cardiovascular disease in patients with chronic kidney
J. Nephrol. 46: 488–497. disease. J. Am. Soc. Nephrol. 16: 1803–1810.
3 Canadian Erythropoietin Study Group (1990). Association 7 Chang, J.M., Chen, S.C., Huang, J.C. et al. (2014). Anemia
between recombinant human erythropoietin and quality of and left ventricular hypertrophy wit renal function decline
life and exercise capacity of patients receiving and cardiovascular events in chronic kidney disease. Am J
hemodialysis. BMJ 200: 573–578. Med Sci 347: 183–189.
4 Wheeler, D.C., London, G.M., Parfrey, P.S. et al. (2014). 8 Besarab, A., Bolton, W.K., Browne, J.K. et al. (1998). The
Effects of cinacalcet on atherosclerotic and non- effects of normal as compared with low hematocrit values
atherosclerotic cardiovascular events in patients receiving in patients with cardiac disease who are receiving
hemodialysis: The EValuation of Cinacalcet Therapy to hemodialysis and epoetin. N. Engl. J. Med. 339: 584–590.
9 Parfrey, P., Foley, R., Wittreich, B. et al. (2005). The with chronic kidney disease: a network meta-analysis.
Canadian European Study Group. Double-blind Chochrane Database Syst. Rev. (12) Art. No.: CD 010590.
comparison of full and partial anemia correction in doi:https://doi.org/10.1002/14651858.CD010590.
incident hemodialysis patients without symptomatic 23 Saglimbene, V., Palmer, S.C., Ruospo, M. et al. (2017).
heart disease. J. Am. Soc. Nephrol. 16: 2180–2189. Continuous erythropoiesis receptor activator (CERA) for
10 Drueke, T.B., Locatelli, F., Clyne, N. et al. (2006). the anaemia of chronic kidney disease. Cochrane
Normalization of hemoglobin level in patients with chronic Database Syst. Rev. 8 Art. No.: CD009904. doi:https://doi.
kidney disease and anemia. N. Engl. J. Med. 355: 2071–2084. org/10.1002/14651858.CD 009904.
11 Singh, A.K., Szczech, L., Tang, K.L. et al. (2006). 24 Collister, D., Komenda, P., Hiebert, B. et al. (2016). The
Correction of anemia with epoetin alfa in chronic kidney effect of erythropoietin-stimulating agents on health-
disease. N. Engl. J. Med. 355: 2085–2098. related quality of life in anemia of chronic kidney
12 Pfeffer, M.A., Burdmann, E.A., Chen, C.Y. et al. (2009). disease: a systematic review and meta-analysis. Ann.
The TREAT investigators. A trial of darbepoetin alfa in Intern. Med. 164: 472–478.
type 2 diabetes and chronic kidney disease. N. Engl. J. 25 Foley, R.N., Curtis, B.M., and Parfrey, P.S. (2009).
Med. 361: 2019–2032. Erythropoietin therapy, hemoglobin targets, and quality
13 Macdougall, I.C., Provenzano, R., Sharma, A. et al. (2013). of life in healthy hemodialysis patients: a randomized
Peginesatide for anemia in patients with chronic kidney trial. Clin. J. Am. Soc. Nephrol. 4: 726–733.
disease not receiving dialysis. N. Engl. J. Med. 368: 320–332. 26 Lewis, E.F., Pfeffer, M.A., Feng, A. et al. (2011).
14 Fishbane, S., Schiller, B., Locatelli, F. et al. (2013). Darbepoetin alfa impact on health status in diabetes
Peginesatide in patients with anemia undergoing patients with kidney disease: a randomized trial. Clin. J.
hemodialysis. N. Engl. J. Med. 368: 307–319. Am. Soc. Nephrol. 6: 845–855.
15 Macdougall, L.C., White, C., Anker, S.D. et al. (2018). 27 US Food and Drug Administration. Omontys
Intravenous iron in patients undergoing maintenance (peginesatide) injection by affymax and takeda: recall of
hemodialysis. N. Engl. J. Med. https://doi.org/10.1056/ all lots – serious hypersensitivity reactions. http://www.
NEJMoal 1810742. fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlerts
16 US Renal Data System (2014, ). 2014 Annual Data Report: forHumanMedicalProducts. (accessed 11 March 2013).
Atlas of End-Stage Renal Disease in the United States. 28 Kidney Disease: Improving Global Outcomes (KDIGO)
Bethesda, MD: National Institutes of Health, National Anemia Work Group (2012). KDIGO clinical practice
Institute of Diabetes and Digestive and Kidney Diseases. guideline for anemia in chronic kidney disease. Kidney
17 Parfrey, P.S. (2011). Critical appraisal of randomized Int. Suppl. 2: 279–355.
controlled trials of anemia correction in patients with 29 Shepshelovich, D., Rozen-Zvi, B., Avni, T. et al. (2016).
renal failure. Curr. Opin. Nephrol. Hypertens. 20: 177–181. Intravenous versus oral iron supplementation for the
18 Foley, R.N., Curtis, B.M., and Parfrey, P.S. (2008). treatment of anemia in CKD: an updated systematic
Hemoglobin targets and blood transfusions in review and meta-analysis. Am. J. Kidney Dis. 68: 677–690.
hemodialysis patients without symptomatic cardiac 30 Roger, S.D., Tio, M., Park, H.C. et al. (2017). Intravenous
disease receiving erythropoietin therapy. Clin. J. Am. Soc. iron and erythropoiesis-stimulating agents in
Nephrol. 3: 1669–1675. haemodialysis: a systematic review and meta-analysis.
19 Solomon, S.D., Uno, H., Lewis, E.F. et al. (2010). Nephrology (Carlton) 22: 969–976.
Erythropoietic response and outcomes in kidney disease 31 Besarab, A., Chernyavskaya, E., Motylev, I. et al. (2016).
and type 2 diabetes. N. Engl. J. Med. 363: 1146–1155. Roxadustat (FG-4592): correction of anemia in incident
20 Palmer, S.C., Navaneethan, S.C., Craig, J.C. et al. (2010). dialysis patients. J. Am. Soc. Nephrol. 27: 1225–1233.
Meta-analysis: erythropoiesis-stimulating agents in patients 32 Brigandi, R.A., Johnson, B., Oei, C. et al. (2016). A novel
with chronic kidney disease. Ann. Intern. Med. 153: 23–33. hypoxia-inducible factor-prolyl hydroxylase inhibitor
21 Palmer, S.C., Saglimbene, V., Craig, J.C. et al. (2014). (GSK1278863) for anemia in CKD; a 28-day, phase 2A
Darbepoetin for the anaemia of chronic kidney disease. randomized trial. Am. J. Kidney Dis. 67: 861–871.
Cochrane Database Syst. Rev. Art. No.: CD009297, 33 Pergola, P.E., Spinowitz, B.S., Hartman, C.S. et al. (2016).
doi:https://doi.org/10.1002/14651858.CD 009297 volume 3. Vadudustat, a novel oral HIF stabilizer, provides effective
22 Palmer, S.C., Saglimbene, V., Mavridis, D. et al. (2014). anemia treatment in nondialysis-dependent chronic
Erythropoiesis-stimulating agents for anaemia in adults kidney disease. Kidney Int. 90: 1115–1122.
549
34
Dyslipidemia in Chronic Kidney Disease

Elani Streja1,2 and Dan A Streja3,4
1
Division of Nephrology, Department of Medicine, Harold Simmons Center for Kidney Disease Research and Epidemiology, UC Irvine School of Medicine, Orange, CA, USA
2
Tibor Rubin VA Medical Center, Long Beach, CA, USA
3
David Geffen School of Medicine at UCLA, Los Angeles, CA, USA
4
Division of Endocrinology, Diabetes and Metabolism, VA Healthcare System, West Los Angeles VA Medical Center, Los Angeles, CA, USA
I ntroduction dyslipidemia in CKD patients, and the prognosis of

dyslipidemia for CV and mortality outcomes may shift as
Cardiovascular (CV)-related mortality is the leading cause of CKD progresses. However, according to the most recent
death in patients with chronic kidney disease (CKD) [1]. American College of Cardiology and American Heart
However, dyslipidemia and treatment of lipids in CKD are Association (ACC/AHA) guidelines [3], presence of CKD
distinct from that of the general population. A deeper under- is considered a coronary risk equivalent and is itself an
standing of how lipids behave in CKD and the results of trial indication for treatment. Therefore, in patients with CKD,
data regarding CKD outcomes or lipid-lowering therapy in measuring lipid levels may be more useful to determine
CKD patients are needed in order to understand how to adherence and response to therapy, rather than as a meas-
properly treat patients with CKD. In this chapter, we will ure to ascertain disease risk and indicate treatment.
review lipid metabolism, describe changes in lipid particles
and metabolism as CKD progresses, the role of these mark-
ers in CKD, and the impact of dyslipidemia and treatment E
pidemiology
with lipid-altering therapies across different stages of CKD.
In the general population, TC, LDLc, and TG are strongly
and positively associated with incident CV events [4–6],
Dyslipidemia Diagnosis while HDLc has been associated with a lower risk of CV
The Executive Summary of the Third Report of the National events. However, reverse associations between these lipid
Cholesterol Education Program (NCEP) Expert Panel on markers and CV events have been described in segments of
Detection, Evaluation, and Treatment of High Blood CKD patients, a so-called “reverse epidemiology.”
Cholesterol in Adults (Adult Treatment Panel III) [2] In nondialysis dependent chronic kidney disease (NDD-
provides a description of dyslipidemia in patients. CKD), the value of lipid levels in predicting cardiovascular
Dyslipidemia may be defined by a total cholesterol (TC) disease (CVD) appeared to attenuate across worsening
200 mg/dl, low-density lipoprotein cholesterol (LDLc) stages of CKD, although some studies showed conflicting
100 mg/dl, high-density lipoprotein cholesterol (HDLc) results. Tonelli et al. studied 836 060 adults, including
<40 mg/dl, and triglycerides (TG) 150 mg/dl. These lipid 47 092 NDD-CKD patients (estimated glomerular filtration
measurements are typically used in conjunction with other rate [eGFR] 15–59 ml/min/1.73m2) from the Alberta Kidney
comorbid measures to ascertain the 10-year risk of a CV Disease Network and found that the association between
event to evaluate indication of treatment for dyslipidemia. higher LDLc and risk of myocardial infarction (MI) was
As will be discussed below, alterations in lipid levels across weaker for people with lower baseline eGFR despite that
progressing CKD may alter these threshold definitions of group having a higher absolute risk of MI [7]. Bowe et al.
550 Dyslipidemia in Chronic Kidney Disease
similarly found an interaction between eGFR category and correlated significantly with inflammatory markers serum
HDLc for mortality risk where lower eGFR attenuated the interleukin (IL)-6 level and C-reactive protein (CRP) level as
lower risk conferred by intermediate HDLc for risk of death well as with several measures of nutritional status and risk
in a very large male cohort of US veterans [8]. The study also of death [17]. In patients with more severe MICS as indi-
found a u-shaped association between HDLc and mortality cated by a high CRP and low albumin, the levels of TG and
risk across all eGFR groups, where higher mortality risk was other lipid markers were also significantly decreased [18].
observed for patients with HDLc >50 mg/dl. Moreover, Studies have demonstrated the impact of MICS on the
Navaneethan et al. further examined HDLc mortality asso- relationship between lipids and mortality in dialysis
ciations in NDD-CKD US veteran patients and reported that patients. In a cohort of 823 incident dialysis patients, Liu
HDLc >60 mg/dl was associated with lower all-cause, CV, et al. showed that a 40 mg/dl increment in baseline TC was
malignancy-related, and non-CV/nonmalignancy-related associated with a lower risk of all-cause mortality overall
deaths among women but not in men [9]. Navaneethan et al. and in the presence of inflammation/malnutrition, but a
also found a shallower u-shaped relationship between TG higher mortality risk in the absence of inflammation/mal-
and mortality for NDD-CKD patients age 65 years and older, nutrition in models adjusted for age, race, and sex [19].
but a direct l inear relationship in younger patients [10]. The Further adjustment for traditional CV risk factors, dialysis
Atherosclerosis Risk In the Communities (ARIC) study also modality, comorbidity, and inflammatory markers attenu-
reported an association of lipids with coronary heart disease ated the inverse association but strengthened the positive
similar to that of the general population across eGFR groups association. In keeping with these findings, another study
in 807 NDD-CKD patients [11]; however, the study was only showed that in dialysis patients, lower serum TC was a sig-
powered to adjust for four demographic covariates and may nificant predictor of death, but higher TC was associated
have residual confounding. Conversely, another small study with higher death risk in a subgroup of patients with serum
of 840 stage 3 and 4 NDD-CKD patients in the Modification albumin 4.5 g/dl [20]. In the Taiwan Renal Registry Data
of Diet in Renal Disease (MDRD) cohort showed that the System, in nonatherosclerotic cardiovascular disease
significant association of continuous TC with all-cause mor- (non-ASCVD), both TC >250 mg/dl and <150 mg/dl were
tality was attenuated with covariate adjustment, and there associated with a higher risk of mortality compared to the
was no association of TG, HDLc, and non-HDLc with all- reference (TC 150–200 mg/dl); however, in the ASCVD
cause or CV mortality [12]. Kovesdy et al. also studied the patients, only TC <150 mg/dl had a higher risk [21]. In a
association between all four lipids and mortality in cohort of hemodialysis patients, each 50 mg/dl excess in TG
1012 NDD-CKD patients in a single-center veteran study was associated with a progressively lower risk of all-cause
and found that compared to the highest quartile of lipid lev- and CV mortality in patients with a waist circumference
els, lower lipid levels (TC, LDLc, TG, and HDLc) were asso- <95 cm but with a progressively higher risk in patients with
ciated with higher all-cause mortality [13]. However, these a waist circumference above this threshold [22].
associations were attenuated after adjustment for demo- Kilpatrick et al. also examined the association of lipid
graphics and abolished after additional adjustment for markers with mortality in consideration of MICS in a large
markers of malnutrition and inflammation complex (or dialysis cohort [23]. Both higher TC and LDLc showed a para-
cachexia) syndrome (MICS). Furthermore, in the African doxical association with better survival. Hyper-triglyceridemia
American Study of Kidney Disease and Hypertension (>200 mg/dl) also showed a similar trend. The association
(AASK) study, higher TC was associated with the CV end- between a low serum LDLc (<70 mg/dl), which was preva-
point in non-MICS NDD-CKD patients, but not in MICS lent among almost 50% of hemodialysis patients in the cohort,
NDD-CKD patients [14]. and higher all-cause death risk was robust to multivariate
The concept of this reverse epidemiology relationship adjustment, including MICS markers. In subgroup analyses,
between lipids and outcomes in CKD has been attributed to these paradoxical associations persisted among most sub-
MICS. The MICS concept was introduced by Kalantar-Zadeh groups, although they tended to be stronger among hypoal-
et al. for dialysis patients [15] and subsequently defined by a buminemic (<3.8 mg/dl) patients and those with a lower
specific diagnosis description [16]. The description of the dietary protein intake (<1 g/kg/day). Conversely, in African
syndrome includes medical history of weight loss, anorexia, American patients, a high serum LDLc (>100 mg/ml) was
gastrointestinal symptoms, reduced activity, comorbidity, associated with a higher adjusted CV death hazard ratio. The
and long dialysis vintage as well as objective examination authors of the study cautioned that the so-called reverse epi-
findings (decreased subcutaneous fat, muscle wasting, low demiology observations of lipids with mortality in dialysis
body mass index [BMI], low albumin, and low total iron patients should not be considered causal or lead to the nihil-
binding capacity). In a subsequent publication from the istic conclusion of no clinical benefit from cholesterol lower-
same group, the malnutrition-inflammation score (MIS) ing therapy. Another small study similarly showed lower TC
Dyslipidemia as a Precursor of CK 551
predicted all-cause and CV mortality in hemodialysis the association between MetS and mortality in peritoneal
patients [24]. Other studies in hemodialysis patients have dialysis was attenuated after adjustment for CRP. One small
also shown that higher TG, TG/HDLc ratio, non-HDLc, and study has shown that higher very-low-density lipoprotein
non-HDLc/HDLc ratio were inversely associated with mor- cholesterol (VLDLc) was associated with higher mortality
tality risk in hemodialysis patients in models adjusted for risk in peritoneal dialysis patients even after adjustment for
markers of malnutrition and inflammation [25, 26]. In the MICS covariates [32]; however, additional studies are
same cohort, higher levels of HDLc [27] and increases in needed to support this different finding.
HDLc [28] were associated with higher mortality risk in In transplanted patients, elevated levels of TC or its sub-
hemodialysis patients in analysis with MICS adjustment. A fractions and elevated TG levels have not been shown to be
further study comparing HDLc-mortality associations in East associated with increased risk for patient mortality or
Coast versus West Coast Hispanic dialysis patients showed allograft loss [33], possibly due to the variability of lipopro-
the relationship of higher HDLc with mortality differed tein concentrations and of the factors affecting graft sur-
according to coast and may be explained by geographical vival. Across CKD stages, the relationship of lipids with
variances in racial-ethnic and genetic distinctions [29]. mortality and CV outcomes is paradoxical to that observed
Associations of lipid markers with outcomes in peritoneal in the general population.
dialysis patients have conflicting results. In 1053 continuous
ambulatory peritoneal dialysis (CAPD) patients, lower TC
levels were associated with a higher risk of all-cause mortal- Dyslipidemia as a Precursor of CKD
ity, including in those taking or not taking lipid-modifying
medications [30]. Additionally, in a cohort of 749 Korean CKD is a frequent complication of type 2 diabetes, and type
incident peritoneal dialysis patients, lower TC, LDLc, TG, 2 diabetes is the leading cause of CKD [1]. The majority of
and HDLc were associated with a higher all-cause mortality patients with type 2 diabetes also have features of MetS,
risk, and lower TC and LDLc were associated with higher including obesity, hypertension, elevated TGs, and low
CV mortality risk. However, these associations may be HDLc [2, 34]. In patients with MetS, dyslipidemia occurs
impacted by MICS as well. In a study of nondiabetic CAPD due to an increased production and decreased catabolism
patients, metabolic syndrome (MetS), characterized by ele- of VLDLc, thereby resulting in an increase in the fraction
vated TG and lower HDLc, was a predictor of mortality in of small dense low-density lipoprotein (sd-LDL) and catab-
patients with serum albumin >3.0 g/dl, but not in those with olism of HDL attributable to insulin resistance (see
albumin <3.0 g/dl [31]. Another study similarly showed that Figure 34.1 for a basic comparison of density and lipid
Increasing lipid content
Chylomicron
Remnants
Non-HDL cholesterol
VLDL
IDL
LDL
SD
LDL
HDL
Increasing density
Figure 34.1 Serum lipoproteins: relative density and lipid content. HDL, high-density lipoprotein; IDL, intermediate density
lipoproteins; LDL, low-density lipoprotein; SD LDL, small dense low-density lipoprotein; VLDL, very low-density lipoprotein.
c ontent across serum lipoproteins, and Figure 34.2 for a increases in TG concentration, changes in LDL subfraction
summary of MetS impact on VLDL). distribution, and decreases in HDLc concentration.
Patients with MetS also have an increased likelihood of
developing CKD [35, 36] and the strength of this associa-
tion seems to increase with the number of MetS compo- hanges in Triglyceride-carrying
C
nents present [37]. Both high TG and low HDLc have been Lipoprotein Particles in NDD-CKD
independently associated with the risk of incident
CKD [38]. The higher quartile of TG/HDLc ratio at base- Similar to MetS, dyslipidemia of CKD patients without sig-
line was significantly associated with greater decline in nificant proteinuria is characterized by hypertriglyceri-
eGFR and increase in urinary protein excretion after demia, elevated plasma levels of VLDLc, intermediate
adjustment for confounding factors [39, 40]. We conclude density lipoprotein (IDL), and chylomicron remnants
that changes in the concentration and composition of lipo- (Figure 34.1), and additionally low plasma HDLc and its
proteins characteristic of MetS and type 2 diabetes predict main apolipoprotein (Apo), ApoA-I [41]. While in MetS the
incident CKD and consequently may be transferred to the primary cause of hypertriglyceridemia is an increased pro-
CKD population and become a feature of CKD. duction of VLDL TGs attributable to the increased flow of
free fatty acids [42, 43], in NDD-CKD without proteinuria
a decreased VLDL metabolism appears to be the main
yslipidemia in Nonproteinuric
D cause [44, 45].
NDD-CKD and Hemodialysis The average TG concentration increases with CKD stage
and an inverse association between eGFR and TGs was
Dyslipidemia in nonproteinuric NDD-CKD patients is described [10, 46–48]. The pathophysiology of the defect in
characterized by the same features as dyslipidemia in MetS: TG metabolism in CKD is attributable primarily to an
increase in ApoC-III resulting in a decreased catabolic rate
of VLDL and IDL [49], with ApoC-III concentration as the
only independent predictor of clearance defects (Figure 34.3).
Metabolic Syndrome
• Obesity
On VLDL particles, ApoC-III is an inhibitor of lipoprotein
• Hypertension Increased VLDL production lipase (LPL) activity responsible for VLDL catabolism. In
• Elevated TG
Decrease VLDL catabolism CKD, the ApoC-III fractional catabolic rate is also
• Low HDL
• Insulin Resistance
decreased [50], leading to the increase in its concentration.
VLDL
(or Diabetes) The decreased TG removal in CKD is also in part attrib-
utable to a decrease in LPL function, although the contri-
bution is not as high as that of ApoC-III [51]. In addition,
hypertriglyceridemia of nonproteinuric NDD-CKD is
Figure 34.2 Metabolic syndrome impact on lipid metabolism.
HDL, high-density lipoprotein; TG, triglycerides; VLDL, very mainly attributable to a lipolytic defect due to a progressive
low-density lipoprotein. increase in ApoC-III as CKD progresses.
Endogenous Pathway
Liver ApoC-I
Apo B
100
VLDL-1 ApoC-II Activates LPL
ApoC-III X Inhibits LPL

Cholesterol
Apo
E
triglycerides
X Inhibits LPL ANGPTL3 Lipo
protein
Lipase
(LPL)
2 free fatty acids

1 monoacylglycerol
Figure 34.3 Endogenous pathway of VLDL metabolism. ANGPTL3, angiopoietin-like protein 3; APO, apolipoprotein; LPL, lipoprotein
lipase; VLDL, very low-density lipoprotein.
Nephrotic Syndrom 553
hanges in Triglyceride-carrying
C Since the majority of patients with nonproteinuric CKD
Lipoprotein Particles Common to and hemodialysis have hypertriglyceridemia, the spectrum
of LDL particles is shifted toward smaller and denser parti-
NDD-CKD and Hemodialysis
cles [64, 65]. As with TG concentrations, sd-LDL is signifi-
cantly associated with CKD stage and BMI [66]. In
Additional pathophysiological mechanisms leading to
hemodialysis patients, even in normotriglyceridemic
dyslipidemia have been described in hemodialysis patients,
patients, there are changes in the composition of lipoproteins
although it is not clear to what extent these mechanisms also
indicating a shift toward denser subfractions in the absence
operate in earlier stages of CKD. In hemodialysis patients,
of hypertriglyceridemia [67], while in hypertriglyceridemia
concentrations of VLDLc ApoC-II/C-III ratio were lower
there is an increase in particle number indicating increased
(attributed to higher ApoC-III concentrations) even in the
atherogenicity for the same amount of LDL cholesterol [68].
presence of normal TG [52]. In addition, studies described
Increased oxidative stress is a major contributor to the
the presence of an inhibitor of LPL in the plasma of hemodi-
increased atherosclerosis and CV morbidity and mortality
alysis patients [53, 54]. Another contributor to hemodialysis-
found in uremia [69]. In different stages of CKD there is a
related changes in LPL is the progressive decrease in the
marked increase in oxidized LDL [70] or of antibodies
presence of an activating factor for LPL in the HDL fraction
against oxidized LDL [71].
of the serum [55]. Hemodialysis patients also have a defect
Another characteristic of LDL spectrum lipoproteins in
in the removal of chylomicron remnants [56, 57], resulting
CKD is an increased concentration of lipoprotein (a)
in a prolonged circulation time. These studies could be
(Lp(a)), a lipoprotein subclass associated with atherogenic
interpreted as a severe defect in postprandial lipoprotein
risk. Lp(a) levels have been found to be elevated across the
metabolism, specifically a decrease in remnant uptake by
entire spectrum of eGFR in CKD patients [72]; kidney
the liver, in hemodialysis patients. Other studies have shown
function was inversely correlated with Lp(a) levels, but this
that the function of lipases is reduced by increased levels of
relationship is present only in nonproteinuric CKD and for
the angiopoietin-like protein 3 (ANGPTL3) and 4 which is
subjects with large Apo(a) isomorphs [73]. The increase in
markedly increased in hemodialysis [58, 59]. LPL level
Lp(a) concentration in CKD is attributable to a decreased
seems to decrease with dialysis duration [60], indicating a
removal rate [74]. Elevated Lp(a) is associated with a more
time effect on the lipolytic deficiency. The use of heparin
rapid eGFR decline [75].
(which detaches LPL from the endothelium and releases it
into the circulation) in hemodialysis has an additional
impact on atherogenicity [61]. After a hemodialytic
yslipidemia in Nephrotic Syndrome
D
procedure with heparin, readily available LPL in circulation
readily metabolizes VLDL, resulting in a significant
and Peritoneal Dialysis
reduction of TG concentration and an increase in VLDL
In nephrotic syndrome and peritoneal dialysis, additional
atherogenic remnants. The authors interpreted their data as
pathophysiological mechanisms are added to those
showing that heparin administration during hemodialysis
described for nonproteinuric CKD and hemodialysis.
leads to a massive production in atherogenic particles
because of the acute stimulation of LPL activity not followed
by an efficient removal of lipoprotein remnants. Nephrotic Syndrome
Early studies described an increase in the production of lipo-

hanges in LDL Common to
C proteins in nephrotic syndrome in proportion with albumi-
Nonproteinuric NDD-CKD and nuria [76–78]. However, this conclusion was questioned
Hemodialysis when an improved technology was used [79] and showed
that VLDL Apo B100 levels were primarily increased as a
Although LDLc concentrations in nonproteinuric CKD consequence of a decrease in fractional catabolic rate.
patients are similar to those of normal controls, there are However, LDL Apo B100 synthesis rate was significantly
important differences in LDL kinetics in nonproteinuric increased in nephrotic patients compared to controls and
CKD and hemodialysis patients as compared to control was greater than that of VLDL Apo B100 in some patients,
patients. The fractional clearance rate of LDL is markedly suggesting that LDL may be, in part, derived from a direct
diminished and possibly decreases with the decrease in secretory pathway. These data indicate that increased VLDL
eGFR [62]. However, since LDL and IDL levels are regulated in nephrotic patients results from a decreased catabolism,
by both clearance as well as production, there is no clear while increased LDL results from increased synthesis.
relationship between clearance and concentration [63]. However, patients with nephrotic syndrome with no
ypertriglyceridemia have low fractional catabolic rates of

h changes may explain why the relationship of HDL with
LDL Apo B and no overproduction of LDL Apo B [80]. A outcomes in kidney disease patients changes across
receptor mediated uptake of LDL defect plays an important progressing disease [27, 28].
role in regulating LDL Apo B concentration [81].
Patients with nephrotic syndrome were shown to display
VLDL enriched with cholesteryl esters (CE), with a CE/TG HDL Structural Changes
ratio approximately twice that of controls and plasma
cholesteryl ester transport protein (CETP) concentration In 1966, Lewis et al. described that in anephric patients,
significantly increased compared to controls. CETP is HDL is very low, and it increases after transplantation [92].
involved in HDL/VLDL metabolism and mediates the In uremic patients, hypertriglyceridemia and low LPL
transfer of CE in HDL to VLDL in exchange for TGs. (signifying a reduced lipid metabolism) are associated with
Treatment with corticosteroids or angiotensin receptor higher HDLc [93]. Inflammation contributes to the
blockers reduced proteinuria and LDLc and normalized reduction in HDL concentration in end-stage renal disease
plasma CETP [82–84]. The drop in CETP correlated with (ESRD) [94]. ApoA-I, the main HDL lipoprotein, changes
the decrease in LDLc inversely with serum concentrations of CRP and the
Other authors identified a role for elevated proprotein inflammatory marker α 1 acid glycoprotein (α 1AG). Apo
convertase subtilisin/kexin type 9 (PCSK9) in the A-I fractional catabolic rate was 50% increased in hemodi-
hyperlipidemia of nephrotic syndrome [85, 86] and PCSK9 alysis patients, whereas the production rate remained
is corrected by dialysis and normalized after renal unchanged [95]. Oxidized HDL is also increased in dialysis
transplant [87]. PCSK9 is an enzyme that mediates the patients [96].
catabolism of the LDL receptor, and thereby its inhibition The protein composition of HDL is also changed, with a
results in an increased LDL receptor number. Plasma higher retinol-binding protein 4, serum amyloid A1, and
PCSK9 level is directly and strongly correlated to total and lipoprotein-associated phospholipase A2, and lower
LDL cholesterol concentrations. vitronectin [97–99]. The composition changes also correlate
with the observed impaired ability of uremic HDL to pro-
mote cholesterol efflux from macrophages.
P
eritoneal Dialysis Another characteristic change in HDL composition in
CKD, but not in nephrotic syndrome, is lecithin cholesterol
In 1987, Sniderman et al. showed that compared to hemo- acyltransferase (LCAT) deficiency [100, 101]. Paroxonase 1
dialysis patients, CAPD patients have higher TC, LDLc, (PON1), a strong anti-atherogenic enzyme of HDL, activity
TG, Apo B, LDL-ApoB, and VLDL-ApoB [88, 89]. In addi- has also been shown to be diminished in CKD [102]. PON1
tion, the clearance of LDL in CAPD patients has been concentration is significantly and positively associated
shown to be markedly decreased in comparison to that of with HDLc and Apo A1 [103]. PON1 activity was also
control subjects [90]. Elevated levels of plasma VLDL and shown to be lower in CAPD compared with
reduced levels of plasma HDL were maintained in these hemodialysis [104]. PON1 activity in CKD subjects
patients throughout 5 years of CAPD, whereas the initially predicted a higher risk of incident long-term adverse CV
increased LDL levels showed a tendency toward events [105].
normalization [91]. All plasma lipoproteins (VLDL, IDL,
LDL, and HDL) are present in the peritoneal effluent. The
peritoneal protein clearance correlates positively with HDL Functional Changes
plasma levels of TG and LDL, and negatively with plasma
HDL. In addition, an inverse correlation was observed To demonstrate the functional changes of HDL in CKD,
between plasma levels of HDL and its peritoneal clearance. the biological activity of high-density lipoprotein particles
The continuous peritoneal loss of HDL and the in patients with chronic kidney disease (HDL-CKD) is
hypertriglyceridemia contribute most to the persistent low compared with the same activity in healthy controls (HDL-
plasma levels of HDL in CAPD patients. H). HDL-CKD, but not HDL-H, promoted endothelial
superoxide production, substantially reduced nitric oxide
bioavailability in endothelial cells in vitro, and increased
HDL Changes in CKD arterial blood pressure in mice [106]. HDL-H, but not
HDL-CKD, inhibits the production of inflammatory
In CKD, HDL undergoes a number of structural and func- cytokines by stimulated peripheral monocytes [107].
tional changes. Researchers have explained that these Compared with HDL-H, HDL-CKD increased the cytokine
Lipid-altering Trials in CKD Patient 555
response and disrupted macrophage chemotaxis in human hyperlipidemia should be balanced against the risk of new
umbilical vein endothelial cells [108]. When the efflux onset diabetes [122, 123].
capacity of HDL was tested directly, HDL-CKD from
dialysis patients was dramatically less effective than
normal HDL-H in accepting cholesterol from macro- ammalian Target of Rapamycin
M
phages [109]. Statins had no effect on the impairment of Inhibitors
HDL-mediated efflux. Cholesterol acceptor capacity of
HDL-CKD from dialysis patients was confirmed to be The drugs from this class are sirolimus and everolimus,
markedly reduced compared with that of HDL-H [110]. and they have a similar metabolic profile [124]. They are
almost invariably used together with calcineurin inhibitors.
Conversion of transplanted patients from TAC to sirolimus
yslipidemia in Patients with Renal
D resulted in a significant increase in TC, LDLc, and TG [125,
Transplant 126]. This effect is the result of reduced catabolism of Apo
B100-containing lipoproteins [127]. During sirolimus upti-
There is also a marked variability of lipoprotein concentra- tration, TG, Apo B100, and ApoC-III levels increased in a
tions in renal transplant patients [111]. This lipoprotein dose-dependent manner [128]. Sirolimus expands the free
variability is at least in part attributable to medications used fatty acid pool, and therefore the incorporation of a labeled
for graft survival, and also to the intensity of statin therapy. fatty acid, [13 C4]-palmitate, into TG of VLDL, IDL, and
LDL was decreased. The authors suggested that sirolimus-
related increased free fatty acid flow is due to an altered
P
rednisone insulin signaling pathway resulting in an increased hyper-
triglyceridemia. Hyperlipidemia induced by sirolimus will
Prednisone in transplanted patients treated with cyclo- also vary with the genetic polymorphism of adenosine
sporin (CsA) resulted in increases in HDL cholesterol [112, triphosphate (ATP)-Binding Cassette (ABC)B1-3435,
113]. High-dose prednisone resulted in decreased Lp(a) ABCB1-1236, and IL-10-1082 genes [129].
levels 1 week after transplantation [114]. As steroid doses
were gradually tapered, Lp(a) concentrations subsequently
increased. Transplanted patients treated with alternate day Lipid-altering Trials in CKD Patients
steroids had a significant decrease in both serum TG and
TC [115]. In 2011, the highly anticipated results of the largest study
to date examining the effectiveness of statin therapy in
patients with CKD were published. The Study of Heart and
Calcineurins Renal Protection (SHARP) included 9270 CKD patients
(4650 assigned to combined statin plus ezetimibe therapy
The drugs from this class are CsA and tacrolimus (TAC), and 4620 patients on placebo) [130] aged 40 years or older
and each have a very different metabolic profile. When with no known history of MI or revascularization; 3056 of
transplanted patients are treated with prednisone, with or the patients were on dialysis at the time of study entry.
without CsA, the patients receiving CsA demonstrated Unlike other prior statin trials that only included NDD-
increased plasma concentrations of LDLc [116]. CsA blood CKD stage 3 patients or dialysis patients, SHARP notably
levels are inversely associated with serum Lp(a) levels in expanded the criteria to include not only those two
renal transplant recipients; this association is independent groups but additionally patients with NDD-CKD stage 4
of prednisone therapy [117]. CsA therapy has opposing (41% eGFR 15–29 ml/min/1.73m2) and stage 5 (20%
effects on plasma Lp(a) and LDLc accumulations. eGFR < 15 ml/min/1.73m2). The study had enrollment sites
Replacing CsA with TAC, in one study of transplanted worldwide and included 72% White, 3% Black, 23% Asian,
patients, resulted in significant decreases in TC, LDLc, and and 3% other or unspecified patients. The mean ± SD age of
ApoB levels [118]. In other studies, a switch from CsA to enrolled patients was 62 ± 12 years, and 31% of patients
TAC significantly decreased TG, ApoB, LDLc, and TC were age 70 or older. The study’s primary outcome of inter-
levels [119, 120]. Changes in HDLc in these studies were est was first major atherosclerotic event (nonfatal MI or
variable. In a small randomized trial of CsA versus TAC in coronary death, non-hemorrhagic stroke, or any arterial
renal transplantation, the CsA treatment group had revascularization procedure). Compared to patients on
significant increases in TC, TG, LDLc, and HDLc [121]. placebo, patients in the simvastatin plus ezetimibe treated
However, the advantages of TAC over CsA in terms of arm had a 17% reduction in major atherosclerotic events,
25% reduction in nonhemorrhagic stroke, and 21% reduc- did not reach statistical significance (RR 0.90, 95% CI 0.75,
tion in arterial revascularization procedures. However, 1.08), many clinicians have interpreted this finding to show
there was no difference in the number of coronary deaths, that the combined simvastatin plus ezetimibe therapy com-
and differences in the number of nonfatal MIs did not reach pared to placebo failed to reduce major outcomes in dialysis
statistical significance. Subgroup analysis revealed the outpatients. Two other larger trials previously evaluated the
come risk reductions were larger in patients with higher benefit of initiating statin therapy in dialysis patients. The
baseline BMI, TC, and LDLc. Strikingly, there was an incre- Deutsche Diabetes Dialyse Studie (4D) [133] enrolled 1255
mentally lower benefit as CKD progressed with lower eGFR type 2 diabetic patients on maintenance hemodialysis
levels or higher excreted microalbumin; however, subgroup (median dialysis vintage 8 months) and randomized them
analysis did not show significant heterogeneity across these to receive either 20 mg of atorvastatin daily or placebo.
groups. In the SHARP study, there was no excess risk of Lipid-lowering benefit was observed in the statin-treated
adverse events for patients on statin therapy. group after only 4 weeks of the trial, where LDL was reduced
Soon after the results of the SHARP trial were published, by 42% in statin-treated patients. However, after a median
meta-analysis studies pooled the results of statin trials in follow-up of 4 years there was no difference in primary end-
CKD patients (including SHARP) to determine statin effi- points (composite of death from cardiac causes, nonfatal
cacy. Palmer et al., as part of the Cochrane collaboration, MI, and stroke). When outcomes were evaluated separately,
reviewed over 50 studies with over 45 000 CKD patients and statin patients had a reduced risk of all cardiac events, no
found that statin therapy in NDD-CKD reduced the risk of difference in all-cause mortality risk, but a higher risk of
major CV events by 18%, risk of all-cause mortality by 21%, fatal stroke. The Study to Evaluate the Use of Rosuvastatin
risk of CV mortality by 23%, and risk of MI by 45% [131]. in Subjects on Regular Hemodialysis: An Assessment of
The study was not able to find an association between statin Survival and Cardiovascular Events (AURORA) [134] also
therapy and stroke prevention. Although the meta-analysis investigated the effects of statin (rosuvastatin 10 mg/day)
was not able to thoroughly evaluate potential harms from versus placebo therapy; this study included 2776 hemodi-
statin therapy due to lack of systematic reporting of adverse alysis patients (median dialysis vintage 3.5 years) aged
events across studies, no significant association between 50–80 years. Similar to 4D, after 3 months the LDL in the
statin therapy and adverse events, including elevated cre- statin-treated group had reduced by 43% compared to 2% in
atinine kinase, liver function abnormalities, and cancer, the placebo group. However, also similar to 4D, after a fol-
was found in the limited data that was available. Similarly, low-up of 3.8 years there was no benefit of statin in reducing
Hou et al. also conducted a meta-analysis that included 31 the primary endpoint (death from CV causes, nonfatal MI,
trials and nearly 50 000 patients with CKD and found that or nonfatal stroke). Statins also had no effect on the indi-
statin therapy reduced CV events by 23%, coronary events vidual components of the primary endpoint and no signifi-
by 18%, and all-cause and CV deaths by 9%, but no significant effect on all-cause mortality.
cant associations were found for stroke or adverse events, The Hou et al. meta-analysis also examined associations
including liver function or muscular disorders [132]. of statin therapy with outcomes in dialysis patients (includ-
Moreover, Hou et al. corroborated the findings of the ing the above three trials) and determined that although
SHARP trial in demonstrating a decreasing benefit of statin there was a lack of a significant association for mortality
therapy across worsening CKD stage. The meta-analysis outcomes in the dialysis patients, there was no heterogene-
found that for CKD stage 3 (which also included two small ity of effect for all-cause mortality or CV death between
studies of CKD stage 2 patients), stage 4 (which included dialysis and NDD-CKD patients. However, associations did
one other study aside from SHARP), and stage 5 NDD- differ for coronary event and stroke outcomes, which found
CKD (SHARP only) CV outcomes were reduced by 31%, no benefit of statin therapy for dialysis patients. There was
22%, and 18%, respectively, for statin-treated patients com- also no heterogeneity of effect between dialysis and NDD-
pared to placebo patients. CKD for adverse event outcomes, including liver disorders,
creatinine kinase elevation, new cancer onset, muscle pain,
and serious adverse events [132].
Statins in Hemodialysis
The SHARP trial also included 3023 patients on dialysis, of Statins in Peritoneal Dialysis
which 84% were hemodialysis patients. The study showed
there was no heterogeneity of effect for the efficacy of the Clinical trials examining statin initiation in peritoneal dialysis
treatment between dialysis and NDD-CKD patients. are limited. SHARP included only 496 peritoneal dialysis
However, because the risk ratio for the dialysis subgroup patients and failed to find a significant benefit of statin therapy
Statins and Progression of CK 557
in reducing the primary endpoint, although estimates trended groups. A post hoc analysis of ALERT found benefit of fluv-
toward statin benefit (RR 0.70, 95% CI 0.46, 1.08). However, a astatin in reducing an alternative primary endpoint of car-
propensity score matched retrospective United States Renal diac death or definite nonfatal MI. In subgroup analysis for
Data System (USRDS) cohort study of 1053 peritoneal dialysis this endpoint, they also found significant benefit of fluvas-
patients found that use of lipid-modifying medication (mostly tatin in patients who were younger, nondiabetic, nonsmok-
statins) was associated with 26% and 33% lower risk of all- ers, and without pre-existing CVD. However, the authors
cause and CV mortality, respectively; this was particularly the caution that this subgroup analysis is exploratory and over-
case in patients with elevated cholesterol levels [135]. lapping confidence intervals demonstrated no heterogene-
Additionally, a propensity score matched cohort analysis of ity of effect across comparison strata [141]. Another post
1024 Korean peritoneal dialysis patients found statin use was hoc analysis determined that 32% of patients enrolled in
associated with 45% lower mortality risk [136]. As discussed ALERT had MetS and risk of major adverse CV events were
above, peritoneal dialysis patients have higher levels of ApoB reduced by 53% in this group, thereby potentially identify-
containing lipoproteins [137], and thereby statin benefit may ing a group of renal transplant recipients who may benefit
be different in peritoneal patients compared to hemodialysis from statin therapy. When the study investigators con-
patients. Additional trials examining statin initiation in peri- ducted an additional post hoc analysis on patients who
toneal dialysis patients and reduction of CV events should be were less than 4.5 years after renal transplant at study base-
undertaken. line, they found risk of cardiac death and nonfatal MI was
The conclusion drawn from the dialysis trials as summa- reduced by 56% and concluded the analysis supported intro-
rized by meta-analysis was that statin therapy appears to be duction of statin therapy early in the post-transplant
ineffective in reducing CV events when initiated in patients period [142]. Lastly, an extension study offered open label
already on dialysis. As will be discussed below, these trials fluvastatin for two additional years to patients who com-
provided the evidence to determine guidelines suggesting pleted the ALERT trial and included 1652 (92%) of ALERT
not to initiate statin therapy in dialysis patients, but to con- trial participants. Patients randomized to fluvastatin had a
tinue therapy after transition to dialysis when statins were 29% reduction in cardiac death or definite nonfatal MI [143].
already started in late-stage CKD. Although continuation of A meta-analysis by Palmer et al. summarized statin trials in
statin therapy has not been examined in a clinical trial, a renal transplant recipients, including the ALERT study, and
recent cohort study [138] of US veteran patients transition- suggested that statin therapy may reduce CV endpoints,
ing to dialysis found that continuation of statin therapy however the current data lacks precision to confirm any
after dialysis transition was associated with a 28% lower risk benefit and further trials are needed [144].
of all-cause mortality and an 18% lower risk of CV mortality
in adjusted analysis compared to patients that discontinued
statin use. As noted in the invited commentary for this Statins and Progression of CKD
study, this analysis is supportive of statin continuation, but
not sufficient “until high-quality randomized clinical trials The benefit of statin therapy in reducing progression of
evaluating this question are conducted” [139]. CKD has been examined in most clinical trials; however,
the results have been conflicting. In the meta-analysis by
Palmer et al. no conclusions could be made about whether
Statins in Renal Transplant statin therapy slowed CKD progression. Although the anal-
ysis found a significant benefit of statin therapy on decreas-
The use of statins for prevention of CV events in renal trans- ing proteinuria, there was no benefit found for reducing the
plant patients has to date only been tested in one clinical number of patients transitioning to ESRD. This latter analy-
trial. The Assessment of LEscol in Renal Transplantation sis was based solely on the results of the SHARP trial.
(ALERT) trial randomized 2102 renal transplant recipients Another meta-analysis similarly found a benefit of statins
to fluvastatin 80 mg/day versus placebo and compared the in reducing proteinuria outcomes in patients with urinary
risk of the primary endpoint (cardiac death, nonfatal MI, or protein excretion of 30 mg/day or more [145]. They also
coronary revascularization) [140]. After a mean of 5.1 years, found that CKD patients receiving statin therapy compared
LDLc was lowered by 32% in statin-treated patients; how- to placebo had less of an eGFR reduction by study end;
ever, risk reduction for the primary endpoint was not sig- however, these results seem largely influenced by one study
nificant. There were, however, fewer cardiac deaths or as no benefit was observed in the other nine studies
nonfatal MI events in the fluvastatin group than in the included in the analysis. A more recent meta-analysis by
placebo group. Coronary intervention procedures and other Sanguankeo et al. supported these latter results, finding
secondary endpoints did not differ significantly between that statin therapy was associated with a modestly slower
eGFR change per year as compared to placebo. They further atorvastatin, simvastatin, lovastatin, and pitavastatin, and
found that these results were driven by significant findings statin doses should be lowered in patients treated with
for high-intensity statins as no significant difference in cyclosporine and fluvastatin, pravastatin, and rosuvastatin.
change in eGFR was found for low or moderate intensity
statins. However, in this meta-analysis they found the data
were insufficient to study impacts on proteinuria. Other Statin Dose
studies have claimed differences in preserving renal func-
tion according to statin type or brand [146–148]. To date, Caution for adverse events in CKD patients such as muscle
the data on the effects of statin on progression of CKD and disorders may lead clinicians to prescribe lower dose therapy.
protein excretion are variable and modest. Additional data One post hoc analysis [152] in 2978 Japanese NDD-CKD
are needed to conclude whether statins provide renal pro- patients without history of ischemic heart disease or stroke
tection in patients with CKD. and moderate hypercholesterolemia evaluated the benefit of
low-dose statin therapy (pravastatin 10–20 mg daily) versus
placebo and found reduction in coronary heart disease
tatins and Adverse Outcome/
S events, all-cause mortality, and stroke. However, results of
Safety in CKD meta-analyses in CKD patients showed that benefit of statin
is related to the strength of LDL lowering and therefore
Statin safety may be of particular concern in CKD patients stronger therapy may still be indicated. Most clinical trial
who typically also have higher comorbidity and polyphar- data included in the above-mentioned meta-analyses [132] of
macy, and may have a higher rate of metabolite accumula- CKD patients included trials where patients received moder-
tion due to decreased renal function and uremia-related ate dose statin therapy versus placebo. The SHARP trial used
muscle injury that could increase risk of myopathy [149]. a combination of moderate dose statin plus ezetimibe, lead-
As mentioned above, results of the meta-analyses evaluating to more intense lipid-lowering therapy. To date, two post
ing the safety of statin therapy in CKD patients (both non- hoc analyses of randomized control trials have examined the
dialysis and dialysis) demonstrated no excess risk of adverse benefit of more aggressive lipid lowering with high-dose sta-
events such as elevated creatinine kinase, liver function tin in CKD patients. The Treating to New Targets (TNT) study
problems, muscle disorders, or cancer onset with statin [153] randomized patients to atorvastatin 80 mg/day (high
treatment. Palmer et al., however, noted lack of systematic dose statin) or 10 mg/day (moderate dose statin) and found
reporting may contribute to the inability to thoroughly eval- the major CV endpoint was reduced by 32% for 3107 CKD
uate adverse events. A recent statement from the ACC fur- (mostly stage 3) patients and reduced by 15% in patients with
ther supported findings of no excess risks and summarized normal eGFR (median follow-up of 5 years). Moreover, the
concerns of statin safety [150]. The report found risk of study reported no difference in safety concerns between the
events including hepatotoxicity, rhabdomyolysis, and new CKD and normal eGFR groups. In the Justification for the
onset of diabetes to be very low at 0.2% or less per year and Use of statins in Prevention; an Intervention Trial Evaluating
found no convincing evidence of risk of cancer, cataracts, Rosuvastatin (JUPITER) primary prevention trial [154],
cognitive dysfunction, peripheral neuropathy, erectile dys- high-dose statin (rosuvastatin 20 mg) was compared with pla-
function, or tendonitis with statin use. In CKD and in par- cebo among men and women free of CVD who had LDLc
ticular in both hemodialysis and peritoneal dialysis patients, <130 mg/dl and elevated high-sensitivity CRP. The post hoc
statins appeared to be safe even in consideration of the poly- analysis compared CV events in 3267 patients with moderate
pharmacy in these patients. Although there was no excess CKD at study entry and 14 528 with normal eGFR. After a
risk reported for myopathy, one randomized clinical trial median follow-up of 1.9 years, statin-treated CKD patients
found CKD was a risk factor for myopathy in patients had a 45% reduction in the primary endpoint and a 44%
treated with simvastatin 80 mg. The statin safety report reduction in all-cause mortality. Additionally, in CKD
therefore provides a table summarizing how to modulate patients there were no differences between statin therapy
statin therapy with CKD progression and maximum dose and placebo for adverse events. However, statin-treated CKD
per statin type in patients with mild to moderate CKD and patients had a statistically significant but not clinically differ-
ESRD to avoid myopathy risks. The report additionally ent higher HbA1c at 24 months, but no differences in fasting
addresses drug–drug interaction of cyclosporine and statin glucose were observed. Although results in CKD patients
in renal transplant patients. Cyclosporine has been shown were not specifically discussed, in another post hoc analysis
to increase blood levels of statins, which may increase risk of the JUPITER trial, Ridker et al. found that treatment with
of myopathy [151]. According to statin safety guidelines, statin therapy was associated with a fast time to diagnosis of
cyclosporine should be avoided in patients treated with diabetes [155].
Cholestero 559
Statins and Diabetes Risk bile acid binder in CKD patients and lowered LDLc as well
as increasing HDLc [171]. Colestilan, a phosphate binder
A number of meta-analyses have summarized statin trials used in Japan, was shown to be equivalent to sevelamer in
in evaluating risk of incident diabetes and found statin phosphate binding and LDLc lowering efficacy [172] and
treatment was significantly associated with an increased similar to simvastatin in LDLc lowering [173]. The benefit
risk [156–160]. The magnitude of risk may differ depend- of sevelamer in reducing adverse outcomes in CKD patients
ing on diabetes definition, statin dose [158, 159, 161] dura- has been demonstrated [174–176]. One small study com-
tion of exposure to statin therapy [162], and statin type [160, pared sevelamer versus calcium acetate in atorvastatin-
163] and may further be explained by bias in competing treated dialysis patients and found greater LDL lowering in
risk of mortality in placebo treated patients [164]. The sevelamer-treated patients, but no difference in coronary
mechanism of action by which statins increase the risk of artery calcium progression between treatment arms after
incident diabetes is not clear, and several mechanisms have 1 year [177]. Whether bile acid resins, including sevelamer,
been postulated [165–168]. To date, no study has directly confer CV benefit in CKD patients already on statin therapy
examined risk of diabetes incidence in statin-treated needs to be further evaluated.
patients with CKD. Authors of the cited studies [156–160]
agree that risk of diabetes incidence does not outweigh the Ezetimibe
benefits of CV risk protection and improved survival with
statin therapy. However, the overall safety of statins (in Ezetimibe is another cholesterol-lowering medication that
particular high-dose statin) in CKD patients still needs to acts similarly to bile acid resins by decreasing cholesterol
be addressed in future studies. absorption. When added to statin therapy, ezetimibe can
reduce cholesterol levels by an additional 23–24% on average.
The Improved Reduction of Outcomes: Vytorin Efficacy
Nonstatin Lipid Lowering in CKD International Trial (IMPROVE-IT) [178] evaluated the effect
of ezetimibe combined with simvastatin, as compared with
Although statin therapy is the main course of treatment for simvastatin alone, in 18 144 patients who had an acute coro-
dyslipidemia, a number of studies have evaluated the nary syndrome and whose LDL cholesterol values were
potential benefits of nonstatin lipid-lowering therapies in within guideline recommendations. Over a median follow-
CKD. The nonstatin lipid-lowering medications include up of 6 years, patients on the combined therapy had signifi-
those targeted at lowering cholesterol (bile acid binding cantly greater benefit of LDL lowering and lower risk of the
resins, ezetimibe, PCSK9 inhibitors) and those altering composite endpoint of nonfatal MI, unstable angina requir-
lowering triglyceride and/or elevating HDL levels (fibrates, ing rehospitalization, coronary revascularization ( 30 days
niacin, omega (n–3) fatty acids). after randomization), or nonfatal stroke. A post hoc analysis
of IMPROVE-IT evaluated associations across eGFR levels
and found patients with normal eGFR had a 12% risk reduc-
Cholesterol tion, while those with moderate CKD had a 13% risk reduc-
tion of the primary endpoint [179]. The analysis found no
Bile Acid Binding Resins
difference in change in mean eGFR over the 7 years of follow-
Bile acid binding resins, including cholestyramine, colestipol, up between study arms, yet also found no excess of adverse
and colesevelam, are another cholesterol-lowering medica- risk in the combined therapy patients. The results of this post
tion often given to statin-intolerant patients. They are less hoc analysis combined with the SHARP trial results demon-
potent than statins at decreasing TC and LDLc, and have strate the benefits of the combined statin plus ezetimibe ther-
been associated with an increase in TG level [169]. The Lipid apy without any concerns of added risk in CKD.
Research Clinics Coronary Primary Prevention Trial (LRC-
CPPT) randomized 3806 men free of coronary artery disease
PCSK9 Inhibitors
with elevated LDLc (>175 mg/dl) to cholestyramine with an
intended dose of 24 g/day versus placebo [170]. After 7.1 years PCSK9 inhibitors have shown remarkable results in LDL
there was a significant 19% reduction in the incidence of non- lowering and CV benefits [180, 181].
fatal MI or coronary death. The average serum creatinine in The Long-term Safety and Tolerability of Alirocumab in
enrolled patients was 1.03 mg/dl and effect modification by High Cardiovascular Risk Patients with Hypercholesterolemia
CKD was not conducted. Not Adequately Controlled with Their Modifying Therapy
The phosphate binder, sevelamer, used in CKD patients (ODYSSEY LONG TERM) study randomized patients with
to lower serum phosphorous levels was shown to act as a high risk of CV events, LDLc 70 mg/dl, and receiving
t reatment with statins at the maximum tolerated dose to e volocumab versus placebo across baseline CKD
alirocumab versus placebo [181]. At week 24, the mean per- stages [185]. The analysis showed that the FOURIER trial
centage change in LDLc from baseline for alirocumab versus was composed of 29%, 55%, and 16% of patients with eGFR
placebo was 62%, and the treatment effect remained over the 90, 60–89, and < 60 ml/min/1.73m2, respectively. LDLc
78 weeks of follow-up. A post hoc analysis found that the rate reduction with evolocumab versus placebo at 48 weeks, rel-
of CV events (death from coronary heart disease, nonfatal ative risk of the primary endpoint, and adverse events
MI, fatal or nonfatal ischemic stroke, or unstable angina including eGFR decline was similar across CKD groups.
requiring hospitalization) was 48% lower in alirocumab- They also found that absolute relative risks for the second-
treated patients versus placebo. Patients with eGFR <60 ml/ ary endpoint (composite of CV death, MI, or stroke) were
min/1.73m2 for 3 months or more at the screening visit were larger among those with eGFR <60 ml/min/1.73m2, as
included in the study as having a high risk of CV events; how- compared to patients with eGFR 90 ml/min/1.73m2.
ever, patients with eGFR <30 ml/min/1.73m2 were excluded. PCSK9 inhibitors may have a great future in treating dys-
A larger trial called ODYSSEY OUTCOMES randomized lipidemia in patients with CKD, although concerns have
18 924 patients aged 40 years or older to alirocumab versus been raised regarding the affordability of this drug. Further
placebo if they had acute coronary syndrome (acute MI or studies are needed, in particular in dialysis and renal trans-
unstable angina) within 12 months prior to randomization, plant patients.
were on high-intensity statin, and had an LDLc 70 mg/dl.
Patients with eGFR <30 ml/min/1.73m2 were again excluded.
At 4 months, alirocumab patients had a 56% greater LDLc Triglyceride/HDL
reduction, which was maintained after the remaining course
Fibrates
of follow-up. Over a median follow-up of 2.8 years,
alirocumab-treated patients had a 15% lower risk of the pri- CKD patients with elevated TG may need additional or
mary endpoint (coronary heart disease death, nonfatal MI, other lipid-lowering therapy as statins have been shown to
ischemic stroke, or unstable angina requiring hospitaliza- only modestly reduce TG. Fibrates (including gemfibrozil
tion). Patients also had a lower risk of all-cause death, but not and fenofibrate) have been used in clinical practice to
CVD or coronary heart disease death. The results of this trial decrease TG levels, modestly improve HDL, and reduce CV
were presented at the 2018 American Cardiology events. The Helsinki Heart Study [186] and the Veterans
Conference [182]. A post hoc analysis included eight of the Affairs High-Density Lipoprotein Cholesterol Intervention
earlier phase 3 ODYSSEY trials and found that among 4629 Trial (VA-HIT) [187] demonstrated a reduction in CV
hypercholesterolemic patients (including 10% with impaired events (fatal and nonfatal MI, and nonfatal MI or death
renal function), LDLc lowering and safety were comparable from coronary causes, respectively) in fibrate-treated
in CKD and non-CKD patients [183]. To date, information patients compared to placebo. However, in the larger
comparing clinical outcomes across CKD stages have not Fenofibrate Intervention and Event Lowering in Diabetes
been presented for alirocumab. (FIELD) trial, which included 9795 patients, fenofibrate
The Further Cardiovascular Outcomes Research with did not reduce the primary endpoint (coronary heart dis-
PCSK9 Inhibition in Subjects with Elevated Risk ease death or nonfatal MI) but did significantly reduce
(FOURIER) trial then tested the impact of evolocumab nonfatal MI by 24%, total CV disease events by 11%, and
when added to high-intensity or moderate-intensity statin coronary revascularizations by 21% [188]. The authors of
therapy in 27 564 patients with clinically evident ASCVD the trial speculated that higher rates of off-protocol initia-
and LDL 70 mg/dl [180]. At 48 weeks follow-up, the mean tion of statin therapy in placebo-allocated patients may
percentage reduction in LDLc with evolocumab, as com- have masked a moderately larger treatment benefit. A
pared with placebo, was 59% (95% CI: 0.58–0.60), with 87%, meta-analysis addressed data concerning patients with
67%, and 42% of patients in the evolocumab-treated group CKD enrolled in VA-HIT and FIELD as well as other stud-
achieving LDLc <70, <40, and <25 mg/dl, respectively. ies (10 studies, 16 869 CKD patients) [189] and found
After a median follow-up of 2.2 years, evolocumab versus fibrates improved lipid profiles and reduced major CV
placebo reduced the risk of the primary composite endpoint events and CV deaths but not all-cause mortality and stroke
(CV death, MI, stroke, hospitalization for unstable angina, in patients with mild to moderate CKD. The analysis also
or coronary revascularization) by 15%. Patients with eGFR reported fibrates reduced albuminuria and reversibly
20 ml/min/1.73m2 were included in the study [184]. At the reduced serum creatinine; however, it found no clear effect
2018 American Society of Nephrology Kidney Week of preventing progression to ESRD.
Conference, Charytan et al. presented the results of a post In clinical practice, fibrates are typically recommended
hoc analysis comparing the efficacy and safety of in addition to statin therapy. The Action to Control CV Risk
Triglyceride/HD 561
in Diabetes (ACCORD) study evaluated the added benefit (>200 mg/dl) and lower HDL (<32 mg/dl) that had a
of fenofibrate versus placebo to statin in 5518 type 2 significant decrease in primary events, suggesting more
diabetic patients being treated with open-label simvastatin extreme thresholds for enrollment should be used for nia-
and found no benefit for fibrates in reducing CV out- cin therapy [199]. Another post hoc analysis evaluated the
comes [190]. In a post hoc analysis, which included 1910 efficacy of niacin in patients with stage 3 CKD and found
patients with mild to moderate CKD and 3554 with normal no benefit in CV outcomes and a higher all-cause mortality
eGFR, fenofibrate conferred no benefit in patients with risk in niacin-treated patients [200].
CKD for primary or secondary outcomes [191]. However, The Treatment of HDL to Reduce the Incidence of
in the non-CKD patients alone, the fenofibrate patients Vascular Events (HPS-THRIVE) trial enrolled 25 673 statin-
had a 36% reduction of CV mortality and a 44% lower rate treated patients with ASCVD and randomized them to nia-
of fatal and nonfatal congestive heart failure. Some studies cin/laropiprant versus placebo [201, 202]. The study failed
have cautioned that the combined therapy may lead to to show any benefit of the niacin therapy in reducing major
safety concerns, such as rhabdomyolysis [192], but phar- CV events. Moreover, patients in the niacin therapy group
macokinetic studies failed to show any changes in statin had a higher risk of adverse events, including disturbances
kinetic parameters when fenofibrate was tested against in diabetes control, diabetes incidence, and adverse events
each statin separately [193–195]. Current Kidney Disease: related to gastrointestinal and musculo-skeletal systems,
Improving Global Outcomes (KDIGO) guidelines suggest skin, infection, and bleeding. Of note, the study included
that combined statin and fibrate therapy should not be patients with a wide range of lipid levels who may not be
used in patients with renal impairment due to concerns of indicated for niacin therapy, including over 49% of trial
adverse events and statin should be prescribed if trying to patients with an HDL 43 mg/dl and 74% with TG levels of
choose between the two medication classes [196]. However, <151 mg/dl.
if indicated, Canadian guidelines suggest that fibrates Although a small pharmacokinetic study demonstrated
alone may be used in patients with mild CKD at lower that niacin may not need dose adjustments in patients with
doses (one-third of the normal dose) [197]. Additional CKD [203], current KDIGO guidelines have stipulated that
studies may be needed to further evaluate whether the nicotinic acid has not been well enough studied in CKD,
potential benefit of fibrate added to statin therapy in CKD and given the risks of toxicity, it is currently not recom-
patients outweighs any potential risks. mended for treatment of high TG [196]. Two review arti-
cles have summarized the beneficial properties of niacin,
including anti-inflammatory properties, reducing serum
Niacin
phosphorous levels, and preventing HDL dysfunctional-
Similar to fibrates, niacin is used to increase HDLc and ity [204, 205]. This suggests potential benefit of niacin ther-
reduce TG concentration. Early trials demonstrated that apy in patients with CKD, although further studies are
niacin added to statin monotherapy resulted in a reduction needed.
of carotid intima thickness, a marker of atherosclerosis.
However, clinical trials failed to show CV benefit and
Omega (n-3) Fatty Acids
introduced concerns regarding niacin safety. The
Atherothrombosis Intervention in Metabolic Syndrome Omega-3 fatty acids have also been used to target lowering
with Low HDL/High Triglycerides and Impact on Global of TG levels. The active ingredients include eicosapentanoic
Health Outcomes (AIM-HIGH) study enrolled 3414 acid (EPA) and docosahexaenoic acid (DHA). In addition
patients with coronary artery disease (men below 40 mg/dl to dietary recommendations to consume fish rich in
and women below 50 mg/dl) and high TG (over 150 mg/dl) omega-3 fatty acids, there are two prescription strength
with LDLc lowered to 40–80 mg/dl with simvastatin and medications on the US market: Lovaza, containing 375 mg
ezetimibe, and randomized them to additional niacin or DHA and 465 mg EPA, and Vascepa, containing 1 g purified
placebo therapy [198]. The primary endpoint evaluated EPA. A study found that the EPA-only Vascepa decreased
was the first event of the composite of death from coronary LDLc, while the DHA-containing Lovaza increased it [206].
heart disease, nonfatal MI, ischemic stroke, hospitalization Clinical trials of omega-3 fatty acids for CV endpoints
for an acute coronary syndrome, or symptom-driven have yielded mixed results. The Gruppo Italiano per lo
coronary or cerebral revascularization. After 36 months of Studio della Sopravvivenza nell’Infarto Miocardico-
follow-up, despite lipid improvement, there was no clinical Prevenzione (GISSI-Prevenzione) study enrolled 11 324
benefit in reducing the primary endpoint from the addition survivors of MI and randomized them to Lovaza 1 g/day
of niacin to statin therapy. However, a post hoc analysis versus placebo [207, 208]. After 3.5 years, there was a sig-
identified a subgroup of 439 patients with elevated TG nificant 15% reduction in the combined endpoint of death,
nonfatal MI, and stroke. Similar results were observed in a with possible combination with ezetimibe [3]. For patients
study of over 6000 heart failure patients [209]. However, transitioning to dialysis, continuation of statin therapy is
additional studies that examined the clinical benefit of recommended. However, because the trials failed to show
omega 3-fatty acids as an add-on therapy to statins failed benefit of statin initiation in dialysis patients, initiation of
to show clinical benefit in patients with a history of MI statin in dialysis is not recommended. This is an update
[210–212]. The Japan EPA Lipid Intervention Study (JELIS) from previous guidelines, which did not have a recommen-
randomized 18 645 patients with or without pre-existing dation regarding initiating or continuing statins in hemodi-
coronary artery disease and a serum cholesterol >250 mg/ alysis patients [222]. Similar to the current ACC/AHA
dl to statin plus either 1800 mg EPA or placebo [213]. After guidelines, KDIGO guidelines recommend initiation of sta-
4.6 years, there was a significant reduction in the risk of tin treatment with a statin or statin/ezetimibe combination
major coronary events and nonfatal coronary events (19% in adults aged 50 years with eGFR <60 ml/min/1.73m2, but
each), a 28% decrease in the risk of unstable angina, and a not treated with chronic dialysis or kidney transplantation.
20% reduction in stroke in treated patients [214, 215]. Renal Statin initiation is also recommended in NDD-CKD non-
patients were excluded from the study. A more recent trial transplanted adult patients aged 18–49 years if they have
randomized 8179 statin-treated patients with established known coronary disease, diabetes mellitus, prior ischemic
CVD, diabetes, or other risk factors and a fasting TG of stroke of >10% estimated 10-year risk of coronary death, or
135–499 mg/dl and LDLc of 41–100 mg/dl to Vascepa or nonfatal MI. Similar to the ACC/AHA guidelines, KDIGO
placebo. The study found that EPA-treated patients had a guidelines also suggest statins not be initiated, but should be
25% lower risk of the primary endpoint (CV death, nonfatal continued in dialysis-dependent CKD. They also suggest
MI, nonfatal stroke, coronary revascularization, or unstable treatment with statin in adult kidney transplant recipients.
angina) [216]. Patients with CKD, except those requiring Current ACC/AHA guidelines do not have suggestions for
dialysis, were included in the trial and stratification by renal transplant recipients. For statin therapy dosing in
eGFR showed no heterogeneity of effect. CKD patients, where statins are recommended, KDIGO
Hemodialysis patients have markedly lower levels of guidelines provide a table which has dosing recommenda-
plasma EPA and DHA when compared with NDD-CKD tions per statin type that corresponds with ACC/AHA
patients [217]. DHA has been shown to be an independent guidelines of moderate statin doses. The ACC/AHA guide-
predictor of all-cause mortality [218] and of sudden car- lines indicate there is a need for randomized control trials
diac death [219] in hemodialysis patients. A small trial testing the efficacy of higher dose versus moderate dose
randomized 206 hemodialysis patients to low-dose omega- statins in patients with risk-enhancing factors, which
3 fatty acids versus placebo, and found a small but signifi- includes CKD. A recent cohort study in US veterans demon-
cant reduction in the number of MIs (four versus 13, strated no difference in the association of statin dose with
P = 0.036) in the treated versus placebo arms [220]. The mortality outcomes across stages of NDD-CKD [223].
Fish Oil Inhibition of Stenosis in Hemodialysis Grafts For TG lowering, the KDIGO guidelines advise therapeu-
(FISH) study enrolled 201 patients and randomized them tic lifestyle changes, and caution against the use of statins
to fish oil (four 1 g pills per day) versus placebo for in combination with fibrates in CKD patients due to
12 months. In the fish oil group, there were half as many increased risk of adverse events. The 2013 ACC/AHA
thromboses, fewer corrective interventions, and improved guidelines state that gemfibrozil should not be used in
CV event-free survival [221]. combination with statin, and for TG lowering a fenofibrate
Although the benefits of omega-3 fatty acids are promis- can be prescribed with dose alterations according to CKD
ing in CKD and dialysis patients, further studies in larger stage: not to exceed 54 mg/day in patients with eGFR
trials are needed. 30–59 ml/min/1.73m2 and to discontinue when eGFR is
persistently <30 ml/min/1.73m2. Although there is cur-
rently a lack of data to support any guideline recommenda-
G
uidelines tion, niacin and omega-3 fatty acids may be used in place of
fibrates, with monitoring of potential adverse reactions;
Guidelines advising on dyslipidemia slightly differ whether however, further studies in CKD patients are needed.
approaching from a nephrology angle using KDIGO guide-
lines [196] or a cardiology angle using ACC/AHA guidelines
(Table 34.1). According to the 2018 ACC/AHA guidelines, Conclusion
CKD with an eGFR <60 ml/min/1.73m2 is considered a risk-
enhancing factor, and thereby adults aged 40–75 years with The lipoprotein profile in CKD patients is variable depend-
LDLc 70–189 mg/dl who are at 10-year ASCVD risk of 7.5% ing on CKD stage, other patient characteristics including
or higher should be treated with a moderate intensity statin nutritional status, and management of the patient including
Table 34.1 Recommendations for treatment of dyslipidemia.
Treatment Recommendation Certainty of evidence Strength of recommendation
Statins for NDD-CKD All NDD-CKD patients aged 50–80 years should be treated with a High for total mortality, CV Strong.
patients, age <80 years high-dose statin or a moderate dose statin ± ezetimibe irrespective of mortality, and MACE Level A Class I.
serum cholesterol. Younger patients should be treated if they have
documented ASCVD, diabetes, or estimated 10-year risk of coronary
death or nonfatal MI >10%.
Statins for hemodialysis Statins should not be initiated in hemodialysis patients. Statins should High for noninitiation; low for Strong for noninitiation. Level A Class
patients be continued in transition to hemodialysis for predialysis patients. continuation of statins III, no benefit.
Weak for continuation. Level B Class IIa.
Statins for peritoneal Statins may be initiated in peritoneal dialysis patients when Weak and only for total mortality Weak for total mortality. Level C
dialysis patients cholesterol is elevated (>225 mg/dl). Class IIb.
Statins for renal Statins should be initiated in renal transplant patients, although there Weak for initiation and only for Weak for initiation. Level C Class IIa.
transplant patients are increased risks due to drug interactions. MACE; high for drug interaction
Statins for patients age Statin initiation or continuation could be considered if the patient or Weak for initiation or continuation Weak evidence. Level C Class IIa.
>80 years family agrees and the long-term benefit exceeds the risks. Informed
consent discussion is necessary.
Fibrates Fibrates are considered not effective as an add-on to statins in CKD Weak for use with triglycerides Weak evidence.
patients unless triglycerides are >500 mg/dl. >500 mg/dl Level C Class IIb.
Niacin Niacin should not be used for treatment of dyslipidemia in CKD Moderate for mortality Moderate evidence. Level C Class III no
patients. benefit/harm.
Omega-3 fatty acids EPA should be used in NDD-CKD patients with stage 3 CKD in Moderate for total and CV Moderate evidence. Level B Class I.
addition to statin therapy if triglycerides are >135 mg/dl mortality and MACE
Ezetimibe Ezetimibe should be added to statin therapy in CKD patients if the High for CV mortality and MACE Strong evidence. Level A Class I.
LDL cholesterol was not decreased by 50% and/or if LDL cholesterol is
not at goal.
Bile acid binding resins Bile acid binding resins should not be used for dyslipidemia in CKD Weak Low evidence. Class C Level III possible
patients. harm.
PCSK9 inhibitors PCSK9 inhibitors should be used in patients with stage 3 CKD and Strong for CV mortality and MACE Strong evidence. Class A Level I.
ASCVD who have a LDL cholesterol >70 mg/dl on maximum tolerated
dose of statin + ezetimibe.
ASCVD, atherosclerotic cardiovascular disease; CKD, chronic kidney disease; CV, cardiovascular; EPA, eicosapentanoic acid; LDL, low-density lipoprotein; MACE, major adverse
cardiovascular event; MI, myocardial infarction; NDD-CKD, nondialysis dependent chronic kidney disease; PCSK9, proprotein convertase subtilisin/kexin type 9.
0005152422.INDD 563 09-12-2022 16:05:51

type of dialysis and drugs aimed at prevention of graft rejec- ued after patients transition to dialysis. The clinical benefit
tion. Consequently, the ability of the different lipoproteins to of statin currently outweighs the risk of adverse events in
predict adverse outcomes is limited. Interpretation of the CKD patients. Further study is needed to ascertain the effi-
levels should be individualized. Moderate dose statins cacy and possible risk of higher dose statins or nonstatin
should be initiated in patients with NDD-CKD and contin- lipid-altering medications in CKD patients.
Acknowledgement
We would like to acknowledge Yousif Arif for his assis-

tance in proofing our chapter.
References
1 Saran, R., Robinson, B., Abbott, K.C. et al. (2018). US 10 Navaneethan, S.D., Schold, J.D., Arrigain, S. et al. (2012).
renal data system 2017 annual data report: Epidemiology Serum triglycerides and risk for death in stage 3 and stage
of Kidney Disease in the United States. Am. J. Kidney Dis. 4 chronic kidney disease. Nephrol. Dial. Transplant. 27
71 (3S1): A7. (8): 3228–3234.
2 Expert Panel on Detection E, Treatment of High Blood 11 Muntner, P., He, J., Astor, B.C. et al. (2005). Traditional
Cholesterol in A (2001). Executive summary of the third and nontraditional risk factors predict coronary heart
report of the National Cholesterol Education Program disease in chronic kidney disease: results from the
(NCEP) expert panel on detection, evaluation, and atherosclerosis risk in communities study. J. Am. Soc.
treatment of high blood cholesterol in adults (adult Nephrol. 16 (2): 529–538.
treatment panel III). JAMA 285 (19): 2486–2497. 12 Chawla, V., Greene, T., Beck, G.J. et al. (2010).
3 Grundy, S.M., Stone, N.J., Bailey, A.L. et al. (2018). AHA/ Hyperlipidemia and long-term outcomes in nondiabetic
ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ chronic kidney disease. Clin. J. Am. Soc. Nephrol. 5 (9):
ASPC/NLA/PCNA guideline on the Management of 1582–1587.
Blood Cholesterol: a report of the American College of 13 Kovesdy, C.P., Anderson, J.E., and Kalantar-Zadeh, K.
Cardiology/American Heart Association Task Force on (2007). Inverse association between lipid levels and
Clinical Practice Guidelines. J. Am. Coll. Cardiol.: 2019 mortality in men with chronic kidney disease who are
Jun 25;73(24):e285–e350. not yet on dialysis: effects of case mix and the
4 Wong, N.D. and Levy, D. (2013). Legacy of the malnutrition-inflammation-cachexia syndrome. J. Am.
Framingham heart study: rationale, design, initial Soc. Nephrol. 18 (1): 304–311.
findings, and implications. Glob. Heart 8 (1): 3–9. 14 Contreras, G., Hu, B., Astor, B.C. et al. (2010).
5 Multiple risk factor intervention trial research group Malnutrition-inflammation modifies the relationship of
(1982). Risk factor changes and mortality results. cholesterol with cardiovascular disease. J. Am. Soc.
Multiple Risk Factor Intervention Trial Research Group. Nephrol. 21 (12): 2131–2142.
JAMA 248 (12): 1465–1477. 15 Kalantar-Zadeh, K., Ikizler, T.A., Block, G. et al. (2003).
6 Emerging Risk Factors, C., Di Angelantonio, E., Sarwar, Malnutrition-inflammation complex syndrome in dialysis
N. et al. (2009). Major lipids, apolipoproteins, and risk of patients: causes and consequences. Am. J. Kidney Dis. 42
vascular disease. JAMA 302 (18): 1993–2000. (5): 864–881.
7 Tonelli, M., Muntner, P., Lloyd, A. et al. (2013). 16 Kalantar-Zadeh, K., Kopple, J.D., Block, G., and
Association between LDL-C and risk of myocardial Humphreys, M.H. (2001). A malnutrition-inflammation
infarction in CKD. J. Am. Soc. Nephrol. 24 (6): score is correlated with morbidity and mortality in
979–986. maintenance hemodialysis patients. Am. J. Kidney Dis. 38
8 Bowe, B., Xie, Y., Xian, H. et al. (2016). High density (6): 1251–1263.
lipoprotein cholesterol and the Risk of all-cause mortality 17 Rambod, M., Bross, R., Zitterkoph, J. et al. (2009).
among U.S. veterans. Clin. J. Am. Soc. Nephrol. 11 (10): Association of Malnutrition-Inflammation Score with quality
1784–1793. of life and mortality in hemodialysis patients: a 5-year
9 Navaneethan, S.D., Schold, J.D., Walther, C.P. et al. prospective cohort study. Am. J. Kidney Dis. 53 (2): 298–309.
(2018). High-density lipoprotein cholesterol and causes of 18 Bowden, R.G. and Wilson, R.L. (2010). Malnutrition,
death in chronic kidney disease. J. Clin. Lipidol. 12 (4): inflammation, and lipids in a cohort of dialysis patients.
1061–1071, e7. Postgrad. Med. 122 (3): 196–202.
Reference 565
9 Liu, Y., Coresh, J., Eustace, J.A. et al. (2004). Association

1 continuous ambulatory peritoneal dialysis. Nephrol. Dial.
between cholesterol level and mortality in dialysis Transplant. 25 (2): 599–604.
patients: role of inflammation and malnutrition. JAMA 32 Xie, X., Zhang, X., Xiang, S. et al. (2017). Association of
291 (4): 451–459. very low-density lipoprotein cholesterol with all-cause
20 Iseki, K., Yamazato, M., Tozawa, M., and Takishita, S. and cardiovascular mortality in peritoneal dialysis.
(2002). Hypocholesterolemia is a significant predictor of Kidney Blood Press. Res. 42 (1): 52–61.
death in a cohort of chronic hemodialysis patients. 33 Schaeffner, E.S., Fodinger, M., Kramar, R. et al. (2006).
Kidney Int. 61 (5): 1887–1893. Prognostic associations between lipid markers and
21 Lin, Y.C., Lin, Y.C., Chen, H.H. et al. (2016). Different outcomes in kidney transplant recipients. Am. J. Kidney
effect of hypercholesterolemia on mortality in Dis. 47 (3): 509–517.
hemodialysis patients based on coronary artery disease or 34 Eckel, R.H., Grundy, S.M., and Zimmet, P.Z. (2005). The
myocardial infarction. Lipids Health Dis. 15 (1): 211. metabolic syndrome. Lancet 365 (9468): 1415–1428.
22 Postorino, M., Marino, C., Tripepi, G. et al. (2011). 35 Kurella, M., Lo, J.C., and Chertow, G.M. (2005).
Abdominal obesity modifies the risk of Metabolic syndrome and the risk for chronic kidney
hypertriglyceridemia for all-cause and cardiovascular disease among nondiabetic adults. J. Am. Soc. Nephrol.
mortality in hemodialysis patients. Kidney Int. 79 (7): 16 (7): 2134–2140.
765–772. 36 Townsend, R.R., Anderson, A.H., Chen, J. et al. (2011).
23 Kilpatrick, R.D., McAllister, C.J., Kovesdy, C.P. et al. Metabolic syndrome, components, and cardiovascular
(2007). Association between serum lipids and survival in disease prevalence in chronic kidney disease: findings
hemodialysis patients and impact of race. J. Am. Soc. from the Chronic Renal Insufficiency Cohort (CRIC)
Nephrol. 18 (1): 293–303. study. Am. J. Nephrol. 33 (6): 477–484.
24 Tsirpanlis, G., Boufidou, F., Zoga, M. et al. (2009). Low 37 Thomas, G., Sehgal, A.R., Kashyap, S.R. et al. (2011).
cholesterol along with inflammation predicts morbidity Metabolic syndrome and kidney disease: a systematic
and mortality in hemodialysis patients. Hemodial. Int. 13 review and meta-analysis. Clin. J. Am. Soc. Nephrol. 6
(2): 197–204. (10): 2364–2373.
25 Chang, T.I., Streja, E., Ko, G.J. et al. (2018). Inverse 38 Bae, J.C., Han, J.M., Kwon, S. et al. (2016). LDL-C/apoB
association between serum non-high-density lipoprotein and HDL-C/apoA-1 ratios predict incident chronic
cholesterol levels and mortality in patients undergoing kidney disease in a large apparently healthy cohort.
incident hemodialysis. J. Am. Heart Assoc. 7 (12): Atherosclerosis 251: 170–176.
e009096. 39 Zoppini, G., Negri, C., Stoico, V. et al. (2012). Triglyceride-
26 Chang, T.I., Streja, E., Soohoo, M. et al. (2017). high-density lipoprotein cholesterol is associated with
Association of Serum Triglyceride to HDL cholesterol microvascular complications in type 2 diabetes mellitus.
ratio with all-cause and cardiovascular mortality in Metabolism 61 (1): 22–29.
incident hemodialysis patients. Clin. J. Am. Soc. Nephrol. 40 Tsuruya, K., Yoshida, H., Nagata, M. et al. (2015). Impact
12 (4): 591–602. of the triglycerides to high-density lipoprotein cholesterol
27 Moradi, H., Streja, E., Kashyap, M.L. et al. (2014). ratio on the incidence and progression of CKD: A
Elevated high-density lipoprotein cholesterol and longitudinal study in a large Japanese population. Am. J.
cardiovascular mortality in maintenance hemodialysis Kidney Dis. 66 (6): 972–983.
patients. Nephrol. Dial. Transplant. 29 (8): 1554–1562. 41 Keane, W.F., Tomassini, J.E., and Neff, D.R. (2013). Lipid
28 Chang, T.I., Streja, E., Soohoo, M. et al. (2018). Increments abnormalities in patients with chronic kidney disease:
in serum high-density lipoprotein cholesterol over time implications for the pathophysiology of atherosclerosis. J.
are not associated with improved outcomes in incident Atheroscler. Thromb. 20 (2): 123–133.
hemodialysis patients. J. Clin. Lipidol. 12 (2): 488–497. 42 Chan, D.C., Barrett, P.H., and Watts, G.F. (2004). Lipoprotein
29 Moradi, H., Abhari, P., Streja, E. et al. (2015). Association transport in the metabolic syndrome: pathophysiological
of serum lipids with outcomes in Hispanic hemodialysis and interventional studies employing stable isotopy and
patients of the West versus East Coasts of the United modelling methods. Clin. Sci. 107 (3): 233–249.
States. Am. J. Nephrol. 41 (4–5): 284–295. 43 Ginsberg, H.N., Zhang, Y.L., and Hernandez-Ono, A.
30 Habib, A.N., Baird, B.C., Leypoldt, J.K. et al. (2006). The (2005). Regulation of plasma triglycerides in insulin
association of lipid levels with mortality in patients on resistance and diabetes. Arch. Med. Res. 36 (3):
chronic peritoneal dialysis. Nephrol. Dial. Transplant. 21 232–240.
(10): 2881–2892. 44 Kaysen, G.A. (2006). Metabolic syndrome and renal
31 Park, J.T., Chang, T.I., Kim, D.K. et al. (2010). Metabolic failure: similarities and differences. Panminerva Med. 48
syndrome predicts mortality in non-diabetic patients on (3): 151–164.
5 Wang, X., Belani, S., Coyne, D.W. et al. (2012). Very low
4 59 Stegmayr, B. (2017). Dialysis procedures alter metabolic
density lipoprotein metabolism in patients with chronic conditions. Nutrients 9 (6): 548.
kidney disease. Cardiorenal Med. 2 (1): 57–65. 60 Mekki, K., Prost, J., Remaoun, M. et al. (2009). Long
46 Wang, Y., Qiu, X., Lv, L. et al. (2016). Correlation between term hemodialysis aggravates lipolytic activity reduction
serum lipid levels and measured glomerular filtration rate and very low density, low density lipoproteins
in Chinese patients with chronic kidney disease. PLoS composition in chronic renal failure patients. BMC
One 11 (10): e0163767. Cardiovasc. Disord. 9: 41.
47 Kittiskulnam, P., Thokanit, N.S., Katavetin, P. et al. 61 Barbagallo, C.M., Noto, D., Cefalu, A.B. et al. (2015).
(2018). The magnitude of obesity and metabolic Heparin induces an accumulation of atherogenic
syndrome among diabetic chronic kidney disease lipoproteins during hemodialysis in normolipidemic
population: a nationwide study. PLoS One 13 (5): end-stage renal disease patients. Hemodial. Int. 19
e0196332. (3): 360–367.
48 Soohoo, M., Moradi, H., Kovesdy, C.P. et al. (2019). Serum 62 Horkko, S., Huttunen, K., Korhonen, T., and Kesaniemi,
Triglycerides and Mortality Risk across Kidney Disease Y.A. (1994). Decreased clearance of low-density
Stage. J. Clin. Lipidol. 13 (5): 744–753.e15. lipoprotein in patients with chronic renal failure. Kidney
49 Chan, D.T., Dogra, G.K., Irish, A.B. et al. (2009). Chronic Int. 45 (2): 561–570.
kidney disease delays VLDL-apoB-100 particle 63 Ikewaki, K., Schaefer, J.R., Frischmann, M.E. et al. (2005).
catabolism: potential role of apolipoprotein C-III. J. Lipid Delayed in vivo catabolism of intermediate-density
Res. 50 (12): 2524–2531. lipoprotein and low-density lipoprotein in hemodialysis
50 Ooi, E.M., Chan, D.T., Watts, G.F. et al. (2011). Plasma patients as potential cause of premature atherosclerosis.
apolipoprotein C-III metabolism in patients with chronic Arterioscler. Thromb. Vasc. Biol. 25 (12): 2615–2622.
kidney disease. J. Lipid Res. 52 (4): 794–800. 64 Austin, M.A., Breslow, J.L., Hennekens, C.H. et al. (1988).
51 Arnadottir, M., Thysell, H., Dallongeville, J. et al. (1995). Low-density lipoprotein subclass patterns and risk of
Evidence that reduced lipoprotein lipase activity is not a myocardial infarction. JAMA 260 (13): 1917–1921.
primary pathogenetic factor for hypertriglyceridemia in 65 Rajman, I., Harper, L., McPake, D. et al. (1998).
renal failure. Kidney Int. 48 (3): 779–784. Low-density lipoprotein subfraction profiles in chronic
52 Wakabayashi, Y., Okubo, M., Shimada, H. et al. (1987). renal failure. Nephrol. Dial. Transplant. 13 (9):
Decreased VLDL apoprotein CII/apoprotein CIII ratio 2281–2287.
may be seen in both normotriglyceridemic and 66 Filler, G., Taheri, S., McIntyre, C. et al. (2018). Chronic
hypertriglyceridemic patients on chronic hemodialysis kidney disease stage affects small, dense low-density
treatment. Metabolism 36 (9): 815–820. lipoprotein but not glycated low-density lipoprotein in
53 Murase, T., Cattran, D.C., Rubenstein, B., and Steiner, G. younger chronic kidney disease patients: a cross-sectional
(1975). Inhibition of lipoprotein lipase by uremic plasma, study. Clin. Kidney J. 11 (3): 383–388.
a possible cause of hypertriglyceridemia. Metabolism 24 67 Homma, K., Homma, Y., Shiina, Y. et al. (2013). Skew of
(11): 1279–1286. plasma low-and high-density lipoprotein distributions to
54 Cheung, A.K., Parker, C.J., Ren, K., and Iverius, P.H. less dense subfractions in normotriglyceridemic chronic
(1996). Increased lipase inhibition in uremia: kidney disease patients on maintenance hemodialysis
identification of pre-beta-HDL as a major inhibitor in treatment. Nephron Clin. Pract. 123 (1–2): 41–45.
normal and uremic plasma. Kidney Int. 49 (5): 68 Bowden, R.G., Wilson, R.L., and Beaujean, A.A. (2011).
1360–1371. LDL particle size and number compared with LDL
55 Cwiklinska, A., Cackowska, M., Wieczorek, E. et al. (2018). cholesterol and risk categorization in end-stage renal
Progression of chronic kidney disease affects HDL impact disease patients. J. Nephrol. 24 (6): 771–777.
on lipoprotein lipase (LPL)-mediated VLDL lipolysis 69 Himmelfarb, J., Stenvinkel, P., Ikizler, T.A., and Hakim,
efficiency. Kidney Blood Press. Res. 43 (3): 970–978. R.M. (2002). The elephant in uremia: oxidant stress as a
56 Nestel, P.J., Fidge, N.H., and Tan, M.H. (1982). Increased unifying concept of cardiovascular disease in uremia.
lipoprotein-remnant formation in chronic renal failure. Kidney Int. 62 (5): 1524–1538.
N. Engl. J. Med. 307 (6): 329–333. 70 Mahrooz, A., Zargari, M., Sedighi, O. et al. (2012).
57 Weintraub, M., Burstein, A., Rassin, T. et al. (1992). Severe Increased oxidized-LDL levels and arylesterase activity/
defect in clearing postprandial chylomicron remnants in HDL ratio in ESRD patients treated with hemodialysis.
dialysis patients. Kidney Int. 42 (5): 1247–1252. Clin. Invest. Med. 35 (3): E144–E151.
58 Baranowski, T., Kralisch, S., Bachmann, A. et al. (2011). 71 Maggi, E., Bellazzi, R., Gazo, A. et al. (1994).
Serum levels of the adipokine fasting-induced adipose Autoantibodies against oxidatively-modified LDL in
factor/angiopoietin-like protein 4 depend on renal uremic patients undergoing dialysis. Kidney Int. 46 (3):
function. Horm. Metab. Res. 43 (2): 117–120. 869–876.
Reference 567
2 Kronenberg, F., Kuen, E., Ritz, E. et al. (2000).

7 85 Jin, K., Park, B.S., Kim, Y.W., and Vaziri, N.D. (2014).
Lipoprotein(a) serum concentrations and Plasma PCSK9 in nephrotic syndrome and in peritoneal
apolipoprotein(a) phenotypes in mild and moderate renal dialysis: a cross-sectional study. Am. J. Kidney Dis. 63 (4):
failure. J. Am. Soc. Nephrol. 11 (1): 105–115. 584–589.
73 Hopewell, J.C., Haynes, R., and Baigent, C. (2018). The 86 Kwakernaak, A.J., Lambert, G., Slagman, M.C. et al.
role of lipoprotein (a) in chronic kidney disease. J. Lipid (2013). Proprotein convertase subtilisin-kexin type 9 is
Res. 59 (4): 577–585. elevated in proteinuric subjects: relationship with
74 Frischmann, M.E., Kronenberg, F., Trenkwalder, E. et al. lipoprotein response to antiproteinuric treatment.
(2007). in vivo turnover study demonstrates diminished Atherosclerosis 226 (2): 459–465.
clearance of lipoprotein(a) in hemodialysis patients. 87 Konarzewski, M., Szolkiewicz, M., Sucajtys-Szulc, E.
Kidney Int. 71 (10): 1036–1043. et al. (2014). Elevated circulating PCSK-9 concentration
75 Lin, J., Khetarpal, S.A., Terembula, K. et al. (2015). in renal failure patients is corrected by renal replacement
Relation of atherogenic lipoproteins with estimated therapy. Am. J. Nephrol. 40 (2): 157–163.
glomerular filtration rate decline: a longitudinal study. 88 Sniderman, A., Cianflone, K., Kwiterovich, P.O. Jr. et al.
BMC Nephrol. 16: 130. (1987). Hyperapobetalipoproteinemia: the major
76 McKenzie, I.F. and Nestel, P.J. (1968). Studies on the dyslipoproteinemia in patients with chronic renal failure
turnover of triglyceride and esterified cholesterol in treated with chronic ambulatory peritoneal dialysis.
subjects with the nephrotic syndrome. J. Clin. Invest. 47 Atherosclerosis 65 (3): 257–264.
(7): 1685–1695. 89 Attman, P.O., Samuelsson, O.G., Moberly, J. et al. (1999).
77 Joven, J., Villabona, C., Vilella, E. et al. (1990). Apolipoprotein B-containing lipoproteins in renal failure:
Abnormalities of lipoprotein metabolism in patients the relation to mode of dialysis. Kidney Int. 55 (4):
with the nephrotic syndrome. N. Engl. J. Med. 323 (9): 1536–1542.
579–584. 90 Horkko, S., Huttunen, K., and Kesaniemi, Y.A. (1995).
78 Berg, A.L., Nilsson-Ehle, P., and Arnadottir, M. (1999). Decreased clearance of low-density lipoprotein in uremic
Beneficial effects of ACTH on the serum lipoprotein patients under dialysis treatment. Kidney Int. 47 (6):
profile and glomerular function in patients with 1732–1740.
membranous nephropathy. Kidney Int. 56 (4): 1534–1543. 91 Kagan, A., Bar-Khayim, Y., Schafer, Z., and Fainaru, M.
79 de Sain-van der Velden, M.G., Kaysen, G.A., Barrett, H.A. (1990). Kinetics of peritoneal protein loss during CAPD:
et al. (1998). Increased VLDL in nephrotic patients results II. Lipoprotein leakage and its impact on plasma lipid
from a decreased catabolism while increased LDL results levels. Kidney Int. 37 (3): 980–990.
from increased synthesis. Kidney Int. 53 (4): 994–1001. 92 Lewis, L.A., Zuehlke, V., Nakamoto, S. et al. (1966). Renal
80 Vega, G.L., Toto, R.D., and Grundy, S.M. (1995). regulation of serum alpha-lipoproteins. Decrease of
Metabolism of low density lipoproteins in nephrotic alpha-lipoproteins in the absence of renal function. N.
dyslipidemia: comparison of hypercholesterolemia alone Engl. J. Med. 275 (20): 1097–1100.
and combined hyperlipidemia. Kidney Int. 47 (2): 93 Shoji, T., Nishizawa, Y., Nishitani, H. et al. (1992).
579–586. Impaired metabolism of high density lipoprotein in
81 Warwick, G.L., Packard, C.J., Demant, T. et al. (1991). uremic patients. Kidney Int. 41 (6): 1653–1661.
Metabolism of apolipoprotein B-containing lipoproteins 94 Kaysen, G.A., Dalrymple, L.S., Grimes, B. et al. (2014).
in subjects with nephrotic-range proteinuria. Kidney Int. Changes in serum inflammatory markers are associated
40 (1): 129–138. with changes in apolipoprotein A1 but not B after the
82 Moulin, P., Appel, G.B., Ginsberg, H.N., and Tall, A.R. initiation of dialysis. Nephrol. Dial. Transplant. 29 (2):
(1992). Increased concentration of plasma cholesteryl 430–437.
ester transfer protein in nephrotic syndrome: role in 95 Okubo, K., Ikewaki, K., Sakai, S. et al. (2004). Abnormal
dyslipidemia. J. Lipid Res. 33 (12): 1817–1822. HDL apolipoprotein A-I and A-II kinetics in hemodialysis
83 Zhang, C., Yao, M., Wang, X. et al. (2007). Effect of patients: a stable isotope study. J. Am. Soc. Nephrol. 15 (4):
hypoalbuminemia on the increased serum cholesteryl ester 1008–1015.
transfer protein concentration in children with idiopathic 96 Honda, H., Hirano, T., Ueda, M. et al. (2016). High-
nephrotic syndrome. Clin. Biochem. 40 (12): 869–875. density lipoprotein subfractions and their oxidized
84 Krikken, J.A., Waanders, F., Dallinga-Thie, G.M. et al. subfraction particles in patients with chronic kidney
(2009). Antiproteinuric therapy decreases LDL- disease. J. Atheroscler. Thromb. 23 (1): 81–94.
cholesterol as well as HDL-cholesterol in non-diabetic 97 Rubinow, K.B., Henderson, C.M., Robinson-Cohen, C.
proteinuric patients: relationships with cholesteryl ester et al. (2017). Kidney function is associated with an altered
transfer protein mass and adiponectin. Expert Opin. Ther. protein composition of high-density lipoprotein. Kidney
Targets 13 (5): 497–504. Int. 92 (6): 1526–1535.
98 Shao, B., de Boer, I., Tang, C. et al. (2015). A cluster of 113 Lloveras, J., Senti, M., Puig, J.M. et al. (1997). Effect of
proteins implicated in kidney disease is increased in time elapsed since transplantation on lipid and
high-density lipoprotein isolated from hemodialysis lipoprotein abnormalities in renal transplant patients:
subjects. J. Proteome Res. 14 (7): 2792–2806. role of maintenance prednisone and gender influence.
99 Holzer, M., Birner-Gruenberger, R., Stojakovic, T. et al. Transplant. Proc. 29 (1–2): 212–215.
(2011). Uremia alters HDL composition and function. 114 von Ahsen, N., Helmhold, M., Eisenhauer, T. et al.
J. Am. Soc. Nephrol. 22 (9): 1631–1641. (1996). Decrease in lipoprotein(a) after renal
100 Pahl, M.V., Ni, Z., Sepassi, L. et al. (2009). Plasma transplantation is related to the glucocorticoid dose. Eur.
phospholipid transfer protein, cholesteryl ester transfer J. Clin. Investig. 26 (8): 668–675.
protein and lecithin:cholesterol acyltransferase in 115 Curtis, J.J., Galla, J.H., Woodford, S.Y. et al. (1982).
end-stage renal disease (ESRD). Nephrol. Dial. Effect of alternate-day prednisone on plasma lipids in
Transplant. 24 (8): 2541–2546. renal transplant recipients. Kidney Int. 22 (1): 42–47.
101 Calabresi, L., Simonelli, S., Conca, P. et al. (2015). 116 Azrolan, N., Brown, C.D., Thomas, L. et al. (1994).
Acquired lecithin:cholesterol acyltransferase deficiency Cyclosporin a has divergent effects on plasma LDL
as a major factor in lowering plasma HDL levels in cholesterol (LDL-C) and lipoprotein(a) [Lp(a)] levels in
chronic kidney disease. J. Intern. Med. 277 (5): 552–561. renal transplant recipients. Evidence for renal
102 Dantoine, T.F., Debord, J., Charmes, J.P. et al. (1998). involvement in the maintenance of LDL-C and the
Decrease of serum paraoxonase activity in chronic renal elevation of Lp(a) concentrations in hemodialysis
failure. J. Am. Soc. Nephrol. 9 (11): 2082–2088. patients. Arterioscler. Thromb. 14 (9): 1393–1398.
103 Samouilidou, E., Kostopoulos, V., Liaouri, A. et al. 117 Senti, M., Puig, J.M., Lloveras, J. et al. (1997). Effect of
(2016). Association of lipid profile with serum PON1 cyclosporine on serum lipoprotein(a) levels in renal
concentration in patients with chronic kidney disease. transplant recipients. Transplant. Proc. 29 (5): 2404–2405.
Ren. Fail. 38 (10): 1601–1606. 118 Cofan, F., Cofan, M., Campos, B. et al. (2005). Effect of
104 Holzer, M., Schilcher, G., Curcic, S. et al. (2015). Dialysis calcineurin inhibitors on low-density lipoprotein
modalities and HDL composition and function. J. Am. oxidation. Transplant. Proc. 37 (9): 3791–3793.
Soc. Nephrol. 26 (9): 2267–2276. 119 Seymen, P., Yildiz, M., Turkmen, M.F. et al. (2009).
105 Kennedy, D.J., Tang, W.H., Fan, Y. et al. (2013). Effects of cyclosporine-tacrolimus switching in
Diminished antioxidant activity of high-density posttransplantation hyperlipidemia on high-density
lipoprotein-associated proteins in chronic kidney lipoprotein 2/3, lipoprotein a1/b, and other lipid
disease. J. Am. Heart Assoc. 2 (2): e000104. parameters. Transplant. Proc. 41 (10): 4181–4183.
106 Speer, T., Rohrer, L., Blyszczuk, P. et al. (2013). 120 Artz, M.A., Boots, J.M., Ligtenberg, G. et al. (2002).
Abnormal high-density lipoprotein induces endothelial Randomized conversion from cyclosporine to tacrolimus
dysfunction via activation of toll-like receptor-2. in renal transplant patients: improved lipid profile and
Immunity 38 (4): 754–768. unchanged plasma homocysteine levels. Transplant.
107 Weichhart, T., Kopecky, C., Kubicek, M. et al. (2012). Proc. 34 (5): 1793–1794.
Serum amyloid a in uremic HDL promotes 121 Ichimaru, N., Takahara, S., Kokado, Y. et al. (2001).
inflammation. J. Am. Soc. Nephrol. 23 (5): 934–947. Changes in lipid metabolism and effect of simvastatin in
108 Kaseda, R., Jabs, K., Hunley, T.E. et al. (2015). renal transplant recipients induced by cyclosporine or
Dysfunctional high-density lipoproteins in children with tacrolimus. Atherosclerosis 158 (2): 417–423.
chronic kidney disease. Metabolism 64 (2): 263–273. 122 Vincenti, F., Friman, S., Scheuermann, E. et al. (2007).
109 Yamamoto, S., Yancey, P.G., Ikizler, T.A. et al. (2012). Results of an international, randomized trial comparing
Dysfunctional high-density lipoprotein in patients on glucose metabolism disorders and outcome with
chronic hemodialysis. J. Am. Coll. Cardiol. 60 (23): cyclosporine versus tacrolimus. Am. J. Transplant. 7 (6):
2372–2379. 1506–1514.
110 Meier, S.M., Wultsch, A., Hollaus, M. et al. (2015). Effect 123 Silva, H.T. Jr., Yang, H.C., Meier-Kriesche, H.U. et al.
of chronic kidney disease on macrophage cholesterol (2014). Long-term follow-up of a phase III clinical trial
efflux. Life Sci. 136: 1–6. comparing tacrolimus extended-release/MMF,
111 Massy, Z.A. and Kasiske, B.L. (1996). Post-transplant tacrolimus/MMF, and cyclosporine/MMF in de novo
hyperlipidemia: mechanisms and management. J. Am. kidney transplant recipients. Transplantation 97 (6):
Soc. Nephrol. 7 (7): 971–977. 636–641.
112 Hilbrands, L.B., Demacker, P.N., Hoitsma, A.J. et al. 124 Kasiske, B.L., de Mattos, A., Flechner, S.M. et al. (2008).
(1995). The effects of cyclosporine and prednisone on Mammalian target of rapamycin inhibitor dyslipidemia
serum lipid and (apo)lipoprotein levels in renal transplant in kidney transplant recipients. Am. J. Transplant. 8 (7):
recipients. J. Am. Soc. Nephrol. 5 (12): 2073–2081. 1384–1392.
Reference 569
25 El-Agroudy, A.E., Alarrayed, S.M., Al-Ghareeb, S.M.

1 138 Streja. E., Gosmanova, E., Molnar M.Z. (2018).
et al. (2017). Efficacy and safety of early tacrolimus Association of Continuation of Statin Therapy Initiated
conversion to sirolimus after kidney transplantation: Before Transition to Chronic Dialysis Therapy With
long-term results of a prospective randomized study. Mortality After Dialysis Initiation. JAMA Network Open.
Indian J. Nephrol. 27 (1): 28–36. 1(6):e182311.
126 Fischereder, M., Graeb, C., Kruger, B. et al. (2006). 139 Edmonston, D., Wolf, M., and Pencina, M. (2018).
Conversion from calcineurin inhibitors to sirolimus in Statins at dialysis transition-supportive but not
patients with chronic renal allograft dysfunction. sufficient. JAMA Netw. Open 1 (6): e182411.
Transplant. Proc. 38 (5): 1295–1297. 140 Holdaas, H., Fellstrom, B., Jardine, A.G. et al. (2003).
127 Hoogeveen, R.C., Ballantyne, C.M., Pownall, H.J. et al. Effect of fluvastatin on cardiac outcomes in renal
(2001). Effect of sirolimus on the metabolism of transplant recipients: a multicentre, randomised,
apoB100 – containing lipoproteins in renal transplant placebo-controlled trial. Lancet 361 (9374): 2024–2031.
patients. Transplantation 72 (7): 1244–1250. 141 Jardine, A.G., Fellstrom, B., Logan, J.O. et al. (2005).
128 Morrisett, J.D., Abdel-Fattah, G., and Kahan, B.D. Cardiovascular risk and renal transplantation: post hoc
(2003). Sirolimus changes lipid concentrations and analyses of the Assessment of Lescol in Renal
lipoprotein metabolism in kidney transplant recipients. Transplantation (ALERT) Study. Am. J. Kidney Dis. 46
Transplant. Proc. 35 (3 Suppl): 143S–150S. (3): 529–536.
129 Sam, W.J., Chamberlain, C.E., Lee, S.J. et al. (2012). 142 Holdaas, H., Fellstrom, B., Jardine, A.G. et al. (2005).
Associations of ABCB1 and IL-10 genetic polymorphisms Beneficial effect of early initiation of lipid-lowering
with sirolimus-induced dyslipidemia in renal transplant therapy following renal transplantation. Nephrol. Dial.
recipients. Transplantation 94 (9): 971–977. Transplant. 20 (5): 974–980.
130 Baigent, C., Landray, M. J., Reith C., et al. (2011). The 143 Holdaas, H., Fellstrom, B., Cole, E. et al. (2005).
effects of lowering LDL cholesterol with simvastatin Long-term cardiac outcomes in renal transplant
plus ezetimibe in patients with chronic kidney disease recipients receiving fluvastatin: the ALERT extension
(Study of Heart and Renal Protection): a randomised study. Am. J. Transplant. 5 (12): 2929–2936.
placebo-controlled trial. Lancet. 377(9784):2181–92. 144 Palmer, S.C., Navaneethan, S.D., Craig, J.C. et al. (2014).
131 Palmer, S.C., Navaneethan, S.D., Craig, J.C. et al. (2014). HMG CoA reductase inhibitors (statins) for kidney
HMG CoA reductase inhibitors (statins) for people with transplant recipients. Cochrane Database Syst Rev. (1)
chronic kidney disease not requiring dialysis. Cochrane Art. No.: CD005019.
Database Syst. Rev. (5) (Art. No.: CD007784). 145 Geng, Q., Ren, J., Song, J. et al. (2014). Meta-analysis of
132 Hou, W., Lv, J., Perkovic, V. et al. (2013). Effect of statin the effect of statins on renal function. Am. J. Cardiol.
therapy on cardiovascular and renal outcomes in 114 (4): 562–570.
patients with chronic kidney disease: a systematic 146 Hanai, K., Babazono, T., Takemura, S. et al. (2015).
review and meta-analysis. Eur. Heart J. 34 (24): Comparative effects of statins on the kidney function in
1807–1817. patients with type 2 diabetes. J. Atheroscler. Thromb. 22
133 Wanner, C., Krane, V., Marz, W., et. al. (2005). (6): 618–627. https://pubmed.ncbi.nlm.nih.gov/25476755/
Atorvastatin in patients with type 2 diabetes mellitus 147 de Zeeuw, D., Anzalone, D.A., Cain, V.A. et al. (2015).
undergoing hemodialysis. N Engl J Med. 353(3):238–48. Renal effects of atorvastatin and rosuvastatin in patients
134 Fellstrom, B. C., Jardine, A. G., Schmieder, R. E., et al. with diabetes who have progressive renal disease
Rosuvastatin and cardiovascular events in patients (PLANET I): a randomised clinical trial. Lancet Diabetes
undergoing hemodialysis. N Engl J Med. Endocrinol. 3 (3): 181–190.
360(14):1395–407. 148 Kimura, S., Inoguchi, T., Yokomizo, H. et al. (2012).
135 Goldfarb-Rumyantzev, A.S., Habib, A.N., Baird, B.C. Randomized comparison of pitavastatin and pravastatin
et al. (2007). The association of lipid-modifying treatment on the reduction of urinary albumin in
medications with mortality in patients on long-term patients with type 2 diabetic nephropathy. Diabetes
peritoneal dialysis. Am. J. Kidney Dis. 50 (5): 791–802. Obes. Metab. 14 (7): 666–669.
136 Lee, J.E., Oh, K.H., Choi, K.H. et al. (2011). Statin 149 Wanner, C. (1996). HMG-CoA reductase inhibitor
therapy is associated with improved survival in incident treatment in renal insufficiency. Nephrol. Dial.
peritoneal dialysis patients: propensity-matched Transplant. 11 (10): 1951–1952.
comparison. Nephrol. Dial. Transplant. 26 (12): 150 Newman, C.B., Preiss, D., Tobert, J.A. et al. (2019).
4090–4094. Statin safety and associated adverse events: a scientific
137 Attman, P.O., Samuelsson, O., Johansson, A.C. et al. statement from the American Heart Association.
(2003). Dialysis modalities and dyslipidemia. Kidney Int. Arterioscler. Thromb. Vasc. Biol. 39(2):e38–e81 doi:
Suppl. (84): S110–S112. https://doi.org/10.1161/ATV.0000000000000073.
51 Launay-Vacher, V., Izzedine, H., and Deray, G. (2005).

1 165 Jahangir, E., Fazio, S., and Sampson, U.K. (2013).
Statins’ dosage in patients with renal failure and Incident diabetes and statins: the blemish of an
cyclosporine drug-drug interactions in transplant undisputed heavy weight champion? Br. J. Clin.
recipient patients. Int. J. Cardiol. 101 (1): 9–17. Pharmacol. 75 (4): 955–958.
152 Nakamura, H., Mizuno, K., Ohash, Y., et al. (2009). 166 Chan, D.C., Pang, J., and Watts, G.F. (2015).
Pravastatin and cardiovascular risk in moderate chronic Pathogenesis and management of the diabetogenic
kidney disease. Atherosclerosis. 206(2):512–7. effect of statins: a role for adiponectin and coenzyme
153 Shepherd, J., Kastelein, J.P., Bittner, V., et al. (2008). Q10? Curr. Atheroscler. Rep. 17 (1): 472.
Intensive lipid lowering with atorvastatin in patients 167 Yaluri, N., Modi, S., Lopez Rodriguez, M. et al. (2015).
with coronary heart disease and chronic kidney disease: Simvastatin impairs insulin secretion by multiple
the TNT (Treating to New Targets) study. J Am Coll mechanisms in MIN6 cells. PLoS One 10 (11):
Cardiol. 51(15):1448–54. e0142902.
154 Ridker, P. M., Danielson, E., Fonseca, F. A. H., et al. 168 Baker, W.L., Talati, R., White, C.M., and Coleman, C.I.
(2008). Rosuvastatin to prevent vascular events in men (2010). Differing effect of statins on insulin sensitivity in
and women with elevated C-reactive protein N Engl J non-diabetics: a systematic review and meta-analysis.
Med. 359(21):2195–207. Diabetes Res. Clin. Pract. 87 (1): 98–107.
155 Ridker, P.M., Pradhan, A., MacFadyen, J.G. et al. (2012). 169 Bays, H.E. (2011). Colesevelam hydrochloride added to
Cardiovascular benefits and diabetes risks of statin background metformin therapy in patients with type 2
therapy in primary prevention: an analysis from the diabetes mellitus: a pooled analysis from 3 clinical
JUPITER trial. Lancet 380 (9841): 565–571. studies. Endocr. Pract. 17 (6): 933–938.
156 Rajpathak, S.N., Kumbhani, D.J., Crandall, J. et al. (2009). 170 LRC-CPPT (1984). The lipid research clinics coronary
Statin therapy and risk of developing type 2 diabetes: a primary prevention trial results. I. Reduction in
meta-analysis. Diabetes Care 32 (10): 1924–1929. incidence of coronary heart disease. JAMA 251 (3):
157 Mills, E.J., Wu, P., Chong, G. et al. (2011). Efficacy and 351–364.
safety of statin treatment for cardiovascular disease: a 171 Chertow, G.M., Burke, S.K., Dillon, M.A., and
network meta-analysis of 170,255 patients from 76 Slatopolsky, E. (1999). Long-term effects of sevelamer
randomized trials. QJM 104 (2): 109–124. hydrochloride on the calcium x phosphate product and
158 Preiss, D., Seshasai, S.R., Welsh, P. et al. (2011). Risk of lipid profile of haemodialysis patients. Nephrol. Dial.
incident diabetes with intensive-dose compared with Transplant. 14 (12): 2907–2914.
moderate-dose statin therapy: a meta-analysis. JAMA 172 Locatelli, F., Spasovski, G., Dimkovic, N. et al. (2014).
305 (24): 2556–2564. The effects of colestilan versus placebo and sevelamer in
159 Navarese, E.P., Buffon, A., Andreotti, F. et al. (2013). patients with CKD 5D and hyperphosphataemia: a
Meta-analysis of impact of different types and doses of 1-year prospective randomized study. Nephrol. Dial.
statins on new-onset diabetes mellitus. Am. J. Cardiol. Transplant. 29 (5): 1061–1073.
111 (8): 1123–1130. 173 Wanner, C., Marz, W., Varushchanka, A. et al. (2014).
160 Sattar, N., Preiss, D., Murray, H.M. et al. (2010). Statins Dyslipidemia in chronic kidney disease: randomized
and risk of incident diabetes: a collaborative meta- controlled trial of colestilan versus simvastatin in
analysis of randomised statin trials. Lancet 375 (9716): dialysis patients. Clin. Nephrol. 82 (3): 163–172.
735–742. 174 Jamal, S.A., Vandermeer, B., Raggi, P. et al. (2013).
161 Steen, D.L. and Bhatt, D.L. (2014). Statin potency Effect of calcium-based versus non-calcium-based
associated with incident diabetes in a real-world phosphate binders on mortality in patients with chronic
evaluation. Evid. Based Med. 19 (2): 68. kidney disease: an updated systematic review and
162 Zaharan, N.L., Williams, D., and Bennett, K. (2013). meta-analysis. Lancet 382 (9900): 1268–1277.
Statins and risk of treated incident diabetes in a primary 175 St Peter, W.L., Liu, J., Weinhandl, E., and Fan, Q. (2008).
care population. Br. J. Clin. Pharmacol. 75 (4): A comparison of sevelamer and calcium-based
1118–1124. phosphate binders on mortality, hospitalization, and
163 Vallejo-Vaz, A.J., Kondapally Seshasai, S.R., Kurogi, K. morbidity in hemodialysis: a secondary analysis of the
et al. (2015). Effect of pitavastatin on glucose, HbA1c dialysis clinical outcomes revisited (DCOR) randomized
and incident diabetes: a meta-analysis of randomized trial using claims data. Am. J. Kidney Dis. 51 (3):
controlled clinical trials in individuals without diabetes. 445–454.
Atherosclerosis 241 (2): 409–418. 176 Boaz, M., Katzir, Z., Schwartz, D. et al. (2011). Effect of
164 Blackburn, D.F., Chow, J.Y., and Smith, A.D. (2015). sevelamer hydrochloride exposure on carotid intima
Statin use and incident diabetes explained by bias rather media thickness in hemodialysis patients. Nephron Clin.
than biology. Can. J. Cardiol. 31 (8): 966–969. Pract. 117 (2): c83–c88.
Reference 571
177 Qunibi, W., Moustafa, M., Muenz, L. R., et al. (2008). A 189 Jun, M., Zhu, B., Tonelli, M. et al. (2012). Effects of
1-year randomized trial of calcium acetate versus fibrates in kidney disease: a systematic review and
sevelamer on progression of coronary artery meta-analysis. J. Am. Coll. Cardiol. 60 (20): 2061–2071.
calcification in hemodialysis patients with comparable 190 Group AS, Ginsberg, H.N., Elam, M.B. et al. (2010).
lipid control: the Calcium Acetate Renagel Evaluation-2 Effects of combination lipid therapy in type 2 diabetes
(CARE-2) study. Am J Kidney Dis. 51(6):952–65. mellitus. N. Engl. J. Med. 362 (17): 1563–1574.
178 Cannon, C.P., Blazing, M.A., Giugliano, M.P., et al. 191 Papademetriou, V., Lovato, L., Tsioufis, C. et al. (2017).
(2015) Ezetimibe Added to Statin Therapy after Acute Effects of high density lipoprotein raising therapies on
Coronary Syndromes. N Engl J Med. 372(25):2387–97. Cardiovascular outcomes in patients with type 2
179 Stanifer, J.W., Charytan, D.M., White, J. et al. (2017). diabetes mellitus, with or without renal impairment:
Benefit of Ezetimibe added to simvastatin in reduced The Action to Control Cardiovascular Risk in Diabetes
kidney function. J. Am. Soc. Nephrol. 28 (10): 3034–3043. Study. Am. J. Nephrol. 45 (2): 136–145.
180 Sabatine, M.S., Giugliano, R.P., Keech, A.C. et al. (2017). 192 Pierce, L.R., Wysowski, D.K., and Gross, T.P. (1990).
Evolocumab and clinical outcomes in patients with Myopathy and rhabdomyolysis associated with
cardiovascular disease. N. Engl. J. Med. 376 (18): lovastatin-gemfibrozil combination therapy. JAMA 264
1713–1722. (1): 71–75.
181 Robinson, J.G., Farnier, M., Krempf, M. et al. (2015). 193 Pan, W.J., Gustavson, L.E., Achari, R. et al. (2000). Lack
Efficacy and safety of alirocumab in reducing lipids and of a clinically significant pharmacokinetic interaction
cardiovascular events. N. Engl. J. Med. 372 (16): between fenofibrate and pravastatin in healthy
1489–1499. volunteers. J. Clin. Pharmacol. 40 (3): 316–323.
182 Schwartz, J.S., Szarek, M., Bhatt, D.L., et al. (2018). The 194 Bergman, A.J., Murphy, G., Burke, J. et al. (2004).
ODYSSEY OUTCOMES Trial: topline results, Simvastatin does not have a clinically significant
alirocumab in patients after acute coronary syndrome. pharmacokinetic interaction with fenofibrate in
Presented at American College of Cardiology-67th humans. J. Clin. Pharmacol. 44 (9): 1054–1062.
Scientific Sessions, Orlando, FL (2018). 195 Whitfield, L.R., Porcari, A.R., Alvey, C. et al. (2011).
183 Toth, P.P., Dwyer, J.P., Cannon, C.P. et al. (2018). Effect of gemfibrozil and fenofibrate on the
Efficacy and safety of lipid lowering by alirocumab in pharmacokinetics of atorvastatin. J. Clin. Pharmacol. 51
chronic kidney disease. Kidney Int. 93 (6): 1397–1408. (3): 378–388.
184 Sabatine, M.S., Giugliano, R.P., Keech, A. et al. (2016). 196 Tonelli, M. and Wanner, C. (2014). Kidney Disease:
Rationale and design of the further cardiovascular Improving Global Outcomes Lipid Guideline
outcomes research with PCSK9 inhibition in subjects Development Work Group M. Lipid management in
with elevated risk trial. Am. Heart J. 173: 94–101. chronic kidney disease: synopsis of the Kidney Disease:
185 Charytan, D.M. and Sabatine, M.S. (2018). FR-OR114. Improving Global Outcomes 2013 clinical practice
Efficacy and Safety of Evolocumab in CKD: Data from guideline. Ann. Intern. Med. 160 (3): 182.
the FOURIER Trial. Presented at: American Society of 197 McPherson, R., Frohlich, J., Fodor, G. et al. (2006).
Nephrology Annual Conference (San Diego, CA, 2018). Canadian Cardiovascular Society Position Statement –
186 Frick, M.H., Elo, O., Haapa, K. et al. (1987). Helsinki recommendations for the diagnosis and treatment of
heart study: primary-prevention trial with gemfibrozil dyslipidemia and prevention of cardiovascular disease.
in middle-aged men with dyslipidemia. Safety of Can. J. Cardiol. 22 (11): 913–927.
treatment, changes in risk factors, and incidence of 198 Investigators, A.-H., Boden, W.E., Probstfield, J.L. et al.
coronary heart disease. N. Engl. J. Med. 317 (20): (2011). Niacin in patients with low HDL cholesterol
1237–1245. levels receiving intensive statin therapy. N. Engl. J. Med.
187 Rubins, H.B., Robins, S.J., Collins, D. et al. (1999). 365 (24): 2255–2267.
Gemfibrozil for the secondary prevention of coronary 199 Guyton, J.R., Slee, A.E., Anderson, T. et al. (2013).
heart disease in men with low levels of high-density Relationship of lipoproteins to cardiovascular events:
lipoprotein cholesterol. Veterans Affairs High-Density the AIM-HIGH trial. J. Am. Coll. Cardiol. 62: 1580–1584.
Lipoprotein Cholesterol Intervention Trial Study Group. 200 Kalil, R.S., Wang, J.H., de Boer, I.H. et al. (2015). Effect
N. Engl. J. Med. 341 (6): 410–418. of extended-release niacin on cardiovascular events and
188 Keech, A., Simes, R.J., Barter, P. et al. (2005). Effects of kidney function in chronic kidney disease: a post hoc
long-term fenofibrate therapy on cardiovascular events analysis of the AIM-HIGH trial. Kidney Int. 87 (6):
in 9795 people with type 2 diabetes mellitus (the FIELD 1250–1257.
study): randomised controlled trial. Lancet 366 (9500): 201 Group HTC (2013). HPS2-THRIVE randomized
1849–1861. placebo-controlled trial in 25 673 high-risk patients of
ER niacin/laropiprant: trial design, pre-specified muscle events in hypercholesterolaemic patients (JELIS): a

and liver outcomes, and reasons for stopping study randomised open-label, blinded endpoint analysis.
treatment. Eur. Heart J. 34 (17): 1279–1291. https:// Lancet 369 (9567): 1090–1098.
pubmed.ncbi.nlm.nih.gov/23444397/ 214 Tanaka, K., Ishikawa, Y., Yokoyama, M. et al. (2008).
202 Group HTC, Landray, M.J., Haynes, R. et al. (2014). Reduction in the recurrence of stroke by
Effects of extended-release niacin with laropiprant in eicosapentaenoic acid for hypercholesterolemic patients:
high-risk patients. N. Engl. J. Med. 371 (3): 203–212. subanalysis of the JELIS trial. Stroke 39 (7): 2052–2058.
203 Reiche, I., Westphal, S., Martens-Lobenhoffer, J. et al. 215 Oikawa, S., Yokoyama, M., Origasa, H. et al. (2009).
(2011). Pharmacokinetics and dose recommendations of Suppressive effect of EPA on the incidence of coronary
Niaspan(R) in chronic kidney disease and dialysis events in hypercholesterolemia with impaired glucose
patients. Nephrol. Dial. Transplant. 26 (1): 276–282. metabolism: Sub-analysis of the Japan EPA Lipid
https://pubmed.ncbi.nlm.nih.gov/20562093/ Intervention Study (JELIS). Atherosclerosis 206 (2):
204 Streja, E., Kovesdy, C.S., Streja, D.A. et al. (2015). Niacin 535–539.
and the progression of chronic kidney disease. Am. J. 216 Bhatt, D.L., Steg, P.G., Miller, M. et al. (2018).
Kidney Dis. 65 (5): 785–798. Cardiovascular Risk reduction with Icosapent ethyl for
205 Taketani, Y., Masuda, M., Yamanaka-Okumura, H. et al. hypertriglyceridemia. N. Engl. J. Med. 380 (1): 11–22.
(2015). Niacin and chronic kidney disease. J. Nutr. Sci. 217 Nakamura, N., Fujita, T., Kumasaka, R. et al. (2008).
Vitaminol. 61 (Suppl): S173–S175. Serum lipid profile and plasma fatty acid composition in
206 Ito, M.K. (2015). A comparative overview of prescription hemodialysis patients – comparison with chronic
omega-3 fatty acid products. P T. 40 (12): 826–857. kidney disease patients. in vivo 22 (5): 609–611.
207 Corrao, G., Zambon, A., Bertu, L. et al. (2004). Lipid 218 Hamazaki, K., Terashima, Y., Itomura, M. et al. (2011).
lowering drugs prescription and the risk of peripheral Docosahexaenoic acid is an independent predictor of
neuropathy: an exploratory case-control study using all-cause mortality in hemodialysis patients. Am. J.
automated databases. J. Epidemiol. Community Health Nephrol. 33 (2): 105–110.
58 (12): 1047–1051. 219 Friedman, A.N., Yu, Z., Denski, C. et al. (2013). Fatty
208 Marchioli, R., Barzi, F., Bomba, E. et al. (2002). Early acids and other risk factors for sudden cardiac death in
protection against sudden death by n-3 polyunsaturated patients starting hemodialysis. Am. J. Nephrol. 38 (1):
fatty acids after myocardial infarction: time-course 12–18.
analysis of the results of the Gruppo Italiano per lo 220 Svensson, M., Schmidt, E.B., Jorgensen, K.A. et al.
Studio della Sopravvivenza nell’Infarto Miocardico (2006). N-3 fatty acids as secondary prevention against
(GISSI)-Prevenzione. Circulation 105 (16): 1897–1903. cardiovascular events in patients who undergo chronic
209 Tavazzi, L., Maggioni, A.P., Marchioli, R. et al. (2008). hemodialysis: a randomized, placebo-controlled
Effect of n-3 polyunsaturated fatty acids in patients with intervention trial. Clin. J. Am. Soc. Nephrol. 1 (4):
chronic heart failure (the GISSI-HF trial): a randomised, 780–678.
double-blind, placebo-controlled trial. Lancet 372 221 Lok, C.E., Moist, L., Hemmelgarn, B.R. et al. (2012).
(9645): 1223–1230. Effect of fish oil supplementation on graft patency and
210 Rauch, B., Schiele, R., Schneider, S. et al. (2010). cardiovascular events among patients with new
OMEGA, a randomized, placebo-controlled trial to test synthetic arteriovenous hemodialysis grafts: a
the effect of highly purified omega-3 fatty acids on top randomized controlled trial. JAMA 307 (17): 1809–1816.
of modern guideline-adjusted therapy after myocardial 222 Stone, N.J., Robinson, J.G., Lichtenstein, A.H. et al.
infarction. Circulation 23;122(21):2152–9. (2014). 2013 ACC/AHA guideline on the treatment of
211 Kromhout, D., Giltay, E.J., and Geleijnse, J.M. (2010). N-3 blood cholesterol to reduce atherosclerotic
fatty acids and cardiovascular events after myocardial cardiovascular risk in adults: a report of the American
infarction. N. Engl. J. Med. 363 (21): 2015, 2026. College of Cardiology/American Heart Association Task
212 Galan, P., Kesse-Guyot, E., Czernichow, S. et al. (2010). Force on Practice Guidelines. Circulation 129 (25 Suppl
Effects of B vitamins and omega 3 fatty acids on 2): S1–S45.
cardiovascular diseases: a randomised placebo 223 Walther, C.P., Richardson, P.A., Virani, S.S. et al. (2018).
controlled trial. BMJ 341: c6273. Association between intensity of statin therapy and
213 Yokoyama, M., Origasa, H., Matsuzaki, M. et al. (2007). mortality in persons with chronic kidney disease.
Effects of eicosapentaenoic acid on major coronary Nephrol. Dial. Transplant. 35 (2): 312–319.
573
35
Chronic Kidney Disease and Hypertension

Branko Braam
Division of Nephrology, Department of Medicine, University of Alberta, Canada
I ntroduction below, with the revision of several guidelines as a conse-

quence. With increasing severity of kidney disease, the
Hypertension (HTN) is not only highly prevalent in prevalence of HTN becomes progressively higher such that
patients with chronic kidney disease (CKD), it is also one most patients with end-stage renal disease (ESRD) will
of the most prevalent causes of progressive kidney injury become hypertensive.
and end-stage kidney disease (ESKD). This makes HTN a Besides uncontrolled HTN, it seems relevant to consider
central issue in the treatment of the patient with CKD. resistant HTN (rHTN) [2]. Several definitions are relevant
Although it is clear that the renin angiotensin system here. The term “apparent treatment-resistant hyperten-
(RAS), the sympathetic nervous system (SNS), and extra- sion” is defined as BP not at goal when the patient is pre-
cellular fluid volume (ECFV) expansion play a pivotal role scribed 3 classes of antihypertensive medication or at goal
in the pathophysiology of HTN in CKD, HTN remains after prescription of 4 classes. Of note, the most recent
uncontrolled in many patients. Clinically, blood pressure position paper on rHTN by the American Heart Association
(BP) measurements in CKD are more complex than in does not mention an actual BP level, but defines rHTN as
essential hypertension because of the frequently disturbed the BP of a hypertensive patient that remains elevated
diurnal rhythms. Angiotensin converting enzyme inhibi- “above goal” [3]. It is quite likely that over the next few
tors (ACEis) and angiotensin receptor blockers (ARBs) are years the BP level for defining rHTN could be adjusted as a
mainstay therapies for HTN in CKD to prevent progression result of the new insights about HTN appearing in the lit-
of kidney failure. Diuretics are underutilized but can sub- erature (see below).
stantially contribute to better BP control in CKD. It is not The prevalence of apparent treatment-rHTN amounts to
clear whether newer therapies like baroreceptor stimula- 2–10% in the general population [4, 5]. Data from the
tion and kidney denervation have utility in CKD. We pro- Chronic Renal Insufficiency Cohort (CRIC) study indicate
pose an approach to the treatment of HTN in CKD based that the prevalence is substantially higher in patients with
on the pathophysiology of HTN in CKD. For information CKD [6]. In the study, 40% of participants had apparent
about management of HTN in transplant recipients, and treatment-rHTN, with half of them not at target BP on
patients on hemodialysis or peritoneal dialysis, refer to 3 medications and the other half at target BP on 4 medi-
Chapters 22 and 42. cations. This leads to questioning how many patients have
“true” rHTN. Judd and Calhoun defined true rHTN as “a
properly measured office BP >140/90 mmHg with a mean
E
pidemiology 24-hour ambulatory BP >130/80 mmHg in a patient con-
firmed to be taking 3 antihypertensive medications” [7].
In this chapter, HTN in CKD is defined as having a BP of The difference with the definition of apparent resistant
>130/80 mmHg or being on antihypertensive treatment [1]. hypertension is that here, BP is “properly measured” and
This threshold is based on the recent literature, as discussed not derived from sources where the assessment of BP is ill
574 Chronic Kidney Disease and Hypertension
Clinic BP 2,819
mmHg
49.5%
1,641
28.8%
Concordant control
Concordant lack of control
“White-coat”
Masked hypertension
140/90
837 396
14.7% 7.0%
130/80 24-hour ambulatory BP

mmHg
Figure 35.1 Prevalence of masked and white-coat hypertension in hypertensive patients with chronic kidney disease. Source:
Reproduced with permission [13]. On the x axis, patients are divided into two groups: blood pressure (BP) <130/80 mmHg with
24-hour ambulatory BP or ≥130/80 mmHg. On the y axis, patients are also divided into two groups: clinic BP <140/90 mmHg and
≥140/90 mmHg. There are two groups where 24-hour ambulatory BP and clinic BP are discordant. One is the well-known white-coat
hypertension: patients with normal BP at home (assessed with the 24-hour ambulatory BP assessment) and elevated in the clinic
(assessed with the clinic BP). Less well recognized is masked hypertension: patients with normal BP in the clinic and elevated BP at
home. Source: Gorostidi et al. [13]. © 2013, Elsevier.
defined, and that medication is “confirmed” and not “pre- Assessment of Blood Pressure
scribed.” While CKD is identified as a reason for true rHTN,
it is currently not clear whether the prevalence of true In the context of clinical evaluation of the patient with CKD,
rHTN is higher or lower than in people without CKD. It four different assessments of BP are being used: in-office
does not take away the relevance to investigate adherence if manual, multiple in-office automated, home, and ambula-
BP targets are not achieved with 3 antihypertensive tory measurements. Unfortunately, the level of BP is still
medications. often only derived from in-office manual measurements.
A last issue is masked HTN, which is HTN that is pre- These are notorious for inaccurate assessment of BP for a
sent when the patient is at home, but is not measured as number of reasons, including white coat HTN (i.e. stress
such in the office. The prevalence of masked HTN is esti- associated with a clinic visit), as well as the inability to diag-
mated to be about 7–42% of hypertensive patients [8, 9] nose masked or nighttime HTN. Multiple in-office auto-
and is associated with an elevated cardiovascular risk [10]. mated measurements, if done appropriately with at least five
By its nature, masked HTN can only be detected by ambu- measurements with the patient left alone seated in the right
latory or home BP measurements. In a recent study with position in a quiet room, do resemble daytime ambulatory
CKD patients, Agarwal defined masked HTN as having BP measurements. Although obviously more preferably
controlled HTN in the office (BP <140/90 mmHg), with than a single in-office measurement, it takes more time in
elevated BP at home. Assessed by ambulatory BP measure- the clinic, and masked or nighttime HTN still cannot be
ments >135/80 mmHg during daytime, >130/80 mmHg diagnosed. Since both masked and nighttime HTN are more
during the entire day, or >120/70 mmHg during the night, prevalent in CKD patients [8, 9], this is obviously relevant.
overall prevalence of masked uncontrolled HTN was 56%. Home BP measurements have a number of benefits, includ-
Assessed using average weeklong home BP measurements ing the possibility to have multiple measurements over the
>135/85 mmHg, prevalence was similarly high at 51% [11]. day and during work and weekend days. This facilitates the
A recent study from Europe reported a lower prevalence of assessment of BP variability, including potential weaning of
19% using ambulatory BP measurements [12]. These stud- medication actions during some time of the day (e.g. becom-
ies strongly emphasize the importance of home and ambu- ing hypertensive before taking medications in the morning),
latory measurements. This is further illustrated in and it enables the diagnosis of masked, but not nighttime
Figure 35.1. hypertension. Furthermore, it engages patients in their care.
Pathophysiolog 575
Performing ambulatory BP measurements is more labor Table 35.1 Pros and cons of different approaches to assess
intensive, can be psychologically troubling for the patient, or blood pressure.
even a bit painful, yet it provides day- and nighttime BP
measurements, shows variations, allows coupling BP to Automated
Issue Office BP office BP Home BP ABPM
symptoms, and avoids patient selection. Given the high
prevalence of HTN in CKD, a strong call is being made to use Affected by Yes Possible No No
ambulatory BP measurements more frequently in the assess- white-coat
ment of BP in the CKD patient because of the issues with phenomenon
high prevalence of white coat, masked, nighttime, and vari- Able to detect No No Yesa Yesa
able hypertension. The pros and cons of different BP assess- white-coat
hypertension
ments are summarized in Table 35.1.
Able to detect No No Yesa Yesa
masked
hypertension
Pathophysiology Easily available Yes Possible Yes Possible
Low cost Yes No Yes No
Three key mechanisms form an important basis for HTN in
CKD: activation of the RAS and the SNS, and ECFV expan- a
When office measurements are also available.
sion. In addition, there are dozens of other BP modulating BP, blood pressure; ABPM, ambulatory blood pressure
measurements.
systems involved in the pathophysiology of HTN in CKD.
Studies from the 1970s demonstrated elegantly that when
renin levels are related to volume status, the baroreceptor diminished the activation of the SNS, as assessed by mus-
curve is shifted upward: at any level of ECFV, renin is cle sympathetic nerve activity [20]. Since then, sympathetic
higher in CKD patients compared to controls. This acti- hyperactivity has been shown in nonproteinuric CKD as
vated RAS is more prominent in polycystic kidney well as polycystic kidney disease. The hyperreactivity
disease [14], kidney involvement in scleroderma [15], as depends at least in part on the RAS, since in CKD patients
well as renovascular disease. The mechanisms by which ACE inhibition leads to normalization of sympathetic
angiotensin II leads to HTN are not only the classically activity [21]. Surgical renal denervation has been shown to
described vasoconstriction and increase in aldosterone reduced BP in experimental CKD as well as in humans [22].
release but also encompass direct increases in proximal Denervation using an intravascular device has been less
tubular reabsorption [16] and a number of indirect effects successful; however, technique improvements could
like increased oxidative stress with all its consequences and resolve this [23]. Of note, sympathetic activation not only
activation of the SNS. It has been nicely demonstrated that affects BP control but is also involved in a large number of
in CKD patients, ACE inhibition partly corrects the left processes, including glucose metabolism and insulin resist-
shift of the baroreceptor curve. ance, endothelial dysfunction and vascular stiffness, and
Volume overload is the consequence of a too excessive immune modulation [24].
sodium intake and the impaired ability of the kidney to Besides this classical triad of mechanisms for HTN, CKD,
excrete sodium at normal arterial pressures. Sodium intake and ESRD come with several other disturbances affecting
to a certain extent is modifiable, yet in advanced kidney BP regulation. Structural changes to the vasculature,
disease normotension cannot be maintained even at very including the classical Monkenberg media sclerosis, lead to
low levels of sodium intake [17]. Effectively the pressure increased vascular stiffness and peripheral resistance. This
natriuresis curve has shifted to the right. At a tubular level, will affect both the BP and the pulse wave propagation and
the activated RAS will enhance sodium reabsorption and reflection, leading to increased cardiac afterload.
the nephron is meanwhile unable to suppress its fractional Accumulation of the endogenous inhibitor of nitric oxide
reabsorption sufficiently. If glomerular filtration rate synthase (i.e. asymmetric dimethyl arginine [ADMA]), due
(GFR) decreases to 10% of normal, fractional sodium excre- to diminished breakdown, has been identified as a
tion would need to increase from about 1% to 10% to functional issue leading to high BP [25]. Functional abnor-
achieve sodium balance. It has become clear that volume malities are also suspected in many BP-regulating systems,
overload is prevalent in CKD and hemodialysis patients including the sensitivity to and handling of, among other
now that reliable bio-impedance assessments are widely factors, natriuretic peptides, eiconasoids, and endothelin.
available [18, 19]. None of these potential mechanisms have currently been
SNS activity is increased in CKD and ESRD. Initial obser- translated into therapy backed up by solid clinical
vations by Converse demonstrated that nephrectomy investigation.
D
ifferential Diagnosis A strong body of evidence supports the use of antihyper-
tensive therapy, specifically with ACEi or ARBs, in patients
A conundrum in the evaluation of the patient with HTN with diabetic nephropathy (Table 35.3). In 1993, the first
and CKD is that once CKD has developed, renal parenchy- large randomized control trial (RCT) to support this com-
mal HTN can obscure other forms of secondary HTN. This pared captopril versus placebo in preventing doubling of
is difficult to decipher; however, it remains important if creatinine or reaching ESRD as primary outcome in patients
one wants to treat the HTN effectively and decrease the with diabetic nephropathy [30]. The study demonstrated a
progression of kidney failure. For example, if a primary decrease of 48% in reaching the outcome over the 3 years of
hyperaldosteronism has led to long-standing HTN and tar- follow-up. This study was followed by the Irbesartan in
get organ damage including CKD, it is important that this Diabetic Nephropathy Trial (IDNT) studying patients with
is diagnosed and treated accordingly. However, diagnostic type 2 diabetes, which demonstrated a better protection by
algorithms for primary hyperaldosteronism in moderate to Irbesartan for kidney disease progression than placebo or
advanced CKD do not have a solid evidence base, so one amlodipine [34] for a composite endpoint of doubling of
falls back on clinical evaluation and judgment: (relative) creatinine or all-cause mortality. Similar findings were
hypokalemia, alkalosis, and treatment resistance. If reported in the Reduction of Endpoints in NIDDM with the
strongly elevated, a Plasma Renin activity (PRA) level Angiotensin II Antagonist Losartan (RENAAL) trial; here,
might be helpful. Visualization of the adrenal glands and a 16% reduction in risk to reach a composite renal outcome
adrenal vein sampling will require an open discussion (doubling of creatinine or reaching ESKD) and death was
between care provider and patient. observed in patients with diabetic nephropathy treated with
Renal artery stenosis also leads to diagnostic and thera- losartan versus placebo [35]. A post hoc analysis of the
peutic dilemmas. Following the negative results of the RENAAL trial for renal outcomes separately indicated a
STAR, ASTRAL, and CORAL trials about angioplasty and benefit for patients with a high pulse pressure but was not
stent placement regarding treatment BP regulation and straightforward in its interpretation [38].
renal function [26–28], there is a strong reluctance to assess Regarding nondiabetic nephropathy, there is also evi-
whether renal artery stenosis is present. If there is a suspi- dence that BP lowering slows progression of kidney dis-
cion of renal artery stenosis, there are at least three reasons ease. The aim of the African American Study of Kidney
to do additional investigations toward confirmation, Disease and Hypertension (AASK) trial was to compare
despite these trials. First, fibromuscular dysplasia can pre- intensive versus less intensive BP lowering in African
sent with impaired renal function and HTN, and is poten- Americans with hypertensive kidney injury [36]. While
tially resolvable [29]. Second, the three trials did not the trial could not conclude that more intensive BP lower-
address whether patients with a rapid deterioration of ing had additional benefit to prevent the major outcomes
renal function or with flash pulmonary edema could ben- (rate change of eGFR and progression of CKD or all-cause
efit from angioplasty and stent placement. Third, perform- mortality), this trial concluded that there was specific ben-
ing additional investigations could be performed if the efit of an ACEi above a beta-blocker or calcium channel
kidney failure is unexplained. Although not very common,
renal artery stenosis can present with a nephrotic range
proteinuria. If kidney function is stable and BP is control- Table 35.2 Secondary causes of hypertension to be considered
in the patient with CKD.
lable, however, the three trials provide strong evidence not
to do additional diagnostic testing or an intervention.
System Cause
Table 35.2 lists a number of diagnoses that should be
considered before HTN in a patient with CKD is labeled as Vascular Atherosclerotic renal artery stenosis
renal parenchymal HTN. Fibromuscular dysplasia
Aortic coarctation
Endocrine Primary hyperaldosteronism
Secondary hyperaldosteronism
oals of Treatment of Hypertension
G
Cushing’s syndrome
in the Patient with CKD Congenital adrenal hyperplasia
Tubular transport Liddle’s syndrome
Prevention of progression of CKD by HTN is one goal in
Gordon’s syndrome
the treatment of the HTN. There has been overwhelming
Other Obstructive sleep apnea
evidence accumulated over the last 25 years supporting
Severe obesity
that fact that BP control reduces the progression of CKD.
Table 35.3 Major trials on hypertension treatment in patients with CKD.
Number of
randomized Duration of
Study patients Primary aim Study population Study design Outcomes evaluated follow-up Results References
The effect of 409 To determine whether Age: 18–49; IDDM for DBRCT in 30 centers; Primary: doubling of Median: Risk reduction for the [30]
angiotensin- captopril has at least 7 years; serum captopril 25 mg PO TID baseline serum 3 yr primary endpoint was
converting kidney-protecting creatinine 2.5 mg/dl; (n = 207), placebo (n = 202); creatinine to at least 48% in the captopril
enzyme properties urine protein excretion goal SBP <140 mmHg, DBP 2.0 mg/dl group
inhibition on independent of its 500 mg/24 h <90 mmHg Secondary: length of Risk reduction was 50%
diabetic effect on BP in time to the combined for the combined
nephropathy diabetic nephropathy endpoints of death, endpoint in the captopril
dialysis, and group
transplantation and
changes in kidney
function
The effects of Study 1:585; To test whether Age: 18–70; serum RCT in 15 centers Primary outcome: Mean: No difference in decline [31]
dietary protein Study 2:255 restricting protein and creatinine (M) 1.4–7.0, Groups: usual ( 107 mmHg) decline in GFR 2.2 years in GFR in both studies
restriction and controlling HTN delay (F) 1.2–7.0 mg/dl; MAP or low MAP ( 92 mmHg); measured by Among patients with
BP control on the progression of 125 mmHg; Study usual or low-protein diet I-iothalamate higher baseline
the progression kidney disease 1–GFR 25–55 ml/min, (Study 1) and low and very clearance, stratified by proteinuria, the low-BP
of CKD (MDRD) Study 2–GFR 13–24 ml/ low protein diet (Study 2) BP and proteinuria group had a significantly
min groups slower rate of GFR
Insulin-requiring decline in both studies
diabetics and those
with proteinuria >10 g/
day were excluded
Effect of ramipril 352 To test the hypothesis Age: 18–70; Cr Cl DBRCT Primary outcome: rate Stratum 1: Stratum 1: GFR decline [32, 33]
on decline in that glomerular 20–70 ml/min with Stratum 1: urine protein of GFR decline Median: per month not significantly
GFR and risk of protein traffic, and its variation <30% in 1–2.9 g/24 h (n = 186) measured by iohexol 31 mo different; significant
terminal kidney modification by an previous 3 months; Stratum 2: urine protein clearance reduction in proteinuria
failure in ACEi, influences urine protein excretion 3 g/24 h (n = 166) Secondary outcomes: Stratum 2: and progression to ESRD
proteinuric, kidney disease >1 g/24 h for at least urinary protein Mean: in the ramipril group
nondiabetic progression 3 months; no ACEi for Randomized to ramipril
(1.25 mg) or placebo excretion and serum 16 mo Stratum 2: Decrease in
nephropathy at least 2 months before creatinine the GFR decline per
(REIN) study Dose increased every concentration month was significantly
2 weeks for a goal DBP lower in the ramipril than
<90 mmHg Intention-to-treat
analyses the placebo group (0.53
Other agents used as needed vs. 0.88 ml/min, P = 0.03)
to achieve goal BP Significant reduction in
proteinuria in the
ramipril group that
correlated inversely with
GFR decline
(Continued)
c35.indd 577 09-12-2022 16:07:28

Number of
Renoprotective 1715 To determine whether Age: 30–70; type 2 DBRCT in 210 clinical Primary endpoint: Mean: Risk of the primary [34]
effect of the use of an ARB or a diabetes; HTN–BP centers; randomized to composite of a 2.6 years composite endpoint in
irbesartan in CCB would provide >135/85 or use of irbesartan (n = 579, dose doubling of the the irbesartan group was
patients with protection against the antihypertensive titration 75–300 mg/day); baseline serum 20% and 23% lower than
nephropathy due progression of agents; urinary protein amlodipine (n = 567, dose creatinine, onset of the placebo (P = 0.02)
to type 2 diabetes nephropathy due to excretion of at least titration 2.5–10 mg/day); or ESRD, or death and amlodipine
(IDNT) type 2 diabetes beyond 900 mg/24 h; serum placebo (n = 569) Secondary outcome: (P = 0.006) groups
that attributable to the creatinine: (M) combined respectively
lowering of BP 1.2–3.0 mg/dl, (F) cardiovascular No significant differences
1–3 mg/dl endpoint in death rates from any
cause or the CV
composite endpoint
Effects of BP 1513 To assess the role of Age: 31–70, type 2 DBRCT, multinational study, Primary composite Mean: The losartan group had a [35]
level on the angiotensin-II- diabetes and randomized to losartan endpoint: time to 3.4 years 16% risk reduction in the
progression of receptor antagonist nephropathy (urine (50–100 mg daily) or placebo, doubling of serum primary endpoint
diabetic losartan in patients albumin/creatinine stratified by baseline creatinine, ESRD, or (P=0.02 vs. placebo).
nephropathy with type 2 diabetes ratio of 300 mg/g or proteinuria (urine albumin- death, whichever Similar rates of death and
(RENAAL) and nephropathy proteinuria of 0.5 g/ to-creatinine ratio occurred first cardiovascular morbidity
day); serum creatinine <2000 mg/g or 2000 mg/g) Secondary endpoints: and mortality in the two
1.3–3.0 mg/dl Additional open-label agents composite of morbidity groups
used as needed for goal BP and mortality from Losartan had a 35%
<140/90 mmHg cardiovascular causes, reduction in proteinuria
proteinuria, and the (P < 0.001, compared to
rate of progression of placebo)
kidney disease
Effect of BP 1094 To compare effects of Age: 17–80; African 3 × 2 factorial design; Primary endpoint: rate 3–6.4 years There was no significant [36, 37]
lowering and two levels of BP Americans; HTN; GFR randomized to usual of change in GFR difference in mean GFR
antihypertensive control and three 20–65 ml/min; (102–107 mmHg) or lower (GFR slope–acute and slope from baseline
drug class on antihypertensive drug nondiabetics (<92 mmHg) MAP goal chronic) through 4 years in the
progression of classes on GFR Treatment with metoprolol, Secondary clinical two BP groups and the
hypertensive decline in African 50–200 mg/day; ramipril, composite outcome: drug group comparisons
kidney disease Americans with HTN 2.5–10 mg/day; or GFR reduction by 50% The secondary clinical
(AASK) amlodipine, 5–10 mg/day or by 25 ml/min from composite outcome was
assigned in 2 : 2 : 1 ratio baseline, ESRD, or similar in the two BP
Additional open-labeled death groups, but the ramipril
agents added sequentially Other secondary group had a 22%
outcomes: urine protein reduction compared to
excretion, all CVE the other drug groups
c35.indd 578 09-12-2022 16:07:28

Effects of BP 1513 To evaluate the impact Age: 31–70; type 2 DBRCT, multinational study, Primary composite Mean: Baseline SBP of 140- [38]
level on of baseline and diabetes and randomized to losartan endpoint: time to 3.4 years 159 mmHg, compared to
progression of treated SBP, DBP, and nephropathy (urine (50–100 mg daily) or placebo, doubling of serum SBP <130 mmHg,
diabetic PP to assess the albumin/creatinine stratified by baseline creatinine, ESRD, or increased risk for ESRD
nephropathy specific effect of ratio of 300 mg/g or proteinuria (urine albumin- death, whichever or death by 38%
(post hoc losartan and the proteinuria of 0.5 g/ to-creatinine ratio occurred first (P = 0.05)
analyses of implications of day); serum creatinine <2000 mg/g or 2000 mg/g) Patients randomized to
RENAAL) dihyropyridine 1.3–3.0 mg/dayl Additional open-label agents the losartan group with a
calcium channel used as needed for goal BP baseline PP >90 mmHg
blockers as concurrent <140/90 mmHg had a 35.5% risk
therapy on composite reduction for ESRD or
and individual death (P = 0.02)
outcomes in patients compared to the placebo
with type 2 diabetes group
and nephropathy
BP control for 338 To assess the effect of Age: 18–70; nondiabetic Multicenter RCT Primary outcome: time Median: No significant difference [39]
renoprotection intensified vs. nephropathy with All received background to ESRD over 19 months in the progression to
in patients with conventional BP persistent proteinuria ACEi therapy with ramipril 36 months follow-up ESRD in the two BP
nondiabetic control on progression (>1 g/24 h for at least Randomized to conventional Secondary outcome: to groups (P = 0.99)
chronic kidney to ESRD in patients 3 months); proteinuria (DBP <90 mmHg) or compare the effects of No difference in the
disease (REIN with nondiabetic of 1-3 g if Cr Cl <45 ml/ intensified (BP two different levels of secondary outcomes
-2) proteinuric min, and proteinuria <130/80 mmHg) BP control BP control on GFR between the two groups
nephropathies 3 g if Cr Cl <70 ml/ decline, residual
min Add-on therapy with
felodipine in the intensified proteinuria, and fatal
BP group and nonfatal CVE
Kidney 33 357 To determine whether, Age: 55; stage 1 or 2 DBRCT, post hoc analyses, Clinical kidney Mean: No significant difference [40]
outcomes in in high-risk HTN with at least one randomized to receive outcomes: 4.9 years in the outcomes in
high-risk hypertensive patients other CAD risk factor; chlorthalidone (12.5–25 mg/ development of ESRD patients taking
hypertensive with a reduced GFR, serum creatinine day), amlodipine (2.5–10 mg/ or death from kidney chlorthalidone compared
patients treated treatment with a CCB <2 mg/dl day), or lisinopril (10–40 mg/ disease to those who received
with an ACEi or or an ACEi lowers the day) in a 1.7 : 1 : 1 ratio Composite endpoint of lisinopril or amlodipine,
a CCB vs. a incidence of kidney Goal BP <140/90 in each arm ESRD or 50% decline irrespective of the
diuretic disease outcomes by adding open-label agents in GFR from baseline; baseline GFR
(ALLHAT post compared with as needed mean GFR during
hoc analyses) treatment with a Stratified by baseline eGFR study follow-up
diuretic
(Continued)
c35.indd 579 09-12-2022 16:07:28

Number of
Effect of a lower 840 To assess the effects of Age: 18–70; serum Long term follow-up of Outcomes: kidney Approx. The adjusted hazard [41]
target BP on the a low target BP on creatinine (M) 1.4–7.0, patients from 1993 to 2000 failure and a 7 years ratios in the low
progression of kidney failure and (F) 1.2–7.0 mg/dl; MAP after the MDRD study ended composite outcome of (1993– compared with the usual
kidney disease: all-cause mortality 125 mmHg; Study A: (see above for MDRD study kidney failure or 2000) target BP group were 0.68
long-term GFR 25–55 ml/min, design) all-cause mortality, and 0.77 for development
follow-up of the Study B: GFR 13–24 ml/ through extended of kidney failure and the
MDRD study (post min follow-up composite outcome,
hoc analyses) respectively
In the Action to 4678 Intense vs. standard Type 2 diabetes Selection of the 2 × 2 design Primary outcome: Mean No clear cardiovascular [42]
Control therapy in patients mellitus; hemoglobin of the ACCORD study: composite endpoint, follow-up: benefit of intense vs.
Cardiovascular with type 2 diabetes A1C 7.5% standard (SBP <140 mmHg) including both 4.7 years standard therapy could
Risk in Diabetes Age 40 years with vs. intense BP (SBP nonfatal and fatal CVD be demonstrated,
blood pressure CVD; age 55 years with <120 mmHg) control; events (first although the risk of
trial any of the following: standard intense diabetes occurrence of nonfatal stroke was reduced by
(ACCORD-BP); atherosclerosis, control myocardial infarction, 38% (HR 0.623, 95% CI
kidney subgrou albuminuria, LVH, 2 nonfatal stroke, and 0.361-1.074)
p analysis CV risk factors cardiovascular death)
(dyslipidemia, HTN,
smoking, or obesity)
Exclusion criteria: BMI
>45 kg/m2; creatinine
>1.5 mg/dl (132.6 umol/l)
Systolic Blood 9361 Intensive treatment Age: >50 years; SBP Open-label RCT in 201 sites Primary outcome: Median SBP of less than [43]
Pressure (SBP <120 mmHg) vs. 130 mmHg and composite outcome of follow-up: 120 mmHg, as compared
Intervention standard (SBP increased myocardial infarction, 3.26 years with less than 140 mmHg,
Trial (SPRINT) <140 mmHg) cardiovascular risk, but acute coronary resulted in lower rates of
treatment without diabetes syndrome not resulting fatal and nonfatal major
in myocardial cardiovascular events and
infarction, stroke, acute death from any cause,
decompensated heart although significantly
failure, or death from higher rates of some
cardiovascular causes adverse events were
Kidney outcome in observed in the intensive-
participants with CKD at treatment group
baseline: composite of a No difference in kidney
decrease in the eGFR of outcome in intense BP
50% or more (confirmed treatment vs. standard
by a subsequent group
laboratory test) or the
development of ESRD
ACEi: angiotensin-converting enzyme inhibitor; BMI: body mass index; BP: blood pressure; CCB: calcium channel blocker; CKD; CrCl: creatinine clearance; (also cr cl); CV: cardiovascular; CVD: cardiovascular
disease; CVE: cardiovascular event; DBP: diastolic blood pressure; DBRTC: double-blind randomized controlled trial; ESRD: end-stage renal disease; GFR: glomerular filtration rate; (also eGFR: estimated glomerular
filtration rate;); HTN: hypertension; IDDM: insulin dependent diabetes mellitus; M/F: male/female; MAP: mean arterial pressure; PP: pulse pressure; RCT: randomized controlled trial; SBP: systolic blood pressure.
Source: Modified from chapter on this subject in the previous edition (authors: Satyan and Agarwal).
c35.indd 580 09-12-2022 16:07:28

Goals of Treatment of Hypertension in the Patient with CK 581
blocker in this setting. The smaller Ramipril Efficacy In lower incidence of the composite outcome (defined as
Nephropathy (REIN) study addressed a similar question myocardial infarction, other acute coronary syndromes,
in proteinuric, nondiabetic nephropathy [32]. It studied stroke, heart failure, or death from cardiovascular causes).
whether BP lowering with an ACEi or with other therapy The hazard ratio with intensive treatment was 0.75 (95% CI
had comparable effect on proteinuria, slowing down kid- 0.64–0.89). A post hoc analysis revealed that in about 2000
ney disease progression and preventing ESRD at the same patients with CKD stages 3–4 the decrease in cardiovascu-
level of BP control. Again, ACEi more so than other ther- lar outcome was similar, with a hazard ratio of 0.72 (95% CI
apy decreased progression and prevented ESRD in the 0.63–1.05), and the hazard ratio for all-cause mortality was
group with the lower level of proteinuria (the group with 0.72 (95% CI 0.53–0.99) [46]. This was not accompanied by
the higher level of proteinuria was discontinued because more adverse events. Regarding the renal outcome, the
of strong benefit of ACEi during an interim analysis). In composite of 50% decrease in eGFR from baseline or
the patients receiving the ACEi, the percentage reduction ESRD was not different between the intensive and less
in proteinuria was inversely correlated with decline in intensive groups (HR 0.90; 95% CI 0.44–1.83). As expected
GFR. This review of trials is not exhaustive and more trials because of the larger drop in BP, the intensive group had a
can be found in the literature, all stressing the importance slightly higher rate of change in eGFR at 6 months into the
of BP control to prevent CKD progression. study (−0.47 vs. −0.32 ml/min per 1.73 m2 per year,
The other very important goal of treatment of HTN is the P < 0.03). So, in this hallmark trial, more intensive treat-
prevention of cardiovascular disease. In the Atherosclerosis ment of patients with cardiovascular risk and CKD for
Risk in Communities (ARIC) study cohort, progressive car- more intensive BP lowering led to lower cardiovascular
diovascular risk was reported in patients with mild to events, was not associated with higher adverse events, and
severe kidney disease that was independent from a large was similar in kidney outcome. In light of the above, for
number of other cardiovascular risk factors [44]. That this cardiovascular more than kidney treatment goals, a lower
is competing with the risk to reach ESRD was demon- BP target in CKD patients seems preferable.
strated in an epidemiological study showing that 1% of
>11 000 patients with CKD stage 3 reached kidney
Lifestyle Changes and Blood Pressure Control
replacement therapy, meanwhile 24% of the cohort passed
in CKD
away from cardiovascular disease; for CKD stage 4 these
numbers were 18 and 46%, respectively [45]. This very Dietary Sodium
clearly underscores the need to take both renal and cardio- Despite the fact that ECFV expansion due to high sodium
vascular protection into consideration for treatment. As intake and impaired sodium handling by the kidneys in CKD
mentioned, the IDNT and RENAAL trials support treat- are considered key factors in the pathogenesis of HTN in
ment of HTN for a composite renal and all-cause mortality CKD patients, information on the role of sodium intake both
outcome, the latter mainly being cardiovascular mortality. in the progression of kidney disease and on cardiovascular
The Action to Control Cardiovascular Risk in Diabetes- event rate in CKD patients is limited. Regarding progression
Blood Pressure (ACCORD-BP) trial investigated the effect of kidney disease, Lambers Heerspink et al. demonstrated in
of intense versus standard therapy in patients with type 2 an analysis of data pooled from the RENAAL and IDNT tri-
diabetes, but it excluded patients with a serum creatinine als that a low sodium intake potentiated on the effect of angi-
level of more than 1.5 mg per deciliter (132.6 μmol/l). In otensin blocker not only to slow CKD progression but also to
the trial patients with mild to moderate CKD (stages 1–3B, decrease cardiovascular events [47]. Further to cardiovascu-
n = 4678) no clear cardiovascular benefit of intense versus lar outcomes, Mills et al. demonstrated that the higher uri-
standard therapy could be demonstrated, although the risk nary sodium excretion from multiple 24-hour urine
of stroke was reduced by 38% (HR 0.623; 95% CI 0.361– collections in the CRIC study was associated with increased
1.074). The authors concluded that, given that the trial was risk of cardiovascular events (composite of congestive heart
not designed and powered to this purpose, an effect of failure, stroke, and myocardial infarction) [48]. There are no
intense BP control in patients with CKD could not be large-scale RCTs on the influence of dietary sodium restric-
excluded [42]. tion specifically aimed at CKD patients. It is necessary to
In the Systolic Blood Pressure Intervention Trial mention the Dietary Approaches to Stop Hypertension
(SPRINT), more than 9000 patients were randomized to (DASH) study since it provides solid evidence on the effect of
intensive versus less intensive BP treatment where average dietary sodium restriction to lower BP; however, this study
systolic BP of 121 and 136 mmHg, respectively, were was not targeted to assess progression of CKD [49]. In 2015,
achieved [43]. When the trial was stopped early, more a Cochrane review was published summarizing the effects of
intensive treatment was associated with a significantly dietary sodium restriction on renal sodium excretion (to
Table 35.4 Lifestyle management in the treatment of hypertension in chronic kidney disease (KDIGO Clinical Practice Guidelines).
Strength of
Management Recommendation Certainty of evidence recommendation
Lifestyle modification Lower or maintain salt intake to <90 mmol/day (equivalent to C, low 1

Sodium/salt <2 g sodium/day or < 5 g sodium chloride/day)

Weight loss Achieve or maintain a healthy weight (BMI 20–25) D, very low 1

Exercise Aim to exercise for at least 30 minutes five times per week D, very low 1
evaluate whether the intervention was successful) and on disease are lacking. One small trial (n = 30) randomized
kidney function [50]. Included in the analysis were eight CKD patients to daily, 30-minute, aerobic exercise or
studies, mostly nonrandomized, with the number of subjects usual care but failed to show that this slowed down pro-
ranging from 14 to 52 per arm. The three short-term gression of CKD [54]. Another observational study
(<4 weeks) and four out of the five long-term ( 4 weeks) showed a 2.8% difference in eGFR decline between phys-
studies indeed showed a significant decrease in urinary ically active people versus those who were sedentary [55].
sodium excretion. Regarding kidney function, low dietary This illustrates the need for adequately powered trials
sodium did not decrease eGFR (n = 2 studies), creatinine with sufficient follow-up to determine the potential ben-
clearance (n = 3), renal plasma flow (n = 1), or filtration frac- efits of exercise to slow down progression of CKD. Of
tion (n = 1); obviously the number of studies is very limited. note, exercise has been shown to decrease cardiovascular
BP and protein excretion were reduced by dietary sodium event rates in numerous other patient populations with
restriction. Only one study measured a decrease (0.8 l) in cardiovascular risk.
ECFV by reduction of sodium intake [51]; unfortunately, no A comprehensive review of lifestyle changes for the
changes in creatinine or eGFR were reported, so changes in treatment of BP can be found elsewhere [56] (table 35.4).
volume status could not be related to changes in renal func-
tion. Although the data is lacking to strongly support that ACEi and ARBs versus Other Medication in
dietary sodium restriction in patients with CKD leads to a Hypertension in CKD
decrease in CKD progression and cardiovascular events, the
There is abundant literature about the actions of angioten-
lack of evidence should not be considered as evidence that
sin II, besides intrinsic renal autoregulation, to maintain
such intervention could not be useful (table 35.4).
glomerular pressure when kidney perfusion pressure is
decreased. This is achieved by preferential efferent arteri-
Obesity and Weight Loss
olar vasoconstriction. Conversely, blockade of angiotensin
There are numerous studies in the literature about weight loss
II action by ACE inhibition or AT1 receptor antagonists can
and its effects on BP, whether by dietary or pharmacological
decrease glomerular pressure by a decrease in BP, but also
intervention or by bariatric surgery. There is also quite
by a preferential efferent vasodilation. In addition, blockade
some evidence to support that obesity can affect the func-
of angiotensin II can decrease the glomerular ultrafiltration
tion of the kidney [52]. Nevertheless, there is no solid evi-
coefficient. Not surprisingly, it has been shown that angio-
dence supporting that weight loss has a beneficial effect on
tensin in many different models of kidney failure can
kidney disease progression in CKD patients. Regarding
decrease kidney damage and proteinuria more so than
diabetic kidney disease, our current knowledge falls short
other agents, particularly calcium antagonists [57].
on how lifestyle, medical, and surgical interventions can
It is therefore rather surprising that clinical data on kid-
decrease CKD progression and whether there is a thresh-
ney protection of RAS inhibitors versus other agents is
old weight loss required for benefit [53] (table 35.4).
conflicting. There are several trial results suggesting that
RAS inhibitors are more effective in decreasing progres-
Exercise sion of nondiabetic and diabetic proteinuric and nonpro-
Exercise can lower BP, reportedly to a similar extent as a teinuric kidney disease compared to calcium channel
normal dose of an antihypertensive medication. However, blockers.
adequately powered RCTs reporting the independent There are several considerations around this issue of
effects of exercise to slow down progression of kidney effectiveness of RAS inhibitors versus other drugs. First,
Blood Pressure Targe 583
RAS inhibitors if dosed appropriately lead to an acute management of HTN [62]. The most recent European
decrease in glomerular filtration, due to the mechanism Society of Hypertension/European Society of Cardiology
explained above, and the absence of good autoregulation to guidelines now advise a target of 130–139 mmHg for sys-
offset the drop in BP in CKD patients. Administration of tolic BP but do not mention a target for diastolic BP [63].
calcium antagonists has the opposite effect as they lead to The debate about the right BP target is ongoing.
an increase in glomerular filtration. If progression of CKD
is studied, one should either study the changes in GFR A Therapeutic Approach to Treatment of
after the initial changes in BP and glomerular filtration, or Hypertension in CKD
one should study the GFR after discontinuation of the
Although a physiology-based approach for more individually
study drugs. This is nicely illustrated in the report by
targeted treatment of HTN is not currently standard of care,
Bakris, where he demonstrated a clear increase in GFR
such an approach may be considered in the management of
after discontinuation of the ACEi [58]. A related problem is
patients who are identified as having truly resistant and
that the exact same BP was not achieved in all studies.
difficult-to-control HTN. To effectively target a specific patho-
While in the IDNT study, for example, the difference was
physiological system, it is necessary to first demonstrate that
not more 2 mmHg [34], one cannot exclude that this
the system is activated (using plasma hormone levels, physi-
explains (the absence) of differences between two study
ologic measurements, or function tests). To date, key regula-
drugs. A third issue is whether other medications were
tory systems are not fully delineated and a comprehensive
allowed to be changed to control BP, which could introduce
panel of “probes” is not yet available. With these limitations,
a confounder. Altogether the consensus seems to be that in
assessment and targeting of several major regulatory path-
diabetic and nondiabetic proteinuric kidney disease block-
ways in HTN development is considered in this section.
ade of the RAS is the preferred first-line approach, whereas
Measurement of plasma renin activity or measuring the
in nonproteinuric disease this is less conclusive.
acute fall in BP after captopril administration may be used
to predict BP response to RAS blockade [64]. Beyond the
Blood Pressure Target use of standard diuretic agents at standard doses, evalua-
tion of volume status and urinary salt excretion could be
No definitive conclusion can be reached about the best BP used to assess a particular patient’s salt-avidity and individ-
target for patients with CKD. On the one hand, there are no ual response to a diuretic agent and dosage. Evaluation of
studies that have reached a solid conclusion about the best volume status and salt intake and excretion is not routinely
BP level to prevent the progression of proteinuric and non- performed and is technically difficult. Application of objec-
proteinuric CKD. As mentioned, the SPRINT post hoc tive estimates of volume status using methods such as bio-
assessment of the effect of BP level on the progression of impedance may be a consideration. Measurement of
CKD did not reveal any benefit of lower BP goals [46]. That hemodynamic parameters and volume status using thoracic
said, the trial was not powered to identify any benefit for bio-impedance has been demonstrated to be an effective
specific etiologies of kidney failure. Similarly, the AASK approach to improve BP control in several small trials [65].
trial could not demonstrate an additional benefit of intense A next important step would be to monitor salt intake and
versus standard treatment of BP in African Americans with excretion by 24-hour urine sampling. Alternatively, a “salt
nondiabetic hypertensive kidney injury [36]. The calculator” has been developed, which consists of a survey
ACCORD-BP trial was unable to show this in patients with of ~25 questions and provides an estimate of daily salt
CKD stage 1–3B with diabetes [42]. Although there is no intake [66]. There is a body of evidence supporting the
RCT comparing standard versus strict BP control to slow importance of increased activity of the SNS as one of the
progression of CKD induced by IgA nephropathy, several key factors in HTN in general and in hypertensive kidney
observational studies support this [59, 60]. The Halt disease in particular [21, 67]. Unfortunately, the activity of
Progression of Polycystic Kidney Disease (HALT-PKD) trial the SNS cannot be easily measured. Methods such as heart
compared strict versus standard BP control in autosomal rate variability, plasma norepinephrine and catechola-
dominant polycystic kidney disease (ADPKD) patients mines [68], and muscle sympathetic nerve activity [21] are
(n = 558). Intense BP control did not affect the primary out- not suitable for daily practice, and it is unclear whether
come, percentage change in kidney volume, but it decreased these methods can reliably predict response to therapy.
the progression of CKD [61]. The clear observations from BP responses to mineralocorticoid receptor inhibition
the SPRINT study have triggered a large dispute about the reportedly are not related to plasma aldosterone concentra-
BP targets in patients with CKD. This has led to lowering of tions or aldosterone-to-renin ratios in patients already
the target to 130/80 mmHg in the American College of treated for hypertension [69, 70]. Only in untreated patients
Cardiology/American Heart Association guidelines for the was response to mineralocorticoid receptor inhibition linked
to plasma aldosterone levels [71] and inversely to plasma BP after addition of a low dose of spironolactone in patients
renin activity [72]. Regarding patients with rHTN, plasma with rHTN of 16.6/7.0, 23.9/9.7, and 26.0/10.7 mmHg at 1, 3,
aldosterone, aldosterone-to-renin ratios, and 24-hour urine and 6 months after starting the drug [73].
aldosterone levels were higher in those with rHTN versus Figure 35.2 illustrates an algorithm that can be followed
controls [65]. Moreover, another study reported decreases in to assess the effectiveness of targeted therapy for the
Assess baseline condition:

- shortness of breath, respiratory rate
- blood pressure, body weight, heart rate, edema
- plasma K, Bicarbonate, Creatinine, renin or PRA,
adosterone, ECG
- 24h urine sodium excretion
- urine protein to creatinine and/or albumin to
creatinine ratio
RAS ACEi/ARB medication
Assess response to ACEi/ARB:

if no change in BP, renin/PRA, plasma K or
- decrease in blood pressure? creatine and/or acute ACEi/test is positive:
- increase in renin or PRA?
- discuss adherence with patient
- increase in plasma K and creatinine?
- increase dose of ACEi/ARB
- negative acute ACEi test?
ECFV diuretic plus low salt diet
Assess response to diuretic: if remaining edema, no decrease in weight,

- decrease in blood pressure? no change in respiratory symptoms:
- decrease in body weight? - discuss adherence with patient
- decrease in edema? - increase dose of diuretic
- decrease in respiratory symptoms? - add or switch class of diuretic
SNS (alpha/+) beta blocker
Assess response to beta blocker: if no change in BP and/or heart rate:

- decrease in blood pressure? - discuss adherence with patient
- decrease in heart rate? - increase dose of drug
Aldo aldosterone antagonist
Assess response to aldosterone antagonist:

- decrease in blood pressure?
- increase in plasma K?
Figure 35.2 Physiology-based approach to treating hypertension in patients with CKD. Plasma K: plasm potassium concentration;
PRA: plasma renin activity; ECG: electrocardiogram; RAS: renin-anigotensin system; ACEi: angiotensin-converting enzyme inhibitor;
ARB: angiotensin receptor blocker; BP: blood pressure; ECFV: extracellular fluid volume; SNS: sympathetic nervous system;
Aldo: aldosterone.
Reference 585
hysiological driving forces of BP as discussed. Until more

p patients with CKD and to prevent progression of CKD
evidence becomes available to support the use of such tech- itself. BP assessment is more complicated and needs spe-
niques in clinical practice, the proposed approaches can be cial attention in patents with CKD. Since CKD by itself is
considered in cases of individual patients with resistant and an important determinant of disturbed BP control, second-
difficult-to-treat HTN to maximize targeted BP therapy. ary causes of HTN can easily be overlooked in this group.
Inhibition of the RAS and volume control are key to BP
management, next to lifestyle measures. Last but not least,
C
onclusion
BP treatment of the hypertensive CKD patient could bene-
Hypertension is not only prevalent in patients with CKD, it fit from a physiology-based approach.
is also a key factor to control cardiovascular events in
R
eferences
Kidney Disease Outcomes Quality I (2004). K/DOQI

1 9 Drawz, P.E., Brown, R., De Nicola, L. et al. (2018).
clinical practice guidelines on hypertension and Variations in 24-hour BP profiles in cohorts of patients
antihypertensive agents in chronic kidney disease. Am. J. with kidney disease around the world: the I-DARE study.
Kidney Dis. 43: S1–S290. Clin. J. Am. Soc. Nephrol. 13: 1348–1357.
2 Braam, B., Taler, S.J., Rahman, M. et al. (2017). 10 Liu, J.E., Roman, M.J., Pini, R. et al. (1999). Cardiac and
Recognition and management of resistant hypertension. arterial target organ damage in adults with elevated
Clin. J. Am. Soc. Nephrol. 12: 524–535. ambulatory and normal office blood pressure. Ann.
3 Carey, R.M., Calhoun, D.A., Bakris, G.L. et al. and on Intern. Med. 131: 564–572.
behalf of the American Heart Association Professional/ 11 Agarwal, R., Pappas, M.K., and Sinha, A.D. (2016).
Public Education and Publications Committee of the Masked uncontrolled hypertension in CKD. J. Am. Soc.
Council on Hypertension; Council on Cardiovascular and Nephrol. 27: 924–932.
Stroke Nursing; Council on Clinical Cardiology; Council 12 Scheppach, J.B., Raff, U., Toncar, S. et al. (2018). Blood
on Genomic and Precision Medicine; Council on pressure pattern and target organ damage in patients
Peripheral Vascular Disease; Council on Quality of Care with chronic kidney disease. Hypertension 72: 929–936.
and Outcomes Research; and Stroke Council (2018). 13 Gorostidi, M., Sarafidis, P.A., de la Sierra, A. et al. (2013).
Resistant hypertension: detection, evaluation, and Differences between office and 24-hour blood pressure
management: a scientific statement from the American control in hypertensive patients with CKD: a
Heart Association. Hypertension 72: e53–e90. 5,693-patient cross-sectional analysis from Spain. Am. J.
4 Egan, B.M., Zhao, Y., Axon, R.N. et al. (2011). Uncontrolled Kidney Dis. 62: 285–294.
and apparent treatment resistant hypertension in the 14 Chapman, A.B., Johnson, A., Gabow, P.A., and Schrier,
United States, 1988 to 2008. Circulation 124: 1046–1058. R.W. (1990). The renin-angiotensin-aldosterone system
5 Muntner, P., Davis, B.R., Cushman, W.C. et al. and and autosomal dominant polycystic kidney disease. N.
ALLHAT Collaborative Research Group (2014). Treatment- Engl. J. Med. 323: 1091–1096.
resistant hypertension and the incidence of cardiovascular 15 Stone, R.A., Tisher, C.C., Hawkins, H.K., and Robinson,
disease and end-stage renal disease: results from the R.R. (1974). Juxtaglomerular hyperplasia and
antihypertensive and lipid-lowering treatment to prevent hyperreninemia in progressive systemic sclerosis
heart attack trial (ALLHAT). Hypertension 64: 1012–1021. complicated acute renal failure. Am. J. Med. 56:
6 Thomas, G., Xie, D., Chen, H.Y. et al. CRIC Study 119–123.
Investigators (2016). Prevalence and prognostic 16 Cogan, M.G. (1990). Angiotensin II: a powerful controller
significance of apparent treatment resistant hypertension of sodium transport in the early proximal tubule.
in chronic kidney disease: report from the chronic renal Hypertension 15: 451–458.
insufficiency cohort study. Hypertension 67: 387–396. 17 Koomans, H.A., Roos, J.C., Dorhout Mees, E.J., and
7 Judd, E. and Calhoun, D.A. (2014). Apparent and true Delawi, I.M. (1985). Sodium balance in renal failure. A
resistant hypertension: definition, prevalence and comparison of patients with normal subjects under
outcomes. J. Hum. Hypertens. 28: 463–468. extremes of sodium intake. Hypertension 7: 714–721.
8 Bobrie, G., Chatellier, G., Genes, N. et al. (2004). 18 Kalainy, S., Reid, R., Jindal, K. et al. (2015). Fluid volume
Cardiovascular prognosis of “masked hypertension” expansion and depletion in hemodialysis patients lack
detected by blood pressure self-measurement in elderly association with clinical parameters. Can. J. Kidney
treated hypertensive patients. JAMA 291: 1342–1349. Health Dis. 2: 54.
19 Hecking, M., Moissl, U., Genser, B. et al. (2018). Greater 33 Ruggenenti, P., Perna, A., Gherardi, G. et al. (1999).
fluid overload and lower interdialytic weight gain are Renoprotective properties of ACE-inhibition in non-
independently associated with mortality in a large diabetic nephropathies with non-nephrotic proteinuria.
international hemodialysis population. Nephrol. Dial. Lancet 354: 359–364.
Transplant. 33: 1832–1842. 34 Lewis, E.J., Hunsicker, L.G., Clarke, W.R. et al. (2001).
20 Converse, R.L. Jr., Jacobsen, T.N., Toto, R.D. et al. (1992). Renoprotective effect of the angiotensin-receptor
Sympathetic overactivity in patients with chronic renal antagonist irbesartan in patients with nephropathy
failure. N. Engl. J. Med. 327: 1912–1918. due to type 2 diabetes. N. Engl. J. Med. 345:
21 Ligtenberg, G., Blankestijn, P.J., Oey, P.L. et al. (1999). 851–860.
Reduction of sympathetic hyperactivity by enalapril in 35 Brenner, B.M., Cooper, M.E., de Zeeuw, D. et al. (2001).
patients with chronic renal failure. N. Engl. J. Med. 340: Effects of losartan on renal and cardiovascular outcomes
1321–1328. in patients with type 2 diabetes and nephropathy. N. Engl.
22 Smithwick, R.H. and Thompson, J.E. (1953). J. Med. 345: 861–869.
Splanchnicectomy for essential hypertension; results in 36 Wright, J.T. Jr., Bakris, G., Greene, T. et al. (2002). Effect
1,266 cases. J. Am. Med. Assoc. 152: 1501–1504. of blood pressure lowering and antihypertensive drug
23 Bhatt, D.L., Kandzari, D.E., O’Neill, W.W. et al. and class on progression of hypertensive kidney disease:
SYMPLICITY HTN-3 Investigators (2014). A controlled results from the AASK trial. JAMA 288: 2421–2431.
trial of renal denervation for resistant hypertension. N. African American Study of Kidney Disease and
Engl. J. Med. 370: 1393–1401. Hypertension Study Group
24 Grassi, G., Arenare, F., Pieruzzi, F. et al. (2009). 37 Agodoa, L.Y., Appel, L., Bakris, G.L. et al., African
Sympathetic activation in cardiovascular and renal American Study of Kidney Disease and Hypertension
disease. J. Nephrol. 22: 190–195. Study Group (2001). Effect of ramipril vs amlodipine on
25 Vallance, P., Leone, A., Calver, A. et al. (1992). renal outcomes in hypertensive nephrosclerosis: a
Accumulation of an endogenous inhibitor of nitric oxide randomized controlled trial. JAMA 285: 2719–2728.
synthesis in chronic renal failure. Lancet 339: 572–575. 38 Bakris, G.L., Weir, M.R., Shanifar, S. et al. and RENAAL
26 Investigators, A., Wheatley, K., Ives, N. et al. (2009). Study Group (2003). Effects of blood pressure level on
Revascularization versus medical therapy for renal-artery progression of diabetic nephropathy: results from the
stenosis. N. Engl. J. Med. 361: 1953–1962. RENAAL study. Arch. Intern. Med. 163: 1555–1565.
27 Bax, L., Woittiez, A.J., Kouwenberg, H.J. et al. (2009). 39 Ruggenenti, P., Perna, A., Loriga, G. et al. (2005).
Stent placement in patients with atherosclerotic renal Blood-pressure control for renoprotection in patients with
artery stenosis and impaired renal function: a randomized non-diabetic chronic renal disease (REIN-2): multicentre,
trial. Ann. Intern. Med. 150: 840–848. W150–1. randomised controlled trial. Lancet 365: 939–946.and
28 Cooper, C.J., Murphy, T.P., Cutlip, D.E. et al. (2014). REIN-2 Study Group
Stenting and medical therapy for atherosclerotic renal- 40 Rahman, M., Pressel, S., Davis, B.R. et al. (2005). Renal
artery stenosis. N. Engl. J. Med. 370: 13–22. outcomes in high-risk hypertensive patients treated with
29 Slovut, D.P. and Olin, J.W. (2004). Fibromuscular an angiotensin-converting enzyme inhibitor or a calcium
dysplasia. N. Engl. J. Med. 350: 1862–1871. channel blocker vs a diuretic: a report from the
30 Lewis, E.J., Hunsicker, L.G., Bain, R.P., and Rohde, R.D. antihypertensive and lipid-lowering treatment to prevent
(1993). The effect of angiotensin-converting-enzyme heart attack trial (ALLHAT). Arch. Intern. Med. 165:
inhibition on diabetic nephropathy. The Collaborative 936–946.
Study Group. N. Engl. J. Med. 329: 1456–1462. 41 Sarnak, M.J., Greene, T., Wang, X. et al. (2005). The effect
31 Klahr, S., Levey, A.S., Beck, G.J. et al. (1994). The effects of a lower target blood pressure on the progression of
of dietary protein restriction and blood-pressure control kidney disease: long-term follow-up of the modification
on the progression of chronic renal disease. Modification of diet in renal disease study. Ann. Intern. Med. 142:
of Diet in Renal Disease Study Group. N. Engl. J. Med. 342–351.
330: 877–884. 42 Papademetriou, V., Zaheer, M., Doumas, M. et al. and
32 The GISEN Group (Gruppo Italiano di Studi ACCORD Study Group (2016). Cardiovascular outcomes
Epidemiologici in Nefrologia) (1997). Randomised in action to control cardiovascular risk in diabetes:
placebo-controlled trial of effect of ramipril on decline in impact of blood pressure level and presence of kidney
glomerular filtration rate and risk of terminal renal disease. Am. J. Nephrol. 43: 271–280.
failure in proteinuric, non-diabetic nephropathy. Lancet 43 SPRINT Reasearch Group, Wright, J.T. Jr., Williamson,
349: 1857–1863. J.D. et al. (2015). A randomized trial of intensive versus
Reference 587
standard blood-pressure control. N. Engl. J. Med. 373: management: what are the data? Kidney Int. 79:
2103–2116. 1061–1070.
44 Manjunath, G., Tighiouart, H., Ibrahim, H. et al. (2003). 57 Griffin, K.A. and Bidani, A.K. (2006). Progression of
Level of kidney function as a risk factor for renal disease: renoprotective specificity of renin-
atherosclerotic cardiovascular outcomes in the angiotensin system blockade. Clin. J. Am. Soc.
community. J. Am. Coll. Cardiol. 41: 47–55. Nephrol. 1: 1054–1065.
45 Keith, D.S., Nichols, G.A., Gullion, C.M. et al. (2004). 58 Bakris, G.L. and Weir, M.R. (2000). Angiotensin-
Longitudinal follow-up and outcomes among a converting enzyme inhibitor-associated elevations in
population with chronic kidney disease in a large serum creatinine: Is this a cause for concern? Arch.
managed care organization. Arch. Intern. Med. 164: Intern. Med. 160: 685–693.
659–663. 59 Kanno, Y., Okada, H., Saruta, T., and Suzuki, H. (2000).
46 Cheung, A.K., Rahman, M., Reboussin, D.M. et al. and Blood pressure reduction associated with preservation of
SPRINT Research Group (2017). Effects of intensive BP renal function in hypertensive patients with IgA
control in CKD. J. Am. Soc. Nephrol. 28: 2812–2823. nephropathy: a 3-year follow-up. Clin. Nephrol. 54:
47 Lambers Heerspink, H.J., Holtkamp, F.A., Parving, H.H. 360–365.
et al. (2012). Moderation of dietary sodium potentiates 60 Osawa, Y., Narita, I., Imai, N. et al. (2001). Determination
the renal and cardiovascular protective effects of of optimal blood pressure for patients with IgA
angiotensin receptor blockers. Kidney Int. 82: nephropathy based on renal histology. Hypertens. Res. 24:
330–337. 89–92.
48 Mills, K.T., Chen, J., Yang, W. et al. and Chronic Renal 61 Schrier, R.W., Abebe, K.Z., Perrone, R.D. et al. and
Insufficiency Cohort Study Investigators (2016). Sodium HALT-PKD Trial Investigators (2014). Blood pressure in
excretion and the risk of cardiovascular disease in early autosomal dominant polycystic kidney disease. N.
patients with chronic kidney disease. JAMA 315: Engl. J. Med. 371: 2255–2266.
2200–2210. 62 Whelton, P.K., Carey, R.M., Aronow, W.S. et al. (2018).
49 Sacks, F.M., Svetkey, L.P., Vollmer, W.M. et al. and 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/
DASH-Sodium Collaborative Research Group (2001). ASPC/NMA/PCNA guideline for the prevention,
Effects on blood pressure of reduced dietary sodium and detection, evaluation, and Management of High Blood
the dietary approaches to stop hypertension (DASH) diet. Pressure in adults: executive summary: a report of the
DASH-Sodium Collaborative Research Group. N. Engl. J. American College of Cardiology/American Heart
Med. 344: 3–10. Association task force on clinical practice guidelines.
50 McMahon, E.J., Campbell, K.L., Bauer, J.D., and Mudge, Circulation 138: e426–e483.
D.W. (2015). Altered dietary salt intake for people with 63 Williams, B., Mancia, G., Spiering, W. et al. and ESC
chronic kidney disease. Cochrane Database Syst. Rev.: Scientific Document Group (2018). 2018 ESC/ESH
CD010070. guidelines for the management of arterial hypertension.
51 McMahon, E.J., Bauer, J.D., Hawley, C.M. et al. (2013). A Eur. Heart J. 39: 3021–3104.
randomized trial of dietary sodium restriction in CKD. J. 64 Waeber, B., Gavras, I., Brunner, H.R. et al. (1982).
Am. Soc. Nephrol. 24: 2096–2103. Prediction of sustained antihypertensive efficacy of
52 Chade, A.R. and Hall, J.E. (2016). Role of the renal chronic captopril therapy: relationships to immediate
microcirculation in progression of chronic kidney injury blood pressure response and control plasma renin
in obesity. Am. J. Nephrol. 44: 354–367. activity. Am. Heart J. 103: 384–390.
53 Docherty, N.G., Canney, A.L., and le Roux, C.W. (2015). 65 Gaddam, K.K., Nishizaka, M.K., Pratt-Ubunama, M.N.
Weight loss interventions and progression of diabetic et al. (2008). Characterization of resistant hypertension:
kidney disease. Curr. Diab. Rep. 15: 55. association between resistant hypertension, aldosterone,
54 Eidemak, I., Haaber, A.B., Feldt-Rasmussen, B. et al. and persistent intravascular volume expansion. Arch.
(1997). Exercise training and the progression of chronic Intern. Med. 168: 1159–1164.
renal failure. Nephron 75: 36–40. 66 Arcand, J., Abdulaziz, K., Bennett, C. et al. (2014).
55 Robinson-Cohen, C., Littman, A.J., Duncan, G.E. et al. Developing a web-based dietary sodium screening tool for
(2014). Physical activity and change in estimated GFR personalized assessment and feedback. Appl. Physiol.
among persons with CKD. J. Am. Soc. Nephrol. 25: Nutr. Metab. 39: 413–414.
399–406. 67 Klein, I.H., Ligtenberg, G., Neumann, J. et al. (2003).
56 Hedayati, S.S., Elsayed, E.F., and Reilly, R.F. (2011). Sympathetic nerve activity is inappropriately increased in
Non-pharmacological aspects of blood pressure chronic renal disease. J. Am. Soc. Nephrol. 14: 3239–3244.
8 de Champlain, J., Cousineau, D., and Lapointe, L. (1980).

6 71 Egan, B.M., Zhao, Y., Li, J. et al. (2013). Prevalence of
Evidences supporting an increased sympathetic tone and optimal treatment regimens in patients with apparent
reactivity in a subgroup of patients with essential treatment-resistant hypertension based on office blood
hypertension. Clin. Exp. Hypertens. 2: 359–377. pressure in a community-based practice network.
69 Mahmud, A., Mahgoub, M., Hall, M., and Feely, J. (2005). Hypertension 62: 691–697.
Does aldosterone-to-renin ratio predict the 72 Schersten, B., Thulin, T., Kuylenstierna, J. et al. (1980).
antihypertensive effect of the aldosterone antagonist Clinical and biochemical effects of spironolactone
spironolactone? Am. J. Hypertens. 18: 1631–1635. administered once daily in primary hypertension.
70 Parthasarathy, H.K., Alhashmi, K., McMahon, A.D. et al. Multicenter Sweden study. Hypertension 2: 672–679.
(2010). Does the ratio of serum aldosterone to plasma renin 73 Engbaek, M., Hjerrild, M., Hallas, J., and Jacobsen, I.A.
activity predict the efficacy of diuretics in hypertension? (2010). The effect of low-dose spironolactone on resistant
Results of RENALDO. J. Hypertens. 28: 170–177. hypertension. J. Am. Soc. Hypertens. 4: 290–294.
589
36
Chronic Kidney Disease-Mineral and Bone Disorder

Suetonia C. Palmer1, Giovanni F.M. Strippoli2, Jonathan C. Craig3, and Donald A. Molony4
1
Department of Medicine, University of Otago Christchurch, Christchurch, New Zealand
2
Department of Emergency and Organ Transplantation University of Bari Bari, Italy and School of Public Health University of Sydney Sydney, NSW, Australia
3
College of Medicine and Public Health Flinders University, Adelaide, Australia
4
Division of Renal Diseases and Hypertension and Center for Clinical Research and Evidence-Based Medicine, University of Texas Houston Medical School, Houston, USA
I ntroduction The therapeutic approach to CKD-MBD that considers

treatment of abnormal serum PTH and phosphorus levels,
Chronic kidney disease-mineral and bone disorder (CKD- and that maintains serum calcium levels to prevent skel-
MBD) is commonly encountered in the later stages of CKD. etal and cardiovascular morbidity, is based principally on
CKD-MBD is a systemic disorder manifest by abnormal observational studies. High-quality evidence arising from
calcium, phosphorus, parathyroid hormone (PTH), and randomized controlled trials (RCTs) to inform clinical
vitamin D metabolism; abnormal bone turnover, minerali- decision-making for CKD-MBD and transplant-related
zation, volume, growth, or strength; and vascular and soft bone disease is limited. Treatment efficacy is generally
tissue calcification. Clinical features of CKD-MBD, such as evaluated based on surrogate biomarkers of bone metab-
accelerated cardiovascular disease and skeletal pain and olism, strength, and density, and serum biochemical
deformity, are associated with premature mortality and markers. Current therapies for CKD-MBD rely on a com-
impaired health-related quality of life (QoL) and symp- bination of approaches, including hormone replacement
toms. Recipients of a kidney transplant experience changes where hormone deficiencies are present (e.g. 25-hydroxy-
in bone metabolism and bone quality related to drug vitamin D and 1,25 dihydroxy-vitamin D) and suppres-
effects, particularly corticosteroid therapy. sion where excesses occur (e.g. PTH and phosphorus) and
A range of treatments are used to correct the metabolic direct modulation of serum and/or whole-body phosphorus
abnormalities observed with CKD-MBD. These include and calcium.
phosphate binding agents, vitamin D compounds and ana- Treatment recommendations regarding CKD-MBD in
logs, calcimimetics, dietary restriction, parathyroidectomy, CKD stages 1–4 have been based on expert opinion and the
and bisphosphonates. Although CKD-MBD is central to the extrapolation of evidence from stage CKD 5D Table 36.1.
care of moderate to severe CKD (stages 3b-5), much of the There are also no well-designed clinical trials in CKD of
rationale for treatment has been based on biological reason- any stage that have evaluated concurrent modulation of
ing rather than empiric evidence. The prevailing rationale combination therapies (e.g. phosphate binders along with
for treatment has been that correcting abnormalities in bone vitamin D and calcimimetics) on the long-term outcomes
and parathyroid metabolism can prevent complications of mortality and nonbone-related morbidity, particularly
(specifically, vascular calcification and skeletal fragility and cardiovascular complications. Short-term trials of combi-
deformity) and thereby improve patient survival and QoL. nation therapy with vitamin D and cinacalcet are available
CKD-MBD is associated with bone demineralization and, for surrogate outcomes such as serum PTH.
more importantly, with skeletal symptoms and nonosseous Recent systematic reviews of trials have shown that there
calcification, including calcification of the vasculature. The is often low or very low certainty that treatment of CKD-
latter is thought to contribute to the increased premature MBD makes a difference to the outcomes that directly
cardiovascular mortality that is observed in CKD. impact on patients including cardiovascular complications,
590 Chronic Kidney Disease-Mineral and Bone Disorder
Table 36.1 Treatment for CKD-MBD.
Strength of
Vitamin D We suggest vitamin D compounds can be used to lower serum Low or very low (mortality, Weak [1, 2]
parathyroid hormone levels for patients with CKD. Patients with cardiovascular events, skeletal
CKD (any stage) may reasonably choose not to receive vitamin symptoms, quality of life)
D compounds based on the very low certainty that treatment Moderate (serum parathyroid
makes a difference to longer term outcomes, including quality of hormone levels)
life, skeletal symptoms, growth and bone deformities, and
cardiovascular morbidity and mortality.
Calcimimetic We suggest calcimimetic agents can be used to lower serum High (mortality, hypocalcaemia, Weak [3]
therapy parathyroid hormone levels or prevent parathyroidectomy for parathyroidectomy in patients
patients treated with dialysis. Patients with CKD (any stage) treated with dialysis)
may reasonably choose not to receive calcimimetic therapy as Moderate (nausea in patients
treatment makes no difference to mortality and causes treated with dialysis)
gastrointestinal side-effects and hypocalcemia. Low or very low (CKD stages 3–5)
Phosphate We suggest phosphate binder therapy may be used to lower Moderate (gastrointestinal Weak [4, 5]
binders serum phosphorus levels in patients with CKD. Patients with side-effects and hypercalcemia)
CKD (any stage) may reasonably choose not to receive Low or very low (mortality,
phosphate binder therapy based on the very low certainty that cardiovascular events, skeletal
treatment makes a difference to skeletal, cardiovascular, or symptoms, parathyroidectomy,
quality of life outcomes. The specific choice of phosphate binder quality of life)
may be determined by patient preferences for avoidance of
specific adverse effects.
Dietary We suggest dietary phosphorus restriction may be used in Very low Weak [6]
restrictions patients with CKD to lower serum phosphorus levels. Patients
with CKD may reasonably choose not to restrict dietary
phosphorus based on the very low certainty that dietary
modification makes any difference to mortality, cardiovascular
events, or skeletal complications.
Serum We suggest that treatment of mineral and bone disorder is not Very low Weak [7]
biochemical guided by specific serum levels of parathyroid hormone,
targets phosphorus, or calcium by phosphorus product based on the
very low certainty that this approach makes a difference to
longer term outcomes, including quality of life, skeletal
symptoms, growth and bone deformities, and cardiovascular
complications and mortality.
Transplant We suggest that bisphosphonates may be used to prevent Low Weak [8]
bone disease fracture and bone pain. Patients with a kidney transplant may
reasonably choose not to receive bisphosphonate therapy based
on the low certainty that treatment makes any difference to
skeletal complications and low certainty of safety information.
skeletal symptoms and deformity, growth in children, and D

efinition
QoL. Based on these findings, the choice of treatment should
be based on patient preferences for improvement of specific CKD-MBD is a systemic disorder of mineral and bone
outcomes and the avoidance of side effects, together with metabolism caused by impaired kidney function with one
treatment availability and cost. or more of the following abnormalities: dysregulated cal-
This chapter examines the evidence for management of cium, phosphorus, PTH, or vitamin D metabolism; altered
CKD-MBD in adults, including transplant-related bone bone turnover, mineralization, volume, linear growth or
disease. strength; or vascular or other soft tissue calcification [10].
Epidemiolog 591
CKD-MBD usually develops over several years, with an of patients with a GFR <20 ml/min experience abnormal
asymptomatic phase that is manifest by hormone and serum phosphorus and levels while 15–20% of such
serum abnormalities, including progressive increases in patients experience abnormal serum calcium levels.
serum phosphorus and PTH. As the condition becomes Serum PTH levels increase with decreasing GFR levels.
symptomatic, patients can experience bone fragility caus- High serum levels (>7.4 pmol/l) are present in 12% of
ing pain or fracture, disability and impaired QoL, and vas- those with GFR>80 ml/min, 17% of those with GFR 70
cular and valvular calcification causing cardiovascular and 79 ml/min, 21% of those with GFR between 60 and
complications, hospital admissions, and premature death. 69 ml/min, and 56% of those with GFR <60 ml/min [12].
In clinical practice during the 1980s and 1990s, disorders of Serum 1,25-dihydroxy-vitamin D3 concentrations tend to
bone and mineral metabolism caused by CKD were consid- increase in parallel with decreases in GFR, with levels of
ered to be largely limited to bone abnormalities, hence the 22 pg/ml experienced by approximately half of patients
term “renal osteodystrophy.” Renal osteodystrophy describes with a GFR between 20 and 29 ml/min.
the pathological changes in bone morphology associated with Determining the prevalence of bone morphology
CKD and is characterized by bone histomorphometry changes changes caused by CKD is more challenging to assess than
based on bone biopsy. Renal osteodystrophy includes a range serum biomarkers due to the invasive nature of the bone
of bone disorders including osteitis fibrosis cystica (high turn- biopsy, potential coexistence of osteopenia or osteoporosis
over with mineralization defect and normal bone volume), in older patients, and the range of treatments used to man-
adynamic bone disease (low turnover with normal minerali- age CKD-MBD. Abnormal bone metabolism is observed in
zation and low or normal bone volume), osteomalacia (low nearly all patients with CKD stage 5D (requiring dialysis)
turnover with abnormal mineralization and low to medium and most patients with CKD stages 3–5. Adynamic bone
bone volume), and mixed uremic osteodystrophy (high turno- disease, in which low bone turnover with normal minerali-
ver with mineralization defect and normal bone volume). zation occurs (with diminished osteoblast activity, bone
The broader term, CKD-MBD, was developed by a consen- formation, and activation) is present in 10–71% of patients
sus workshop in 2005 to recognize the diverse range of sys- with CKD stage 5D and 5–49% of patients with CKD stages
temic manifestations of altered mineral metabolism [11]. As 3–4 [13]. Due to secular trends in treatment of CKD-MBD
a result of this conference, the term renal osteodystrophy and increasing comorbidity associated with CKD, the prev-
was recommended as a term used exclusively to define alter- alence of adynamic bone disease appears to be increasing.
ations in bone morphology associated with CKD, with Medial vascular calcification is highly prevalent in
reporting on histomorphometry based on a unified classifi- patients with CKD and inversely correlates with GFR [14].
cation system including parameters of bone turnover, min- In young patients treated with dialysis, coronary artery cal-
eralization, and volume (TMV). The term CKD-MBD was cification, measured by electron beam computed tomogra-
recommended to describe the broader clinical syndrome phy, was present in 80% of patients between 20 and 30 years
manifest by abnormal bone and mineral metabolism and/or old, and nearly doubled over a period of 20 months [15]. In
skeletal calcification. a small study in 48 patients with a GFR between 17 and
For each patient, the clinical manifestations of CKD- 55 ml/min, 90% of patients demonstrated aortic vascular
MBD are varied and change over time. The variable pheno- calcification, suggesting the majority of patients with CKD
typic expression of CKD-MBD leads to the challenge of stages 3–5 have arterial calcific changes [16].
identifying and using a single and reproducible measure to Recipients of a kidney transplant lose bone rapidly and
define CKD-MBD and assess response to treatment. early after transplantation from body sites rich in trabecu-
lar bone [17]. Bone mineral density may decline by 4–10%
in the first 6 months, with further decreases of 0.4–4.5%
E
pidemiology between 6 and 12 months [18, 19]. Devolution of parathy-
roid gland overactivity decreases after transplantation, but
The heterogeneity in clinical and biochemical abnormalities PTH levels may remain elevated in 45% of kidney trans-
of CKD-MBD makes it difficult to determine its exact preva- plantation recipients for up to 2 years [20]. Bone histomor-
lence in patients with CKD. The prevalence also depends phometry shows increased trabecular separation and
both on the clinical characteristic used to identify the pres- reduced number, delayed mineralization and lower miner-
ence of the syndrome and the stage and duration of CKD. alizing surface, and higher osteoid surfaces with lower
Serum calcium and phosphorus levels tend to remain bone volume [21].
within normal physiological limits (>2.1 mmol/l and This figure 36.1 illustrates some of the complex relation-
<1.5 mmol/l, respectively) until the glomerular filtration ships between calcium, phosphorus, vitamin D, and PTH
rate (GFR) is below 30 ml/min [12]. Approximately 40% that determine normal mineral metabolism and some of the
ways in which these relationships may be altered in CKD P

athophysiology
leading to the development of CKD-MBD. Two processes
mediate the early events in the pathogenesis of CKD-MBD Renal osteodystrophy was recognized as a significant and
in early CKD: the renal and systemic responses to excess progressive disease in patients with CKD stage 5 soon after
phosphorus and the fall in active 1,25 dihydroxy-vitamin D effective renal replacement therapy became generally
levels. The number of functioning nephron units declines available in North America, Europe, and Asia. The disease
as kidney function worsens. The maintenance of normal
state was recognized principally as a demineralizing and
serum phosphorus levels in the presence of CKD is depend-
debilitating bone disease leading to gross skeletal deforma-
ent on an increase in phosphate excretion per nephron.
tion, hypocalcemia, and hyperphosphatemia. It was
High PTH and low active vitamin D levels inhibit tubular
observed to be more prevalent in patients with prolonged
phosphate transport by decreasing sodium-dependent
kidney failure, and especially among those with chronic
phosphate cotransporter activity. Fibroblast growth factor
diseases of the renal tubulo-interstitium [22]. Bone biopsy
(FGF)-23 and the FGF-23 FGFR-Klotho complex may be
studies demonstrated that the degree of disordered bone
important modulators of renal phosphate excretion. In
architecture was associated directly with the degree of
addition, FGF-23 appears to reduce 1,25(OH)2 vitamin D in
part by decreased 1-α-hydroxylase activity. The fall in hyperparathyroidism and that the latter correlated in some
1-α-hydroxylase activity in CKD is most likely secondary to studies to 25(OH)-vitamin D deficiency or to 1,25(OH)2-
multiple mechanisms. Active vitamin D is also reduced in vitamin D deficiency [23–26].
many patients because of insufficient 25 hydroxyvitamin Abnormalities of bone and mineral metabolism start to
D. Low levels of active vitamin D reduce calcium absorp- occur in CKD stages 2 and 3 when the GFR is between 40
tion from the gastrointestinal tract and stimulate secretion and 70 ml/min. Even before clinical changes are observed
of PTH. Additionally, low calcium levels stimulate PTH in serum phosphorus levels, injury to kidney tissue stimu-
secretion. Consequently, calcium levels are partially lates circulating signals that modify osteocyte function. An
restored by release of calcium from the bone. Excess PTH, overview of the pathogenesis of CKD-MBD is shown in
hyperphosphatemia, and vitamin D deficiency each appear Figure 36.1.
to contribute to the development of the systemic and vas- A progressive increase in fibroblast growth factor 23
cular manifestations of CKD-MBD. (FGF23) occurs as GFR falls due to increased secretion by
calcium-sensing receptor
PTH secretion
Excess PTH
+ – Excess PTH
Renal osteodystrophy +
Vascular calcification
calcium Systemic manifestations
FGF23 1,25 di-hydroxy-vitamin D3

phosphorus load
FGFR1: Klotho-α
Klotho-α
Circulating bone
formation inhibitors
1-α hydroxylase
nephron number 25-hydroxy-vitamin D3

net renal phosphorus clearance
Figure 36.1 Pathogenesis of secondary hyperparathyroidism and CKD-MBD.

Pathophysiolog 593
osteocytes and osteoblasts, loss of secretory inhibition,

and impaired renal excretion [27]. FGF23 acts together High
with a coreceptor, the transmembrane protein Klotho-α,
OF
to suppress PTH gene expression and secretion [28].
Klotho-α is also expressed in the proximal renal tubule,
Bone
acting to regulate tubular phosphorus transport, together MILD
volume
HPT
with modulation of 1α-hydroxylase and 24-hydroxylase,
involved in the activation steps of vitamin D3 [29]. The
maintenance of normal serum phosphorus levels in the OM
presence of CKD is dependent on an increase in the frac- MUO

Low
tion of filtered phosphate reabsorbed per nephron. High AD
al
PTH and low active vitamin D levels inhibit tubular phos-
tio rm
za No
phate transport by decreasing sodium-dependent phos-
n
phate cotransporter activity and increasing phosphate
ti
Mi al
excretion per nephron. In addition, FGF-23 appears to
ra
m
ne
or
Low
reduce 1,25(OH)2 vitamin D3 in part via decreased
n
Turnover
Ab
1-α-hydroxylase activity. High
The fall in 1-α-hydroxylase activity in CKD is most
likely secondary to multiple mechanisms. Active vitamin Figure 36.2 Changes in bone architecture and metabolism in
CKD-MBD. AD, adynamic bone disease. HPT, hyperparathyroidism.
D is lower in many patients because of insufficient MUO, mixed uremic osteodystrophy. OM, osteomalacia.
25-hydroxy-vitamin D. 25-hydroxy-vitamin D (calcidiol) is (Permissions needed from [11]). Source: Moe S et al. [11].
the substrate for 1-α hydroxylase action and can be © 2006, Elsevier.
reduced in CKD due to lower sun exposures, lower skin
synthesis of cholecalciferol with uremic-pigmentation, collagen deposition from fibroblastic osteoprogenitors.
and dietary restrictions. Low levels of active vitamin D Osteoid, the nonmineralized structure within bone, is
(1,25-dihydroxy-vitamin D3) result in reduced calcium expanded, resulting in loss of the usual three-dimensional
absorption through the gastrointestinal tract and stimu- structure of bone architecture. Mineralization of bone is
late secretion of PTH. Additionally, low serum calcium accelerated, leading to altered bone strength and form. This
levels lead to increases in PTH secretion through exocyto- process of abnormal and increased bone osteoid generation
sis due to lower activation of the calcium-sensing recep- and mineralization is known as high bone turnover renal
tor. In response, serum calcium levels are partially osteodystrophy (or osteitis fibrosa). Low-bone turnover dis-
restored by release of calcium from the bone. ease is marked by reduced bone turnover, suppressed bone
The progressive decrease in 1,25-hydroxylation of vita- formation, and impaired bone mineralization, leading to
min D3 through decreasing 1-α hydroxylase activity in accumulated nonmineralized bone matrix and increased
kidney tissue, mediated in part by the counter-regulatory osteoid volume. A mixed osteodystrophy results from high
actions of FGF-23, is an important factor in the develop- PTH levels together with impaired mineralization. Bone for-
ment of CKD-MBD [30]. Decreasing 1,25-hydroxy-vitamin mation is variable and can be high or low. Increased number
D3 levels and the consequent hypocalcemia are key stimu- and activity of osteoclasts is seen together with accumulated
lants of PTH secretion. A complex interplay between vita- woven osteoid, and variable mineralization depending on
min D deficiencies, low serum calcium, high serum the presence of woven or lamellar bone. Vascular smooth
phosphorus, elevated FGF-23 levels which reduce calci- muscle cells are located in the media of vessel walls and
triol activity, together with impaired intestinal absorption contract to regulate vascular tone.
of dietary and supplemental vitamin D, and altered die- In the presence of uremic serum and inorganic phosphate,
tary patterns caused by uremia, are the hallmarks of high glucose, oxidized lipids, and cytokines, vascular smooth
CKD-MBD. muscle cells can dedifferentiate into osteocytic-like cells.
Abnormalities in bone due to CKD include changes in This osteogenic phenotype increases expression of bone-
bone turnover, mineralization, and volume (Figure 36.2), forming transcription factors and matrix proteins [31]. The
leading to pain, fracture, and deformity. transformed vascular smooth muscle cells secrete collagen
Excessive circulating PTH increases bone turnover. and noncollagenous proteins within the vessel intima and
Disordered osteoblast function results in impaired collagen media. Incorporation of calcium and phosphorus within
release and the generation of woven bone structure. Fibrosis matrix vesicles initiates mineralization, leading to vascular
occurs in the peritrabecular and marrow contours due to calcification [32].
Figure 36.2 shows the prevalence of combinations of hypercalcemia, and extraskeletal calcification may prompt
bone volume, turnover and mineralization. Clusters of consideration of surgical parathyroidectomy. Serum PTH
characteristics can identify commonly encountered dis- levels may be poorly predictive of bone histological
crete pathological abnormalities that occur in CKD-MBD. changes, although values at the extremes of the range can
For example, high bone volume, with normal mineraliza- be specific for high and low bone turnover. In a retrospec-
tion and high bone turnover, characterizes osteitis fibrosa tive study, the best cut-point to discriminate low from high
(OF) or advanced hyperparathyroid bone disease. bone turnover was an intact PTH <11.8 pmol/l while the
Transplant bone disease is characterized by pre-existing best cut-point to discriminate high from low bone turnover
CKD-MBD, persistent hyperparathyroidism despite nor- was >36 pmol/l [39]. There may be some utility in serial
malization of kidney function, hypovitaminosis D, and measures of PTH to detect a rising concentration that
immunosuppression effects, particularly glucocorticoid might prompt identification of reversible causes including
treatment [33]. high dietary phosphorus, low dialysis dose, or vitamin D
deficiency, or indicate the need to consider parathyroidec-
tomy if symptoms arise.
D
iagnostic Tests
Serum Phosphorus, Calcium, and Vitamin D
As CKD-MBD has protean manifestations that can be spe- Serum phosphorus and calcium are commonly measured
cific to the individual patients and vary based on severity of at regular intervals in clinical care. The serum levels are
kidney disease and treatment, none of the diagnostic tests not diagnostic of a specific process and should be inter-
used for clinical decision-making measure all aspects of preted in the context of other findings (including symp-
CKD-MBD. The diagnostic approach can be based on spe- toms, treatment, and serum PTH levels). Measurement of
cific clinical probability of abnormalities and as a guide to 25-hydroxy-vitamin D3 may identify vitamin D deficiency
treatment. The diagnostic tests to evaluate CKD-MBD are (<50 nmol/l) or insufficiency (50–75 nmol/l). Circulating
outlined in Table 36.2. In general, diagnosis and assess- levels do not necessarily correlate with kidney function.
ment involve evaluation of biochemical abnormalities and
bone and vascular involvement. To date, there are no sys-
Bone
tematic summaries of diagnostic test performances of clin-
ical methods used to evaluate CKD-MBD. Bone tissue in CKD-MBD has abnormal turnover (balance
between production and resorption), mineralization, and
volume. A variety of measures have been used to assess for
Biochemical Abnormalities
renal osteodystrophy and vary in terms of specificity, the
Clinical practice guidelines recommend biochemical range of information provided about bone turnover, miner-
markers including serum PTH, calcium, phosphorus, and alization, and volume, and the availability of diagnostic
bone-specific alkaline phosphatase to guide management tests within routine clinical practice.
and provide indirect information about high and low bone
turnover [10]. Biochemical markers are not diagnostic of Bone-specific Alkaline Phosphatase
the clinical complications of CKD-MBD (bone disease, vas- Alkaline phosphatase (ALP) circulates as several isoforms,
cular occlusion, and symptoms) but are simple to obtain including liver-specific and bone-specific ALP. Nearly all
and can indicate responses to therapeutic interventions. (95%) of circulating ALP is either bone-or liver-derived in
equal ratios. Bone-specific ALP above 20 ng/ml has high
Serum PTH sensitivity and specificity for identifying patients with high
Serum PTH rises progressively as kidney function declines bone turnover and excluding those with low bone turno-
and is often elevated above the normal range (1.1– ver [36]. Total ALP (including liver isoforms) has lower
7.4 pmol/l) in stages 3 and 4 CKD. PTH is the most com- specificity for high (specificity 90%) and low (specificity
monly used biochemical marker to indicate likely bone 50%) bone turnover at a cut-point of 200 IU/l.
turnover. A specific target range has not been identified at
which clinical outcomes are reduced, and clinical practice Biomarkers of Collagen and Bone Metabolism
guidelines suggest a broad range of appropriate values, and Activity
consistent with this lack of evidence. High PTH levels Serum biomarkers such as osteocalcin, osteoprotegerin,
(>91.2 pmol/l) combined with complications including TRAP-5b, pyridinoline, deoxypyridinoline, procollagen
refractory itch, skeletal pain or malformation, persistent type 1 amino-terminal extension peptides, and C-terminal
Diagnostic Test 595
Table 36.2 Summary of diagnostic tests in CKD-MBD.
Biochemical
Parathyroid hormone ●● Measures circulating intact parathyroid ●● Optimal level unknown in CKD
hormone ●● Rising level can prompt for assessment of high
●● Normal 1.1–7.4 pmol/l phosphate levels or intake, hypocalcemia, and/or
●● PTH fragments accumulate in CKD and may vitamin D deficiency
interfere with some assays ●● Low levels may indicate risk of low bone turnover
●● Serum PTH levels poorly predict bone histology
●● High PTH levels together with symptoms (high serum
calcium, symptomatic bone disease, refractory
pruritus, extraskeletal calcification or calciphylaxis,
unexplained myopathy) may prompt consideration of
surgical parathyroidectomy
Calcium ●● Measures circulating calcium concentrations ●● Can be low due to decreased 1,25-di-hydroxy-vitamin
but not ionized levels D3 levels, phosphatemia, and blunted bone response
●● Needs adjustment for serum albumin to PTH
●● Not indicative of total body calcium burden ●● Can be elevated secondary to treatment and/or
tertiary hyperparathyroidism
●● Low levels associated with tetany and cardiac
arrhythmias; high levels associated with kidney
stones, mood disorders, acute kidney injury, and
consitpation
Phosphorus ●● Measures circulating serum phosphorus ●● Optimal level unknown in CKD

concentrations. ●● Interpretation in context of other biomarkers and
●● Influenced by food intake, adherence to and patient’s treatment and symptomatic state is
timing of medication, and interval from last appropriate
dialysis
●● Serial measures most appropriate to guide
therapy
25-hydroxy-vitamin D3 ●● Identifies vitamin D deficiency (<50 nmol/l) ●● Not directly correlated with GFR
or insufficiency (50–75 nmol/l) [34] ●● 10–50% of patients with CKD stages 3 and 4% have
levels >75 nmol/l
●● Can guide therapeutic replacement although high
certainty evidence of improved clinical outcomes not
available
Bone
Alkaline phosphatase ●● 95% of serum alkaline phosphatase (ALP) in ●● Lower biological variability than PTH
healthy adults is either bone ALP or liver ALP ●● Markedly high or low levels of bone-specific may
isoforms in approximately equal parts [35] predict bone turnover
●● Testing for bone-specific ALP isoform can ●● Elevated bone-specific ALP (20 μg/l) and PTH
provide predictive information for bone (>22 pmol/l) has 100% sensitivity and 80% specificity
histology and activity for high turnover bone disease [36]
●● Combining bone-specific ALP <20 μg/l and PTH
<22 pmol/l provides 100% sensitivity and 100%
specificity for low turnover bone disease [36]
Bone-derived turnover ●● Includes markers of synthesis (procollagen ●● Uncertain diagnostic utility in CKD
markers of collagen type I C-terminal propeptide) and breakdown ●● Variable renal clearances and undefined reference
synthesis and (type I collagen cross-linked telopeptide, ranges for CKD mean these tests are not currently
breakdown pyridinoline, deoxypyridinoline) recommended in routine diagnostic work-up for
CKD-MBD
(Continued)
Bone biopsy ●● Measures bone microarchitecture, turnover, ●● Invasive, requires expertise to perform, process, and
and mineralization interpret; rarely performed in clinical care
●● Considered gold-standard for assessment of ●● Limited usefulness for monitoring treatment response
renal osteodystrophy due to invasiveness
Dual-energy X-ray ●● Measures bone mineral density (BMD) in ●● Inexpensive, noninvasive, and low radiation exposure
absorptiometry (DXA) grams per square centimeter ●● Limited information about bone quality or turnover
●● Low bone mineral density may predict fracture in any
stage of CKD [37, 38]
High-resolution ●● Measures three-dimensional microstructure ●● Low radiation

peripheral quantitative of cortical and trabecular bone to generate ●● Requires specific scanner, not available in many
computed tomography microstructural analysis clinical settings
(HR-pQCT) ●● Evaluates bone mechanical competence ●● Limited standardized reference data
●● Evaluates distal tibia and radius ●● Limited evidence of validation against bone
●● Measures volumetric bone mineral density histomorphometry
●● Not in routine clinical care
High-resolution ●● Measures bone microarchitecture ●● No ionizing radiation

magnetic resonance ●● Evaluates distal tibia, radius, femur ●● Routine scanning equipment
imaging (HR-MRI) ●● Limited evidence of validation against bone
histomorphometry
●● Complex post-processing analysis
●● Not in routine clinical care
Cardiovascular
Lateral X-ray lumbar ●● Semi-quantitative estimation of calcification ●● Sensitivity 62%, specificity 78% for coronary artery
abdominal aorta of abdominal aorta calcification if X-ray score >7
●● Simplicity
Computed tomography ●● Electron bean (EBCT) and multislice (MSCT) ●● Radiation

●● Coronary artery calcification measured as ●● Not able to differentiate between intimal and medial
Agatston score is calculated as the product of calification
a calcified plaque area by a coefficient related ●● Limited information about severity of obstructive
to the peak CT density measured within the coronary artery disease
same plaque
Echocardiogram ●● Identification of valvular abnormalities ●● Readily available

●● Functional cardiac information
●● No radiation
crosslinks have no role in clinical care due to their lack of tests measure indirect markers of bone metabolism or
diagnostic evidence and renal elimination, leading to diffi- turnover, density, or structural composition. Routinely
cult interpretation of their diagnostic test performance in used diagnostic approaches for CKD-MBD would opti-
the presence of CKD [11]. mally identify specific functional or structural bone abnor-
malities that can be ameliorated by therapy to prevent
Bone Structure and Quality adverse patient outcomes. However, not all fractures or
Diagnostic tests for renal osteodystrophy would ideally bone symptoms have the same pathogenesis and CKD-
examine the essential properties of bone: strength for load- MBD is characterized by highly heterogeneous changes to
bearing, flexibility to absorb energy, compression, and ten- bone density, architecture, mineralization, and volume.
sion without fracture, and volume and weight to facilitate Standard diagnostic approaches to fracture risk secondary
mobility [40]. Assessing the quality of bone structure to to osteoporosis may have different diagnostic utility in the
bear mechanical load without injury is limited as diagnostic present of CKD.
Diagnostic Test 597
In this section, the diagnostic strategies to examine bone Quantitative Computed Tomography
density and bone structure and quality are discussed. In addition to DXA, BMD can be quantified using quantita-
tive computed tomography (QCT), which provides a volu-
Bone Imaging metric assessment of bone density, differentiates between
Bone Mineral Density cortical and trabecular bone, and is less prone to interfer-
A key purpose of diagnostic evaluation of renal bone dis- ence from bone deformity and overlying nonosseus calcifi-
ease is to predict and guide management based on patient- cation. QCT can assess BMD at the hip and vertebra
relevant outcomes, including risks of skeletal deformity, peripheral body sites, and involves high radiation expo-
loss of height, bone pain, and low trauma fracture. sure. Peripheral QCT (pQCT) is an application of QCT that
Assessing bone mineral density (BMD) is recommended by can evaluate BMD at distal sites, such as the radius or tibia.
the United States Preventive Task Force [41] as a compo- High-resolution (HR) peripheral QCT (HR-pQCT) is
nent of screening for osteoporosis in women older than similar to QCT, although higher spatial resolution allows
65 years and younger women at higher risk of fracture. The HR-pQCT to measure BMD in the distal extremities (spe-
role of BMD assessment to guide management in CKD- cifically the radius and tibia). In patients with stage 5D,
MBD is more uncertain. HR-pQCT at the radius measuring cortical bone structures
Hyperparathyroidism may lead to sclerosis and thicken- (area, density, and thickness) provided the highest AUCs
ing of trabecular bone (and increased BMD) with concur- (0.73, 0.72, and 0.72, respectively) to predict any incident
rent accelerated resorption of cortical bone (with reduced radiological spinal or low trauma fracture [43]. The predic-
BMD). Consequently, BMD may poorly predict fracture tive value of trabecular bone parameters derived from HR-
risk and lead to administration of anti-osteoporotic therapy pQCT was lower than for cortical bone. In a study
that adversely impacts on bone metabolism. comparing vertebral bone density obtained by QCT with
A previous consensus workshop on the definition and bone histomorphometry in 26 patients treated with hemo-
evaluation of CKD-MBD in 2006 cited a study in 70 patients dialysis, vertebral BMD by QCT was correlated with tra-
treated with hemodialysis, in which mid-radius but not becular bone volume, while QCT for vertebral BMD had an
femoral neck and lumbar spine BMD was decreased and AUC of 0.76 for bone volume by bone biopsy [44].
correlated with time on dialysis [42]. While BMD at all sites In addition to measurement of BMD in distal sites, HR-
(mid-radius, femoral neck, lumbar) was negatively corre- pQCT can provide information about bone microstructure
lated with bone-specific ALP and serum-intact PTH, there including cortical density, area, and thickness, and trabec-
was no association with fracture occurrence, suggesting ular number and architecture. In 91 patients with CKD
BMD may have limited utility in CKD-MBD. (32 with fracture and 59 without fracture), HR-pQCT of
BMD measurement using dual-energy X-ray absorpti- the radius in those with fracture demonstrated lower corti-
ometry (DXA) is suggested by the 2017 updated Kidney cal area and thickness, and lower trabecular number and
Disease: Improving Global Outcomes (KDIGO) guide- greater trabecular separation [45]. The discrimination for
lines for patients in whom knowledge of fracture risk fracture by HR-pQCT parameters varied by duration of
will inform decision-making [10]. This is consistent with CKD.
findings in a longitudinal cohort study involving 131
patients with CKD stage 5 treated with hemodialysis [43]. Magnetic Resonance Imaging
A BMD T-score −2.5 at the total hip had a sensitivity of High-resolution MRI (HR-MRI) and micro-MRI provide
88.5% and specificity of 65.6% for incident low trauma 3D information of bone morphology and trabecular struc-
fractures or morphometric spine fractures, while the tures at peripheral body regions and avoids the need for
diagnostic test performance of lumbar spine BMD was ionizing radiation [46]. These tools may provide noninva-
somewhat lower (81.3 and 68.6%). In a second study of sive evaluation of renal osteodystrophy, but remain largely
485 hemodialyzed patients who were followed for 5 years limited to research settings.
for radiographic spine or any clinical fracture, the sensi-
tivity and specificity for any fracture was highest for total Bone Biopsy
hip BMD T score −2.7 (66.7% and 84.9%) for patients The bone biopsy remains the gold standard for detailed
with a serum PTH <23.3 pmol/l, and had lower predic- information about bone metabolism, but remains limited
tive utility in those patients with a serum PTH above the in clinical practice due to the invasive nature of the
same cut-point [37]. procedure and the expertise required to conduct, process,
There are no high-quality studies evaluating the perfor- and analyze the information provided. The bone biopsy
mance of DXA for BMD to predict clinical bone outcomes can allow the direct visualization of bone to evaluate
after kidney transplantation. bone turnover, mineralization, and volume, according to
a standardized nomenclature based on these components of 275 adults treated with hemodialysis, 77.8% had abdomi-
(turnover/mineralization/volume; TMV) [11]. Unlike nal aortic calcification with a median score of 6 [51]. Evidence
biochemical markers, bone biopsy can distinguish high for the diagnostic utility of abdominal X-ray to predict clini-
turnover bone disease from other forms of bone disease, cally relevant disease in CKD is sparse. In 90 patients treated
including mixed osteodystrophy, low bone turnover (ady- with hemodialysis, a cut-off vascular calcification score of
namic bone disease, osteomalacia), and bone pathologies 6.5 on abdominal radiography had a sensitivity of 85% and
caused by specific etiologies. specificity of 57% for coronary artery disease [52].
The bone biopsy has not been recommended for routine
care, but may assist specific management in the presence Computed Tomography
of inconsistency among diagnostic tests, unexplained frac- Computed tomography (electron beam [Electron Beam
ture or bone pain, severe progressive vascular calcification, Computed Tomography] or multislice [Multi-slice computed
unexplained hypercalcemia, possible aluminum toxicity, tomography]) provides quantitative assessment of coronary
definitive histological diagnosis before parathyroidectomy, artery calcification. The degree of coronary artery calcifica-
and prior to bisphosphonate treatment. tion can be calculated using the Agatston score, calculated
Bone biopsy after kidney transplantation may identify by multiplying the area of each calcified lesion by the pixel
patients who have low bone turnover, for whom antire- intensity of the lesion [53]. The scores for each lesion are
sorptive therapies may increase risks of complications, then summed to provide an overall score, reflecting burden
although there are no studies evaluating the diagnostic test and intensity of calcification. Calcification scores in adults of
performance of bone histomorphometry in the setting of 0–10, 11–100, >100, and >400 represent low, intermediate,
kidney transplantation [33]. progression from intermediate to high, and highest risk,
respectively [54]. Patients treated with dialysis have coronary
Fracture Risk Assessment artery calcification scores 10-fold higher (approximately 4000
The Fracture Risk Assessment Tool (FRAX) incorporates on average) than age-matched patients [55].
demographic variables (age, sex), clinical factors (weight, In a study of young patients with CKD stage 5 (ages 7 to
height), and risk factors for factor (previous facture, smok- 30 years), nearly all (88%) patients older than 20 years of age
ing, glucocorticoids, rheumatoid arthritis, alcohol intake, had evidence of coronary calcification with EBCT, com-
secondary osteoporosis) with or without BMD to estimate a pared with 5% of similarly aged people without CKD stage
ten-year probability of fracture [47]. 5 [15]. In a small study in 29 patients with CKD stage 5, a
Evidence supporting FRAX as a predictive tool in CKD is CT-based coronary artery calcium score >500 had sensitivity
limited. The performance of the FRAX score has been and specificity of 50% and 80% for detecting coronary artery
tested in a Canadian multicenter study of patients over stenosis 50% luminal narrowing on percutaneous angiog-
40 years with CKD stage 3 [48]. FRAX with BMD has mod- raphy. This suggests that CT calcification has a low detection
erate–low discriminatory power to predict fracture (AUC rate for coronary artery stenosis, but may have moderate test
0.69). In a study involving recipients 1-year on average after performance for exclusion of coronary artery narrow-
kidney transplantation, the FRAX score with bone mineral ing [56]. The current role of coronary artery calcification
density had limited predictive utility for major fracture scores to detect luminal coronary artery occlusion or for
(AUC 0.62) [49]. clinical-decision making is uncertain [57].
Vascular Calcification Echocardiography

Diagnostic tests for vascular calcification in patients with There are limited data on the diagnostic utility of echocar-
CKD include assessment for abdominal aortic and valvular diography for clinical decision-making about cardiac val-
calcification. vular calcification specific to the setting of CKD.
Plain Radiography
Abdominal aortic calcification can be identified by a lateral
abdominal radiograph (to avoid the overlay of the aorta
P
rognosis
and the vertebral structures). X-rays can be assessed for the
Skeletal Complications (Fracture, Pain,
presence of vascular calcification. Plain radiography is a
Deformity)
simpler, cheaper test for calcification that limits radiation
exposure compared with computed tomography. In addi- Several large-scale studies have demonstrated increased
tion, semiquantitative measures of severity can be applied risks of fracture for patients with CKD stage 5D. Hip frac-
to provide a calcification score (range 0–24) [50]. In a study tures measured using ICD-9 codes occur at an incidence of
Treatmen 599
8 per 1000 person-years for men and 14 per 1000 person- studies, while there is evidence that the risk of death
years for women, suggesting a relative risk approximately increases by 18% for every 1 mg/dl increase in serum phos-
fourfold that of the general population [58]. Prior kidney phorus (0.32 mmol/l), there is little evidence that serum
transplantation increases fracture risk, which becomes levels of calcium and PTH are associated with all-cause or
attenuated over time after surgery, while among transplant cardiovascular mortality [7]. Similarly, there are no robust
recipients fracture risk is higher among those with a pro- cohort data to show that alterations in serum concentra-
longed duration of dialysis prior to transplantation. The tions over time are linked to clinical outcomes, or that ther-
incidence of fracture is higher among those with low or apeutic modulation of serum phosphorus levels improve
high serum PTH levels and for those receiving psychoac- clinical outcomes.
tive or corticosteroid medications [59, 60]. Bone-related abnormalities including altered bone min-
In nearly two million community-living adults, and eral density, fracture, high or low bone turnover diseases
using diagnostic and procedure codes to evaluate fracture and extra-osseus calcification are all associated with worse
risk during 4.4 years of follow-up, there appears to be no clinical outcomes, including cardiovascular events, hospi-
increased risk of hip, wrist, or vertebral fracture among talization, and death. As with serum parameters, evidence
those with an estimated GFR below 60 ml/min per that treatment modification or prevention improves QoL
1.73 m2 [61]. It is possible that evidence of higher fracture or survival is awaited, while associations with clinical out-
risk associated with CKD is due to confounding by comor- comes may represent confounding due to kidney function
bidity, including inflammation and BMD [62]. and comorbidity that cannot be accounted for by statistical
Risks of fracture after kidney transplantation are highly adjustment [70].
heterogeneous, ranging from 3 to 100 fractures for every
1000 person-years [63]. Differences between studies are
likely to relate to study population differences, type of T
reatment
fracture, and duration of dialysis before transplantation.
Overall, the 10-year cumulative incidence of hip fracture Historically, treatment of CKD-MBD focused on manage-
for kidney transplantation is 1.7%, while the 3-year non- ment of renal osteodystrophy and secondary hyperparathy-
vertebral fracture incidence (1.6%) may be lower than the roidism. The goals of therapy were to prevent symptoms or
general population with a previous nonvertebral fracture provide symptom relief, and control the mineral abnormal-
(2.3%) [64]. ities of high serum phosphorus, low serum calcium, and
The incidence of bone pain and skeletal deformity with progressive increases in serum PTH. More recently, due to
CKD is less frequently reported than fracture, likely due to the recognition that serum calcium, phosphorus and PTH
the availability of diagnostic and procedural codes to cap- level, and vascular calcification are associated with mortal-
ture fracture-related events for longitudinal analyses com- ity and premature cardiovascular events, treatment goals
pared with the challenge of systematic collection of were expanded to preventing premature death and cardio-
symptoms and patient-reported outcome measures. In a vascular complications. Treatment to correct abnormal
community cross-sectional study among 1118 patients serum levels of calcium and phosphorus has demonstrated
with CKD, bone and joint pain was the most frequently efficacy across many randomized trials, while evidence of
experienced symptom [65], although this has not been a high certainty that such treatment makes a difference to
consistent finding in the literature [66, 67]. Similarly, there clinical outcomes including fracture, mortality and cardio-
are insufficient systematic data recorded on skeletal pain vascular events has remained elusive despite a wealth of
or deformity for recipients of a kidney transplant. clinical research.
Mortality and Cardiovascular Events Phosphate Lowering

The protean manifestations of CKD-MBD, including bio- Phosphate binders reduce phosphorus absorption from
chemical abnormalities, vascular calcification, and frac- dietary content through binding within the gut (Box 36.1).
ture, are associated with mortality, hospitalization, and Given the strong association between elevated serum phos-
cardiovascular events. The rationale for therapeutic control phorus that occurs in the later stages of CKD with vascular
of serum PTH and phosphorus levels has been, in part, evi- and metastatic calcification, phosphate binders have been
dence of the “dose-dependent” association between higher universally adopted into routine care and were recom-
serum phosphorus and PTH concentrations with all-cause mended in early guidelines [71], despite there being no
mortality, as demonstrated in two seminal studies by Block high-quality evidence that these agents prevent symptoms
et al. [68, 69]. However, in a systematic review of 47 cohort or bone and cardiovascular complications of CKD.
with calcium-based binders, although confidence in this

Box 36.1 Agents used as Phosphate Binders
evidence is low due to important methodological limita-
Mineral salts tions in the available trials. It is also uncertain whether
●● Calcium carbonate sevelamer lowers mortality, calcium biners increase mor-
●● Calcium acetate tality, both, or neither.
●● Calcium citrate Sevelamer and lanthanum binder classes probably incur
●● Magnesium hydroxide less hypercalcemia than calcium-based binders. The
longer-term consequences of less hypercalcemia and cal-
●● Magnesium carbonate
cium exposure with noncalcium binders is not deducible
Metal-based compounds
from randomized trials.
●● Aluminum hydroxide Adverse effects appear specific to specific phosphate
●● Lanthanum carbonate binder classes. Sevelamer ranks as the highest probability
●● Iron salts of causing constipation, while lanthanum ranks highest
Nonabsorbed polymers and resins for nausea and iron binders rank highest for diarrhea.
Phosphate binders account for one-half of the daily pill
●● Sevelamer hydrochloride
burden for patients treated with dialysis, which is associ-
Sevelamer carbonate
ated with lower health-related QoL [81].
●●
Magnesium carbonate, magnesium hydroxide, and

other magnesium salts bind phosphorus. In a short-term
Aluminum study magnesium carbonate lowered calcium ingestion
Aluminum hydroxide came into widespread use in the mid- from phosphate binding agents [82]. In a network meta-
to late 1970s, although subsequently drew concern because analysis, there was little or no high-quality evidence per-
of case series that associated aluminum accumulation with taining to patient-centered outcomes for magnesium-based
neurotoxicity and other tissue toxicities, well-studied disease binders [5].
outbreaks associated with parenteral aluminum from It remains uncertain when, along the spectrum of CKD,
dialysate, experimental models of aluminum toxicity, and phosphate binders should be initiated, what level of serum
the observation that chelation therapy appeared to modify phosphorus should be targeted to improve outcomes, and
the natural history of toxicity in cases attributed to oral alu- whether all phosphate binders are equivalent in terms of
minum exposure [72–78]. Evidence supporting the view that long-term outcomes. There is currently no evidence based
aluminum salts may be harmful and less effective than on randomized trials that specific target levels of serum
calcium-based binders infer that there is not a level of alu- phosphorus improve clinical endpoints. In a trial involving
minum exposure that is safe during long-term use [78, 79]. patients with CKD (stages 3b and 4) increases in coronary
calcification artery calcium volume particularly among
Nonaluminum Phosphate Binders patients at caution may be needed for the use of phosphate-
Numerous RCTs of phosphate binders have been con- binding therapy for patients with normal serum phospho-
ducted, although most evaluate therapy for a few weeks or rus levels [83].
months and compare newer phosphate binders against
calcium-containing salts [80]. Placebo-controlled trials Dietary Modification
have small sample sizes and are designed to prove the effi- Clinical practice guidelines suggest lowering dietary intake
cacy of therapy to lower serum phosphorus. of phosphorus to control serum phosphorus [10]. A
A network meta-analysis of RCTs evaluating phosphate Cochrane review identified there is low-quality evidence
binders, except for aluminum salts, provides evidence for that dietary phosphorus restriction makes any difference to
comparative efficacy and safety against other phosphate clinical outcomes, although dietary modification did lower
binder classes or placebo or standard therapy in the absence serum phosphorus by 0.18 mmol/l on average [84]. In
of head-to-head trials (Table 36.3) [7]. Compared with pla- uncontrolled studies, there is no evidence of an association
cebo, all phosphate binder classes lower serum phospho- between phosphate intake (as estimated by 24-hour excre-
rus, and iron-based binders lead to greater reductions than tion) and serum phosphorus [85] or cardiovascular disease
sevelamer, lanthanum, and calcium. Despite this, it is very and mortality [86].
uncertain whether phosphate binders prevent death, frac-
ture, or need for parathyroidectomy when compared with Dialysis Duration and Frequency
either calcium-based binders or placebo. Sevelamer (car- Frequent or longer hours dialysis lowers serum phospho-
bonate or hydrochloride) may prevent death compared rus levels. In the Frequent Hemodialysis Network (FHN)
Table 36.3 Summary of findings: phosphate binder therapy.
Absolute number of patients affected

No. of participants per 1000 patients treated for 1 year Relative risk (95% Certainty of the
Outcomes (no. of studies) (95% CI) confidence interval) evidence (GRADE) Conclusion
Sevelamer versus calcium-based binder

All-cause mortality 6376 (20) 122 fewer (158 to 52 fewer) 0.39 (0.21 to 0.74) Low Sevelamer may prevent death
●○○○
Cardiovascular death 2712 (4) 100 fewer (138 fewer to 61 more) 0.33 (0.08 to 1.41) Very low It is uncertain whether sevelamer makes any
○○○○ difference to cardiovascular mortality
Fracture -(−) — No studies Absent No studies were found that evaluated the impact
of sevelamer on fracture
Parathyroidectomy -(−) — No studies Absent No studies were found that evaluated the impact
of sevelamer on parathyroidectomy
Constipation 7862 (27) 12 more (0 to 38 more) 2.12 (1.01 –4.45) Moderate Sevelamer probably leads to constipation
●●○○ compared with calcium-based binders
Nausea 7265 (26) 2 fewer (10 fewer to 16 more) 0.84 (0.34–2.08) Low It is uncertain whether sevelamer leads to
●○○○ nausea
Hypercalcemia 5159 (21) 76 fewer (82–62 fewer) 0.14 (0.07 –0.29) Moderate Sevelamer probably leads to less frequent
●●○○ hypercalcemia than calcium-based binders
Lanthanum versus calcium-based binder

All-cause mortality 6376 (20) Not estimable 0.78 (0.16–3.72) Very low It is uncertain whether lanthanum decreases
○○○○ mortality
Cardiovascular death -(−) — No studies Absent No studies were found that studied the impact of
lanthanum on cardiovascular mortality
of lanthanum on fracture
of lanthanum on parathyroidectomy
Constipation 7862 (27) 18 fewer (38 fewer to 18 more) 0.70 (0.37–1.30) Low It is uncertain whether lanthanum leads to
●○○○ constipation
Nausea 7265 (26) 144 more (0 to 456 more) 2.18 (1.00–4.74) Low Lanthanum may lead to nausea
●○○○
Hypercalcemia 5159 (21) 109 fewer (116–90 fewer) 0.09 (0.03–0.25) Moderate Lanthanum probably leads to less frequent
●●○○ hypercalcemia than calcium-based binders
(Continued)
c36.indd 601 09-12-2022 16:07:45

Absolute number of patients affected

No. of participants per 1000 patients treated for 1 year Relative risk (95% Certainty of the
Outcomes (no. of studies) (95% CI) confidence interval) evidence (GRADE) Conclusion
Iron-based binders versus calcium-based binder

All-cause mortality 6376 (20) Not estimable 0.37 (0.09 to 1.60) Very low It is uncertain whether iron-based binders
○○○○ decrease mortality
Cardiovascular death 2913 (5) Not estimable 0.50 (0.007 to 33) Very low It is uncertain whether iron-based binders
○○○○ decrease cardiovascular mortality
of iron-based binders on fracture
of iron-based binders on fracture
Constipation 7265 (26) 8 fewer (18 fewer to 15 more) 0.67 (0.28–1.61) Low It is uncertain whether iron-based binders lead
●○○○ to constipation
Nausea 7265 (26) Not estimable 0.53 (0.17–1.69) Low It is uncertain whether iron-based binders lead
●○○○ to nausea
Hypercalcemia 5159 (21) Not estimable 0.10 (0.007–1.30) Low It is uncertain whether iron-based binders lead
●○○○ to less frequent hypercalcemia
CI: confidence interval. Data for all stages of chronic kidney disease (1–5, including 5D) are combined as there were few studies involving only people with CKD stages 1–4. Treatment
estimates are drawn from a systematic review with network meta-analysis of randomized controlled trials [5]. GRADE assessment of the certainty of the evidence [9]: High: This research
provides a very good indication of the likely effect. The likelihood that the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect.
The likelihood that the effect will be substantially different* is moderate. Low: This research provides some indication of the likely effect. However, the likelihood that it will be substantially
different* is high. Very low: This research does not provide a reliable indication of the likely effect. The likelihood that the effect will be substantially different* is very high. *Substantially
different = a large enough difference that it might affect decision-making.
c36.indd 602 09-12-2022 16:07:45

Treatmen 603
Daily Trial, dialysis six times per week lowered serum kidney failure in a dose-dependent manner [94, 95]. Lower
phosphorus by 0.18 mmol/l compared with dialysis three calcitriol doses suppressed PTH with lower risks of a hyper-
times per week [87]. Similarly, frequent nocturnal hemodi- calcemic response [96, 97]. Much of the evidence support-
alysis lowered serum phosphorus levels by 0.45 mmol/l ing vitamin D treatment derives from these proof-of-concept
compared with conventional hemodialysis [88]. However, studies and longitudinal cohort studies demonstrating an
whether these changes in serum phosphorus translate into association between vitamin D therapy and lower mortality,
improved clinical outcomes is uncertain. although this observation has been questioned on the basis
In summary, these data suggest phosphate binder therapy that a survival advantage with treatment may reflect
may be used to lower serum phosphorus levels in patients unmeasured confounding related to treatment selection
with CKD. Patients with CKD (any stage) may reasonably and indication [98–101].
choose not to receive phosphate binder therapy based on the In 2009, two Cochrane reviews summarized the evidence
very low certainty that treatment makes a difference to skel- for vitamin D therapy for patients with CKD stages 1–5 not
etal, cardiovascular, or QoL outcomes. There is limited evi- requiring dialysis [1] and those with CKD stage 5D [2]
dence dietary modification or higher dose dialysis to lower (summarized in Table 36.4), respectively. Evidence for
serum phosphorus improves clinical endpoints. The specific beneficial effects of vitamin D therapy is restricted to
choice of phosphate binder may be determined by patient effects on bone histology and serum PTH, as trials do not
preferences for the avoidance of specific adverse effects have sufficient statistical power to provide definitive evi-
together with considerations of drug availability and cost. dence on clinical outcomes.
In patients with CKD stages 1–5 and stage 5D, there is
very low certainty that vitamin D therapy makes any differ-
Lowering Serum Parathyroid Hormone
ence to mortality, commencement of dialysis, need for par-
and Maintaining Serum Calcium
athyroidectomy, bone pain, or fracture. Vitamin D therapy
There are five principal modalities of treatment to treat lowers PTH by 5.6 pmol/l for patients with CKD stages 1 to
elevated serum PTH levels and maintain serum calcium 4 and 22 pmol/l, for those with stage 5D, while incurring
levels. These include (i) vitamin D compounds and syn- hypercalcemia. Two randomized trials have demonstrated
thetic analogs, (ii) calcimimetic agents, (iii) combinations oral activated vitamin D does not improve cardiac function
of pharmacological interventions (vitamin D plus calcimi- for patients with CKD stages 3–5 during 1 year of treat-
metic), (iv) lowering dialysate calcium, and (v) surgical ment [102, 103].
parathyroidectomy Box 36.2.
Vitamin D Analogs
Vitamin D Vitamin D therapy is limited by hypercalcemia and hyper-
The natural history of untreated CKD-MBD is character- phosphatemia, leading to the development of structurally
ized by marked skeletal disability and hypocalcemia modified derivatives with altered affinity for vitamin D
causing tetany. Early studies demonstrated that binding proteins and vitamin D receptor to ameliorate
25-hydroxy-cholecalciferol and calcitriol increased serum treatment-related adverse effects on calcium and phospho-
calcium, suppressed PTH, and stabilized bone histol- rus. These include paricalcitol, 22-oxacalcitriol, doxercal-
ogy [89–93], but promptly incurred hypercalcemia, elevated ciferol, and falecalcitriol. Based on a systematic review of
renal calcium excretion, and accelerated progression of randomized trials, there is limited evidence of different
effects of vitamin D analogs on hypercalcemia, hyperphos-
phatemia, PTH levels, or clinical outcomes including bone
Box 36.2 Vitamin D Agents pain, parathyroidectomy, or mortality compared with calci-
Cholecalciferol (vitamin D3) triol or other vitamin D compounds [104].
Ergocalciferol (vitamin D2)
Calcifediol (25-hydroxy-vitamin D3) Nutritional Vitamin D
24,25-dihydroxy-vitamin D3 Vitamin D3 (cholecalciferol) is synthesized in the skin from
Calcitriol (1,25-dihydroxy-vitamin D3) cholesterol through a chemical reaction dependent on
Alfacalcidol (1-hydroxy-vitamin D3) UVB radiation derived from sunlight. Vitamin D2 is syn-
Paricalcitol (19-nor-1,25-dihydroxy-vitamin D2) thesized from radiation of ergosterol, a compound derived
Maxacalcitol (22-oxa-calcitriol) from the mold, ergot. In a systematic review of RCTs in
Doxercalciferol (1-alpha-hydroxy-vitamin D2) 2010, five trials of cholecalciferol for 6 months on average
Falecalcitriol (26,27-hexfluoro-calcitriol) in patients with all stages of CKD were identified [105]. In
low to moderate quality evidence, cholecalciferol (20 000 IU
Table 36.4 Summary of findings: vitamin D therapy.
No. of participants Absolute effect per 1000 Relative risk (95% Certainty of the
Outcomes (no. of studies) patients treated for 1 yr (95% CI) confidence interval) evidence (GRADE) Conclusion

All-cause mortality 477 (4) Not estimable 1.40 (0.38–5.15) Very low It is uncertain whether vitamin D prevents
○○○○ mortality
Fracture 76 (1) Not estimable Not estimable Very low It is uncertain whether vitamin D prevents
○○○○ fracture
Parathyroidectomy 76 (1) Not estimable Not estimable Very low It is uncertain whether vitamin D prevents
○○○○ parathyroidectomy
Bone pain -(−) Not estimable Not estimable Absent No studies were found that evaluated the impact
of vitamin D on bone pain
Commencing dialysis 301 (4) Not estimable 0.76 (0.36–1.62) Very low It is uncertain whether vitamin D prevents the
○○○○ need to commence dialysis
Hypercalcemia 712 (4) 24 more (3 to 96 more) 3.04 (1.17–7.90) Moderate Vitamin D probably causes hypercalcemia
●●○○
Chronic kidney disease stage 5D

All-cause mortality 233 (5) Not estimable 1.34 (0.34–5.24) Very low It is uncertain whether vitamin D decreases
○○○○ mortality.
Fracture 181 (4) Not estimable 1.00 (0.06 –15.4) Very low It is uncertain whether vitamin D prevents
○○○○ fracture
Parathyroidectomy 133 (2) Not estimable 0.82 (0.05–12.5) Very low It is uncertain whether vitamin D prevents
○○○○ parathyroidectomy
Bone pain 109 (4) Not estimable 0.29 (0.03–2.63) Very low It is uncertain whether vitamin D prevents bone
○○○○ pain
Hypercalcemia 182 (5) 33 more (2 fewer to 180 more) 3.80 (0.90–16.1) Low Vitamin D may cause hypercalcemia
●○○○
CI: confidence interval. Treatment estimates are drawn from two Cochrane reviews [1, 2]. GRADE assessment of the certainty of the evidence [9]: High: This research provides a very good
indication of the likely effect. The likelihood that the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood that the
effect will be substantially different* is moderate. Low: This research provides some indication of the likely effect. However, the likelihood that it will be substantially different* is high. Very
low; This research does not provide a reliable indication of the likely effect. The likelihood that the effect will be substantially different* is very high. *Substantially different = a large enough
difference that it might affect decision-making.
c36.indd 604 09-12-2022 16:07:45

Treatmen 605
weekly to 25 000 IU once monthly) reduces serum PTH experience nausea. Additional prespecified subgroup
levels to some extent (3.5 pmol/l), without evidence of analyses are reported from the Evaluation of Cinacalcet
increases in serum calcium or phosphorus. Calcifediol Hydrochloride Therapy to Lower Cardiovascular Events
(25-hydroxy-vitamin D3) is converted to calcitriol by 1α- (EVOLVE) trial, in which cinacalcet was evaluated against
hydroxylase predominantly in the kidney, remaining under placebo in nearly 4000 patients with CKD stage 5D [113]. A
feedback regulation. In evidence from two short-term ran- treatment-age interaction led to speculation that cinacalcet
domized trials (26 weeks or less), calcifediol lowers serum may be efficacious among older adults over 65 years of age,
PTH levels in a dose-dependent manner for patients with although caution is needed with this interpretation, as the
hyperparathyroidism and vitamin D insufficiency [106, EVOLVE study was not specifically designed to test this
107]. The effects of nutritional vitamin D supplementation hypothesis [114]. Further trials are specifically needed
for patients with CKD is unknown. among older patients to determine whether cinacalcet low-
In summary, vitamin D compounds can be used to lower ers mortality in CKD stage 5D.
serum PTH levels for patients with CKD. Patients with Evidence for parenteral etelcalcetide in CKD is limited
CKD (any stage) may reasonably choose not to receive vita- to a phase 2 dose-escalation trial [115], pooled data from
min D compounds based on the very low certainty that two placebo-controlled randomized clinical trials [116],
treatment makes a difference to longer term outcomes and a noninferiority trial comparing etelcalcetide with
including skeletal symptoms, bone deformities and symp- cinacalcet [117], all involving hemodialysis patients.
toms, or cardiovascular complications and mortality. Etelcalcetide suppresses PTH levels, while leading to
some patients experiencing muscle spasm, nausea, or
Calcimimetic Therapy vomiting during 26 weeks of treatment. Etelcalcetide
probably decreases serum PTH levels to a similar extent
Calcimimetic therapy mimcs the action of calcium on the
to cinacalcet. Effects of etelcalcitide on patient-level out-
calcium-sensing receptor in parathyroid tissue and sup-
comes are uncertain.
presses PTH release. Two calcimimetic agents are approved
The single-arm BONAFIDE study has characterized the
for use in the USA by the US Food and Drug Administration
bone response to cinacalcet therapy in 77 patients with CKD
(FDA) for treatment of secondary hyperparathyroidism in
stage 5D, serum PTH >33 pmol/l, and biopsy-proven high
CKD stage 5D: oral cinacalcet and parenteral etelcalcitide.
bone turnover, after 6–12 months of cinacalcet treat-
Cinacalcet is approved for marketing authorization in
ment [118]. Cinacalcet may normalize bone metabolism,
Europe [108]. Randomized trials have demonstrated cina-
although two participants developed adynamic bone disease
calcet lowers elevated PTH levels by 33 pmol/l on average
concomitant with lower PTH levels (<16.5 pmol/l) and one
for patients with CKD [109–111] Box 36.3.
patient with hypophosphatemia developed osteomalacia.
A meta-analysis of RCTs comparing cinacalcet with pla-
Similarly, Lien et al. evaluated the effects of cinacalcet on
cebo was published in 2013 involving 18 studies in 7446
BMD over 26 weeks in 14 patients randomly assigned to con-
adults with CKD (predominantly CKD Stage 5D) [112]. A
ventional therapy or cinacalcet [119]. Cinacalcet may
summary of the findings is shown in Table 36.5. Given at
slightly increase BMD at the proximal femur. Although
30–180 mg per day for a median of 6.5 months, there is high
these two studies raise the hypothesis that cinacalcet may
certainty that cinacalcet makes no difference to all-cause
improve bone histology, a high-quality trial of clinical bone
mortality for patients with CKD stage 5D and has uncertain
outcomes such as fracture, deformity, or disability is needed
effects on cardiovascular death. For patients with CKD
to provide evidence for clinical decision-making.
stages 3–5, evidence for mortality was of low certainty. It is
uncertain whether cinacalcet prevents fracture in CKD stage
5D and this outcome has not been evaluated in people with Surgical Parathyroidectomy
earlier stages of CKD. Cinacalcet leads to slightly fewer par- Numerous case series and cohort studies have demon-
athyroidectomy operations for people with CKD stage 5D. strated that surgical parathyroidectomy reduces PTH lev-
Cinacalcet causes hypocalcemia for 60 patients in every els in dialysis patients with secondary hyperparathyroidism.
1000 treated for 1 year, and 150 more patients probably Prospective surgical cohorts have also described the rate
of complications, the “optimal” surgical technique to
increase the probability of maintaining target PTH levels
Box 36.3 Calcimimetic Agents postoperatively, and the effects of parathyroidectomy on
Cinacalcet the bone component of CKD-MBD, on postoperative
Etelcalcitide serum calcium levels, and the need for vitamin D supple-
mentation [120–124].
Table 36.5 Summary of findings: calcimimetic therapy.
No. of Absolute effect per Certainty of

participants 1000 patients treated Relative the evidence
Outcomes (no. of studies) for 1 year (95% CI) effect (GRADE) Conclusion

All-cause 458 (2) 18 fewer per 1000 0.29 Low It is uncertain whether calcimimetic
mortality (23 fewer to 12 more) (0.06–1.48) ●○○○ therapy prevents death
Parathyroidectomy -(−) — No studies Absent No studies were found that evaluated the
impact of calcimimetic therapy on
parathyroidectomy
Fracture -(−) — No studies Absent No studies were found that evaluated the
impact of calcimimetic therapy on fracture
Hypocalcemia 449 (2) 310 more per 1000 31.9 Very low It is uncertain whether calcimimetic
(43 to 1910 more) (5.3–193) ○○○○ therapy leads to hypocalcemia
Nausea 449 (2) 126 more per 1000 2.26 Low Calcimimetic therapy may lead to nausea
(29 to 295 more) (1.29–3.95) ●○○○
Chronic kidney disease stage 5D
All-cause 6502 (9) 6 fewer per 1000 0.97 High Calcimimetic therapy results in little or no
mortality (22 fewer to 10 more) (0.89–1.05) ●●●● difference in mortality
Parathyroidectomy 4893 (5) 3 fewer per 1000 0.49 High Calcimimetic therapy leads to slightly
(4 fewer to 3 fewer) (0.40–0.59) ●●●● fewer parathyroidectomy operations
Fracture 3965 (2) 6 fewer per 1000 0.59 Very low It is uncertain whether calcimimetic
(8 fewer to 16 more) (0.13–2.60) ○○○○ therapy prevents fracture
Hypocalcemia 6415 (12) 60 more per 1000 6.98 High Calcimimetic therapy leads to
(41–85 more) (5.10–9.53) ●●●● hypocalcemia
Nausea 6450 (12) 153 more per 1000 2.02 Moderate Calcimimetic therapy probably leads to
(68 to 272 more) (1.45–2.81) ●●○○ nausea
CI; confidence interval. GRADE assessment of the certainty of the evidence [9]: High: This research provides a very good indication of the likely
effect. The likelihood that the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect.
The likelihood that the effect will be substantially different* is moderate. Low: This research provides some indication of the likely effect.
However, the likelihood that it will be substantially different* is high. Very low: This research does not provide a reliable indication of the likely
effect. The likelihood that the effect will be substantially different* is very high. *Substantially different = a large enough difference that it might
affect decision-making.
Surgical parathyroidectomy has been advocated princi- Dialysate Calcium

pally as a means of controlling severe hyperparathyroidism It was recognized more than four decades ago that eleva-
in CKD stage 5D who are refractory to medical manage- tions in serum calcium levels in dialysis-dependent patients
ment. There have been, to date, no long-term RCTs com- reduced PTH levels. This led to adoption of higher dialysate
paring parathyroidectomy to other modalities of controlling calcium concentrations as the standard of care. In the
PTH in end-stage kidney disease (ESKD) patients with 1970s, dialysate calcium levels increased from 2.5 mEq/l
respect to mortality and significant morbidities, including (and roughly equivalent to the normal serum calcium
pathologic fractures and cardiovascular events. level) to 1.50–1.75 mmol/l to assist management of hypoc-
Parathyroidectomy has not been evaluated either with alcemia, elevated serum PTH levels, and bone disease. As
observational studies or with an RCT as a modality of treat- calcium salts replaced aluminum to bind phosphorus and
ment for secondary hyperparathyroidism in patients with maintain serum calcium, together with the introduction of
CKD stages 3–5 not treated with dialysis. Surgical parathy- vitamin D compounds, higher dialysate calcium concen-
roidectomy has not been evaluated long-term in comparison trations were linked to hypercalcemia and dialysate
to medical approaches such as the use of calcimimetics, calcium concentrations were lowered.
although the latter medical approach has displaced surgical More recently, attention has been directed toward the
parathyroidectomy in some areas of the world. risks associated with an increased exogenous calcium
Treatmen 607
burden, principally based on low-quality observational combination). Therefore, it can be challenging to decide
data, including the oversuppression of parathyroid func- on the most appropriate clinical management of fractures
tion and the development of low-turnover bone disease, as in the presence of CKD, as treatment will depend on the
well as vascular calcification. Increasing dialysate calcium underlying predominant pathology that may include oste-
has been associated with mortality across 307 facilities and oporosis, osteomalacia, high bone turnover, or adynamic
17 236 patients in the Dialysis Outcomes and Practice bone disease. The likely etiology of a fragility fracture or
Patterns Study (DOPPS) [125]. low BMD is related with estimated GFR. For patients in an
The KDIGO guideline for management of CKD-MBD estimated GFR above 30 ml/min per 1.73 m2, the predomi-
suggests a dialysate calcium concentration between 1.25 and nant pathology, in the absence of elevated serum phospho-
1.50 mmol/l is used [10]. Emerging nonrandomized rus or PTH levels, is likely to be osteoporosis. For patients
evidence indicates lower dialysate calcium concentrations in an estimated GFR below 30 ml/min per 1.73 m2, other
(<1.25 mmol/l) are associated with hospitalization for heart bone disorders may predominate and CKD-MBD should
failure, hypocalcemia, and intradialytic hypotension [126]. In be excluded before a diagnosis of osteoporosis is
addition, lower dialysate calcium is associated with lower probable.
serum calcium, increased serum phosphorus and PTH levels, Evidence is emerging for the effects of antiresorptive med-
and increased use of phosphate binders, vitamin D, and calci- ications on BMD and fracture risk in CKD stages 3–5D,
mimetics, indicating that low dialysate calcium may aggra- although the evidence of impact on patient-centered out-
vate secondary hyperparathyroidism. Observational data also comes is very low quality. Possible treatments include bis-
indicate that higher dialysate calcium concentrations phosphonates, receptor activator of nuclear factor-κB ligand
(>1.75 mmol/l) are associated with higher mortality [127]. (RANKL) inhibition (denusomab) and selective estrogen
In a small RCT, low dialysate calcium (1.3 mmol/l) may receptor modulators (SERM). Post hoc analyses of treatment
slightly lower serum calcium levels (by 0.12 mmol/l) and effects in people with CKD enrolled in studies in the wider
increase serum PTH levels (by 16 pmol/l) and prescribing population and trials specifically evaluating therapy in the
of calcimimetics and vitamin D therapies compared with setting of CKD have become available.
high dialysate calcium (1.6 or 1.75 mmol/l). It is uncertain
whether low dialysate calcium makes any difference to vas- Bisphosphonates
cular calcification, left ventricular mass, or bone mineral Bisphosphonates are used widely for the treatment and/or
density [128]. In a second randomized clinical trial in 225 prevention of osteoporosis and its complications in the
patients treated with hemodialysis and with serum PTH general population (Box 36.4).
levels 33 pmol/l, it was uncertain whether a dialysate cal- These agents are eliminated via renal excretion [134,
cium of 1.25 mmol/l made any difference to coronary artery 135]. The efficacy of these agents in the treatment of
calcification, bone histomorphometry, or adverse events osteoporosis in people without CKD is confirmed with
compared with a dialysate calcium of 1.75 mmol/l [129]. high-certainty evidence within multiple trials [136]. The
use of bisphosphonates in individuals with CKD is less
Low Bone Mineral Density well studied, resulting in low certainty evidence, and
largely limited to prevention of BMD loss for recipients
Adults with CKD experience an increased risk of fractures of a kidney transplant [137]. There is no evidence of
due to altered bone strength. The incidence of hip fracture is high certainty that bisphosphonates modify clinical out-
approximately 7 per 1000 person-years among men and nearly comes such as fracture or safety for people with CKD.
14 per 1000 person-years among women, representing a rela-
tive risk that is fourfold that of the general population for both
sexes [58]. As in the general population, hip fracture is associ- Box 36.4 Bisphosphonates (shown in ascending
ated with high rates of death and hospitalization [130]. In order of potency [potency relative to etidronate])
patients with CKD, the degree of demineralization is corre-
lated with the degree (stage) and/or duration of CKD [131, Etidronate [1]
132]. Additionally, demineralization may be accelerated in Clodronate [10]
patients who have CKD caused by pathologies typically Pamidronate [100]
treated with steroids (e.g. glomerulonephrititis, lupus nephri- Alendronate [500]
tis, etc.) and those characterized by chronic metabolic acidosis Ibandronate [1000]
associated with renal tubular acidosis [133]. Risedronate [2000]
The presence of a fragility fracture or low BMD in a per- Zolendronate [10000]
son with CKD can be due to osteoporosis or CKD-MBD (or
The effects of alendronate on BMD for women enrolled in dialysis and osteoporosis or severe osteopenia, raloxifene
the Fracture Intervention Trial (FIT) were examined post 60 mg daily may increase lumbar spine BMD [147]. Similarly,
hoc in participants with CKD [138]. Women with an esti- in a trial involving 60 women with CKD stage 5 and 5D,
mated GFR < 45 ml/min per 1.73 m2 experienced increases raloxifene treatment for 8 months may increase BMD signifi-
in BMD (5.6%, 95% CI 4.8–6.5%) and may have experienced cantly at the lumbar spine [148]. In a post hoc analysis of a
fewer fractures. There are no trials specifically designed to multicenter randomized placebo-controlled trial, raloxifene
evaluate adverse events or risk of adynamic bone disease may increase vertebral BMD, but has uncertain effects on
with bisphosphonates in CKD (any stage). In a pooled analy- fracture and adverse events among women with an esti-
sis of nine trials evaluating risedronate 5 mg daily to prevent mated GFR below 60 ml/min per 1.73 m2 [149]. Thus, there
or treat osteoporosis or steroid-induced bone disease, is no high-quality evidence that antiresorptive treatment
patients with CKD stages 3 and 4 were identified post hoc. strategies, including raloxifene, are effective or safe in people
Risedronate has uncertain effects on adverse events, kidney with CKD and low bone mineral density.
function, or rates of low bone turnover [139]. In summary, these data do not lead to a recommendation
In a small RCT, patients treated with dialysis receiving about treatment for low BMD for patients with CKD.
etidronate (a low potency bisphosphonate) for 24 weeks
experienced retardation of coronary artery calcification
and lower aortic calcification compared with standard Target Levels of Serum Parathyroid Hormone,
therapy [140]. In contrast, alendronate therapy for adults Phosphorus, and Calcium
with CKD stages 3 and 4 has uncertain effects on aortic
Although specific target ranges for serum phosphorus and
vascular calcification and kidney function [141].
PTH have been previously suggested by clinical practice
Bisphosphonates have been reported to cause acute kid-
guidelines [71], there are no RCTs that evaluate whether
ney injury, which may occur after some delay and is not
these serum targets make any difference to clinical out-
always fully reversible [142, 143].
comes [7]. Accordingly, the KDIGO guidelines provide weak
recommendations based on low-quality evidence (Box 36.5).
Denusomab
In summary, these data do not allow a recommendation
Denusomab is a human monoclonal antibody that binds
that treatment for CKD-MBD should include specific
and inhibits the actions of receptor RANKL, which is
serum targets for phosphorus or PTH.
expressed by osteoblastic cells and mediates osteclastic
bone resorption. Denusomab increases BMD [144] and
prevents fracture [145] in older women with low bone den- Box 36.5 Target serum levels of parathyroid
sity. Preliminary studies indicate that the pharmacokinet- hormone, phosphorus, and calcium suggested
ics of denusomab are not affected by kidney function, in clinical practice guidelines
although in a single-dose study among 55 people with CKD
stages 1–5D treatment was complicated by low serum cal- Serum CKD Kidney Disease: Improving Global
cium levels and sometimes required hospitalization for biochemical stage Outcomes (KDIGO) (2017) [10]
parameter
parenteral calcium supplementation [145].
Phosphorus 3 and 4 Toward normal range (2C)
At present, there are no prospective randomized trials
5D Toward normal range (2C)
designed to evaluate the benefits and harms of denusomab in
Calcium 3 and 4 Avoid hypercalcemia (2C)
people with CKD and low BMD. In a post hoc analysis of the
5D Avoid hypercalcemia (2C)
FREEDOM trial, denosumab may prevent vertebral fractures
Intact 3 Optimal level is not known
in women with CKD stage 3, but treatment effectiveness was parathyroid Patients should be assessed when
very uncertain among women with CKD stage 4 [146]. hormone levels above upper limit of normal (2C)
Although the authors concluded that adverse events were
4 Optimal level is not known
not increased, there was no information given about individ-
Patients should be assessed when
ual adverse events, particularly low serum calcium levels. levels above upper limit of normal (2C)
5D Optimal level is not known
Raloxifene Patients should be assessed when
Raloxifene is an oral SERM that increases BMDs Three trials levels above upper limit of normal (2C)
provide information about raloxifene for women with low
Grade for strength of recommendation: level 1, strong’
bone density and CKD, but evidence is limited by uncer-
level 2, weak. Grade for quality of evidence: A, high, B,
tainty of effects on in nonvertebral bone, and very low cer-
moderate, C, low, D, very low.
tainty about safety and fracture endpoints. In 50 women on
Treatmen 609
Bone Disease After Kidney Transplantation Evidence for treatment of bone disease after kidney
transplantation is generally of low or very low certainty
In a recently updated Cochrane review, 50 randomized trials based on the methodological limitations in randomized
have evaluated bisphosphonates, vitamin D, teriparatide trials. As the rates of fracture are low in available trials and
(synthetic PTH), denusomab, cinacalcet, parathyroidec- existing trials involve small numbers of participants, exist-
tomy, and calcitonin [150]. Bisphosphonates are the most ing research studies lack statistical power to provide high-
commonly assessed treatment approach for prevention of quality evidence that treatment makes a difference to
fracture after kidney transplantation (Table 36.6). bone-related patient outcomes.
Table 36.6 Summary of findings: treatments for bone disease after kidney transplantation.
No. of Absolute effect Relative Certainty of

participants (per 1000 patients effect the evidence
Outcomes (no. of studies) treated for 1 year) (95% CI) (GRADE) Conclusion
Bisphosphonates
Fracture 765 (13) 38 fewer (61 fewer 0.62 Low Bisphosphonates may prevent bone fracture for
to 1 more) (0.38–1.01) ●○○○ recipients of a kidney transplant
Bone pain 153 (3) 133 fewer (160–12 0.20 Low Bisphosphonates may prevent bone pain for
fewer) (0.04–0.93) ●○○○ recipients of a kidney transplant
Spinal deformity -(−) — No studies Absent No studies were found that evaluated the
impact of bisphosphonate therapy on spinal
deformity
Graft loss 332 (6) Not estimable 1.01 (0.42 Very low It is very uncertain whether bisphosphonate
to 2.44) ○○○○ therapy makes any difference to graft loss
All-cause 606 (10) Not estimable 0.62 (0.23 Very low It is very uncertain whether bisphosphonate
mortality to 1.63) ○○○○ therapy makes any difference to all-cause
mortality
Nausea -(−) — No studies Absent One study was found that evaluated nausea
with bisphosphonate therapy
Hypocalcemia 207 (4) Not estimable 5.59 Very low Bisphosphonate therapy may incur
(1.00–31.1) ○○○○ hypocalcemia
Vitamin D
Fracture 299 (5) Not estimable 0.96 Very low It is very uncertain whether vitamin D therapy
(0.10–8.94) ○○○○ makes any difference to fracture
Bone pain -(−) — No studies Absent No studies were found that evaluated the
impact of vitamin D therapy on bone pain
Spinal deformity -(−) — No studies Absent No studies were found that evaluated the
impact of vitamin D therapy on spinal
deformity
Graft loss -(−) — No studies Absent One study was found that evaluated graft loss
with vitamin D therapy
All-cause 232 (3) Not estimable 0.49 Very low It is very uncertain whether vitamin D therapy
mortality (0.03–9.22) ○○○○ makes any difference to mortality
Nausea -(−) — No studies Absent No studies were found that evaluated whether
vitamin D therapy leads to nausea
Hypercalcemia 465 (7) 66 more (15 fewer 2.09 Very low It is very uncertain whether vitamin D therapy
to 300 more) (0.84–5.22) ○○○○ leads to hypercalcemia
CI: confidence interval. Evidence is drawn from a 2017 Cochrane review update (in peer review) [150]. GRADE assessment of the certainty of
the evidence [9]: High: This research provides a very good indication of the likely effect. The likelihood that the effect will be substantially
different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood that the effect will be substantially
different* is moderate. Low: This research provides some indication of the likely effect. However, the likelihood that it will be substantially
different* is high. Very low: This research does not provide a reliable indication of the likely effect. The likelihood that the effect will be
substantially different* is very high. *Substantially different = a large enough difference that it might affect decision-making.
Systematic radiological assessment for vertebral fracture including patients, caregivers, health providers, and
is the predominant fracture-related outcome in existing researchers, lifestyle interventions and those interven-
randomized trials of bone therapies in kidney transplantations that can improve long-term transplant outcomes,
tion. Treatment in randomized trials is commenced at the prevention, and QoL were research priorities in
time of transplantation in about half of all trials and within CKD [151]. There is currently insufficient data to recom-
the first 3 years in the remaining trials. Based on 13 studies, mend lifestyle or exercise interventions for CKD-MBD-
involving 765 patients and reporting fracture events for 58 related outcomes [152].
participants, bisphosphonates given for about 12 months
may prevent fracture (spinal or low trauma) and bone pain.
The safety profile of bisphosphonate treatment and the C
onclusion
efficacy and safety of vitamin D for bone protection among
kidney transplant recipients is very uncertain [150]. CKD-MBD is a frequent complication for patients with
Evidence for denosumab, teriparatide, and cinacalcet is moderate to severe CKD and is a central component of spe-
currently insufficient to inform decision-making. cialist nephrology care. In addition, bone loss occurs rap-
In summary, these data do not allow one to recom- idly in the first few months after kidney transplantation.
mend specific treatments for bone disease after kidney No RCTs or large cohort studies have provided evidence
transplantation. of high certainty to support practices of monitoring serum
biochemical markers, bone disease or vascular and non-
vascular calcification. At this time, there is insufficient
Patient and Caregiver Treatment Preferences
high-quality evidence that treatments improve patient-
Currently, there are no studies evaluating patient and car- centered outcomes. At present, treatment for CKD-MBD
egiver preferences for treatment or outcomes specifically in can be guided by shared decision-making based on patient
CKD-MBD. In a national workshop involving stakeholders preferences and therapeutic safety and cost.
R
eferences
Palmer, S., McGregor, D., Craig, J. et al. (2009). Vitamin D

1 6 Liu, Z., Su, G., Guo, X. et al. (2015). Dietary interventions
compounds for people with chronic kidney disease not for mineral and bone disorder in people with chronic
requiring dialysis. Cochrane Database Syst. Rev. (4) (Art. kidney disease. Cochrane Database Syst. Rev. (9) (Art. No.:
No.: CD008175). DOI: https://doi.org/10.1002/14651858. CD010350). DOI: https://doi.org/10.1002/14651858.
CD008175. CD010350.pub2.
2 Palmer, S., McGregor, D., Craig, J. et al. (2009). Vitamin D 7 Palmer, S., Hayen, A., Mackaskill, P. et al. (2011).
compounds for people with chronic kidney disease Serum levels of phosphorus, parathyroid hormone, and
requiring dialysis. Cochrane Database Syst. Rev. (4) (Art. calcium and risks of death and cardiovascular disease
No.: CD005633). DOI: https://doi.org/10.1002/14651858. in individuals with chronic kidney disease: a
CD005633.pub2. systematic review and meta-analysis. JAMA 305 (11):
3 Ballinger, A., Palmer, S., Nistor, I. et al. (2014). 1119–1127.
Calcimimetics for secondary hyperparathyroidism in 8 Palmer, S., McGregor, D.O., and Strippoli, G.F. (2005).
chronic kidney disease patients. Cochrane Database Syst. Interventions for preventing bone disease in kidney
Rev. (12) (Art. No.: CD006254). DOI: https://doi. transplant recipients. Cochrane Database Syst. Rev. (2)
org/10.1002/14651858.CD006254.pub2. (Art. No.:Cd005015).
4 Navaneethan, S., Palmer, S., Vecchio, M. et al. (2011). 9 Hultcrantz, M., Rind, D., Akl, E.A. et al. (2017). The
Phosphate binders for preventing and treating bone disease GRADE Working Group clarifies the construct of
in chronic kidney disease patients. Cochrane Database Syst. certainty of evidence. J. Clin. Epidemiol. 87: 4–13.
Rev. (2) (Art. No.: CD006023). DOI: https://doi. 10 Kidney Disease: Improving Global Outcomes (KDIGO)
org/10.1002/14651858.CD006023.pub2. CKD-MBD Update Work Group (2017). KDIGO 2017
5 Palmer, S., Gardner, S., Tonelli, M. et al. (2016). Phosphate- clinical practice guideline update for the diagnosis,
binding agents in adults with CKD: a network meta- evaluation, prevention, and treatment of chronic kidney
analysis of randomized trials. Am. J. Kidney Dis. 68 (5): disease-mineral and bone disorder (CKD-MBD). Kidney
691–702. Int. Supp. 7: 1–59.
Reference 611
11 Moe, S., Drüeke, T., Cunningham, J. et al. (2006). 24 Clair, F., Leenhardt, L., Bourdeau, A. et al. (1987). Effect
Definition, evaluation, and classification of renal of calcitriol in the control of plasma calcium after
osteodystropy: a position statement from Kidney Disease: parathyroidectomy. A placebo-controlled, double-blind
Improving Global Outcomes (KDIGO). Kidney Int. 69 study in chronic hemodialysis patients. Nephron 46 (1):
(11): 1945–1953. 18–22.
12 Levin, A., Bakris, G., Molitch, M. et al. (2007). Prevalence 25 Andress, D.L., Norris, K.C., Coburn, J.W. et al. (1989).
of abnormal serum vitamin D, PTH, calcium, and Intravenous calcitriol in the treatment of refractory
phosphorus in patients with chronic kidney disease: osteitis fibrosa of chronic renal failure. N. Engl. J. Med.
results of the study to evaluate early kidney disease. 321 (5): 274–279.
Kidney Int. 71: 31–38. 26 Hamdy, N.A., Kanis, J.A., Beneton, M.N. et al. (1995).
13 Bover, J. and Cozzolino, M. (2011). Mineral and bone Effect of alfacalcidol on natural course of renal bone
disorders in chronic kidney disease and end-stage renal disease in mild to moderate renal failure. BMJ 310 (6976):
disease patients: new insights into vitamin D receptor 358–363.
activation. Kidney Int. Supp. 1 (4): 122–129. 27 Larsson, T., Nisbeth, U., Ljunggren, O. et al. (2003).
14 Budoff, M., Rader, D., Reilly, M. et al. (2011). Relationship Circulating concentration of FGF-23 increases as renal
of estimated GFR and coronary artery calcification in the function declines in patients with chronic kidney disease,
CRIC (Chronic Renal Insufficiency Cohort) Study. Am. J. but does not change in response to variation in phosphate
Kidney Dis. 58: 519–526. intake in healthy volunteers. Kidney Int. 64 (6):
15 Goodman, W., Goldin, J., Kuizon, B. et al. (2000). 2272–2279.
Coronary-artery calcification in young adults with 28 Ben-Dov, I.Z., Galitzer, H., Lavi-Moshayoff, V. et al.
end-stage renal disease who are undergoing dialysis. N. (2007). The parathyroid is a target organ for FGF23 in
Engl. J. Med. 342 (20): 1478–1483. rats. J. Clin. Invest. 117 (12): 4003–4008.
16 Toussaint, N., Lau, K., Strauss, B. et al. (2008). 29 Hu, M.C., Shi, M., Zhang, J. et al. (2010). Klotho: a novel
Associations between vascular calcification, arterial phosphaturic substance acting as an autocrine enzyme in
stiffness and bone mineral density in chronic the renal proximal tubule. FASEB J. 24 (9): 3438–3450.
kidney disease. Nephrol. Dial. Transplant. 23 (2): 30 Gutierrez, O., Isakova, T., Rhee, E. et al. (2005). Fibroblast
586–593. growth factor-23 mitigates hyperphosphatemia but
17 Julian, B., Laskow, D., Dubovsky, J. et al. (1991). Rapid accentuates calcitriol deficiency in chronic kidney
loss of vertebral mineral density after renal disease. J. Am. Soc. Nephrol. 16 (7): 2205.
transplantation. N. Engl. J. Med. 325: 544–550. 31 Hruska, K., Mathew, S., Memon, I., and Saab, G. (2009).
18 Malluche, H., Monier-Faugere, M.-C., and Herberth, J. The pathogenesis of vascular calcification in the Chronic
(2010). Bone disease after renal transplantation. Nat. Rev. Kidney Disease Mineral and Bone Disorder (CKD-MBD).
Nephrol. 6: 32–40. Semin. Nephrol. 29 (2): 156–165.
19 Brandenburg, V., Politt, D., Ketteler, M. et al. (2004). Early 32 Moe, S. and Chen, N. (2008). Mechanisms of vascular
rapid loss followed by long-term consolidation calcification in chronic kidney disease. J. Am. Soc.
characterizes the development of lumbar bone mineral Nephrol. 19 (2): 213–216.
density after kidney transplantation. Transplantation 77: 33 Bouquegneau, A., Salam, S., Delanaye, P. et al. (2016).
1566–1571. Bone disease after kidney transplantation. Clin. J. Am.
20 Lou, I., Foley, D., Odorico, S.K. et al. (2015). How well Soc. Nephrol. 11 (7): 1282–1296.
does renal transplantation cure hyperparathyroidism? 34 Holick, M., Binkley, N., Bischoff-Ferrari, H. et al. (2011).
Ann. Surg. 262 (4): 653–659. Evaluation, treatment, and prevention of vitamin D
21 Neves, C.L., dos Reis, L.M., Batista, D.G. et al. (2013). deficiency: an Endocrine Society clinical practice
Persistence of bone and mineral disorders 2 years after guideline. J. Clin. Endocrinol. Metab. 96 (7): 1911–1930.
successful kidney transplantation. Transplantation 96 (3): 35 Sardiwal, S., Magnusson, P., Goldsmith, D., and Lamb, E.
290–296. (2013). Bone alkaline phosphatase in CKD-mineral and
22 Cundy, T., Hand, D., Oliver, D. et al. (1985). Who gets bone disorder. Am. J. Kidney Dis. 62 (4): 810–822.
renal bone disease before beginning dialysis? Br. Med. J. 36 Ureña, P., Hruby, M., Ferreira, A. et al. (1996). Plasma
290: 271–275. total versus bone alkaline phosphatase as markers of
23 Ghazali, A., Fardellone, P., Pruna, A. et al. (1999). Is low bone turnover in hemodialysis patients. J. Am. Soc.
plasma 25-(OH)vitamin D a major risk factor for Nephrol. 7 (3): 506–512.
hyperparathyroidism and Looser’s zones independent of 37 Iimori, S., Mori, Y., Akita, W. et al. (2012). Diagnostic
calcitriol? Kidney Int. 55 (6): 2169–2177. usefulness of bone mineral density and biochemical
markers of bone turnover in predicting fracture in CKD end-stage renal disease on hemodialysis: a cross-sectional
stage 5D patients--a single-center cohort study. Nephrol. observational study. Ther. Adv. Chronic Dis. 6 (3): 84–96.
Dial. Transplant. 27 (1): 345–351. 52 de Bie, M., Buiten, M., Rotmans, J. et al. (2017). Abdominal
38 Yenchek, R.H., Ix, J.H., Shlipak, M.G. et al. (2012). Bone aortic calcification on a plain X-ray and the relation with
mineral density and fracture risk in older individuals significant coronary artery disease in asymptomatic
with CKD. Clin. J. Am. Soc. Nephrol. 7 (7): 1130–1136. chronic dialysis patients. BMC Nephrol. 18: 82.
39 Sprague, S.M., Bellorin-Font, E., Jorgetti, V. et al. (2016). 53 Agatston, A.S., Janowitz, W.R., Hildner, F.J. et al. (1990).
Diagnostic accuracy of bone turnover markers and bone Quantification of coronary artery calcium using ultrafast
histology in patients with CKD treated by dialysis. Am. J. computed tomography. J. Am. Coll. Cardiol. 15 (4): 827–832.
Kidney Dis. 67 (4): 559–566. 54 Rumberger, J.A. (2008). Coronary artery calcium
40 Seeman, E. and Delmas, P. (2006). Bone quality --the scanning using computed tomography: clinical
material and structural basis of bone strength and recommendations for cardiac risk assessment and
fragility. N. Engl. J. Med. 354 (21): 2250–2261. treatment. Semin. Ultrasound CT MR. 29 (3): 223–229.
41 U. S. Preventive Services Task Force (2011). Screening for 55 Braun, J., Oldendorf, M., Moshage, W. et al. (1996).
osteoporosis: U. S. preventive services task force Electron beam computed tomography in the evaluation
recommendation statement. Ann. Intern. Med. 154 (5): of cardiac calcification in chronic dialysis patients. Am. J.
356–364. Kidney Dis. 27 (3): 394–401.
42 Ureña, P., Bernard-Poenaru, O., Ostertag, A. et al. (2003). 56 Park, M.Y., Choi, S.J., Kim, J.K. et al. (2011). Use of
Bone mineral density, biochemical markers and skeletal multidetector computed tomography for evaluating
fractures in haemodialysis patients. Nephrol. Dial. coronary artery disease in patients undergoing dialysis.
Transplant. 18 (11): 2325–2331. Nephrology (Carlton) 16 (3): 285–289.
43 West, S.L., Lok, C.E., Langsetmo, L. et al. (2015). Bone 57 Lentine, K.L., Costa, S.P., Weir, M.R. et al. (2012). Cardiac
mineral density predicts fractures in chronic kidney disease evaluation and management among kidney and
disease. J. Bone Miner. Res. 30 (5): 913–919. liver transplantation candidates: a scientific statement
44 Carvalho, A.B., Carneiro, R., Leme, G.M. et al. (2013). from the American Heart Association and the American
Vertebral bone density by quantitative computed College of Cardiology Foundation: endorsed by the
tomography mirrors bone structure histomorphometric American Society of Transplant Surgeons, American
parameters in hemodialysis patients. J. Bone Miner. Society of Transplantation, and National Kidney
Metab. 31 (5): 551–555. Foundation. Circulation 126 (5): 617–663.
45 Nickolas, T.L., Stein, E., Cohen, A. et al. (2010). Bone 58 Alem, A.M., Sherrard, D.J., Gillen, D.L. et al. (2000).
mass and microarchitecture in CKD patients with Increased risk of hip fracture among patients with
fracture. J. Am. Soc. Nephrol. 21 (8): 1371–1380. end-stage renal disease. Kidney Int. 58 (1): 396–399.
46 Moorthi, R. and Moe, S. (2013). Recent advances in the 59 Jadoul, M., Albert, J.M., Akiba, T. et al. (2006). Incidence
non-invasive diagnosis of renal osteodystrophy. Kidney and risk factors for hip or other bone fractures among
Int. 84 (5): 886–894. hemodialysis patients in the Dialysis Outcomes and
47 Leslie, W., Lix, L., Langsetmo, L. et al. (2011). Practice Patterns Study. Kidney Int. 70 (7): 1358–1366.
Construction of a FRAX® model for the assessment of 60 Coco, M. and Rush, H. (2000). Increased incidence of hip
farcture probability in Canada and implications for fractures in dialysis patients with low serum parathyroid
treatment. Osteoporos Int. 22 (3): 817–827. hormone. Am. J. Kidney Dis. 36 (6): 1115–1121.
48 Naylor, K.L., Garg, A.X., Zou, G. et al. (2015). 61 Elliott, M.J., James, M.T., Quinn, R.R. et al. (2013).
Comparison of fracture risk prediction among individuals Estimated GFR and fracture risk: a population-based
with reduced and normal kidney function. Clin. J. Am. study. Clin. J. Am. Soc. Nephrol. 8 (8): 1367–1376.
Soc. Nephrol. 10 (4): 646–653. 62 Ensrud, K.E., Parimi, N., Fink, H.A. et al. (2014).
49 Naylor, K.L., Leslie, W.D., Hodsman, A.B. et al. (2014). Estimated GFR and risk of hip fracture in older men:
FRAX predicts fracture risk in kidney transplant comparison of associations using cystatin C and
recipients. Transplantation 97 (9): 940–945. creatinine. Am. J. Kidney Dis. 63 (1): 31–39.
50 Kauppila, L.I., Polak, J.F., Cupples, L.A. et al. (1997). New 63 Naylor, K.L., Li, A.H., Lam, N.N. et al. (2013). Fracture
indices to classify location, severity and progression of risk in kidney transplant recipients: a systematic review.
calcific lesions in the abdominal aorta: a 25-year Transplantation 95 (12): 1461–1470.
follow-up study. Atherosclerosis 132 (2): 245–250. 64 Naylor, K.L., Jamal, S.A., Zou, G. et al. (2016). Fracture
51 Kraus, M., Kalra, P., Hunter, J. et al. (2015). The incidence in adult kidney transplant recipients.
prevalence of vascular calcification in patients with Transplantation 100 (1): 167–175.
Reference 613
5 Senanayake, S., Gunawardena, N., Palihawadana, P. et al.

6 79 Salusky, I.B., Foley, J., Nelson, P., and Goodman, W.G.
(2017). Symptom burden in chronic kidney disease; a (1991). Aluminum accumulation during treatment with
population based cross sectional study. BMC Nephrol. 18 aluminum hydroxide and dialysis in children and young
(1): 228. adults with chronic renal disease. N. Engl. J. Med. 324 (8):
66 Murtagh, F.E., Addington-Hall, J.M., Edmonds, P.M. et al. 527–531.
(2007). Symptoms in advanced renal disease: a cross- 80 Tonelli, M., Pannu, N., and Manns, B. (2010). Oral
sectional survey of symptom prevalence in stage 5 phosphate binders in patients with kidney failure. N.
chronic kidney disease managed without dialysis. J. Engl. J. Med. 362 (14): 1312–1324.
Palliat. Med. 10 (6): 1266–1276. 81 Chiu, Y.W., Teitelbaum, I., Misra, M. et al. (2009). Pill
67 Murtagh, F.E., Addington-Hall, J., and Higginson, I.J. burden, adherence, hyperphosphatemia, and quality of
(2007). The prevalence of symptoms in end-stage renal life in maintenance dialysis patients. Clin. J. Am. Soc.
disease: a systematic review. Adv. Chronic Kidney Dis. 14 Nephrol. 4 (6): 1089–1096.
(1): 82–99. 82 Spiegel, D.M., Farmer, B., Smits, G., and Chonchol, M.
68 Block, G., Hulbert-Shearon, T., Levin, N., and Port, F. (2007). Magnesium carbonate is an effective phosphate
(1998). Association of serum phosphorus and calcium x binder for chronic hemodialysis patients: a pilot study. J.
phosphate product with mortality risk in chronic Ren. Nutr. 17 (6): 416–422.
hemodialysis patients: a national study. Am. J. Kidney Dis. 83 Block, G., Wheeler, D., Persky, M. et al. (2012). Effects of
31 (4): 607–617. phosphate binders in moderate CKD. J. Am. Soc. Nephrol.
69 Block, G., Klassen, P., Lazarus, J. et al. (2004). Mineral 23: 1407–1415.
metabolism, mortality, and morbidity in maintenance 84 Liu, Z., Su, G., Gui, X. et al. (2015). Dietary interventions
hemodialysis. J. Am. Soc. Nephrol. 15 (8): 2208–2218. for mineral and bone disorder in people with chronic
70 Wang, M., Obi, Y., Streja, E. et al. (2017). Association of kidney disease. Cochrane Database Syst. Rev. (9) (Art. No.:
parameters of mineral bone disorder with mortality in CD010350). DOI: https://doi.org/10.1002/14651858.
patients on hemodialysis according to level of residual CD010350.pub2.
kidney function. Clin. J. Am. Soc. Nephrol. 12 (7): 85 Houston, J., Smith, K., Isakova, T. et al. (2013).
1118–1127. Associations of dietary phosphorus intake, urinary
71 National Kidney Foundation (2003). K/DOQI clinical phosphate excretion, and fibroblast growth factor 23 with
practice guidelines for bone metabolism and disease in vascular stiffness in chronic kidney disease. J. Ren. Nutr.
chronic kidney disease. Am. J. Kidney Dis. 42 (4 Suppl 3): 23 (1): 12–20.
S1–S201. 86 Selamet, U., Tighiouart, H., Sarnak, M.J. et al. (2016).
72 Alfrey, A. (1993). Aluminum toxicity in patients with Relationship of dietary phosphate intake with risk of
chronic renal failure. Ther. Drug Monit. 15: 593–597. end-stage renal disease and mortality in chronic kidney
73 de Wolff, F. (1985). Toxicological aspects of aluminum disease stages 3-5: The Modification of Diet in Renal
poisoning in clinical nephrology. Clin. Nephrol. 24 (Suppl Disease Study. Kidney Int. 89 (1): 176–184.
1): S9–S14. 87 The FHN Trial Group (2010). In-center hemodialysis six
74 Canata-Andía, J. and Fernández-Martín, J. (2002). The times per week versus three times per week. N. Engl. J.
clinical impact of aluminum overload in renal failure. Med. 363: 2287–2300.
Nephrol. Dial. Transplant. 17 (Suppl 2): 17–20. 88 Rocco, M.V., Lockridge, R.S. Jr., Beck, G.J. et al. (2011).
75 Andreoli, S., Bergstein, J., and Sherrard, D. (1984). The effects of frequent nocturnal home hemodialysis: the
Aluminum intoxication from aluminum-containing Frequent Hemodialysis Network Nocturnal Trial. Kidney
phosphate binders in children with azotemia Int. 80 (10): 1080–1091.
not undergoing dialysis. N. Engl. J. Med. 310: 89 Berl, T., Berns, A.S., Hufer, W.E. et al. (1978). 1,25
1079–1084. dihydroxycholecalciferol effects in chronic dialysis. A
76 Parkinson, I., Ward, M., Feest, T. et al. (1979). double-blind controlled study. Ann. Intern. Med. 88 (6):
Fracturing dialysis osteodystrophy and dialysis 774–780.
encephalopathy. An epidemiological survey. Lancet i 90 Bordier, P., Zingraff, J., Gueris, J. et al. (1978). The effect
(8113): 406–409. of 1alpha(OH)D3 and 1alpha,25(OH)2D3 on the bone in
77 Berlyne, G.M., Ben-Ari, J., Pest, D. et al. (1970). patients with renal osteodystrophy. Am. J. Med. 64 (1):
Hyperaluminaemia from aluminum resins in renal 101–107.
failure. Lancet 2 (7671): 494–496. 91 DeLuca, H.F. and Avioli, L.V. (1970). Treatment of renal
78 Wrong, O.M. and Swales, J.D. (1970). Hyperaluminaemia osteodystrophy with 25-hydroxycholecalciferol. Arch.
form aluminium resins. Lancet 2 (7683): 1130–1131. Intern. Med. 126 (5): 896–899.
92 Brickman, A.S., Coburn, J.W., and Norman, A.W. (1972). in patients with chronic kidney disease: the PRIMO
Action of 1,25-dihydroxycholecalciferol, a potent, randomized controlled trial. JAMA 307 (7): 674–684.
kidney-produced metabolite of vitamin D, in uremic 104 Palmer, S.C., McGregor, D.O., Macaskill, P. et al. (2007).
man. N. Engl. J. Med. 287 (18): 891–895. Meta-analysis: vitamin D compounds in chronic kidney
93 Brickman, A.S., Coburn, J.W., Massry, S.G., and disease. Ann. Intern. Med. 147 (12): 840–853.
Norman, A.W. (1974). 1,25 Dihydroxy-vitamin D3 in 105 Kandula, P., Dobre, M., Schold, J.D. et al. (2011).
normal man and patients with renal failure. Ann. Intern. Vitamin D supplementation in chronic kidney disease: a
Med. 80 (2): 161–168. systematic review and meta-analysis of observational
94 Christiansen, C., Rodbro, P., Christensen, M.S., and studies and randomized controlled trials. Clin. J. Am.
Hartnack, B. (1981). Is 1,25-dihydroxy-cholecalciferol Soc. Nephrol. 6 (1): 50–62.
harmful to renal function in patients with chronic renal 106 Sprague, S.M., Crawford, P.W., Melnick, J.Z. et al.
failure? Clin. Endocrinol. (Oxf). 15 (3): 229–236. (2016). Use of extended-release calcifediol to treat
95 Christiansen, C., Roodbro, P., Christensen, M.S. et al. secondary hyperparathyroidism in stages 3 and 4
(1980). Decreased renal function in association with chronic kidney disease. Am. J. Nephrol. 44 (4):
administration of 1,25-dihydroxyvitamin D3 to patients 316–325.
with stable, advanced renal failure. Contrib. Nephrol. 18: 107 Sprague, S.M., Silva, A.L., Al-Saghir, F. et al. (2014).
139–146. Modified-release calcifediol effectively controls
96 Ritz, E., Kuster, S., Schmidt-Gayk, H. et al. (1995). secondary hyperparathyroidism associated with vitamin
Low-dose calcitriol prevents the rise in 1,84-iPTH D insufficiency in chronic kidney disease. Am. J.
without affecting serum calcium and phosphate in Nephrol. 40 (6): 535–545.
patients with moderate renal failure (prospective 108 European Medicines Agency (2017). EPAR summary for
placebo-controlled multicentre trial). Nephrol. Dial. the public. Mimpara. Cincalcet. http://www.ema.
Transplant. 10 (12): 2228–2234. europa.eu/ema/index.jsp?curl=pages/medicines/
97 Sprague, S.M. and Moe, S.M. (1992). Safety and efficacy human/medicines/000570/human_med_000903.
of long-term treatment of secondary jsp&mid=WC0b01ac058001d124 (access 22 May 2021).
hyperparathyroidism by low-dose intravenous calcitriol. 109 Lindberg, J.S., Culleton, B., Wong, G. et al. (2005).
Am. J. Kidney Dis. 19 (6): 532–539. Cinacalcet HCl, an oral calcimimetic agent for the
98 Dobrez, D.G., Mathes, A., Amdahl, M. et al. (2004). treatment of secondary hyperparathyroidism in
Paricalcitol-treated patients experience improved hemodialysis and peritoneal dialysis: a randomized,
hospitalization outcomes compared with calcitriol- double-blind, multicenter study. J. Am. Soc. Nephrol. 16
treated patients in real-world clinical settings. Nephrol. (3): 800–807.
Dial. Transplant. 19 (5): 1174–1181. 110 Charytan, C., Coburn, J.W., Chonchol, M. et al. (2005).
99 Kovesdy, C.P., Ahmadzadeh, S., Anderson, J.E., Cinacalcet hydrochloride is an effective treatment for
and Kalantar-Zadeh, K. (2008). Association of secondary hyperparathyroidism in patients with CKD
activated vitamin D treatment and mortality in not receiving dialysis. Am. J. Kidney Dis. 46 (1): 58–67.
chronic kidney disease. Arch. Intern. Med. 168 (4): 111 Strippoli, G.F., Tong, A., Palmer, S.C. et al. (2006).
397–403. Calcimimetics for secondary hyperparathyroidism in
100 Zheng, Z., Shi, H., Jia, J. et al. (2013). Vitamin D chronic kidney disease patients. Cochrane Database
supplementation and mortality risk in chronic kidney Syst. Rev. (4) (Art. No.:Cd006254).
disease: a meta-analysis of 20 observational studies. 112 Palmer, S.C., Nistor, I., Craig, J.C. et al. (2013).
BMC Nephrol. 14: 199. Cinacalcet in patients with chronic kidney disease: a
101 Tentori, F., Albert, J.M., Young, E.W. et al. (2009). The cumulative meta-analysis of randomized controlled
survival advantage for haemodialysis patients taking trials. PLoS Med. 10 (4): e1001436.
vitamin D is questioned: findings from the Dialysis 113 Chertow, G.M., Block, G.A., Correa-Rotter, R. et al.
Outcomes and Practice Patterns Study. Nephrol. Dial. (2012). Effect of cinacalcet on cardiovascular disease in
Transplant. 24 (3): 963–972. patients undergoing dialysis. N. Engl. J. Med. 367 (26):
102 Wang, A.Y., Fang, F., Chan, J. et al. (2014). Effect of 2482–2244.
paricalcitol on left ventricular mass and function in 114 Parfrey, P.S., Drueke, T., Block, G.A. et al. (2015). The
CKD--the OPERA trial. J. Am. Soc. Nephrol. 25 (1): effects of cinacalcet in older and younger patients on
175–186. hemodialysis: the evaluation of cinacalcet HCl therapy
103 Thadhani, R., Appelbaum, E., Pritchett, Y. et al. (2012). to lower cardiovascular events (EVOLVE) trial. Clin. J.
Vitamin D therapy and cardiac structure and function Am. Soc. Nephrol. 10 (5): 791–799.
Reference 615
115 Bell, G., Huang, S., Martin, K.J., and Block, G.A. (2015). 127 Kim, H.W., Kim, S.H., Kim, Y.O. et al. (2015). Impact of
A randomized, double-blind, phase 2 study evaluating dialysate calcium concentration on clinical outcomes in
the safety and efficacy of AMG 416 for the treatment of incident hemodialysis patients. Medicine (Baltimore) 94
secondary hyperparathyroidism in hemodialysis (40): e1694.
patients. Curr. Med. Res. Opin. 31 (5): 943–952. 128 Masterson, R., Blair, S., Polkinghorne, K.R. et al. (2017).
116 Block, G.A., Bushinsky, D.A., Cunningham, J. et al. Low versus high dialysate calcium concentration in
(2017). Effect of etelcalcetide vs placebo on serum alternate night nocturnal hemodialysis: a randomized
parathyroid hormone in patients receiving hemodialysis controlled trial. Hemodial. Int. 21 (1): 19–28.
with secondary hyperparathyroidism: two randomized 129 Ok, E., Asci, G., Bayraktaroglu, S. et al. (2016).
clinical trials. JAMA 317 (2): 146–155. Reduction of dialysate calcium level reduces progression
117 Block, G.A., Bushinsky, D.A., Cheng, S. et al. (2017). of coronary artery calcification and improves low bone
Effect of etelcalcetide vs cinacalcet on serum turnover in patients on hemodialysis. J. Am. Soc.
parathyroid hormone in patients receiving hemodialysis Nephrol. 27 (8): 2475–2486.
with secondary hyperparathyroidism: a randomized 130 Tentori, F., McCullough, K., Kilpatrick, R.D. et al. (2014).
clinical trial. JAMA 317 (2): 156–164. High rates of death and hospitalization follow bone fracture
118 Behets, G.J., Spasovski, G., Sterling, L.R. et al. (2015). Bone among hemodialysis patients. Kidney Int. 85 (1): 166–173.
histomorphometry before and after long-term treatment 131 Klawansky, S., Komaroff, E., Cavanaugh, P. Jr. et al.
with cinacalcet in dialysis patients with secondary (2003). Relationship between age, renal function and
hyperparathyroidism. Kidney Int. 87 (4): 846–856. bone mineral desnity in the US population. Osteoporos.
119 Lien, Y., Silva, A., and Whittman, D. (2005). Effects of Int. 14: 570–576.
cinacalcet on bone mineral density in patients with 132 Cunningham, J., Sprague, S., Canata-Andía, J. et al.
secondary hyperparathyroidism. Nephrol. Dial. (2004). Osteoporosis in chronic kidney disease. Am. J.
Transplant. 20 (6): 1232–1237. Kidney Dis. 43 (3): 566–571.
120 Palazzo, F. and Sadler, G. (2004). Minimally invasive 133 Domrongkitchaiporn, S., Pongsakul, C.,
parathyroidectomy. Br. Med. J. 328 (7444): 849–850. Stitchantrakul, W. et al. (2001). Bone mineral density
121 Sejean, K., Calmus, S., Durand-Zaleski, I. et al. (2005). and histology in distal renal tubular acidosis. Kidney
Surgery versus medical follow-up in patients with Int. 59: 1086–1093.
asymptomatic primary hyperparathyroidism: a decision 134 Rodan, G. and Fleisch, H. (1986). Bisphosphonates:
analysis. Eur. J. Endocrinol. 153 (6): 915–927. mechanism of action. J. Clin. Invest. 97 (12): 2692–2696.
122 Coen, G., Calabria, S., Bellinghieri, G. et al. (2001). 135 Saha, H., Castren-Kortekangas, P., Ojanen, S. et al.
Parathyroidectomy in chronic renal failure: short-and (1994). Pharmacokinetics of clodronate in renal failure.
long-term results on parathyroid function, blood J. Bone Miner. Metab. 9 (12): 1953–1958.
pressure and anemia. Nephron 88 (2): 149–155. 136 MacLean, C., Newberry, S., Maglione, M. et al. (2008).
123 Gagne, E.R., Urena, P., Leite-Silva, S. et al. (1992). Systematic review: comparative effectiveness of treatments
Short-and long-term efficacy of total parathyroidectomy to prevent fractures in men and women with low bone
with immediate autografting compared with subtotal density or osteoporosis. Ann. Intern. Med. 148 (3): 197–213.
parathyroidectomy in hemodialysis patients. J. Am. Soc. 137 Wilson, L.M., Rebholz, C.M., Jirru, E. et al. (2017).
Nephrol. 3 (4): 1008–1017. Benefits and harms of osteoporosis medications in
124 Lorenz, K., Ukkat, J., Sekulla, C. et al. (2006). Total patients with chronic kidney disease: a systematic review
parathyroidectomy without autotransplantation for and meta-analysis. Ann. Intern. Med. 166 (9): 649–658.
renal hyperparathyroidism: experience with a qPTH- 138 Jamal, S.A., Bauer, D.C., Ensrud, K. et al. (2007).
controlled protocol. World J. Surg. 30 (5): 743–751. Alendronate treatment in women with normal to severely
125 Young EW, Albert JM, Satayathum S, Goodkin DA, impaired renal function: an analysis of the fracture
Pisoni RL, Akiba T, Akizawa T, Kurokawa K, Bommer J, intervention trial. J. Bone Miner. Res. 22 (4): 503–508.
Piera L, Port FK. Predictors and consequences of altered 139 Miller, P., Roux, C., Boonen, S. et al. (2005). Safety and
mineral metabolism: The Dialysis Outcomes and efficacy of risedronate in patients with age-related
Practice Patterns Study. Kidney International reduced renal function as estimated by the Cockcroft
2005;67(3):1179–1187. and Gault method: a pooled analysis of nine clinical
126 Brunelli, S.M., Sibbel, S., Do, T.P. et al. (2015). Facility trials. J. Bone Miner. Res. 20: 2105–2115.
dialysate calcium practices and clinical outcomes 140 Aryoshi, T., Eishi, K., Sakamoto, I. et al. (2006). Effect of
among patients receiving hemodialysis: a retrospective etirdronic acid on arterial calcification in dialysis
observational study. Am. J. Kidney Dis. 66 (4): 655–665. patients. Clin. Drug Invest. 26: 215–222.
141 Toussaint, N.D., Lau, K.K., Strauss, B.J. et al. (2010). 147 Hernández, E., Valera, R., Alonzo, E. et al. (2003).
Effect of alendronate on vascular calcification in CKD Effects of raloxifene on bone metabolism and serum
stages 3 and 4: a pilot randomized controlled trial. Am. lipids in postmenopausal women on chronic
J. Kidney Dis. 56 (1): 57–68. hemodialysis. Kidney Int. 63 (6): 2269–2274.
142 Markowitz, G., Appel, G., Fine, P. et al. (2001). 148 Haghverdi, F., Farbodara, T., Mortaji, S. et al. (2014).
Collapsing focal and segmental glomerulosclerosis Effect of raloxifene on parathyroid hormone in
following treatment with high-dose pamidronate. J. Am. osteopenic and osteoporotic postmenopausal women
Soc. Nephrol. 12 (6): 1164–1172. with chronic kidney disease stage 5. Iran. J. Kidney Dis.
143 Perazella, M.A. and Markowitz, G.S. (2008). 8 (6): 461–466.
Bisphosphonate nephrotoxicity. Kidney Int. 74 (11): 149 Ishani, A., Blackwell, T., Jamal, S.A. et al. (2008). The
1385–1393. effect of raloxifene treatment in postmenopausal women
144 McClung, M.R., Lewiecki, E.M., Cohen, S.B. et al. with CKD. J. Am. Soc. Nephrol. 19 (7): 1430–1438.
(2006). Denosumab in postmenopausal women with 150 Palmer, S., Chung, E., McGregor, D. et al. (2017).
low bone mineral density. N. Engl. J. Med. 354 (8): Interventions for preventing bone disease in kidney
821–831. transplant recipients. Cochrane Database Syst. Rev.
145 Cummings, S.R., San Martin, J., McClung, M.R. et al. 10.1002/14651858.CD005015.pub4 Article number
(2009). Denosumab for prevention of fractures in CD005015
postmenopausal women with osteoporosis. N. Engl. J. 151 Tong, A., Crowe, S., Chando, S. et al. (2015). Research
Med. 361 (8): 756–765. priorities in CKD: report of a National Workshop Conducted
146 Jamal, S.A., Ljunggren, O., Stehman-Breen, C. et al. in Australia. Am. J. Kidney Dis. 66 (2): 212–222.
(2011). Effects of denosumab on fracture and bone 152 Heiwe, S. and Jacobson, S.H. (2011). Exercise training
mineral density by level of kidney function. J. Bone for adults with chronic kidney disease. Cochrane
Miner. Res. 26 (8): 1829–1835. Database Syst. Rev. (10) (Art. No.:Cd003236).
617
37
Nutritional Disorders in Chronic Kidney Disease: Pathophysiology, Detection,

and Treatment
Pablo Molina1,2, Belén Vizcaíno1, Emma Huarte3, Luis M. Pallardó1, and Juan J. Carrero2,4
1
Department of Nephrology, Hospital Universitario Dr Peset, Universitat de València, Valencia, Spain
2
European Renal Nutrition Working Group of the European Renal Association – European Dialysis Transplant Association, Parma, Italy
3
Department of Nephrology, Hospital San Pedro, Logroño, La Rioja, Spain
4
Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden
I ntroduction [GH] and insulin-like growth factor 1 [IGF-1] deficiencies,

hyperinsulinism, hyperparathyroidism) [23–37].
Nutritional status is one of the main modifiable factors that Nutritional interventions should be a cornerstone strat-
affect the prognosis and evolution of patients with acute egy in the management of CKD patients, both as a renopro-
and chronic kidney disease (CKD) (Table 37.1). In CKD, tective antiproteinuric measure in the predialysis stage,
dietary factors may have an influence on important clinical and to prevent overweight and malnutrition in all stages of
outcomes, including disease progression, quality of life, CKD, the latter especially in later stages. Given the number
morbidity, and mortality. Due to the kidney’s unique role of factors involved in the nutritional and metabolic status
in nutrient metabolism, patients with kidney failure are of the CKD patient, adequate dietary management will
unable to maintain adequate nutrient homeostasis. As kid- require a combination of therapeutic measures that mini-
ney disease progresses, there is a reduction in the excretion mize energy-protein depletion and favor the anabolic stim-
of solutes, which results in sodium and volume overload, ulus. Many of these measures, such as the correction of
hyperkaliemia, hyperphosphatemia, and metabolic acido- acidosis and the modification of renal replacement ther-
sis. In addition, uremic toxins accumulation leads to apy, go beyond simple caloric or dietary support [38, 39].
reduced appetite with increased risk for suboptimal energy The optimal approach to nutrition is not well known, and
and nutrient intake, altered hormone regulation, and randomized controlled trials (RCTs) are lacking and have
inflammation, all together exacerbating the risk for the so- yielded conflicting results. This chapter examines the evi-
called protein-energy wasting (PEW) syndrome. This syn- dence for management of nutrition in adults with CKD
drome is defined as a state in which there is a decrease in stages 3–5, including dialysis and transplant patients, focus-
protein and energy stores, usually associated with a lower ing mainly on CKD progression and PEW treatment.
functional ability to adapt to stress situations and worse Evidence for nutritional interventions for CKD mineral and
patient survival [18–22]. bone disorders (CKD-MBD), and for children with kidney
Many factors contribute to the nutritional status of the failure is discussed in Chapters 36 and 38, respectively.
CKD patient, such that there is no definitive marker to
diagnose PEW nor a single treatment to reverse this state.
Nutritional Requirements for Patients
In the CKD patient, PEW can be produced both by under-
with Kidney Disease
nutrition and by other factors unrelated to nutrient intake,
such as the appearance of intercurrent catabolic processes, Because a “kidney diet” can have many restrictions, adher-
the loss of nutrients through urine or dialysate, metabolic ence to such a diet can be difficult and troublesome. In
acidosis, and endocrine alterations (e.g. growth hormone addition, the optimal diet for CKD patients varies depending
Table 37.1 Nutritional interventions for CKD.
Strength of
Dietary interventions, We suggest dietary interventions can be used to improve health-related quality of life, blood High (blood pressure control) Weak [1]
including dietary pressure control, eGFR, and nutritional status assessed by albumin levels. Patients with CKD Moderate (serum albumin levels)
patterns and (any stage) may reasonably choose not to receive dietary interventions based on the very low Low (health-related quality of life,
nutritional counseling certainty that treatment makes a difference to longer term outcomes including survival, eGFR)
cardiovascular morbidity, and development of ESRD.
Very low (mortality, cardiovascular
events, progression to ESRD)
Dietary protein We suggest dietary protein restriction can be used to reduce the risk for development of ESRD High (mortality) Weak [2–5]
restriction, including in both nondiabetic and diabetic patients with CKD stages 3–5, and to reduce both mortality Moderate (development of ESRD)
LPD and VLPD and risk for development of ESRD in type 1 diabetic patients with ND-CKD. Patients with Very low (development of PEW)
CKD may reasonably choose not to receive dietary protein restriction as treatment makes no
difference to mortality or PEW development.
Dietary salt restriction We suggest dietary salt restriction can be used to reduce blood pressure, antihypertensive High (blood pressure, proteinuria) Weak [6, 7]
medication dosage, proteinuria, and edema for patients with CKD. Very low (mortality, cardiovascular
Patients with CKD may reasonably choose not to receive dietary salt restriction based on the morbidity, development of ESRD)
very low certainty that treatment makes a difference to longer term outcomes including
health-related quality of life, survival, cardiovascular morbidity, and the development of ESRD.
Dietary caloric We suggest dietary caloric restriction can be used to reduce proteinuria and blood pressure, as Moderate (blood pressure, Weak [8]
restriction to reduce well as hyperfiltration in ND-CKD patients. Patients with CKD may reasonably choose not to proteinuria)
weight receive dietary caloric restriction based on the very low certainty that treatment makes a Very low (development of ESRD)
difference to the development of ESRD.
Probiotics We suggest probiotics supplementation may be used to reduce p-cresyl sulfate levels. Patients Very low Weak [9]
supplementation with CKD may reasonably choose not to receive probiotics supplementation based on the very
low certainty that treatment makes a difference to longer term outcomes including health-
related quality of life, survival, cardiovascular morbidity, and the development of ESRD.
Oral nutritional We suggest oral nutritional supplements may be used to improve nutritional status by Very low Weak [10]
supplements increasing serum albumin levels and body mass index in dialysis patients, without influence
on serum potassium and phosphorus levels. Patients with CKD (any stage) may reasonably
choose not to receive probiotics supplementation based on the very low certainty that
treatment makes a difference to survival.
Intradialytic parenteral We suggest intradialytic parenteral nutrition may be tested to improve nutritional status in Very low Weak [11]
nutrition hemodialysis patients with PEW if nutrient requirements can be achieved in combination
with oral intake. Hemodialysis patients with PEW may reasonably choose not to receive
intradialytic parenteral nutrition based on the very low certainty that treatment makes a
difference to longer term outcomes including survival and morbidity.
0005152425.INDD 618 09-12-2022 16:10:52

Megestrol acetate We suggest megestrol acetate may be tested in dialysis patients to reverse poor appetite and Very low Weak [12]
improve nutritional status if other treatment options are unavailable and for less than 24 weeks.
Dialysis patients with poor appetite may reasonably choose not to receive megestrol acetate
based on safety concerns and the very low certainty that treatment makes a difference to longer
term outcomes including health-related quality of life, survival, and morbidity.
Fish oil supplements We suggest fish oil supplements may be used in dialysis patients to alleviate chronic High (hypertriglyceridemia) Weak [13]
inflammation status, depression symptoms, and hypertriglyceridemia, and reduce Moderate (C-reactive protein levels)
cardiovascular events and the risk of arterio-venous-graft events. Dialysis patients may Low (depression symptoms,
reasonably choose not to receive fish oil supplements based on the very low certainty that cardiovascular events, arterio-
treatment makes a difference to longer term outcomes including health-related quality of life venous-graft events)
and survival.
Very low (health-related quality of
life, mortality, ND-CKD)
l-Carnitine We suggest l-carnitine supplements may be tested in dialysis patients to alleviate chronic Very low Weak [14, 15]
supplementation inflammation status. Dialysis patients may reasonably choose not to receive l-carnitine
supplements based on the very low certainty that treatment makes a difference to longer term
outcomes including health-related quality of life, survival, and comorbidity.
Statins We suggest statins may be used in dialysis patients to alleviate chronic inflammation status and High (C-reactive protein levels, Weak [16]
reduce cholesterol levels. Dialysis patients may reasonably choose not to receive statins based cholesterol levels)
on the moderate to very low certainty that treatment makes a difference to longer term Moderate (survival and
outcomes including health-related quality of life, survival, and cardiovascular events. cardiovascular events)
Very low (health-related quality of
life, ND-CKD)
CKD, chronic kidney disease; LPD, low protein diet; VLPD, very low protein diet; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; ND-CKD, nondialysis chronic
kidney disease.
GRADE assessment of the certainty of the evidence [17]: High: This research provides a very good indication of the likely effect. The likelihood that the effect will be substantially different* is
low. Moderate: This research provides a good indication of the likely effect. The likelihood that the effect will be substantially different* is moderate. Low: This research provides some
indication of the likely effect. However, the likelihood that it will be substantially different* is high. Very low: This research does not provide a reliable indication of the likely effect. The
likelihood that the effect will be substantially different* is very high. *Substantially different = a large enough difference that it might affect decision-making.
0005152425.INDD 619 09-12-2022 16:10:52

620 Nutritional Disorders in Chronic Kidney Disease: Pathophysiology, Detection, and Treatment
upon their nutritional needs, the CKD stage, the type of rogression as well as ameliorating uremic symptoms [41].
p
renal replacement therapy, and the presence of other However, the nutrient content of each type of diet tends to
comorbidities such as PEW, obesity, hypertension, be steady and constant, and individual dietary habits usu-
proteinuria, or heart failure. Patient preferences, their ally remain unchanged over time, therefore the overall
cultural eating habits, and their economic status are dietary pattern may be more decisive for patient outcomes
additional factors to be kept in mind for improving their than an excess or deficiency of specific nutrients [42].
compliance with dietary recommendations, which should
be monitored regularly. Overall Recommendations
The current Kidney Disease Outcomes Quality Initiative
Need for Individualization (K/DOQI) guidelines recommend a protein intake ranging
Too many restrictions should be avoided as they may result from 0.6 to 1.2 g/kg body weight per day according to CKD
in a reduced nutrient intake. Dietary modifications are stage, of which more than half should be of high biological
seldom required for patients with estimated glomerular value. The recommended energy intake is 35 kcal/kg body
filtration rate (eGFR) 60 ml/min/1.73 m2. Such patients weight per day for those who are younger than 60 years of
should be advised to implement the same dietary recom- age, and 30–35 kcal/kg/day for individuals who are 60 years
mendations as for the general population, consisting of low or older [43]. These guidelines do not refer to the recom-
sodium and refined sugar, avoidance of red and processed mended intake of carbohydrates, lipids, or most micronu-
meats, and high content of fruits, vegetables, legumes, fish, trients, with the exception of recommendations for calcium
poultry, and whole grains. In later stages of CKD, however, and phosphorus intake. The European guidelines give
the diet needs to be modified (Figure 37.1) [40]. Historically, recommendations regarding the intake of vitamins,
research and guideline recommendations have been minerals, and trace elements, although they recognize that
focused mainly on modifying single micro-or macronutri- in the absence of RCTs such recommendations cannot be
ents intake, such as protein, sodium, potassium, or considered true clinical guidelines, but rather reflect only
phosphate. Providing specific nutrients may prevent and the opinion of experts [44]. Overall, many of the nutritional
treat PEW, whereas limiting the intake of certain nutrients requirements indicated for the CKD population are simple
may prevent obesity, hypertension, and proteinuria, extrapolations based on the recommendations made for
reducing the risk of cardiovascular disease and CKD the general population, with the added classic limitations
Prevent
overweight Maintain desirable weight Maintain desirable weight Prevent obesity Maintain desirable weight
Consider dietary protein Consider dietary protein

Promote healthy diet: restriction (stage 3b) restriction
Promote healthy diet:
Delay CKD high content of fruits,
high content of fruits,
vegetables, legumes, fish, Dietary and lifestyle Dietary and lifestyle
progression poultry, and whole grains; modifications: limiting salt and modifications: limiting salt and
Dietary and lifestyle vegetables, legumes, fish,
modifications: limiting salt and poultry, and whole grains;
low salt and refined sugar processed foods with processed foods with
processed foods with low salt and refined sugar
intakes; avoid red and phosphorous-based additives, phosphorous-based additives,
phosphorous-based additives, intakes; avoid red and
processed meats and encouraging home-cooked and encouraging home-cooked
and encouraging home-cooked processed meats
meals from fresh ingredients meals from fresh ingredients
meals from fresh ingredients
Ameliorate (preferably plant-based foods) (preferably plant-based foods)
uremic
symptoms Manage electrolytes and fluids Manage electrolytes and fluids
Ensure adequate Ensure adequate caloric Ensure adequate caloric High-protein diet Ensure adequate
Prevent caloric intake
Maintain skeletal
intake, especially if low dietary
protein intake is prescribed
intake, especially if low dietary caloric intake
Maintain skeletal
protein intake is prescribed
PEW muscle stores Maintain skeletal muscle stores Maintain skeletal muscle stores Ensure adequate caloric intake muscle stores
Maintain skeletal muscle stores
Early CKD Moderate CKD Advanced CKD ESRD Transplantation
(stages 1–2) (stage 3) (stages 4–5) (stage 5D)
Figure 37.1 Schematic representation of the relative importance of each of the main nutritional goals across the CKD stages,
including preventing overweight, delaying the development of ESRD, decreasing uremic symptoms, and preventing PEW. CKD, chronic
kidney disease; PEW, protein-energy wasting.
Introductio 621
in potassium intake, sodium, and phosphorus, which have total daily energy intake [59]. This additional source of
prevailed for more than 40 years in patients on dialysis. calories should be taken into consideration when assess-
However, many of these classic recommendations, ing energy intake in this population.
which include restricting the intake of fruits, vegetables, or ●● In kidney transplant patients, the caloric intake should
dairy products, are probably not adequate for current be adapted to individual needs according to physical
patient characteristics (i.e. old and fragile) and treatments, activity and presence of low weight or obesity.
which range from conservative management to more ●● Any intercurrent catabolic episode may require increas-
intensive dialysis regimens, including nocturnal or short- ing energy (and protein intake) independently of CKD
daily hemodialysis (HD) [45]. Conversely, what has stage or renal replacement therapy (Table 37.2).
recently been proposed is an increase in the intake of fiber,
antioxidants, and phytochemicals, which could have a very
Protein Requirements and Source of Protein Intake
positive impact on the nutritional and cardiovascular sta-
For ND-CKD stable patients, the recommended protein
tus of CKD patients [46, 47]. In this sense, recommenda-
intake is 0.6–0.8 g/kg/day, of which at least half should be
tions should not only be based on the quantity but also on
of high biological value [43, 48]. This restriction is gener-
the quality of each food, so that the choice, for example, of
ally well tolerated, limiting waste-product accumulation
unprocessed foods may allow us to abandon unnecessary
while simultaneously preserving nutrition. Benefits of
restrictions that can affect the quality of life of patients [45].
lower protein diets include slowing CKD progression and
In addition, stable patients with precise dietary indica-
reducing uremic symptoms [56, 58, 60]. Reduced protein
tions, both those with CKD and those on dialysis or trans-
intake is also associated with favorable laboratory and met-
planted, can develop an intercurrent acute process that
abolic effects, including reduction of uremic toxins and
requires substantially increasing their energy and protein
acid load, with better control of hyperkalemia and hyper-
requirements.
phosphatemia [61, 62]. Figure 37.2 summarizes the poten-
Table 37.2 shows the nutritional requirements for adult
tial benefits of dietary protein restriction.
patients with nondialysis CKD (ND-CKD), HD, peritoneal
Nevertheless, the optimal degree of protein restriction
dialysis (PD), and transplant, according to the main clini-
remains unclear, and there is no consensus on the recom-
cal guidelines and consensus papers [43, 44, 48–55].
mended level of protein restriction according to the degree
of kidney dysfunction. There are basically two alternatives:
Energy Requirements
the so-called low protein diet (LPD, 0.6–0.8 g/kg/day) and
Adequate caloric intake – by avoiding processed carbohy-
the very low protein diet (VLPD, 0.3–0.4 g/kg/day) supple-
drates and saturated fat intake, and by increasing complex
mented with 7–15 g/day of ketoanalogs and essential
carbohydrates and mono-and polyunsaturated fats – must
amino acids. In both cases, patients should be carefully
be maintained to avoid malnutrition, and conversely to
monitored for adequate caloric intake (30–35 kcal/kg/day)
prevent overweight and obesity, given their contribution to
that allows a positive nitrogen balance and so that proteins
the cardiovascular risk and kidney disease progression.
are not used as a caloric source. [41, 62] The 2012 Kidney
Due to the increased resting energy expenditure (REE) sec-
Disease, Improving Global Outcomes (KDIGO) guidelines
ondary to inflammation and comorbidities linked to CKD,
for the management of CKD recommended lowering
as well as for preserving a neutral or positive nitrogen bal-
dietary protein intake to 0.8 g/kg/day in CKD stage 4–5
ance, current guidelines recommend a caloric intake of
subjects with diabetes (grade 2C recommendation) and
30–35 kcal/kg/day, which is higher than for the general
without diabetes (grade 2B recommendation), and avoid-
population. However, this general recommendation should
ing a high protein intake (>1.3 g/kg/day) in adults with
be individualized according to the patient’s profile, includ-
CKD at risk of progression (recommendation grade 2C). It
ing age, lean body mass (which is the primary determinant
considers the restriction of 0.6 g/kg/day nutritionally
of energy expenditure), physical activity, and the type of
adequate given that it facilitates better control of uremic
kidney disease [56–58].
complications [63, 64].
●● In ND-CKD patients under a low protein diet (LPD), an There is no consensus on whether the source of proteins
energy intake of 35 kcal/kg/day may be recommended to differently impacts on CKD progression risk. Although
ensure a neutral nitrogen balance. By contrast, in seden- recent observational studies have suggested that plant pro-
tary, elderly, and obese patients the energy contribution teins may have more reno-protective effects than animal
should provide around 30 kcal/kg/day. proteins [41], confounding factors inherent to a diet rich in
●● In PD patients, about 300 kcal may be provided in each protein from plant sources (i.e. a higher intake of vitamins
glucose-based exchange, contributing up to 30% of the and antioxidants) make it difficult to draw conclusions [42].
Table 37.2 Summary of nutritional requirements for patients with kidney failure.
Early CKD (stages1–2) CKD stages 3–5 Hemodialysis Peritoneal dialysis Transplantation
Energy (kcal/kg/ 30–35 30–35 30–35 30–35 (including kcal from 30–35
day)a dialysate)
Protein (g/kg/ 0.8 0.6–0.8 1.2 if IHD 1.2 0.8
day)b 1.0 if intercurrent catabolic 0.3–0.4 in selective patients with 1.5 if intercurrent 1.5 if peritonitis 1.4 for the first 4 weeks after
episode CKD 5 and conservative treatment catabolic episode transplantation or if high
0.8 (+1 g/day of high biological Keto/amino acid supplementation 1.7 if CRRT dose of prednisone is
value protein for each g/day of may be considered together with required
urinary protein loss) if dietitian support 0.6–0.8 if CKD stages 3–5T
nephrotic syndrome 1.0 if intercurrent catabolic
episode
Fiber (g/day) 25–38 25–38 25–38 25–38 25–38
Sodium (mg/day) <2.0–2.4 <2.0–2.4 <2.0–2.4 <2–2.4 <2–2.4
Potassium (mEq/ No potassium restriction 51–102 40–70 Does not usually need Potassium restriction should
day)c Potassium restriction should be If long interdialytic restriction be prescribed based on the
prescribed based on serum interval restriction Patients with hypokalemia serum potassium levels
potassium levels must be should increase their
mandatory potassium intake
Vitamin D (IU/ 600–800 600–800 600–800 600–800 600–800
day)
Calcium (g/day) 800–1000 800–1000 800–1000 800–1000 800–1000
Phosphorus (mg/ 800–1000 800–1000 <800 <800 800–1300
day)d
Magnesium (mg/ 200–300 200–300 200–300 200–300 200–300
day)
Iron (mg/day)e 10–18 10–18 10–18 10–18 10–18
Zinc (mg/day) f 15 15 15 15 15
CKD, chronic kidney disease; IHD, intermittent hemodialysis; CRRT, continuous renal replacement therapy; T, transplant.
a
Carbohydrates, which provide 40–60% of the daily energy intake, should be natural (nonrefined) and complex with a high-fiber content; 35 kcal/kg ideal body weight/day for those who are
less than 60 years of age and 30–35 kcal/kg ideal body weight/day for individuals 60 years or older and sedentary people.
b
At least 50% of dietary protein should be of high biological value.
c
First avoid high-potassium foods with poor nutritional value and correct other causes of hyperkaliemia such as metabolic acidosis or renin-angiotensin-aldosterone system inhibitors before
restricting healthy foods.
d
Limiting processed foods with phosphorous-based additives and encouraging home-cooked meals from fresh ingredients (preferably plant-based foods) should be the first-line interventions
for phosphorus restriction; the next-line strategy may be to exclude that dietary protein intake is higher than recommended, and limiting foods with high phosphorus to protein ratios.
e
10 mg/day for males and nonmenstruating females; supplementary iron should be given to all hemodialysis patients treated with an erythropoiesis-stimulating agent (ESA) to maintain
adequate serum transferrin and serum ferritin levels, aimed to achieve a target hemoglobin (Hb), except for those receiving the iron intravenously.
f
Zinc supplementation of 50 mg zinc element per day for 3–6 months should be considered in hemodialysis patients with a chronic inadequate protein/energy intake and symptoms evoking
zinc deficiency (impaired taste or smell, skin fragility, impotence, peripheral neuropathy).
0005152425.INDD 622 09-12-2022 16:10:54

Introductio 623
Low protein diet

Microbiome modulation
↓ Pro
teolyti
c
↑ Sacch
arolyti
↓ Phosphate c
Nephroprotection
H burden
↓ Acid load of food 1. Vasoconstriction of
N H
H afferent arteriole →
↑ hyperfiltration →
↓ proteinuria
2. Downregulate renal ↓ Uremic toxins
inflammation by ↑ KLF15, ↓ FGF-23
(urea, p-cresol,
↓ Metabolic 3. ↓ mesangial proliferation, ↓PTH
indoxyl sulfate)
acidosis glomerulosclerosis, and ↓ Vascular
renal fibrosis. calcification
CKD-MBD
↓ Uremic signs
Preservation of ↓ CKD progression & symptoms
nutritional status ↓ Cardiovascular
disease
-Protein stores
& body weight ↓ Anorexia
preservation
-↑ albumin ↓ Peripheral resistance to insulin
Figure 37.2 Potential benefits of low protein diet. CKD, chronic kidney disease; CKD-MBD, chronic kidney disease-mineral and bone
disorder; KLF15, Kruppel-like factor-15; FGF-23, fibroblast growth factor-23; PTH, parathyroid hormone. Source: Adapted from: Wang
et al. [60] and Chan et al. [58].
Patients transitioning to dialysis therapy with incremental rogression [62]. For dialysis patients, it may help to
p
dialysis, and especially those on prevalent dialysis, should decrease thirst, avoiding an excessive interdialytic weight
increase their protein intake to 1.2–1.4 g/kg/day to prevent gain and fluid overload, which are closely related to cardio-
PEW [60]. Protein losses into the dialysate, increased vascular events and survival [65, 66].
energy expenditure, and inflammation related to the dialy-
sis procedure are the main reasons for this higher protein Potassium Intake Recommendations
requirement [20, 59]. Patients in any CKD stage with exist- Most CKD patients do not require aggressive dietary potas-
ing or imminent PEW due to intercurrent catabolic epi- sium restriction until advanced stages. K/DOQI guidelines
sodes may require an increase in protein intake (Table 37.2). recommend no potassium restriction for patients with
For kidney transplant recipients, during the first 4 weeks CKD stages 1–2, while a limited intake from 2 to 4 g/day
after transplantation and whenever high doses of steroids (51–102 mEq/day) is recommended for patients with CKD
are used, a protein intake of at least 1.4 g/kg/day is recom- stages 3–4 [43]. However, given that potassium is associ-
mended. Due to the lack of available studies in this group ated in the general population with lower systolic BP, lower
of patients, it has been suggested those with normal kidney risk of stroke, and better bone mineral density, dietary
function should follow similar recommendations to the potassium restriction should be prescribed based on serum
general population. In transplant patients with chronic potassium levels [67–69]. Moreover, it has been suggested
graft dysfunction, it is recommended to provide the same that high potassium foods with poor nutritional value (i.e.
protein restriction diet as in patients with ND-CKD [54]. bran products, chocolate, or salt substitutes) should be
voided and other causes of hyperkaliemia, such as meta-
Sodium Intake and Fluid Balance bolic acidosis or renin-angiotensin-aldosterone system
A modest sodium restriction (<2.0–2.4 g/day of Na <5–6 g/ inhibitors, corrected before restricting healthy foods such
day of NaCl) is recommended for the management of CKD as fruits and vegetables [64]. For ND-CKD patients with
patients [44, 62]. The benefits of salt restriction may chronic hyperkalemia, obtaining a 24-hour urine for potas-
include lower blood pressure (BP) and proteinuria, sium may help to determine whether decreased kidney
improved cardiovascular outcomes, and slower CKD excretion or excessive potassium intake is the cause of
hyperkalemia. If the urine potassium content is much rapid CKD progression. A daily fiber intake of 25–30 g/day
higher than 50 mEq/24 h, reducing potassium intake or more for CKD patients may be suggested, as this amount
should be effective for controlling high potassium levels. is similar to the recommendation for the general popula-
For HD patients, the long interdialytic interval may be a tion [62]. Because of potassium restriction in advanced
particularly critical period for severe hyperkalemia appear- CKD, fruits and vegetables intake tends to be low, and die-
ance, and potassium intake restriction may be mandatory, tary fiber intake is often very low [80]. Plant-based foods,
given the association of hyperkalemia with arrhythmia and with special attention to fruits and vegetables that are low
higher mortality in this population [70]. in potassium, are an important source of dietary fiber that
should be recommended in moderation to CKD patients, as
Calcium, Phosphorus, and Vitamin D Intake long as the potassium levels allow [62].
Recommendations
Although K/DOQI guidelines suggest limiting total calcium
Definition of Protein-energy Wasting
intake to 1500 mg/day, including medication sources [43],
there have been concerns about the safety of this recom- In the uremic milieu, a state of hypercatabolism, inflam-
mendation, which may lead to a marked positive calcium mation, and undernutrition leads to the development of
balance and the potential development of cardiovascular PEW syndrome. This syndrome represents a maladaptive
calcifications [71]. Several experts recommend limiting metabolic state characterized by a decrease in both protein
total calcium intake to 800–1000 mg/day or less [64]. deposits and energy reserves that appears in patients with
Given the role of serum phosphate in the pathogenesis of acute or chronic kidney failure as part of a mechanism of
vascular calcification and the link between phosphate lev- defense of the body in response to a stress situation. PEW
els and survival in both the general and CKD populations, is a consequence of various factors that usually appear as
clinical practice guidelines suggest limiting dietary phos- kidney function decreases, including a spontaneous
phate intake in the treatment of hyperphosphatemia (grade decrease in nutrient intake and various metabolic and hor-
2D recommendation) [72], with a maximum dietary phos- monal alterations that lead to a depletion of proteins
phate intake of 800–1000 mg/day [43, 44]. Limiting pro- (especially skeletal muscle), and that does not recover
cessed foods with phosphorous-based additives and only with diet. Table 37.3 shows the diagnostic categories
encouraging home-cooked meals from fresh ingredients defined by the International Society of Renal Nutrition
(preferably plant-based foods) should be the first-line inter- and Metabolism for diagnosing PEW, including biochemi-
ventions for phosphorus restriction [64]. The next-line cal criteria, low body mass, reduced muscle mass, and low
strategy may be to exclude excessive dietary protein intake, protein or energy intakes [20]. One diagnostic criterion
as well as limiting foods with high phosphorus to protein needs to be met in at least three of the four categories for
ratios [41]. the diagnosis of PEW, so no single marker can diagnose
The optimal dietary vitamin D intake and the optimal the syndrome.
levels of 25-hydroxyvitamin D for CKD patients remain
controversial [72, 73]. Based on the recommendations Epidemiology of Protein-energy Wasting
made for the general population, vitamin D intake for CKD PEW is a common phenomenon across the spectrum of
patients is recommended at 600–800 IU/day [44, 64]. acute kidney injury (AKI) and CKD; prevalence increases
Although some interventional studies have showed some in patients on dialysis and the longer they remain on dialy-
benefits of vitamin D supplementations in proteinuria and sis [81]. However, the exact evaluation of the prevalence of
nutritional status in CKD patients, no definite conclusions PEW in the kidney population has been historically ham-
can be drawn yet from this emerging evidence [74, 75]. pered by several factors, including the lack of a standard-
Moreover, to date no clear benefit on skeletal outcomes can ized definition for malnutrition, the variability of existing
be concluded from the vitamin D administration in CKD evaluation tools, the limited sample sizes of most of the
populations [73, 75]. studies, and the different prevalence of malnutrition in the
general population reflecting economic and social dispari-
Fiber Intake Recommendations ties between different geographical regions [82]. In a recent
Although the recommended dietary fiber intake for CKD meta-analysis from contemporary studies (2000–2014)
patients is not even discussed in current nutritional including more than 50 subjects with kidney disease, the
guidelines, recent data suggest that fiber intake seems to prevalence of PEW by Subjective Global Assessment (SGA)
provide similar [76–78], if no greater [79], benefits to those or Malnutrition-Inflammation Score (MIS) was estimated
demonstrated in the general population, including to range from 60% to 82%, 11% to 54%, 28% to 54%, and 28%
improved survival, lower systemic inflammation, and less to 54% for AKI, CKD stages 3–5, dialysis, and transplant
Pathophysiolog 625
Table 37.3 The ISRNM criteria for diagnosing PEW.
Criteria Comments
Serum chemistry
Serum albumin <3.8 g/dl Nonspecific. It may be influenced by fluid status,
inflammation, proteinuria, dialysis procedure,
intercurrent illnesses, liver disease, etc.
Serum prealbumin (transthyretin) <30 g/dl Only appropriate for maintenance dialysis patients
(levels may vary according to GFR level for patients
with nondialysis CKD).
Serum cholesterol <100 mg/dl It may be influenced by lipid-lowering medication.
Body mass
BMI <23 kg/m2 Weight is influenced by fluid overload and does not
distinguish fat from muscle.
Unintentional nonedematous weight loss over
time: 5% over 3 months or 10% over 6 months
Total body fat percentage <10%
Muscle mass
Muscle wasting: reduced muscle mass of 5% over 3 months
or 10% over 6 months
Reduced mid-arm muscle circumference area (reduction Measurement must be performed by a trained
>10% in relation to 50th percentile of reference population) anthropometrist.
Creatinine appearance May be influenced by both muscle mass and meat
intake.
Dietary intake
Unintentional low DPI <0.8 g/kg/day for at least 2 months Can be assessed by dietary diaries and interviews, or
for dialysis patients or <0.6 g/kg/day for patients with CKD by calculation of nPCR or nPNA.
stages 2–5
Unintentional low DEI <25 kcal/kg/day for at least Can be assessed by dietary diaries and interviews.
2 months
BMI, body mass index; CKD, chronic kidney disease; DEI, dietary energy intake; DPI, dietary protein intake; GFR, glomerular filtration rate;
ISRNM, International Society of Renal Nutrition and Metabolism; nPCR, normalized protein catabolic rate; nPNA, normalized protein nitrogen
appearance.
Note: At least one measure from at least three of the criteria should be met in order to classify a patient as having PEW. Source: Adapted from:
Fouque D, Kalantar-Zadeh K, Kopple J, et al. A proposed nomenclature and diagnostic criteria for protein-energy wasting in acute and chronic
kidney disease. Kidney Int 2008; 73: 391–398.
patients, respectively (Figure 37.3) [83]. Another recent solutes and subsequent accumulation of toxic products
study using the four PEW diagnostic criteria showed a (including protein waste products, phosphate, sodium) and
prevalence of more than 50% of moderate or severe PEW in water. This leads to acidemia, hyper-azotemia, hyperphos-
the HD population [84]. phatemia, and fluid retention. The kidneys attempt to com-
pensate for damage through hyperfunction (hyperfiltration)
of the remaining functional nephrons, and activation of
Pathophysiology several compensative mechanisms, such as secondary
hyperparathyroidism (SHPT) or volume expansion, which
Nutritional Disorders in CKD
may be harmful for the body. As a result of all these
The kidney functions in maintaining nutrient homeostasis changes, several nutrient disorders arise as kidney func-
and the nutritional disorders that appear in CKD are tion declines, including metabolic acidosis, systemic
depicted in Figure 37.4. During the course of CKD, the kid- inflammation, and hormonal dysregulation that have been
neys progressively lose their ability to maintain nutrient attributed to the development of hypercatabolism and risk
homeostasis, with a resulting reduction in the excretion of for negative nitrogen balance [85].
Crude PEW prevalence

No data
< 20%
20–40%
41–60%
61–80%
28–54% (interquartile range) of maintenance dialysis patients
> 80%
present with Protein-Energy Wasting (PEW) worldwide.t
Figure 37.3 Prevalence of PEW among patients undergoing maintenance dialysis worldwide reported from studies published
between 2000 and 2014. Color gradation reflects PEW prevalence in all included studies from each country (weighted averages within
countries). PEW, protein-energy wasting. Source: From: Carrero et al. [83].
Kidney functions in nutrient homeostasis Nutritional disorders in CKD
1. Metabolic regulation of acid-base balance. 1. ↓ Acid excretion→Metabolic acidosis
Systemic
2. Regulation of nitrogen balance via the synthesis, inflammation ↑ Protein
filtration, and reabsorption of amino acids and the catabolism
elimination of urea, uric acid, and creatinine.
2. Negative nitrogen balance.
Anorexia. Suboptimal protein
PEW
3. Excretion of uremic toxins and low molecular weight and energy intake
proteins, including insulin, GH, and leptin.
3. Altered appetite signaling.
4. Glucose homeostasis via insulin degradation,
glucose excretion, and reabsorption. 4. Insulin resistance.
Altered glucose regulation.
5. Metabolic regulation of sodium and water balance. 5. Overhydration Gut edema
6. 1-hydroxylation of 25-hydroxycholecalciferol to 6. SHPT, CKD-MBD

1,25-dihydroxycholecalciferol. ↓ Muscle mass
Figure 37.4 Role of the kidney in maintaining nutrient homeostasis and nutritional disorders in CKD. CKD, chronic kidney disease;
CKD-MBD, chronic kidney disease-mineral and bone disorder; GH, growth hormone; PEW, protein-energy wasting, SHPT, secondary
hyperparathyroidism.
Pathophysiolog 627
Causes of Protein-energy Wasting nutrient intake. The progressive accumulation of uremic

toxins, enteric hormones, adipokines (leptin, visfatin), and
Several factors contribute to the deterioration of nutri-
inflammatory cytokines that appear as kidney disease pro-
tional status in CKD patients. PEW can occur due to inad-
gresses are the main factors that have been related to
equate nutrient intake, as well as other factors not related
decreased appetite [86, 87]. Underdialysis is an additional
to diet, such as comorbidities, appearance of intercurrent
factor that contributes to anorexia and dysgeusia, so ade-
catabolic processes, metabolic acidosis, and several endo-
quate dialysis is essential to preserve the nutritional status
crine disorders associated with CKD, such as deficit of GH
of HD patients. Several RCTs have shown that an increase
and IGF-1, hyperinsulinism, SHPT, or hyperglucagonemia
in the dose of diffusive or convective dialysis allows an
(Figure 37.5) [37]. Although clearance of uremic toxins by
increase in nutrient intake. In a randomized study, an
dialysis is able to partially reverse anorexia related to ure-
increase in the dialysis dose from Kt/V 0.8 to 1.3 increased
mic syndrome, the dialysis technique itself can contribute
the protein intake (estimated by the rate of normalized pro-
to PEW due both to the forced loss of nutrients by the
tein catabolic rate [nPCR]) at 3 months, whereas the con-
dialysate and to direct catabolic and inflammatory
trol group did not show changes in Kt/V and nPCR [88]. In
effects [27]. The consequences of PEW are reduction of
another trial comparing high-flux HD with on-line hemo-
muscle mass, increased risk of infection, and cardiovascu-
diafiltration, the administration of a convective volume of
lar disease, which results in lower quality of life and sur-
23.6 l allowed an increase in protein intake with a tendency
vival of CKD patients [37].
to increase appetite [89]. Similar results have been observed
with other more intensive dialysis regimens, such as short-
Decreased Protein and Energy Intake daily dialysis or long nocturnal dialysis [90, 91].
Uremic syndrome is associated with loss of appetite and The usual dietary restrictions in CKD patients, the rec-
various gastrointestinal disorders that lead to a decrease in ommendations for cooking foods to reduce the content of
Anorexia GFR
Chewing
difficulties Uremic
toxins
Dietary restriction Nutrient Diabetes
Polypharmacy intake CV disease..
Inflammation
Depression
Metabolic disorders
-Inhibition of GH/IGF-1 axis
Nutrient loss -Insulin resistance
during Anabolism -Hypogonadism
dialysis PEW
Anabolis o -Vit D defficiency/SHPT
mo Catabolism -Metabolic acidosis
Catabolism o
o
INFECTION SARCOPENIA CV DISEASE
Quality of live
Survival
Figure 37.5 Pathophysiology of PEW. CV, cardiovascular; GFR, glomerular filtration rate; GH, growth hormone; IGF-1, insulin-like
growth factor 1; PEW, protein-energy wasting; SHPT, secondary hyperparathyroidism. Source: Adapted from: Carrero et al. [37].
electrolytes, as well as the use of phosphate binders during to PEW syndrome in CKD patients through multiple mech-
meals alter the palatability of food and its absorption, anisms (Figure 37.6) [93–95]. In addition to inducing
which contributes to undernutrition of this population. reduced appetite, inflammation generates increased energy
Even restricting fluids to reduce interdialytic weight gain expenditure and exerts direct and indirect catabolic effects.
in dialysis patients can decrease caloric intake, since many Tumor necrosis factor alpha (TNFα) promotes the loss of
solid foods have a high water content. lean mass through the promotion of catabolic pathways,
Oral and gastrointestinal tract problems modify the such as nuclear factor kappa B (NF-kB) and the ubiquitin-
quantity and quality of the food ingested. Some of these proteasome system. Indirectly, interleukin-6 (IL-6) pro-
factors are very frequent, such as dry mucous membranes, motes muscle catabolism by contributing to the inhibition
glossitis, mucosal inflammation, and palatability prob- and resistance of anabolic pathways.
lems; these can create difficulties in swallowing and food The inhibition of the GH/IGF-1 axis (due to a lower
perception. Periodontitis and lack of teeth can make chew- expression of IGF-1 receptors in the muscle and negative
ing difficult, forcing the patient to choose foods that are hypothalamic feedback secondary to the retention of GH,
easy to chew and possibly of poor nutritional value, espe- IGF-1, and its carrier protein), the resistance to insulin,
cially fiber [92]. and the inhibition of testosterone synthesis are the main
Finally, psychological and social aspects such as depres- altered anabolic pathways in CKD patients, which gener-
sion, loneliness, or limited ability to perform activities of ates a reduction in muscle mass and strength [96, 97].
daily living also contribute to potential anorexia or diffi- Metabolic acidosis contributes to protein hypercatabolism,
culty in ensuring food availability [37]. mainly by activating the ubiquitin–proteasome system and
other muscle proteolytic pathways, decreasing hepatic syn-
Catabolic Effect of Uremia, Decreased Anabolism, thesis of albumin, and increasing insulin resistance, which
and Comorbidities of the CKD Patient in kidney failure is preferentially muscular, favoring the
Hypercatabolism is the other main component of PEW [20, onset of sarcopenic obesity [98, 99]. Finally, the typical
37]. Potential causes of its occurrence include the accumu- comorbidity of the CKD patient also contributes to hyper-
lation of uremic toxins, microinflammation, and metabolic catabolism through several factors including anemia, water
alterations associated with the uremic milieu and the overload, cardiovascular disease, or SHPT (Figure 37.7) [37].
patient’s comorbidities. Systemic inflammation contributes The coexistence of diabetes is especially relevant, given
Inflammation
Blockage of
melanocortin-4
receptors IL-6
TNFα
⊕ Ø Insulin resistance
Peripheral effects in organs ⊕ Ubiquitin– Ø

involved in nutrient intake proteasome GH/IGF-1
(mucositis, gastritis, NF -kB system axis
periodontitis…) Testosterone
synthesis
Appetite ↑ Energy Direct muscle
loss expenditure protein breakdown
Indirect muscle
protein breakdown
↑ Protein
Undernutrition
catabolism
PEW
Figure 37.6 Effects of inflammation in the development of PEW. GH, growth hormone; IL-6, interleukin-6; IGF-1, insulin-like growth
factor-1; NF-κB, nuclear factor kappa B; PEW, protein-energy wasting; TNFα, tumor necrosis factor alpha. Source: Adapted from: Lecker
SH et al. [93] and Carrero et al. [94].
Pathophysiolog 629
Sedentary
lifestyle
Anemia
Sarcopenia, inflammation,
↓ Physical activity, ↓ physical activity, frailty,
frailty, cardiovascular disease
high output heart failure
Unwillingness to eat,
anorexia, inability to
↑ Inflammatory cytokines,
obtain/prepare food,
gut edema,
inflammation
↓ physical activity,
↓ physical activity
↑ pain Depression
PEW
Overhydration
Anorexia,
inflammation,
Cachexia, inflammation, ↑ REE
glucocorticoid release, ↑ REE,
↑ sympathetic nervous system, low vitamin D status,
↑ circulating angiotensin II, glucose intolerance, Gastroparesis,
↑ insulin resistance sarcopenia, insulin resistance,
cardiovascular disease sarcopenia Infections
Cardiovascular Diabetes
disease
Secondary
hyperparathryroidism
CKD-MBD
Figure 37.7 Comorbidities linked to CKD that contribute to PEW. CKD, chronic kidney disease; CKD-MBD, chronic kidney disease-
mineral and bone disorder; PEW, protein-energy wasting; REE, resting energy expenditure. Source: Adapted from: Carrero et al. [37].
that it is associated with gastroparesis, obesity, and muscle than in HD, which can contribute to development of sarco-
loss with relative preservation of fat mass. penic obesity, and difficult gastric emptying, which may
limit the intake of food and supplements due to early
Loss of Nutrients and Catabolic Effects of Dialysis satiety [105–108].
The loss of nutrients and the inflammatory stimulus pro- The dialysis procedure produces an inflammatory stimu-
duced during dialysis contribute to the appearance of PEW. lus with cytokine activation, which increases muscle degra-
The dialysis procedure is a perfect example of the integra- dation and PEW, especially if the microbiological quality of
tion of malnutrition and catabolism and how PEW is car- the dialysate is low, if the membrane is bioincompatible, or
ried out (Figure 37.8). Amino acids, peptides, and proteins if the filters are reused. Finally, patients on HD are at greater
are lost through the dialysate in both HD and PD. Plasma risk of loss of residual renal function, especially if high
amino acid concentrations are reduced approximately 20% ultrafiltration rate is prescribed, which is associated with
on average after a single HD session, which decreases the decreased appetite and an increase in basal metabolism
availability of substrate for the synthesis of muscle pro- and inflammation, aggravating PEW [37]. Both volume and
teins. All this, along with the decrease in nutrient intake, a high ultrafiltration rate may also generate intestinal
leads to a lower availability of nutrients, which generates hypoxia by a mechanism similar to that which occurs with
undernutrition and an increase in protein catabolism with intradialytic myocardial stunning. This hypoxia increases
the breakdown of muscle proteins to compensate for these the permeability of the intestinal mucosa to bacterial lipo-
losses, which contributes to the appearance of PEW [27, saccharides, favoring its translocation into the enterocyte
37, 101–104]. and increasing the inflammatory state [109, 110].
Patients undergoing PD have additional characteristics The combination of hypermetabolism and undernutri-
inherent to the technique that modify their nutritional sta- tion may explain why standard HD does not prevent the
tus, including forced absorption of glucose, which provides appearance of PEW. Although initially HD improves the
300–600 kcal/day, greater protein losses by the dialysate appetite of the incident patients and a progressive increase
Anorexia
Inadequate Kt/V
↑Inflammation
Clean blood returned to Endotoxin back-filtration
patient Bioincompatible membranes
↑ Basal metabolic rate

Hypercatabolism
FILTER
Loss of residual
renal function
Blood
PEW
Endotoxin
from patient leakage
with waste products
Fresh Undernutrition
dialysate
Intestinal
stunning
Used dialysate
(containing toxins
and excess salts) Excessive ultrafiltration rate
Losses of amino acids, protein,

and vitamins into the dialysate
Figure 37.8 Effects of dialysis-related factors in the development of PEW. PEW, protein-energy wasting. Source: Adapted from: Laville
and Fouque [100].
in dry weight can be observed, this increase in weight is that can affect nutritional status, such as phosphate binders
due to the increase in fat deposits, while the muscle mass or the use of corticosteroids, and the comorbidities and the
decreases progressively as the patient remains in this socioeconomic status of the patient [112]. In addition, the
modality of renal replacement therapy [111]. presence of proteinuria, the residual renal function, and the
type of dialysis are factors that may affect the nutritional
status of CKD patients.
iagnosis: Assessment
D Measures that can be routinely collected under daily
of Nutritional Status practice conditions are included in the so-called screening
tools. Questions that are common to all screening tools
The nutritional assessment of CKD patients requires com- include asking about appetite and involuntary body weight
bining different dietary, anthropometric, and biochemical changes within a given time frame. Screening tools are
criteria. Due to the lack of reference values of many of these inexpensive and quick and easy to use, so they can be per-
nutritional markers for this population, it is advisable to formed by any healthcare professional with the result of
make repeated measurements of the different markers to identifying patients who are at risk for poor nutritional sta-
assess their trend over time and to be able to detect subtle tus and may require further assessment [112]. Typical
changes in nutritional status. Overhydration is another issue screening tools are the Nutrition Screening Tool, Mini-
for assessment of nutritional status, as it increases the intra- Nutrition Assessment, and Malnutrition Universal
and interindividual variability of these measures [20]. Screening Tool/Malnutrition Screening Tool. There are a
To date, there are no systematic summaries of screening few studies that have assessed their reliability in CKD
and diagnostic tests to evaluate nutritional status among patients, observing a moderate sensitivity [113, 114].
CKD patients. Overall, evidence for recommending the use of screening
tools among CKD populations is very limited.
Patient Interview
Anthropometry and Body Composition
The clinical history allows collection of key information
related to nutritional status, including: the presence of Weight and Body Mass Index
symptoms such as nausea, vomiting, or involuntary weight These measures have the advantages of simplicity and
loss, which are associated with high risk for PEW, treatments low cost, but they do not distinguish the different body
Diagnosis: Assessment of Nutritional Statu 631
c omponents (water, muscle, or fat). For HD patients, history and physical examination to obtain a score that
weight should always be assessed post-dialysis, taking into classifies the nutritional status of the patient from well-
account the possible overhydration of the patient [115]. nourished to very malnourished. It collects information on
weight changes, dietary intake, persistent gastrointestinal
Mid-arm Circumference symptoms, functional capacity, nutritional needs, and
This measure is simple to perform, correlating well with physical examination [104]. It takes 15 minutes to complete
lean mass [116]. by a health professional who has been trained to perform it.
Several studies have demonstrated the validity and reliabil-
Skinfold Thickness ity of the 7-point SGA score among CKD patients (includ-
Subcutaneous fat evaluation by skinfold thickness of tri- ing ND-CKD, HD, and PD) and its correlation with other
cipital or bicipital and subscapular folds allows estimating measures of nutritional status, including lean body mass
the fat mass in adults with CKD stages 1–5 and post- and serum albumin levels [122–125].
transplant with a high correlation with dual energy X-ray
absorptiometry (DEXA). Therefore, it is postulated as the Malnutrition-Inflammation Score
method of choice for evaluating body fat among CKD pop- Malnutrition-Inflammation Score (MIS) is a modified ver-
ulations [117, 118]. sion of the SGA, and incorporates three new objective
parameters: body mass index (BMI), serum albumin, and
Bioelectrical Impedance Analysis transferrin or iron binding capacity. It was designed in an
Bioelectrical impedance analysis (BIA) determines the attempt to reduce the subjectivity of the SGA by using a
electrical impedance or the resistance to the flow of an quantitative scoring system. Each of the referred items are
electrical current through the tissues of the body. Due to its scored from 0 to 3. A MIS score of 0–5 is considered nor-
water and electrolyte content, lean tissue is a good electri- mal, 6–17 represents mild/moderate malnutrition, and
cal conductor, while fat, bone, and skin are very poor con- 18–30 indicates severe malnutrition. MIS was designed
ductors and offer a high resistance to electric currents. specifically for the HD population [126], although it has
Using algorithms, BIA can calculate body water, body fat, also been validated in ND-CKD, PD, and kidney transplant
and fat-free mass. Advantages include its ease of use and patients [127–129].
portability, as well as its sensitivity to detect changes in the
short-term; it is highly recommended that BIA be available
Biochemistry
in all dialysis units. However, consideration must be given
to limitations due to the state of hydration of the patient; Albumin
overhydration, so frequent in CKD patients, overestimates Serum albumin concentration is the most used nutritional
lean mass and underestimates fat mass. Thus, repeated parameter in CKD patients due to its reproducibility and
measures should be taken at a constant time post-dialysis low cost, as well as its capacity to respond to nutritional
to obtain a more reliable assessment of body composition interventions [130]. Several studies have supported that
with BIA [112, 119]. serum albumin concentration is a sensitive measure of
nutritional status, being associated with other common
Handgrip Strength nutritional status markers in CKD. Although evidence sug-
This measure is a surrogate marker of lean body mass and gests that it is a strong predictor of mortality, it has a half-
is an indicator of superior strength of the body muscle. It life of 20 days, which means that its response to acute
has higher correlations with nutritional status and inflam- changes in nutrient intake is delayed. Another limitation is
matory markers, and was more predictive of mortality than its low specificity, since albumin is influenced by inflam-
muscle mass measured by DEXA. Its limitations are the mation, stress, and overhydration [131–135]. An albumin
lack of standardized protocols (with respect to body pos- level below 3.8 g/dl is suggestive of PEW [20].
ture, selection/position of the hand) and the lack of refer-
ence measurements in the CKD population [116, 120, 121]. Prealbumin
The main advantage of prealbumin (or transthyretin) is its
half-life of 2 days, which makes it more sensitive to short-
Nutritional Assessment Scores
term changes in nutritional status and/or acute response to
Subjective Global Assessment nutritional therapy than albumin. However, levels are
Subjective Global Assessment (SGA) is the most used affected by kidney function, inflammatory status, and
nutritional questionnaire in CKD patients, and combines intercurrent diseases [131]. Prealbumin levels below
subjective clinical findings and objectives of the clinical 30 mg/dl are suggestive of PEW [20].
Serum Creatinine Assessment of Food Intake

This may be an adequate surrogate measure of muscle
There are three methods to assess the food intake of pro-
mass in dialysis patients as long as residual kidney func-
teins, calories, and nutrients in CKD patients. This should
tion is minimal or absent. There are no defined cut-off
be done periodically and/or whenever PEW is suspected.
points, so longitudinal assessment is suggested for each
patient, with the advantage that it is useful for detecting
Twenty-four-hour Dietary Recalls
individual changes in the short term [130, 133]. In addi-
Through an interview, the patient is asked to recall food
tion, serum creatinine determination allows for the calcu-
intake during the previous 24 hours. The main limitation of
lation of the creatinine index [135].
this method is that it extrapolates the intake of a day as the
habitual intake of the patient, together with the high risk
Creatinine Index
of bias, especially when the patient may have cognitive
If bioimpedance testing is not available, lean mass can be
disorders or is not cooperative. The main advantages are
estimated using the kinetic model of creatinine, which is a
the speed of its execution (it can even be done by telephone)
reliable method to monitor muscle mass and protein nutri-
and the lack of need for record-keeping by the
tional status in dialysis patients. It is based on the principle
patient [137–139].
that the generation of creatinine is proportional to muscle
mass in stable patients in steady state. The creatinine index
Food Diaries and Records
is defined as the normalized creatinine production rate,
The patient (or caregiver) is required to record food intake
which is equal to the sum of the creatinine excretion rate
for 3 or 7 days, including a day of HD if this is the case and
(creatinine in the dialysate + urinary creatinine) and the
a weekend day. From this, a dietitian estimates the patient’s
rate of metabolism of creatinine in a state of metabolic
food intake from food composition tables with the help of
equilibrium. To avoid post-dialysis determination of blood
computer programs [140]. The most important limitations
creatinine and dialysate, different formulas have been
are food intake not recorded completely or accurately, and
developed to calculate the creatinine index from baseline
the impossibility of detecting seasonal variations in
creatinine levels, Kt/V, age, and sex [135].
intake [137].
Cholesterol
Although a blood cholesterol level of less than 100 mg/dl is Food Frequency Questionnaires
a criterion for PEW [3], its use as a nutritional marker is In these, a series of questions ask about the frequency and
limited given that it is negatively influenced by lipid- amount of consumption of several common foods during
lowering treatment and positively influenced by the last 6–12 months; calculation of the amount of food
proteinuria [130]. ingested is done by multiplying the frequency by the spe-
cific amount for the measure provided. With questions
C-Reactive Protein such as “How many times a week do you eat fruit?”, it may
Although not a nutritional parameter, the determination of be easier to perform than the 3- or 7-day food diary,
serum levels of C-reactive protein (CRP) allows monitoring although it is not as reliable as such records [140]. One
of the inflammatory state in the CKD patient in a reproduc- advantage is that it gives a view of food intake over a longer
ible way. Given the great variability of CRP, repeated test- period of time, which is why it is especially useful in epide-
ing is required to assess its trend in each patient [130]. miological studies. However, it is of little use at the
individual level or when analyzing small groups of
Protein Equivalent of Nitrogen Appearance patients [137].
The kinetic model of urea has been used to estimate pro-
tein intake in HD patients in a steady state [133]. The pro-
Assessment of Resting Energy Expenditure
tein equivalent of nitrogen appearance (PNA or nPCR) is
calculated based on the increase of nitrogen in the interdi- Knowledge concerning energy requirements of CKD
alysis period. Catabolic or anabolic processes may overesti- patients is imperative to providing a sufficient amount of
mate or underestimate, respectively, the measurement of energy to maintain adequate nutritional status [141].
protein intake by this method. If the patient has residual Although indirect calorimetry is the most accurate method
kidney function, the urea lost in the urine should be taken for measuring resting energy expenditure (REE), it is not
into account [130]. For ND-CKD patients, protein intake usually available in the clinical routine [142]. In the
may be estimated from urinary urea excretion with absence of indirect calorimetry, several predictive energy
Maroni’s formula [136]. equations based on simple bedside parameters have been
Treatmen 633
proposed. However, it must be noted that prediction equa- down progression in the patient with ND-CKD, alleviating
tions used in normal individuals (i.e. Harris and Benedict’s the protein hypercatabolism associated with the uremic
equation) overestimate REE in the ND-CKD population, milieu in the dialysis patient, or controlling the hyper-
although the errors may be minimized in the presence of kalemia in the patient under conservative treatment.
comorbidities [142]. Conversely, several equations have Although extensive literature has shown that high
been specifically developed for dialysis patients and vali- mortality risk in CKD patients is related in part to the
dated in this population [143]. nutritional derangements linked to kidney failure, there is
not enough evidence to address the exact role of different
dietary interventions in the prognosis of the CKD
rognosis: Consequences of PEW in
P population. Whereas undernutrition may increase the risk
Mortality, Morbidity, and Quality for PEW, overnutrition may lead to sodium and volume
of Life overload, high BP, hyperkalemia, hyperphosphatemia, and
accumulation of toxic metabolites of protein degradation.
Multiple observational studies have shown that PEW Treatment to decrease proteinuria and BP and improve
markers are independently associated with worse sur- several markers of PEW, such as serum albumin levels,
vival [39, 144–146], decreased quality of life [147], increased have demonstrated efficacy across many RCTs, while evi-
risk of infections, and complications related to cardiovas- dence to hard clinical outcomes including survival, cardio-
cular disease, with greater risk of hospitalization and vascular events, or CKD progression remains inconclusive.
healthcare costs [148, 149]. Moreover, the consequences of However, this lack of evidence does not necessarily indi-
PEW are so rapid and negative in the prognosis of CKD cate that nutritional interventions have no effect on CKD
patients that this modifies the usual and logical relation- outcomes. Moreover, given that nutritional therapy includ-
ship between risk factors and poor evolution [150]. A clear ing dietary counseling and supplements is safe and cost-
example is the association between surrogates of overnu- effective, it should be considered to improve the poor
trition (i.e. obesity or hyperlipidemia) and mortality. In prognosis of CKD patients.
dialysis patients, obesity (versus normal weight) is associ-
ated with better survival, whereas the opposite effect is Dietary Interventions for Survival, Quality of
observed in the general population. The most likely expla- Life, and Overall Health of Adults with CKD
nation is that patients undergoing dialysis usually die from
the short-term consequences of PEW and do not live long Dietary interventions are routinely recommended in CKD
enough to die of cardiovascular disease associated with patients given the potential ability to influence clinical out-
obesity. Additionally, increased fat deposits could protect comes in this population. On the basis of evidence from
the PEW patient for longer [151]. nonrandomized studies in CKD patients, and especially
Although RCTs demonstrating the causal association from randomized studies in the general population, it has
between undernutrition and hypermetabolism with adverse been suggested that healthy dietary patterns may prevent
patient outcomes in CKD patients are lacking, several epi- hard outcomes, including survival, cardiovascular events,
demiological studies have explored this issue, showing that and health-related quality of life [155].
nutritional interventions in hypoalbuminemic HD patients However, the optimal approach to nutrition in CKD
may be effective in reducing hospitalizations rates and mor- patients is not well known, and the few RCTs conducted in
tality [152–154]. All these data support the hypothesis that this population have yielded conflicting results. In 2017, a
PEW is not only a risk factor, but also a treatable etiological Cochrane review evaluated the benefits and harms of
factor in preventing mortality and morbidity, so that nutri- dietary interventions in CKD patients, including dietary
tional therapy can result in multiple improved clinical out- patterns and nutritional counseling, with all-cause mortal-
comes throughout the spectrum of CKD. ity, major adverse cardiovascular events, and health-related
quality of life being the primary outcomes [1]. The review
evaluated 17 RCTs with 1639 adult CKD patients, including
Treatment 10 studies with 1130 people with ND-CKD stages 1–5, three
studies with 341 dialysis patients, and four studies with
People with kidney disease represent an extremely hetero- 168 kidney transplant recipients (summarized in Table 37.4).
geneous group of patients in whom nutritional recommen- Eleven studies (n = 900) evaluated dietary counseling and
dations may differ greatly depending on the type of six studies evaluated dietary patterns (n = 739), including
nephropathy, its form of onset (acute or chronic), and the the Mediterranean diet (two studies, n = 355), a
therapeutic goals being pursued, as, for example slowing carbohydrate-restricted, low-iron, polyphenol-enriched diet
Table 37.4 Summary of findings: dietary interventions for overall health of adults with CKD.
Absolute number of
participants 1000 patients treated the evidence
a
Outcomes (no. of studies) for one year (95% CI) Relative effect (95% CI) (GRADE) Conclusion
Primary outcomes
Death (all causes): 371 (4) Not estimable 1.59 (0.60–4.21) Very low It is uncertain whether dietary counseling
dietary counseling ●○○○ impacts on all-cause mortality
Death (all causes): 170 (1) Not estimable 0.50 (0.22–1.12) Very low It is uncertain whether CR-LIPE diet
CR-LIPE diet ●○○○ impacts on all-cause mortality
Major adverse 62 (1) Not estimable 6.58 (0.35–122.21) Very low It is uncertain whether dietary counseling
cardiovascular events: ●○○○ impacts on major adverse cardiovascular
dietary counseling events
Health-related quality 119 (2) — The MD in SF-36 score was 11.46 higher (7.73 Low Dietary counseling may have clinically
of life: dietary higher to 15.18 higher) in the intervention groups ●●○○ important increases in the SF-36 quality of
counseling compared to control life score
Secondary outcomes
ESRD: dietary 232 (2) 0.3 fewer (1 person 0.53 (0.26–1.07) Very low It is uncertain whether dietary interventions
counseling and in every 3000 treated ●○○○ impact on risks of ESRD
CR-LIPE diet for one year avoiding
ESRD)
Worsening nutrition 230 (2) Not estimable 0.40 (0.05–3.37) Very low It is uncertain whether dietary counseling
(subjective global ●○○○ impacts on risks of worsening nutrition as
assessment): dietary measured by subjective global assessment
counseling
eGFR: dietary 219 (5) — The SMD in eGFR was 1.08 higher (0.20 higher to Very low Dietary interventions may increase eGFR
counseling, 1.97 higher) with dietary intervention compared ●○○○
Mediterranean diet, to standard care
and fruits and
vegetables
Serum creatinine: 112 (3) — The MD in creatinine was 0.83 μmol/l higher Very low It is uncertain whether dietary interventions
dietary counseling (−16.57 lower to 18.23 higher) with dietary ●○○○ impact on creatinine levels
and Mediterranean intervention compared to standard care
diet
Systolic blood 167 (3) — The MD in systolic BP was −9.26 mmHg lower Moderate Dietary interventions probably reduce
pressure: dietary (−13.48 lower to −5.04 lower) with dietary ●●●○ systolic BP
counseling and fruits intervention compared to standard care
and vegetables
0005152425.INDD 634 09-12-2022 16:11:12

Diastolic blood 95 (2) — The MD in diastolic BP was −8.95 mmHg lower Moderate Dietary counseling probably reduces
pressure: dietary (−10.69 lower to −7.21 lower) with dietary ●●●○ diastolic BP
counseling intervention compared to standard care
Energy intake: dietary 340 (4) — The SMD in energy intake was 1.54 higher Very low It is uncertain whether dietary counseling
counseling (−0.87 lower to 3.95 higher) with dietary ●○○○ impacts on energy intake
intervention compared to standard care
Energy intake: 40 (1) — The SMD in energy intake was 1.86 higher (1.11 Very low Mediterranean diet may increase energy
Mediterranean diet higher to 2.61 higher) with Mediterranean diet ●○○○ intake
compared to standard care
Energy intake: high 12 (1) — The SMD in energy intake was −0.65 lower Very low It is uncertain whether high nitrogen and
nitrogen and low (−1.82 lower to 0.53 higher) with high nitrogen ●○○○ low carbohydrate diet impacts on energy
carbohydrate diet and low carbohydrate diet compared to standard intake
care
Body weight: dietary 454 (6) — The MD in body weight was −0.44 kg lower Very low It is uncertain whether dietary interventions
counseling, fruits and (−1.46 lower to 0.58 higher) with dietary ●○○○ impact on body weight
vegetables, CR-LIPE, intervention compared to standard care
and high nitrogen and
low carbohydrate diet
BMI: dietary 119 (2) — The MD in BMI was −1.70 kg/m2 lower Very low It is uncertain whether dietary counseling
counseling (−5.23 lower to 1.82 higher) with dietary ●○○○ impacts on BMI
Arm circumference: 149 (2) — The MD in arm circumference was 0.37 higher Very low It is uncertain whether dietary counseling
dietary counseling (−0.39 lower to 1.12 higher) with dietary ●○○○ impacts on arm circumference
Serum albumin: 541 (6) — The MD for the effect of serum albumin levels Low Dietary interventions may increase albumin
dietary counseling, was 0.16 mg/dl higher (0.07 higher to 0.24 higher) ●●○○ levels
Mediterranean diet, with dietary intervention compared to control or
and CR-LIPE standard care
BMI, body mass index; BP, blood pressure; CR-LIPE, carbohydrate-restricted, low-iron-available, polyphenol-enriched; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal
disease; SF-36, short-form 36-item health survey.
Treatment estimates are drawn from a Cochrane review [1]. GRADE assessment of the certainty of the evidence [17]: High: This research provides a very good indication of the likely effect.
The likelihood that the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood that the effect will be substantially
different* is moderate. Low: This research provides some indication of the likely effect. However, the likelihood that it will be substantially different* is high. Very low: This research does not
provide a reliable indication of the likely effect. The likelihood that the effect will be substantially different* is very high. *Substantially different = a large enough difference that it might affect
decision-making.
a
Effects are reported as the relative risk (RR) and 95% confidence interval (CI) for binary outcomes and mean difference (MD) and 95% CI for continuous outcomes (standardized mean
difference [SMD] if different scales were used).
0005152425.INDD 635 09-12-2022 16:11:13

(one study, n = 191), increased fruit and vegetable intake in CKD progression. The primary results of the largest trial
(two studies, n = 181), and a high protein/low carbohydrate to date, the Modification of Diet in Renal Disease (MDRD),
diet (one study, n = 12). Studies related to dietary manage- which analyzed the effect of protein restriction and inten-
ment of specific dietary factors such as protein, sodium, or sive control of BP on the rate of loss of GFR (measured by
phosphate were excluded. According to this systematic iothalamate clearance) in 840 ND-CKD patients, did not
review, dietary interventions have uncertain effects on mor- support the hypothesis of benefit of protein restriction on
tality and cardiovascular events among CKD patients as the progression of kidney failure [168]. It is possible that
studies were not designed to examine these outcomes. In the MDRD study duration was too short to fully appreciate
very-low-quality evidence, dietary interventions had uncer- the effect of protein restriction; this and other limitations
tain effects on the development of end-stage renal disease could have allowed classifying this RCT as inconclusive
(ESRD). Conversely, dietary interventions may increase rather than negative [169]. Indeed, subsequent secondary
health-related quality of life, eGFR, and serum albumin, analyses performed by some of the MDRD investigators
and reduce several risks factors for cardiovascular events concluded that the balance of evidence was more consist-
such as BP and serum LDL cholesterol levels. Whether ent with the hypothesis of beneficial effect from the reduc-
these favorable effects on these surrogate endpoints would tion of the proteins in the diet rather than with the
improve clinical outcomes for CKD patients remains hypothesis of nonbenefit, with this benefit being additive
uncertain [1]. to that obtained by the BP control. The restriction of pro-
teins slowed the deterioration of GFR, mainly in more
advanced stages and in those cases of more rapid loss of
Nutritional Interventions to Slow
renal function. These findings would justify the recom-
Progression of CKD
mendation of an LPD in ND-CKD patients, and its safety,
Dietary Protein Restriction when accompanied by nutritional advice and systematic
According to the hyperfiltration theory, the amount of pro- monitoring, is also supported by the evidence [169, 170].
tein intake in current diets, which is generally higher than In 2007 and 2018, two Cochrane reviews summarized
our daily needs, could explain the appearance of glomeru- the evidence for protein restriction for diabetic and
lar sclerosis that accompanies aging (0.75–1.5 ml/min/year nondiabetic kidney diseases, respectively [3, 5]. A summary
from the age of 40) [156–159]. Conversely, dietary protein of the findings is shown in Table 37.6. Evidence for
restriction may protect against CKD progression by beneficial effects of protein restriction for nondiabetic
hemodynamic-mediated mechanisms, including intraglo- patients with ND-CKD stages 3–5 is restricted to effects of
merular pressure reduction, as well as by changes in VLPD on progression to ESRD. Conversely, when LPDs
cytokine expression and matrix synthesis (Figure 37.2) [62]. were compared with normal protein diets, there was little
Nearly 20 RCTs have assessed the effects of protein or no difference in the number of people who progressed to
restriction on kidney outcomes of interest, including CKD dialysis. LPDs or VLPDs probably do not influence death;
progression and the need for maintenance dialysis, pro- however, there are limited data on other adverse effects
teinuria, phosphate levels, acidemia, and BP. These studies such as weight differences and PEW. Most studies reported
have been summarized in several meta-analyses [2–5, 160– that adherence to diet was satisfactory, but quality of life
165]. Most of these have been conducted recently [2–4], was not formally assessed in any studies. The authors
which indicates there remains great interest regarding pro- encouraged conduct of new studies evaluating the adverse
tein restriction in CKD patients [62, 166, 167]. effects and the impact on quality of life of dietary protein
Several practical approaches have been described for restriction before these dietary approaches be recom-
protein restriction, with 0.6–0.8 g/kg of weight per day mended for widespread use [3].
(LPD) being most frequently recommended for ND-CKD Restricted protein intake appeared to slow progression of
patients stages 3–5. However, a greater restriction of less diabetic kidney disease, but not by much on average. Only
than 0.6 g/kg/day (VLPD) supplemented with essential one study explored all-cause mortality and ESRD as end-
amino acids or their ketoacids has also been tested. The points in type 1 diabetic patients, showing a relative risk
safety and implementation of protein restriction diets (RR) of 0.23 for patients assigned to LPD. However, pooling
should be accompanied by ensuring adequate energy of the seven RCTs of patients with type 1 diabetes resulted
intake (30–35 kcal/kg/day) and regular nutritional checks. in a nonsignificant reduction in the decline of GFR of
Overall, the balance of evidence suggests a benefit of die- 0.1 ml/min/month (95% confidence interval [CI] −0.1 to
tary protein restriction (summarized in Table 37.5). 0.3) in the LPD group. For type 2 diabetes, one trial showed
However, results of some RCTs have not consistently a small insignificant improvement in the rate of decline of
demonstrated that protein restriction would be beneficial GFR in the protein-restricted group and a second found a
Table 37.5 Meta-analyses and systematic reviews of diet protein restriction for kidney outcomes (RCTs only).
No. of
Author studies Sample Outcome measures Methods Conclusions
Fouque 6 n = 890 No. of deaths due to Difference in protein intake >0.2 g/kg/ LPDs slowed progression to ESRD.
et al. [160] ND-CKD with renal cause or need day in both groups
varying degrees of to initiate Follow-up of at least 1 year
renal failure hemodialysis
treatment
Pedrini 5 n = 1413 Calculation of RR LPD of 0.4–0.6 g/kg/day Protein restriction slowed progression of renal disease in
et al. [161] ND-CKD with of progression of Follow-up for 18–36 months diabetic and nondiabetic patients
moderate degree of renal disease
renal failure
Kasiske 13 n = 1919 Differences in rate Measurement of protein intake with Protein restriction reduced the rate of decline in eGFR,
et al. [162] ND-CKD (degree of of fall of GFR or urea excretion and nitrogen although the magnitude of effects was weak [by only
renal failure was creatinine clearance Analysis of results by intent-to-treat 0.53 ml/min/year (95% CI 0.08–0.98 ml/min/year)]
unreported) between control Measurement of disease progression The effect of dietary protein restriction was relatively
groups and groups by GFR or creatinine clearance greater among diabetic versus nondiabetic patients
treated with LPDs (5.4 ml/min/year, 95% CI 0.3–10.5 ml/min/year, P < 0.05)
Follow-up for 6–36 months
Fouque 7 n = 1494 Renal death (death In theory, difference in protein intake Reducing protein intake overall reduced renal deaths by
et al. [163] ND-CKD stages 3–5 of any cause, between higher and restricted protein about 39% (P = 0.006)
Only nondiabetic requirement to start intake groups was approximately To spare one extra renal death, 17 patients need to be
patients dialysis, or kidney 0.35 g/kg/day in almost all the studies. treated with a low protein diet for approximately 2 years
transplant) However, based on urinary collection A subgroup analysis of the “start of dialysis” event was
of protein waste products, the actual also highly significant, with an OR of 0.56 in favor of a
reduction in protein intake between restricted protein intake (P < 0.01)
groups in each study was less than
expected (between 0.2 and 0.35 g/kg/
day).
Follow-up of at least 1 year
Robertson 12 (nine n = 585 Biochemistry (GFR), Actual protein intake in the The RR of ESRD or death was 0.23 (95% CI 0.07–0.72) for
et al. [5] RCTs, three Only diabetic patients ESRD (all-cause intervention groups ranged from 0.7 to patients assigned to an LPD
before and (type 1 : 322; type mortality, nutritional 1.1 g/kg/day Pooling of the seven RCTs in patients with type 1 diabetes
after 2 : 263) with status, health-related Interventions of reduced or modified resulted in a nonsignificant reduction in the decline of
studies) nephropathy (UAER QOL, and costs were protein intake 4 months GFR of 0.1 ml/min/months (95% CI −0.1 to 0.3) in the
300 mg/day) also analyzed) LPD group
For type 2 diabetes, one trial showed a small insignificant
improvement in the rate of decline of GFR in the
protein-restricted group and a second found a similar
decline in both the intervention and control groups
A specific recommendation of the necessary protein level to
achieve this outcome was not possible
(Continued)
0005152425.INDD 637 09-12-2022 16:11:13

No. of
Author studies Sample Outcome measures Methods Conclusions
Fouque 10 (also n = 1494 Renal death (death Protein intake ( 0.8 g/kg/day) vs. Reducing protein intake overall reduced renal deaths by
et al. [165] included CKD stages 3–5 of any cause, moderate (0.6 g/kg/day) to severe about 32% (P = 0.0002)
three Only nondiabetic requirement to start protein restriction (0.3 g/kg/day) Subanalysis (seven studies) found that reduced protein
crossover patients dialysis, or kidney regardless of supplementation with intakes between 0.3 and 0.6 g/kg/day compared to higher/
studies) transplant) amino acids or keto-acids free protein intakes resulted in a significant reduction in
renal deaths (37%, P = 0.0009)
The optimal level of protein intake cannot be determined
based on this review
Rughooputh 15 n = 1965 Change in mean The assessment of protein intake was Moderate protein diet restriction slowed CKD progression
et al. [166] CKD stages 1–5 GFR based on reported urinary indices in nondiabetic and type 1 diabetic patients (−1.50 ml/
Pooling the mean values for all studies, min/1.73 m2/year, 95% CI −2.73 to 0.26, P = 0.0175), but
the achieved protein intake was 0.83 not in type 2 diabetic patients (−0.17 ml/min/1.73 m2/year,
(0.15) g/kg/day in the experimental 95% CI −1.88 to 1.55, P = 0.85)
arm and 1.07 (0.17) g/kg/day in the
control arm, based solely on urinary
indices
Follow-up of at least 1 year
Rhee et al. [2] 16 (RCTs n = 2771 Rate of progression LPD (<0.8 g/kg/day) vs. higher protein Risk of progression to ESRD was 4% lower in those who
with less CKD stages 3–5 (one to ESRD (all-cause diets (>0.8 g/kg/day) received LPD
than 30 study with 60 new mortality, nutritional VLPD (<0.4 mg/kg/day) vs. LPD LPD (<0.8 g/kg/day) was also associated with higher serum
participants ESRD patients on status, and (0.4–0.8 mg/kg/day) bicarbonate levels, lower phosphorus levels, lower azotemia,
were peritoneal dialysis bicarbonate and and a trend toward lower rates of all-cause mortality
excluded) with residual kidney phosphorus levels VLPDs (protein intake <0.4 g/kg/day) were associated
function was also were also analyzed) with greater preservation of kidney function and
included) reduction in the rate of progression to ESRD
Safety and adherence to a LPD was not inferior to a
normal protein diet, and there was no difference in the
rate of malnutrition or protein-energy wasting
0005152425.INDD 638 09-12-2022 16:11:13

Hahn 17 n = 2996 End or change in LPD (0.5–0.6 g/kg/day) compared with LPD compared with a normal protein intake did not
et al. [3] ND-CKD stages 3–5 GFR, ESRD a normal protein intake ( 0.8 g/kg/ impact on the incidence of ESRD (RR 1.05, 95% CI
(all-cause mortality day) 0.73–1.53) and on the outcome of final or change in GFR
and nutritional VLPD (0.3–0.4 g/kg/day) compared (SMD −0.18, 95% CI −0.75 to 0.38)
status were also with LPD (0.5–0.6 g/kg/day) VLPD comparted with LPD reduced the number who
analyzed) Follow-up of at least 1 year reach ESRD (RR 0.65, 95% CI 0.49–0.85) with no influence
on the final or change in GFR (SMD 0.12, 95% CI −0.27 to
0.52)
It was uncertain whether LPD or VLPD alter survival
Twelve studies reported no evidence of protein-energy
wasting in their study participants, while three studies
reported small numbers of participants in each group
with protein-energy wasting. It was uncertain whether
LPD or VLPD alter final body weight.
Most studies reported that adherence to diet was
satisfactory
Quality of life was not formally assessed in any studies
Yan et al. [4] 19 n = 2492 Kidney failure The difference in protein intake LPD reduced the risk of kidney failure (OR = 0.59, 95%
CKD stages 1–5 (two events, rate of between protein restriction and control CI 0.41–0.85) and ESRD (OR = 0.64, 95% CI 0.43–0.96),
studies with 78 and change in eGFR per groups was at least 0.2 g/kg/day but did not produce a clear beneficial effect for all-cause
60 peritoneal dialysis year, and changes in Follow-up of at least 24 weeks death events (OR = 1.17, 95% CI 0.67–2.06)
patients were also proteinuria The change in the MD for the rate of decline in the eGFR
included) (all-cause mortality, was significant (MD −1.85, P = 0.001), and for the
nutritional status, proteinuria reduction (MD −0.44, P = 0.02)
and phosphorus LPD also reduced the serum phosphorus concentration
levels were also (MD −0.37, 95% CI −0.5 to −0.24) and BMI (MD −0.61,
analyzed) 95% CI −1.05 to −0.17). However, the change in albumin
presented no significant difference between two groups
(MD 0.23, 95% CI −0.51 to 0.97).
BMI, body mass index; CI, confidence interval; CKD, chronic kidney disease, ESRD, end-stage renal disease; eGFR, estimated glomerular filtration rate; LPD, low protein diet; MD, mean
difference; ND-CKD, nondialysis chronic kidney disease; OR, odds ratio; QOL, quality of life; RCT, randomized controlled trial; RR, relative risk; SMD, standardized mean difference; UAER,
urinary albumin excretion rate; VLDP, very-low protein diet. Source: Hahn D, Hodson EM, Fouque D (2018); Robertson L, Waugh N, Robertson A.(2007); Hultcrantz M et al.(2017).
0005152425.INDD 639 09-12-2022 16:11:13

Table 37.6 Summary of findings: dietary protein restriction for adults with CKD.
Absolute number of
participants 1000 patients treated the evidence
a
Low protein diet versus normal protein diet for nondiabetic adults with CKD stages 3–4
Death (all 1680 (5) 13 fewer (27 fewer to 0.77 (0.51–1.18) Moderate LPD probably does not influence death
causes) 10 more) ●●●○
ESRD 1814 (6) 7 more (39 fewer to 1.05 (0.73–1.53) Low LPD may make little difference to the number
76 more) ●●○○ of people who progress to ESRD
End of study or 1680 (8) Not estimable The SMD for end or change in GFR was 0.18 lower Very low It is uncertain whether LPD impacts on the
change in GFR (0.75 lower to 0.38 higher) with LPD compared to ●○○○ outcome of final or change in GFR
normal protein diet
End of study 223 (2) Not estimable The MD for end body weight was 3.09 kg lower Very low It is uncertain whether LPD impacts on final
body weight (5.02 lower to 1.16 lower) with LPD compared to ●○○○ body weight
normal protein diet
Quality of life —(−) Not estimable Not estimable Absent No studies were found that evaluated quality
of life
Very low protein diet versus low or normal protein diet for nondiabetic adults with CKD stages 4–5
Death (all 681 (6) 10 more (15 fewer to 1.26 (0.62–2.54) Moderate VLPD compared with LPD probably makes
causes) 60 more) ●●●○ little or no difference to death
ESRD 1010 (10) 165 fewer (69–233 0.65 (0.49–0.85) Moderate VLPD probably reduces number of people
fewer) ●●●○ who progress to ESRD
End of study or 456 (6) Not estimable The SMD for end or change in GFR was 0.12 higher Very low It is uncertain whether VLPD compared with
change in GFR (0.27 lower to 0.52 higher) with VLPD compared to ●○○○ LPD impacts on the outcome of final or
LPD change in GFR
End of study 291 (4) Not estimable The MD for end body weight was 1.40 kg higher Very low It is uncertain whether VLPD impacts on final
body weight (3.40 lower to 6.21 higher) with VLPD compared to LPD ●○○○ body weight
Development 2373 (15) Not estimable 1.31 (0.42–4.13) Low Protein restriction may make little difference
of PEW ●●○○ to the number of people who develop PEW
of life
0005152425.INDD 640 09-12-2022 16:11:13

Low protein diet versus normal protein diet for diabetic renal disease
Death (all 84 (1) Not estimable 0.23 (0.07–0.72) Low LPD may prevent death or progression to
causes) or ●●○○ ESRD in type 1 diabetic patients
ESRD: LPD vs.
normal protein
diet in type 1
diabetic
patients
Change in 222 (7) Not estimable There was a nonsignificant improvement in GFR of Low LPD may make little difference to the change
GFR: LPD vs. 0.1 ml/min/month (95% CI −0.1 to 0.3) in the LPD ●●○○ in GFR in type 1 diabetic patients
normal protein group
diet in type 1
diabetic
patients
Change in 83 (2) Not estimable Not estimable Very low It is uncertain whether LPD impacts on
GFR: LPD vs. ●○○○ change in GFR in type 2 diabetic patients
normal protein
diet in type 2
diabetic
patients
of life
CKD, chronic kidney disease; ESRD, end-stage renal disease; GFR, glomerular filtration rate; LPD, low protein diet; VLDP, very-low protein diet; PEW, protein-energy wasting.
Treatment estimates are drawn from two Cochrane reviews [3, 5]. GRADE assessment of the certainty of the evidence [17]: High: This research provides a very good indication of the likely
effect. The likelihood that the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood that the effect will be
substantially different* is moderate. Low: This research provides some indication of the likely effect. However, the likelihood that it will be substantially different* is high. Very low: This
research does not provide a reliable indication of the likely effect. The likelihood that the effect will be substantially different* is very high. *Substantially different = a large enough difference
that it might affect decision-making.
a
Source: Hahn D, Hodson EM, Fouque D (2018); Robertson L, Waugh N, Robertson A.(2007); Hultcrantz M et al.(2017).
0005152425.INDD 641 09-12-2022 16:11:13

similar decline in both the intervention and control groups. Dietary Caloric Restriction
No data on the effects of LPDs on health-related quality of Among both the general population and ND-CKD patients,
life in this population were found. According to the an extensive body of evidence from epidemiological studies
authors, and due to the variability amongst patients, there indicates that obesity and central fat distribution are inde-
might perhaps be a 6-month therapeutic trial of protein pendent risk factors for CKD progression and the develop-
restriction in all individuals with diabetic kidney disease, ment of ESRD [175, 176]. Insulin resistance, obesity-mediated
with continuation only in those who responded best [5]. hypertension, glomerular hyperfiltration, renin-angiotensin-
aldosterone system activation, glomerular hypertrophy, and
Dietary Salt Restriction inflammation are the potential mechanisms by which
The restriction of salt in the diet reduces BP and reduces obesity may increase CKD progression [177]. This excess
proteinuria (independently of the effect on the BP) by a risk associated with obesity may be reversible by achieving
direct effect on the renin-angiotensin-aldosterone system, sustained weight loss through diet or bariatric surgery in the
potentially improving cardiovascular outcomes and CKD general population [178–180]. However, clear evidence on
progression to ESRD. However, the lack of long-term RCTs the harms and benefits of avoiding excessive caloric intake
assessing the effectiveness and safety of dietary salt restric- and reducing weight in obese people with ND-CKD is still
tion on CKD progression and survival precludes any firm lacking.
conclusions [62]. Nonrandomized evidence in obese patients with CKD has
A 2015 Cochrane review summarized the evidence for suggested that weight loss may lead to better BP control and
salt restriction from eight RCTs of 258 adults with CKD reduction of the obesity-related glomerular hyperfiltration
(including six studies with ND-CKD patients, one with and proteinuria [180, 181]. Similar data were shown in two
dialysis patients, and one with kidney transplant recipi- RCTs assessing the effect of hypocaloric diet in 17 obese
ents) that compared two or more levels of salt intake [6]. patients and 30 overweight patients, respectively, with pro-
Unfortunately, the short follow-up of the included studies teinuric nephropathies [182, 183]. Meta-analysis of both
(1–26 weeks) did not allow for establishing the effect of RCTs showed a significant overall reduction in urinary
sodium intake on primary endpoints such as cardiovascu- protein excretion of −1.65 g/24 h (95% CI −3.21 to −0.08),
lar and all-cause mortality, cardiovascular events, or CKD whereas eGFR remained stable during follow-up [8].
progression. However, low salt interventions were associ- In summary, although there is limited evidence of the
ated with significant reduction in 24-hour sodium excre- potential reversibility of the harmful hemodynamic effects,
tion (range 52–141 mmol) (eight studies, 258 patients, proteinuria, and progression of kidney disease related to
mean difference [MD] −105.86 mmol/day, 95% CI −119.20 obesity, weight reduction through diet and lifestyle
to −92.51), systolic BP (eight studies, 258 patients, MD modifications could be considered as a component of the
−8.75 mmHg, 95% CI −11.33 to −6.16), and diastolic BP therapeutic regimen of obese patients with ND-CKD [177].
(eight studies, 258 patients, MD −3.70 mmHg, 95% CI
−5.09 to −2.30), allowing a significant reduction in antihy- Fiber Intake and Probiotics
pertensive treatment dosage (two studies, 52 patients, RR Evidence is emerging for the effects of fiber intake on CKD
5.48, 95% CI 1.27–23.66). Although data regarding urinary progression and mortality in CKD stages 3–5, although
protein excretion was not able to be meta-analyzed, a sig- only a few small interventional uncontrolled studies have
nificant reduction in proteinuria in relation to a low salt addressed the effects of fiber intake in renal outcomes. In a
diet was noted in all the RCTs that reported this single-blind crossover study of 13 ND-CKD patients with
outcome [171–174]. Salt restriction significantly increased an eGFR less than 50 ml/min/1.73 m2, supplementing the
plasma renin activity and serum aldosterone, whereas one diet with fiber significantly reduced p-cresol levels and
study reported a 56% reduction in the risk of edema [171]. improved stool frequency [184]. In another uncontrolled
There was no significant change in eGFR (two studies, 68 trial, the addition of 23 g/day of fiber incorporated into the
patients, MD −1.14 ml/min/1.73 m2, 95% CI −4.38 to 2.11) usual diets of 13 ND-CKD patients was associated with a
or creatinine clearance (three studies, 85 patients, MD significant reduction in blood urea nitrogen and creatinine
−4.60 ml/min, 95% CI −11.78 to 2.57) in relation to low salt levels, increasing the eGFR from 29.6 ± 3.5 to 32.5 ± 3.6 ml/
interventions. min/1.73 m2 after 4 weeks of fiber intervention [78]. These
The effect of moderate salt restriction for lowering BP studies suggest with very-low-quality evidence that con-
and proteinuria was confirmed in a 2018 meta-analysis of suming a highly fermentable fiber, such as inulin, might
RCTs comparing the effect of low versus high salt intake, reduce uremic toxins generation, potentially through the
which involved 11 studies of 738 patients with ND-CKD inhibition of proteolytic and enhancement of saccharolytic
stages 1–4 [7]. microbiome.
Treatmen 643
Supporting this is a placebo-controlled RCT involving 3 years showed a slower progression of kidney disease in
30 patients with CKD stages 3–4 where total plasma the group of patients allocated to dietary alkali treatment
p-cresol concentration was reduced by 40% after taking a with base-producing fruits and vegetables or bicarbonate
synbiotic for 4 weeks [185]. It remains uncertain if increas- versus usual care [190].
ing fiber intake to normalize intestinal microflora could In summary, it is uncertain whether healthy dietary pat-
delay CKD progression to ESRD; even probiotics supple- terns (independent of other healthy lifestyles) make any
mentation has not shown evidence of reducing CKD pro- difference in preventing the development of ESRD among
gression, according to a recent meta-analysis involving patients with CKD, although long-term RCTs are
eight studies of 261 patients with CKD stages 3–5D (sum- warranted [1].
marized in Table 37.7). However, this therapy may reduce
the levels of p-cresol sulfate and elevate the levels of IL-6,
Treatment of PEW in CKD
thereby protecting the intestinal epithelial barrier of
patients with CKD [9]. Prevention and early detection of PEW seems to be essen-
tial for the optimal treatment of the syndrome [48, 166,
Dietary Patterns and CKD Progression 191]. Important to the prevention of PEW is periodic
Although nutritional management in patients with CKD screening for malnutrition, which must be performed at
has been focused on limiting or supplementing the each patient visit to the clinic or hospital. Through an
amounts of specific nutrients, epidemiologic and observa- adequate patient interview, a guided physical examina-
tional evidence suggests that overall dietary pattern might tion, and a basic biochemical analysis that includes blood
be more protective against CKD progression than excess or levels of albumin, urea, creatinine, and CRP, the sus-
deficiency of an isolated nutrient [186–188]. Whereas the pected presence of PEW may be determined and a more
so-called Western diet (a high saturated fat diet, rich in red detailed assessment of the nutritional status should fol-
and processed meat and sweets) has been associated with low, as well as early referral to the dietitian. Symptoms
increased risk of CKD progression and albuminuria [186], and warning signs include loss of appetite, changes in
the Mediterranean diet has been linked to slower kidney bowel habits, the appearance of edema, or involuntary
function decline and favorable effects on cardiovascular weight loss. The routine determination of albumin and
morbidity and mortality [187]. In kidney transplant recipi- CRP allows objective monitoring of the visceral protein
ents, the Dietary Approach to Stop Hypertension (DASH) deposits and the inflammatory state of patients. Through
diet has been associated with lower risk of both kidney the urea clearance (otherwise needed for controlling the
function decline and all-cause mortality [188]. A lower dialysis dose in patients on dialysis), changes in protein
load and bioavailability of phosphorus (vegetable protein), intake can be quantified. In each clinical review, other
reduction of the burden of uremic toxins and salt, better preventive measures to preserve the nutritional status
control of BP and other cardiovascular risk factors, and should be reinforced, including control of the dialysis
reduction of oxidative stress and inflammation are putative dose and metabolic acidosis, continuous nutritional coun-
mechanisms which could explain the benefits derived from seling, and supportive messaging about the benefits of
the Mediterranean diet [47]. However, in a systematic adequate physical exercise tailored to each patient [48].
review of seven cohort studies involving 15 282 patients, six Table 37.8 shows a protocol used to monitor nutritional
studies did not show an independent association between status and treat PEW in CKD patients.
healthy dietary patterns (defined as generally consistent With a frequency that usually ranges between 4 weeks
with higher intake of fruits and vegetables, fish, legumes, and 4 months, depending on the CKD stage (the more
cereals, whole grains, and fiber, and lower in red meat, salt, advanced the stage, the more frequent the nutritional
and refined sugars) and risk of ESRD [155]. assessment), and whenever PEW is suspected in CKD
There is very limited evidence from interventional stud- patients, a more detailed assessment of nutritional status
ies that adherence to healthy dietary patterns might have should be conducted. Recommendations for this assess-
effects on CKD progression. In one RCT, 40 ND-CKD ment include an SGA score to provide a more detailed col-
patients were randomly assigned to nutritional advice lection of the signs and symptoms of malnutrition, the
adapted from a Mediterranean diet or to standard care. determination of prealbumin and transferrin levels, which
After 3 months of follow-up, eGFR and creatinine and urea can reflect changes in nutritional status earlier than albu-
levels remained unchanged in both groups, although it min, performance of bioimpedance if available (otherwise,
may be argued that the intervention time was too short to the determination of the mid-arm circumference as a
assess CKD progression [189]. Conversely, another RCT marker of muscle mass), and measurement of muscle
involving 108 patients with CKD stage 3 followed for strength by hand dynamometry.
Table 37.7 Summary of findings: probiotics supplementation for adults with CKD.
Absolute number of
patients affected per Certainty of
No. of participants 1000 patients treated the evidence
a
Serum 250 (5) Not estimable The MD for serum creatinine was 0.08 mg/dl Very low It is uncertain whether probiotics
creatinine higher (0.13 lower to 0.28 higher) with ●○○○ supplementation impacts on serum creatinine
probiotics compared to placebo levels compared to placebo
Blood urea 164 (3) Not estimable The MD for blood urea was 1.38 mg/dl lower Very low It is uncertain whether probiotics
(9.26 lower to 6.15 higher) with probiotics ●○○○ supplementation impacts on blood urea levels
compared to placebo compared to placebo
p-cresol 125 (3) Not estimable The SMD for p-cresyl sulfate was 0.57 lower Low Probiotics may slightly reduce p-cresyl sulfate
sulfate (0.99 lower to 0.14 lower) with probiotics ●●○○ levels of CKD patients compared to placebo
compared to placebo
Hemoglobin 93 (3) Not estimable The MD for hemoglobin was 0.21 g/dl higher Low Probiotics may slightly improve hemoglobin
(0.48 lower to 0.91 higher) with probiotics ●●○○ levels of CKD patients compared to placebo
compared to placebo
Interleukin-6 134 (3) Not estimable The SMD for interleukin-6 was 0.37 higher Very low It is uncertain whether probiotics
(0.03 higher to 0.72 higher) with probiotics ●○○○ supplementation impacts on interleukin-6 levels
compared to placebo compared to placebo
C-reactive 106 (3) Not estimable The MD for C-reactive protein was 0.64 mg/dl Very low It is uncertain whether probiotics
protein higher (3.81 lower to 5.09 higher) with ●○○○ supplementation impacts on C-reactive protein
probiotics compared to placebo levels compared to placebo
CKD, chronic kidney disease.

Treatment estimates are drawn from a Cochrane review [9]. GRADE assessment of the certainty of the evidence [17]: High: This research provides a very good indication of the likely effect.
decision-making.
a
Effects are reported as mean difference (MD) and 95% confidence interval (CI) for continuous outcomes (standardized mean difference [SMD] if different scales were used).
Source: Jia L, Jia Q, Yang J, Jia R, Zhang H. (2018); Hultcrantz M, Rind D, Akl EA, Treweek S, Mustafa RA, Iorio A, et al. (2017).
0005152425.INDD 644 09-12-2022 16:11:13

Treatmen 645
Table 37.8 Suggested strategies for early detection and treatment of PEW in CKD patients.
Warning signs Potential interventions
Periodic nutritional screening (in

each clinical visit)
Actual body weight and ideal Unintentional weight loss >5% of If PEW is suspected, a more detailed nutritional
weight ideal body weight or estimated dry assessment and an early referral to the renal dietitian are
weight or weight <85% of ideal mandatory
weight
Dietary intake Loss of appetite Other preventive measures:
Serum albumin <3.8 g/dl 1) Dietary assessments to evaluate whether the intake
provided with the diet is adequate in protein and
C-reactive protein >10 mg/dl caloric contents
Urea clearance to estimate the <1.2 g/kg/day in dialysis 2) Correction of metabolic acidosis
protein intake (and calculate the <0.6 g/kg/day in nondialysis 3) Achievement of adequate dialysis dose
dialysis dose) dependent CKD 4) Good oral hygiene
Creatinine levels to calculate the <20 g/kg/day or significant 5) Treatment of digestive symptoms as needed (nausea/
creatinine index in dialysis reduction vomiting)
patientsa 6) Adequate physical exercise (as tolerated)
7) Management of comorbidities (diabetes, chronic heart
failure, etc.)
8) In-center meals (dialysis)
Detailed assessment (at least two
to four times/year or whenever
PEW is suspected)
Subjective global assessment Score of B or C ●● An etiopathogenic approach:
Prealbumin <30 mg/dl or significative 1) Treatment of inflammatory causes: intercurrent
reduction infections, nonfunctioning renal graft, use of central
Transferrin <200 mg/dl or significative venous catheters
reduction 2) Uremic symptoms: optimize dialysis if needed
Mid-arm circumference Significative reduction 3) Pill burden reduction (i.e. phosphate binders):
Bioelectrical impedance Loss of muscle mass or fat mass consider dose reduction or temporary discontinua-
tion of treatment
Hand dynamometry Loss of strength
4) Treatment for gastroparesis in diabetic patients
Dietary intake assessment Protein intake <1.2 g/kg/day in
●● Dietary counseling
dialysis or <0.6 g/kg/day in
nondialysis dependent CKD ●● Oral nutritional supplements
Energy intake <30–35 kcal/day ●● Tube feeding or percutaneous endoscopic gastrostomy
●● Intradialysis parenteral nutrition (for hemodialysis
patients)
●● Intraperitoneal nutrition (for peritoneal dialysis
patients)
●● Total parenteral nutrition
●● Adjunct therapies: appetite stimulants; anabolic
hormones (androgens, growth hormone); anti-
inflammatory interventions (omega-3)
CKD, chronic kidney disease; PEW, protein-energy wasting.
a
The equation to estimate creatinine index (CI) was developed based on age, gender, predialysis serum creatinine concentrations, and spKt/V urea with
the simplified equation from Cannaud et al. [135] [16.21 + 1.12 × (1 si man; 0 si female) − 0.06 × age (years) − 0.08 × spKt/V + 0.009 × Crpre(μmol/l)]. To
evaluate CI using plasma creatinine concentration in mg/dl, divide the creatinine concentration term by 88.4.
Source: Adapted from: Ikizler TA, Cano NJ, Franch H, et al. Prevention and treatment of protein-energy wasting in chronic kidney disease
patients: a consensus statement by the International Society of Renal Nutrition and Metabolism. Kidney Int. 2013 Dec;84(6):1096–107.
Dietary Counseling restricted to effects on nutritional parameters, as RCTs did

In patients for whom preventive approaches to PEW have not have sufficient statistical power to allow meta-analysis
been insufficient, the first therapeutic step should be inter- of patient-centered outcomes, such as survival, quality of
vention by a nutritionist, who will carry out a dietary life, or cardiovascular events. Given the short follow-up
assessment to evaluate whether the dietary intake is ade- and the small sample sizes of the studies, the quality of the
quate in protein and caloric contents. This intervention evidence according to the meta-analysis was considered
may support initiating therapy with oral nutritional sup- very low.
plements (ONS) [48]. Only one RCT has assessed the effects of protein supple-
The randomized evidence to determine whether provid- mentation on mortality in the dialysis population as a
ing individual nutrition counseling improves nutritional primary endpoint [196]. However, it compared only oral
status in ND-CKD patients is limited to one RCT involving protein supplement with intradialytic parenteral nutrition
56 patients with CKD stages 4–5 [192, 193]. Patients were (IDPN) and did not include a usual care arm, probably
randomly allocated to fortnightly, individualized coun- because of the ethical implications of withholding nutri-
seling on a prescription of 0.75 g/kg/day of protein and tional therapy to malnourished patients. Body weight,
145 kJ/kg/day of energy versus written education material serum albumin, and PNA increased in both groups.
for 3 months. During the 12-week follow-up, the interven- Similarly, there were no differences between the two
tion group had less decrease in body cell mass, greater groups regarding mortality.
increase in energy intake, and greater improvement in SGA The randomized evidence for ONS in PD patients is very
score compared with the control group. limited and shows a higher rate of noncompliance and
Two RCTs have demonstrated dietary counseling may intolerance compared to HD patients [39, 197]. In an open-
improve several nutritional parameters for patients with label RCT, 28 PD patients were allocated to an oral egg
CKD stage 5D [194, 195]. In the largest trial to date, 180 HD albumin-based supplement versus conventional nutritional
patients with baseline albumin levels less than 3.7 g/dl counseling. During a 6-month follow-up, the treatment
were randomized to identify and intervene on specific group had significantly improved serum albumin, calorie
modifiable nutritional barriers (i.e. low nutritional knowl- and protein intake, and nPNA [198]. In a 4-month RCT
edge, decreased appetite, or gastrointestinal symptoms) involving 100 PD patients with malnutrition, calcium
versus usual care; results showed significant improve- caseinate versus usual care resulted in a significant increase
ments in protein and energy intakes and albumin levels in serum albumin levels [199]. Evidence for ONS in renal
regardless of levels of inflammatory markers [193]. transplant and ND-CKD patients is absent, although ONS
According to another, small RCT, intensive nutritional might help to prevent PEW in patients on LPD [200].
counseling might be of greater benefit than nutritional Altogether, very-low-quality evidence suggests that
supplements alone in the management of PEW in HD short-term oral energy or protein/amino acid supplements
patients [195]. may improve nutritional status in dialysis patients, without
influence on serum phosphorus or potassium levels.
Oral Nutritional Supplements In those patients with oral intolerance to ONS, nasogas-
ONS can offer an additional source of protein (0.3–0.4 g/ tric tubes, percutaneous endoscopic gastrostomy, or jeju-
kg/day) and energy (7–10 kcal/kg/day) for achieving the nostomy probes can be used. These nutritional treatments
recommended targets (Table 37.2) in CKD patients with a are usually reserved for patients with severe, postsurgical,
decreased spontaneous nutrient intake. ONS are usually critical anorexia or with problems swallowing due to neu-
recommended to be administered two or three times a day, rological alterations or head and neck surgery. Randomized
always after meals (“as a dessert”) to avoid losing appetite, evidence for enteral tube feeding in the CKD population is
or during the HD session [39]. The nutritional response to absent [201].
therapy is directly correlated with severity of PEW at
baseline [48]. Intradialytic Parenteral Nutrition in HD
The efficacy of oral supplements to improve nutritional Although the enteral route is the initial treatment of choice
status in dialysis patients has been tested in a number of for nutritional supplementation, parenteral provision of
small RCTs, which have been meta-analyzed in a recent nutrients during the HD session can be considered for
systematic review of 15 studies with 589 subjects patients who continue to be malnourished despite ONS or
(Table 37.9) [10]. Compared with placebo or routine care, who are unable to tolerate oral supplementation (e.g. those
provisions of ONS resulted in a significant increase in BMI with severe gastroparesis). Although the optimal duration
and albumin levels, with no effects on phosphorus or of treatment is unknown, a trial of 3–6 months is recom-
potassium levels. Evidence for beneficial effects of ONS is mended to assess the response to IDPN. IDPN is usually
Table 37.9 Summary of findings: oral nutritional supplements for adults with CKD on dialysis.
Absolute number of
participants (no. 1000 patients treated the evidence
Outcomes of studies) for one year (95% CI) Relative effect (95% CI)a (GRADE) Conclusion
Primary outcomes
Death (all —(−) — No studies Absent No studies were found that evaluated the
causes) impact of ONS on survival
Cardiovascular —(−) — No studies Absent No studies were found that evaluated the
events impact of ONS on the appearance of
Infectious —(−) — No studies Absent No studies were found that evaluated the
events impact of ONS on the appearance of infectious
events
Health-related —(−) — No studies Absent No studies were found that evaluated the
quality of life impact of ONS on health-related quality of life
Secondary outcomes
Body mass 376 (9) Not estimable The MD for the effect of BMI was 0.40 kg/m2 higher Low ONS may increase body mass index
index (0.10 higher to 0.71 higher) with ONS compared to ●●○○
control or standard care
Serum 507 (14) Not estimable The MD for the effect of serum albumin levels was Moderate ONS probably increases serum albumin levels
albumin 1.58 mg/dl higher (0.52 higher to 2.63 higher) with ●●●○
ONS compared to control or standard care
BMI, body mass index; CKD, chronic kidney disease; ONS, oral nutritional supplements.
Treatment estimates are drawn from a Cochrane review [10]. GRADE assessment of the certainty of the evidence: [17] High: This research provides a very good indication of the likely effect.
decision-making.
a
Source: Liu PJ, Ma F, Wang QY, He SL. (2018); Hultcrantz M, Rind D, Akl EA, Treweek S, Mustafa RA, Iorio A, et al.(2017).
0005152425.INDD 647 09-12-2022 16:11:13

administered only 3 days per week for 4 hours, thus it is not RCTs to suggest that outcomes are improved by nonspe-
appropriate for severely malnourished patients, in whom cific inflammation-reducing modalities.
total parenteral nutrition may be necessary to provide suf- The treatment of the different comorbidities could also
ficient proteins and calories to treat PEW [48]. help to correct PEW (i.e. adequate metabolic control for
A 2010 systematic review of RCTs compared IDPN with diabetes or antidepressants for mood disorders). Although
usual diet, placebo, or ONS; it included three studies with the quality of evidence for the ethological approach to
220 HD patients [202]. Unfortunately, there were insuffi- PEW is very low or absent, a few small randomized and
cient data to undertake a meta-analysis of whether IDPN quasi-randomized controlled trials have shown the bene-
improves survival, quality of life, or nutritional status. In fits of increasing both solute diffusion and convection on
the largest trial in the review (and to date), 186 malnour- the nutritional status of dialysis patients [88, 89]. Thus,
ished HD patients received ONS with or without 1 year of improving dialysis adequacy, with interventions aimed at
IDPN [196]. The study showed an improvement in enhancing solute clearance, might impact positively on
nutritional markers in both groups, indicating that the PEW, although quality evidence is lacking [48].
addition of IDPN to ONS did not provide additional benefit Other additional therapeutic measures to correct PEW,
over ONS alone; no differences in survival were noted. A most of them considered as experimental, may include the
2017 RCT compared IDPN given three times weekly versus use of appetite stimulants such as megestrol acetate, anti-
dietary counseling in 107 HD patients with PEW. IDPN for inflammatory drugs, and exercise and anabolic agents [48].
16 weeks resulted in a significant increase in mean serum
prealbumin, although without changes in SGA score or
quality of life [203]. Megestrol Acetate
The conclusion then is that although IDPN might be used
in highly selected patients with positive results, it is uncer- Evidence for the use of megestrol acetate as orexigenic is
tain whether IDPN improves quality of life, decreases mor- restricted to dialysis patients. A systematic review of three
tality, or prevents hospitalizations. This is in agreement with RCTs and six observational studies assessing the efficacy
the results of a recently published systematic review that and safety of megestrol acetate was published in 2016 [12].
included both interventional and observational studies [11].
Box 37.1 Nonspecific Inflammation-reducing
Intraperitoneal Infusion of Amino Acids in PD Interventions for Protein-energy Wasting in Patients
Although use of amino acids in the dialysate may be an ade- with Chronic Kidney Disease
quate nutritional intervention for malnourished PD patients
who are unable to tolerate ONS, evidence in this regard is Screen and treat occult infections:
limited to one small RCT [204]. Improvements in overall pro- ●● Urinary tract infections
tein balance (protein synthesis minus protein breakdown) ●● Diabetic foot ulcers
was observed with the use of a dialysate that contained ●● Gut flora dysbiosis
amino acids and glucose versus glucose as control dialysate ●● Tuberculosis
in a crossover study of eight patients over 14 days. Whether Assess for poor dental hygiene and periodontal disease
this dialysis procedure may improve the nutritional status in Routinely assess for volume overload
malnourished PD patients remains uncertain. Screen and treat rheumatologic conditions:
●● Vasculitis
Non-nutritional Interventions for PEW ●● Systemic lupus relapses
●● Rheumatoid arthritis
In addition to correcting any nutritional deficiencies, treat-
Screen for pericarditis
ment of malnourished CKD patients should always include
Embolize or remove nonfunctioning renal allografts
an ethological approach to identify the causes involved in
Optimize vascular access
PEW [48]. Inflammation may be a cause of anorexia and
Replace central venous catheters with arteriovenous fistula
increased energy expenditure, which increases the risk of
●●
●● Remove nonfunctioning arteriovenous grafts

developing PEW. This inflammation can be caused by a
Optimize hemodialysis prescription
wide range of potentially treatable factors (Box 37.1).
Although periodic determination of CRP to help assess the ●● Replace standard dialysate fluid with ultrapure dialysate
Replace bioincompatible membranes with biocompatible
presence of underlying and treatable inflammatory pro- ●●
membranes
cesses would be beneficial [43, 48], this recommendation is
●● Replace standard hemodialysis with on-line
based on epidemiological evidence linking inflammation hemodiafiltration
to poor outcome in the CKD population [130]; there are no
Conclusion 649
All three RCTs were limited by small sample sizes (range Pentoxifylline, anti-TNF agents, and recombinant IL-1
9–22 subjects), short duration (range 8–24 weeks), a high receptor antagonists have also been tested in the treatment
degree of bias, and absence of clinical outcomes such as of inflammation in patients with CKD.
quality of life or hospitalizations. Appetite, serum albumin,
and body weight significantly increased in some but not all
the studies. Altogether, it is uncertain whether megestrol Anabolic Agents
acetate impacts on PEW. In addition, use of megestrol
acetate was associated with significant adverse events, Anabolic agents including recombinant human GH and
which were especially noted in the six observational stud- androgenic steroids may also be effective in CKD patients
ies, and there were no data regarding the safety of using with PEW. In a 4-week follow-up RCT involving 17 mal-
megestrol acetate beyond 24 weeks [12]. nourished dialysis patients, treatment with recombinant
human GH, whose use is approved for pediatric patients
with CKD and short stature, significantly increased weight
Anti-inflammatory Agents gain and IGF-1 and transferrin levels, suggesting that
short-term GH administration might be associated with an
In addition to treating inflammation directly due to a spe- anabolic reaction in malnourished dialysis patients [208].
cific identified source (e.g. an ongoing chronic infection), However, the long-term nutritional benefits of GH have
the use of anti-inflammatory agents in malnourished CKD been not demonstrated [209]. In an RCT involving 29
patients has been tested in several RCTs. A meta-analysis patients undergoing dialysis, treatment with nandrolone
of RCTs assessing the anti-inflammatory effects of fish oil versus placebo for 6 months resulted in a significant
containing omega-3 polyunsaturated fatty acids versus pla- increase in lean body mass associated with functional
cebo or no treatment was published in 2016 [13]. It involved improvement, including a decrease in the time to complete
13 studies with 916 patients with CKD stage 5D on HD, and the walking and stair-climbing test [210]. Despite this posi-
showed that fish oil significantly reduced some CRP and tive effect on protein stores, the long-term efficacy and risk
high-sensitivity CRP but not all inflammatory markers. In for adverse effects of androgenic anabolic steroids remains
addition, fish oil reduced the risk of arteriovenous (AV)- unclear, thus limiting their role in medical practice.
graft events and cardiovascular events and alleviated Overall, evidence on the safety and effectiveness of anti-
depression symptoms, although no effect on albumin levels inflammatory drugs and anabolic agents in CKD patients is
was noted. In 2014, two meta-analyses examined the effects very low, hindered by small studies, with limited follow-up,
of l-carnitine supplementation in HD patients, with mixed and focused primarily on HD patients. Large-scale RCTs
results [14, 15]. Whereas in one meta-analysis (49 RCTs, are needed to confirm these anti-inflammatory effects.
1734 participants) there was a significant decrease in CRP Lastly, although an aerobic and resistance exercise pro-
levels noted [14], the other review (25 studies, 1172 partici- gram in well-nourished CKD patients may improve
pants) indicated this therapy did not result in a significant physical capacity and quality of life [211], no data are
reduction in CRP levels [15]. In a small uncontrolled trial, available regarding the anabolic effect of exercise in CKD
daily supplementation with Brazil nut (the richest known patients with PEW.
food source of selenium) for 3 months resulted in a signifi-
cant decrease in cytokines and other inflammatory and
oxidative markers [205]. Conclusions
Besides the anti-inflammatory effects related to some
nutrient supplementation, some drugs may also exert the Attention to nutritional status is critical as it may impact on
same effects in CKD populations. In 2012, a meta-analysis important clinical outcomes in patients with CKD, including
summarized the evidence for the effect of statin therapy on CKD progression, quality of life, morbidity, and mortality.
inflammatory and nutritional markers in dialysis patients. Although some interventional studies have shown that
Nine RCTs involving 3098 patients were analyzed, with the nutritional interventions may have reno-protective effects
resulting report of a significant decrease in CRP and high- and improve surrogate markers of survival, high-quality
sensitivity CRP levels but no effect on serum albumin evidence is still insufficient. However, this lack of evidence
levels [16]. In a small RCT of 108 HD patients, sevelamer does not necessarily indicate that nutritional interventions
hydrochloride versus calcium acetate lowered high- have no effect on CKD outcomes. Given that nutritional
sensitivity CRP after a 1-year follow-up [206]. An uncon- therapy, including dietary counseling and supplements, is
trolled trial of 24 HD patients noted a reduction in CRP and safe and cost-effective, it should be considered to improve
IL-6 after a 3-month intervention with N-acetylcysteine [207]. the poor prognosis of CKD patients.
R
eferences
1 Palmer, S.C., Maggo, J.K., Campbell, K.L. et al. (2017). 13 He, L., Li, M.S., Lin, M. et al. (2016). Effect of fish oil
Dietary interventions for adults with chronic kidney supplement in maintenance hemodialysis patients: a
disease. Cochrane Database Syst. Rev. 4: CD011998. systematic review and meta-analysis of published
2 Rhee, C.M., Ahmadi, S.F., Kovesdy, C.P., and Kalantar- randomized controlled trials. Eur. J. Clin. Pharmacol. 72
Zadeh, K. (2018). Low-protein diet for conservative (2): 129–139.
management of chronic kidney disease: a systematic 14 Chen, Y., Abbate, M., Tang, L. et al. (2014). L-Carnitine
review and meta-analysis of controlled trials. J. Cachexia supplementation for adults with end-stage kidney
Sarcopenia Muscle 9 (2): 235–245. disease requiring maintenance hemodialysis: a
3 Hahn, D., Hodson, E.M., and Fouque, D. (2018). systematic review and meta-analysis. Am. J. Clin.
Low protein diets for non-diabetic adults with chronic Nutr. 99 (2): 408–422.
kidney disease. Cochrane Database Syst. Rev. 10: 15 Yang, S.K., Xiao, L., Song, P.A. et al. (2014). Effect of
CD001892. L-carnitine therapy on patients in maintenance
4 Yan, B., Su, X., Xu, B. et al. (2018). Effect of diet protein hemodialysis: a systematic review and meta-analysis. J.
restriction on progression of chronic kidney disease: a Nephrol. 27 (3): 317–329.
systematic review and meta-analysis. PLoS One 13 (11): 16 Deng, J., Wu, Q., Liao, Y. et al. (2012). Effect of statins on
e0206134. chronic inflammation and nutrition status in renal
5 Robertson, L., Waugh, N., and Robertson, A. (2007). dialysis patients: a systematic review and meta-analysis.
Protein restriction for diabetic renal disease. Cochrane Nephrology (Carlton) 17 (6): 545–551.
Database Syst. Rev. (4): CD002181. 17 Hultcrantz, M., Rind, D., Akl, E.A. et al. (2017). The
6 McMahon, E.J., Campbell, K.L., Bauer, J.D., and Mudge, GRADE Working Group clarifies the construct of
D.W. (2015). Altered dietary salt intake for people with certainty of evidence. J. Clin. Epidemiol. 87: 4–13.
chronic kidney disease. Cochrane Database Syst. Rev. (2): 18 Ikizler, T.A. and Hakim, R.M. (1996). Nutrition in
CD010070. end-stage renal disease. Kidney Int. 50: 343–357.
7 Garofalo, C., Borrelli, S., Provenzano, M. et al. (2018). 19 Kopple, J.D. (1994). Effect of nutrition on morbidity and
Dietary salt restriction in chronic kidney disease: a mortality in maintenance dialysis patients. Am. J. Kidney
meta-analysis of randomized clinical trials. Nutrients 10 Dis. 24: 1002–1009.
(6): E732. 20 Fouque, D., Kalantar-Zadeh, K., Kopple, J. et al. (2008).
8 Navaneethan, S.D., Yehnert, H., Moustarah, F. et al. A proposed nomenclature and diagnostic criteria for
(2009). Weight loss interventions in chronic kidney protein-energy wasting in acute and chronic kidney
disease: a systematic review and meta-analysis. Clin. J. disease. Kidney Int. 73: 391–398.
Am. Soc. Nephrol. 4 (10): 1565–1574. 21 Kalantar-Zadeh, K., Block, G., McAllister, C.J. et al.
9 Jia, L., Jia, Q., Yang, J. et al. (2018). Efficacy of probiotics (2004). Appetite and inflammation, nutrition, anemia,
supplementation on chronic kidney disease: a systematic and clinical outcome in hemodialysis patients. Am. J.
review and meta-analysis. Kidney Blood Press. Res. 43 (5): Clin. Nutr. 80: 299–307.
1623–1635. 22 Lodebo, B.T., Shah, A., and Kopple, J.D. (2018). Is it
10 Liu, P.J., Ma, F., Wang, Q.Y., and He, S.L. (2018). The important to prevent and treat protein-energy wasting in
effects of oral nutritional supplements in patients with chronic kidney disease and chronic dialysis patients? J.
maintenance dialysis therapy: a systematic review and Ren. Nutr. 28 (6): 369–379.
meta-analysis of randomized clinical trials. PLoS One 13 23 Kalantar-Zadeh, K., Mehrotra, R., Fouque, D. et al.
(9): e0203706. (2004). Metabolic acidosis and malnutrition–
11 Anderson, J., Peterson, K., Bourne, D., and Boundy, E. inflammation complex syndrome in chronic renal failure.
(2019 S1051-2276(18)30278-4. doi: https://doi. Semin. Dial. 17: 445–465.
org/10.1053/j.jrn.2018.11.009). Effectiveness of 24 Bologa, R.M., Levine, D.M., Parker, T.S. et al. (1998).
intradialytic parenteral nutrition in treating protein- Interleukin-6 predicts hypoalbuminemia,
energy wasting in hemodialysis: a rapid systematic hypocholesterolemia, and mortality in hemodialysis
review. J. Ren. Nutr. [Epub ahead of print]. patients. Am. J. Kidney Dis. 32: 107–114.
12 Wazny, L.D., Nadurak, S., Orsulak, C. et al. (2016). The 25 Grodstein, G.P., Blumenkrantz, M.J., and Kopple, J.D.
efficacy and safety of megestrol acetate in protein-energy (1980). Nutritional and metabolic response to
wasting due to chronic kidney disease: a systematic catabolic stress in uremia. Am. J. Clin. Nutr. 33:
review. J. Ren. Nutr. 26 (3): 168–176. 1411–1416.
Reference 651
6 Wolfson, M., Jones, M.R., and Kopple, J.D. (1982). Amino

2 40 Hyun, Y.Y., Lee, K.B., Han, S.H. et al. (2017). Nutritional
acid losses during hemodialysis with infusion of amino status in adults with predialysis chronic kidney disease:
acids and glucose. Kidney Int. 21: 500–506. KNOW-CKD study. J. Korean Med. Sci. 32 (2): 257–263.
27 Ikizler, T.A., Flakoll, P.J., Parker, R.A. et al. (1994). Amino 41 Rysz, J., Franczyk, B., Cialkowska-Rysz, A., and Gluba-
acid and albumin losses during hemodialysis. Kidney Int. Brzozka, A. (2017). The effect of diet on the survival of
46: 830–837. patients with chronic kidney disease. Nutrients 9, 495
28 Chazot, C., Shahmir, E., Matias, B. et al. (1997). Dialytic (5): 1–17.
nutrition: provision of aminoacids in dialysate during 42 Kelly, J.T. and Carrero, J.J. (2017). Dietary sources of
hemodialysis. Kidney Int. 52: 1663–1670. protein and chronic kidney disease progression: the proof
29 Bailey, J.L., Wang, X., England, B.K. et al. (1996). The may be in the pattern. J. Ren. Nutr. 27 (4): 221–224.
acidosis of chronic renal failure activates muscle 43 (2000). Clinical practice guidelines for nutrition in
proteolysis in rats by augmenting transcription of genes chronic renal failure. K/DOQI, National Kidney
encoding proteins of the ATP-dependent ubiquitin– Foundation. Am. J. Kidney Dis. 35: S1–S140.
proteasome pathway. J. Clin. Invest. 97: 1447–1453. 44 Fouque, D., Vennegoor, M., ter Wee, P. et al. (2007). EBPG
30 Pickering, W.P., Price, S.R., Bircher, G. et al. (2002). guideline on nutrition. Nephrol. Dial. Transplant. 22
Nutrition in CAPD: serum bicarbonate and the ubiquitin– (Suppl 2): ii45–ii87.
proteasome system in muscle. Kidney Int. 61: 1286–1292. 45 Piccoli, G.B., Moio, M.R., Fois, A. et al. (2017). The diet
31 Mak, R.H. (1996). Insulin resistance but IGF-I sensitivity and haemodialysis dyad: three eras, four open questions
in chronic renal failure. Am. J. Physiol. 271: F114–F119. and four paradoxes. A narrative review, towards a
32 Moyle, G.J., Daar, E.S., Gertner, J.M. et al. (2004). Growth personalized, patient-centered approach. Nutrients 9 (4):
hormone improves lean body mass, physical E372. https://doi.org/10.3390/nu9040372.
performance, and quality of life in subjects with HIV- 46 Biruete, A., Jeong, J.H., Barnes, J., and Wilund, K.R.
associated weight loss or wasting on highly active (2017). Modified nutritional recommendations to
antiretroviral therapy. J. Acquir. Immune Defic. Syndr. 35: improve dietary patterns and outcomes in hemodialysis
367–375. patients. J. Ren. Nutr. 27: 62–70.
33 Fouque, D., Peng, S.C., Shamir, E. et al. (2000). 47 Chauveau, P., Aparicio, M., Bellizzi, V. et al. (2018).
Recombinant human insulin-like growth factor-1 induces Mediterranean diet as the diet of choice for patients with
an anabolic response in malnourished CAPD patients. chronic kidney disease. Nephrol. Dial. Transplant. 33 (5):
Kidney Int. 57: 646–654. 725–735.
34 Sherwin, R.S., Bastl, C., Finkelstein, F.O. et al. (1976). 48 Ikizler, T.A., Cano, N.J., Franch, H. et al. (2013).
Influence of uremia and hemodialysis on the turnover Prevention and treatment of protein energy wasting in
and metabolic effects of glucagon. J. Clin. Invest. 57: chronic kidney disease patients: a consensus statement
722–731. by the International Society of Renal Nutrition and
35 Kopple, J.D., Cianciaruso, B., and Massry, S.G. (1980). Metabolism. Kidney Int. 84 (6): 1096–1107. https://doi.
Does parathyroid hormone cause protein wasting? org/10.1038/ki.2013.147.
Contrib. Nephrol. 20: 138–148. 49 Cano, N., Fiaccadori, E., Tesinsky, P. et al. (2006). ESPEN
36 Kaplan, A.A., Halley, S.E., Lapkin, R.A. et al. (1995). guidelines on enteral nutrition: adult renal failure. Clin.
Dialysate protein losses with bleach processed Nutr. 25: 295–310.
polysulphone dialyzers. Kidney Int. 47: 573–578. 50 Cano, N.J.M., Aparicio, M., Brunori, G. et al. (2009).
37 Carrero, J.J., Stenvinkel, P., Cuppari, L. et al. (2013). ESPEN guidelines on parenteral nutrition: adult renal
Etiology of the protein-energy wasting syndrome in failure. Clin. Nutr. 28: 401–414.
chronic kidney disease: a consensus statement from the 51 Wright, M. and Jones, C. (2010). Nutrition in CKD. 5e.
International Society of Renal Nutrition and Metabolism UK Renal Association. https://ukkidney.org/sites/renal.
(ISRNM). J. Ren. Nutr. 23: 77–90. org/files/nutrition-in-ckd-5th-edition-1.pdf.
38 Johansen, K.L., Painter, P.L., Sakkas, G.K. et al. (2006). 52 Brown, R.O., Compher, C., and American Society for
Effects of resistance exercise training and nandrolone Parenteral and Enteral Nutrition Board of Directors
decanoate on body composition and muscle function (2010). ASPEN clinical guidelines: nutrition support in
among patients who receive hemodialysis: a randomized, adult acute and chronic renal failure. J. Parenter. Enteral.
controlled trial. J. Am. Soc. Nephrol. 17: 2307–2314. Nutr. 34: 366–377.
39 Kalantar-Zadeh, K., Cano, N.J., Budde, K. et al. (2011). 53 Ruperto López, M., Barril Cuadrado, G., and Lorenzo,
Diets and enteral supplements for improving outcomes in S.V. (2008). Nutrition guidelines for advanced chronic
chronic kidney disease. Nat. Rev. Nephrol. 7: 369–384. kidney disease (ACKD). Nefrologia 28 (Suppl 3): 79–86.
54 Chadban, S., Chan, M., Fry, K. et al. (2010). The CARI 67 Hajjar, I.M., Grim, C.E., George, V., and Kotchen, T.A.
guidelines. Protein requirement in adult kidney transplant (2001). Impact of diet on blood pressure and age-related
recipients. Nephrology (Carlton) (Suppl 1): S68–S71. changes in blood pressure in the US population: analysis
55 Ikizler, T.A., Burrowes, J.D., Byham-Gray, L.D. et al. of NHANES III. Arch. Intern. Med. 161: 589.
(2020). KDOQI Clinical Practice Guideline for Nutrition 68 Appel, L.J., Moore, T.J., Obarzanek, E. et al. (1997). A
in CKD: 2020 Update. Am. J. Kidney Dis. 76 (3 Suppl 1): clinical trial of the effects of dietary patterns on blood
S1–S107. https://doi.org/10.1053/j.ajkd.2020.05.006 pressure. DASH Collaborative Research Group. N. Engl. J.
Erratum in: Am. J. Kidney Dis. 2021 Feb;77(2):308. Med. 336: 1117.
PMID: 32829751. 69 Zhu, K., Devine, A., and Prince, R.L. (2009). The effects
56 Kovesdy, C.P., Kopple, J.D., and Kalantar-Zadeh, K. of high potassium consumption on bone mineral density
(2013). Management of protein-energy wasting in in a prospective cohort study of elderly postmenopausal
non-dialysis-dependent chronic kidney disease: women. Osteoporos. Int. 20: 335.
reconciling low protein intake with nutritional therapy. 70 Yusuf, A.A., Hu, Y., Singh, B. et al. (2016). Serum
Am. J. Clin. Nutr. 97 (6): 1163–1177. potassium levels and mortality in hemodialysis patients:
57 Carrero, J.J. and Cozzolino, M. (2014). Nutritional a retrospective cohort study. Am. J. Nephrol. 44 (3):
therapy, phosphate control and renal protection. Nephron. 179–186.
Clin. Pract. 126 (1): 1–7. 71 Spiegel, D.M. and Brady, K. (2012). Calcium balance in
58 Chan, M., Kelly, J., and Tapsell, L. (2017). Dietary normal individuals and in patients with chronic kidney
modeling of foods for advanced CKD based on general disease on low-and high-calcium diets. Kidney Int. 81
healthy eating guidelines: what should be on the plate? (11): 1116–1122.
Am. J. Kidney Dis. 69 (3): 436–450. 72 Kidney Disease: Improving Global Outcomes (KDIGO)
59 Tennankore, K.K. and Bargman, J.M. (2013). Nutrition CKD-MBD Update Work Group (2017). KDIGO 2017
and the kidney: recommendations for peritoneal dialysis. clinical practice guideline update for the diagnosis,
Adv. Chronic Kidney Dis. 20 (2): 190–201. evaluation, prevention, and treatment of chronic kidney
60 Wang, A.Y., Kalantar-Zadeh, K., Fouque, D. et al. (2018). disease-mineral and bone disorder (CKD-MBD). Kidney
Precision medicine for nutritional management in Int. Suppl. 7: 1–59.
end-stage kidney disease and transition to dialysis. Semin. 73 Molina, P., Carrero, J.J., Bover, J. et al. (2017).
Nephrol. 38 (4): 383–396. Vitamin D, a modulator of musculoskeletal health in
61 Fouque, D. and Aparicio, M. (2007). Eleven reasons to chronic kidney disease. J. Cachexia Sarcopenia Muscle 8
control the protein intake of patients with chronic kidney (5): 686–701.
disease. Nat. Clin. Pract. Nephrol. 3 (7): 383–392. 74 Molina, P., Górriz, J.L., Molina, M.D. et al. (2014). The
62 Kalantar-Zadeh, K. and Fouque, D. (2017). Nutritional effect of cholecalciferol for lowering albuminuria in
management of chronic kidney disease. N. Engl. J. Med. chronic kidney disease: a prospective controlled study.
377 (18): 1765–1776. Nephrol. Dial. Transplant. 29 (1): 97–109.
63 KDIGO (2013). 2012 Clinical practice guideline for the 75 Taskapan, H., Baysal, O., Karahan, D. et al. (2011).
evaluation and management of chronic kidney disease. Vitamin D and muscle strength, functional ability and
Kidney Int. Suppl. 3: 5. balance in peritoneal dialysis patients with vitamin D
64 Stevens, P.E., Levin, A., and Kidney Disease: Improving deficiency. Clin. Nephrol. 76: 110–116.
Global Outcomes Chronic Kidney Disease Guideline 76 Evenepoel, P. and Meijers, B.K. (2012). Dietary fiber and
Development Work Group Members (2013). Evaluation protein: nutritional therapy in chronic kidney disease and
and management of chronic kidney disease: synopsis of beyond. Kidney Int. 81 (3): 227–229.
the kidney disease: improving global outcomes 2012 77 Krishnamurthy, V.M., Wei, G., Baird, B.C. et al. (2012).
clinical practice guideline. Ann. Intern. Med. 158: High dietary fiber intake is associated with decreased
825–830. inflammation and all-cause mortality in patients with
65 Kalantar-Zadeh, K., Regidor, D.L., Kovesdy, C.P. et al. chronic kidney disease. Kidney Int. 81: 300.
(2009). Fluid retention is associated with cardiovascular 78 Salmean, Y.A., Segal, M.S., Langkamp-Henken, B. et al.
mortality in patients undergoing long-term hemodialysis. (2013). Foods with added fiber lower serum creatinine
Circulation 119 (5): 671–679. levels in patients with chronic kidney disease. J. Ren.
66 Ng, J.K., Kwan, B.C., Chow, K.M. et al. (2018). Nutr. 23 (2): e29–e32.
Asymptomatic fluid overload predicts survival and 79 Xu, H., Huang, X., Risérus, U. et al. (2014). Dietary fiber,
cardiovascular event in incident Chinese peritoneal kidney function, inflammation, and mortality risk. Clin. J.
dialysis patients. PLoS One 13 (8): e0202203. Am. Soc. Nephrol. 9 (12): 2104–2110.
Reference 653
0 Kalantar-Zadeh, K., Kopple, J., Deepak, S. et al. (2002).

8 93 Lecker, S.H., Goldberg, A.L., and Mitch, W.E. (2006).
Food intake characteristics of hemodialysis patients as Protein degradation by the ubiquitin-proteasome
obtained by food frequency questionnaire. J. Ren. Nutr. pathway in normal and disease states. J. Am. Soc.
12: 17–31. Nephrol. 17 (7): 1807–1819.
81 Gracia-Iguacel, C., González-Parra, E., Pérez-Gómez, 94 Carrero, J.J., Park, S.H., Axelsson, J. et al. (2009).
M.V. et al. (2013). Prevalence of protein-energy wasting Cytokines, atherogenesis, and hypercatabolism in
syndrome and its association with mortality in chronic kidney disease: a dreadful triad. Semin. Dial. 22
haemodialysis patients in a centre in Spain. Nefrologia 33 (4): 381–386.
(4): 495–505. 95 Carrero, J.J. and Stenvinkel, P. (2010). Inflammation in
82 Gracia-Iguacel, C., González-Parra, E., Barril-Cuadrado, end-stage renal disease–what have we learned in 10
G. et al. (2014). Defining protein-energy wasting years? Semin. Dial. 23 (5): 498–509.
syndrome in chronic kidney disease: prevalence and 96 Feldt-Rasmussen, B. and El Nahas, M. (2009). Potential
clinical implications. Nefrologia 34 (4): 507–519. role of growth factors with particular focus on growth
83 Carrero, J.J., Thomas, F., Nagy, K. et al. (2018). Global hormone and insulin-like growth factor-1 in the
prevalence of protein-energy wasting in kidney disease: a management of chronic kidney disease. Semin. Nephrol.
meta-analysis of contemporary observational studies 29 (1): 50–58.
from the International Society of Renal Nutrition and 97 Carrero, J.J. (2014). Testosterone deficiency at the
Metabolism. J. Ren. Nutr. 28 (6): 380–392. crossroads of cardiometabolic complications in CKD.
84 Moreau-Gaudry, X., Jean, G., Genet, L. et al. (2014). A Am. J. Kidney Dis. 64 (3): 322–325.
simple protein-energy wasting score predicts survival in 98 Hung, A.M. and Ikizler, T.A. (2011). Factors
maintenance hemodialysis patients. J. Ren. Nutr. 24 (6): determining insulin resistance in chronic
395–400. hemodialysis patients. Contrib. Nephrol. 171:
85 Zha, Y. and Qian, Q. (2017). Protein nutrition and 127–134.
malnutrition in CKD and ESRD. Nutrients 9 (3): 1–19. 99 Kraut, J.A. and Kurtz, I. (2005). Metabolic acidosis of
86 Carrero, J.J., Witasp, A., Stenvinkel, P. et al. (2010). CKD: diagnosis, clinical characteristics, and treatment.
Visfatin is increased in chronic kidney disease patients Am. J. Kidney Dis. 45 (6): 978–993.
with poor appetite and correlates negatively with fasting 100 Laville, M. and Fouque, D. (2000). Nutritional aspects in
serum amino acids and triglyceride levels. Nephrol. Dial. hemodialysis. Kidney Int. Suppl. 76: S133–S139.
Transplant. 25 (3): 901–906. 101 Ikizler, T.A., Pupim, L.B., Brouillette, J.R. et al. (2002).
87 Lopes, A.A., Elder, S.J., Ginsberg, N. et al. (2007). Lack of Hemodialysis stimulates muscle and whole body protein
appetite in haemodialysis patients–associations with loss and alters substrate oxidation. Am. J. Physiol.
patient characteristics, indicators of nutritional status Endocrinol. Metab. 282: E107.
and outcomes in the international DOPPS. Nephrol. Dial. 102 Ikizler, T.A. (2005). Effects of hemodialysis on protein
Transplant. 22 (12): 3538–3546. metabolism. J. Ren. Nutr. 15: 39.
88 Lindsay, R.M., Spanner, E., Heidenheim, R.P. et al. (1992). 103 Raj, D.S., Oladipo, A., and Lim, V.S. (2006).
Which comes first, Kt/V or PCR–chicken or egg? Kidney Amino acid and protein kinetics in renal failure: an
Int. Suppl. 38: S32–S36. integrated approach. Semin. Nephrol. 26 (2):
89 Molina, P., Vizcaíno, B., Molina, M.D. et al. (2018). The 158–166.
effect of high-volume online haemodiafiltration on 104 Lim, V.S., Ikizler, T.A., Raj, D.S., and Flanigan, M.J.
nutritional status and body composition: the ProtEin Stores (2005). Does hemodialysis increase protein breakdown?
prEservaTion (PESET) study. Nephrol. Dial. Transplant. Dissociation between whole-body amino acid turnover
https://doi.org/10.1093/ndt/gfx342. [Epub ahead of print]. and regional muscle kinetics. J. Am. Soc. Nephrol. 16 (4):
90 Miller, B.W., Himmele, R., Sawin, D.A. et al. (2018). 862–868.
Choosing home hemodialysis: a critical review of patient 105 Mehrotra, R., Duong, U., Jiwakanon, S. et al. (2011).
outcomes. Blood Purif. 45 (1–3): 224–229. https://doi. Serum albumin as a predictor of mortality in peritoneal
org/10.1159/000485159. [Epub ahead of print]. dialysis: comparisons with hemodialysis. Am. J. Kidney
91 Sikkes, M.E., Kooistra, M.P., and Weijs, P.J. (2009). Dis. 58 (3): 418–428.
Improved nutrition after conversion to nocturnal home 106 Blumenkrantz, M.J., Kopple, J.D., Moran, J.K., and
hemodialysis. J. Ren. Nutr. 19: 494–499. Coburn, J.W. (1982). Metabolic balance studies and
92 Chen, L.P., Chiang, C.K., Chan, C.P. et al. (2006). Does dietary protein requirements in patients undergoing
periodontitis reflect inflammation and malnutrition status continuous ambulatory peritoneal dialysis. Kidney Int.
in hemodialysis patients? Am. J. Kidney Dis. 47 (5): 815–822. 21 (6): 849–861.
107 Westra, W.M., Kopple, J.D., Krediet, R.T. et al. (2007). 120 Wang, A.Y., Sea, M.M., Ho, Z.S. et al. (2005). Evaluation
Dietary protein requirements and dialysate protein of handgrip strength as a nutritional marker and
losses in chronic peritoneal dialysis patients. Perit. Dial. prognostic indicator in peritoneal dialysis patients. Am.
Int. 27 (2): 192–195. J. Clin. Nutr. 81 (1): 79–86.
108 Van, V., Schoonjans, R.S., Struijk, D.G. et al. (2002). 121 Chang, Y.T., Wu, H.L., Guo, H.R. et al. (2011). Handgrip
Influence of dialysate on gastric emptying time in strength is an independent predictor of renal outcomes
peritoneal dialysis patients. Perit. Dial. Int. 22 (1): 32–38. in patients with chronic kidney diseases. Nephrol. Dial.
109 McIntyre, C.W., Harrison, L.E., Eldehni, M.T. et al. Transplant. 26 (11): 3588–3595.
(2011). Circulating endotoxemia: a novel factor in 122 Steiber, A.L., Kalantar-Zadeh, K., Secker, D. et al. (2004).
systemic inflammation and cardiovascular disease in Subjective global assessment in chronic kidney disease:
chronic kidney disease. Clin. J. Am. Soc. Nephrol. 6 (1): a review. J. Ren. Nutr. 14 (4): 191–200.
133–141. 123 Cuppari, L., Meireles, M.S., Ramos, C.I., and Kamimura,
110 Lemesch, S., Ribitsch, W., Schilcher, G. et al. (2016). M.A. (2014). Subjective global assessment for the
Mode of renal replacement therapy determines diagnosis of protein-energy wasting in nondialysis-
endotoxemia and neutrophil dysfunction in chronic dependent chronic kidney disease patients. J. Ren. Nutr.
kidney disease. Sci. Rep. 6: 34534. 24 (6): 385–389.
111 Marcelli, D., Brand, K., Ponce, P. et al. (2016). 124 Vero, L.M., Byham-Gray, L., Parrott, J.S., and Steiber,
Longitudinal changes in body composition in patients A.L. (2013). Use of the subjective global assessment to
after initiation of hemodialysis therapy: results from an predict health-related quality of life in chronic kidney
international cohort. J. Ren. Nutr. 26 (2): 72–80. disease stage 5 patients on maintenance hemodialysis. J.
112 Carrero, J.J., Johansen, K.L., Lindholm, B. et al. (2016). Ren. Nutr. 23 (2): 141–147.
Screening for muscle wasting and dysfunction in 125 Enia, G., Sicuso, C., Alati, G., and Zoccali, C. (1993).
patients with chronic kidney disease. Kidney Int. 90 (1): Subjective global assessment of nutrition in dialysis
53–66. patients. Nephrol. Dial. Transplant. 8 (10): 1094–1098.
113 Afsar, B., Sezer, S., Arat, Z. et al. (2006). Reliability of 126 Kalantar-Zadeh, K., Kopple, J.D., Block, G., and
mini nutritional assessment in hemodialysis compared Humphreys, M.H. (2001). A malnutrition-inflammation
with subjective global assessment. J. Ren. Nutr. 16 (3): score is correlated with morbidity and mortality in
277–282. maintenance hemodialysis patients. Am. J. Kidney Dis.
114 Lawson, C.S., Campbell, K.L., Dimakopoulos, I., and 38 (6): 1251–1263.
Dockrell, M.E. (2012). Assessing the validity and 127 Amparo, F.C., Kamimura, M.A., Molnar, M.Z. et al.
reliability of the MUST and MST nutrition screening (2015). Diagnostic validation and prognostic
tools in renal inpatients. J. Ren. Nutr. 22 (5): 499–506. significance of the malnutrition-inflammation score in
115 Chumlea, W.C. (2004). Anthropometric and body nondialyzed chronic kidney disease patients. Nephrol.
composition assessment in dialysis patients. Semin. Dial. Transplant. 30 (5): 821–828.
Dial. 17 (6): 466–470. 128 Afşar, B., Sezer, S., Ozdemir, F.N. et al. (2006).
116 Isoyama, N., Qureshi, A.R., Avesani, C.M. et al. (2014). Malnutrition-inflammation score is a useful tool in
Comparative associations of muscle mass and muscle peritoneal dialysis patients. Perit. Dial. Int. 26 (6):
strength with mortality in dialysis patients. Clin. J. Am. 705–711.
Soc. Nephrol. 9 (10): 1720–1728. 129 Molnar, M.Z., Keszei, A., and Czira, M.E. (2010).
117 Bross, R., Chandramohan, G., Kovesdy, C.P. et al. (2010). Evaluation of the malnutrition-inflammation score in
Comparing body composition assessment tests in kidney transplant recipients. Am. J. Kidney Dis. 56 (1):
long-term hemodialysis patients. Am. J. Kidney Dis. 55 102–111.
(5): 885–896. 130 Carrero, J.J., Chen, J., Kovesdy, C.P., and Kalantar-
118 Avesani, C.M., Draibe, S.A., Kamimura, M.A. et al. Zadeh, K. (2014). Critical appraisal of biomarkers of
(2004). Assessment of body composition by dual energy dietary intake and nutritional status in patients
X-ray absorptiometry, skinfold thickness and creatinine undergoing dialysis. Semin. Dial. 27 (6): 586–589.
kinetics in chronic kidney disease patients. Nephrol. 131 Ikizler, T.A. (2014). Using and interpreting serum
Dial. Transplant. 19 (9): 2289–2295. albumin and prealbumin as nutritional markers in
119 Chertow, G.M., Lowrie, E.G., Wilmore, D.W. et al. patients on chronic dialysis. Semin. Dial. 27 (6):
(1995). Nutritional assessment with bioelectrical 590–592.
impedance analysis in maintenance hemodialysis 132 Kaysen, G.A., Dubin, J.A., Müller, H.G. et al. (2004).
patients. J. Am. Soc. Nephrol. 6 (1): 75–81. Inflammation and reduced albumin synthesis associated
Reference 655
with stable decline in serum albumin in hemodialysis 147 Rambod, M., Bross, R., Zitterkoph, J. et al. (2009).
patients. Kidney Int. 65 (4): 1408–1415. Association of malnutrition-inflammation score with
133 Kopple, J.D. (2001). National kidney foundation K/DOQI quality of life and mortality in hemodialysis patients: a
clinical practice guidelines for nutrition in chronic renal 5-year prospective cohort study. Am. J. Kidney Dis. 53
failure. Am. J. Kidney Dis. 37 (1 Suppl 2): S66–S70. (2): 298–309.
134 de Roij van Zuijdewijn, C.L., ter Wee, P.M., 148 Kaysen, G.A., Muller, H.G., Young, B.S. et al. (2004). The
Chapdelaine, I. et al. (2015). A comparison of influence of patient-and facility-specific factors on
8 nutrition-related tests to predict mortality in nutritional status and survival in hemodialysis. J. Ren.
hemodialysis patients. J. Ren. Nutr. 25 (5): 412–419. Nutr. 14 (2): 72–81.
135 Canaud, B., Granger Vallée, A., Molinari, N. et al. 149 Dalrymple, L.S., Johansen, K.L., Chertow, G.M. et al.
(2014). Creatinine index as a surrogate of lean body (2010). Infection-related hospitalizations in older
mass derived from urea Kt/V, pre-dialysis serum levels patients with ESRD. Am. J. Kidney Dis. 56 (3): 522–530.
and anthropometric characteristics of haemodialysis 150 Kalantar-Zadeh, K., Kovesdy, C.P., Derose, S.F. et al.
patients. PLoS One 9: e93286. (2007). Racial and survival paradoxes in chronic kidney
136 Maroni, B.J., Steinman, T.I., and Mitch, W.E. (1985). A disease. Nat. Clin. Pract. Nephrol. 3 (9): 493–506.
method for estimating nitrogen intake of patients with 151 Kalantar-Zadeh, K., Rhee, C.M., Chou, J. et al. (2017).
chronic renal failure. Kidney Int. 27 (1): 58–65. The obesity paradox in kidney disease: how to reconcile it
137 Bross, R., Noori, N., Kovesdy, C.P. et al. (2010). Dietary with obesity management. Kidney Int. Rep. 2 (2): 271–281.
assessment of individuals with chronic kidney disease. 152 Lacson, E. Jr., Wang, W., Zebrowski, B. et al. (2012).
Semin. Dial. 23 (4): 359–364. Outcomes associated with intradialytic oral nutritional
138 Singhal, S., Goyle, A., and Gupta, R. (1998). Quantitative supplements in patients undergoing maintenance
food frequency questionnaire and assessment of dietary hemodialysis: a quality improvement report. Am. J.
intake. Natl. Med. J. India 11 (6): 268–275. Kidney Dis. 60 (4): 591–600.
139 Posner, B.M., Martin-Munley, S.S., Smigelski, C. et al. 153 Cheu, C., Pearson, J., Dahlerus, C. et al. (2013).
(1992). Comparison of techniques for estimating Association between oral nutritional supplementation
nutrient intake: the Framingham study. Epidemiology 3 and clinical outcomes among patients with ESRD. Clin.
(2): 171–177. J. Am. Soc. Nephrol. 8 (1): 100–107.
140 Bingham, S.A., Gill, C., Welch, A. et al. (1994). 154 Benner, D., Brunelli, S.M., Brosch, B. et al. (2018).
Comparison of dietary assessment methods in Effects of oral nutritional supplements on mortality,
nutritional epidemiology: weighed records v. 24 h missed dialysis treatments, and nutritional markers in
recalls, food-frequency questionnaires and estimated- hemodialysis patients. J. Ren. Nutr. 28 (3): 191–196.
diet records. Br. J. Nutr. 72 (4): 619–643. 155 Kelly, J.T., Palmer, S.C., Wai, S.N. et al. (2017). Healthy
141 Cuppari, L. and Avesani, C.M. (2004). Energy dietary patterns and risk of mortality and ESRD in CKD:
requirements in patients with chronic kidney disease. J. a meta-analysis of cohort studies. Clin. J. Am. Soc.
Ren. Nutr. 14 (3): 121–126. Nephrol. 12 (2): 272–279.
142 Kamimura, M.A., Avesani, C.M., Bazanelli, A.P. et al. 156 Meyer, T.W., IchicKawa, I., Zatz, R., and Brenner, B.M.
(2011). Are prediction equations reliable for estimating (1983). The renal hemodynamic response to amino acid
resting energy expenditure in chronic kidney disease infusion in the rat. Trans. Assoc. Am. Physicians 96: 76–83.
patients? Nephrol. Dial. Transplant. 26 (2): 544–550. 157 Castellino, P., Coda, B., and De Fronzo, R.A. (1985). The
143 Vilar, E., Machado, A., Garrett, A. et al. (2014). Disease- effect of intravenous amino acid infusion on renal
specific predictive formulas for energy expenditure in hemodynamics in man (Abstract). Kidney Int. 27: 243.
the dialysis population. J. Ren. Nutr. 24 (4): 243–251. 158 King, A.J. and Levey, A.S. (1993). Dietary protein and
144 de Mutsert, R., Grootendorst, D.C., Boeschoten, E.W. renal function. J. Am. Soc. Nephrol. 3: 1723–1737.
et al. (2009). Subjective global assessment of nutritional 159 Brenner, B.M. (1985). Nephron adaptation to renal injury
status is strongly associated with mortality in chronic or ablation. Am. J. Physiol. 249 (3 Pt 2): F324–F337.
dialysis patients. Am. J. Clin. Nutr. 89 (3): 787–793. 160 Fouque, D., Laville, M., Boissel, J.P. et al. (1992).
145 Dong, J., Li, Y., Xu, Y., and Xu, R. (2011). Daily protein Controlled low protein diets in chronic renal
intake and survival in patients on peritoneal dialysis. insufficiency: meta-analysis. BMJ 304: 216.
Nephrol. Dial. Transplant. 26 (11): 3715–3721. 161 Pedrini, M.T., Levey, A.S., Lau, J. et al. (1996). The effect
146 Reese, P.P., Cappola, A.R., Shults, J. et al. (2013). of dietary protein restriction on the progression of
Physical performance and frailty in chronic kidney diabetic and nondiabetic renal diseases: a meta-analysis.
disease. Am. J. Nephrol. 38 (4): 307–315. Ann. Intern. Med. 124: 627.
62 Kasiske, B.L., Lakatua, J.D., Ma, J.Z., and Louis, T.A.

1 174 Vogt, L., Waanders, F., Boomsma, F. et al. (2008). Effects
(1998). A meta-analysis of the effects of dietary protein of dietary sodium and hydrochlorothiazide on the
restriction on the rate of decline in renal function. Am. antiproteinuric efficacy of losartan. J. Am. Soc. Nephrol.
J. Kidney Dis. 31: 954. 19 (5): 999–1007.
163 Fouque, D., Wang, P., Laville, M., and Boissel, J.P. 175 Wang, Y., Chen, X., Song, Y. et al. (2008). Association
(2000). Low protein diets delay end-stage renal disease between obesity and kidney disease: a systematic review
in non-diabetic adults with chronic renal failure. and meta-analysis. Kidney Int. 73 (1): 19–33.
Nephrol. Dial. Transplant. 15: 1986. 176 Evans, P.D., McIntyre, N.J., Fluck, R.J. et al. (2012).
164 Fouque, D. and Laville, M. (2006). Low protein diets for Anthropomorphic measurements that include central fat
chronic kidney disease in non diabetic adults. Cochrane distribution are more closely related with key risk factors
Database Syst. Rev. (2): CD001892. than BMI in CKD stage 3. PLoS One 7 (4): e34699.
165 Rughooputh, M.S., Zeng, R., and Yao, Y. (2015). Protein 177 Eknoyan, G. (2011). Obesity and chronic kidney disease.
diet restriction slows chronic kidney disease progression Nefrologia 31 (4): 397–403.
in non-diabetic and in type 1 diabetic patients, but not 178 Look AHEAD Research Group (2014). Effect of a
in type 2 diabetic patients: a meta-analysis of long-term behavioural weight loss intervention on
randomized controlled trials using glomerular filtration nephropathy in overweight or obese adults with type 2
rate as a surrogate. PLoS One 10 (12): e0145505. diabetes: a secondary analysis of the Look AHEAD
166 Ash, S., Campbell, K.L., Bogard, J., and Millichamp, A. randomised clinical trial. Lancet Diabetes Endocrinol. 2
(2014). Nutrition prescription to achieve positive (10): 801–809.
outcomes in chronic kidney disease: a systematic 179 Chang, A.R., Grams, M.E., and Navaneethan, S.D.
review. Nutrients 6 (1): 416–451. (2017). Bariatric surgery and kidney-related outcomes.
167 Zarazaga, A., García-De-Lorenzo, L., García-Luna, P.P. Kidney Int. Rep. 2 (2): 261–270.
et al. (2001). Nutritional support in chronic renal 180 Chagnac, A., Weinstein, T., Herman, M. et al. (2003). The
failure: systematic review. Clin. Nutr. 20 (4): 291–299. effects of weight loss on renal function in patients with
168 Klahr, S., Levey, A.S., Beck, G.J. et al. (1994). The effects severe obesity. J. Am. Soc. Nephrol. 14 (6): 1480–1486.
of dietary protein restriction and blood-pressure control 181 MacLaughlin, H.L., Sarafidis, P.A., Greenwood, S.A.
on the progression of chronic renal disease. et al. (2012). Compliance with a structured weight loss
Modification of Diet in Renal Disease Study Group. N. program is associated with reduced systolic blood
Engl. J. Med. 330 (13): 877–884. pressure in obese patients with chronic kidney disease.
169 Levey, A.S., Greene, T., Beck, G.J. et al. (1999). Dietary Am. J. Hypertens 25 (9): 1024–1029.
protein restriction and the progression of chronic renal 182 Praga, M., Hernández, E., Andrés, A. et al. (1995). Effects of
disease: what have all of the results of the MDRD study body-weight loss and captopril treatment on proteinuria
shown? Modification of Diet in Renal Disease Study associated with obesity. Nephron 70 (1): 35–41.
Group. J. Am. Soc. Nephrol. 10: 2426–2439. 183 Morales, E., Valero, M.A., León, M. et al. (2003).
170 Levey, A.S., Adler, S., Caggiula, A.W. et al. (1996). Beneficial effects of weight loss in overweight patients
Effects of dietary protein restriction on the progression with chronic proteinuric nephropathies. Am. J. Kidney
of advanced renal disease in the Modification of Diet in Dis. 41 (2): 319–327.
Renal Disease Study. Am. J. Kidney Dis. 27: 652–663. 184 Salmean, Y.A., Segal, M.S., Palii, S.P., and Dahl, W.J. (2015).
171 Slagman, M.C., Waanders, F., Hemmelder, M.H. et al. Fiber supplementation lowers plasma p-cresol in chronic
(2011). Moderate dietary sodium restriction added to kidney disease patients. J. Ren. Nutr. 25 (3): 316–320.
angiotensin converting enzyme inhibition compared 185 Guida, B., Germanò, R., Trio, R. et al. (2014). Effect of
with dual blockade in lowering proteinuria and blood short-term synbiotic treatment on plasma p-cresol levels in
pressure: randomised controlled trial. BMJ 343: d4366. patients with chronic renal failure: a randomized clinical
172 Konishi, Y., Okada, N., Okamura, M. et al. (2001). Sodium trial. Nutr. Metab. Cardiovasc. Dis. 24 (9): 1043–1049.
sensitivity of blood pressure appearing before hypertension 186 Lin, J., Fung, T.T., Hu, F.B., and Curhan, G.C. (2011).
and related to histological damage in immunoglobulin a Association of dietary patterns with albuminuria and
nephropathy. Hypertension 38 (1): 81–85. kidney function decline in older white women: a
173 de Brito-Ashurst, I., Perry, L., Sanders, T.A. et al. (2012). subgroup analysis from the Nurses’ Health Study. Am. J.
A dietitian’s role in the management of blood pressure: Kidney Dis. 57 (2): 245–254.
results of a randomised controlled trial in British 187 Huang, X., Jiménez-Moleón, J.J., Lindholm, B. et al. (2013).
Bangladeshi chronic kidney disease patients [abstract]. Mediterranean diet, kidney function, and mortality in men
Clin. Nutr. Suppl. 7 (1): 168–169. with CKD. Clin. J. Am. Soc. Nephrol. 8 (9): 1548–1555.
Reference 657
88 Osté, M.C.J., Gomes-Neto, A.W., Corpeleijn, E. et al.

1 and caloric supplements in patients with chronic kidney
(2018). Dietary Approach to Stop Hypertension (DASH) failure in predialysis. Comparative study. Rev. Clin. Esp.
diet and risk of renal function decline and all-cause 198 (9): 580–586.
mortality in renal transplant recipients. Am. J. 201 Stratton, R.J., Bircher, G., Fouque, D. et al. (2005).
Transplant. 18 (10): 2523–2533. Multinutrient oral supplements and tube feeding in
189 Mekki, K., Bouzidi-bekada, N., Kaddous, A., and maintenance dialysis: a systematic review and meta-
Bouchenak, M. (2010). Mediterranean diet improves analysis. Am. J. Kidney Dis. 46 (3): 387–405.
dyslipidemia and biomarkers in chronic renal failure 202 Sigrist, M.K., Levin, A., and Tejani, A.M. (2010).
patients. Food Funct. 1 (1): 110–115. Systematic review of evidence for the use of intradialytic
190 Goraya, N., Simoni, J., Jo, C.H., and Wesson, D.E. (2014). parenteral nutrition in malnourished hemodialysis
Treatment of metabolic acidosis in patients with stage 3 patients. J. Ren. Nutr. 20 (1): 1–7.
chronic kidney disease with fruits and vegetables or oral 203 Marsen, T.A., Beer, J., Mann, H., and German IDPN-Trial
bicarbonate reduces urine angiotensinogen and preserves Group (2017). Intradialytic parenteral nutrition in
glomerular filtration rate. Kidney Int. 86 (5): 1031–1038. maintenance hemodialysis patients suffering from
191 Campbell, K.L., Ash, S., Zabel, R. et al. (2009). protein-energy wasting. Results of a multicenter, open,
Implementation of standardized nutrition guidelines by prospective, randomized trial. Clin. Nutr. 36 (1): 107–117.
renal dietitians is associated with improved nutrition 204 Tjiong, H.L., van den Berg, J.W., Wattimena, J.L. et al.
status. J. Ren. Nutr. 19 (2): 136–144. (2005). Dialysate as food: combined amino acid and
192 Campbell, K.L., Ash, S., Davies, P.S., and Bauer, J.D. glucose dialysate improves protein anabolism in renal
(2008). Randomized controlled trial of nutritional failure patients on automated peritoneal dialysis. J. Am.
counseling on body composition and dietary intake in Soc. Nephrol. 16 (5): 1486–1493.
severe CKD. Am. J. Kidney Dis. 51 (5): 748–758. 205 Stockler-Pinto, M.B., Mafra, D., Moraes, C. et al. (2014).
193 Campbell, K.L., Ash, S., and Bauer, J.D. (2008). The Brazil nut (Bertholletia excelsa, H.B.K.) improves
impact of nutrition intervention on quality of life in oxidative stress and inflammation biomarkers in
pre-dialysis chronic kidney disease patients. Clin. Nutr. hemodialysis patients. Biol. Trace Elem. Res. 158 (1):
27 (4): 537–544. 105–112.
194 Leon, J.B., Albert, J.M., and Gilchrist, G. (2006). 206 Ferramosca, E., Burke, S., Chasan-Taber, S. et al. (2005).
Improving albumin levels among hemodialysis patients: Potential antiatherogenic and anti-inflammatory
a community-based randomized controlled trial. Am. J. properties of sevelamer in maintenance hemodialysis
Kidney Dis. 48 (1): 28–36. patients. Am. Heart J. 149 (5): 820.
195 Akpele, L. and Bailey, J.L. (2004). Nutrition counseling 207 Saddadi, F., Alatab, S., Pasha, F. et al. (2014). The effect
impacts serum albumin levels. J. Ren. Nutr. 14 (3): 143–148. of treatment with N-acetylcysteine on the serum levels
196 Cano, N.J., Fouque, D., Roth, H. et al. (2007). of C-reactive protein and interleukin-6 in patients on
Intradialytic parenteral nutrition does not improve hemodialysis. Saudi J. Kidney Dis. Transpl. 25 (1): 66–72.
survival in malnourished hemodialysis patients: a 2-year 208 Iglesias, P., Díez, J.J., Fernández-Reyes, M.J. et al.
multicenter, prospective, randomized study. J. Am. Soc. (1998). Recombinant human growth hormone therapy
Nephrol. 18 (9): 2583–2591. in malnourished dialysis patients: a randomized
197 Teixidó-Planas, J., Ortiz, A., Coronel, F. et al. (2005). controlled study. Am. J. Kidney Dis. 32 (3): 454–463.
Oral protein-energy supplements in peritoneal dialysis: 209 Kotzmann, H., Schmidt, A., Lercher, P. et al. (2003).
a multicenter study. Perit. Dial. Int. 25 (2): 163–172. One-year growth hormone therapy improves
198 González-Espinoza, L., Gutiérrez-Chávez, J., del Campo, granulocyte function without major effects on
F.M. et al. (2005). Randomized, open label, controlled nutritional and anthropometric parameters in
clinical trial of oral administration of an egg albumin- malnourished hemodialysis patients. Nephron Clin.
based protein supplement to patients on continuous Pract. 93 (2): C75–C82.
ambulatory peritoneal dialysis. Perit. Dial. Int. 25 (2): 210 Johansen, K.L., Mulligan, K., and Schambelan, M.
173–180. (1999). Anabolic effects of nandrolone decanoate in
199 Aguirre Galindo, B.A., Prieto Fierro, J.G., Cano, P. et al. patients receiving dialysis: a randomized controlled
(2003). Effect of polymeric diets in patients on trial. JAMA 281 (14): 1275–1281.
continuous ambulatory peritoneal dialysis. Perit. Dial. 211 Rossi, A.P., Burris, D.D., Lucas, F.L. et al. (2014). Effects
Int. 23 (5): 434–439. of a renal rehabilitation exercise program in patients
200 Montes-Delgado, R., Guerrero Riscos, M.A., García- with CKD: a randomized, controlled trial. Clin. J. Am.
Luna, P.P. et al. (1998). Treatment with low-protein diet Soc. Nephrol. 9 (12): 2052–2058.
658
38
Preparation for Dialysis
Oluwatoyin I. Ameh1, Muhammad A. Makusidi2, Aminu M. Sakajiki2, Hamidu M. Liman2,
and Ikechi G. Okpechi3,4
1
Division of Nephrology, Zenith Medical and Kidney Centre, Abuja, Nigeria
2
Division of Nephrology, Usmanu Danfodiyo University Teaching Hospital, Sokoto, Nigeria
3
Division of Nephrology and Hypertension, Groote Schuur Hospital, Cape Town, South Africa
4
Kidney and Hypertension Research Unit, University of Cape Town, Cape Town, South Africa
I ntroduction pidemiology of ESKD and Dialysis

E
Therapies
Chronic kidney disease (CKD) is a progressive disorder associ-
ated with increased morbidity, cost of treatment, and mortal- It is widely known and accepted that ESKD is a noncom-
ity. When identified early, CKD progression to end-stage municable disease with global impact and public health
kidney disease (ESKD) can be delayed or halted. However, importance. ESKD represents the final phase in the spec-
once CKD has significantly advanced (estimated glomerular trum of CKD and is typified by the occurrence of worsened
filtration rate [eGFR] <30 ml/min/1.73 m2), very little can be biochemical and metabolic profiles. Although a greater
done to sufficiently delay progression of disease to avoid the percentage of CKD patients are likely to die from
need for renal replacement therapy (RRT) [1]. The number of cardiovascular-related causes rather than progress to
people with advanced CKD is high globally. For instance, the ESKD [8, 9], those who do progress usually need the intro-
United States Renal Data System (USRDS) 2018 annual report duction of RRT, such as dialysis, for survival.
showed that the overall prevalence of CKD (stages 1–5) in the There has been a steep increase in the prevalence of
United States adult general population was 14.8% in 2013–2016 ESKD since the first forms of RRT were instituted about six
(CKD stage 3 was the most prevalent [6.4%]) [2]. Although decades ago. In the United States, an increasing prevalence
RRT is often inaccessible to many across the world [3], in of ESKD continues to be observed despite plateauing inci-
places where treatment is available it is important that patients dence rates; a recent summary of the United States Renal
are adequately prepared for treatment initiation. In many Data System (USRDS) reports a yearly increase of about
instances, however, patients with CKD will not be prepared for 21 000 ESKD prevalent cases per year [10]. A similar pat-
RRT, especially in developing countries where patients are tern of increase has been observed in Europe [11]. This epi-
often “crash-landers” requiring RRT at time of first contact demiologic trend has far-reaching implications with
with nephrologist [4, 5]. Preparation for dialysis, when carried regards to the initiation and provision of RRT due to
out through a multidisciplinary process, has been shown to be increasing need and resource implications. The global
associated with better patient survival [6] or slower decline in prevalence of ESKD is presently estimated at 0.1%, with
glomerular filtration rate (GFR) compared to patients receiv- known notable inter- and intra-regional differences in the
ing usual care [7]. Strategies for retarding progression of CKD availability and access to RRT. There will be a continued
and treating complications associated with CKD are discussed increase in the incidence and prevalence of ESKD and dial-
in detail in Chapters 31 to 37. This chapter focuses on the ysis therapies due to increasing age, the increasing burden
approach to dialysis modality selection, patient education, and of chronic diseases such as diabetes mellitus and hyperten-
preparation for specific modalities. sion, as well as advances in the treatment of these diseases
Timing of Nephrology Referral (Early vs. Late) and Outcome 659
which promote longevity. In addition, the contribution of iming of Nephrology Referral (Early
T
chronic infectious disease-related ESKD, such as HIV- vs. Late) and Outcomes
associated nephropathy and hepatitis B-and C-associated
kidney disease, to this burden cannot be overlooked. Although there is no consensus on the ideal timing for neph-
Maintenance dialysis modalities have become the more rology referral of the CKD patient, most societal/organiza-
predominant forms of RRT; kidney transplantation tional guidelines recommend that CKD patients should be
remains an underutilized RRT option due to reasons such referred to a nephrologist when GFR is <30 ml/min/1.73 m2
as organ scarcity and the unbearably high costs of setting (i.e. stage 4). Early referral ensures that the choice of care
up and running renal transplant programs, especially for (conservative or active, peritoneal dialysis [PD] or hemodialy-
many developing nations [12]. There were approximately sis [HD]) can be a shared decision-making process rather than
284 patients per million population (pmp) receiving main- an option chosen out of dire necessity [14]. It also ensures that
tenance dialysis in 2010 [13]. Incidence rates (i.e. initiation the discussion about ESKD care with respect to RRT options
rates) of maintenance dialysis have been the highest in such as renal transplantation and the various alternatives of
low-to middle-income countries due to improved access to HD and PD therapies can be established. It has been demon-
dialysis therapies. Prevalence rates are on the increase strated that the timing of nephrology referral influences the
globally; rates have increased, albeit disproportionately, by eventual dialysis modality selected and the creation of the
between 154% and 170%. Prevalence rates have been esti- required vascular access in the circumstances of hemodialytic
mated at >2000 patients pmp, 750–1999 patients pmp, therapies [15]. Early referral of the CKD patient to the neph-
10–99 patients pmp, and <0.1 patients pmp in North rologist provides ample time for permanent access creation
America, Europe (excluding Russia), Central Asia, and and the opportunity to address any complications that might
West Africa, respectively [13]. occur before dialysis becomes needed (Table 38.1).
Table 38.1 Consequences of late referral and benefits of early referral of CKD patients to nephrologist.
Consequences of late referrala of patients with CKD Benefits of early referralb of patients with CKD
●● Inadequate control of diabetes, blood pressure, hyperlipidaemia, ●● Prevention and management of modifiable risk factors of
and malnutrition CKD
●● Inadequate management of anemia and mineral bone disease ●● Optimization of treatment of CKD
●● Rapid deterioration of renal function ●● Delayed progression of CKD
●● Increased hospitalizations and duration of in-hospital stay ●● Access to structured educational programs
●● Increased cardiovascular morbidity and mortality ●● Adaptation of CKD patient to RRT treatment
●● Poor preparation for dialysis ●● Preparation and creation of suitable dialysis access
●● Inadequate psychological preparation and support ●● Avoidance or lesser need for temporary vascular access
●● Loss of opportunity to participate in shared decision-making on ●● Training on selected modality of RRT
dialysis modality and initiation
●● Less chance of initiating RRT with PD ●● Better compliance
●● Increased chance of use of temporary access with associated ●● Increased opportunity for pre-emptive kidney
complications transplantation
●● Increased need for urgent start of dialysis ●● Reduction of cardiovascular morbidity and mortality
●● Increased chance of poor compliance to treatment due to poor ●● Reduction of management costs due to better planning
understanding
●● Lack of pre-transplant evaluation and loss of opportunity for
pre-emptive kidney transplant
●● Excessive management costs
CKD, chronic kidney disease; RRT, renal replacement therapy; PD, peritoneal dialysis.
a
Referral of patients with CKD <3 months before initiation of dialysis.
b
Primary healthcare physician should ideally refer patients with CKD at stage 1 if hematuria or significant proteinuria is present, at stage 2 if
eGFR declines >4 ml/min/year, and at stage 3 for all patients with CKD where eGFR is <60 ml/min/1.73 m2.
Source: Adapted from Karkar [14]. © 2011, Wolters Kluwer.
660 Preparation for Dialysis
I mportance of Patient Educational and patient outcomes [23–26]. Lopez-Vargas et al. [27] in a

and Learning Programs systematic review of educational interventions in CKD
patients that included 26 studies (12 trials, 14 observational
Management of patients with CKD can be complicated studies involving 5403 participants) broadly categorized
given that at early stages CKD is usually an asymptomatic the primary outcomes of educational interventions into
condition. Explaining this diagnosis to a patient, who is two groups: (i) patient-reported outcomes (e.g. quality of
expected to make major lifestyle changes, can pose signifi- life, psychosocial function, knowledge, self-management,
and lifestyle modification) [28–32] and (ii) clinical end
cant challenges in the face of low public awareness of kid-
points (eGFR, dialysis therapy commencement, survival,
ney diseases in general [16, 17]. Several studies have shown
blood pressure, and biochemical markers) [26, 33–36].
that awareness of kidney disease is low amongst patients.
They reported that the characteristics of effective interven-
Tuot et al., using 1999–2008 National Health and Nutrition
tions included those with interactive and practical skills
Examination Survey (NHANES) data in 1852 adults with
teaching sessions (13/15 studies), integrated negotiated
eGFR <60 ml/min/1.73 m2, assessed whether markers of
goal setting (10/13 studies), involved groups of patients
kidney disease that should trigger CKD recognition among
(12/14 studies), their families (4/4 studies), and a multidis-
providers are associated with higher individual CKD
ciplinary team (6/6 studies), and frequent participant/edu-
awareness [18]. They reported that 90% of individuals with
cator encounters (weekly [4/5 studies], monthly [7/7
two to four markers of CKD and 84% of individuals with
studies]) [27]. Kurella Tamura et al. [37] evaluated the role
five or more markers of CKD were unaware of having
of participation in a kidney disease screening and educa-
CKD. Only those with albuminuria had greater odds of
tion program in improving ESKD preparation and survival
CKD awareness than those without (adjusted odds ratio
in 595 adults who developed ESKD after participating in
4.0, P < 0.01) [18].
the National Kidney Foundation Kidney Early Evaluation
Patient education is a process and should begin from the
Program (KEEP). They identified and selected non-KEEP
time when the physician reveals the diagnosis, which should
patients from a national ESKD registry and matched them
be delivered at the patient’s pace of understanding with to KEEP participants based on demographic and clinical
empathy and conveyance of hope [19]. This initial message characteristics. KEEP participants were more likely to see a
should be given in a quiet place that is free of interruptions, nephrologist before ESKD (76.0% vs. 69.3%, P < 0.01), more
and preferably when the patient is accompanied by support- likely to use PD versus HD (10.3% vs. 6.4%, P < 0.01), more
ive family or friend or in the presence of supportive staff or likely to be placed on the transplant waiting list before
care giver (e.g. nurse, counsellor, or social worker) [19, 20]. ESKD (24.2% vs. 17.1%, P < 0.01), and more likely to
Potential dialysis patients need information not only about undergo transplantation (9.7% vs. 6.4%, P < 0.01), although
the mechanics of specific dialysis modalities but also the they were not more likely to undergo pre-emptive trans-
associated features of such therapy to help make informed plantation (1.7% vs. 1.5%, P = 0.7). KEEP participants were
choices. Predialysis CKD education should be individual- slightly more likely to use an arteriovenous (AV) fistula or
ized according to available resources and the level of knowl- graft (23.4% vs. 20.1%, P = 0.09) at the first outpatient dialy-
edge of the patient, and should include tours of dialysis sis and slightly more likely to have a mature or maturing
facilities, meeting patients who are undergoing treatment arteriovenous fistula or graft at the first outpatient dialysis
with different dialysis modalities, use of videos and written (44.0% vs. 39.6%, P = 0.06).
materials, and behavior-changing protocols with small-
group problem-solving activities [21]. Other education pro-
gram contents could include discussion of CKD risk factors,
interpretation of kidney function tests, complications of
odality Selection and Timing
M
CKD, interventions to slow loss of kidney function, dietary of Dialysis Initiation
changes necessitated by disease state, different options for
dialysis and their impact on the individual’s lifestyle, travel
Modality Selection
and time commitments for in-center HD treatment, impor- Counseling CKD patients about RRT options for ESKD is a
tance of preserving upper extremity veins for future dialysis distinct educational process. Patients in late stage 3 and
access, timing of placement of dialysis access, renal trans- stage 4 CKD should be educated on a range of potential
plantation, and financial costs and insurance coverage [22]. options, including home and in-center dialysis modalities,
There is evidence that educational interventions can be a transplantation, and in appropriate cases where RRT cannot
useful tool for improving knowledge on various aspects of be offered, conservative care (Figure 38.1) [38]. In the United
CKD (including risk factors), patient self-management, States, more than 90% of patients initiating RRT commence
Modality Selection and Timing of Dialysis Initiatio 661
Therapeutic options for

ESKD patients
Pre-emptive kidney
Dialysis OR OR Conservative care
transplantation
Home dialysis OR Center HD
PD OR HHD
Figure 38.1 Treatment options for patients with end-stage kidney disease. HD, hemodialysis; HHD, home hemodialysis; PD,
peritoneal dialysis.
therapy using in-center HD, with approximately 8.5% initi- to involve patients in selection of the best treatment option
ated on PD while the rest receive treatment through other suitable to them [44]. In the Empowering Patients on Choices
modalities (home HD [<2%], conservative management for Renal Replacement Therapy (EPOCH-RRT) study, a
[<1%], and pre-emptive kidney transplant [<1%]) [39]. cross-sectional study was designed to assess patients’ per-
There are geographic differences in the modality patients are spectives at the time they choose a dialysis modality [51]. It
likely to initiate RRT on depending on government policy included 180 patients with advanced CKD, either nondialysis
and the country’s economic status [40, 41]. Other factors like dependent or on dialysis therapy (HD and PD) across the
the patient’s clinical (comorbid) conditions and the neph- United States. Maintaining as much independence as possi-
rologist’s familiarity with each technique also play a role on ble, quality and quantity of life, and flexibility in daily sched-
modality selection; often, however, little attention is given to ule were the themes most often reported by patients as
individual patient preferences [42]. important for choice of modality. About half of the HD
Selecting a dialysis modality is a complex decision process patients (47%) believed that the decision to be treated by HD
for the patient. Murray et al. have shown that factors influ- had largely not been their choice; this was only reported by
encing a CKD patient’s participation in the decision-making 3% of PD patients. The study findings supported the need for
included interpersonal relationships; preservation of current interventions to improve shared decision-making on dialysis
well-being, normality, and quality of life, need for control, treatment options, targeting both patients and clinicians.
and personal importance on benefits and risks [43]. Patient Expected patient survival on dialysis is a major factor that
involvement in modality selection through shared decision- influences choice of treatment modality. Several studies
making is advocated to support patients deciding on a modal- have assessed the relationship between dialysis modality
ity of treatment suitable to their clinical condition, lifestyle, and survival; most studies report an initial survival benefit
and social circumstances [44]. However, for shared decision- on PD within the first 2 years of treatment initiation, but a
making to occur, the patient must be willing to be involved in less favorable long-term survival for PD compared with
their dialysis treatment decision. This is not always observed HD [52–54]. In a recent study that used data from 196 076
as studies have shown that up to 27% of patients may prefer a patients within the European Renal Association-European
passive role in treatment decision-making [45–47]. Also, due Dialysis and Transplant Association (ERA-EDTA) Registry
to resource constraints, the choice of modality may be made who started RRT between 1993 and 2012, van de Luijtgaarden
only by physicians (e.g. in South Africa where RRT is et al. reported a reduction of PD incidence and prevalence
rationed) [4]. Barriers to shared decision-making have been with time [55]. They also showed that there was a similar
explored and include time needed to integrate shared 5-year patient survival for PD versus HD in 1993–1997
decision-making in clinical practice, patients preference not (adjusted hazard ratio [HR] 1.02, 95% confidence interval
to be involved in the process of decision-making, and disa- [CI] 0.98–1.06), while patient survival was higher for PD
greement amongst healthcare professionals on asking patients in 2003–2007 (HR 0.91, 95% CI 0.88–0.95). Thus,
patients about their preference (some patients are informa- although initiating RRT on PD was associated with favorable
tion receivers and others are engagers) [48–50]. However, the patient survival when compared with starting on HD treat-
shared decision-making model provides an ideal framework ment, PD was often not selected as initial dialysis modality.
Although home hemodialysis (HHD) is becoming an Table 38.2 Contraindications to specific dialysis modalities.
increasingly available RRT option, patient selection criteria
may confound comparison with other RRT modalities. For Hemodialysis
instance, HHD patients in Australia and New Zealand are Relative contraindications
usually younger and have fewer comorbidities than ●●Difficult vascular access
patients undergoing PD and in-center HD [56, 57]. One –– Severe peripheral arterial vascular
study that included 10 710 patients on incident PD and 706 disease
patients on incident HHD reported that treatment with –– Central vein stenoses or occlusions
HHD was associated with better patient survival than
●● Desired home dialysis with lack of
treatment with PD (5-year survival 85% vs. 44%, respectively, appropriate home environment or partner
log-rank P < 0.001) [57]. Another study showed that home Peritoneal
dialysis (home HD + PD) had a 13% lower mortality risk dialysis
compared to facility HD, while PD showed a 20% lower Absolute contraindications
mortality risk in the early period (<3 years) that is offset by ●● Extensive abdominal adhesions
a 33% greater mortality risk in the late period (>3 years), ●● Hydrothorax or diaphragmatic defects
with no overall net effect [56]. When the attributes of ●● Loss of peritoneal membrane function
dialysis care and the trade-offs that patients considered (ultrafiltration, solute clearance)
when making decisions about dialysis modalities was Relative contraindications
quantified, Walker et al. reported that patients preferred ●● Morbid obesity
home dialysis over facility-based care when increased ●● Severe protein calorie malnutrition
●● Loss of residual renal function

nursing support was available and when longer survival,
●● Bowel disease
wellbeing, and flexibility were expected. In addition, they
showed that sociodemographic factors (age, ethnicity, and –– Ischemic
income) also influenced patients’ choices [58]. –– Chronic inflammatory
Given that the presence of some comorbidities or patient –– Diverticulosis or diverticulitis
features may be a relative or absolute contraindication to
Source: Adapted from Parker and Himmelfarb [59].
the desired dialysis modality (Table 38.2) [59], patients
should be informed about this and guided toward other
options. Patients should also understand that no care plan (eGFR <6–10 ml/min/1.73 m2) and, importantly, for dialysis
is completely inflexible, and modality switching can and initiation to be guided by the patient’s clinical status (pres-
should occur if an initial choice is deemed less than optimal ence of uremic symptoms, nutritional status, and daily life
by the patient and kidney care team. activities) (Table 38.3) [70–74].
The IDEAL study was specifically designed to evaluate
whether the timing of the initiation of maintenance dialy-
Timing of Dialysis Initiation
sis influenced survival among CKD patients [62]. Patients
The optimal time to initiate maintenance dialysis is cur- were randomly assigned to initiate dialysis early (eGFR
rently unknown; many patients are vulnerable during this 10–14 ml/min/1.73 m2) or assigned to the late-start group
transition period as it has been reported that the majority of (eGFR 5–7 ml/min/1.73 m2). The primary outcome was all-
patients are unprepared and ill-informed about commenc- cause mortality. During the follow-up period, 152 of 404
ing dialysis [42]. The factors surrounding timing of initiation patients in the early-start group (37.6%) and 155 of 424 in
of treatment are often multifaceted and are often different the late-start group (36.6%) died (HR with early start
for patients and care providers. For instance, one study group 1.04, 95% CI 0.83–1.30, P = 0.75). The time of dialy-
reported that patients based their choice on a “gut instinct,” sis did not influence secondary events (cardiovascular,
as well as the effects of commencing treatment on quality of infectious) or quality of life. The study concluded that
life and survival, while care providers considered biomedical planned early initiation of dialysis in patients with stage 5
factors and an instinct to prolong life [60]. In the United CKD was not associated with an improvement in survival
States, provider characteristics such as graduating from a or clinical outcomes. Following publication of the IDEAL
nondomestic medical school and greater experience (length study, several observational studies have been published
of time of practice) have been linked with timing of initia- about timing of dialysis initiation [61, 75–77]; most did not
tion of treatment [61]. Various guidelines published after the show benefit with early initiation and some even reported
Initiating Dialysis Early and Late (IDEAL) study [62] have harm when treatment was started early. This suggests that
generally recommended a late commencement of dialysis presence of comorbidities rather than level of eGFR at
Modality Selection and Timing of Dialysis Initiatio 663
Table 38.3 Guideline recommendations on vascular access.
Strength and evidence

Group Year Recommendation statements of recommendation
Canadian Society of 2006 Planning for vascular access, access timing, placement, and GRADE: D (opinion)
Nephrology [63] maturation GRADE: D (opinion)
●● Preserve arm veins suitable for placement of vascular access.
●● Establish AVF when the patient has an eGFR of 15–20 ml/min

and progressive kidney disease.
Preferred access type GRADE: C
●●The preferred type of vascular access is a radio-cephalic native
vessel AVF.
National Kidney 2006 Planning for vascular access, access timing, placement, and GRADE: B
Foundation Kidney Disease maturation
Outcomes Quality Initiative ●● In patients with CKD stage 4 or 5, forearm and upper-arm veins
(NKF-KDOQI) [64] suitable for placement of vascular access should not be used for
venipuncture or for the placement of intravenous catheters,
subclavian catheters, or peripherally inserted central catheter
lines.
●● A fistula should be placed at least 6 months before the
anticipated start of HD treatments. This timing allows for access
evaluation and additional time for revision to ensure a working
fistula is available at initiation of dialysis therapy.
●● A graft should, in most cases, be placed at least 3–6 weeks before
the anticipated start of HD therapy. Some newer graft materials
may be cannulated immediately after placement.
Preferred access type Not graded
●●The access should be placed distally and in the upper
extremities whenever possible. Options for fistula placement
should be considered first, followed by prosthetic grafts if fistula
placement is not possible. Catheters should be avoided for HD
and used only when other options listed are not available.
European Renal Best 2007 Planning for vascular access, access timing, placement, and Evidence level: III
Practice Advisory Board maturation
(ERBP)/European Renal ●● Potential chronic HD patients should be ideally referred to the
Association – European nephrologist and/or surgeon for preparing vascular access when
Dialysis and Transplant they reach stage 4 of their CKD (eGFR <30 ml/min/1.73 m2) or
Association earlier in case of rapidly progressive nephropathy or specific
(ERA-EDTA) [65] clinical conditions such as diabetes or severe peripheral vascular
disease.
Preferred access type Evidence level: III
●● Autogenous AVF should be preferred over AVG and AVG should
be preferred over catheters.
●● The upper extremity AV fistula should be the preferred access
and should be placed as distal as possible.
The Society for Vascular 2008 Planning for vascular access, access timing, placement, and GRADE 1
Surgery [66] maturation recommendation, very
●● We recommend that patients with advanced CKD disease (late low-quality evidence
stage 4, MDRD <20 to 25 ml/min) who have elected HD as their
choice of renal replacement therapy be referred to an access
surgeon in order to evaluate and plan construction of AV access.
Preferred access type GRADE 1

●●We recommend the placement of forearm autogenous AV access recommendation, very
as the first choice for primary access for HD. low-quality evidence
(Continued)

Group Year Recommendation statements of recommendation
Renal Association Clinical 2015 Planning for vascular access, access timing, placement, and GRADE 2
Practice Guideline on maturation recommendation,
Vascular Access for ●● We suggest that all patients that may require HD should have low-quality evidence
Hemodialysis [67] education on forearm vein preservation. GRADE 1
●● We recommend that the exact timing of placement of vascular recommendation,
access will be determined by rate of decline of renal function, moderate quality
comorbidities, and by the surgical pathway. evidence
●●We recommend that all patients with ESKD who commence HD recommendation,
or are on long-term HD should dialyze with an AVF as first high-quality evidence
choice, an AVG as second choice, a tunneled venous catheter as
third choice, and a nontunneled temporary catheter as an
option of necessity.
Japanese Society for 2011 Planning for vascular access, access timing, placement, and Opinion, not graded
Dialysis Therapy [68] maturation
Opinion, not graded
●● If HD has been selected as the treatment for ESKD, the
nephrologist should explain the role and importance of vascular
access in this treatment and refer the patient to an access
surgeon as soon as possible.
●● Vascular access construction should be considered when eGFR
is <15 ml/min/1.73 m2 (CKD stages 4 and 5) as well as
considering clinical conditions; in patients with diabetic
nephropathy, who have a tendency to show overhydration,
vascular access construction should be considered at a higher
eGFR.
●●AVF of the wrist or the anatomical snuff box would be the first recommendation,
choice, but the site of vascular access construction should moderate quality
eventually be determined by a comprehensive evaluation of the evidence
patient’s background, general conditions, and local conditions.
Clinical Practice Guidelines 2018 Planning for vascular access, access timing, placement, and GRADE 1
of the European Society for maturation recommendation,
Vascular Surgery [69] ●● Referral of CKD patients to the nephrologist and/or surgeon for very-low quality
preparing vascular access is recommended when they reach evidence
stage 4 of CKD (eGFR <30 ml/min/1.73 m2), especially in cases GRADE 1
of rapidly progressing nephropathy. recommendation,
●● A permanent vascular access should be created 3–6 months moderate quality
before the expected start of HD treatment. evidence
●●An autogenous AVF is recommended as the primary option for recommendation, high
vascular access. quality evidence
AV, arteriovenous; AVF, arteriovenous fistula; AVG, arteriovenous graft; CKD, chronic kidney disease; ESKD, end-stage kidney disease; eGFR,
estimated glomerular filtration rate; HD, hemodialysis; MDRD, modification of diet in renal disease.
initiation could be associated with patient outcomes et al. reported that in comparison with the late-start group,
[77–79]. Given that residual renal function is a key deter- the overall mortality rate was not higher for the early-start
minant of outcome and is better preserved with PD, some (adjusted HR 1.08, 95% CI 0.96–1.23) or mid-start (adjusted
studies have assessed the effects of early-start compared HR 0.96, 95% CI 0.86–1.09) groups [80]. Their study con-
with late-start dialysis on clinical outcomes in PD clusion, that early initiation of PD therapy (eGFR >10.5 ml/
patients [80, 81]. In a cohort of 8047 incident PD patients min/1.73 m2) is not associated with increased mortality,
who started dialysis therapy in 2001–2009 in Canada, Jain was divergent to most observational studies assessing HD.
Preparation for Hemodialysis and Vascular Acces 665
Put together, it will be more beneficial if choice of dialysis are preferred over CVCs as they have comparatively lower
modality and timing of commencement are determined on risks of infection, thrombosis, or stenosis of central veins [85];
an individual basis that combines patient preference and in addition, they deliver superior blood flow rates which in
lifestyle considerations while integrating medical factors turn contribute to better clearance of uremic wastes. For
and availability of resources and support, rather than com- instance, an AVF provides blood flow rates of up to 500 ml/
parison of survival outcomes [79]. min with relatively low complication rates [86]. In addition,
Preparation for transplantation is addressed elsewhere in AVF is the most cost-effective vascular access option as it is
this textbook (Chapters 56 and 57). Unless there is absolute associated with relatively lower infection rates and need for
certainty about a patient’s prospect for pre-emptive live endovascular revision [87–89]. It is also recommended that
donor kidney transplantation, it is preferable to provide AVFs should preferably be created using more proximal ves-
dialysis preparation in parallel with or as an alternative to sels with the aim of preserving the distal vessels for later use
transplantation evaluation [59]. in the event of a failed proximal AVF (Table 38.3) [63–66, 68,
69]. Some have challenged the assertion that AVFs are supe-
rior or should be the preferred choice of access [90, 91]. They
reparation for Hemodialysis
P argue that various comparisons between access types are sub-
and Vascular Access ject to high degrees of bias due to their observational design
and suggest that such comparisons are unable to account for
An integral component of dialysis is preparation and initia- other patient factors which likely contribute to poor outcomes
tion. Hemodialysis as an extracorporeal form of RRT seen in patients treated with catheters [92, 93].
requires access to the circulatory system (i.e. vascular Although AVF is the preferred vascular access type, its use
access) for blood to be circulated through the extracorporeal at dialysis initiation has shown marked global variability.
membrane (i.e. the dialysis membrane) (Table 38.3). Astor et al. in the Choices for Healthy Outcomes in Caring
Selection of the optimal type of vascular access can be for End-Stage Renal Disease (CHOICE) study demonstrated
challenging as there are various factors that need to be that only about one-third of the proportion of patients
taken into consideration [82]. These elements include referred early to their nephrology service within the prior
health-system associated factors, principal among which 12 months started HD with an AV (AVF/AVG) access [94].
include the timing of referral of the CKD patient from Data from the Dialysis Outcomes and Practice Patterns
primary care to the nephrologist or referral from nephrology Study (DOPPS) [95] indicate that AVF use among incident
care to the vascular access team, patient-related factors, dialysis patients is highest in Japan, where 84% of patients
such as the patient’s preferred choice, age, and comorbidities initiated dialysis with an AVF. This contrasts with 28% of
(e.g. diabetes mellitus, obesity, peripheral vessel disease), patients in the United States and 19% in the Gulf Cooperation
and institutional availability/accessibility to the technical Council countries such as Kuwait and Bahrain; 50–60% of
expertise required in creating and revising the vascular most European countries report dialysis initiation with an
accesses should the need arise. AVF [94]. The Fistula First Breakthrough Initiative (FFBI)
The timing of referral of the CKD patient to the nephrolo- of the Centers for Medicare and Medicaid Services (CMS),
gist significantly influences the type of access incident HD initiated in 2003, is a set of strategies based on the National
patients commence therapy with. Clinical practice guidelines Kidney Foundation Kidney Disease Outcomes Quality
recommend that in preparing CKD patients for HD, arterio- Initiative (NKF-KDOQI) Clinical Practice Guidelines for
venous (AV) access should be created well in advance of the Vascular Access with the aim of promoting more AV vascu-
anticipated start time of HD [63, 66, 67, 74, 83]. The guidelines lar access use among incident and prevalent HD patients. It
also recommend that CKD patients be referred early, with the has set a target of 50% AVF use among incident HD patients.
exact eGFR cut-offs varying among guidelines. They also One of the core strategies of the initiative in attaining this
advocate an even earlier referral for stage 4 CKD patients with objective is the timely referral of the CKD patient to a vascu-
known comorbidities as these may influence the outcome of lar surgeon for AVF assessment and placement [66, 96].
access creation. There are three main vascular access options As well-defined as the objective and benefits of establish-
for HD, including (in order of preference) native arterio- ing an autogenous vascular access route at dialysis initiation
venous fistulae (AVF), arteriovenous grafts (AVG), and central are, there are patients for whom such access may not be a
venous catheters (CVCs). AVFs remain the preferred initial preferred option, for example the elderly. As the global pop-
option over AVGs, while CVCs are the least preferred except ulation of elderly CKD patients increases, the incidence
in special circumstances, for example high-risk of AV compli- rates of ESKD in the elderly population aged 65 years and
cations or the elderly with limited life expectancy or personal above will likewise increase [2]. This translates to an older
preference [84]. Arteriovenous access options (AVG or AVF) population of patients with a high burden of comorbidities,
such as diabetes mellitus, peripheral vascular disease, and of the catheter insertion site, training of the patient on
congestive heart failure, requiring the establishment of vas- the peculiarities of the PD technique such as catheter
cular access to initiate dialysis. Age and comorbidities are handling and care, and carrying out supervised
known negative determinants of autologous AV access mat- exchanges. For the patients whose catheter will not be
uration and functionality [77, 79, 97]. With an all-cause mor- used soon, it can be buried in the anterior abdominal
tality rate of 31% per annum [98], death becomes a competing wall until exteriorization at that time when PD is about
risk for fistula maturation and cost-effectiveness in the to be commenced.
elderly [99]. There are no specific recommendations for vas-
cular access type in the elderly; however, the suitability of
the fistula first initiative has been questioned [100]. C
onclusion
Early referral and involvement of the nephrologist are

Preparation for Peritoneal Dialysis important to enable adequate preparation of CKD patients
who progress to requiring dialysis. A multidisciplinary
As is the case with commencing HD, early referral is key to approach to patient preparation for dialysis is encouraged
preparing the patient who has made the choice of PD. It is and should be individualized on a basis that combines
recommended that potential PD patients should be referred patient preference and lifestyle considerations, while inte-
at the latest 4 months prior to the anticipated start of dialy- grating medical factors and availability of resources and
sis [101]. This period affords the opportunity for PD educa- support. The approach should also include comprehensive
tion and home visits by a home PD team to assess the patient education that allows for shared decision-making
suitability of the patient’s living conditions for PD and the on choice of dialysis modality. Early referral also enables
availability of adequate storage space for PD supplies. In adequate time to prepare for successful creation and matu-
situations where suitability of the home is suboptimal, nec- ration of a native AVF (or other vascular access options) in
essary modifications may be made in time. In addition, patients who have opted for HD. Although guidelines pub-
early referral provides ample time to establish PD access lished after the IDEAL study suggest that dialysis be initi-
and obviates the need for bridging HD therapy. ated late, for patients who have opted to start treatment
A functioning PD catheter is a crucial requirement to with PD or in places where PD is favored dialysis should be
successfully starting PD. Current guidelines suggest that initiated at higher GFR when patients still have residual
PD access be established at least 2 weeks before the start renal function as this have been shown to be associated
of dialysis [102, 103]. This time allows for the healing with better patient outcomes.
References
1 Hazzan, A.D., Halinski, C., Agoritsas, S. et al. (2016). 6 Hemmelgarn, B.R., Manns, B.J., Zhang, J. et al. (2007).
Epidemiology and challenges to the management of Association between multidisciplinary care and survival
advanced CKD. Adv. Chronic Kidney Dis. 23 (4): 217–221. for elderly patients with chronic kidney disease. J. Am.
2 Saran, R., Robinson, B., Abbott, K.C. et al. (2018). US Soc. Nephrol. 18 (3): 993–999.
renal data system 2017 annual data report: epidemiology 7 Bayliss, E.A., Bhardwaja, B., Ross, C. et al. (2011).
of kidney disease in the United States. Am. J. Kidney Dis. Multidisciplinary team care may slow the rate of decline
71 (3s1): A7. in renal function. Clin. J. Am. Soc. Nephrol. 6 (4): 704–710.
3 Liyanage, T., Ninomiya, T., Jha, V. et al. (2015). 8 Foley, R.N., Murray, A.M., Li, S. et al. (2005). Chronic
Worldwide access to treatment for end-stage kidney kidney disease and the risk for cardiovascular disease,
disease: a systematic review. Lancet 385 (9981): renal replacement, and death in the United States
1975–1982. Medicare population, 1998 to 1999. J. Am. Soc. Nephrol.
4 Swanepoel, C.R., Wearne, N., and Okpechi, I.G. (2013). 16 (2): 489–495.
Nephrology in Africa – not yet uhuru. Nat. Rev. Nephrol. 9 9 Keith, D.S., Nichols, G.A., Gullion, C.M. et al. (2004).
(10): 610–622. Longitudinal follow-up and outcomes among a
5 Kilonzo, K.G., Jones, E.S.W., Okpechi, I.G. et al. (2017). population with chronic kidney disease in a large
Disparities in dialysis allocation: an audit from the new managed care organization. Arch. Intern. Med. 164 (6):
South Africa. PLoS One 12 (4): e0176041. 659–663.
Reference 667
10 Saran, R., Robinson, B., Abbott, K.C. et al. (2017). US 26 Wu, I.W., Wang, S.Y., Hsu, K.H. et al. (2009).
renal data system 2016 annual data report: epidemiology Multidisciplinary predialysis education decreases the
of kidney disease in the United States. Am. J. Kidney Dis. incidence of dialysis and reduces mortality–a controlled
69 (3 Suppl 1): A7–A8. cohort study based on the NKF/DOQI guidelines.
11 Heaf, J. (2017). Current trends in European renal Nephrol. Dial. Transplant. 24 (11): 3426–3433.
epidemiology. Clin. Kidney J. 10 (2): 149–153. 27 Lopez-Vargas, P.A., Tong, A., Howell, M., and Craig, J.C.
12 Chapman, J.R. (2013). What are the key challenges we (2016). Educational interventions for patients with
face in kidney transplantation today? Transplant. Res. 2 CKD: a systematic review. Am. J. Kidney Dis. 68 (3):
(Suppl 1): S1. 353–370.
13 Thomas, B., Wulf, S., Bikbov, B. et al. (2015). 28 Campbell, K.L., Ash, S., and Bauer, J.D. (2008). The
Maintenance dialysis throughout the world in years 1990 impact of nutrition intervention on quality of life in
and 2010. J. Am. Soc. Nephrol. 26 (11): 2621–2233. pre-dialysis chronic kidney disease patients. Clin. Nut.
14 Karkar, A. (2011). The value of pre-dialysis care. Saudi J. (Edinburgh, Scotland) 27 (4): 537–544.
Kidney Dis. Transplant. 22 (3): 419–427. 29 Gutierrez Vilaplana, J.M., Zampieron, A., Craver, L., and
15 Arora, P., Obrador, G.T., Ruthazer, R. et al. (1999). Buja, A. (2009). Evaluation of psychological outcomes
Prevalence, predictors, and consequences of late following the intervention ‘teaching group’: study on
nephrology referral at a tertiary care center. J. Am. Soc. predialysis patients. J. Ren. Care 35 (3): 159–164.
Nephrol. 10 (6): 1281–1286. 30 Paes-Barreto, J.G., Silva, M.I., Qureshi, A.R. et al. (2013).
16 White, S.L., Polkinghorne, K.R., Cass, A. et al. (2008). Can renal nutrition education improve adherence to a
Limited knowledge of kidney disease in a survey of low-protein diet in patients with stages 3 to 5 chronic
AusDiab study participants. Med. J. Australia 188 (4): kidney disease? J. Ren. Nut. 23 (3): 164–171.
204–208. 31 Lin, C.C., Tsai, F.M., Lin, H.S. et al. (2013). Effects of a
17 Shan, Y., Zhang, Q., Liu, Z. et al. (2010). Prevalence and self-management program on patients with early-stage
risk factors associated with chronic kidney disease in chronic kidney disease: a pilot study. Appl. Nurs. Res. 26
adults over 40 years: a population study from Central (3): 151–156.
China. Nephrology (Carlton) 15 (3): 354–361. 32 Teng, H.L., Yen, M., Fetzer, S. et al. (2013). Effects of
18 Tuot, D.S., Plantinga, L.C., Hsu, C.Y. et al. (2011). targeted interventions on lifestyle modifications of
Chronic kidney disease awareness among individuals chronic kidney disease patients: randomized controlled
with clinical markers of kidney dysfunction. Clin. J. Am. trial. West. J. Nurs. Res. 35 (9): 1107–1127.
Soc. Nephrol. 6 (8): 1838–1844. 33 Chen, S.H., Tsai, Y.F., Sun, C.Y. et al. (2011). The impact
19 Ptacek, J.T. and Eberhardt, T.L. (1996). Breaking bad of self-management support on the progression of
news. A review of the literature. JAMA 276 (6): 496–502. chronic kidney disease–a prospective randomized
20 Schofield, P.E., Butow, P.N., Thompson, J.F. et al. (2003). controlled trial. Nephrol. Dial. Transplant. 26 (11):
Psychological responses of patients receiving a diagnosis 3560–3566.
of cancer. Ann. Oncol. 14 (1): 48–56. 34 Pagels, A.A., Hylander, B., and Alvarsson, M. (2015). A
21 Golper, T. (2001). Patient education: can it maximize the multi-dimensional support programme for patients with
success of therapy? Nephrol. Dial. Transplant. 16 (Suppl diabetic kidney disease. J. Ren. Care 41 (3): 187–194.
7): 20–24. 35 Jia, T., Bi, S.H., Lindholm, B., and Wang, T. (2012). Effect
22 Saggi, S.J., Allon, M., Bernardini, J. et al. (2012). of multi-dimensional education on disease progression in
Considerations in the optimal preparation of patients for pre-dialysis patients in China. Ren. Fail. 34 (1): 47–52.
dialysis. Nat. Rev. Nephrol. 8 (7): 381–389. 36 Walker, R.C., Marshall, M.R., and Polaschek, N.R. (2014).
23 Devins, G.M., Mendelssohn, D.C., Barre, P.E., and Binik, A prospective clinical trial of specialist renal nursing in
Y.M. (2003). Predialysis psychoeducational intervention the primary care setting to prevent progression of chronic
and coping styles influence time to dialysis in chronic kidney: a quality improvement report. BMC Fam. Pract.
kidney disease. Am. J. Kidney Dis. 42 (4): 693–703. 15: 155.
24 Mehrotra, R. (2011). Bridging the care gap around dialysis 37 Kurella Tamura, M., Li, S., Chen, S.C. et al. (2014).
initiation: is CKD education part of the solution? Am. J. Educational programs improve the preparation for
Kidney Dis. 58 (2): 160–161. dialysis and survival of patients with chronic kidney
25 Lacson, E. Jr., Wang, W., DeVries, C. et al. (2011). Effects disease. Kidney Int. 85 (3): 686–692.
of a nationwide predialysis educational program on 38 Bargman, J.M. (2015). Timing of initiation of RRT and
modality choice, vascular access, and patient outcomes. modality selection. Clin. J. Am. Soc. Nephrol. 10 (6):
Am. J. Kidney Dis. 58 (2): 235–242. 1072–1077.
9 Saran, R., Li, Y., Robinson, B. et al. (2015). US renal data

3 52 McDonald, S.P., Marshall, M.R., Johnson, D.W., and
system 2014 annual data report: epidemiology of kidney Polkinghorne, K.R. (2009). Relationship between dialysis
disease in the United States. Am. J. Kidney Dis. 66 (1 modality and mortality. J. Am. Soc. Nephrol. 20 (1):
Suppl 1) Svii, S1–305. 155–163.
40 Liu, F.X., Gao, X., Inglese, G. et al. (2015). A global 53 Vonesh, E.F., Snyder, J.J., Foley, R.N., and Collins, A.J.
overview of the impact of peritoneal dialysis first or (2006). Mortality studies comparing peritoneal dialysis
favored policies: an opinion. Perit. Dial. Int. 35 (4): and hemodialysis: what do they tell us? Kidney Int. Suppl.
406–420. 103: S3–S11.
41 Li, P.K., Chow, K.M., Van de Luijtgaarden, M.W. et al. 54 Jaar, B.G., Coresh, J., Plantinga, L.C. et al. (2005).
(2017). Changes in the worldwide epidemiology of Comparing the risk for death with peritoneal dialysis and
peritoneal dialysis. Nat. Rev. Nephrol. 13 (2): 90–103. hemodialysis in a national cohort of patients with chronic
42 Song, M.K., Lin, F.C., Gilet, C.A. et al. (2013). Patient kidney disease. Ann. Intern. Med. 143 (3): 174–183.
perspectives on informed decision-making surrounding 55 van de Luijtgaarden, M.W., Jager, K.J., Segelmark, M.
dialysis initiation. Nephrol. Dial. Transplant. 28 (11): et al. (2016). Trends in dialysis modality choice and
2815–2823. related patient survival in the ERA-EDTA Registry over a
43 Murray, M.A., Brunier, G., Chung, J.O. et al. (2009). 20-year period. Nephrol. Dial. Transplant. 31 (1): 120–128.
A systematic review of factors influencing decision- 56 Marshall, M.R., Walker, R.C., Polkinghorne, K.R., and
making in adults living with chronic kidney disease. Lynn, K.L. (2014). Survival on home dialysis in New
Patient Educ. Couns. 76 (2): 149–158. Zealand. PLoS One 9 (5): e96847.
44 Johansson, L. (2013). Shared decision making and patient 57 Nadeau-Fredette, A.C., Hawley, C.M., Pascoe, E.M. et al.
involvement in choosing home therapies. J. Ren. Care 39 (2015). An incident cohort study comparing survival on
(Suppl 1): 9–15. home hemodialysis and peritoneal dialysis (Australia and
45 Deber, R.B., Kraetschmer, N., Urowitz, S., and Sharpe, New Zealand Dialysis and Transplantation Registry).
N. (2007). Do people want to be autonomous patients? Clin. J. Am. Soc. Nephrol. 10 (8): 1397–1407.
Preferred roles in treatment decision-making in several 58 Walker, R.C., Morton, R.L., Palmer, S.C. et al. (2018). A
patient populations. Health Expectations 10 (3): discrete choice study of patient preferences for dialysis
248–258. modalities. Clin. J. Am. Soc. Nephrol. 13 (1): 100–108.
46 Ambigapathy, R., Chia, Y.C., and Ng, C.J. (2016). Patient 59 Parker, M.G. and Himmelfarb, J. (2009). Preparation for
involvement in decision-making: a cross-sectional study Dialysis. In: Evidence-Based Nephrology (eds. J.C. Craig,
in a Malaysian primary care clinic. BMJ Open 6 (1): D.A. Molony and G.F.M. Strippoli), 391.
e010063. 60 Hussain, J.A., Flemming, K., Murtagh, F.E., and Johnson,
47 Levinson, W., Kao, A., Kuby, A., and Thisted, R.A. (2005). M.J. (2015). Patient and health care professional decision-
Not all patients want to participate in decision making. making to commence and withdraw from renal dialysis: a
A national study of public preferences. J. Gen. Intern. systematic review of qualitative research. Clin. J. Am. Soc.
Med. 20 (6): 531–535. Nephrol. 10 (7): 1201–1215.
48 Legare, F., Ratte, S., Gravel, K., and Graham, I.D. (2008). 61 Slinin, Y., Guo, H., Li, S. et al. (2014). Provider and care
Barriers and facilitators to implementing shared decision- characteristics associated with timing of dialysis
making in clinical practice: update of a systematic review initiation. Clin. J. Am. Soc. Nephrol. 9 (2): 310–317.
of health professionals’ perceptions. Patient Educ. Couns. 62 Cooper, B.A., Branley, P., Bulfone, L. et al. (2010). A
73 (3): 526–535. randomized, controlled trial of early versus late initiation
49 Gravel, K., Legare, F., and Graham, I.D. (2006). Barriers of dialysis. N. Engl. J. Med. 363 (7): 609–619.
and facilitators to implementing shared decision-making 63 Jindal, K., Chan, C.T., Deziel, C. et al. (2006).
in clinical practice: a systematic review of health Hemodialysis clinical practice guidelines for the
professionals’ perceptions. Implementation Sci. 1: 16. Canadian Society of Nephrology. J. Am. Soc. Nephrol. 17
50 Bonner, A. and Lloyd, A. (2012). Exploring the (3 Suppl 1): S1–S27.
information practices of people with end-stage kidney 64 Vascular Access 2006 Work Group (2006). Clinical
disease. J. Ren. Care 38 (3): 124–130. practice guidelines for vascular access. Am. J. Kidney Dis.
51 Dahlerus, C., Quinn, M., Messersmith, E. et al. (2016). 48: S176–S247. doi: 10.1053/j.ajkd.2006.04.029. 10.1053/j.
Patient perspectives on the choice of dialysis modality: ajkd.2006.04.029.
results from the empowering patients on choices for renal 65 Tordoir, J., Canaud, B., Haage, P. et al. (2007). EBPG on
replacement therapy (EPOCH-RRT) study. Am. J. Kidney vascular access. Nephrol. Dial. Transplant. 22 (suppl_2):
Dis. 68 (6): 901–910. ii88–ii117.
Reference 669
6 Sidawy, A.N., Spergel, L.M., Besarab, A. et al. (2008). The

6 79 Trinh, E., Chan, C.T., and Perl, J. (2018). Dialysis
Society for Vascular Surgery: clinical practice guidelines modality and survival: done to death. Semin. Dial. 31 (4):
for the surgical placement and maintenance of 315–324.
arteriovenous hemodialysis access. J. Vasc. Surg. 48 (5 80 Jain, A.K., Sontrop, J.M., Perl, J. et al. (2014). Timing of
Suppl): 2s–25s. peritoneal dialysis initiation and mortality: analysis of the
67 Kumwenda, M., Mitra, S., and Reid, C. (2015). UK Canadian Organ Replacement Registry. Am. J. Kidney Dis.
Renal Association Clinical Practice Guideline Vascular 63 (5): 798–805.
Access For Haemodialysis UK Renal Association. 6e. 81 Johnson, D.W., Wong, M.G., Cooper, B.A. et al. (2012).
https://renal.org/wp-content/uploads/2017/06/ Effect of timing of dialysis commencement on clinical
vascular-access.pdf. outcomes of patients with planned initiation of peritoneal
68 Kukita, K., Ohira, S., Amano, I. et al. (2015). 2011 update dialysis in the IDEAL trial. Perit. Dial. Int. 32 (6):
Japanese Society for Dialysis Therapy guidelines of 595–604.
vascular access construction and repair for chronic 82 Woo, K. and Lok, C.E. (2016). New insights into dialysis
hemodialysis. Ther. Apher. Dial. 19 (Suppl 1): 1–39. vascular access: what is the optimal vascular access type
69 Schmidli, J., Widmer, M.K., Basile, C. et al. (2018). and timing of access creation in CKD and dialysis
Editor’s choice – vascular access: 2018 clinical practice patients? Clin. J. Am. Soc. Nephrol. 11 (8): 1487–1494.
guidelines of the European Society for Vascular Surgery 83 Polkinghorne, K.R., Chin, G.K., MacGinley, R.J. et al.
(ESVS). Eur. J. Vasc. Endovasc. Surg. 55 (6): 757–818. (2013). KHA-CARI guideline: vascular access – central
70 Tattersall, J., Dekker, F., Heimburger, O. et al. (2011). venous catheters, arteriovenous fistulae and
When to start dialysis: updated guidance following arteriovenous grafts. Nephrology (Carlton) 18 (11):
publication of the Initiating Dialysis Early and Late 701–705.
(IDEAL) study. Nephrol. Dial. Transplant. 26 (7): 84 Papini, A., Ravani, P., and Quinn, R.R. (2018).
2082–2086. Controversies in vascular access. Curr. Opin. Nephrol.
71 KDIGO. Kidney Disease: Improving Global Outcomes Hypertens 27 (3): 209–213.
(KDIGO) CKD Work Group (2013). KDIGO 2012 85 Twardowski, Z.J. (1995). Percutaneous blood access for
clinical practice guideline for the evaluation and hemodialysis. Semin. Dial. 8 (3): 175–186.
management of chronic kidney disease. Kidney Int. 86 Huijbregts, H.J. and Blankestijn, P.J. (2006). Dialysis
Suppl. 3: 1–150. access–guidelines for current practice. Eur. J. Vasc.
72 Nesrallah, G.E., Mustafa, R.A., Clark, W.F. et al. (2014). Endovasc. Surg. 31 (3): 284–287.
Canadian Society of Nephrology 2014 clinical practice 87 Lok, C.E. and Foley, R. (2013). Vascular access morbidity
guideline for timing the initiation of chronic dialysis. and mortality: trends of the last decade. Clin. J. Am. Soc.
Can. Med. Assoc. J. 186 (2): 112–117. Nephrol. 8 (7): 1213–1219.
73 Watanabe, Y., Yamagata, K., Nishi, S. et al. (2015). 88 Perera, G.B., Mueller, M.P., Kubaska, S.M. et al. (2004).
Japanese society for dialysis therapy clinical guideline for Superiority of autogenous arteriovenous hemodialysis
“hemodialysis initiation for maintenance hemodialysis”. access: maintenance of function with fewer secondary
Ther. Apher. Dial. 19 (Suppl 1): 93–107. interventions. Ann. Vasc. Surg. 18 (1): 66–73.
74 KDOQI Clinical Practice Guideline for Hemodialysis 89 Lok, C.E., Sontrop, J.M., Tomlinson, G. et al. (2013).
Adequacy (2015). 2015 update. Am. J. Kidney Dis. 66 (5): Cumulative patency of contemporary fistulas versus
884–930. grafts (2000-2010). Clin. J. Am. Soc. Nephrol. 8 (5):
75 Rosansky, S.J., Eggers, P., Jackson, K. et al. (2011). Early 810–818.
start of hemodialysis may be harmful. Arch. Intern. Med. 90 Dumaine, C.S., Brown, R.S., MacRae, J.M. et al. (2018).
171 (5): 396–403. Central venous catheters for chronic hemodialysis: is
76 Wright, S., Klausner, D., Baird, B. et al. (2010). Timing of “last choice” never the “right choice”? Semin. Dial. 31 (1):
dialysis initiation and survival in ESRD. Clin. J. Am. Soc. 3–10.
Nephrol. 5 (10): 1828–1835. 91 Murea, M., James, K.M., Russell, G.B. et al. (2014). Risk
77 Zhang, Y., Hu, C., Bian, Z., and Chen, P. (2018). Impact of of catheter-related bloodstream infection in elderly
timing of initiation of dialysis on long-term prognosis of patients on hemodialysis. Clin. J. Am. Soc. Nephrol. 9 (4):
patients undergoing hemodialysis. Exp. Ther. Med. 16 (2): 764–770.
1209–1215. 92 Grubbs, V., Wasse, H., Vittinghoff, E. et al. (2014). Health
78 Murphy, S.W., Foley, R.N., Barrett, B.J. et al. (2000). status as a potential mediator of the association between
Comparative mortality of hemodialysis and peritoneal hemodialysis vascular access and mortality. Nephrol. Dial.
dialysis in Canada. Kidney Int. 57 (4): 1720–1726. Transplant. 29 (4): 892–898.
93 Ravani, P., Palmer, S.C., Oliver, M.J. et al. (2013). the United States. National Institutes of Health,
Associations between hemodialysis access type and National Institute of Diabetes and Digestive and Kidney
clinical outcomes: a systematic review. J. Am. Soc. Diseases. Bethesda, MD.
Nephrol. 24 (3): 465–473. 99 Hall, R.K., Myers, E.R., Rosas, S.E. et al. (2017). Choice
94 Astor, B.C., Eustace, J.A., Powe, N.R. et al. (2001). of hemodialysis access in older adults: a cost-
Timing of nephrologist referral and arteriovenous access effectiveness analysis. Clin. J. Am. Soc. Nephrol. 12 (6):
use: the CHOICE Study. Am. J. Kidney Dis. 38 (3): 947–954.
494–501. 100 Woo, K., Goldman, D.P., and Romley, J.A. (2015). Early
95 Pisoni, R.L., Zepel, L., Port, F.K., and Robinson, B.M. failure of dialysis access among the elderly in the era of
(2015). Trends in US vascular access use, patient fistula first. Clin. J. Am. Soc. Nephrol. 10 (10):
preferences, and related practices: an update from the 1791–1798.
US DOPPS practice monitor with international 101 Lo, W.K., Kwan, T.H., Ho, Y.W. et al. (2008). Preparing
comparisons. Am. J. Kidney Dis. 65 (6): 905–915. patients for peritoneal dialysis. Perit. Dial. Int. 28 (Suppl
96 Sequeira, A., Naljayan, M., and Vachharajani, T.J. 3): S69–S71.
(2017). Vascular access guidelines: summary, rationale, 102 Figueiredo, A., Goh, B.L., Jenkins, S. et al. (2010).
and controversies. Tech. Vasc. Interv. Radiol. 20 (1): 2–8. Clinical practice guidelines for peritoneal access. Perit.
97 Hod, T., Desilva, R.N., Patibandla, B.K. et al. (2014). Dial. Int. 30 (4): 424–429.
Factors predicting failure of AV “fistula first” policy in 103 Dombros, N., Dratwa, M., Feriani, M. et al. (2005).
the elderly. Hemodial. Int. 18 (2): 507–515. European best practice guidelines for peritoneal dialysis.
98 USRDS. (2013). USRDS 2013 Annual Data Report: atlas 3 peritoneal access. Nephrol. Dial. Transplant. 20
of chronic kidney disease and end-stage renal disease in (Suppl 9): ix8–ix12.
671
39
Choice of Hemodialysis or Peritoneal Dialysis for Kidney Replacement Therapy

Marlies Noordzij1 and Raymond T. Krediet2
1
Department of Medical Informatics, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands
2
Division of Nephrology, Department of Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands
I ntroduction discussion. It should be emphasized that all studies on late

referral to a nephrologist are either retrospective studies,
The issue of timing of initiation of dialysis was first raised based on surveys, or prospective cohort studies; neverthe-
by Bonomini et al. in the 1970s [1–3]. They reported the less, the results are strikingly similar. Late referral for
highest survival rate in hemodialysis (HD) patients with a maintenance dialysis is associated with increased mortal-
residual kidney function (measured by creatinine clear- ity [4–12] and decreased quality of life (QoL) [13, 14]. The
ance) between 15 and 21 ml/min. However, criteria for the excess mortality associated with late referral was independ-
selection of patients for dialysis based on residual kidney ent of demographic characteristics, socio-economic status,
function were not provided. The issue of an “earlier” start, and comorbidity. Moreover, late referral influenced the
which means with a higher glomerular filtration rate dialysis modality choice; patients who obtained predialysis
(GFR) than the usual threshold of 5–10 ml/min/1.73 m2, care had a stronger preference for PD than patients
reappeared again in the late 1990s when the process of who did not, the so-called late referrals [15]. In addition,
developing guidelines for dialysis was started. patients who were referred late and started with PD were
The choice of dialysis modality at the initiation of kidney more likely to switch to HD during the first 6 months com-
replacement therapy has also been studied and discussed pared with patients who were referred in a more timely
extensively in recent decades. There is, however, still no manner [16]. Despite all the evidence pointing to a benefi-
final consensus on whether HD or peritoneal dialysis (PD) cial effect of timely referral to a nephrologist, almost one-
treatment modality gives the best results. Consequently, fourth of patients were referred less than 1 month before
both options have to be weighed in individual patients the start of dialysis in a survey of eight different European
according to their specific needs, preferences, and clinical countries [17].
characteristics, with the aim of providing a patient-tailored
kidney replacement therapy.
In this chapter, an overview is given of the current litera- uidelines and Studies on When
G
ture on the optimal timing of dialysis initiation and on the to Initiate Dialysis Treatment
best dialysis modality choice.
The timing of initiation of dialysis has been the subject of
many opinion-based guidelines. A summary of interna-
arly Start or Timely Referral
E tional guidelines is presented in Table 39.1. The US
for Dialysis Initiation National Kidney Foundation Dialysis Outcomes Quality
Initiative (K/DOQI) Workgroup on Peritoneal Dialysis
A distinction should be made between the timing of referral published an opinion-based guideline on the initiation of
to a nephrologist and the timing of initiation of dialysis chronic dialysis treatment in 1997 [22]. The workgroup
treatment. Failure to make this distinction has confused the advised that dialysis should be initiated when renal Kt/Vurea
672 Choice of Hemodialysis or Peritoneal Dialysis for Kidney Replacement Therapy
Table 39.1 Guideline recommendations on the timing of initiation of maintenance dialysis in patients with end-stage kidney disease
(ESKD), based on GRADE.

Group Year Recommendation statement of recommendation
European Renal Best 2011 ●● In patients with eGFR <15 ml/min/1.73 m2, dialysis should be Grade: strong
Practice advisory board considered when there is one or more of the following: recommendation,
(ERBP)/European Renal symptoms or signs of uremia, inability to control hydration high-quality evidence
Association – European status or blood pressure or a progressive deterioration in
Dialysis and Transplant nutritional status. It should be taken into account that the
Association majority of patients will be symptomatic and need to start
(ERA-EDTA) [18] dialysis with eGFR in the range 9–6 ml/min/1.73 m2.
●● High-risk patients (e.g. diabetics and those whose renal Grade: strong
function is deteriorating more rapidly than eGFR 4 ml/ recommendation,
min//1.73 m2 per year) require particularly close supervision. low-quality evidence
Where close supervision is not feasible and in patients whose
uremic symptoms may be difficult to detect, a planned start to
dialysis while still asymptomatic may be preferred.
●● Asymptomatic patients presenting with advanced CKD may Grade: weak
benefit from a delay in starting dialysis in order to allow recommendation,
preparation, planning, and permanent access creation rather low-quality evidence
than using temporary access.
Kidney Disease Improving 2012 We suggest that dialysis be initiated when one or more of the Grade: 2B
Global Outcomes following are present: symptoms or signs attributable to kidney
(KDIGO) [19] failure (serositis, acid-base or electrolyte abnormalities, pruritus);
inability to control volume status or blood pressure; a progressive
deterioration in nutritional status refractory to dietary
intervention; or cognitive impairment. This often but not
invariably occurs in the GFR range between 5 and 10 ml/
min/1.73 m2.
Canadian Society of 2014 For adults (aged >18 yr) with an eGFR of <15 ml/min/1.73 m2, Grade: strong
Nephrology [20] we recommend an “intent-to-defer” over an “intent-to-start- recommendation;
early” approach for the initiation of chronic dialysis. With the moderate quality of
intent-to-defer strategy, patients with eGFR of <15 ml/ evidence
min/1.73 m2 are monitored closely by a nephrologist, and dialysis
is initiated with the first onset of a clinical indication or a decline
in the eGFR to 6 ml/min/1.73 m2 or less, whichever of these
should occur first. Clinical indications for the initiation of
dialysis include the following: symptoms of uremia, fluid
overload, refractory hyperkaliemia or acidemia, or other
conditions or symptoms that are likely to be ameliorated by
dialysis. In the absence of these factors, the eGFR should not
serve as a sole criterion for the initiation of dialysis unless it is
6 ml/min/1.73 m2 or less.
Japanese Society for 2015 The judgment on the time to initiate HD is allowed when a Grade: 1D
Dialysis Therapy [21] residual renal function shows progressive deterioration and
reaction to eGFR <15 ml/min/1.73 m2 in spite of sufficient
optimal conservative treatment. However, the decision of starting
HD should be determined based on a comprehensive assessment
of renal failure symptoms, daily life activities, and nutritional
status, which are not relievable without HD treatment.
National Kidney 2015 The decision to initiate maintenance dialysis in patients who Grade: not graded
Foundation Kidney choose to do so should be based primarily upon an assessment of
Disease Outcomes Quality signs and/or symptoms associated with uremia, evidence of
Initiative protein-energy wasting, and the ability to safely manage
(NKF-KDOQI) [1] metabolic abnormalities and/or volume overload with medical
therapy rather than on a specific level of kidney function in the
absence of such signs and symptoms.
CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; HD, hemodialysis.
Guidelines and Studies on When to Initiate Dialysis Treatmen 673
has fallen to 2.0/week. For a 70-kg man this equals a urea diminish the effects of muscle mass on plasma creatinine is
clearance of 8 ml/min, a GFR of 10 ml/min/1.73 m2, and a the use of creatinine clearance requiring timed urine col-
creatinine clearance of 12 ml/min. A lower Kt/Vurea would lections or the use of a formula that includes an approxi-
only be acceptable when the normalized protein equiva- mation of muscle mass, like the one developed by Cockcroft
lent of nitrogen appearance, an index of protein intake in and Gault (C&G) in which body weight is included [34].
stable patients, was at least 0.8 g/kg of body weight/day Using creatinine clearance, no effect of this parameter at
and the patient was in a good nutritional state. These rec- the start of dialysis on mortality was found between
ommendations were retained in the update published in patients who survived less than 1 year compared to those
2001. The clinical practice guidelines of the Canadian who survived more than 1 year on dialysis [35]. A compari-
Society of Nephrology for treatment of patients with son between 10-year patient survival after the start of dialy-
chronic kidney disease, published in 1999, were not sig- sis between early (C&G >8 ml/min) and late (C&G <8 ml/
nificantly different from the US guidelines but set the tar- min) starters showed a nonsignificant survival difference
get for the initiation of dialysis at a GFR of 12 ml/ in favor of the early start group [36].
min/1.73 m2 when symptoms were present. In addition, With the exception of Traynor et al. [36], none of the
these guidelines state that dialysis should be started above-mentioned studies has taken lead-time bias into
before the GFR drops below 6 ml/min/1.73 m2 [23]. The account. The lead-time is the period that late starters survive
European Best Practice Guidelines for PD recommend without dialysis, when compared with those who start ear-
consideration of initiation of dialysis when the GFR is lier. When this difference is not taken into account, late start-
below 15 ml/min/1.73 m2 in combination with evidence ers will always have a survival disadvantage. The first study
of poorly controlled conservative treatment. An absolute on the effect of lead-time bias was performed in 253 patients
minimum value was set at 6 ml/min/1.73 m2 [24]. from the cohort of the Netherlands Cooperative Study on
At the time these guidelines were published many the Adequacy of Dialysis (NECOSAD) [37]. The patients
patients typically started dialysis at much lower clearances. who started dialysis according to the US guidelines [22] sur-
In one study from the UK, the mean urea clearance was vived longer than those who started at a lower urea clear-
3.9 ml/min [25]. A study from the United States reported ance, but this difference was no longer present after
that 63% of the patients had an estimated GFR between 5 correction for the lead-time. This correction was performed
and 10 ml/min/1.73 m2, and 23% had an estimated value using a program developed for patients with a malignancy.
below 5 [26]. Although the aforementioned guidelines are The other published studies on effects of the lead-time on
not evidence based and are far from current practice, they survival used a simpler approach, that is, analyzing patients
have caused a trend toward earlier initiation of dialysis not from the time of dialysis initiation, but from a prede-
treatment [27, 28]. fined GFR, for instance 20 ml/min/1.73 m2 [36, 38–40]. No
Observational cohort studies on the initiation of dialysis difference in mortality was found after correction for the
suffer two potential problems. First, the method used for lead-time bias in the studies from Scotland [36] and the
GFR determination and, second, the effect of lead-time. United States [39]. A survival difference was present in the
Many cohort studies, often retrospective, used an estima- study from Hong Kong [38], but here a population starting
tion of glomerular filtration rate (eGFR) based on plasma dialysis with a GFR of 9 ml/min/1.73 m2 was compared to
creatinine, age, gender, and race, similar to the modifica- one with a lower GFR, in which dialysis was initiated acutely
tion of diet in renal disease (MDRD) formula [29]. All due to the presence of severe uremic symptoms. The study
investigations using the MDRD formula for eGFR reported from the Netherlands is interesting because effects of lead-
associations between a high MDRD and increased mortal- time correction were analyzed in a large cohort of patients
ity, despite corrections for possible confounders [30–32]. with GFR measurements (mGFR) and eGFR [40]. Results
The use of the reciprocal plasma creatinine value in these for eGFR without lead-time correction showed no difference
analyses, instead of the creatinine clearance, is the most between early and late starters, but after correction early
likely explanation because plasma creatinine is dependent starters had an increased risk of death. Without correction
on muscle mass and on glomerular filtration. The relative for lead-time, mGFR early starters had a somewhat lower
contribution of muscle mass to plasma creatinine will risk of death than late starters, but the difference disap-
therefore be higher in the presence of severely impaired peared completely after lead-time correction.
kidney function. Indeed a relationship between muscle The only randomized controlled trial comparing early ini-
mass and eGFR has been described in end-stage kidney tiation of dialysis with a late start based on GFR estima-
disease (ESKD) patients at the initiation of dialysis, while tions [41] confirmed the results of the observational cohort
follow-up confirmed the association between high eGFR studies discussed above, including when subgroups (e.g.
values and mortality [33]. The obvious solution to avoid or presence or absence of diabetes mellitus) were analyzed
separately. It can therefore be concluded that no fixed GFR investigators showed that patient survival was not different
value to initiate dialysis can be identified, but that the value between the two dialysis treatments during 3 years of fol-
of clinical assessment cannot be underestimated. The devel- low-up. Since then, numerous survival comparisons
opment of severe uremic complaints should be avoided. between HD and PD treatment have been published, but
their findings were inconsistent [43, 44]. Most studies per-
formed in the last decade in Western countries showed
D or PD as Initial Kidney
H either better survival for PD in the first period after the ini-
Replacement Therapy tiation of dialysis [45–49] or similar patient survival for
both treatment groups [50–53]. In contrast, a large study
Dialysis modality choice at the initiation of kidney replace- performed in Korea showed higher mortality rates in PD
ment therapy has been studied and discussed extensively in patients than in HD patients [54]. Figures 39.1 and 39.2
recent decades but still there is no final consensus on which show the dialysis “career” (including switches to other
treatment modality gives the best results. Consequently, modalities and death) of patients who started either HD or
both options have to be weighed in individual patients PD in Europe between 1993 and 2007.
according to their specific needs, preferences, and clinical There are several factors that could potentially explain
characteristics, with the aim of providing a patient-tailored the differences in findings between these studies, including
kidney replacement therapy. Important outcomes for the practice patterns, patient characteristics, and methodologi-
evaluation of initial dialysis modality are patient survival, cal issues.
QoL, and values of biochemical parameters.
Practice Patterns
Patient Survival Differences in dialysis modality-specific practice patterns
between continents or countries may influence modality-
The first study comparing continuous ambulatory PD with specific patient survival. These practice patterns include
HD in incident patients was performed in the UK more general factors such as access to healthcare and organiza-
than 30 years ago [42]. Using Kaplan–Meier analyses, the tion of the healthcare system, but also more specific factors
1993–1997 1998–2002 2003–2007

1.0 1.0 1.0
Death Death Death
0.8 0.8 0.8
49.0% 52.3% 53.1%
0.6 0.6 0.6
HD HD HD
0.4 23.8% 0.4 23.6% 0.4
PD 25.9%
3.5% PD PD
0.2 0.2 3.0% 0.2 3.0%
Transplantation 23.7% Transplantation 21.1% Transplantation 18.0%
0.0 0.0 0.0
0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5
Patient years Patient years Patient years
Figure 39.1 Unadjusted cumulative incidence survival curves for transfers to peritoneal dialysis or transplantation and death for
patients who started hemodialysis in 1993–1997, 1998–2002, and 2003–2007. HD, hemodialysis; PD, peritoneal dialysis.
Source: Van de Luijtgaarden et al. [49] © 2016, Oxford University Press.
1993–1997 1998–2002 2003–2007

1.0 1.0 1.0
0.8 Death 30.6% 0.8 Death 29.8% 0.8 Death 27.6%
PD 7.9% 0.6 PD 8.3% 0.6 PD 8.6%

0.6
0.4 HD 30.8% 0.4 HD 32.5% 0.4 HD 33.9%
0.2 0.2 0.2
Transplantation 30.7% Transplantation 29.4% Transplantation 29.9%
0.0 0.0 0.0
0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5
Patient years Patient years Patient years
Figure 39.2 Unadjusted cumulative incidence survival curves for transfers to hemodialysis or transplantation and death for European
patients who started peritoneal dialysis in 1993–1997, 1998–2002, and 2003–2007. HD, hemodialysis; PD, peritoneal dialysis.
Source: Van de Luijtgaarden et al. [49] © 2016, Oxford University Press.
Patient Surviva 675
that may contribute to the efficiency and quality of the Patient Characteristics
dialysis offered, such as the HD treatment time and vascu-
Several studies from different parts of the world that found
lar access type, peritoneal catheters, and the dialysis fluids
overall similar effects for PD and HD or a beneficial effect
used. However, the results provided in most comparative
of PD over HD also showed that patient survival on PD was
studies do not allow evaluation of the extent to which HD
worse in specific subgroups of patients: those who were
and PD were performed in a state-of the-art manner. For
diabetic and who were older (mostly over 55 years). For
example, in many survival comparisons the type of vascu-
example, investigators from the Canadian Organ
lar access used for HD is not included in the analyses,
Replacement Register who included data on more than
whereas it was shown in a Canadian cohort that the type
35 000 incident dialysis patients found that overall adjusted
of vascular access plays an important role in the relation-
survival remained similar for HD and PD, even in the most
ship between dialysis modality and mortality [55]. Starting
contemporary cohort [60]. After stratification for diabetic
HD with a central venous catheter largely explained the
status, age, and sex, they found that nondiabetic patients in
higher early mortality risk of HD in this study. Another
the youngest age group (<45 years) showed survival bene-
example is the omission of the type of dialysis fluid in the
fits on PD treatment, while there was no difference in mor-
comparison of patient survival. Based on data from the
tality in the older age groups. However, survival on PD was
Taiwan health insurance database, it was demonstrated
worse than on HD among patients with diabetes, in par-
that the inclusion of one icodextrin-based PD fluid in the
ticular in those with higher age, which confirms the find-
usual PD schedule could attenuate the survival disadvan-
ings from previous studies from the United States, Asia,
tage for PD relative to HD in diabetic patients, particularly
and Europe [48, 50, 51, 57, 61, 62].
for those who were elderly [56].
In addition, patient characteristics related to ethnicity
An additional factor that is usually not taken into
could partly explain different study findings across popula-
account in survival comparisons of HD and PD is whether
tions. In the United States, it was shown that dialysis
the start of dialysis was “planned” or “unplanned.” It has
patients who are Hispanic have better survival rates com-
been postulated that patients who need to start dialysis
pared to those who are white, across all age categories [63].
urgently are at a high risk of death, and as they are treated
This survival advantage is also present in African-American
predominantly with HD this could induce selection bias
patients, although it is limited to patients older than
into the comparison of mortality between HD and PD
40 years of age.
patients. A French cohort study showed there was a sig-
To illustrate the variation in findings from different parts
nificant increase (50%) in mortality risk among elderly
of the world, the results of a selection of large renal registry
patients ( 75 years) with an unplanned start of HD when
studies comparing patient survival between HD and PD
compared to those with a planned start, which suggests a
treatment are summarized in Table 39.2.
comparison between the two dialysis modalities would be
more balanced after removing the unplanned HD
starts [57]. Two Canadian studies confirmed these find-
Methodological Issues
ings. In 2011, Quinn et al. showed that HD and PD were
associated with similar survival in incident patients start- Ideally, the decision on which dialysis modality gives the
ing dialysis electively as outpatients [52]. More recently, best outcomes should be based on results of randomized
Wong et al. reported that HD and PD are associated with controlled trials in which the allocation of the dialysis
similar mortality among incident dialysis patients who are modality is not influenced by attitudes or preferences of
eligible for both modalities [53]. They claim that to better the nephrologist and the patient. The first randomized con-
reflect the outcomes for patients who have the opportunity trolled trial comparing HD and PD as initial kidney replace-
to choose between HD and PD in clinical practice, future ment therapy was conducted in the Netherlands between
comparisons of dialysis modality should be restricted to 1997 and 2000 [64]. However, pre-existing opinions held by
patients who are deemed eligible for both modalities. patients and physicians on the best treatment option were
Finally, most recently more and more attention has been often reason for not including patients in the trial. As a
paid to home-based therapies, including both HD and PD. result, the study was underpowered (only 38 patients could
Data from the Australia and New Zealand Dialysis and be included) and thus unsuccessful. Despite this result, a
Transplant Registry (ANZDATA) demonstrated that out- new trial started in China in 2011 (trial registration
comes on dialysis therapy are improving with time, and NCT01413074, http://clinicaltrials.gov); 5 years later, it
this improvement is most marked with home dialysis also was terminated. Consequently, outcomes in HD and
modalities, especially home HD, when compared to con- PD patients can only be compared based on large-scale
ventional in-center HD [58, 59]. observational studies.
Table 39.2 Comparison of mortality between HD and PD patients: a selection of recent large studies based on renal registry data.
Author, publication year

(reference) Country/region Inclusion period N Effect
Asia
Huang, 2008 [51] Taiwan Renal Registry 1995–2002 48 629 Overall similar survival. PD better among
nondiabetic patients and those 55 years; HD
better among diabetic patients and those
>55 years
Kim, 2014 [54] Korean Health Insurance 2005–2008 32 280 Overall higher mortality on PD
Review & Assessment HD better among patients 55 years in all
Service database subgroups except those with no comorbidities
and malignancy (similar survival regardless
of age)
Europe
Van de Luijtgaarden, European Renal 1998–2006 196 076 Overall PD better in first years after starting
2016 [49] Association-European dialysis
Dialysis and Transplant Similar survival for diabetic patients and
Association (ERA-EDTA) those 65 years
Registry
North America
Mehrotra, 2011 [50] United States Renal Data 1996–2004 624 426 Overall similar survival in the most recent
System (USRDS) cohorts
Over time greater improvement in survival
among PD patients relative to HD at all
follow-up periods
Yeates, 2012 [60] Canadian Organ 1991–2004 46 839 Overall similar survival; PD better in first
Replacement Register 18 months, HD better after 36 months
(CORR) Women with diabetes aged 65 years higher
risk on PD
Oceania
McDonald, 2009 [46] The Australia and New 1991–2005 27 015 Overall PD better in the first 12 months after
Zealand Dialysis and starting dialysis, thereafter HD better
Transplant Registry Younger patients without comorbidities had
(ANZDATA) mortality advantage with PD treatment, but
other groups did not
N, number; HD, hemodialysis; PD, peritoneal dialysis.
An important issue regarding treatment comparisons is marginal structural models (e.g. in [46, 47]), and the use of
the difficulty of making causal inference based on observa- treatment propensity scores in statistical models by means
tional studies. The most important weakness of observa- of adjustment, stratification, or matching (e.g. in [46, 49,
tional studies is that selection bias by the clinician – also 59, 62]). The fact that so many different statistical methods
called confounding by indication – may occur. There are are being applied can partly explain the inconsistency in
several strategies to reduce the influence of such selection study results.
bias. Of these, multivariable adjustment for potential con- Mehrotra et al. applied several of these statistical
founders during statistical analysis is most commonly methods to compare the survival of HD and PD patients
applied. In recent years, more and more research groups from the United States Renal Data System (USRDS) [50].
have started to apply advanced statistical methods in addi- They concluded that in the most recent cohort (2002–2004),
tion to the conventional methods of survival analysis (i.e. there was no significant difference in the risk of death
Kaplan–Meier and standard Cox proportional hazards between patients who started on HD or PD. These findings
models) to assess the associations between dialysis modal- were confirmed by investigators from the Canadian Organ
ity and mortality risk [44]. Such methods include Replacement Register who used a similar approach in com-
time-dependent Cox regression models (e.g. in [46, 49]), paring the data of their incident dialysis patient cohort [60].
Reference 677
uality of Life and Biochemical

Q patients who were treated with PD before their transplan-
Parameters tation had a significantly lower incidence of delayed graft
function and a better 5-year patient survival rate than those
Although patient survival is an essential outcome, there treated with HD before transplantation; there were no dif-
are several other important outcomes that can and should ferences in the occurrence of acute rejection and graft
be considered in the decision on which dialysis modality survival [71].
to choose. Many studies comparing all kinds of clinical
outcomes between HD and PD have been performed.
First of all, QoL is probably the most important outcome Conclusion
from the patient’s perspective. However, only a few recent
studies have compared the QoL between HD and PD Both for establishment of the optimal starting point for
patients. A meta-analysis published in 2017, in which dialysis treatment and for dialysis modality comparison,
only seven individual studies could be included, has not the evidence base has needed to come from observational
led to a unanimous conclusion [65]. Only some of the studies, despite their drawbacks, because as yet it has not
seven articles found significant differences between the been possible to perform randomized controlled trials. Up
two modalities; one showed a better QoL for PD patients, to now, these observational studies have indicated that no
while two other studies found the best QoL in patients fixed GFR value to start dialysis can be identified. However,
receiving HD. the value of clinical assessment cannot be underestimated
In addition to QoL, there are many other outcomes that and the development of severe uremic complaints should
can be considered in the choice between the two dialysis be avoided.
modalities. One such outcome is the occurrence of renal Overall, the treatment outcomes for HD and PD are very
anemia. A recent meta-analysis showed that both HD and similar, even when sophisticated statistical methods are
PD have a similar effect on renal anemia in ESKD patients; used. When specific patient subgroups were investigated,
the authors found no significant differences for levels of most studies showed that PD is associated with better sur-
hemoglobin, ferritin, and transferrin saturation index vival for younger (mostly <55 years) and nondiabetic
between the two dialysis modalities [66]. A number of stud- patients, but with worse survival for older and diabetic
ies have shown advantages of PD, which were especially patient subgroups. These results indicate that both HD and
marked in the first years of treatment. These advantages PD can be considered as suitable initial kidney replace-
include a better preservation of residual GFR [67–69] and ment therapy, depending on the patients’ characteristics.
better results after kidney transplantation with regard to Patients should be educated about potential advantages
delayed graft function [70, 71]. A systematic review and and disadvantages of both treatment modalities before the
meta-analysis on the association between pre-transplant start of dialysis, so that they can make a well-informed
dialysis modality and transplant outcomes showed that choice together with their physician.
References
KDOQI Clinical Practice Guideline for Hemodialysis

1 6 Innes, A., Rowe, P.A., Burden, R.P., and Morgan, A.G.
Adequacy (2015). 2015 update. Am. J. Kidney Dis. 66: (1992). Early deaths on renal replacement therapy: the
884–930. need for an early nephrological referral. Nephrol. Dial.
2 Bonomini, V., Albertazzi, A., Vangelista, A. et al. (1976). Transplant. 7: 467–471.
Residual renal function and effective rehabilitation in 7 Jungers, P., Zingraff, J., Albouze, G. et al. (1993). Late
chronic dialysis. Nephron 16: 89–99. referral to maintenance dialysis: detrimental
3 Bonomini, V., Feletti, C., Scolari, M.P., and Stefoni, S. consequences. Nephrol. Dial. Transplant. 8:
(1985). Benefits of early initiation of dialysis. Kidney Int. 28 1089–1093.
(Suppl 17): S57–S59. 8 Khan, I.H., Catto, G.R.D., Edward, N., and MacLeod, A.M.
4 Ratcliffe, P.J., Phillips, R.E., and Oliver, D.O. (1984). Late (1995). Death during the first 90 days of dialysis: a
referral for maintenance dialysis. Br. Med. J. 288: 441–443. case-control study. Am. J. Kidney Dis. 25: 276–280.
5 Campbell, J.D., Ewigman, B., Hosokawa, M., and Van 9 Sesso, R. and Belusco, A.G. (1996). Late diagnosis of
Stone, J.C. (1989). The timing of referral of patients with chronic renal failure and mortality on maintenance
end-stage renal disease. Dial. Transplant. 18: 660–668. dialysis. Nephrol. Dial. Transplant. 11: 2417–2420.
0 Kinchen, K.S., Sadler, J., Fink, N. et al. (2002). The timing

1 24 EBPG Expert Group on Peritoneal Dialysis (2005).
of specialist evaluation in chronic kidney disease and European best practice guidelines for peritoneal dialysis.
mortality. Ann. Intern. Med. 137: 479–486. Nephrol. Dial. Transplant. 20 (Suppl 9): ix3–ix7.
11 Khan, S.S., Xue, J.L., Kazmi, W.H. et al. (2005). Does 25 Tattersall, J., Greenwood, R., and Farrington, K. (1995).
predialysis nephrology care influence patient survival Urea kinetics and when to commence dialysis. Am. J.
after initiation of dialysis? Kidney Int. 67: 1038–1046. Nephrol. 15: 283–289.
12 Dogan, E., Erkoc, R., Sayarlioglu, H. et al. (2005). Effects 26 Obrador, G.T., Arora, P., Kausz, A.T. et al. (1999). Level of
of late referral to a nephrologist in patients with chronic renal function at the initiation of dialysis in the US
renal failure. Nephrology 10: 516–519. end-stage population. Kidney Int. 56: 2227–2235.
13 Sesso, R. and Yoshihiro, M.M. (1997). Time of diagnosis 27 Obrador, G.T. and Pereira, B.J. (1998). Early referral to
of chronic renal failure and assessment of quality of life the nephrologist and timely initiation of renal
in haemodialysis patients. Nephrol. Dial. Transplant. 12: replacement therapy: a paradigm shift in the
2111–2116. management of patients with chronic renal failure. Am.
14 Caskey, F.J., Wordsworth, S., Ben, T. et al. (2003). Early J. Kidney Dis. 31: 348–417.
referral and planned initiation of dialysis: what impact 28 Termorshuizen, F., Korevaar, J.C., Dekker, F.W. et al.
on quality of life? Nephrol. Dial. Transplant. 18: (2003). Time trends in initiation and dose of dialysis in
1330–1338. end-stage renal disease patients in The Netherlands.
15 Jager, K.J., Korevaar, J.C., Dekker, F.W. et al. (2004). The Nephrol. Dial. Transplant. 18: 552–558.
effect of contraindications and patient preference on 29 Levy, A.S., Bosch, J.P., Lewis, J.B. et al. (1999). A more
dialysis modality selection in ESRD patients in The accurate method to estimate glomerular filtration rate
Netherlands. Am. J. Kidney Dis. 43: 891–899. from serum creatinine: a new prediction equation.
16 Owen, W.F. Jr. (2003). Patterns of care for patients with Modification of diet in Renal Disease Study Group. Ann.
chronic kidney disease in the United States: dying for Intern. Med. 130: 461–470.
improvement. J. Am. Soc. Nephrol. 14 (Suppl 2): 30 Beddu, S., Samore, M.H., Roberts, M.S. et al. (2003).
S76–S80. Impact of timing of dialysis on mortality. J. Am. Soc.
17 Lameire, N., van Biesen, W., Dombros, N. et al. (1997). Nephrol. 14: 2305–2312.
The referral pattern of patients with ESRD is a 31 Fink, J.C., Burdick, R.A., Kurth, S.J. et al. (1999).
determinant in the choice of dialysis modality. Perit. Dial. Significance of serum creatinine values in new end-stage
Int. 17 (Suppl 3): S161–S166. renal disease patients. Am. J. Kidney Dis. 34: 694–701.
18 Tattersall, J., Dekker, F., Heimbürger, O. et al. (2011). 32 Stel, V.S., Dekker, F.W., Ansell, D. et al. (2009). Residual
When to start dialysis: updated guidance following renal function at the start of dialysis and clinical
publication of the Initiating Dialysis Early and Late outcomes. Nephrol. Dial. Transplant. 24: 3175–3182.
(IDEAL) study. Nephrol. Dial. Transplant. 26: 2082–2086. 33 Grootendorst, D.C., Michels, W.M., Richardson, J.D. et al.
19 Kidney Disease: Improving Global Outcomes (KDIGO) (2011). The MDRD formula does not reflect GFR in ESRD
CKD Work Group (2013). KDIGO 2012 clinical practice patients. Nephrol. Dial. Transplant. 26: 1932–1937.
guideline for the evaluation and management of chronic 34 Cockcroft, D. and Gault, M. (1976). Prediction of
kidney disease. Kidney Int. Suppl. 3: 1–150. creatinine clearance from serum creatinine. Nephron 16:
20 Nesrallah, G.E., Mustafa, R.A., Clark, W.F. et al. (2014). 31–41.
Canadian Society of Nephrology 2014 clinical practice 35 Biesenbach, G., Hubmann, K., Janko, O. et al. (2002).
guideline for timing the initiation of chronic dialysis. Predialysis management and predictors for early
CMAJ 186: 112–117. mortality in uremic patients who die within one year
21 Watanabe, Y., Yamagata, K., Nishi, S. et al. (2015). after initiation of dialysis therapy. Ren. Fail. 24: 197–205.
Japanese society for dialysis therapy clinical guideline for 36 Traynor, J.P., Simpson, K., Geddes, C.C. et al. (2002).
“hemodialysis initiation for maintenance hemodialysis”. Early initiation of dialysis fails to prolong survival in
Ther. Apher. Dial. 19 (Suppl 1): 93–107. patients with end-stage renal failure. J. Am. Soc. Nephrol.
22 NKF (1997). K/DOQI clinical practice guidelines for 13: 2125–2132.
hemodialysis and peritoneal dialysis adequacy. Am. J. 37 Korevaar, J.C., Jansen, M.A., Dekker, F.W. et al. (2001).
Kidney Dis. 30 (Suppl 2): S67–S136. When to initiate dialysis: effect of proposed US guidelines
23 Churchill, D.N., Blake, P.G., Goldstein, M.B. et al. (1999). on survival. Lancet 358: 1046–1050.
Clinical practice guidelines of the Canadian Society of 38 Tang, S.C.W., Ho, Y.W., Tang, A.W.C. et al. (2007).
Nephrology for treatment of patients with chronic renal Delaying initiation of dialysis till symptomatic uraemia-is
failure. J. Am. Soc. Nephrol. 10 (Suppl 13): S287–S321. it too late? Nephrol. Dial. Transplant. 22: 1926–1932.
Reference 679
9 Crews, D.C., Scialla, J.J., Boulware, L.E. et al. (2014).

3 52 Quinn, R.R., Hux, J.E., Oliver, M.J. et al. (2011).
Comparative effectiveness of early versus conventional Selection bias explains apparent differential mortality
timing of dialysis initiation in advanced CKD. Am. J. between dialysis modalities. J. Am. Soc. Nephrol. 22:
Kidney Dis. 63: 806–815. 1534–1542.
40 Janmaat, C.J., van Diepen, M., Krediet, R.T. et al. (2017). 53 Wong, B., Ravani, P., Oliver, M.H. et al. (2018).
Effect of glomerular filtration rate at dialysis initiation on Comparison of patient survival between hemodialysis
survival in patients with advanced chronic kidney and peritoneal dialysis among patients eligible for both
disease: what is the effect of lead-time bias? Clin. modalities. Am. J. Kidney Dis. 71: 344–351.
Epidemiol. 9: 217–230. 54 Kim, H., Kim, K.H., Park, K. et al. (2014). A population-
41 Cooper, B.A., Branly, P., Bulfone, L. et al. (2010). A based approach indicates an overall higher patient
randomized controlled trial of early versus late initiation mortality with peritoneal dialysis compared to
of dialysis. N. Engl. J. Med. 363: 609–619. hemodialysis in Korea. Kidney Int. 86: 991–1000.
42 Gokal, R., Jakubowski, C., King, J. et al. (1987). Outcome 55 Perl, J., Wald, R., McFarlane, P. et al. (2011).
in patients on continuous ambulatory peritoneal dialysis Hemodialysis vascular access modifies the association
and haemodialysis: a 4-year analysis of a prospective between dialysis modality and survival. J. Am. Soc.
multicentre study. Lancet ii: 1105–1108. Nephrol. 22: 1113–1121.
43 Vonesh EF, Snyder JJ, Foley RN, Collins AJ. Mortality 56 Wang, I.-K., Lind, C.-L., Yen, T.-H. et al. (2018).
studies comparing peritoneal dialysis and hemodialysis: Comparison of survival between hemodialysis and
what do they tell us? Kidney Int. Suppl. 2006;103:S3–S11. peritoneal dialysis patients with end-stage renal disease
44 Noordzij, M. and Jager, K.J. (2012). Survival comparisons in the era of icodextrin treatment. Eur. J. Int. Med. 50:
between haemodialysis and peritoneal dialysis. Nephrol. 69–74.
Dial. Transplant. 27: 3385–3387. 57 Couchoud, C., Moranne, O., Frimat, L. et al. (2007).
45 Termorshuizen, F., Korevaar, J.C., Dekker, F.W. et al. Associations between comorbidities, treatment choice
(2003). Hemodialysis and peritoneal dialysis: comparison and outcome in the elderly with end-stage renal disease.
of adjusted mortality rates according to the duration of Nephrol. Dial. Transplant. 22: 3246–3254.
dialysis: analysis of The Netherlands Cooperative Study 58 Marshall, M.R., Polkinghorne, K.R., Kerr, P.G. et al.
on the Adequacy of Dialysis 2. J. Am. Soc. Nephrol. 14: (2015). Temporal changes in mortality risk by dialysis
2851–2860. modality in the Australian and New Zealand dialysis
46 McDonald, S.P., Marshall, M.R., Johnson, D.W., and population. Am. J. Kidney Dis. 66: 489–498.
Polkinghorne, K.R. (2009). Relationship between 59 Nadeau-Fredette, A.-C., Hawley, C.M., Pascoe, E.M. et al.
dialysis modality and mortality. J. Am. Soc. Nephrol. 20: (2015). An incident cohort study comparing survival on
155–163. home hemodialysis and peritoneal dialysis (Australia and
47 van der Wal, W.M., Noordzij, M., Dekker, F.W. et al. New Zealand Dialysis and Transplantation Registry).
(2010). Comparing mortality in renal patients on Clin. J. Am. Soc. Nephrol. 10: 1397–1407.
hemodialysis versus peritoneal dialysis using a marginal 60 Yeates, K., Zhu, N., Vonesh, E. et al. (2012). Hemodialysis
structural model. Int. J. Biostat. 6 (1): article 2. and peritoneal dialysis are associated with similar
48 Weinhandl, E.D., Foley, R.N., Gilbertson, D.T. et al. outcomes for end-stage renal disease treatment in
(2010). Propensity-matched mortality comparison of Canada. Nephrol. Dial. Transplant. 27: 3568–3575.
incident hemodialysis and peritoneal dialysis patients. J. 61 Van de Luijtgaarden, M.W., Noordzij, M., Stel, V.S. et al.
Am. Soc. Nephrol. 21: 499–506. (2011). Effects of comorbid and demographic factors on
49 van de Luijtgaarden, M.W., Jager, K.J., Segelmark, M. dialysis modality choice and related patient survival in
et al. (2016). Trends in dialysis modality choice and Europe. Nephrol. Dial. Transplant. 26: 2940–2947.
related patient survival in the ERA-EDTA Registry over a 62 Winkelmayer, W.C., Glynn, R.J., Mittleman, M.A. et al.
20-year period. Nephrol. Dial. Transplant. 31: 120–128. (2002). Comparing mortality of elderly patients on
50 Mehrotra, R., Chiu, Y.W., Kalantar-Zadeh, K. et al. (2011). hemodialysis versus peritoneal dialysis: a propensity
Similar outcomes with hemodialysis and peritoneal score approach. J. Am. Soc. Nephrol. 13: 2353–2362.
dialysis in patients with end-stage renal disease. Arch. 63 Rhee, C.M., Lertdumrongluk, P., Streja, E. et al. (2014).
Intern. Med. 171: 110–118. Impact of age, race and ethnicity on dialysis patient
51 Huang, C.C., Cheng, K.-F., and Wu, H.-D.I. (2008). survival and kidney transplantation disparities. Am. J.
Survival analysis: comparing peritoneal dialysis and Nephrol. 39: 183–194.
hemodialysis in Taiwan. Perit. Dial. Int. 28 (Suppl 3): 64 Korevaar, J.C., Feith, G.W., Dekker, F.W. et al. (2003).
S15–S20. Effect of starting with hemodialysis compared with
peritoneal dialysis in patients new on dialysis treatment: 68 Misra, M., Vonesh, E., van Stone, J.C. et al. (2001). Effect
a randomized controlled trial. Kidney Int. 64: 2222–2228. of cause and time of dropout on the residual GFR: a
65 Zazzeroni, L., Pasquinellia, G., Nannib, E. et al. (2017). comparative analysis of the decline of GFR on dialysis.
Comparison of quality of life in patients undergoing Kidney Int. 59: 754–763.
hemodialysis and peritoneal dialysis: a systematic 69 Jansen, M.A.M., Hart, A.A.M., Korevaar, J.C. et al. (2002).
review and meta-analysis. Kidney Blood Press. Res. 42: Predictors of the rate of decline of residual renal function
717–727. in incident dialysis patients. Kidney Int. 62: 1046–1053.
6 Wang, W.-N., Zhang, W.-L., Sun, T. et al. (2017). Effect of
6 70 Van Biesen, W., Vanholder, R., Van Loo, A. et al. (2000).
peritoneal dialysis versus hemodialysis on renal anemia Peritoneal dialysis favorably influences early graft
in renal in end-stage disease patients: a meta-analysis. function after renal transplantation compared to
Ren. Fail. 39: 59–66. hemodialysis. Transplantation 69: 508–514.
67 Moist, L.M., Port, F.K., Orol, S.M. et al. (2000). Predictors 71 Tang, M., Li, T., and Liu, H. (2016). A comparison of
of loss of residual renal function among new dialysis transplant outcomes in peritoneal and hemodialysis
patients. J. Am. Soc. Nephrol. 11: 556–564. patients: a meta-analysis. Blood Purif. 42: 170–176.
681
Index
Note: Page numbers in italics refer to Figures, those in bold refer to Tables.
1‐alpha‐hydroxy‐vitamin ABCD study see Appropriate Blood ACTH see adrenocorticotropic

D2 603 (Box) Pressure Control in Diabetes hormone
1‐hydroxy‐vitamin D3 603 (Box) (ABCD) normotensive study Action in Diabetes and Vascular
1,25‐dihydroxy‐vitamin D3 ACCOMPLISH trial see Avoiding Disease: Preterax and
603 (Box) Cardiovascular events Diamicron MR Controlled
19‐nor‐1,25‐dihydroxy‐vitamin D2 through Combination Evaluation (ADVANCE)
603, 603 (Box) therapy in Patients Living study 373, 374, 382
22‐oxacalcitriol 603 with Systolic Hypertension Action to Control Cardiovascular Risk
24,25‐dihydroxy‐vitamin D3 (ACCOMPLISH) trial in Diabetes (ACCORD)
603 (Box) ACCORD BP trial see Action to trial 332, 340, 373, 374,
25‐hydroxy‐vitamin D3 603 (Box) Control Cardiovascular Risk 560–61
26,27‐hexfluoro‐calcitriol 603, in Diabetes‐Blood Pressure Action to Control Cardiovascular Risk
603 (Box) (ACCORD‐BP) trial in Diabetes‐Blood Pressure
ACCORD trial see Action to Control (ACCORD‐BP) trial 341,
a Cardiovascular Risk in 382, 581, 583
A MAstricht Contrast‐Associated Diabetes (ACCORD) trial Acute Dialysis Quality Initiative
Nephropathy Guideline ACCORDION study 373 (ADQI) 87, 88, 193
(AMACING) trial 152 A‐C‐D drugs 337 acute glomerulonephritis 87,
A Proliferating Inducing Ligand acetazolamide 175 98, 455
(APRIL) 404 acetylcysteine 148, 360 acute interstitial nephritis
AASK study see African American Acetylcysteine for Contrast clinical features 163–4, 165–7
Study of Kidney Disease Nephropathy (ACT) trial 152 course and treatment 168–9
and Hypertension ACHIEVE see Aldosterone bloCkade definition 163
(AASK) study for Health Improvement etiology 163, 164
abatacept EValuation in End‐stage laboratory diagnosis 164–8, 168
in focal segmental Renal Disease (ACHIEVE) pathogenesis 163
glomerulosclerosis acid‐base disturbance correction in acute kidney disease (AKD) 85–91
(FGS) 225 acute tubular necrosis 136 biomarkers 88–90, 89, 91
in lupus nephritis 417 ACT trial see Acetylcysteine for classification 87–88, 87
in minimal change disease 219, Contrast Nephropathy urine cast types with typical
222, 225 (ACT) trial associations 88
bindex1.indd 681 09-12-2022 17:14:43

682 Index
acute kidney disease (AKD) (cont’d) physical examination 127 African Americans 18
urine microscopy scoring prevention 129–35, 130–2, 133–4 APOL‐1 gene mutations 9, 77
systems 88 aminoglycosides 129 atheroembolic renal disease 170
epidemiology 90 amphotericin B 129–35 blood pressure in 581, 583
limitations of SCr to define 88 lipid‐based formulations 135 CKD in 77, 78
overview/definition 85–6, 86 volume expansion (salt dyslipidemia in 550
staging system for 86 loading) 135 ESKD 75–7, 420
acute kidney injury (AKI) 3, 85–91 iodinated radiocontrast dye 135 FSGS in 216
avoidance of 297, 298 piperacillin‐tazobactam 135 IgAN in 254
kidney failure and 33 vancomycin 135 kidney disease risk 9
in pregnancy 445 prognosis 128–9 lupus nephritis 420
as risk factor of chronic kidney mortality after 128–9 masked hypertension in 334
disease 7 renal functional reserve mycophenolate mofetil in
STEC‐HUS and 434 (RFR) and 129 412–13
see also acute tubular renal recovery after 128 patient survival 675
necrosis (ATN) septic ATN 126 screening 55, 56
Acute Kidney Injury‐Epidemiologic treatment 136, 137–8 SLE in 400, 401
Prospective Investigation urine microscopy 127 age
(AKI‐EPI) study 128 urine sodium concentration and autosomal dominant polycystic
Acute Kidney Injury Network fractional excretion of kidney disease (ADPKD)
(AKIN) 86, 190 sodium (FENa) 127 and 288–301
acute renal failure see acute kidney acute tubulo‐interstitial nephritis, chronic kidney disease and 8–9
injury (AKI) NSAIDS and 322 progression to kidney failure
Acute Renal Failure Trial Network acyclovir 126, 175, 178 and 33–4
(ATN) study 195 adalimumab in focal segmental AIM‐HIGH study see
acute renal impairment, NSAIDS glomerulosclerosis Atherothrombosis
and 321–2 (FGS) 222, 225 Intervention in Metabolic
acute respiratory distress syndrome ADQI see Acute Dialysis Quality Syndrome with Low HDL/
(ARDS) 97, 98, 99 Initiative (ADQI) High Triglycerides and
acute tubular injury (ATI) see acute adrenocorticotropic hormone (ACTH) Impact on Global Health
tubular necrosis (ATN) 125 in FSHS 224 Outcomes (AIM‐
acute tubular necrosis (ATN) 96, in membranous nephropathy HIGH) study
123–38, 123, 124 247, 248 AIN see acute interstitial
definition 124 in minimal change nephritis (AIN)
diagnosis 127–8 disease 220, 221 AKI‐EPI study see Acute Kidney
epidemiology 124 Adult Treatment Panel III 549 Injury‐Epidemiologic
histology 124–5 ADVANCE trial see Action in Diabetes Prospective Investigation
history 127 and Vascular Disease: (AKI‐EPI) study
introduction 123–4 Preterax and Diamicron MR AKIKI trial see Artificial Kidney
ischemic ATN 125–6 Controlled Evaluation Initiation in Kidney Injury
phases of 125 (ADVANCE) (AKIKI) trial
renal blood supply 125–6 ADVANCE‐ON study 373, 379 AKIN see Acute Kidney Injury
tubular factors 125 adynamic bone disease 591 Network (AKIN)
nephrotoxic ATN 126, 126 African American Study of Kidney Alberta Kidney Disease
novel biomarkers 127–8 Disease and Hypertension Network 549
NSAIDS and 321–2 (AASK) study 61, 340, 550, albumin 111–12, 631
pathophysiology 124–6 581, 583 dialysis 116
bindex1.indd 682 09-12-2022 17:14:43

Index 683
albumin‐to‐creatinine ratio (ACR) allopurinol 174 cyclophosphamide 467–8,

4, 28 ALMS Group study see Aspreva Lupus 467, 468
albuminuria Management Study (ALMS) definitions 461, 461
AKI and 7 Group study diagnostic tests 465, 465
CKD and 26, 27 alogliptin 377 epidemiology 462–4
definition of chronic kidney ALTITUDE 380 glucocorticoids 468
disease 3, 4 AMACING trial see A MAstricht hematological tests 465‐6
ESKD and 6 Contrast‐Associated histological tests 466
kidney failure and 28–33 Nephropathy Guideline imaging 466
mortality and 5 (AMACING) trial indirect immunofluorescence 466
screening for chronic kidney ambulatory blood pressure induction agents 470
disease 45 monitoring (ABPM) 332, pathogenesis 464
see also African American Study of 333, 334, 343 physiological testing 466
Kidney Disease and Ambulatory Blood Pressure plasma exchange 468‐9, 469
Hypertension; HOPE trial Monitoring and prognosis 472–3
ALCHEMIST see ALdosterone Cardiovascular Events remission induction 467
Antagonist Chronic (MAPEC) study 333 remission maintenance 470,
HEModialysis Interventional American Academy of Family 471, 472
Survival Trial Physicians 49–50 rituximab 469–70, 469, 470
(ALCHEMIST) American College of Rheumatology treatment 463, 466‐7, 467
alcohol, hypertension and 337 (ACR) 400 treatment of relapse 472
ALdosterone Antagonist Chronic American Heart Association/ anemia in chronic kidney disease
HEModialysis Interventional American College of 542–47, 543
Survival Trial Cardiology (ACC/AHA) clinical outcomes in randomized
(ALCHEMIST) 346 guidelines 549 controlled trials 542–44
Aldosterone bloCkade for Health American Society of Nephrology 50 blood transfusions 544
Improvement EValuation in aminoglycosides cardiovascular events 544
End‐stage Renal Disease in acute tubular necrosis 129 harms vs. benefits 544
(ACHIEVE) 346 in CKD 537 Hb levels 542
alemtuzumab 470 amiodarone 488 quality of life 544
alendronate 607 (Box) amlodipine 323, 337, 339, 345, erythropoiesis‐ stimulating agents
ALERT trial see Assessment of LEscol 380, 381 544–6
in Renal Transplantation amoxycillin/clavulanate in UTI 307 hyporesponsiveness 545–6
(ALERT) trial amphotericin B in acute tubular meta‐analyses 546
alfacalcidol 603 (Box) necrosis 129–35 moderate anemia 544–5
alirocumab 382, 560 lipid‐based formulations 135 Canada–Europe Study 545
aliskiren 338 volume expansion (salt loading) 135 cardiovascular risk reduction
alkaline phosphatase, bone and 594 amyloid 205 in early anemia treatment
alkylating agents ANCA‐associated vasculitis 430, with epoetin study 545
in focal segmental 461–73 hemoglobin correction and
glomerulosclerosis 224 ANCA testing 466 renal insufficiency 545
in minimal change disease 221, antibiotics 471 normal hematocrit study 544
221 (Box) antigen specific tests 466 trial to reduce cardiovascular
ALLHAT see Antihypertensive and azathioprine 471 events with aransep
Lipid‐Lowering Treatment biochemical tests 465–6 therapy 545
to Prevent Heart Attack Trial clinical features 464–5 severe anemia 544
(ALLHAT) co‐trimoxazole 472 intravenous iron 546–7
bindex1.indd 683 09-12-2022 17:14:43

684 Index
anemia in chronic kidney disease adult 305, 306, 307, 308, 309 aristolochic acid 75
(cont’d) children 307, 308 arsenic 75
peginesatide 546 anti‐CD40 antibody in focal segmental arterial thromboembolism
prolyl hydroxylase inhibitors 547 glomerulosclerosis (FGS) (ATE) 220
anemia in STEC‐HUS and 434 217–18 arteriovenous fistula (AVF) 66
angioplasty 354 anticoagulants in hypertension 340
in atherosclerotic renovascular in CVD 517, 519–20 Artificial Kidney Initiation in Kidney
disease 357 in renal biopsy 209 Injury (AKIKI) trial 190,
benefits/ disadvantages of 355, anticoagulation strategies 198 191, 192, 193
356, 360, 576 antigen specific tests in ANCA‐ aspirin
angiotensin converting enzyme associated vasculitis 466 in cardiovascular risk 206,
inhibitors (ACEi) 56 Antihypertensive and Lipid‐Lowering 371, 382–3
blood pressure control 296, 519 Treatment to Prevent Heart in pregnancy 419, 452, 453
CA‐AKI and 154 Attack Trial (ALLHAT) stroke and 209
in cast nephropathy 172 337, 338, 381 thrombotic prevention and
for hypertension 573, 582–3 antimalarial therapy in lupus 319–20
in idiopathic immune‐complex nephritis 410–11 Aspreva Lupus Management Study
MPGN 278 antineutrophil cytoplasmic antibody (ALMS) Group study 408
in IgA nephropathy 255, 258 (ANCA) see ANCA‐ Maintenance Study 417, 418
in Ig‐related amyloidosis 488 associated vasculitis Assessment of LEscol in Renal
angiotensin receptor blockers (ARBs) antiplatelet therapy Transplantation (ALERT)
ADPKD and 196 in CVD 515, 519 trial 557
ATN and 124 in diabetic kidney disease 382–3 ASTRAL trial 355, 576
blood pressure and 380 in renal biopsy 209 atacicept 266
CA‐AKI and 154 Antithrombotic Trialists’ (ATT) atheroembolic renal disease 98, 146,
cast nephropathy and 172, 482 collaborators 382 163, 169–70
CKD risk and 50 ANZDATA see Australian and New clinical features and laboratory
glomerulonephritis and 488 Zealand Dialysis and findings 170, 170, 171
hypertension and 338, 453, Transplant Registry course and management 170–1
573, 582–3 (ANZDATA) definition, prevalence, and
membranous nephropathy 237 APACHE III 128 pathogenesis 169–70
MPGN and 277 apolipoprotein L1 (APOL1) gene Atherosclerosis Risk in Communities
nephrotic syndrome and 554 9, 34, 77 (ARIC) study 507, 529,
prerenal failure 96 apparent treatment‐resistant 550, 581
antibiotics hypertension 573 Atherothrombosis Intervention in
in AAV 469, 469 Appropriate Blood Pressure Control Metabolic Syndrome with
in acute interstitial nephritis 163 in Diabetes (ABCD) Low HDL/High
in ANCA‐associated vasculitis 470 normotensive study 379 Triglycerides and Impact on
in ATN 129 APRIL see A Proliferating Inducing Global Health Outcomes
bacterial tonsillitis 260 Ligand (APRIL) 404 (AIM‐HIGH) study 561
in cast nephropathy 488 arachidonic acid metabolism atorvastatin
in CKD 536, 537 pathway 322 in chronic kidney disease 556,
cyst infection and 300 Aranesp 542, 544, 545 558, 559
in MCD 220 ARDS Clinical Trials Network 100 in diabetic nephropathy 382
in STEC‐HUS 435 ARIC study see Atherosclerosis Risk in atrial natriuretic peptide (ANP) in
urinary tract infection Communities (ARIC) study CA‐AKI 151
bindex1.indd 684 09-12-2022 17:14:43

Index 685
attendant hyperkalemia 321 diet modification 296 baroreceptor activation therapy in

atypical hemolytic uremic syndrome protein intake 296 hypertension 340
(aHUS) 425 salt restriction 296 B‐cell activating factor (BAFF) 404
AURORA study 382, 556 fluid intake 296 belimumab 417
Australian and New Zealand Dialysis intracranial aneurysms 297 benazepril/amlodipine vs benazepril/
and Transplant Registry physical activity 297 thiazide 381
(ANZDATA) 445 prognosis 291–4, 291 Bence Jones proteinuria 482
autologous stem cell transplantation statistical prediction models Bence‐Jones proteins 172
(ASCT) 488 293–4 bendroflumethiazide 338
automated oscillometric blood Irazabal (Mayo) BENEDICT study see Bergamo
pressure (AOBP) 332 subclassification 293 Nephrologic Diabetes
autosomal dominant polycystic PRO‐PKD score 294 Complications Trial
kidney disease (ADPKD) treatment of 289, 294–301, 295 (BENEDICT) study
288–301 disease‐modifying treatments Bergamo Nephrologic Diabetes
clinical history 291–2, 292 297–9 Complications Trial
age 291 extrarenal complications (BENEDICT) study 380
clinical events 292 300, 301 beta‐blockers 110, 488
family history 291–2 genetic care 294–5 blood pressure and 345
perinatal history 292, 292 family planning and combined with RAS blockade 339
diagnosis 290–1 preimplantation genetic beta‐lactam therapy 307, 308
genetic testing 290–1, 291 diagnosis (PGD) 295 bile acid binding resins 559
renal ultrasound in absence of genetic counseling 294 biomarkers
family history 290 screening 294–5, 295 in acute kidney disease 90–2, 91, 93
renal ultrasound with positive kidney pain 299–300 in acute tubular necrosis 127–8
family history 290, 290 management of ESKD in 301 in chronic kidney disease 60
epidemiology 288 psychological care acute and conventional biomarkers 60–2, 62
incidence 288 chronic, 295–6 novel biomarkers 63
introduction 288 Avoiding Cardiovascular events of collagen and bone metabolism
investigations 293, 293 through Combination and activity 594–6
eGFR and serum creatinine 293 therapy in Patients Living conventional 60–2
imaging 293 with Systolic Hypertension albuminuria/proteinuria 61, 62
laboratory and imaging (ACCOMPLISH) trial 381 estimated glomerular filtration
tests. 293 AWARD‐7 376 rate 60–1
lack of validity to define “rapid azathioprine 412 hypertension 61
progressor” 293 in ANCA‐associated vasculitis 471 serum bicarbonate/metabolic
pathophysiology 288–90 azotemia 4 acidosis 61–2
prevalence 288 serum phosphate and FGF23 62
prevention and/or minimization of b uric acid 62
renal and cardiovascular B cells in hepatorenal syndrome 110–12
disease 296–7 focal segmental glomerulosclerosis novel 63
avoidance of acute kidney and 219 in predicting risk of DKD 383
injury 297, 298 HCV and 278 bisoprolol 339
blood pressure control, vascular minimal change disease and 219 bisphosphonates
dysfunction, and B lymphocyte stimulator (Blys) 404 AKI secondary to 173
cardiovascular disease Balkan endemic nephropathy 75 bone mineral density 607–8,
296–7 bare metal stent (BMS) 520 607 (box)
bindex1.indd 685 09-12-2022 17:14:43

686 Index
bleeding prediction scores in renal c Canadian European Study Group

biopsy 210 C3 glomerulopathies (C3G) 275–7, 545, 546
blisibimod 266 275, 494–501 Canadian Society of Nephrology
blood pressure associated with monoclonal 50, 673
control in CVD 514, 519 gammopathy 486–7, 497 canagliflozin 375, 519
lowering 379–80 clinical presentation and Canagliflozin and Renal Events in
combination therapy 381 outcome 495–6 Diabetes and Established
pharmacotherapy for 380–1 complement assays 496–7 Nephropathy Clinical
RAAS blockade 380 definitions 494 Evaluation (CREDENCE)
home measurement 332 epidemiology 494–5 trial, 375, 519
in hypertension global approach for 498 CANPREDDICT study 531–34
assessment 575, 575 introduction 494 CANVAS trial 374, 375
control 582 laboratory work‐up 496–7 CANVAS R trial 374
measurement 331–2, 332, pathophysiology 495 Caplan’s syndrome 315
343–4 renal pathological findings 496 CAPTOPRIL study 380
target 583–85, 584 treatment 281–3, 282, 283, cardiac magnetic resonance,
body mass index (BMI) 630–31 498–501, 499–500 CVD and 513
BONAFIDE study 605 cadmium 75, 314–15 cardiorenal syndrome 96, 101
bone 594–8 calcifediol 603 (Box) ACE Inhibitors and ARBs 102
alkaline phosphatase in 594 calcimimetic therapy 605, 605 B‐type natriuretic peptide
architecture and (Box), 606 (bnp) 102–3
metabolism in 593 calcineurin inhibitors diuretics 101
biomarkers of metabolism and (CNIs) 278, 430 natriuretic peptides 102–3
activity 594–6 in dyslipidemia 555 postobstructive diuresis 104
biopsy 597–8 in focal segmental glomerulosclerosis treatment of 101–102
fracture risk assessment 598 (FGS) 224–5 ultrafiltration 103
imaging 597 in minimal change disease 221–2, cardiotrophin‐like cytokine factor 1
structure and quality 596–8 221 (Box) (CLCF1) 217
vascular calcification 598 calcitriol 603 (Box) cardiovascular disease and chronic
bone disease after kidney calcium, serum 603–605 kidney disease 507–20
transplantation in chronic kidney disease‐mineral definition 507
609–10, 609 and bone disorder 594 diagnostic tests 510–13, 511–12
see also chronic kidney disease‐ target levels 608, 608 (box) biochemical abnormalities
mineral and bone disorder calcium acetate/calcium carbonate for 510–12
(CKD‐MBD) hyperphosphatemia B‐type natriuretic
bone mineral density 597, 607–8 calcium channel blockers 339 peptides 510
bortezomib 172, 173, 488 calcium disodium versenate 313 inflammation 510
BP Treatment Trialists calcium ethylenediaminetetraacetic lipid profile 510
Collaboration 340 acid (CaNa2–EDTA) 313 troponin 512
brain natriuretic peptide calcium inhibitors 488 imaging 512–13
(BNP) 376, 510 calcium intake cardiac magnetic
British Society of Rheumatology recommendations 624 resonance 513
(BSR) guidelines 467 calcium oxalate crystal computed tomography 513
B‐type natriuretic peptide nephropathy 178, 179 coronary angiography 513
(BNP) 102–3, 510 Canadian Erythropoietin Study echocardiography 512–13
budesonide 266 Group 544 electrocardiogram 512
bindex1.indd 686 09-12-2022 17:14:44

Index 687
myocardial perfusion Charcot–Marie–Tooth disorder 429 GFR estimating equations

imaging 513 Child–Pugh score 110 and risk 6
positron emission tomography children progression to ESKD 6
(PET) 513 renal biopsy 211 prediction of 61
single‐photon emission routine childhood in pregnancy 444–5
computed tomography vaccinations 536 prevalence of 7–8
(SPECT) 513 urinary tract infection 307, 308 risk factors 6–7
epidemiology 507–8 see also hemolytic uremic syndrome acute kidney injury 7
management 508, 513–20 chlorambucil in combination with diabetes mellitus 7
anticoagulants 517, 519–20 corticosteroids 238 hypertension 6–7
antiplatelet therapy 515, 519 chloroquine 411 in specific subgroups 8–9
blood pressure control 514, 519 chlorothiazide 338 older adults 8–9
coronary revascularization chlorthalidone 337, 338, 381 sex difference in kidney health
518, 520 Choices for Healthy Outcomes in and outcomes 9
glucose control 516, 519 Caring for End‐Stage Renal chronic kidney disease‐mineral and
lifestyle modification 520 Disease (CHOICE) study bone disorder (CKD‐MBD)
statins 515, 519 510, 665 589–610
pathophysiology 508–10 cholecalciferol 603 (Box) bone architecture and
prediction of 61 cholesterol 632 metabolism in 593
prognosis 513 in dyslipidemia 559–60 definition 590–91
Cardiovascular Risk reduction in embolism in renal artery stenosis diagnostic tests 594–8, 595–6
Early Anemia Treatment 360–1 biochemical abnormalities 594
with Epoetin (CREATE) Cholesterol Treatment Trialists’ (CTT) bone 594–8
study 545, 546 Collaboration 381 biomarkers of collagen and
CARI guidelines see Caring for Chronic Kidney Disease Epidemiology bone metabolism and
Australasians with Renal Collaboration activity 594–6
Impairment (CARI) (CKD‐EPI) 371 bone structure and quality
guidelines Chronic Kidney Disease in Children 596–8
Caring for Australasians with Renal Study (CKiD) cohort 62 bone‐specific alkaline
Impairment (CARI) Chronic Kidney Disease Prognosis phosphatase 594
guidelines 208 Consortium (CKD‐PC) vascular calcification 598
CARMELINA trial 377 5, 26 epidemiology 591–2
cast nephropathy 171, 481, 488 chronic kidney disease introduction 589–90
clinical features and laboratory classification 3, 5 pathophysiology 592–94
findings 172, 172 definition 3, 4 prognosis 598–9
course and treatment 172–3 in developing world 8 secondary hyperparathyroidism
definitions and etiology 171–2 dyslipidemia as precursor of and 592
catastrophic antiphospholipid 551–2, 551, 552 treatment for 590, 599–610
syndrome (CAPLS) 429 epidemiology of 3–10 bone disease after kidney
CD46 282 as global public health priority 8 transplantation
CD80 in minimal change disease 219 knowledge gaps and future 609–10, 609
celecoxib 319, 320, 321 directions 9–10 bone mineral density 607–8
hypertension and 323 NSAIDS and 322–3 bisphosphonates 607–8,
Chapel Hill Consensus Conference outcomes 5–6 607 (box)
(CHCC) nomenclature of all‐causes mortality 5 denusomab 608
vasculitis 461 cardiovascular events 5–6 raloxifene 608
bindex1.indd 687 09-12-2022 17:14:44

688 Index
chronic kidney disease‐mineral and Clostridium difficile infection 531 longer‐term adverse 148–9, 149
bone disorder (CKD‐MBD) Cochrane Kidney and Transplant short‐term adverse 148, 148
(cont’d) Group 412, 413 incidence 146–7, 147
calcimimetic therapy 605, 605 Cochrane Renal Group 263–4, 263 nomenclature and definition 146
(Box), 606 Cockcroft–Gault formula 78, pathophysiology 145–6, 145
dialysate calcium 606–7 446, 673 prevention 150–4
surgical colestilan 559 current recommendations 154
parathyroidectomy 605–6 collagen, biomarkers of 594–6 intravenous crystalloid 152–3
patient and caregiver treatment complement‐mediated thrombotic limitations of clinical trials
preferences 610 microangiopathy 429 153–4
phosphate binders 599–603, controversy 437 overview 150
600 (Box) pathophysiology 431–2 pharmacological
aluminum 600 prognosis 433–34 agents 151–2, 151
dialysis duration and treatment 436–7 renal replacement therapies 150
frequency 600–603 dialysis 436 selection of contrast agent 150
dietary modification 600 eculizumab 436 risk factors 147–8
nonaluminum phosphate kidney transplantation 436 patient‐related 147
binders 600, 601–2 plasma therapy 436 procedure‐related 148
serum parathyroid hormone and computed tomography contrast‐induced nephropathy (CIN)
serum calcium 603–605 of bone 597 135, 146, 207
vitamin D 603, 603(Box), 604 CVD and 513 CONVINT trial 189
target levels of serum parathyroid vascular calcification 598 Cooperative North Scandinavian
hormone, phosphorus, and CONSENSUS see Cooperative North Enalapril Survival Study
calcium 608, 608 (box) Scandinavian Enalapril (CONSENSUS) 102
chronic lymphocytic leukemia 479 Survival Study CORAL trial 355, 576
Chronic Renal Insufficiency Cohort (CONSENSUS) coronary angiography, CVD
(CRIC) study 63, 74, 573 continuous ambulatory peritoneal and 513
chronic thrombotic dialysis (CAPD) 551 coronary artery bypass grafting
microangiopathy 494 continuous positive airway pressure (CABG) 520
cinacalcet 605 (CPAP) 333 coronary revascularization in CVD
CKD Epidemiology Collaboration continuous renal replacement therapy 518, 520
(CKD‐EPI) equation 28 (CRRT) 150, 185, 186, 187, corticosteroids
CKD‐Epi eGFR formula 18 188, 189, 191, 192, 194–6, in focal segmental
CKD‐EPI see Chronic Kidney Disease 194, 195 glomerulosclerosis (FGS)
Epidemiology Collaboration continuous veno‐venous 223–4
(CKD‐EPI) hemodiafiltration in IgA nephropathy 263–4, 263
CKD‐PC see Chronic Kidney Disease (CVVHDF) 186, in membranous nephropathy 238,
Prognosis Consortium 188–9, 194–5 239, 242
CKID see Chronic Kidney Disease in continuous veno‐venous hemodialysis in minimal change disease 220–1,
Children Study (CVVHD) 186 220 (box)
(CKiD) cohort continuous veno‐venous co‐trimoxazole in ANCA‐associated
CLCF1 in focal segmental hemofiltration vasculitis 472
glomerulosclerosis (CVVH) 186 COX‐2
(FGS) 217 contrast‐associated acute kidney NSAIDs, effects on renal
clodronate 173, 607 (Box) injury (CA‐AKI) 145–54 prostaglandins 319, 320
clonidine 339 associated outcomes 148–50 renal dysfunction and 321
clopidogrel 383 clinical implications 149–50 C‐reactive protein (CRP) 632
bindex1.indd 688 09-12-2022 17:14:44

Index 689
CREATE study see Cardiovascular cystatin C 4, 6, 18, 90, 314, 376 prevention, treatment, and
Risk reduction in Early cytotoxic agents combined with follow‐up 371–83
Anemia Treatment with corticosteroids antiplatelet agents 382–3
Epoetin (CREATE) study in membranous nephropathy 238, biomarkers in predicting risk
creatinine index 632 240–1, 242 of DKD 383
creatinine‐positive or creatinine‐ blood pressure lowering 379–81
negative subjects 91 d dipeptidyl peptidase 4 inhibitors
CREDENCE trial see Canagliflozin and “D− HUS” (diarrhea negative 376–7
Renal Events in Diabetes and HUS) 429 glucagon‐like peptide‐1 receptor
Established Nephropathy “D+ HUS” (diarrhea positive agonists 376
Clinical Evaluation HUS) 429 glycemic control 372–4
(CREDENCE) trial DAPA‐CKD 375 lifestyle and multifaceted
crescentic Immunoglobin A dapagliflozin 375 approach 371–2
nephropathy (IgAN) 256, daprodustat 547 lipid lowering 381–2
260, 261 darbepoetin (Aranesp) 545 metformin 377–8
CRIC study see Chronic Renal DASH diet 336–7 mineralcorticoid receptors
Insufficiency Cohort DASH study see Dietary Approaches inhibitor 381
(CRIC) study to Stop Hypertension novel interventions 383, 383
cryoglobulinemic glomerulonephritis (DASH) study pharmacotherapy for glucose
484–5, 485 DCVAS study see Diagnostic and lowering 374
cryoglobulinemic nephritis 494 Classification Criteria sodium‐glucose linked co‐
crystalline nephropathies 174–5 for Vasculitis (DCVAS) transport‐2 inhibitors
clinical features 175 study 374–5, 375
course and management 175–9 DECLARE‐TIMI 58 sulfonylureas 378
definition, etiology, and dense deposits disease (DDD) 276–7, thiazolidinediones 378–9
pathogenesis 174–5, 176–7 494, 495, 496 screening and diagnosis 369–71,
laboratory findings 175, 178 denusomab 608 370, 371
crystalloid versus colloid fluids 100 desmopressin in renal biopsy 209–10 vs. diabetic nephropathy 366, 367
cyclophosphamide Deutsche Diabetes Dialyse Studie diabetic nephropathy see diabetic
in ANCA‐associated vasculitis (4D) 556 kidney disease
467–8, 467, 468 DGK epsilon‐associated Diabetic Retinopathy Candesartan
in combination with nephropathy 494 Trial [DIRECT] ‐Prevent 1,
corticosteroids 238 Diabetes Atherosclerosis Intervention and DIRECT‐Protect 1) 380
and methylprednisolone 412 Study (DAIS) 382 diacylglycerol kinase ε (DGKE)‐
with/without azathioprine in IgA diabetes mellitus 369 related TMA 429
nephropathy 264 and chronic kidney disease 7, 35 Diagnostic and Classification Criteria
cyclosporin A dyslipidemia and 559 for Vasculitis (DCVAS)
in focal segmental as risk factor for bacterial study 461
glomerulosclerosis infection 530 dialysate calcium 606–7
(FGS) 224 screening 369 dialysis 658–66
in membranous nephropathy type 1 366, 367 epidemiology of ESKD and 658–9
242, 243 type 2 366, 371, 374, 375, 377, guidelines and studies on initiation
in minimal change disease 221–2 382, 383 of 671–74, 672
cyclosporine 558 see also diabetic kidney disease modality selection 660–62,
in membranous nephropathy diabetic kidney disease 366–84 661, 662
242, 243 incidence and prevalence 366–7 nephrology referral (early vs. late)
in renal transplant patients 558 pathogenesis 367–9, 368 and outcomes 659, 659
bindex1.indd 689 09-12-2022 17:14:44

690 Index
dialysis (cont’d) environmental exposures 75, 76 in renal transplant 555

patient educational and learning socioeconomic factors 74 statins
programs 660 diagnosis of CKD among 77–9 adverse outcome/safety
in pregnancy 445, 447–51 awareness 78 in CKD 558
preparation for hemodialysis and diagnostic tools 78–9 and diabetes risk 559
vascular access 665–6 other challenges 79 dosage 558
preparation for peritoneal screening 78 in hemodialysis 556
dialysis 668 epidemiology 72–3, 73 in peritoneal dialysis 556–7
timing of initiation 662–65, treatment 79–80 progression of CKD 557–8
663–64, 671 access to optimal care 79 in renal transplant 557
see also hemodialysis; peritoneal availability of RRT 79–80 triglyceride‐carrying lipoprotein
dialysis early prevention programs 79 particles in NDD‐CKD
Dialysis Outcomes and Practice health literacy 80 552, 553
Patterns Study (DOPPS) disseminated intravascular coagulation triglyceride/HDL 560–62
607, 665 (DIC) in hemolytic uremic fibrates 560–61
diclofenac 321 syndrome 432 niacin 561
Dietary Approaches to Stop DOPPS see Dialysis Outcomes and omega (n‐3) fatty acids 561–62
Hypertension (DASH) Practice Patterns see also lipid lowering therapies;
study 582 Study (DOPPS) statin therapy
dietary caloric restriction 642 doxazosin 337, 339
dietary modification doxercalciferol 603 (Box) e
in autosomal dominant polycystic drug‐eluting stent (DES) 520 early prevention programs 79
kidney disease 296 dulaglutide 376 echocardiography
in chronic kidney disease‐mineral dyslipidemia in chronic kidney CVD and 512–13
and bone disorder disease 549–64 vascular calcification 598
(CKD‐MBD) 600 calcineurins 555 eculizumab 430
dietary patterns 643, 644 cholesterol 559–60 for C3 glomerulopathy in native
dietary protein restriction 636–42, bile acid binding resins 559 and transplant kidneys
637–9, 640–41 ezetimibe 559 (adults) 283
dietary salt restriction 642 PCSK9 Inhibitors 559–60 in complement‐mediated TMA 436
digitalis 488 diagnosis 549 in IC‐MPGN and C3G 501
dihydropyridine calcium channel epidemiology 549–51 in membranous nephropathy
blocker see amlodipine guidelines 562, 563 242–8
diltiazem 339 HDL changes in CKD 554–5 in MPGN 501
dipeptidyl peptidase 4 inhibitors in LDL changes 553 in STEC‐HUS 435
diabetic kidney disease lipid‐altering trials 555–6 edetate calcium disodium 313
376–7 mammalian target of rapamycin Effect of Early vs Delayed Initiation of
dipyridamole 264–6 inhibitors 556 Renal Replacement Therapy
direct oral anticoagulants in nephrotic syndrome 553 on Mortality in Critically Ill
(DOAC) 520 in nonproteinuric NDD‐CKD and Patients With Acute Kidney
disability‐adjusted life years hemodialysis 552 Injury (ELAIN) trial
(DALYs) 8, 46 nonstatin lipid lowering 192, 193
disadvantaged populations in CKD 559 Effect of Irebesrtan in the
72–80 peritoneal dialysis 553, 554 Development of Diabetic
definitions and theoretical as precursor of CKD 551–2, Nephropathy in Patients
framework 73–7, 75 551, 552 With T2DM (IRMA‐2)
biological factors 75–7 prednisone 555 trial 380
bindex1.indd 690 09-12-2022 17:14:44

Index 691
Efficacy and Safety of Finerenone in environmental factors associated epoietin 544

Subjects With Type 2 with CKD 74 Ergocalciferol (vitamin D2) 603 (Box)
Diabetes Mellitus and environmental nephrotoxins erythropoiesis‐ stimulating agents in
Diabetic Kidney Disease 311–16, 312 anemia 542, 544–6
(FIDELIO‐DKD) 385 treatment of 313 Escherichia coli 126, 305
Efficacy and Safety of Finerenone in epidemiology Escherichia coli‐related HUS (STEC‐
Subjects With Type 2 acute tubular necrosis 124 HUS, typical HUS) 425
Diabetes Mellitus and the ANCA‐associated vasculitis 461–4 estimated glomerular filtration
Clinical Diagnosis of autosomal dominant polycystic rate (eGFR)
Diabetic Kidney Disease kidney disease cross‐sectional surveys of 16
(FIGARO‐DKD) 385 (ADPKD) 288 end‐stage renal disease
ELAIN trial see Effect of Early vs C3 glomerulopathies (C3G) 494–5 (ESRD) and 14
Delayed Initiation of Renal chronic kidney disease (CKD) etelcalcitide 605
Replacement Therapy on 3–10 ethylene glycol poisoning 179
Mortality in Critically Ill and cardiovascular disease etidronate 607 (Box)
Patients With Acute Kidney 507–8 EULAR/ERA‐EDTA Guidelines
Injury (ELAIN) trial and hypertension 573–74, 574 410, 418
elderly people see age; Evaluation of and kidney failure 26–7 EuLite trial 173
Losartan in the Elderly mineral and bone Euro‐Lupus Nephritis Trial
(ELITE) study disorder 591–2 (ELNT) 412
electrocardiogram, CVD and 512 in CVD 507–8 European League Against
ELITE study see Evaluation of disadvantaged populations Rheumatism (EULAR) 467
Losartan in the Elderly 72–3, 73 European League Against
(ELITE) study dyslipidemia 549–51 Rheumatism and European
ELIXA trial 376 ESKD and dialysis therapies Renal Association–European
ELNT see Euro‐Lupus Nephritis 658–9 Dialysis and Transplant
Trial (ELNT) focal segmental glomerulosclerosis Association (EULAR/
empagliflozin 375 215–17 ERA‐EDTA) 410, 418
EMPA‐Kidney 375 hemolytic uremic syndrome 430 European Myeloma network
EMPA‐REG trial 374 immune complex‐mediated guidelines (2015) 173
Empowering Patients on Choices membranoproliferative European Renal Association–
for Renal Replacement glomerulonephritis European Dialysis and
Therapy (EPOCH‐RRT) (IC‐MPGN) 494–5 Transplant Association
study 661 minimal change disease 215–17 (ERA‐EDTA) 467, 661
end‐stage kidney disease (ESKD) 3 in pregnancy 444–5 Evaluation of Cinacalcet
in African Americans 75–7, 420 acute kidney injury in 445 Hydrochloride Therapy to
autosomal dominant polycystic chronic kidney disease in 444–5 Lower Cardiovascular
kidney disease dialysis in 445 Events (EVOLVE) trial 605
(ADPKD) in 301 transplantation in 445 everolimus 299, 555
and dialysis therapies 658–9 renal biopsy 207 evolocumab 382
estimated glomerular filtration rate reverse 549, 550 EVOLVE trial see Evaluation of
(eGFR) and 14 urinary tract infection 305–6 Cinacalcet Hydrochloride
end‐stage renal disease (ESRD) see eplerenone 339 Therapy to Lower
end‐stage kidney EPOCH‐RRT study see Empowering Cardiovascular Events
disease (ESKD) Patients on Choices for (EVOLVE) trial
Engerix 535 Renal Replacement Therapy exercise, hypertension and 337, 582
Enterobacteriaceae 305 (EPOCH‐RRT) study ezetimibe 382, 559
bindex1.indd 691 09-12-2022 17:14:44

692 Index
f Fluid and Catheter Treatment Trial fractional excretion of sodium (FENa)

FACTT see Fluid and Catheter (FACTT) 136 in acute tubular
Treatment Trial (FACTT) fluid balance 623 necrosis 127
falecalcitriol 603, 603(Box) fluid intake in autosomal dominant Fracture Intervention Trial (FIT) 608
family planning in autosomal polycystic kidney Fracture Risk Assessment Tool
dominant polycystic kidney disease 296 (FRAX) 598
disease 295 fluoroquinolones in UTI 307 fractures 598–9
Fanconi syndrome 311, 482 fluvastatin 558 FRAX see Fracture Risk Assessment
febuxostat 174 focal segmental glomerulosclerosis Tool (FRAX)
fendrix 535 (FGS) 214–27, 429 FREEDOM trial 608
fenofibrate 382, 560 clinical features 216 Frequent Hemodialysis Network
Fenofibrate Intervention and Event definition 214 (FHN) Daily Trial 600–603
Lowering in Diabetes pathogenesis of 218 fresolimumab in focal segmental
(FIELD) trial 382, 560 pathophysiology 217–19, 218 glomerulosclerosis
fenoldopam in CA‐AKI 151 immune disorders 219 (FGS) 225
fertility and CKD 445–6 B cells 219 furosemide 481
FHN trial see Frequent Hemodialysis T cells 219 in CA‐AKI 151
Network (FHN) Daily Trial others 219 furosemide stress test (FST) 92, 193
fiber intake 642–3 putative circulating permeability Further Cardiovascular Outcomes
fibrates 560–61 factors 217–19 Research with
FIDELIO‐DKD see Effect of anti‐CD40 antibody 217–18 PCSK9 Inhibition in
Irebesrtan in the CLCF1 217 Subjects with Elevated Risk
Development of Diabetic IL‐13 218 (FOURIER) trial 560
Nephropathy in Patients microRNA 218–19
With T2DM (IRMA‐2) suPAR 217 g
trial 380 plasmapheresis 225–6 GDCN see Glomerular Disease
FIELD trial see Fenofibrate pregnancy 227 Collaborative Network
Intervention and Event treatment of 216, 223–5, 226 (GDCN) Registry
Lowering in Diabetes alkylating agents 224 gemfibrozil 560, 562
(FIELD) trial calcineurin inhibitors genetic care in autosomal dominant
FIGARO‐DKD see Efficacy and Safety (CNIs) 224–5 polycystic kidney
of Finerenone in Subjects CsA 224 disease 294–5
With Type 2 Diabetes MMF 225 genetic counseling in autosomal
Mellitus and the Clinical sirolimus 225 dominant polycystic kidney
Diagnosis of Diabetic TAC 224–5 disease 294
Kidney Disease corticosteroids 223–4 genetic screening in autosomal
(FIGARO‐DKD) monoclonal antibodies 222 dominant polycystic kidney
fish oil in IgA nephropathy (Box), 225 disease 294–5, 295
266, 266 abatacept 225 genome‐wide association studies 75
Fish Oil Inhibition of Stenosis in adalimumab 225 GISSU‐Prevenzione see Gruppo
Hemodialysis Grafts (FISH) fresolimumab 225 Italiano per lo Studio della
study 562 RTX 225 Sopravvivenza nell’Infarto
Fistula First Breakthrough Initiative summary 225, 226 Miocardico‐
(FFBI) (CMS) 665 food diaries and records 632 Prevenzione study
FIT see Fracture Intervention food frequency questionnaires 632 GKHA see Global Kidney Health
Trial (FIT) fostamatinib 266 Atlas (GKHA)
bindex1.indd 692 09-12-2022 17:14:44

Index 693
glibenclamide 378 GRADE framework see Grading of hematuria

gliclazide 378 Recommendations isolated invisible, IgA
glipizide 378 Assessment, Development nephropathy and 260
Global Burden of Disease Project and Evaluation (GRADE) recurrent visible, IgA nephropathy
21, 46–7 framework and 259–60
Global Kidney Health Atlas (GKHA) Grading of Recommendations Hemodiafe study 177
47, 207 Assessment, Development hemodialysis 343–5, 671–77
Glomerular Disease Collaborative and Evaluation (GRADE) choice of BP measurement 343–4
Network (GDCN) framework 26, 410 as initial kidney replacement
Registry 224 renal biopsy 205–12 therapy 674
glomerular disorders Working Group criteria 135 nonpharmacological measures
with nonorganized immunoglobulin Grams MCMC model 65 345, 345
deposits 485–6 granulomatous interstitial nephritis patient survival 674–7, 674
with organized immunoglobulin (GIN) 163 methodological issues 675–6
deposits 483–85 Gruppo Italiano per lo Studio della patient characteristics 675, 676
without immunoglobulin deposits Sopravvivenza nell’Infarto practice patterns 674–5
486–7 Miocardico‐Prevenzione quality of life and biochemical
glomerular filtration rate (GFR) in (GISSI‐Prevenzione) parameters 676
definition of chronic kidney study 561 preparation for 665–6
disease 3, 4 guanfacine 339 hemolytic uremic syndrome 425–37
glomerular tip lesion 214 atypical 425
glomerulonephritis, NSAIDS h clinical manifestation 432
and 322 Halt Progression of Polycystic Kidney diagnosis 427–9, 432–3
glomerulonephritis with organized Disease (HALT‐PKD) epidemiology 430
microtubular monoclonal trial 297, 583 pathophysiology 430–32
immunoglobulin deposits HALT‐PKD trial see Halt Progression prognosis 433–34
(GOMMID) 484, 485 of Polycystic Kidney Disease treatment 434–7
glucagon‐like peptide‐1 receptor (HALT‐PKD) trial see also complement‐mediated
agonists in diabetic kidney HBV‐associated TMA; STEC‐HUS
disease 376 membranoproliferative hemorrhagic colitis 429
glucarpidase (carboxypeptidase) 179 glomerulonephritis 278–81 Henoch–Schönlein purpura 254, 260
glucocorticoids in ANCA‐associated HCV‐associated immune‐complex‐ heparin 180
vasculitis 468 mediated 274–5, 274 hepatic cysts in autosomal dominant
glucose control treatment 278–81 polycystic kidney disease
in CVD 516, 519 HDL in CKD 554, 560–62 (ADPKD) 300
in diabetic kidney disease 374 health literacy 80 hepatitis B, prevention of
glucose‐6‐phosphate dehydrogenase Heart and Soul Study 63 infection 535
deficiency 174 Heart Outcomes Prevention hepatitis C 494
glyburide 378 Evaluation (HOPE) hepatorenal syndrome 96, 107–16
glycemic control in diabetic kidney trial 380 AKI and HRS‐AKI
disease 372–4 HOPE‐3 332 criteria 110, 111
GOMMID see glomerulonephritis heavy‐chain deposition disease arterial vasodilation 107–8
with organized microtubular (HCDD) 479, 481, 485 clinical features 110
monoclonal heavy metals 311–15 definition and staging of AKI 107
immunoglobulin deposits HELLP syndrome 430 diagnostic approach 107,
(GOMMID) Helsinki Heart Study 560 109–10, 111
bindex1.indd 693 09-12-2022 17:14:44

694 Index
hepatorenal syndrome (cont’d) patient populations 340–3 hyporeninemic hypoaldosteronism,

management 112 advanced CKD 342–3 NSAIDS and 323–4
nonpharmacological diabetic CKD 341, 344
therapy 115–16 dialysis 343 i
liver transplantation 115 intensive BP lowering 340, ibandronate 607 (Box)
renal replacement therapy 116 341, 342 ICER 50
transjugular intrahepatic nondiabetic CKD 340–1, 342–3 IDEAL study 662, 666
portosystemic shunt pharmacotherapy 338–40, 346–7 IDEAL‐ICU trial see Initiation of
(tips) 115–16 calcium channel blockers 339 Dialysis Early Versus
pathogenesis 107, 108 combination choice 339 Delayed in the Intensive
pharmacological therapy 112–15 drug choice 337–8 Care Unit (IDEAL‐ICU) trial
aim of 115 guidelines and variability 346 idiopathic nephrotic syndrome (INS)
vasoconstrictors 113–15, 114 kidney transplantation 346 see focal segmental
midodrine 113–15 newer agents 339 glomerulosclerosis; minimal
norepinephrine 113–15 renin angiotensin system change disease
octreotide 113–15 blockade 338 IDNT see Irbesartan in Diabetic
terlipressin 113 second‐line drugs 339 Nephropathy Trial (IDNT)
prevention 116 thiazide and thiazide‐like IgA nephropathy 254–66, 429–30
reduced cardiac output 108–9 diuretics 338 corticosteroids in 263–4, 263
renal vasoconstriction 109 reason for treatment 331–4 cyclophosphamide with/without
systemic inflammation 109 research needs 346–7 azathioprine 264
treatment recommendations box, as risk factor of chronic kidney epidemiology 254
based on GRADE, for disease 6–7 evidence in adults 258–9, 259
AKI‐HRS 112 (Box) treatment in CKD 334–40 fish oil in 266, 266
urinary biomarkers 110–12 see also blood pressure; immunosuppressive
herpes zoster, prevention of hemodialysis; hypertension treatments 263
infection 535–6 and CVD; masked with isolated invisible hematuria
high bone turnover renal hypertension; peritoneal and proteinuria 260
osteodystrophy 593 dialysis management flowchart 256
HIV 73, 429 hypertension and CVD 573–85, 574 MMF and mizoribine in
Hodgkin disease 214, 217 blood pressure 265–6, 265
home BP measurement 332 assessment 575, 575 with nephrotic syndrome 262
home hemodialysis (HHD) 662 control 582 other therapies 266
HOT trial see Hypertension Optimal target 583–85, 584 Oxford Classification of IgA
Treatment (HOT) trial definition 573 nephropathy 258
hydralazine 339 differential diagnosis 576, 576 pathophysiology 255
hydrochlorothiazide (HCTZ), epidemiology 573–74, 574 prognosis 255–7, 257
hydroxychloroquine 411 introduction 573 adult IgAN 255–6, 257
hyperdynamic sepsis 126 pathophysiology 575–6 pediatric IgAN 256
hyperkalemia, NSAIDS and 323–4 treatment goals 576–83, 577–80 prognostic factors 256–7
hypertension 331–47 Hypertension in the Very Elderly Trial with rapidly declining GFR 260–2
blood pressure measurement (HYVET) 340 evidence 261, 261
331–2, 332 Hypertension Optimal Treatment recommendation 260–1
blood pressure variability 333 (HOT) trial 380 recommendations for adults 258
nocturnal pattern 333 hypocomplementemic persistent with recurrent visible hematuria
NSAIDS and 323 glomerulonephritis 273 259–60
bindex1.indd 694 09-12-2022 17:14:44

Index 695
rituximab 266 infection and CKD 526–38, 527 iodinated radiocontrast dye in acute
with slowly progressive renal complications 530–34 tubular necrosis 135
impairment 262–3 AKI and progression to ESKD Irazabal (Mayo) subclassification in
treatment 255, 258 531–34 autosomal dominant
IgA vasculitis (IgAV) 254 cardiovascular events 534 polycystic kidney disease
Ig‐related amyloidosis 479, 488 mortality 530–31, 532–33 (ADPKD) 293
IL‐13 in focal segmental introduction 526 irbesartan 339, 380, 581
glomerulosclerosis prevention 534–6 Irbesartan in Diabetic Nephropathy
(FGS) 218 hepatitis B 535 Trial (IDNT) 380, 581,
imatinib 278 herpes zoster 535–6 582, 583
immune complex‐mediated influenza 534–5 IRMA‐2 trial see Effect of Irebesrtan
membranoproliferative live vaccines 536 in the Development of
glomerulonephritis (IC‐ pneumococcal vaccine 534 Diabetic Nephropathy in
MPGN), primary 494–501 routine childhood Patients With T2DM
clinical presentation and outcome vaccinations 536 (IRMA‐2) trial
495–6 vaccination: summary 536 iron, intravenous in anemia 546–7
definitions 494 risk of developing 526–9, ischemic nephropathy 354
epidemiology 494–5 528, 529 itai‐itai disease 315
global approach for 498 susceptibility to 529–30, 531
introduction 494 treatment 536–7 j
laboratory work‐up 496–7 antibiotics 537 Japan EPA Lipid Intervention Study
complement assays 496–7 other considerations 537 (JELIS) 562
monoclonal gammopathy 497 infliximab 470 JELIS see Japan EPA Lipid
pathophysiology 495 influenza A and B 429 Intervention Study (JELIS)
renal pathological findings 496 prevention of infection 534–5 JUPITER see Justification for the Use
treatment 498–501, 499–500 Initiation of Dialysis Early Versus of statins in Prevention – an
immunoglobulin‐related amyloidosis Delayed in the Intensive Intervention Trial
483, 484 Care Unit (IDEAL‐ICU) Evaluating Rosuvastatin
immunotactoid glomerulonephritis trial 190, 191, 192, (JUPITER) primary
484, 485 insulin glargine 376 prevention trial
Improved Reduction of Outcomes: insulin‐like growth factor‐binding Justification for the Use of statins in
Vytorin Efficacy protein 7 (IGFBP7) 128 Prevention – an Intervention
International Trial interleukin‐18 (IL‐18) 111, Trial Evaluating
(IMPROVE‐IT) 381–2, 559 116, 127–8 Rosuvastatin (JUPITER)
IMPROVE‐IT see Improved Reduction intermittent hemodialysis (IHD) primary prevention
of Outcomes: Vytorin 185, 193–4 trial 558
Efficacy International Trial International Myeloma Working
(IMPROVE‐IT) Group (IMWG) criteria 171 k
IMWG see International Myeloma International Society of Kaiser Permanente renal registry 46
Working Group (IMWG) Nephrology 72 Kaplan–Meier analyses 674
indapamide 338 Renal Pathology Society (RPS) KDIGO see Kidney Disease Improving
indinavir crystal nephropathy 178 Classification of lupus Global Outcomes (KDIGO)
indirect immunofluorescence in nephritis 400, 405 KHA‐CARI ADPKD Guidelines 297
ANCA‐associated intracranial aneurysms in autosomal kidney biopsy in pregnancy 446
vasculitis 466 dominant polycystic kidney kidney cyst hemorrhage and
indomethacin 320 disease (ADPKD) 297 rupture 300
bindex1.indd 695 09-12-2022 17:14:44

696 Index
Kidney Disease Improving Global kidney replacement therapy see liver‐type fatty acid binding
Outcomes (KDIGO) hemodialysis; peritoneal (L‐FABP) 111
45, 418 dialysis lixisenatide 376
AKI guidelines 88, 91, 92, 128, kidney transplantation 489 LNP023 266
136, 146, 190, 191, 192, 193 bone disease after 609–10, 609 lobular glomerulonephritis see
catheter length 197 in complement‐ membranoproliferative
CKD classification 3, 4, 7, 64 mediated TMA 436 glomerulonephritis
CKD heat map 27, 27 dysfunction 206 Long‐term Safety and Tolerability of
CKD and infection 527 dyslipidemia and 555 Alirocumab in High
CKD‐MBD 607 hypertension and 346 Cardiovascular Risk Patients
color‐coded CKD prognosis chart in pregnancy 445 with Hypercholesterolemia
61, 62 impact on outcomes 451 Not Adequately Controlled
Controversy Conference on outcomes after 451 with Their Modifying
ADPKD 297 renal biopsy 211 Therapy (ODYSSEY
CR, definition of 236 Kimmelstiel–Wilson lesions 366 LONG TERM)
on cyclophosphamide 467 Klebsiella spp. 305 study 559–60
glomerulofiltration rate 534 Kumamoto trial 372 Look AHEAD (Action for Health in
glomerulonephritis guidelines 214 Diabetes) multicentre
guidelines (2012) 50, 65, 66, 88, l trial 372
410, 546, 561, 562, 621 lanreotide 299 losartan 380
guidelines (2017) 597 lanthanum 600 lovastatin in chronic kidney
hypertension 346, 574 lead 75, 311–13 disease 558
iron, in anemia management 546 LEADER study 376 lovaza 561
Lipid Management guidelines 510 left ventricular mass index low‐bone turnover disease 593
MCD 221 (LVMI) 509 low‐osmolal contrast 150
on SLE 410 lenalidomide 172, 488 LRC‐CPPT see Lipid Research Clinics
vaccinations 535, 536 leucovorin 179 Coronary Primary
Kidney Disease Outcome Quality lifestyle modification in CVD 520 Prevention Trial
Initiative (KDOQI) 50 life‐years gained (LYG) 50, 55 (LRC‐CPPT)
guidelines 534, 620, 623 light‐ and heavy‐chain deposition LUNAR study see Lupus Nephritis
Kidney Disease Quality of Life disease (LHCDD) 485 Assessment with Rituximab
(KDQoL) 544 light‐chain associated Fanconi (LUNAR) study
kidney failure risk equations (KFREs) syndrome 482 lupus nephritis 400–420
36–8, 39, 64, 64, 65, 66 light‐chain deposition disease clinical presentations 407–8, 408
kidney failure risk prediction models (LCDD) 479, 485, 486 definition 400
36–8, 37, 38–9, 38, 39 linagliptin 377 diagnosis 405–9
kidney failure, epidemiology of 26–7 Lipid Research Clinics Coronary epidemiology 400–401
Kidney Health Australia –Caring for Primary Prevention Trial histological classification 401,
Australasians with Renal (LRC‐CPPT) 559 405–7, 405
Impairment (KHA‐CARI) lipids history 401
guidelines 50 in CKD patients 555–6 introduction 400
kidney injury molecule‐1 (KIM‐1) in diabetic kidney disease 381–2 kidney biopsy in 408–9
63, 127 liraglutide 376 pathophysiology 401–405
kidney pain in autosomal dominant lisinopril 337, 381 prognosis 409
polycystic kidney liver transplantation in hepatorenal treatment 409–20, 402–3
disease 299–300 syndrome 115 all patients 410–11
bindex1.indd 696 09-12-2022 17:14:44

Index 697
Class III and IV disease, alone or Malnutrition Universal Screening membranous nephropathy 235–49
combined with Class V Tool/Malnutrition Screening adrenocorticotropic hormone
411–17, 418, 427–9 Tool 630 247, 248
biologic therapies 413–17 Malnutrition‐Inflammation Score clinical manifestations 235–6
corticosteroids alone 411 (MIS) 624, 631 immunosuppressive therapy
cyclophosphamide or mammalian target of rapamycin 237–48, 238
azathioprine plus inhibitors, dyslipidemia high‐risk patients 242–8
corticosteroids 411–12 and 555 corticosteroids 239, 242
induction versus maintenance mannitol in CA‐AKI 151 cyclosporine 242, 243
therapy 411 MAPEC study 333 cytotoxic agents combined
mycophenolate mofetil plus markers of kidney function in with corticosteroids
corticosteroids 412–13 pregnancy 446 240–1, 242
pure membranous (Class V) Markov Chain Monte Carlo (MCMC) eculizumab 242–8
417–18, 427–9 model 64–5, 65 mycophenolate mofetil
special situations 418–20 Markov simulation models 50 242, 244
pregnancy and lupus Maroni’s formula 632 rituximab 242, 245
nephritis 419 masked hypertension 333–4, low‐risk patients 237
renal replacement 334, 335–6 medium‐risk patients 238–42
therapy 419–20 MASTER trial 334 corticosteroids 238, 239
very severe or nonresponsive maxacalcitol (22‐oxa‐calcitriol) cytotoxic agents combined
lupus nephritis 418 603 (Box) with corticosteroids
withdrawal of maintenance membrane cofactor protein (MCP) 282 238, 240–1
therapy 419 membranoproliferative cyclosporine A 242, 243
targets for 410 glomerulonephritis mycophenolate mofetil 244
Lupus Nephritis Assessment with (MPGN) 272–84, 429 tacrolimus 242
Rituximab (LUNAR) clinical presentation 277–8 nonimmunosuppressive
study 413 complement‐mediated 275–6, 275 therapy 237
HBV‐associated blood pressure 237
m immune‐complex‐ lipid lowering 237
Madrid Acute Renal Failure Study mediated 278–81 thromboembolism
Group 124 HCV‐associated immune‐complex‐ prophylaxis 237
magnetic resonance imaging of mediated 274–5, 274 introduction 235
bone 597 treatment 278–81 natural history 235
MAINRITSAN trial 472 histology 272–3 pathogenesis 236
MAINTAIN Nephritis Trial 413 immune‐complex‐mediated phospholipase A2 receptor 236
Maintenance of Remission using 273–4, 274 thrombospondin type‐1 domain‐
Rituximab in Systemic treatment 278, 279, 280 containing, 7A
ANCA‐associated infectious agents associated (THSD7A) 236
Vasculitis (MAINRITSAN) with 274 predicting factors 236
trial 472 natural history 277–8 response measurements 236–7
malaria 73 novel classification 273, 273 treatment 237–48, 238
malignancies, AKI associated with pathophysiology 273, 273 disease and treatment summary
171–3 treatment 278–83 248–9, 248
malnutrition and inflammation HBV‐or HCV‐associated 278–81 immunosuppressive therapy
complex (or cachexia) idiopathic immune‐complex 237–48, 238
syndrome (MICS) 550 278, 279, 280 high‐risk patients 242–8
bindex1.indd 697 09-12-2022 17:14:44

698 Index
membranous nephropathy (cont’d) MMF 222 monoclonal immunoglobulin

low‐risk patients 237 RTX 222 deposition disease (MIDD)
medium‐risk patients 238–42 TAC 222 478, 479, 485–6
nonimmunosuppressive corticosteroids 220–1, 220 (box) multidrug resistant organisms
therapy 237 Mini‐Nutrition Assessment 630 (MDROs) 536
blood pressure 237 minoxidil 339 Multi‐Ethnic Study of
lipid lowering 237 mixed uremic osteodystrophy 591 Atherosclerosis 63
thromboembolism mizoribine in IgA nephropathy multiple myeloma 126, 171–2,
prophylaxis 237 265–6, 265 205, 478
MEPEX trial 468 Model for End‐stage Liver Disease mycophenolate mofetil (MMF)
mercury 75, 313–14 (MELD) score 110 in acute interstitial nephritis 169
mesangiocapillary glomerulonephritis Modification of Diet in Renal birth defects 455
see membranoproliferative Disease (MDRD) 18, 237, in focal segmental glomerulosclerosis
glomerulonephritis 550, 673 (FGS) 225
Mesoamerican nephropathy 75 equation (MDRD‐75) 28, in IgA nephropathy 262,
metformin in diabetic kidney 78, 88, 371 265–6, 265
disease 377–8 molecular adsorbent recirculating in membranous
methotrexate 126 system (MARS) and albumin nephropathy 242, 244
methylmalonic aciduria and the dialysis 116 in minimal change
homocystinuria type C Molidustat 547 disease 222
protein (MMACHC) 429 Monitorización Ambulatoria de la in MPGN 278, 279, 498–501
microangiopathic hemolytic anemia Presión Arterial y Eventos Mycophenolate Mofetil versus
(MAHA) 425, 487 Cardiovasculares 333 Cyclophosphamide for
microRNA in focal segmental Monkenberg media sclerosis 576 Remission Induction of
glomerulosclerosis monoclonal antibodies 278 ANCA‐Associated
(FGS) 218–19 in focal segmental Vasculitis (MYCYC)
midodrine 107 glomerulosclerosis trial 470
in hepatorenal syndrome 113–15 (FGS) 222 (Box), 225 MYCYC trial see Mycophenolate
mineralcorticoid receptors inhibitor in monoclonal B‐cell Mofetil versus
diabetic kidney disease 381 lymphocytosis 479 Cyclophosphamide for
minimal change disease 214–27 monoclonal gammopathies 273, Remission Induction of
clinical features 216 274, 478–84 ANCA‐Associated Vasculitis
definition 214 C3 glomerulopathy associated (MYCYC) trial
epidemiology 215–17 with 486–7 myeloma cast nephropathy 172
introduction 214 complement and cytokine myocardial perfusion imaging,
pathogenesis of 218 activation associated CVD and 513
pathophysiology 217–19, 218 with 481 MYRE trial 173
CD80 219 in renal biopsy 211
immune disorders 219 thrombotic microangiopathy n
putative circulating permeability associated with 487 N‐acetylcysteine (NAC) 649
factors 217–19 monoclonal gammopathy of renal in CA‐AKI 152
treatment of 215, 220–2, 223 significance (MGRS) 478, National Health and Nutrition
alkylating agents 221, 221 (Box) 482, 483, 484 Examination
calcineurin inhibitors (CNIs) monoclonal gammopathy of Survey(NHANES) 46,
221–2, 221 (Box) undetermined significance 530, 660
CsA 221–2 (MGUS) 478 NHANES III 45, 47, 312
bindex1.indd 698 09-12-2022 17:14:45

Index 699
National Institute of Health and Care nonsteroidal anti‐inflammatory drugs biochemistry 631–2
Excellence (NICE) 50 319–24 food intake 632
National Kidney Foundation COX‐2 NSAIDs, effects on renal megestrol acetate 648–9
(NKF) 50 prostaglandins 320 nutritional assessment scores 631
National Kidney Foundation Kidney gastrointestinal complications 319 nutritional status assessment
Disease Outcomes Quality glomerular filtration rate 320 630–33
Initiative (NKF/KDOQI) natriuresis and diuresis 321 pathophysiology 625–30
3, 4, 665 nonselective and COX‐2‐selective patient interview 630
National Kidney Foundation Kidney NSAIDs, clinical syndromes resting energy expenditure 632–3
Early Evaluation Program associated with 321–2 treatment 633–48
(KEEP) 660 acute renal impairment and dietary interventions for survival,
nephrolithiasis 300 acute tubular necrosis quality of life, and overall
nephrotic syndrome 321–2 health of adults 633–36,
dyslipidemia in 553–54 acute tubulo‐interstitial 634–5
IgA nephropathy and 262 nephritis 322 nutritional interventions to slow
indication for renal biopsy 205 chronic kidney disease 322–3 progression 636–43
NSAIDS and 324 glomerulonephritis 322 dietary caloric restriction 642
nephrotoxic medications, hyperkalemia and dietary patterns 643, 644
discontinuation of, as ATN hyporeninemic dietary protein restriction
treatment 136 hypoaldosteronism 323–4 636–42, 637–9, 640–41
nephrotoxic monoclonal entire hypertension 323 dietary salt restriction 642
immunoglobulins 481 nephrotic syndrome 324 fiber intake and probiotics
nephrotoxic monoclonal free heavy renal papillary necrosis 322 642–3
chains 481 salt and water retention 323 see also protein‐energy wasting
nephrotoxic monoclonal free light postoperative use of 320–1, 321 (PEW) syndrome
chains 479–81, 480 renal blood flow 320 nutritional requirements 617–24
nesiritide in cardiorenal renal dysfunction in COX‐2 calcium, phosphorus, and
syndrome 102 trials 321 vitamin D intake
Netherlands Cooperative Study on the renal prostaglandins, recommendations 624
Adequacy of Dialysis actions of 320 energy requirements 621
(NECOSAD) 673 renal side‐effects of 319 fiber intake recommendations 624
neutrophil gelatinase‐associated renin release and potassium need for individualization
lipocalin (NGAL) 63, homeostasis 321 620, 620
91, 127–8 norepinephrine 107 overall recommendations
niacin 561 in hepatorenal syndrome 113–15 620–21, 622
nifedipine 339 North American IgA Nephropathy potassium intake recommendations
niitrofurantoin in UTI 307, 308 Trials 262 623–24
non‐Hodgkin lymphomas. 478 N‐terminal pro‐brain natriuretic protein requirements and source of
nonimmunosuppressive therapy in peptide (NT‐proBNP) 479 protein intake 621–23, 623
lupus nephritis 410 Nutrition Screening Tool 630 sodium intake and fluid
nonmuscle myosin heavy chain‐IIA nutritional assessment scores 631 balance 623
(NMMHC‐IIA) 219 nutritional disorders in chronic nutritional status assessment 630–33
nonproteinuric NDD‐CKD, kidney disease 617–49, anthropometry and body
dyslipidemia and 552 618–19 composition 630–31
nonstatin lipid lowering in anabolic agents 649 bioelectrical impedance
CKD 559 anti‐inflammatory agents 648–9 analysis 631
bindex1.indd 699 09-12-2022 17:14:45

700 Index
nutritional status assessment (cont’d) treatment 487–9 as initial kidney replacement

handgrip strength 631 clone‐directed therapy 488 therapy 674
mid‐arm circumference 631 indication of therapy 487 patient survival 674–7, 674
skinfold thickness 631 symptomatic measures 488–9 methodological issues 675–6
weight and body mass index see also entries under monoclonal patient characteristics 675, 676
630–31 gammopathy practice patterns 674–5
parathyroid hormone, serum 603–5 quality of life and biochemical
o target levels 608, 608 (box) parameters 676
obesity 582 parathyroidectomy 605–6 preparation for 665–6
obstructive nephropathy 103–4 paricalcitol 603, 603(Box) PEXIVAS trial see Plasma Exchange
ocrelizumab 416–17 PATHWAY trial see Prevention And and Glucocorticoids for
octreotide 107, 299 Treatment of Hypertension Treatment of Anti‐
in hepatorenal syndrome With Algorithm‐based Neutrophil Cytoplasm
113–15 therapy (PATHWAY) Antibody (ANCA)
ODYSSEY LONG TERM PCSK9 Inhibitors 559–60 Associated Vasculitis
study 559–60 pediatric IgA nephropathy (PEXIVAS) trial
omega (n‐3) fatty acids 561–62 corticosteroids in 264 phlorizin 374
omeprazole 321 cyclophosphamide with/without phosphate binders 599–603,
OMS721 266 azathioprine 264–5 600 (Box)
ONTARGET 380 evidence in 259 aluminum 600
opportunity costs 50 with isolated invisible hematuria dialysis duration and frequency
osteitis fibrosa 593 and proteinuria 260 600–603
osteitis fibrosis cystica 591 MMF and mizoribine dietary modification 603
osteomalacia 591 265–6, 265 nonaluminum phosphate
Oxford Classification of IgA with nephrotic syndrome 262 binders 600, 601–2
Nephropathy 258 prognosis 256 phospholipase A2 receptor
oxypurinol 174 with rapidly declining in membranous nephropathy 236
GFR 261–2 phosphorus
p recommendations for 258 nutritional intake
pain in chronic kidney disease‐ with slowly progressive renal recommendations 624
mineral and bone disorder impairment 262–3 target levels 608, 608 (box)
598–9 with visible hematuria 259, 260 serum in chronic kidney disease‐
pamidronate 173, 607 (Box) pediatric nephrotic syndrome and mineral and bone
paraprotein‐associated kidney adult FSGS 217 disorder 594
disorders 478–89 prognosis 219 physical activity
diagnosis 482–3 AKI 220 in autosomal dominant polycystic
crystal‐storing histiocytosis 483 infection 220 kidney disease 297
light‐chain cast thromboembolic events 220 in CKD 520
nephropathy 482, 482 peginesatide 542, 546 PICARD study see Program to
light‐chain proximal pentoxifylline 649 Improve Care in Acute
tubulopathy 482, 483 percutaneous renal denervation in Renal Disease
tubulointerstitial lesions 482 hypertension 340 (PICARD) study
diagnostic evaluation 479 peripheral blood flow cytometry 479 pioglitazone 378–9
glomerular lesions 483–87 peritoneal dialysis 343, 554, 671–77 piperacillin‐tazobactam in acute
pathophysiology 479–81 acute 185 tubular necrosis 135
in renal biopsy 211 in dyslipidemia 553, 554 PIRRT/SLED 186, 189, 196
bindex1.indd 700 09-12-2022 17:14:45

Index 701
pitavastatin in chronic kidney preeclampsia in patients urinary indices 98–9, 98

disease 558 with CKD 446 urine output 98
plain radiography of vascular pregnancy in patients see also cardiorenal syndrome;
calcification 598 with CKD 446 obstructive nephropathy
plasma exchange (PE) therapy 429 epidemiology 444–5 PRESERVE trial see Prevention of
in ANCA‐associated vasculitis acute kidney injury 445 Serious Adverse Events
468‐9, 469 chronic kidney disease 444–5 Following Angiography
in complement‐mediated dialysis 445 (PRESERVE) trial
TMA 436 transplantation 445 PRESION study 321
Plasma Exchange and Glucocorticoids fertility and CKD 445–6 Prevention And Treatment of
for Treatment of Anti‐ focal segmental Hypertension With
Neutrophil Cytoplasm glomerulosclerosis and 227 Algorithm‐based therapy
Antibody (ANCA) introduction 444 (PATHWAY) 339
Associated Vasculitis lupus nephritis and 419 PATHWAY‐2 trial 339
(PEXIVAS) trial 469 pathophysiology 445–6 Prevention of Contrast Renal Injury
Pneumocystis jirovecii 410, 472 physiological changes with with Different Hydration
POEMS syndrome 481, 487 pregnancy 445 Strategies (POSEIDON)
polyclonal immunoglobulin prognosis 446–51 trial 153
deposition disease 479 kidney function 447, 448–50 Prevention of Serious Adverse Events
polycystic kidney disease 4 maternal kidney disease 446–7 Following Angiography
polycystic liver disease 300 pregnancy and dialysis 447–51 (PRESERVE) trial
pomalidomide 488 pregnancy outcomes after kidney 152, 153
POSEIDON trial 153 transplantation 451 PROactive study see PROspective
positron emission tomography (PET), primary disease 447 pioglitAzone Clinical Trial
CVD and 513 transplant outcomes 451 (PROactive) study
potassium renal biopsy 211 probiotics 642–3
hypertension and 334 preimplantation genetic diagnosis Program to Improve Care in Acute
nutritional intake (PGD) in autosomal Renal Disease (PICARD)
recommendations 623–24 dominant polycystic kidney study 123, 124
prasugrel 383 disease 295 proliferative glomerulonephritis
pravastatin 558 prepregnancy counseling and with monoclonal
prealbumin 631 planning 451–2 immunoglobulin deposits
PRECISION study 319 prerenal azotemia 96 (PGNMID) 478, 481,
predialysis 4 prerenal failure 96–103 486, 487
prednisolone in MCD 220 causes 97 prolonged‐intermittent renal
prednisone crystalloid versus colloid replacement therapy
in dyslipidemia 555 fluids 100 (PIRRT) 185
in MCD 220 CVP and PAOP 97–8 prolyl hydroxylase inhibitors
preeclampsia in patients diagnosis 97 (PHI) 542
with CKD 446 goal‐directed hemodynamic in anemia 547
pregnancy 444–56 management and fluid PRO‐PKD score in autosomal
acute kidney injury in 445 therapy 99–100 dominant polycystic kidney
diagnosis 446 high‐chloride versus balanced disease (ADPKD) 294
kidney biopsy 446 fluids 100–101 PROspective pioglitAzone Clinical
markers of kidney function in treatment 99 Trial (PROactive)
pregnancy 446 urinalysis 98 study 378
bindex1.indd 701 09-12-2022 17:14:45

702 Index
prostaglandins, renal, actions of 320 r RENAAL trial see Reduction of

effects on COX‐2 NSAIDs on 320 radiocontrast toxicity in renal artery Endpoints in NIDDM with
protein‐energy wasting (PEW) stenosis 360–1 the Angiotensin II
syndrome 617, 623, 624 raloxifene 608 Antagonist Losartan
causes of 627–30, 629 ramipril 380 (RENAAL) trial
catabolic effect of uremia Ramipril Efficacy In Nephropathy renal artery stenosis
628–9, 628, 629 (REIN) study 581 cholesterol embolism and
decreased protein and energy Randomized Evaluation of Normal radiocontrast
intake 627–8 versus Augmented Level toxicity 360–1
loss of nutrients and catabolic (RENAL) Replacement correction of, versus drug
effects of dialysis 629–30 Therapy study 195 treatment 355, 356
definition 624–5, 625 Randomized Evaluation of Sodium exclusion of renoparenchymatous
epidemiology of 624–5, 626 Dialysate Levels on Vascular disease 358–60, 360
mortality, morbidity, and quality of Events [RESOLVE] 345 improving results of
life 633 Randomized Olmesartan and correction 360
non‐nutritional Interventions 648, Diabetes Microalbuminuria pathophysiology ofs 354
648 (Box) Prevention (ROADMAP) quantification of stenosis and
treatment in CKD 643–48, 645 trial 380 estimation of functional
dietary counseling 646 Randomized Trial of Daily Oral vs relevance 358
intradialytic parenteral nutrition Pulse Cyclophosphamide as screening for 355–8, 359
in HD 646–8 Therapy for ANCA‐ renal artery thrombosis 180
intraperitoneal infusion of amino Associated Systemic renal biopsy 205–12
acids in PD 648 Vasculitis (CYCLOPS) complications 210–11
oral nutritional supplements trial 467 monoclonal gammopathies and
646, 647 RAP1GAP 219 paraprotein disease 211
protein equivalent of nitrogen rasburicase 174 pediatrics 211
appearance 632 RECORD study 379 pregnancy 211
protein intake Reduction of Endpoints in NIDDM transplant biopsies 211
in autosomal dominant polycystic with the Angiotensin II contraindications 207
kidney disease Antagonist Losartan epidemiology 207
(ADPKD) 296 (RENAAL) trial 380, indications 205–7, 206
restriction of, dietary 636–42, 581, 582 post‐procedure care 210
637–9, 640–41 REIN study 581 preprocedure care 208–10
protein‐to‐creatinine ratio (PCR) 28 relapse, definition 215 anticoagulants 209
proteinuria, isolated invisible, IgA steroid dependent 215 antiplatelet agents 209
nephropathy and 260 REMAIN study see Remission‐ bleeding prediction scores 210
Proteus mirabilis 305 Maintenance Therapy in desmopressin 209–10
proximal tubulopathy 488 Systemic Vasculitis investigations 209
psychological care in autosomal (REMAIN) study prebiopsy evaluation 208–9
dominant polycystic kidney remission, procedure 207–8
disease 295–6 definition 214–15 imaging guidance 207–8
complete 214 needle size 208
q partial 215 positioning 207
quality‐adjusted life years (QALYs) Remission‐Maintenance Therapy in practitioner 208
50, 55 Systemic Vasculitis sample preparation 208
quinine 430 (REMAIN) study 472 technique 207
bindex1.indd 702 09-12-2022 17:14:45

Index 703
renal functional reserve (RFR) and renin‐angiotensin‐aldosterone system eGFR 28

acute tubular necrosis 129 (RAAS) blockade hypertension 34–5
renal infarction 180 therapy 45 race 34
renal insufficiency 4 renoparenchymatous disease sex 34
renal osteodystrophy 591 358–60, 360 Risk, Injury, Failure, Loss, and
renal papillary necrosis, NSAIDS renovascular azotemia 354 End‐stage (RIFLE)
and 322 renovascular disease 354–61 classification system for
renal replacement therapy 185–98 terminology 354 AKI 87–8
access to 72 renovascular hypertension 354 RITUXILUP regimen 416
continuous (CRRT) 150, 185, 186, reverse epidemiology 549, 550 rituximab 214, 278, 413, 488
187, 188, 189, 191, 192, RIFLE (Risk, Injury, Failure, Loss, in ANCA‐associated
194–6, 194, 195 and End‐stage) vasculitis 469–70, 469, 470
circuit considerations 197–8 criteria 87–8, 124 for C3 glomerulopathy 282
anticoagulation strategies 198 risedronate 607 (Box) in focal segmental glomerulosclerosis
dialysis fluids/solutions and risk prediction in chronic kidney (FGS) 225
buffers 198 disease 60–7, 61 in IgA nephropathy 266
dialysis vascular access 197 AKI in predicting future CKD 63 in membranous nephropathy
dialyzer membranes 197–8 CKD risk scores 63–4 242, 245
dosage 193–6 eGFR and cardiovascular risk 67 in minimal change disease 222
CRRT dose 194–6, 194, 195 future considerations 67 in pregnancy 455
intermittent hemodialysis incidence 60 Rituximab for the Treatment of
dose 193–4 kidney failure risk equations 64, Wegener’s Granulomatosis
PIRRT/SLED dose 196 64, 65, 66 and Microscopic Polyangiitis
global incidence 19 prediction models 64–7, 65 (RAVE) trial 469
in hepatorenal syndrome 116 dialysis access and modality rituximab versus cyclophosphamide
indications for 189–90 planning 66–7 in ANCA‐associated renal
and lupus nephritis 419–20 interdisciplinary care 66 vasculitis (RITUXIVAS)
modalities 186, 187 predicting cardiovascular disease trial 469
choice of 186–9, 188, 188 in CKD patients 67 RITUXIVAS trial 469
timing of dialysis initiation 190–3 transition from primary care to ROADMAP trial see Randomized
conventional/delayed/late nephrology 65–6 Olmesartan and Diabetes
RRT 190–1 proteinuria/albuminuria and Microalbuminuria
definition 190 cardiovascular risk 67 Prevention (ROADMAP) trial
early acute RRT initiation 192 see also biomarkers rofecoxib 319, 320, 323
early/preemptive RRT 191–2 risk stratification 26–40 rosiglitazone 379
renal replacement therapy (RRT) epidemiology of CKD and kidney rosuvastatin 558
registries 16, 18 failure 26–7 in CA‐AKI 152
renal transplantation see kidney pathophysiology of CKD roxadustat 547
transplantation Progression 27–8
renal vein thrombosis 180 prognostic factors for progression to s
renalism 149 kidney failure 28–36, 29–32 sacubutril 339
renin angiotensin system acute kidney injury 33 Safety of Estrogens in Lupus
blockade 338 age 33–4 Erythematosus National
renin‐angiotensin system inhibitors 27 albuminuria 28–33 Assessment–SLE Disease
Renin‐Angiotensin System Study cardiovascular disease 35–6 Activity Index
(RASS) 380 diabetes mellitus 35 (SELENA‐SLEDAI) 417
bindex1.indd 703 09-12-2022 17:14:45

704 Index
salt single‐photon emission computed dosage 558

in CKD 520 tomography (SPECT), in hemodialysis 556
restriction, dietary 642 CVD and 513 in peritoneal dialysis 556–7
in autosomal dominant sirolimus 299, 555 progression of CKD 557–8
polycystic kidney in focal segmental in renal transplant 557, 558
disease 296 glomerulosclerosis STEC‐HUS 425, 429
retention, NSAIDS and 323 (FGS) 225 management 435
Sapphire study 128 sitagliptin (TECOS) 377 pathophysiology 431
saxagliptin (SAVOR‐TIMI) 377 Sjögren’s disease nephropathy 494 prevention 434
schistosomiasis 73 skeletal complications 598–9 prognosis 433
screening for chronic kidney SLED/PIRRT 195 therapy 434–5
disease 44–56 sleep apnea, nocturnal antibiotics 435
cost‐effectiveness ratios 44 hypertension and 333 controversy 435
cost‐effectiveness studies 50–6, SOAP study 191 emerging therapy and future
54, 55 sodium perspectives 435
current guideline recommendations in acute tubular necrosis 127 IV hydration 434–5
for 49–50, 51–3 dietary 582 steroids resistant nephrotic syndrome
detection and prevention hypertension and 334 (SRNS) 217
programs 45–7 intake 623 Streptococcus pneumoniae 429
CKD detection program 47, 49 sodium‐glucose linked co‐transport‐1 Study of Heart and Renal Protection
high‐risk populations 47, 49 and ‐2 (SGLT1 and 2) (SHARP) trial 382, 555,
national capacity for CKD inhibitor in diabetic kidney 556, 557, 558, 559
screening 47, 48 disease 374–5, 375 Subjective Global Assessment
methods 45 sodium‐glucose transporter inhibitors (SGA) 624, 631
opportunistic screening 44 (SGLT2i) 339 sulfadiazine crystal nephropathy 178
targets for 44–5 soluble urokinase‐type plasminogen sulfamethoxazole in UTI 307, 308
secondary hyperparathyroidism activator receptor sulfonylureas in diabetic kidney
(SHPT) 625 (suPAR) 63, 217 disease 378
SELENA‐SLEDAI 417 solvent nephropathy 316 suPAR in focal segmental
sepsis 99 somatostatin analogs 299 glomerulosclerosis
septic shock 99 sparsentan 266 (FGS) 217
serum creatinine 632 spironolactone 339 Supportive Versus
sevelamer 559, 600 SPRINT see Systolic Blood Pressure Immunosuppressive
SF 36 544 Intervention Trial (SPRINT) Therapy for Progressive IgA
Shiga toxin neutralizing agents in Standard vs. Accelerated Initiation of Nephropathy (STOP‐IgAN)
STEC‐HUS 435 RRT in Acute Kidney Injury trial 263
Shiga toxin‐producing Escherichia (STAART‐AKI), 192–3 surveillance of chronic kidney
coli‐related HUS see Staphylococcus 305 disease 14–22
STEC‐HUS Staphylococcus aureus 464 active 16, 17
Shigella dysenteriae 429 Staphylococcus saprophyticus 305 benefits of 14
silicon 315–16 STAR trial 576 cross‐sectional surveys
silicoproteinosis 315 statins of eGFR 16
simultaneous liver‐kidney (SLK) 115 adverse outcome/safety global burden 18–21
simvastatin 382 in CKD 558 health systems data 19, 20
in chronic kidney disease 558 in CVD 515, 519 high‐risk populations, testing of
plus ezetimibe 382, 555 and diabetes risk 559 19, 20
bindex1.indd 704 09-12-2022 17:14:45

Index 705
measurement error 18 Taiwan’s National Health Insurance complement‐mediated 429, 430

outreach screening 16 Research Database 534 diagnosis 427–9, 432–3
population‐based surveys Tamm‐Horsfall protein 172, 488 diacylglycerol kinase ε
19–21, 21 terlipressin 107 (DGKE)‐related 429
routine health records 16–18 in hepatorenal syndrome 113 nomenclature 425, 426
types of 14–16, 15 thalidomide 172, 488 primary
surveillance, definition of 14 The Study to Evaluate the Use of noncomplement‐
SUSTAIN‐6 376 Rosuvastatin in Subjects on mediated 429
Swedish Heart Failure Registry and Regular Hemodialysis: An TMA spectrum 425–30
the Asian Sudden Cardiac Assessment of Survival and ticagrelor 383
Death in Heart Failure Cardiovascular Events tissue inhibitor of metalloproteinase‐2
registry 507 (AURORA) 556 (TIMP‐2) 128
sympathetic renal nerve ablation in theophylline in CA‐AKI 151 tissue inhibitor of metalloproteinase, 2
hypertension 339–40 Therapeutic Evaluation of Steroids in and insulin‐like growth
Symplicity HTN‐3 trial 340 IgA Nephropathy Global factor‐binding protein, 7
systemic lupus erythematosus (TESTING) study 264 (TIMP‐2*IGFBP‐7) 91, 193
(SLE) 400, 429 therapeutic plasma exchange tolvaptan 297–9, 300
diagnosis 404 (TPE) 172–3 TORC1 inhibitors 299
nephritis in pregnancy 446, 447 thiazide‐like diuretics 338 trandolapril 380
renal biopsy and 205 thiazolidinediones in diabetic kidney transjugular intrahepatic
Systemic Lupus International disease 378–9 portosystemic shunt (TIPS)
Collaborative Clinics thrombocytopenia 425 in hepatorenal syndrome
(SLICC) [1] criteria 400 STEC‐HUS and 434 115–16
systemic sclerosis 429 thrombocytopenic purpura Translational Research Investigating
Systolic Blood Pressure Intervention (TTP) 425–9 Biomarkers and End Points
Trial (SPRINT) 332, 338, thrombospondin type‐1 domain‐ for Acute Kidney Injury
340, 519, 581, 583 containing 7A (THSD7A) (TRIBE‐AKI)
Systolic Hypertension in Europe in membranous nephropathy 236 consortium 126
(Syst‐Eur) trial 379 thrombotic microangiopathy TREAT see Trial to Reduce
Systolic Hypertension in the Elderly (TMA) and 430 cardiovascular Events with
Program (SHEP) 340 treatment 451–5 Aranesp Therapy (TREAT)
CKD in pregnancy 452 Treating to New Targets (TNT)
t dialysis management 455 study 558
T cells hypertension 453 Treatment of HDL to Reduce the
focal segmental immunosuppression 453–55 Incidence of Vascular Events
glomerulosclerosis and 219 management of transplanted (HPS‐THRIVE) trial 561
minimal change disease 219 women 455 Trial to Evaluate Cardiovascular and
tacrolimus optimize maternal health pre Other Long‐term Outcomes
in focal segmental conception 452–3 with Semaglutide in Subjects
glomerulosclerosis pregnancy management 453, 454 with Type 2 Diabetes trial
(FGS) 225 prepregnancy counseling and (SUSTAIN‐6) 376
in membranous nephropathy 242 planning 452–3 Trial to Reduce cardiovascular Events
in minimal change disease 221 thrombotic microangiopathy (TMA) with Aranesp Therapy
in pregnancy 455 association with monoclonal (TREAT) 542, 544, 545, 546
Taiwan Renal Registry Data gammopathy 487 triglyceride 560–62
System 550 classification 425 trimethoprim in UTI 307
bindex1.indd 705 09-12-2022 17:14:45

706 Index
troponin T 479 urokinase plasminogen activator vasopressin receptor antagonists in

tubulointerstitial disease 4 surface receptor (uPAR) 217 autosomal dominant
tubulointerstitial fibrosis 27–8 uromodulin (UMOD) 63, 479–81 polycystic kidney disease
Cairo–Bishop diagnostic criteria US National Kidney Foundation (ADPKD) 297–8
173, 173 Dialysis Outcomes Quality venous thromboembolism
management 174–9 Initiative (K/DOQI) (VTE) 220
pathophysiology 173–4 Workgroup on Peritoneal verapamil 339, 380
tumor lysis syndrome 126, 172, Dialysis 671 Veterans Affairs Diabetes Trial
173, 173 US Preventive Services Task Force (VADT) 372, 374
twenty‐four‐ hour dietary recalls 632 (USPSTF) 49 Veterans Affairs High‐density
lipoprotein Intervention
u v Trial (VA‐HIT) 382, 560
UK Clinical Practice Research VA/NIH Acute Renal Failure Trial vitamin D 603
Datalink 378 Network (ATN) study 195 agents 603 (Box), 604
UK Obstetric Surveillance Study 445 vaccinations 536 analogs 603
UK Renal Association 50 live 536 in chronic kidney disease‐mineral
UKPDS 372 pneumococcal 534 and bone disorder 594
United States Renal Data System routine childhood 536 intake recommendations 624
(USRDS) 557, 658 vadadustat 547 nutritional 603–5
uric acid crystal nephropathy 178 VADT see Veterans Affairs Diabetes vitamin D3 (cholecalciferol)
urinary dipstick 28 Trial (VADT) 603 (Box)
urinary neutrophil gelatinase‐ valdecoxib 319
associated lipocalin valsartan 339 w
(uNGAL) 111–12 vancomycin in acute tubular Waldenström macroglobulinemia
urinary tract infection, necrosis 135 478, 482
adult 305–9, 306 VA‐NEPHROND 380 warfarin sodium 180
epidemiology 305–6 Vascepa 561 water retention, NSAIDS and 323
introduction 305 vasoconstrictors in hepatorenal Wegener’s granulomatosis 315
pathophysiology 306–7 syndrome 113–15, 114 weight loss 582
prevention 308–9 Vasodilation in the Management of
prognosis 307–9 Acute Congestive Heart z
treatment 306, 307–8 Failure trial 102 zoledronate 173, 607 (Box)
bindex1.indd 706 09-12-2022 17:14:45

09-12-2022 15:10:23
ffirs_Vol2.indd 1
Second Edition
VOLUME II
Edited by
Jonathan C. Craig, MBChB, DipCH, MMed(Clin Epi), PhD, FAHMS

Matthew Flinders Distinguished Professor
Vice President and Executive Dean, College of Medicine and Public Health
Flinders University
Adelaide, Australia
Donald A. Molony, MD
Professor of Medicine
Distinguished Teaching Professor of the University of Texas System
Division of Renal Diseases and Hypertension AND
Center for Clinical Research and Evidence-based Medicine
McGovern Medical School University of Texas, Houston, TX, USA
Giovanni F.M. Strippoli, MD, PhD, MPH, MM (Epi)

Professor of Nephrology, Department of Emergency and Organ Transplantation – University of Bari
Bari, Italy;
Adjunct Professor of Epidemiology, School of Public Health
Sydney, NSW, Australia
With section editors
Aminu Bello Liz Lightstone

Mark Canney Paul Palevsky
Sara Davison Suetonia Palmer
Carmel Hawley Susan Samuel
David Johnson Allison Tong
Adeera Levin Germaine Wong
ffirs_Vol2.indd 3 09-12-2022 15:10:23

This edition first published 2023
© 2023 John Wiley & Sons Ltd
Edition History
Blackwell Publishing Ltd (1e, 2009)
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material
from this title is available at http://www.wiley.com/go/permissions.
The right of Jonathan C. Craig, Donald A. Molony, and Giovanni F.M. Strippoli to be identified as the authors of the editorial material in this
work has been asserted in accordance with law.
Registered Offices
John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
Editorial Office
9600 Garsington Road, Oxford, OX4 2DQ, UK
For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.
Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that appears in standard print versions
of this book may not be available in other formats.
Limit of Liability/Disclaimer of Warranty

The contents of this work are intended to further general scientific research, understanding, and discussion only and are not intended and
should not be relied upon as recommending or promoting scientific method, diagnosis, or treatment by physicians for any particular patient.
In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating
to the use of medicines, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert
or instructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indication of usage and
for added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no
representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties,
including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or
extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or
product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors
endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with
the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be
suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this
work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be
liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
Library of Congress Cataloging-in-Publication Data

Names: Molony, Donald A., editor. | Craig, Jonathan C., editor. |
Strippoli, Giovanni F.M., editor.
Title: Evidence-based nephrology / edited by Donald A. Molony, Jonathan C.
Craig, Giovanni F.M. Strippoli; with section editors, Aminu Bello [and 11 others].
Ýescription: Second edition. | Hoboken, NJ: Wiley-Blackwell, 2021. |
Includes bibliographical references and index.
Identifiers: LCCN 2020043376 (print) | LCCN 2020043377 (ebook) | ISBN
9781119105923 (cloth) | ISBN 9781119105930 (adobe pdf) | ISBN
9781119105947 (epub)
Subjects: MESH: Kidney Diseases | Evidence-Based Medicine–methods
Classification: LCC RC903 (print) | LCC RC903 (ebook) | NLM WJ 300 |
DDC 616.6/1–dc23
LC record available at https://lccn.loc.gov/2020043376
LC ebook record available at https://lccn.loc.gov/2020043377
Cover Design: Wiley

Cover Images: Jose Luis Calvo/Shutterstock, SEBASTIAN KAULITZKI/SCIENCE PHOTO LIBRARY
Set in 9.5/12.5pt STIXTwoText by Straive, Pondicherry, India
ffirs_Vol2.indd 4 09-12-2022 15:10:23

v
Contents
VOLUME I
Preface ix
Part 1 Epidemiology 1
1 An Introduction to the Epidemiology of Chronic Kidney Disease 3
2 The Surveillance and Burden of Chronic Kidney Disease 14
3 Progression of Chronic Kidney Disease: An Evidence-based Approach to Risk Stratification 26
4 Screening for Chronic Kidney Disease 44
5 Risk Prediction in Chronic Kidney Disease 60
6 Chronic Kidney Disease in Disadvantaged Populations 72
Part 2 Acute Kidney Injury 85
7 Overview / Definition, Classification, and Epidemiology of Acute Kidney Disease 87
8 Pre-Renal Failure and Obstructive Disease 96
9 Hepatorenal Syndrome 107
10 Acute Tubular Necrosis 123
11 Iodinated Contrast and Acute Kidney Injury 145
12 Miscellaneous Etiologies of Acute Kidney Injury 163
13 Renal Replacement Therapy in Acute Kidney Injury 185
Part 3 Primary Diseases of the Kidney 203
14 Renal Biopsy 205
15 Minimal Change Disease and Focal Segmental Glomerulosclerosis in Adults 214
16 Membranous Nephropathy 235
ftoc_V2.indd 5 09-12-2022 15:13:33

vi Contents
17 IgA Nephropathy in Adults and Children 254
18 Membranoproliferative Glomerulonephritis 272
19 Autosomal Dominant Polycystic Kidney Disease 288
20 Urinary Tract Infections 305
21a Toxic Nephropathies: Environmental Agents and Metals 311
21b Toxic Nephropathies: Nonsteroidal Anti-Inflammatory Drugs 319
Part 4 Secondary Diseases of the Kidney 329
22 Hypertension 331
23 Renovascular Disease 354
24 Secondary Diseases of the Kidney: Diabetic Nephropathy 366
25 Lupus Nephritis 400
26 Hemolytic Uremic Syndrome 425
27 Pregnancy 444
28 ANCA-associated Vasculitis 461
29 Paraprotein-associated Kidney Disorders 478
30 Primary Immune Complex mediated Membranoproliferative Glomerulonephritis and

C3 Glomerulopathies: A Clinical Approach 494
Part 5 Chronic Kidney Disease and Complications 505
31 Cardiovascular Disease and CKD 507
32 Infection and CKD 526
33 Treatment of Anemia in Chronic Kidney Disease 542
34 Dyslipidemia in Chronic Kidney Disease 549
35 Chronic Kidney Disease and Hypertension 573
36 Chronic Kidney Disease-Mineral and Bone Disorder 589
37 Nutritional Disorders in Chronic Kidney Disease: Pathophysiology, Detection, and Treatment 617
38 Preparation for Dialysis 658
39 Choice of Hemodialysis or Peritoneal Dialysis for Kidney Replacement Therapy 671
Index 681
ftoc_V2.indd 6 09-12-2022 15:13:33

Contents vii
VOLUME II
Preface ix
Part 6 Hemodialysis 1
40 Modalities of Extracorporeal Therapy 3
41 Dialysis Dose and Adequacy for Hemodialysis 20
42 General Management of the Hemodialysis Patient 34
43 Infections in Hemodialysis Patients 48
44 Vascular Access for Hemodialysis 66
Part 7 Peritoneal Dialysis 91
45 Small Solute Clearance in Peritoneal Dialysis 93
46 Salt and Water Balance 113
47 Solutions 127
48 Peritoneal Dialysis: Infections 138
49 Urgent-start Peritoneal Dialysis 156
50 Peritoneal Dialysis Catheter Insertion 170
Part 8 Supportive Care 179
51 Overview of Kidney Supportive Care 181
52 Symptoms 194
53 Prognostication in Advanced Chronic Kidney Disease 208
54 Advance Care Planning 216
55 Conservative Kidney Management and Dialysis Withdrawal 227
Part 9 Transplantation 247
56 Evaluation of the Living Donor Kidney 249
57 The Impact of Deceased Donor Quality and Outcomes after Kidney Transplantation 257
58 Early Medical and Surgical Complications After Kidney Transplantation 271
59 Infections After Kidney Transplantation 294
ftoc_V2.indd 7 09-12-2022 15:13:33

viii Contents
Part 10 Electrolytes and Acid-Base Disorders 301
60 Electrolyte Disorders 303
61 Metabolic Evaluation and Prevention of Kidney Stone Disease 322
Part 11 Pediatrics 333
62 Growth, Nutrition, and Development 335
63 Bone Disease in Children with Chronic Kidney Disease 356
64 Anemia 379
65 Peritoneal Dialysis in Children 399
66 Hemodialysis 412
67 Urinary Tract Infections in Children 426
68 Henoch–Schonlein Purpura Glomerulonephritis and IgA Nephropathy in Children 439
69 Hereditary Nephritis in Children 451
70 Shigatoxin-related Hemolytic-uremic Syndrome 463
Part 12 Patient-centered Care and Outcomes 473
71 Shared Decision-making 475
72 Fatigue 488
73 Depression 499
74 Pain 517
75 Neurocognitive Disorders 539
76 Pruritus 551
77 Sexual Dysfunction 570
78 Family and Caregiver Support 584
Index 590
ftoc_V2.indd 8 09-12-2022 15:13:33

ix
Preface
The synthesis of the totality of evidence in kidney disease methods assist the summarizing and interpreting of evi-
was the original challenge we undertook when we con- dence in key areas of decision-making, which is useful to
ceived of the first edition of this evidence-based nephrol- patients, healthcare providers, and policy-makers.
ogy (EBN) textbook, thereby providing students of Using these methods, organizations including the
nephrology and practicing clinicians with a single conveni- Cochrane Kidney and Transplant group and several guide-
ent source of clinical evidence that had been passed line agencies have committed on a large scale to summa-
through an evidence-based filter. More than 12 years have rize the evidence and provide timely updates of established
passed since the first edition, and so it is timely to revisit and emerging new evidence, making research findings
the challenge in the form of a second edition. The reasons more manageable and reliable for end users. Much of this
for a second edition are principally three: the scale of new work has been included in this edition of the textbook. We
available data is substantial, newer methods of research hope this will allow the end-user to both access best evi-
synthesis have been developed, and there is a renewed dence from this diversity of resources as summarized by
focus on consumer engagement in general, including in chapter authors and to become more familiar with how the
the evaluation and management of kidney disease. newer techniques and resources contribute to a practice
that is informed by best evidence.
Scale of new data
Compared to the time of the first edition, the amount of Consumer engagement
new data is impressive. There are today more than 15 000 Recently, there has been increasing recognition of the criti-
randomized controlled trials in nephrology published in cal importance of engaging the end-users in research to
approximately 30 000 reports, 11 300 more than in 2009, ensure relevance and uptake. In particular there have been
resulting in more confidence and precision around the esti- large-scale efforts to engage patients in all stages of
mates of intervention effects and new evidence in areas research, including the evidence synthesis process and the
that were previously mostly evidence-free. These trials dissemination of findings. For example, the Cochrane
have also been summarized in systematic reviews, with Kidney and Transplant group now has a patient editor,
over 6500 additional reviews in Medline since the first edi- whose role includes translating scientific outputs into plain
tion. Much of this new evidence is incorporated in three language summaries to simplify the “medical jargon” and
new sections of this textbook and multiple either new or make evidence easily accessible to people without a medi-
entirely updated chapters. cal background, as well as prioritizing review topics. The
Standardized Outcomes in Nephrology (SONG) initiative
Novel methods was launched in 2014 to establish the core outcomes to be
Methods to assess the methodological quality of both sys- reported in all trials in chronic kidney disease, based on a
tematic reviews and other study designs, including nonran- consensus among patients, caregivers, health profession-
domized (case-control or cohort studies) and diagnostic test als, and policy-makers. This initiative has emphasized
studies, have developed since the first edition. In addition, patient-centered research and improved clinical outcomes
the number of syntheses of cohort studies has increased in nephrology, ensuring that the outcomes measured and
and the technique of network meta-analysis for comparing reported in clinical trials and other forms of research are
more than two interventions both directly (head to head) relevant and meaningful to patients and increase the
and indirectly (via a common comparator) has become acceptability, transparency, and generalizability of the
more developed and is commonly encountered. These results in this population. New studies assessing these
fpref_V2.indd 9 09-12-2022 15:12:41

x Preface
utcomes are now being developed, with advancement of

o and comprehensiveness has added complexity in the meth-
knowledge which could be incorporated into this edition of ods and may lead to residual uncertainty in the estimate of
the textbook. intervention effects. Finally, we focused on empiric evi-
In short, compared to the previous version published in dence for management decisions and so pathophysiologi-
2009, this updated textbook includes 20 new chapters and cal mechanisms of chronic kidney disease and its treatment
three new sections, covering supportive care and patient- were not considered in detail. We have attempted to include
centered care and outcomes. In addition, every chapter economic evaluations where particularly relevant for
that was in the first edition has been extensively updated. national and international management guidelines.
Specifically, the new textbook covers epidemiology, acute In the true spirit of evidence-based medicine, we hope
kidney injury, primary diseases of the kidney, secondary that this edition of Evidence-based Nephrology will con-
diseases of the kidney, chronic kidney disease and compli- tinue to push the specialty toward greater reliance on the
cations, hemodialysis, chronic kidney disease stage 5, peri- totality of evidence and the generation and utilization of
toneal dialysis, supportive care, transplantation, evidence that is the least biased and the most precise to
electrolytes and acid-base disorders, and patient-centered inform clinical decision making. To the material detailed in
care and outcomes for both adult and pediatric patients. all the chapters we would expect that our readers will add,
This effort was undertaken by existing and new authors through their own judicious application of evidence-based
and section editors, whom we would like to thank for their medicine principles, their local context, and incorporate
extensive work. their patients values and preferences, given that objective
Giovanni Strippoli has joined as the third co-editor of evidence is rightly only one consideration in the delivery of
Evidence-based Nephrology. Altogether, we hope that we true patient-centered care. Additionally, a textbook like
have produced an even better evidence-based nephrology this can at best be only one of several resources for the
tool, where each chapter provides a clear foundation of the evidence-based medicine practitioner, and should be sup-
topic that is supported by the best current evidence. plemented with current methodologically rigorous and
We believe that this updated textbook more broadly cov- transparent clinical practice guidelines. We hope this effort
ers available evidence and addresses crucial clinical ques- will provide a core resource for the evidence-based neph-
tions regarding the treatment and care of people with all rology practitioner who is otherwise limited by time con-
stages of chronic kidney disease, including people under- straints from researching every question that may arise
going any form of dialysis (hemodialysis or peritoneal dial- daily in the care of patients.
ysis), those requiring kidney transplantation, and pediatric We certainly wish that students and evidence-based
patients. nephrology practitioners will benefit from this updated
Inherently a textbook of this nature will always manifest edition in the search for answers to questions that arise in
potential limitations. We acknowledge that a textbook is daily care and their ambition to deliver the best care
unable to collect all new evidence in real time, and, like all possible.
of healthcare internationally, the COVID-19 pandemic did
impact this book, specifically the publication timelines as Jonathan C. Craig
contributors had to contend with major challenges. The Donald A. Molony
inclusion of nonrandomized studies to maximize relevance Giovanni F.M. Strippoli
fpref_V2.indd 10 09-12-2022 15:12:41

xi
List of Contributors
Diego Aguilar Bourne L. Auguste

PRS Population Health, Clinical Trial Service Division of Nephrology
Unit and Epidemiological Studies Unit, BHF Sunnybrook Health Sciences Centre
Centre of Research Excellence, Nuffield Department Toronto
of Population Health Canada
University of Oxford
UK Sevcan A. Bakkaloğlu
Department of Pediatric Nephrology
Gazi University
Kathleen E. Altemose School of Medicine, Ankara
Department of Pediatrics, Division of Pediatric Turkey
Nephrology & Hypertension
Penn State College of Medicine Joanne M. Bargman
Hershey Department of Medicine, Division of Nephrology
USA University of Toronto
Toronto, Canada
Oluwatoyin I. Ameh
Jonathan Barratt
The John Walls Renal Unit, Leicester General Hospital,
Zenith Medical and Kidney Centre
and Department of Cardiovascular Sciences
Abuja
University of Leicester
Nigeria
UK
Sharon Phillips Andreoli
Nathan T. Beins
Department of Pediatrics
James Whitcomb Riley Hospital for Children
Indiana University School of Medicine
Indianapolis, IN
USA
Aminu K. Bello
Chaisiri Angkurawaranon Division of Nephrology and Immunology, Department
Department of Family Medicine, Faculty of Medicine of Medicine
Chiang Mai University, Chiang Mai, Thailand University of Alberta
Thailand Edmonton, Alberta
Canada
Meredith A. Atkinson William M. Bennett

Department of Pediatrics, Division of Pediatric Legacy Transplant Services
Nephrology Legacy Good Samaritan Medical Center
Johns Hopkins University School of Medicine Portland, OR
Baltimore, MD, USA USA
flast_V2.indd 11 09-12-2022 15:13:05

xii List of Contributors
Meha Bhatt Ben Caplin

Division of Nephrology, Department of Medicine Department of Renal Medicine
Cumming School of Medicine University College London
University of Calgary UK
Alberta, Canada
Andres Cardenas
Bhadran Bose Institut de Malalties Digestives i Metabòliques,
Nepean Hospital University of Barcelona
Kingswood, NSW, Australia; Spain and
University of Sydney, Sydney, NSW, Australia Institut d’Investigacions Biomèdiques August Pi-Sunyer y
Ciber de Enfermedades Hepáticas y Digestivas
Branko Braam
Barcelona
Spain
Canada
Juan J. Carrero
European Renal Nutrition Working Group of the
Frank Brennan
European Renal Association – European Dialysis
Departments of Nephrology and Palliative Care
Transplant Association
St George Hospital
Parma
Sydney
Italy and
Australia
Department of Medical Epidemiology and Biostatistics,
Karolinska Institutet
Frank Bridoux
Stockholm
Department of Nephrology, Centre Hospitalier
Sweden
Universitaire et Université de Poitiers
France and Department of Immunology CNRS UMR7276
Kerri Cavanaugh
Université de Limoges, Limoges
Division of Nephrology & Hypertension
France and
Centre de Référence Amylose AL et Autres Maladies par
Vanderbilt University Medical Center
Dépôt d’Immunoglobulines Monoclonales
Nashville, TN, USA
Poitiers
France Christopher T. Chan
Victoria Briggs Toronto General Hospital
Department of Nephrology Toronto
Calderdale and Huddersfield NHS Foundation Trust Canada
Mark Brown Katharine L. Cheung

St. George Hospital Division of Nephrology, Department of Medicine
University of New South Wales The University of Vermont
Sydney Burlington
Australia USA
Neil Boudville Yeoungjee Cho

University of Western Australia, Medical Department of Nephrology
School, Sir Charles Gairdner Hospital, Nedlands, WA, University of Queensland at Princess Alexandra Hospital
Australia Woolloongabba, Brisbane, Australia
Mark Canney David Collister

Division of Nephrology Department of Medicine, Division of Nephrology
University of British Columbia McMaster University
Vancouver, BC, Hamilton
Canada Canada
flast_V2.indd 12 09-12-2022 15:13:05

List of Contributors xiii
Tess E Cooper Simon J Davies

Sydney School of Public Health Renal Research Group
The University of Sydney, Sydney Faculty of Medicine and Applied Clinical Sciences
Australia Keele University
and Newcastle-under-Lyme
Centre for Kidney Research, The Children’s Hospital at UK
Westmead, Westmead, Australia
Sara N. Davison
Sara A. Combs
Department of Medicine, Divisions of Nephrology and
Division of Nephrology and Immunology
Palliative Care
University of New Mexico School of Medicine
Edmonton, AB, Canada
Albuquerque, NM,
USA Mary Amanda Dew
Department of Psychiatry
Kelsey Connelly
University of Pittsburgh School of Medicine and
Max Rady College of Medicine,
Medical Center, Pittsburgh, PA, USA
Winnipeg, Manitoba Meghan J. Elliott
Canada Division of Nephrology, Department of Medicine,
Cumming School of Medicine
Michael J. Connor, Jr University of Calgary, Alberta
Division of Pulmonary, Allergy, Canada and
Critical Care, & Sleep Medicine Department of Community Health Sciences, Cumming
Department of Medicine School of Medicine
Emory University School of Medicine University of Calgary
Atlanta, GA, USA Alberta
and Canada
Division of Renal Medicine
Department of Medicine Fabrizio Fabrizi
Emory University School of Medicine Division of Nephrology
Atlanta, GA, USA Maggiore Hospital and IRCCS Foundation
Milano
Cecile Couchoud Italy
REIN Registry
Department Agence de la biomédecine Fadi Fakhouri
Saint Denis – La Plaine Department of Nephrology and Immunology
France Centre Hospitalier Universitaire de Nantes
France
Jonathan C Craig
College of Medicine and Public Health Kevin W. Finkel
Flinders University Division of Renal Diseases and Hypertension
Adelaide, Australia McGovern Medical School
Houston, TX
Neera K. Dahl USA
Section of Nephrology
Yale University School of Medicine Fred Finkelstein
New Haven, CT, Department of Medicine
USA Yale University
New Haven, CT, USA
Matthew J. Damasiewicz
Monash Health and Monash University James Fotheringham
Clayton, Victoria, School of Health and Related Research Regent Court,
Australia University of Sheffield
flast_V2.indd 13 09-12-2022 15:13:05

xiv List of Contributors
Martin Gallagher Camilla S. Hanson

Director of the Renal & Metabolic Division Sydney School of Public Health
The George Institute for Global Health University of Sydney
Newtown, NSW, Australia Sydney, NSW, Australia and
Amit X. Garg The Children’s Hospital at Westmead
Western University Westmead, NSW, Australia
Institute for Clinical Evaluative Sciences Western Facility
Victoria Hospital Carmel M Hawley
London, ON, Canada Department of Nephrology, Princess Alexandra Hospital
University of Queensland
Michael J. Germain Brisbane, Australia
Department of Medicine, Division of Nephrology
Baystate Medical Center Swapnil Hiremath
Tufts University Department of Medicine
Springfield, MA, USA University of Ottawa
Canada
Pere Ginès
Institut de Malalties Digestives i Metabòliques, Hospital Htay Htay
Clínic Department of Renal Medicine
University of Barcelona Singapore General Hospital
Spain and Singapore
Ciber de Enfermedades Hepáticas y Digestivas Emma Huarte
Barcelona Department of Nephrology
Spain Hospital San Pedro
La Rioja, Spain
David S. Goldfarb
Nephrology Section, New York Harbor Kwaifa S. Ibrahim
VA Medical Center, and Nephrology Department of Medicine
Division, NYU Grossman School of Wuse District General Hospital
Medicine, New York, Abuja, Nigeria
New York, USA
Georgina Irish
Silviu Grisaru Central and Northern Adelaide Renal and Transplantation
Alberta Children’s Hospital and University of Calgary Service
Calgary, Alberta, Canada Royal Adelaide Hospital Adelaide,
South Australia, Australia and
Vanessa Grubbs Department of Medicine
Department of Medicine, University of Adelaide
University of California Adelaide, South Australia, Australia
San Francisco, CA, USA
Masao Iwagami
Talia Gutman
Department of Health Services Research, University of
Tuskuba, Ibaraki,
The University of Sydney
Japan
NSW, Australia and
and
The Centre for Kidney Research
Department of Non-Communicable
Disease Epidemiology, London School of Hygiene &
NSW, Australia
Tropical Medicine, London, UK
flast_V2.indd 14 09-12-2022 15:13:05

List of Contributors xv
Matthew T. James David W Johnson

Division of Nephrology, Department of Medicine Department of Nephrology
Cumming School of Medicine University of Queensland at Princess
University of Calgary Alexandra Hospital
Alberta Woolloongabba, Brisbane, Australia
Canada and
Department of Community Health Sciences, Cumming Angela Ju
School of Medicine Sydney School of Public Health
University of Calgary University of Sydney
Alberta, Canada NSW, Australia and
Meg Jardine Children’s Hospital at Westmead
NHMRC Clinical Trials Centre Westmead, NSW, Australia
Camperdown Adrià Juanola
Australia and Institut de Malalties Digestives i Metabòliques,
Concord Repatriation General Hospital University of Barcelona
Australia Spain and
Sarbjit Vanita Jassal Ciber de Enfermedades Hepáticas y Digestivas
Division of Nephrology Barcelona
Department of Medicine Spain
University of Toronto, Toronto, ON
Canada
Peter G Kerr
Vincent Javaugue Professor and Director of Nephrology
Department of Nephrology Monash Health and Monash University
Centre Hospitalier Universitaire et Clayton, Victoria,
Université de Poitiers Australia
France and
Department of Immunology CNRS UMR7276 Myda Khalid
Université de Limoges Department of Pediatrics
Limoges Division of Pediatric Nephrology and Hypertension
France and James Whitcomb Riley Hospital for Children
Centre de Référence Amylose AL et Autres Maladies par Indiana University Medical Center
Dépôt d’Immunoglobulines Monoclonales Indianapolis, IN, USA
Poitiers Mubeen M. Khan
France Division of Renal Diseases and Hypertension
McGovern Medical School
Shilpanjali Jesudason Houston, TX
Central and Northern Adelaide Renal and Transplantation USA
Service
Royal Adelaide Hospital Amrit Kirpalani
Adelaide, South Australia, Australia and Division of Nephrology and Cell Biology Program
Department of Medicine Research Institute
University of Adelaide The Hospital for Sick Children
Adelaide, South Australia, Australia Toronto, ON, Canada
flast_V2.indd 15 09-12-2022 15:13:05

xvi List of Contributors
Jay L. Koyner Kelly Li

Section of Nephrology St. George Hospital
Department of Medicine University of New South Wales
University of Chicago, Chicago, IL Sydney
USA Australia
Raymond T. Krediet
Christoph Licht
Division of Nephrology and Cell Biology Program
Amsterdam UMC
Research Institute
University of Amsterdam
The Hospital for Sick Children
The Netherlands
Toronto, ON
Mark Lambie Canada and
Renal Research Group, Faculty of Medicine and Applied Department of Paediatrics
Clinical Sciences University of Toronto, Toronto, ON
Keele University Canada
Newcastle-under-Lyme
UK Wai H. Lim
Nicholas G. Larkins Sir Charles Gairdner Hospital
Department of Nephrology, Perth Children’s Hospital Perth
Nedlands Australia and
Western Australia School of Medicine
Australia and Discipline of Internal Medicine
School of Paediatrics and Child Health University of Western Australia
University of Western Australia Perth
Nedlands, Western Australia, Australia
Australia
Hamidu M. Liman
Nelson Leung Division of Nephrology
Division of Nephrology, Hematology Usmanu Danfodiyo University Teaching Hospital
Department of Internal Medicine Sokoto
Mayo Clinic, Rochester, MN Nigeria
USA
Kathleen D. Liu
Asaf Lebel Division of Nephrology, Departments of Medicine and
Division of Nephrology and Cell Biology Program Anesthesia
Research Institute University of California
The Hospital for Sick Children San Francisco, CA
Toronto, ON, Canada USA
Adeera Levin Jayme E. Locke

Division of Nephrology Department of Surgery, Division of Transplantation
University of British Columbia University of Alabama at Birmingham, Birmingham, AL
Vancouver, BC USA
Canada
Charlotte Logeman
Guisen Li
Renal Department
Sydney, NSW, Australia and
Institute of Nephrology
Sichuan Provincial People’s Hospital
Medical School of UESTC
Westmead, NSW, Australia
China
flast_V2.indd 16 09-12-2022 15:13:05

List of Contributors xvii
Jicheng Lv Piergiorgio Messa

Peking University First Hospital Division of Nephrology
Beijing Maggiore Hospital and IRCCS Foundation
China Milano
Italy and
Bryan Ma University School of Medicine
Division of Nephrology, Department of Medicine Milano
Cumming School of Medicine Italy
University of Calgary
Alberta Pablo Molina
Canada Department of Nephrology, Hospital Universitario Dr Peset
Robert MacGinley Spain and
Department of Renal Medicine European Renal Nutrition Working Group of the European
Eastern Health Renal Association – European Dialysis Transplant Association
Box Hill Parma
Australia Italy
and Renal Medicine Department
Eastern Health Clinical School Donald A. Molony
Monash University Division of Renal Diseases and Hypertension and Center
Melbourne for Clinical Research and Evidence-Based Medicine
Australia University of Texas Houston Medical School
Houston
Magdalena Madero
USA
Department of Nephrology,
National Institute of Cardiology David W. Mudge
Mexico Department of Nephrology
University of Queensland at Princess Alexandra Hospital
Muhammad A. Makusidi Woolloongabba, Brisbane,
Division of Nephrology Australia
Usmanu Danfodiyo University Teaching Hospital
Sokoto Elmi Muller
Nigeria Faculty of Medicine and Health Science & Dept of Surgery
tellenbosch University
Karine E. Manera
Sydney School of Public Health Alexandra Munt
University of Sydney Department of Renal Medicine
Australia and Westmead Hospital
Centre for Kidney Research Western Sydney Local Health District
The Children’s Hospital at Westmead Sydney
Sydney Australia and
Australia Michael Stern Laboratory for Polycystic Kidney Disease
Christina Mariyam Joy
Australia
Washington University in St. Louis
USA Margaux N. Mustian
Department of Surgery, Division of Transplantation
Stephen McAdoo
University of Alabama at Birmingham, Birmingham, AL
Imperial College London
USA
London, UK
Masaomi Nangaku
Helen McDonald
Department of Infectious Disease Epidemiology
London School of Hygiene & Tropical Medicine
Bunkyo-ku, Tokyo
London, UK
Japan
flast_V2.indd 17 09-12-2022 15:13:05

xviii List of Contributors
Patrizia Natale Ali Olyaei

Department of Emergency and Organ Transplantation Department of Pharmacy
University of Bari Oregon State University/Oregon Health & Science University
Bari, Italy Portland, OR
and USA
The University of Sydney Mohamed A. Osman
Sydney, Australia Division of Nephrology and Immunology,
Sharon J. Nessim University of Alberta
Division of Nephrology Edmonton, Alberta
Jewish General Hospital Canada
McGill University
Montreal Marlies Ostermann
Canada Department of Critical Care
Guy’s & St Thomas’ Hospital
London
Javier A. Neyra
UK
Division of Nephrology, Bone and Mineral Metabolism
University of Kentucky Luis M. Pallardó
Lexington Department of Nephrology
USA Hospital Universitario Dr Peset
Spain
Thu T. Nguyen
Suetonia C. Palmer
Auckland City Hospital
Auckland
University of Otago Christchurch
New Zealand
Christchurch,
New Zealand
Dorothea Nitsch
Department of Non-Communicable Patrick S. Parfrey
Disease Epidemiology Faculty of Medicine
London School of Hygiene & Tropical Medicine Memorial University
London, UK St. John’s, NL,
Canada
Marlies Noordzij
Department of Medical Informatics Henry Pleass
Amsterdam UMC Specialty of Surgery
University of Amsterdam Faculty of Medicine and Health
The Netherlands University of Sydney and
Centre for Transplant and Renal Research
Ikechi G. Okpechi Westmead Hospital
Division of Nephrology and Hypertension Sydney
Groote Schuur Hospital Australia and
Cape Town, South Africa Department of Surgery
and Surgical Research and Education Centre
Kidney and Hypertension Research Unit Westmead Hospital
University of Cape Town Sydney
Cape Town, South Africa Australia
flast_V2.indd 18 09-12-2022 15:13:05

List of Contributors xix
Kevan R. Polkinghorne Gopala Rangan

Department of Nephrology Department of Renal Medicine
Monash Medical Centre Westmead Hospital
Melbourne, Victoria Western Sydney Local Health District
Australia Sydney, Australia and
Michael Stern Laboratory for Polycystic
Carol Pollock Kidney Disease
Sydney School of Public Health Westmead Institute for Medical Research
The University of Sydney University of Sydney
Sydney, Australia Australia
and
Kolling institute of Medical Research Amanda DeMauro Renaghan
St Leonards, Australia Division of Nephrology
University of Virginia Health System
Evgenia Preka Charlottesville
Southampton Children’s Hospital USA
University Hospital Southampton NHS Foundation Trust
Southampton Lesley Rees†
UK Paediatric Nephrology
Great Ormond Street Hospital NHS Foundation
Maria Prendecki Trust and UCL Great Ormond Street Institute of
Imperial College London Child Health
London, UK London, UK
Joseph B. Pryor Matthew A. Roberts

Department of Internal Medicine Eastern Health Clinical School
University of Washington, Seattle Monash University
WA, USA Melbourne, VIC
Australia
Lonnie Pyne
Department of Medicine Mitchell H. Rosner
Division of Nephrology Division of Nephrology
McMaster University University of Virginia Health System
Hamilton Charlottesville, VA
Canada USA
Qi Qian
Marinella Ruospo
Division of Nephrology and Hypertension
Mayo Clinic School of Medicine
Australia
USA
Kannaiyan S. Rabindranath Fahad Saeed

Renal Unit Departments of Medicine and Public Health
Waikato Hospital Divisions of Nephrology and Palliative Care
Hamilton University of Rochester Medical Centre
New Zealand Rochester, NY, USA
Jörg Radermacher Valeria Saglimbene

Department of Nephrology Sydney School of Public Health
Johannes Wesling Klinikum Minden University of Sydney, NSW
UK- RUB, Minden, Germany Australia
flast_V2.indd 19 09-12-2022 15:13:05

xx List of Contributors
Aminu M. Sakajiki Emily J. See

Division of Nephrology Department of Intensive Care
Usmanu Danfodiyo University Teaching Hospital Austin Health
Sokoto, Nigeria Melbourne
Australia and
Joshua Samuels School of Medicine
The McGovern Medical School University of Melbourne
University of Texas HSC Melbourne
Houston, TX, USA Australia
Irene Sangadi Nikhil Shah

Department of Renal Medicine Division of Nephrology and Immunology, Department of
Westmead Hospital Medicine
Western Sydney Local Health District University of Alberta
Sydney Edmonton, Alberta
Australia and Canada
Soroush Shojai
Division of Nephrology and Immunology, Department of
Australia
Medicine
Sayan Saravanabavan
Edmonton, Alberta
Canada
Westmead Hospital
Sydney, Australia and Badri Shrestha
Michael Stern Laboratory for Polycystic Kidney Disease Department of Renal Medicine
Westmead Institute for Medical Research Sheffield Teaching Hospitals NHS Foundation Trust
University of Sydney Sheffield
Australia S Yorkshire, UK
Judy Savige Rukshana Shroff

Department of Medicine (Melbourne Health) and Great Ormond Street Hospital for Children
Northern Health NHS Foundation Trust, London
The University of Melbourne UK
Australia
Brendan Smyth
Jane Schell Department of Renal Medicine
Section of Palliative Care and Medical Ethics, Division of St George Hospital, Kogarah
Renal-Electrolyte Australia
Department of Medicine and
University of Pittsburgh School of Medicine NHMRC Clinical Trials Centre
UPMC Health System University of Sydney
Pittsburgh, PA, Camperdown, Australia
USA
Belinda Stallard
Rebecca Schmidt Department of Nephrology
Section of Nephrology, Department of Medicine Princess Alexandra Hospital
West Virginia University School of Medicine Brisbane
Morgantown, WV Australia
USA
flast_V2.indd 20 09-12-2022 15:13:05

List of Contributors xxi
Dan A Streja Mathew Tabinor

David Geffen School of Medicine at UCLA Renal Research Group, Faculty of Medicine and Applied
Los Angeles, CA Clinical Sciences
USA and Keele University
Division of Endocrinology, Diabetes and Metabolism Newcastle-under-Lyme
VA Healthcare System UK
West Los Angeles VA Medical Center
Los Angeles, CA Tetsuhiro Tanaka
USA Division of Nephrology and Endocrinology
Elani Streja Bunkyo-ku, Tokyo
Division of Nephrology, Department of Medicine Japan
Harold Simmons Center for Kidney Disease Research and
Navdeep Tangri
Epidemiology
Max Rady College of Medicine
UC Irvine School of Medicine
Orange, CA, USA and
Winnipeg, Manitoba
Tibor Rubin VA Medical Center
Canada
Long Beach, CA
USA
J. Pedro Teixeira
Giovanni F.M. Strippoli Divisions of Nephrology and Pulmonary, Critical Care,
Department of Emergency and Organ Transplantation and Sleep Medicine
University of Bari University of New Mexico School of Medicine
Bari, Italy Albuquerque, NM, USA
and
School of Public Health Allison Tong
University of Sydney Sydney School of Public Health
Sydney, NSW, Australia University of Sydney
NSW, Australia and
Guobin Su Centre for Kidney Research
Department of Nephrology, Guangdong Provincial Children’s Hospital at Westmead
Hospital of Chinese Medicine Westmead, NSW,
The Second Affiliated Hospital Australia
Guangzhou University of Chinese Medicine
Hernán Trimachi
Guangzhou City
Nephrology Service and Kidney Transplant Unit
Guangdong Province
Hospital Británico de Buenos Aires
China
Buenos Aires
and
Argentina
National Clinical Research Center for Kidney Disease
State Key Laboratory of Organ Failure Research Katie Trinh
Guangdong Provincial Clinical Research Center for Department of Nephrology
Kidney Disease Nepean Hospital
and Kingswood, NSW, Australia;
Department of Nephrology University of Sydney, Sydney, NSW, Australia
Nanfang Hospital, Southern Medical University
Guangzhou city, Guangdong Province David J Tunnicliffe
China Global Health – Health Systems and Policy Sydney School of Public Health
Department of Global Public Health, Karolinska Institutet The University of Sydney, Sydney, NSW
Stockholm, Sweden Australia and
and Centre for Kidney Research
Department of Medical Epidemiology and Biostatistics The Children’s Hospital at Westmead
Karolinska Institutet Westmead, NSW
Stockholm, Sweden Australia
flast_V2.indd 21 09-12-2022 15:13:05

xxii List of Contributors
Mark Unruh Li Wang

Division of Nephrology Renal Department
University of New Mexico School of Medicine Institute of Nephrology
Albuquerque, NM Sichuan Provincial People’s Hospital
USA and Medical School of UESTC
New Mexico VA Health Care System China
Albuquerque, NM
USA Bradley A Warady
Tomoko Usui
Bunkyo-ku, Tokyo
Japan
Steven D. Weisbord
Renal Section and Center for Health Equity Research and
Anita van Zwieten
Promotion
VA Pittsburgh Healthcare System and Renal Electrolyte
Division
Australia
University of Pittsburgh School of Medicine
Pittsburgh
Andrea K. Viecelli USA
Department of Nephrology
Princess Alexandra Hospital Martin Wilkie
Brisbane, Queensland, Department of Renal Medicine, Sheffield Teaching
Australia Hospitals NHS Foundation Trust
Sheffield, S Yorkshire, UK
Anitha Vijayan
Division of Nephrology Annette Wong
Washington University in St. Louis Department of Renal Medicine
USA Westmead Hospital
Belén Vizcaíno Sydney Australia
Department of Nephrology and
Hospital Universitario Dr Peset Michael Stern Laboratory for Polycystic Kidney Disease
Universitat de València Westmead Institute for Medical Research
Spain University of Sydney
Australia
Michael Walsh
Department of Medicine, Division of Nephrology Muh Geot Wong
McMaster University, Hamilton The University of Sydney
Canada and Sydney, NSW, Australia
Department of Health Research Methods and
Evaluation and Impact Department of Renal Medicine
McMaster University Concord Repatriation General Hospital
Hamilton Concord, NSW, Australia
Canada and
Population Health Research Institute See Cheng Yeo
Hamilton Health Sciences/McMaster University Department of Renal Medicine
Hamilton Tan Tock Seng Hospital
Canada Singapore
flast_V2.indd 22 09-12-2022 15:13:05

List of Contributors xxiii
Jennifer Zhang Yuemiao Zhang

Department of Renal Medicine, Westmead Hospital Peking University First Hospital
Western Sydney Local Health District, Sydney Beijing
Australia and China
Australia
flast_V2.indd 23 09-12-2022 15:13:05

1
Hemodialysis
Part 6
3
40
Modalities of Extracorporeal Therapy

Matthew J. Damasiewicz1, Muh Geot Wong2,4, and Peter G. Kerr3
1
Monash Health and Monash University, Clayton, Victoria, Australia
2
Department of Renal Medicine, Concord Repatriation General Hospital, Concord, NSW, Australia
3
Professor and Director of Nephrology, Monash Health and Monash University, Clayton, Victoria, Australia
4
The University of Sydney, Sydney, NSW, Australia
I ntroduction solvent drag by the passage of water across the membrane,

and therefore greater membrane flux or permeability to
Despite the complexity of hemodialysis (HD), the theoreti- water. Of note, the relative differences in clearance between
cal principles are relatively simple (Box 40.1). Blood and all of these modalities of blood purification are large,
dialysis fluid are circulated on the opposite sides of a semi- although the absolute differences relative to native kidney
permeable membrane (the dialysis membrane), which per- function are small (Box 40.2).
mits the passage of water and smaller solutes, but restricts The two other practiced forms of extracorporeal blood
the passage of larger molecules. The provision of clinically purification, hemofiltration and acetate-free biofiltration,
adequate HD is for the most part dependent on water qual- are seldom used and will not be discussed in this chapter.
ity, dialysate composition, dialysis access, the presence and In this chapter, we summarize the published studies of
extent of treatment monitoring, and delivery and modality hemodialysis membrane type and hemodialysis modality,
of blood purification. and provide recommendations for nephrologists responsi-
Dialysis membranes are one of two key components that ble for hemodialysis care.
define the modality of blood purification, the other being
the hydrostatic pressure gradient across the membrane E
pidemiology
itself (Table 40.1). The most important operating character-
istic of dialysis membranes is their permeability to water. The origins of modern extracorporeal techniques for the
Low flux HD, high flux HD, and hemodiafiltration (HDF) treatment of end-stage kidney disease (ESKD) can be traced
are modalities characterized by increasing convective clear- back to the early twentieth century, where rudimentary
ance of larger solutes in addition to the usual diffusive devices were used to remove fluid and solutes from blood in
clearance of small solutes. The convective process relies on animal experiments. However, these early efforts were sig-
nificantly limited by the insufficient effectiveness of the
blood purification treatment. Modern efforts using HD
Box 40.1 Dialysis Key Features became possible in the 1960s, with technological advances
allowing improved dialysis membrane quality and mass
Dialysis modality Membrane qualities Solute removal production, along with improvements in anticoagulation
and dialysis access.
●● Low-flux HD ●● Synthetic versus ●● Diffusion
cellulosic Today HD is routinely applied to patients as an indefinite
●● High-flux HD ●● Convection
●● HDF ●● Low-flux, high-flux, ●● Adsorption
life-sustaining treatment for ESKD. Worldwide there are
super-flux currently in excess of three million patients receiving HD
and this number is projected to increase significantly,
HD, hemodialysis; HDF, hemodiafiltration.
mostly due to the increased access to care in developing
4 Modalities of Extracorporeal Therapy
Table 40.1 Membrane type and modality.
Strength of
Hemodialysis (HD) We suggest a synthetic, biocompatible, high-flux dialysis Low Weak

membrane can be used for hemodialysis rather than a
low-flux membrane, to reduce cardiovascular mortality,
although providers and patients may reasonably choose a
low flux membrane due to the low certainty of effects on
cardiovascular mortality and other outcomes.
Hemodiafiltration We suggest HDF may be used rather than conventional Low (cardiovascular Weak
(HDF) HD to reduce cardiovascular mortality and reduce mortality) or very low
frequency of intradialytic hypotensive events, although (dialysis-related
patients may reasonably choose not to due to the low hypotension)
certainty evidence for improved clinical outcomes.
that the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood that the
effect will be substantially different* is moderate. Low: This research provides some indication of the likely effect. However, the likelihood that it
will be substantially different* is high. Very low: This research does not provide a reliable indication of the likely effect. The likelihood that the
Box 40.2 Relative Small and Large Solute Clearance for Different Stages of CKD and Dialysis
Equivalent small uremic toxin Equivalent large uremic toxin

Native renal or RRT Fluid control (%)a clearance (EKRc)b clearance (EKRc)c
Low-flux HD 5–10 10–15 3–3.5

High-flux HD 5–10 10–15 3.5–4
High-volume HDF 5–10 10–15 4–5
PD 100 8–12 1–2
Intensive HD 10–30 20–30 5–10
Severe CKD 100 15–30 15–30
Moderate CKD 100 30–60 30–60
Normal kidneys 100 > 60 > 60
CKD, chronic kidney disease; PD, peritoneal dialysis; RRT, renal replacement therapy.
a
Hours per week of ultrafiltration as a percentage of 168.
b
EKRc for urea, ml/min.
c
EKRc for β2-microglobulin, ml/min.
nations. For most patients routine HD involves three treat- been generally classified based on their composition (cel-
ment sessions per week, of 4–5 hours of treatment per ses- lulosic vs. synthetic) and permeability to water (high vs.
sion, although practice can and does vary considerably low flux). Super-flux membranes, with markedly
between countries. Most treatments are performed in hos- increased water permeability and middle molecule clear-
pital or free-standing community facilities, with a small ance, have become commercially available and are
percentage occurring as home-based HD. defined as medium cut-off (MCO) and high cut-off (HCO)
membranes [1]. Dialysis membranes are not physiologi-
cally inert and make contact with the patient’s blood,
P
athophysiology
therefore they can elicit active biological responses in the
patient. Although there is no formal definition of “bioin-
Dialysis Membranes
compatibility,” such responses have long been recognized
Dialysis membranes permit the movement and removal as device-related and clearly linked to adverse clinical
of solutes and water during dialysis. Membranes have outcomes.
Pathophysiolog 5
The Evolution of Dialysis Membranes water crosses the membrane at a given transmembrane
pressure (defined in ml/h/mmHg). The importance of flux is
The first modern HD membranes were cellulose based, thin-
not the permeability of the water per se, but the fact that
walled, and functionally classified as low flux. Despite their
greater flux of water across the membrane during the course
functionality, there were two significant disadvantages of
of HD or HDF treatment results in greater passage of larger
these membranes. First, the natural, cellulosic fibers were
molecules (nominally β2-microglobulin) through the dialy-
immune-reactive mainly as a result of exposed hydroxyl
sis membrane by the process of solvent drag. Informally,
groups [2, 3] activating inflammatory pathways such as com-
both of these operating characteristics are now used to dis-
plement and inducing inflammatory mediators. This
criminate low-from high-flux membranes. However, thresh-
resulted in a classical inflammatory response syndrome,
olds that define degrees of “flux” remain somewhat arbitrary.
reduced dialysis tolerance, and dialysis-related cachexia.
In the HEMO study, a high-flux dialysis membrane was
Second, they only permitted the passage of small solutes
defined as Kuf > 14 ml/h/mmHg and β2-microglobulin clear-
(MW < 500 Da). This inability to remove larger molecules (in
ance of >20 ml/min, whereas low-flux membranes were
particular β2-microglobulin, MW 11 800 Da) was implicated
defined as β2-microglobulin clearance of <10 ml/min [9].
in the subsequent development of dialysis-related amyloido-
More recent definitions normalize the Kuf to the membrane
sis, especially in very long-term HD patients [4].
surface area (expressed in ml//h/mmHg/m [1]) and define
The evolution of membranes in general occurred in two
β2-microglobulin in terms of the sieving coefficient (SC),
directions, first to improve “bioincompatibility” and sec-
which is a ratio of the solute in the filtrate compared to its
ond to improve flux. Development largely focused on the
plasma concentration. The Membrane Permeability
large-scale production of synthetic membranes made from
Outcome (MPO) study defined high-flux as β2-microglobulin
polysulfone, polyamide, polyarylethersulfone, and polym-
SC >0.6 and Kuf 20 ml/mmHg/h [10]. The European
ethylmethacrylate [5]. Such plastics are generally, but not
Dialysis (EUDIAL) Working Group felt that a characteristic
always, more biocompatible and less biologically active.
of middle molecule clearance should be incorporated in a
They can be spun into fibers with more consistent pore size
modern definition of “high flux” in addition to the ultrafil-
and geometry compared to cellulose, allowing more pre-
tration coefficient of a membrane.
cise molecular weight (MW) cut-off and therefore more
The agreed EUDIAL definition of a high-flux membrane
predictable clearance of larger molecules [6]. Early syn-
is based on the MPO study and is Kuf = 20 ml/h/
thetic membranes were considerably thicker than low-flux
mmHg/m [2] and β2-microglobulin SC > 0.6 [11]. It is
membranes, causing a degree of impediment to diffusion-
important to note that most of the studies comparing
based clearance. However, modern synthetic membranes
high- and low-flux, and HD and HDF did not use this
are thinner, with excellent operating characteristics.
definition.
High-flux membranes have increased clearance of mid-
dle molecules (nominally β2-microglobulin), which in turn
has been associated with a reduced incidence of symptoms Middle Molecule Clearance and Super-flux
on dialysis and long-term complications such as dialysis- Membranes
related amyloidosis [7]. Initial concerns about the exces-
The chronic inflammatory response seen in maintenance
sive cost of high-flux membranes have been largely negated
HD patients has been linked to accelerated atherosclerosis,
in developed countries with near-universal adoption. In
endothelial dysfunction, increased advanced glycosolation
addition, decreased dialyzer reuse has led to large volume
end-product levels, protein-energy wasting, cachexia, and
production and further decreased costs. However, these
secondary immune deficiency [12–19]. These in part medi-
remain pertinent issues in developing countries with com-
ate the increased all-cause and cardiovascular disease
paratively nascent dialysis infrastructure. The routine use
(CVD) mortality seen in ESKD patients when compared to
of endotoxin filters in modern dialysis circuits, and overall
age-matched individuals without kidney disease. This
improvements in dialysis water quality and water purifica-
chronic inflammatory response is multifactorial and can be
tion technology have largely mitigated concerns about
attributed to the plastic consumables used during dialysis,
back-filtration of dialysate contaminants to the patient’s
dialysis membranes, disinfecting solutions, and water qual-
blood [8].
ity as well as the presence of endotoxins and uremia. All of
these can directly or indirectly induce pro-inflammatory
The Clinical Relevance of “Flux” cytokines, many of which are large-middle molecules
(TNF-α, IL-6, and IL-10), themselves poorly cleared with
Formally, “flux” is defined by the ultrafiltration coefficient conventional high-flux HD [20]. The contribution of a dial-
(Kuf ) of a dialysis membrane, defined as the rate at which ysis membrane to this process can be viewed in terms of its
contribution to inflammation (bioincompatibility) and albumin (68 000 D). In theory these allow for more effective
potential for clearance of middle molecules during HD. The middle and large-middle molecule clearance, without
aim of increasing middle molecule clearance during HD excessive albumin loss [27]. What constitutes acceptable
has driven the development of super-flux membranes. loss of albumin during HD remains unclear. Some albumin
Middle molecules are a heterogeneous group of com- loss may in fact be beneficial due to the removal of protein-
pounds broadly defined as having a MW of 500–60 000 D [21]. bound toxins. Super-flux membranes may provide us with
This includes the archetypal middle molecule marker β2- a form of “enhanced” HD (when considered in terms of
microglobulin and a plethora of other compounds that large-middle molecule clearance). The potential benefits of
include cytokines, uremic toxins, growth factors, and hor- this new class of membranes, particularly when compared
mones, which are generally further subclassified based on to existing convective therapies, requires further evaluation
MW. Modern high-flux membranes are excellent at remov- in clinical studies.
ing β2-microglobulin (MW 11800 D), but are poor at remov-
ing large-middle molecules (MW > 15 000 D) [22]. Therapies
which combine convection with diffusion such as HDF (see D
ialysis Modality
below) have become an increasingly accepted method for
increasing middle molecule clearance. Similarly, recent HD is based on the diffusion of solutes across a semiperme-
developments in membrane technology have enabled the able membrane and is highly effective at fluid removal, cor-
development of super-flux membranes which are capable of recting acidosis and electrolyte disturbances, and removing
clearing molecules with a MW of up to 50 000 D, which small solutes and toxins. HD is less effective at removing
approximates and potentially surpasses the clearances pro- middle and large-middle molecules, as effective diffusion
vided by HDF [23, 24]. is inversely related to MW. Convective therapies have been
High cut-off membranes (HCO) were used clinically in developed to overcome this problem by increasing the
the setting of kidney failure due to multiple myeloma to transport of larger molecules across dialysis membranes.
enhance the removal of kappa and lambda light chains HDF differs to conventional dialysis in that it involves the
(large-middle molecules) [25, 26]. These have a wide exchange of large volumes of fluid with the patient. Early
distribution of pore sizes and show variable clinical efficacy convective therapies such as hemofiltration were subse-
in the treatment of cast nephropathy. However, their use is quently combined with diffusion-based therapies in HDF,
associated with increased albumin loss and significant which is the most commonly used convective therapy
electrolyte abnormalities, e.g. hypophosphatemia. MCO across the world. Whilst HD remains the most commonly
membranes have been developed with a narrower range of used dialysis modality worldwide, the uptake of HDF is
distribution of pore sizes (Figure 40.1), below the MW of increasing, particularly in Europe and Japan [28, 29].
Albumin
Conventional HD
Large MM MCO
(65kDa)
(15kDa) Conventional HD involves the use of a single-pump dialy-
sis machine. Blood is drawn from the patient, passed
through a semiporous hollow-fiber dialyzer of a given
Number of pores
High flux
membrane type, and returned to the patient. At the same
time, the dialysis solution (dialysate) is passed in a
Low flux counter-current fashion through the dialyzer, on the other
HCO side of the dialysis membrane to the blood, in a single-pass
fashion (meaning dialysate passes the membrane only
once and then is discarded). The removal of excess extra-
cellular fluid from the patient occurs by passage of the
ultrafiltered extracellular fluid across the semiporous
Pore size membrane from the blood compartment to the dialysate
Figure 40.1 Schematic of pore size in dialysis membranes. compartment. This is predominantly controlled by hydro-
MCO membranes (in pink) have been specifically designed to static pressures but is also dependent on the porosity of
remove large middle molecule (compared to high-flux) the membrane and its ultrafiltration coefficient. High-flux
membranes without excessive albumin loss (compared to
membranes allow for the movement of more fluid than is
HCO membranes). HCO, high cut-off; MCO, medium cut-off;
MM: middle molecules. Source: From Wolley et al., CJASN desired, and this is balanced by back-filtration of dialysate
2018. ASN. back into the blood. In modern dialysis machines this is
Dialysis Modalit 7
achieved with the use of ultrafiltration monitors which HDF

allow the setting of net fluid removal during each dialysis
HDF differs from standard HD by incorporating a highly
session. Solute removal is dependent on molecular size
porous dialysis membrane that allows very high convective
and weight, hydration radius, osmotic gradient, and elec-
volumes. On-line HDF requires the use of a double-pump
trical charge. Different dialyzer membranes, of different
dialysis machine, utilizing a highly porous dialysis mem-
pore size and composition, will allow the passage of sol-
brane that allows loss of high-volume ultrafiltrate into the
utes at different rates, and more porous membranes allow
dialysate compartment. The volume of ultrafiltrate lost is
the removal of larger molecules (see above). The predomi-
replaced by the infusion of ultrapure dialysate (UPD) into
nant mechanism of removal, especially for smaller sol-
the patient’s blood using the second, substitution fluid
utes, is movement through diffusion from blood into the
blood pump (Figure 40.3). The typical substitution vol-
dialysate. The other mechanism is convection, which
umes achieved in a standard post-dilution HDF should
occurs when dissolved solutes are removed as ultrafiltra-
exceed 20 l, and any proposed benefit of HDF over high
tion occurs (also referred to as “solvent drag”). This latter
flux relies on effective convection volumes of at least
mechanism is more important for larger molecules and is
18–20 l. The European Dialysis Working Group (EUDIAL)
dependent on the porosity of the membrane. In conven-
has revised the definition of HDF incorporating both mem-
tional HD, most solute removal is by dialysis and ultrafil-
brane properties (high flux as defined above) and convec-
tration volumes are generally in the order of 6–8 l
tive volumes of least 20% of the total blood volume
maximum (balanced by back-filtration within the dia-
processed with fluid balance maintained by infusion of
lyzer; Figure 40.2).
substitution fluid into the patient’s blood [11].
Blood
200
compartment
pressure
180
Hydrostatic pressure, mm Hg
160
140 Dialysate
compartment
pressure
40
Net
pressure Back filtration
20
–20
Distance along dialyzer blood path
Polysulfone membrane
QB = 200 ml/min
QD = 400 ml/min
QF = 0
Figure 40.2 Schematic of fluid removal along a dialysis hollow fiber membrane. The net ultrafiltration volume is controlled by
balancing the pressures, in effect by controlling negative pressure on the dialysate side of the membrane. Source: From Streicher, E., &
Schneider, H. (n.d.). The Development of a Polysulfone Membrane. In Highly Permeable Membranes (pp. 1–13). © 1985, Karger
Publishers.
Substitution fluid pump
3rd ultrafilter
1st and 2nd ultrafilter
Blood pump
HD controller
Figure 40.3 Schematic of the circuit used in HDF. This diagram represents predilution fluid replacement. Postdilution differs in the
site of the entry of the replacement fluid (as the blood leaves the dialyzer, rather than prior to entry to the dialyzer).
The various types of HDF differ by the site of the dialysate monitoring of fluid balance to ensure the balance of
replacement, which can be pre or post the dialyzer (or pre removed and replaced fluid equates with the desired fluid
and post as “mixed” HDF), with differences around solute loss for the session. The second is the microbiological
removal and the potential for excessive hemoconcentration. purity of the dialysate, with separate standards existing for
Predilution HDF (infusing the substitution fluid before the HDF and HD, respectively.
blood enters the dialyzer) dilutes the concentration gradient
across the dialyzer, potentially decreasing small molecule
Water Quality
clearance, but in clinical practice this is offset by greater
ultrafiltration volumes (compared to postdilutional HDF), Historically, poor water quality (from a variety of environ-
with the ensuing solvent drag resulting in similar clearances mental, chemical, and microbiological factors) has been
overall [30]. In post-dilution HDF the dialysate is infused implicated in bio-incompatibility and adverse patient out-
downstream of the dialyzer (after dialysis has occurred), comes [33]. The quality of dialysis water has seen incre-
which optimizes the clearance of solutes through diffusion, mental improvements with advances in water purification,
and the high ultrafiltration rates limit the back-filtration that disinfection methods, and routine monitoring. The provi-
occurs in high-flux HD [31]. A limitation of this modality is sion of high-quality dialysis fluids is now readily achieva-
that at high ultrafiltration rates there is excessive hemocon- ble in all modern dialysis units. The provision of
centration within the dialyzer fibers, causing elevated trans- high-quality dialysis water is arguably even more impor-
membrane pressures, reduced clearances, and clotting of the tant in the provision of HDF, given the direct infusion of
dialysis circuit. To prevent this problem, the filtration frac- substitution fluid into the blood line. The advent of convec-
tion is generally limited to 20–25% of the total blood flow tive therapies has resulted in a new, more rigorous dialysis
rate [32]. In modern dialysis machines, mixed dilution HDF fluid quality standard, UPD, which is used for substitution
allows for the infusion of dialysate both before and after the fluids in HDF. The differences between standard and UPD
dialyzer, and the ratio of the infusion rates can be balanced dialysis fluids are defined by the International Organization
to optimize clearances and minimize the adverse conse- for Standardization (ISO) in terms of terms of bacterial
quences of hemoconcentration. However, most observa- load, the total viable count in colony forming units per mil-
tional and trial data are based on postdilution HDF. liliter, and total the endotoxin load in endotoxin forming
The large convective component of HDF greatly enhances units per milliliter (Box 40.3).
solute removal, but any improvement over conventional HD
is mostly limited to middle and large-middle molecules.
Putative Benefits of HDF
As large volumes of dialysate are infused directly into
blood lines, additional treatment monitoring considera- It has been a long-standing supposition that the improved
tions are needed. The first is the need for more accurate middle-molecule clearances (see above) observed with
Treatmen 9
Box 40.3 Allowable Levels for Bacteria and Endotoxins in Dialysis Fluids
Standard dialysis fluid Ultrapure dialysis fluid

Contaminant Maximum allowable level Actional level (~ 50% of maximum level) Maximum allowable level
TVC (CFU/ml) <100 50 <0.1
Endotoxin (EU/ml) <0.5 <0.25 <0.03
CFU, colony forming units; TVC, total viable count; EU, endotoxin units.
Source: Adapted from ANSI/AAMI/ISO. ISO 11663 : 2009. Quality of dialysis fluid for haemodialysis and related therapies.
HDF should translate to improved patient outcomes. dialysis sessions found no difference in wall motion abnor-
Adding large convective volumes to standard (high-flux) malities and a number of cardiac parameters between
HD improves the clearance of middle molecules, especially high-flux HD and HDF [49]. The mechanisms underpin-
those in the 15 000–25 000 MW range, which includes a ning any benefit for HDF have not been clearly elucidated
number of cytokines such as TNF-a, IL-1, IL-10, and but may relate to removal of cardioactive/vasoactive
IL-6 [34–36]. These have long been implicated in dialytic agents, such as interleukins and other vasoactive peptides.
instability, bio-incompatibility, and poor dialysis outcomes, Alternatively, it may relate to thermal stability. This con-
so their potential removal by HDF is an area of significant cept suggests that the large volume of infused dialysate
clinical interest. In the CONTRAST study, IL-6 and prevents overheating during dialysis, thus preventing
C-reactive protein (CRP) levels were lower in the HDF vasodilatation.
cohort compared to the conventional (low-flux) HD
cohort [37]. However, in the FRENCHIE study there were
no significant differences between HDF and high-flux HD
T
reatment
over 24 months in serum levels of IL-6, IL-10, TNF-alpha,
High-flux Versus Low-flux on Cardiovascular
NT- N-terminal pro brain naturetic peptide (ProBNP), or
Outcomes and Survival
high sensitivity cardiac troponin T (hs-cTNT) [38]. These
results need further validation, and increased removal of A number of observational studies support an association
cytokines may itself not be adequate if the stimulus for between the use of high-flux membranes and improved
their induction is not addressed. Molecules in the large- survival outcomes in hemodialysis [50–53]. Three large
middle molecule spectrum with MW 25 000–60 000 D RCTs were designed to provide further clinical evidence
remain poorly cleared by all HDF and high-flux HD thera- and these include the HEMO trial, MPO and the Multiple
pies. The clearance of protein-bound solutes is generally Interventions Related to Dialysis Procedures in Order to
not improved by HDF [39]. Reduce Cardiovascular Morbidity and Mortality in HD
Cardiovascular instability remains a frequent clinical Patients (known as the EGE study) [10, 54, 55]. The details
problem in patients on HD, and intradialytic hypotension of these studies are outlined in Table 40.2.
has been increasingly associated with stunning of the myo- A Cochrane meta-analysis pooled data from 33 rand-
cardium and other vital organs such and the gut [40–43]. omized controlled trials and included 3820 participants.
Anecdotally it appears that HDF may confer increased car- The results of the analysis are outlined in Table 40.3 [56].
diovascular stability, but considerable debate continues as Cardiovascular mortality was proportionally reduced by
to whether this is indeed the case. Early observational and 17% in individuals receiving high-flux membrane dialysis
interventional studies provided mixed results with regards compared to low flux dialysis (five studies, 2612 partici-
to cardiovascular stability and outcomes [44–46]. The pants: risk ratio [RR] 0.83, 95% confidence interval [CI]
results of two separate meta-analyses found that HDF 0.70 to 0.99) across a median follow up of 24 months.
reduces the incidence of symptomatic hypotension (com- However, high-flux membranes did not reduce all-cause
pared to HD), with no significant difference in cardiovas- mortality, infection-related mortality, or hospital admis-
cular events or cardiovascular mortality [47, 48]. The sions compared to low-flux membranes. While the use of
recent FRENCHIE study, a randomized controlled trial high-flux membranes reduced predialysis β2-microglobulin
(RCT) of older patients (>65 years), did not demonstrate a levels, insufficient data were available to reliably estimate
positive benefit for HDF over HD for the primary outcome the effects of membrane flux on carpal tunnel syndrome or
of intradialytic tolerance. A small study using functional amyloid-related arthropathy. In relation to surrogate out-
cardiac magnetic resonance imaging (MRI) studies during comes, the data demonstrates that it is unlikely there is any
Table 40.2 Major randomized controlled trial of high-flux versus low-flux membrane dialysis on clinical outcomes.
Study Comparators N Follow-up Primary results Subanalysis Comments
HEMO [9, 54] 2–2 factorial study: 1846 0.8–6.6 years No difference in primary and composite RR of cardiac death for high-flux Patients are not
2003, 15 USA high-flux HD vs. (mean 4.24) secondary endpoints HD vs. low-flux HD 0.80, 95% CI blinded to the
centers low-flux HD and low Lost to follow-up: Primary endpoint: RR for all-cause 0.65–0.99, P = 0.042 intervention
dose (achieved Kt/V 0/1846 mortality high-flux vs. low-flux HD 0.92, RR for composite of first cardiac Dialyzer reuse was
1.08) vs. high dose (Kt/V 95% CI 0.81–1.05, P = 0.23 hospitalization or cardiac death permitted
1.4) Secondary endpoints: (i) first for high-flux HD vs. low-flux HD
hospitalization for cardiac causes or 0.87, 95% CI 0.76–1.00, P = 0.045
death from any cause, (ii) first Survival benefits were seen in the
hospitalization for infection or death high-flux HD if dialysis vintage
from any cause, (iii) first 15% decrease in >3.7 years RR 0.68, 95% CI
the serum albumin level or death from 0.53–0.86, P = 0.001
any cause, and (iv) all hospitalizations Lower predialysis β2-
not related to vascular access microglobulin in the high-flux HD
arms
No difference in health-related
quality of life between the groups
MPO [10] Parallel RCT: high-flux 738 3–7.5 years No difference in primary and secondary Reduction in mortality in patients Enrolled incident
2008, 59 vs. low-flux HD Lost to follow-up: endpoints with serum albumin < 4g/l was patients only
centers in Italy (stratified by serum 191/738 Primary endpoints: all-cause mortality, significantly lower in the Initially inclusion
albumin > or 4 g/dl) HR 0.76, 95% CI 0.56 to 1.04, P = 0.091 high-flux HD (p=0.032). RR 0.49, criteria was serum
with a minimal dialysis Secondary endpoints: hospitalization due 95% CI 0.28–9.87, observed in albumin <4 g/l
dose of Kt/V 1.2 to all causes, infections, or vascular high-flux HD Due to slow
access related issues Survival benefit in those with recruitment, protocol
Lower accumulation of predialysis diabetes (combined type 1 and was amended to
β2-microglobulin in high-flux HD group type 2 diabetes mellitus) observed include patients with
in high-flux group serum albumin >4 g/l
EGE 2013, 8 2 × 2 factorial design, 740 1–50 months No difference in primary endpoint High-flux HD group had lower CV Participants with
centers, randomized to high-flux (mean Primary endpoints: composite of fatal event in those with AVF and substantial residual
Turkey vs. low-flux, and 35.2 months) and nonfatal CV events (myocardial among diabetic subjects function were
ultrapure versus infarction, stroke, unstable angina The highest overall survival was excluded from this
standard dialysate requiring hospitalization, and observed in patients receiving study, this may
revascularization): HR 0.73, 95% CI dialysis using an AVF, treated increase the reported
0.49–1.08, P = 0.12 with a combination of high-flux benefit from high-flux
Secondary endpoint: overall mortality HD using ultrapure dialysate HD
Lower β2-microglobulin and time-
averaged CRP in HF groups
ACM, all-cause mortality; AVF, arteriovenous fistula; CI, confidence interval; CRP, C-reactive protein; CV, cardiovascular; HF, high-flux membrane dialyzer; LF, low-flux membrane dialyzer.
Source: Data from Palmer et al. [56].
c40.indd 10 09-12-2022 16:18:39

Dialysis Modalit 11
Table 40.3 Summary of findings: high-flux HD compared to low-flux HD.
High-flux hemodialysis vs. low-flux hemodialysis
Absolute number (95% CI)

No. of of patients affected per Quality of the
participants 1000 treated patients for Relative effect evidence
Outcomes (studies) 1 year (95% CI) (GRADE) Conclusion
All-cause mortality 2915 (10) Five fewer deaths 0.95 Low It is unlikely that high flux
(15 more to 20 fewer) (0.87–1.04) ●●○○ membranes reduce all-cause
mortality
Infection-related 2547 (3) Not estimated 0.90 Low It is unlikely that high-flux
mortality (0.71–1.14) ●●○○ membranes reduce infection-
related mortality
Cardiovascular 2612 (5) 15 fewer deaths (0–30 0.83 Low High-flux membranes may reduce
mortality fewer) (0.70–0.99) ●●○○ CV mortality
All hospital admissions 74 (2) Not estimated 0.79 Very low It is uncertain whether high-flux
(0.52–1.20) ●○○○ membranes reduce all-cause
hospital admissions
Dialysis-related 84 (1) Not estimated 3.00 Very low It is uncertain whether high-flux
hypotension (0.13–71.61) ●○○○ membranes reduce intradialytic
hypotension
Quality of life 1835 (2) Not estimable Not estimable Very low It is uncertain whether high-flux
●○○○ membranes have benefits on
health-related quality of life
domains
Dialysis adequacy 155 (1) Not estimated Very low It is uncertain whether high-flux
leading to trial ●○○○ membranes have benefits on
withdrawal dialysis adequacy compared to
low-flux membranes
Residual renal function RRF at end Not estimated 1.10 ml/min Very low It is uncertain whether high-flux
(RRF) of study: (0.80–1.30) ●○○○ membranes help preserve residual
30 (1) renal function
Rate of −0.13 ml/min/
decline: month (− 0.17
20 (1) to – 0.09)
β2-microglobulin 20 (1) Not estimated 0.06 Very low It is uncertain whether high-flux
complication: carpal (0.00–0.90) ●○○○ membranes reduce β2-
tunnel syndrome microglobulin-related carpal tunnel
syndrome
β2-microglobulin 20 (1) Not estimated 0.06 Very low It is uncertain whether high-flux
complication: (0.00–0.90) ●○○○ membranes reduce β2-
arthropathy microglobulin-related arthropathy
Adverse events and 557 (13) Not estimated Not estimable Very low It is uncertain whether high-flux
safety ●○○○ membranes reduce adverse events
in dialysis
All data derived from the 2013 Cochrane Review [56].

significant benefit for high -flux membranes compared to with the increased clinical use of MCO membranes [20].
low-flux membranes in relation to urea clearance or serum The efficacy of MCO membranes has been shown in small
lipid levels. The data suggest that if 1000 individuals with studies, with improved clearance rates of complement fac-
ESKD were treated with high-flux HD there would be no tor D (MW 24 000 D), α1-microglobulin (MW 33 000 D), and
reduction in all-cause mortality or hospitalization rates, the chitinase-like protein YKL-40 (MW 40 000 D) compared
but 15 fewer deaths might occur attributed to cardiovascu- to high-flux HD [27]. Dialysis with MCO membranes has
lar disease (CVD) causes. been associated with reduced activation of the renin-
angiotensin system and decreased expression of proinflam-
matory cytokines (TNF-a and IL-6) [59, 60].
Super-flux Membranes
These findings need further evaluation in prospective stud-
At present there is limited clinical experience with the use ies and correlation with clinical outcomes. An early example
of MCO membranes. Evidence is limited to nonrand- of such a study is REMOVAL-HD [61]. This is a prospective,
omized studies. investigator-led, open-label study which aims to address the
The putative benefits outlined above are offset by con- safety, efficacy, and impact on patient-centered outcomes
cerns of albumin loss and back-filtration, and both need with the use of MCO membranes in 85 prevalent HD patients.
further evaluation. Albumin losses during a MCO HD ses- The primary endpoint is the change in predialysis concentra-
sion can approximate 3 g, but no significant change in tions of serum albumin, with additional biochemical and
serum albumin levels has been demonstrated and at pre- patient-centered secondary endpoints. The outcomes will
sent the clinical significance of these findings remains inform the design of future clinical trials which need to com-
unclear [57, 58]. Similarly, in vitro studies have shown no pare MCO membranes not only to high-flux membranes, but
increase in back-filtration, but this may become apparent perhaps more importantly to convective therapies.
International Society Guidelines: High Versus Low Flux
Guideline Year Recommendation
KDOQI [62] 2015 ●● We recommend the use of biocompatible, either high-or low-flux, hemodialysis
membranes for intermittent hemodialysis. High-flux dialyzers should be used
preferentially, but if there are cost restraints, patients with diabetes mellitus,
low albumin, and longer dialysis vintage should be prioritized.
Japanese Society for Dialysis 2015 ●● High-performance membrane dialyzers should be used. These were defined in
Therapy Clinical 2005 as a β2-microglobulin clearance of at least 10 ml/min and updated in 2013
Guideline [63] to include albumin permeability and adsorption capability.
CARI-KHA [64] 2013 ●● We recommend that high-flux membranes be used to remove molecules such as
β2-microglobulin because they have been shown to achieve lower serum levels.
●● We suggest there are possible survival benefits from high-flux membranes for
some groups, such as those on dialysis for more than 3.7 years, those with a
serum albumin below 40 g/l, and diabetics
Renal Association Clinical 2011 ●● We suggest that hemodialyzers with synthetic and modified cellulose
Practice Guidelines [65] membranes should be used instead of unmodified cellulose membranes.
●● We suggest that high-flux dialyzers should be used instead of low-flux dialyzers
to provide HD. Evidence of improved patient survival with the use of high-flux
membranes is restricted to incident patients, who have lower serum albumin
concentrations (<40 g/l) or have diabetes mellitus, and prevalent patients, who
have been on HD for more than 3.7 years.
European Best Practice 2010 Update to ●● Synthetic high-flux membranes should be used to delay long-term
Guidelines [66] the 2007 complications of hemodialysis therapy in patients at high risk (serum
Guidelines albumin < 40 g/l)
●● In view of underlying practical considerations and the observation of a
reduction of an intermediate marker (β2-microglobulin), synthetic high-flux
membranes should be recommended even in low-risk patients.
The Canadian Society of Nephrology Clinical Practice Guidelines published in 2006 provide no clear recommendation. KDIGO does not provide
guideline on dialysis prescription or membranes.
Treatmen 13
Discussion 0.72–1.05), although it probably reduced cardiovascular

mortality (six studies, 2889 participants, RR 0.75, 95% CI
The choice of dialysis membrane for HD remains a difficult
0.61–0.92). There was evidence that HDF reduced nonfatal
question to answer based on hard evidence alone. Our pref-
cardiovascular mortality, infection related mortality or
erence reflects the default position in most higher-income
hospitalization. A reduction in dialysis-related hypotension
countries, that of a high-flux, synthetic dialysis membrane.
with convective therapy was reported in one study (906
This is based on the putative benefits (improved biocom-
participants, RR 0.72, 95% CI 0.66–0.80). There was no
patibility, greater cardiovascular stability, and improved
significant benefit in terms of health-related quality of life,
middle molecule clearance). The widespread uptake of
although the evidence was considered of very low quality.
synthetic membranes, along with reduced dialyzer reuse,
Convective therapies significantly reduced predialysis β2-
have greatly reduced unit cost. Similarly, concerns about
microglobulin and increased dialysis dose (Kt/V urea), but
back-filtration and endotoxemia have been largely obvi-
results across the studies were heterogeneous and analysis
ated by improvements in water quality. In countries where
limited to comparing HDF with HD yielded similar results.
the availability of high-flux membranes remains limited,
In absolute terms, it was suggested that convective
their use should be preferentially directed to patients with
therapies may prevent 25 cardiovascular deaths for every
clinical characteristics that were associated with evidence
1000 patients treated. In terms of surrogate outcomes there
of benefit, namely high cardiovascular risk. The choice
was a small reduction in predialysis β2-microglobulin
between particular synthetic membranes is unlikely to be
levels (1813 participants, −5.55 mg/dl, 95% CI −9.11 to
clinically significant, as any differences are probably small,
−1.98), however there was no evidence that HDF made any
and this has not been (or is likely to be) compared in head-
difference to predialysis blood pressure.
to-head clinical trials. Finally, we await further clinical trial
In a subsequent analysis, pooled individual participant
data pertaining to MCO membranes to assess any role in
data from the 4 major RCTs, which were designed to
evidence-based clinical practice.
examine the effects of HDF on mortality endpoints, was
used to study the effects of HDF on mortality outcomes in
Evidence for HDF vs. HD ESKD [73]. In the overall analysis HDF was associated
with a proportional reduction in all-cause mortality by 14%
Much of the purported mortality benefit for HDF when (95% CI 1–24%) and cardiovascular mortality by 23% (95%
compared to HD came from early observational studies [67– CI 3–39%). There was no difference in subgroups, such as
71]. These studies mostly compared postdilution HDF, diabetics, but a closer scrutiny of convection volumes and
albeit with variably defined effective convection volumes, mortality revealed that the benefit for HDF and all-cause
to either low- or high-flux HD. The studies were mortality remained significant only in those patients who
heterogeneous in terms of the population studied as well as received convection volumes (>23 l/1.73 m per session [2]).
the type and quality of dialysis provided. As such these are Although this analysis has the advantage of a larger num-
particularly prone to the bias and confounding inherent to ber of patients (n = 2793), the original trial issues remain,
observational data, but the majority of these studies especially in regard to the differences in achieved convec-
showed a mortality benefit for HDF when compared to tion volumes and HD type, and that these are non-
conventional HD. randomized analyses. The major studies and meta-analyses
The four main RCTs comparing HDF to HD are summa- all suggest a mortality benefit if higher convection volumes
rized in Table 40.4 [37, 38, 72, 74]. These studies failed to are achieved. However, the ability to “achieve” high con-
provide clear answers, but in general did not demonstrate vective volumes may simply reflect a well-functioning
the mortality benefits hypothesized from the early observa- arterio-venous fistula, reasonably high blood flow rate, and
tional trials. However, a consistent trend emerged from the the absence of major CVD disease. Therefore, patients
secondary analyses, where a higher convection volume was achieving higher volumes may in fact be “better patients,”
associated with reduced all-cause and CVD mortality. biasing the high convective group to better outcomes.
A 2015 Cochrane systematic review and meta-analysis Further trials are underway to further evaluate these ques-
included the three main RCTs (CONTRAST, ESHOL, and tions. The CONVINCE trial, which is ongoing in Europe, is a
Turkish HDF), but also trials involving hemofiltration randomized multicenter study involving 1800 patients com-
(HF) and acetate-free biofiltration [47] (Table 40.5). In paring HDF to high-flux HD, with mortality as the primary
general, the risk of bias was classified as high, resulting in endpoint (Netherlands Trial Registry, 7138). Careful attention
a low level of confidence in the estimated treatment effects. is being paid to convection volumes, with a prespecified mini-
Convective dialysis probably did not decrease all-cause mum of 23 l per session (body surface area [BSA] adjusted).
mortality (11 studies, 3396 participants, RR 0.87, 95% CI In the meantime, the concept of “internal convection” has
Table 40.4 Randomized controlled trials of HDF versus HD examining mortality as an endpoint (all involve postdilution HDF only).
Study Comparators N Follow-up Primary results Subanalysis Comments
2010, 27 Italian Centres LF HD, HF, or HDF 146 (70 HD, 36 1.5 years No difference HDF vs Mortality not the primary
(46). HF, 40 HDF) HD endpoint
Underpowered
Randomization was between HD
and convective therapy
Randomization imbalances
CONTRAST, 2013, LF HD vs. HDF 714 3 years Mortality, no benefit All-cause mortality benefit Benefit was in a post hoc analysis
Netherlands (37) if UF volume > 21.95 L. Not prespecified
No QoL benefit. Included some twice-weekly
dialysis
Turkish HDF, 2013 (74) HF HD vs. HDF 782 2 years Mortality, no benefit CV and overall survival Benefit was in a post hoc analysis
benefit if UF Not prespecified
volume > 17.4 L
ESHOL, Spain, 2013 HF HD vs. HDF (NB 906 3 years Mortality benefit for Prespecified requirement for UF
(72) 8% low-flux HD in HDF, all-cause HR volume >18 l
control arm) 0.70; CV mortality HR Not clearly randomized,
0.67 significant differences between
groups that potentially influenced
outcome
40% drop-outs
FRENCHIE, 2017, HF HD vs. HDF 381, over age 120 days for primary Tolerance of dialysis No benefit in primary Underpowered
France (38) 65 years outcomes, 2 years for sessions analysis All-cause and cardiovascular
secondary outcomes Possibly some benefit at mortality not different, including
“session” level accounting for convective volume
No difference in quality of life
Some predilution HDF included
ACM, all-cause mortality; AVF, arteriovenous fistula; CRP, C-reactive protein; CV, cardiovascular; HD, hemodialysis; HDF, hemodiafiltration; HF, high-flux membrane dialyzer; LF, low-flux
membrane dialyzer; QoL, quality of life.
c40.indd 14 09-12-2022 16:18:40

Treatmen 15
Table 40.5 Summary of findings: HDF (convective therapies) versus HD (diffusive therapies).
Hemodiafiltration (convective modalities) therapy versus conventional hemodialysis (diffusive modalities):

outcomes of interest
Absolute number of Quality of

patients affected per 1000 the
No. of participants patients treated for 1 year Relative effect evidence
Outcomes (studies) (95% CI) (95% CI) (GRADE) Conclusion
All-cause mortality 3396(11) 26 fewer (56 fewer to 0.87 Low HDF has little or no effect on
10 more) (0.72–1.05) ●●○○ all-cause mortality
Infection-related 2793 (4) Two fewer (eight fewer to 0.94 Very low It is uncertain whether HDF
mortalitya seven more) (0.68–1.30) ●○○○ reduces infection-related
mortality
Cardiovascular mortality 2889 (6) 25 fewer (39 to 8 fewer) 0.75 Low HDF may reduce
(0.61–0.92) ●●○○ cardiovascular mortality
Nonfatal cardiovascular 714 (1) 18 more (18 fewer to 1.14 Very low It is uncertain whether HDF
events 64 more) (0.86–1.50) ●○○○ reduces non-fatal
All hospital admissions 1688 (2) Not estimable 1.21 Very low It is uncertain whether HDF
(0.12–12.05) ●○○○ reduces hospital admissions
Dialysis-related 906 (1) Not estimablec 0.72 Low HDF probably reduces the risk
hypotensionb (0.66–0.80) ●●○○ of dialysis hypotension
Quality of life 988(8) Not estimable Not estimable Very low HDF/convective therapies
●○○○ have uncertain effects on
quality of life
Change in dialysis 2919 (5) Four fewer (19 fewer to 0.87 Low It is uncertain whether HDF
therapy 40 more) (0.30–2.52) ●●○○ results is a lower cross-over to
another dialysis therapy
Residual renal function . . .(..) ..(. . .) No studies Absent No studies found that
(RRF) evaluated the impact of HDF
on residual renal function
β2-microglobulin 67(1) Not estimable Not estimable Very low It is uncertain whether HDF
complication: carpal ●○○○ reduces carpal tunnel
tunnel syndrome syndrome incidence
Adverse events and . . .(..) ..(. . .) No studies Absent No studies found that
safety evaluated the impact of HDF
on adverse events and safety
The results where possible are derived from the Cochrane systematic review, Nistor et al. [47]; for some outcomes other systematic reviews are
referenced and indicated where relevant.
a
Metrics from Peters [73], 2016 individual pooled meta-analysis.
b
Metric described is from Nistor et al. (Cochrane systematic review), but in addition three of four aggregate study meta-analyses also showed a
benefit (consistent across all meta-analyses).
c
Estimate using Maduell et al. [72], events per 1000 patient years suggest a possible reduction by 2585 events per 1000 patient dialysis years
(3188–1875 events fewer).
re-emerged with the development of MCO membranes with International Society Guidelines
high Kuf values, with high obligatory ultrafiltration and there-
It is important to note the HDF is not practiced in the
fore back-filtration rates when used in HD mode. This has
United States and the uptake of HDF has been led by parts
some similarities to HDF, but all of the volume transfer occurs
of Europe and Japan. Being a comparably novel therapy
within the dialyzer (as ultrafiltration and back-filtration).
HDF has not been considered in the Canadian Society of
It has been argued that this approach may well provide many
Nephrology Clinical Practice, KDIGO, KDOQI, and KHA-
of the benefits of HDF without the need for additional
CARI guidelines.
equipment. This of course remains as yet unproven.
Guideline Year Recommendation
Japanese Society for Dialysis Therapy 2015 ●● HDF should enhance the removal of small proteins, reduce the production
Clinical Guideline [63] of inflammatory cytokines, and improve patients’ prognoses.
●● HDF should be considered a therapy for nonspecific symptoms, including
itching, joint pain, malaise, and poor appetite. Patients with dialysis
hypotension should be considered for treatment with HD using a high-flux
dialyzer and ultrapure dialysate.
Renal Association Clinical Practice 2011 ●● Both modalities are effective extracorporeal techniques for established renal
Guidelines [65] failure but hemodiafiltration can provide higher rates of removal of small
and middle molecules and may lower the risk of developing complications
due to dialysis-related amyloidosis.
●● We suggest that high-flux HD using ultrapure water provides noninferior
patient outcomes to hemodiafiltration.
European Best Practice Guidelines [75] 2007 ●● To exploit the high permeability of high-flux membranes, on-line HDF or
HF should be considered. The exchange volumes should be as high as
possible, with consideration of safety.
C
onclusion The choice of dialysis modality remains a controversial
topic, and the practice is highly variable across the world.
Despite the putative benefits of high-flux membranes such The large treatment benefits of convective therapies seen
as improved middle molecule clearance and better biocom- in early observational studies were not confirmed by subse-
patibility, large RCT trails have failed to show a clear mor- quent RCTs, although meta-analyses of intervention stud-
tality benefit with the use of high-flux dialysis membranes. ies suggest there may be a benefit in relation to
However, secondary analyses revealed a potential benefit cardiovascular mortality and reducing the frequency of
in those with diabetes, low serum albumin, and longer intradialytic hypotensive events. Moreover, the hypothesis-
dialysis vintage. generating data suggesting higher convective volumes may
The increased adoption of high-flux membranes in the be beneficial require further investigation in robust trials.
developed world has been driven in part by the presence of In summary, in modern dialysis units with ready availa-
these clinical characteristics in many HD patients, as well bility of high-quality dialysis water, synthetic, high-flux
as decreasing application of dialyzer reuse, which has dialysis membranes are now the accepted standard. The
resulted in a greater scale of manufacturing and decreased choice of dialysis modality remains less clear, and for most
costs. Therefore, high-flux membranes are the default patients these therapies are likely to be of equivocal clinical
choice in most developed countries, but modified cellulosic benefit. Whilst we believe that HDF should not be rou-
membranes and low-flux membranes remain in use in tinely offered to all patients, it should be considered in
many parts of the world and remain an appropriate choice patients with ongoing intradialytic instability and CVD,
in certain clinical situations. but only if adequate convection volumes can be achieved.
R
eferences
Ronco, C. and Clark, W.R. (2018). Haemodialysis

1 5 Kerr, P.G. and Huang, L. (2010). Review: membranes for
membranes. Nat. Rev. Nephrol. 14: 394–410. haemodialysis. Nephrology (Carlton) 15: 381–385.
2 Falkenhagen, D., Bosch, T., Brown, G.S. et al. (1987). A 6 Jindal, K.K., McDougall, J., Woods, B. et al. (1989). A study
clinical study on different cellulosic dialysis membranes. of the basic principles determining the performance of
Nephrol. Dial. Transplant. 2: 537–545. several high-flux dialyzers. YAJKD 14: 507–511.
3 Pereira, B.J., King, A.J., Poutsiaka, D.D. et al. (1993). 7 van Ypersele de Strihou, C., Jadoul, M., Malghem, J. et al.
Comparison of first use and reuse of cuprophan (1991). Effect of dialysis membrane and patient’s age on
membranes on interleukin-1 receptor antagonist and signs of dialysis-related amyloidosis. The Working Party on
interleukin-1 beta production by blood mononuclear cells. Dialysis Amyloidosis. Kidney Int. 39: 1012–1019.
Amer. J. Kidney Dis. 22: 288–295. 8 Weber, V., Linsberger, I., Rossmanith, E. et al. (2004).
4 Wehle, B. et al. (1993). Beta 2-microglobulin and granulocyte Pyrogen transfer across high-and low-flux hemodialysis
elastase. Nephrol. Dial. Transplant. 8 (Suppl 2): 20–24. membranes. Artif. Organs 28: 210–217.
Reference 17
9 Eknoyan, G. et al. (2002). Effect of dialysis dose and membranes: a randomized in vivo study. Blood Purif. 36:
membrane flux in maintenance hemodialysis. N. Engl. J. 287–294.
Med. 347: 2010–2019. 25 Hutchison, C.A. et al. (2012). Immunoglobulin free light
10 Locatelli, F. et al. (2009). Effect of membrane chain levels and recovery from myeloma kidney on
permeability on survival of hemodialysis patients. J. Am. treatment with chemotherapy and high cut-off
Soc. Nephrol. 20: 645–654. haemodialysis. Nephrol. Dial. Transplant. 27: 3823–3828.
11 Tattersall, J.E. and Ward, R.A. EUDIAL Group (2013). 26 Hutchison, C.A. et al. (2009). Treatment of acute renal
Online haemodiafiltration: definition, dose quantification failure secondary to multiple myeloma with
and safety revisited. Nephrol. Dial. Transplant. 28: chemotherapy and extended high cut-off hemodialysis.
542–550. Clin. J. Am. Soc. Nephrol. 4: 745–754.
12 Hartman, J. and Frishman, W.H. (2014). Inflammation 27 Kirsch, A.H. et al. (2017). Performance of hemodialysis
and atherosclerosis: a review of the role of with novel medium cut-off dialyzers. Nephrol. Dial.
interleukin-6 in the development of atherosclerosis and Transplant. 32: 165–172.
the potential for targeted drug therapy. Cardiol. Rev. 22: 28 Masakane, I. et al. (2015). An overview of regular dialysis
147–151. treatment in Japan (as of 31 December 2013). Ther. Apher.
13 Jefferis, B.J.M.H. et al. (2011). Interleukin 18 and Dial. 19: 540–574.
coronary heart disease: prospective study and systematic 29 Sichart, J.-M. and Moeller, S. (2011). Utilization of
review. Atherosclerosis 217: 227–233. hemodiafiltration as treatment modality in renal
14 Prasad, A., Bekker, P., and Tsimikas, S. (2012). Advanced replacement therapy for end-stage renal disease
glycation end products and diabetic cardiovascular patients – a global perspective. Contrib. Nephrol. 175:
disease. Cardiol. Rev. 20: 177–183. 163–169.
15 Meerwaldt, R. et al. (2005). Skin autofluorescence, a 30 Canaud, B. et al. (2004). On-line hemodiafiltration as
measure of cumulative metabolic stress and advanced routine treatment of end-stage renal failure: why pre-or
glycation end products, predicts mortality in mixed dilution mode is necessary in on-line
hemodialysis patients. J. Am. Soc. Nephrol. 16: hemodiafiltration today? Blood Purif. 22 (Suppl 2): 40–48.
3687–3693. 31 Ledebo, I. and Blankestijn, P.J. (2010).
16 Kato, S. et al. (2008). Aspects of immune dysfunction in Haemodiafiltration-optimal efficiency and safety. NDT
end-stage renal disease. Clin. J. Am. Soc. Nephrol. 3: Plus 3: 8–16.
1526–1533. 32 Teatini, U., Steckiph, D., and Romei Longhena, G. (2011).
17 Kaizu, Y. et al. (2003). Association between inflammatory Evaluation of a new online hemodiafiltration mode with
mediators and muscle mass in long-term hemodialysis automated pressure control of convection. Blood Purif. 31:
patients. Am. J. Kidney Dis. 42: 295–302. 259–267.
18 Gupta, J. et al. (2012). Association between albuminuria, 33 Damasiewicz, M.J., Polkinghorne, K.R., and Kerr, P.G.
kidney function, and inflammatory biomarker profile in (2012). Water quality in conventional and home
CKD in CRIC. Clin. J. Am. Soc. Nephrol. 7: 1938–1946. haemodialysis. Nat. Rev. Nephrol. 8: 725–734.
19 Garibotto, G. et al. (2006). Peripheral tissue release of 34 Lin, C.-L. et al. (2003). Reduction of advanced glycation
interleukin-6 in patients with chronic kidney diseases: end product levels by on-line hemodiafiltration in
effects of end-stage renal disease and microinflammatory long-term hemodialysis patients. Am. J. Kidney Dis. 42:
state. Kidney Int. 70: 384–390. 524–531.
20 Wolley, M., Jardine, M., and Hutchison, C.A. (2018). 35 Panichi, V. et al. (2008). Chronic inflammation and
Exploring the clinical relevance of providing increased mortality in haemodialysis: effect of different renal
removal of large middle molecules. Clin. J. Am. Soc. replacement therapies. Results from the RISCAVID study.
Nephrol. 13: 805–814. Nephrol. Dial. Transplant. 23: 2337–2343.
21 Vanholder, R. et al. (2003). Review on uremic toxins: 36 Carracedo, J. et al. (2006). On-line hemodiafiltration
classification, concentration, and interindividual reduces the proinflammatory CD14+CD16+ monocyte-
variability. Kidney Int. 63: 1934–1943. derived dendritic cells: a prospective, crossover study. J.
22 Leypoldt, J.K. (2000). Solute fluxes in different treatment Am. Soc. Nephrol. 17: 2315–2321.
modalities. Nephrol. Dial. Transplant. 15 (Suppl 1): 3–9. 37 Grooteman, M.P.C. et al. (2012). Effect of online
23 Gondouin, B. and Hutchison, C.A. (2011). High cut-off hemodiafiltration on all-cause mortality and
dialysis membranes: current uses and future potential. cardiovascular outcomes. J. Am. Soc. Nephrol. 23:
Adv. Chronic Kidney Dis. 18: 180–187. 1087–1096.
24 Kneis, C. et al. (2013). Elimination of middle-sized 38 Morena, M. et al. (2017). Treatment tolerance and
uremic solutes with high-flux and high-cut-off patient-reported outcomes favor online hemodiafiltration
compared to high-flux hemodialysis in the elderly. Kidney 53 Port, F.K. et al. (2001). Mortality risk by hemodialyzer
Int. 91: 1495–1509. reuse practice and dialyzer membrane characteristics:
39 Locatelli, F., Carfagna, F., Del Vecchio, L., and La Milia, results from the usrds dialysis morbidity and mortality
V. (2018). Haemodialysis or haemodiafiltration: that is study. Am. J. Kidney Dis. 37: 276–286.
the question. Nephrol. Dial. Transplant. 33: 1896–1904. 54 Cheung, A.K. et al. (2003). Effects of high-flux
40 CW, M.I. (2010). Haemodialysis-induced myocardial hemodialysis on clinical outcomes: results of the HEMO
stunning in chronic kidney disease – a new aspect of study. J. Am. Soc. Nephrol. 14: 3251–3263.
cardiovascular disease. Blood Purif. 29: 105–110. 55 Asci, G. et al. (2013). The impact of membrane
41 McIntyre, C. and Crowley, L. (2016). Dying to feel better: permeability and dialysate purity on cardiovascular
the central role of dialysis-induced tissue hypoxia. Clin. J. outcomes. J. Am. Soc. Nephrol. 24: 1014–1023.
Am. Soc. Nephrol. 11: 549–551. 56 Palmer, S.C. et al. (2012). High-flux versus low-flux
42 Meyring-Wösten, A. et al. (2016). Intradialytic hypoxemia membranes for end-stage kidney disease. Cochrane
and clinical outcomes in patients on hemodialysis. Clin. J. Database Syst. Rev. (35) Art. No.: CD005016. doi:
Am. Soc. Nephrol. 11: 616–625. 10.1002/14651858.cd005016.pub2.
43 Eldehni, M.T., Odudu, A., and McIntyre, C.W. (2015). 57 Zickler, D. et al. (2016). High cut-off dialysis in chronic
Randomized clinical trial of dialysate cooling and haemodialysis patients reduces serum procalcific activity.
effects on brain white matter. J. Am. Soc. Nephrol. 26: Nephrol. Dial. Transplant. 31: 1706–1712.
957–965. 58 Belmouaz, M. et al. (2018). Comparison of hemodialysis
44 Mion, M. et al. (1992). Haemodiafiltration in high- with medium cut-off dialyzer and on-line
cardiovascular-risk patients. Nephrol. Dial. Transplant. 7: hemodiafiltration on the removal of small and middle-
453–454. sized molecules. Clin. Nephrol. 89 (2018): 50–56.
45 Maduell, F. et al. (1999). Change from conventional 59 Trojanowicz, B. et al. (2018). Modulation of leucocytic
haemodiafiltration to on-line haemodiafiltration. angiotensin-converting enzymes expression in patients
Nephrol. Dial. Transplant. 14: 1202–1207. maintained on high-permeable haemodialysis. Nephrol.
46 Locatelli, F. et al. (2010). Hemofiltration and Dial. Transplant. 33: 34–43.
hemodiafiltration reduce intradialytic hypotension in 60 Zickler, D. et al. (2017). Medium cut-off (MCO)
ESRD. J. Am. Soc. Nephrol. 21: 1798–1807. membranes reduce inflammation in chronic dialysis
47 Nistor, I. et al. (2015). Haemodiafiltration, haemofiltration patients-a randomized controlled clinical trial. PLoS One
and haemodialysis for end-stage kidney disease. Cochrane 12: e0169024.
Database Syst. Rev. (17) Art. No.: CD006258. DOI https:// 61 Krishnasamy, R. et al. (2018). Design and methods of the
doi.org/10.1002/14651858.CD006258.pub2. REMOVAL-HD study: a tRial evaluating mid cut-off
48 Susantitaphong, P., Siribamrungwong, M., and Jaber, B.L. value membrane clearance of albumin and light chains in
(2013). Convective therapies versus low-flux hemodialysis haemodialysis patients. BMC Nephrol. 19: 89.
for chronic kidney failure: a meta-analysis of randomized 62 National Kidney Foundation (2015). KDOQI clinical
controlled trials. Nephrol. Dial. Transplant. 28: 2859–2874. practice guideline for hemodialysis adequacy: 2015
49 Buchanan, C. et al. (2017). Intradialytic cardiac magnetic update. Am. J. Kidney Dis. 66: 884–930.
resonance imaging to assess cardiovascular responses in a 63 Watanabe, Y. et al. (2015). Japanese Society for Dialysis
short-term trial of hemodiafiltration and hemodialysis. J. Therapy Clinical Guideline for ‘maintenance
Am. Soc. Nephrol. 28: 1269–1277. hemodialysis: hemodialysis prescriptions’. Ther. Apher.
50 Hornberger, J.C., Chernew, M., Petersen, J., and Garber, Dial. 19: 67–92.
A.M. (1992). A multivariate analysis of mortality and 64 Kerr, P.G. and Toussaint, N.D. (2013). Kidney health
hospital admissions with high-flux dialysis. J. Am. Soc. Australia caring for Australasians with renal impairment.
Nephrol. 3: 1227–1237. KHA-CARI guideline: dialysis adequacy (haemodialysis):
51 Koda, Y. et al. (1997). Switch from conventional to dialysis membranes. Nephrology 18: 485–488.
high-flux membrane reduces the risk of carpal tunnel 65 Mactier, R., Hoenich, N., and Breen, C. (2011). Renal
syndrome and mortality of hemodialysis patients. Kidney association clinical practice guideline on haemodialysis.
Int. 52: 1096–1101. Nephron Clin. Pract. 118 (Suppl 1): c241–c286.
52 Woods, H.F. and Nandakumar, M. (2000). Improved 66 Tattersall, J. et al. (2010). High-flux or low-flux dialysis: a
outcome for haemodialysis patients treated with high- position statement following publication of the
flux membranes. Nephrol. Dial. Transplant. 15 (Suppl 1): membrane permeability outcome study. Nephrol. Dial.
36–42. Transplant. 25: 1230–1232.
Reference 19
7 Canaud, B. et al. (2006). Mortality risk for patients receiving

6 72 Maduell, F. et al. (2013). High-efficiency postdilution
hemodiafiltration versus hemodialysis: European results online hemodiafiltration reduces all-cause mortality
from the DOPPS. Kidney Int. 69: 2087–2093. in hemodialysis patients. J. Am. Soc. Nephrol. 24:
68 Vilar, E. et al. (2009). Long-term outcomes in online 487–497.
hemodiafiltration and high-flux hemodialysis: a comparative 73 Peters, S.A.E. et al. (2016). Haemodiafiltration and
analysis. Clin. J. Am. Soc. Nephrol. 4: 1944–1953. mortality in end-stage kidney disease patients: a pooled
69 Mercadal, L. et al. (2016). Hemodiafiltration versus individual participant data analysis from four
hemodialysis and survival in patients with ESRD: the randomized controlled trials. Nephrol. Dial. Transplant.
French Renal Epidemiology and Information Network 31: 978–984.
(REIN) registry. Am. J. Kidney Dis. 68: 247–255. 74 Ok, E. et al. (2013). Mortality and cardiovascular
70 Locatelli, F. et al. (2017). Mortality risk in patients on events in online haemodiafiltration (OL-HDF)
hemodiafiltration versus hemodialysis: a ‘real-world’ compared with high-flux dialysis: results from the
comparison from the DOPPS. Nephrol. Dial. Transplant. Turkish OL-HDF Study. Nephrol. Dial. Transplant.
33: 683–689. 28: 192–202.
71 Masakane, I., Kikuchi, K., and Kawanishi, H. (2017). 75 Tattersall, J. et al. (2007). EBPG guideline on dialysis
Evidence for the clinical advantages of predilution strategies. Nephrol. Dial. Transplant. 22 (Suppl 2):
on-line hemodiafiltration. Contrib. Nephrol. 189: 17–23. ii5–ii21.
20
41
Dialysis Dose and Adequacy for Hemodialysis

Brendan Smyth1,2, Carmel M Hawley3, and Meg Jardine2,4
1
Department of Renal Medicine, St George Hospital, Kogarah, Australia
2
NHMRC Clinical Trials Centre, University of Sydney, Camperdown, Australia
3
Department of Nephrology, Princess Alexandra Hospital, University of Queensland, Brisbane, Australia
4
Department of Nephrology, Concord Repatriation General Hospital, Australia
Introduction Definition of Dialysis Adequacy: Meets

Patient Needs
Dialysis is a laborious and expensive process for both It is important to distinguish between the narrow defini-
patients and healthcare providers. Naturally, all stakehold- tion of adequacy as the “minimum acceptable urea clear-
ers are interested in knowing “How much is enough?” The ance” and the broader concept of the “quantum and quality
definition of “adequacy” – the necessary amount of dialy- of dialysis (and associated therapies) that the patient
sis – continues to evolve as our understanding of the impact requires.” The two are not interchangeable. The former
of dialysis on patient quality of life and survival advances. measure – in the form of the urea-reduction ratio (URR) or
The initial approach of Scribner and colleagues was to pro- its normalized derivation (Kt/V) – retains importance as a
vide sufficient dialysis to alleviate the patient’s uremic useful and easily interpretable reference point. The latter
symptoms [2]. Then, with the observation that urea clear- recognizes that dialysis performs many functions – clear-
ance was associated with survival, the focus shifted to ance of solutes (ranging from 60 to 60,000 daltons in mass)
delivering sufficient dialysis to achieve urea clearance tar- and control of sodium, water balance, and blood pressure –
gets using a normalized urea clearance formula: Kt/V. and forms part of a broader suite of therapies that also aim
More recently, as the limitations of a “urea-centric” to maintain bone and mineral homeostasis and red blood
approach have become apparent, there has been growing cell production. No single measurement can combine all of
interest in dialysis frequency and duration, and in the pres- these parameters. Moreover, urea-based adequacy does not
ervation of residual renal function. Finally, with an aging consider the broader impacts of dialysis therapy on quality
dialysis population, in which multimorbidity is the norm, of life or the potential for dialysis-related complications
an appreciation of patient goals and quality of life is help- such as access dysfunction or reduction in residual renal
ing to add nuance to patient–clinician discussions such function. A true definition of adequacy therefore has to be
that they extend beyond biochemical targets and mortality. multidimensional and flexible enough to accommodate
Our approach in this chapter will be to consider dialysis differing patient characteristics and goals of therapy.
adequacy as a multidimensional concept, incorporating a We propose a hierarchy of needs for dialysis recipients
comprehensive assessment of the patient’s needs, symp- ranging from basic physiological survival to achievement of
toms, and goals, and within which minimum urea- the best possible quality of life and longevity. Inspired by
clearance standards provide a useful, but far from definitive, Maslow [3], this approach places the needs of the patient on
benchmark.
Epidemiolog 21
Table 41.1 Assessment of level of evidence for intensity of hemodialysis therapy.
Strength of
Minimum If <4 h per session is prescribed in a three times per week Very low Weak
dialysis hours schedule, we suggest a comprehensive assessment of
adequacy, including rates of volume removal, is performed
regularly to avoid under-dialysis.
Increased We suggest increasing dialysis hours for willing patients High (hyperphosphatemia, pill Weak
dialysis hours based on evidence suggesting improvements in quality of burden)
life, hyperphosphatemia, pill burden, and blood pressure. Moderate (left ventricular
We suggest that longer hours may also be a useful way to hypertrophy)
treat patients with high ultrafiltration rate requirements, Low (quality of life, blood
uncontrolled hypertension, poor phosphate control, and/or pressure control)
to improve some aspects of quality of life.
Dialysis We suggest that more frequent dialysis sessions (four to six Moderate to high Weak
frequency per week) be offered to patients wishing to improve blood (hyperphosphotaemia, pill
pressure, improve serum phosphate or reduce pill burden. It burden)
may also be recommended in patients with left ventricular Moderate (blood pressure, left
hypertrophy and may have small benefits for some aspects ventricular hypertrophy)
of quality of life. Patients taking on this therapy should be Low (quality of life, vascular
advised of the possible risks of increased vascular access access events and residual renal
events and more rapid loss of residual renal function. function)
Targeting urea We recommend achieving at least an spKt/V of 1.2. We Low Weak
clearance suggest that this not be used as the primary measure of
dialysis adequacy.
Limiting We suggest targeting ultrafiltration rates below 10 ml/h/kg. Very low Weak
volume
Removal
Incremental We suggest that incremental dialysis be undertaken only Very low Weak
dialysis within a structured program including regular reassessment
of residual renal function and dialysis adequacy.
that the effect will be substantially different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood that the
effect will be substantially different* is moderate. Low: This research provides some indication of the likely effect. However, the likelihood that it
will be substantially different* is high. Very low: This research does not provide a reliable indication of the likely effect. The likelihood that the
a continuum (Figure 41.1). Adequacy of dialysis therapy is 3.8 million in 2020 and rise to 5.4 million by 2030 [4]. This
then judged from the patient perspective in relation to the burden is not distributed equally and rates of dialysis and
degree to which these needs are met. This schema aims to transplant uptake vary dramatically around the globe, from
encourage clinicians to go beyond urea clearance by focusing over 2000 per million population in Japan to as low as 2.8
on biological parameters known to be associated with per million population in Rwanda [5]. These figures are
improved patient outcome and on a patient-centered concept also likely to hide as many as 9.7 million individuals with
of wellbeing. In general, “more” dialysis – whether increased ESKD who are unable to receive renal replacement therapy
frequency or duration – may be helpful for many patients in due to resource limitations [4]. For those fortunate enough
achieving higher-level needs but cannot be considered “ade- to have access to dialysis, hemodialysis is the dominant
quate” without reference to the individual patient. modality, with the proportion of dialysis delivered as
hemodialysis estimated to be 89% in 2013 [6]. This also var-
ies from region to region, with the proportion on hemodi-
Epidemiology alysis being as low as 20.6% in Hong Kong and greater than
99% in some countries [7].
The number of individuals receiving renal replacement Three sessions per week is near universal in high-
therapy is growing steadily, it is expected to reach income countries, with less than 5% of patients receiving
22 Dialysis Dose and Adequacy for Hemodialysis
Consider quality of life and related goals; consider patient wishes regarding
longevity and fitness for transplantation; minimize dietary restrictions and pill Achieve health and
burden; manage specific situations requiring high dialysis doses such as well-being
pregnancy.
Optimize biochemical, hemodynamic and fluid status; minimize rapid Optimize surrogate
ultrafiltration; management of left ventricular hypertrophy outcomes
Avoid inadequate small molecule clearance (achieve

Clear small
Kt/V> 1.2 for thrice weekly dialysis) and intra-dialytic solutes
hypotension
Treat immediate symptoms of ESKD including overt

Sustain
Uraemia, symptomatic volume overload and gross
life
biochemical abnormalities
Figure 41.1 Hierarchy of needs for the hemodialysis patient.
two or fewer sessions per week [8]. More frequent dialysis Clearance

is also uncommon, although its prevalence varies from
14% in Sweden to <1% in the United States, Japan, and
France [8]. In contrast, in 2011 up to 26% of patients in
China were dialyzing twice weekly [9]. Although accurate
data is lacking, twice-weekly dialysis is believed to be C
common practice in many low-and middle-income coun-
tries. Average session length also varies globally, with the
B
mean value in most countries lying between 3.5 and P
4 hours [8–10]. Other differences in dialysis delivery
should be noted, with high-income countries typically
providing high-flux dialysis with no dialyzer reuse. In con- A
trast, low-flux dialysis and dialyzer reuse is common in
many middle-income countries [11]. This is likely to con- Time
tribute to poorer outcomes in these countries, although
Figure 41.2 Clearance vs. time curves for solutes during a
the relative impact of these practices on morbidity is not
single dialysis session. A represents small solute clearance (e.g.
well described. urea). B and C represent larger solutes of varying weight (e.g.
middle molecules). P (dashed line) represents clearance of a
sequestered small solute (e.g. phosphate).
Pathophysiology illustrates the three concepts central to understanding the

prescription of dialysis time and frequency:
In high-income countries, where the use of high-efficiency
and high-flux dialyzers is now standard, the question of dial- 1) Solute clearance is greatest at the start of dialysis when
ysis adequacy is essentially a discussion of the relationship the concentration gradients are highest [12]. For this
of solute and water clearance with dialysis time and fre- reason, more frequent dialysis sessions (e.g. six vs.
quency. Detailed discussion of dialyzer characteristics and three times per week) allow more time on the steepest
the impact of blood and dialysate flow rates can be found part of the removal curve, increasing small solute clear-
elsewhere [12] and is beyond the scope of this chapter. ance even if total dialysis time remains the same [13].
Figure 41.2 demonstrates the relationship between rate 2) As solute size increases, the rate of diffusion decreases.
of solute clearance on dialysis and dialysis session time. It Thus, as solute size increases the concentration gradient
Diagnosi 23
between blood and dialysate changes less rapidly, lead-

Body water
ing to greater “gain” in clearance from longer sessions. A
As a result, total dialysis time is more important than
frequency for middle-molecule clearances.
3) “Sequestered” solutes display a biphasic clearance. B
Initially, the rate of clearance from the first compart-
ment (blood or extracellular space) may decline rela-
tively rapidly as the concentration gradient C
diminishes. This is followed by the mobilization of
solute from secondary compartments which main-
tains the concentration gradient between blood and Time
dialysate, and leads to a second phase of clearance in
Figure 41.3 Pattern of volume changes with different dialysis
which the rate of clearance is maintained over the
schedules. (A) Thrice weekly 8 hours, (B) thrice weekly 4 hours, (C)
later hours of the dialysis session. The classic example frequent dialysis 2.4 hours × five sessions per week. Arrows
of this phenomenon is phosphate. Up to four com- indicate dialysis sessions.
partments have been proposed to adequately model
intradialytic phosphate clearance [14, 15]. In all mod-
treatment. We present here an indication of the informa-
els, the initial clearance from the blood compartment
tion required to facilitate an adequacy assessment. An
is rapid, but beginning in the second hour of dialysis
introduction to principles of urea kinetic modeling (UKM)
the rate of clearance plateaus as sequestered solute is
is also provided as this remains a useful component of the
mobilized from less accessible compartments. The
adequacy assessment.
result is that phosphate clearance can be improved
with both increased session frequency (maximizing
the time with rapid clearance) and increasing dialysis Comprehensive Clinical Assessment
time (as clearance does not diminish over longer dial-
The comprehensive clinical assessment should include the
ysis sessions).
following:
Water “clearance” – i.e. ultrafiltration – is distinct from 1) Symptoms
other parameters. The rate of ultrafiltration required to a) Uremic symptoms, such as nausea, poor appetite,
maintain euvolemia is determined by fluid intake, urine pruritis, and restless leg syndrome.
output, and the interaction between dialysis frequency and b) Symptoms of over- or under-hydration, including
session length. More frequent sessions minimize fluctua- dyspnea, edema, thirst, and orthostatic dizziness.
tions in volume status and longer sessions permit slower c) Adverse symptoms of dialysis, including cramps on
rates of ultrafiltration. Both large intradialytic weight gain dialysis and post-dialysis fatigue.
and rapid ultrafiltration are associated with increased mor- d) Common comorbid symptoms, such as pain, dis-
bidity and mortality [16, 17]. Increasing session length (if turbed sleep, and limitations to mobility
frequency remains the same) tends to reduce the rate of 2) Physical examination
ultrafiltration. Increasing session frequency (if total dialy- a) Volume status and blood pressure
sis hours remains the same) may reduce the size of intra- b) Access assessment
dialytic weight gain (Figure 41.3). 3) Dialysis parameters
a) Ideal body weight and interdialytic weight gain
b) Pre-and postdialysis blood pressure
Diagnosis c) Kt/V urea
4) Laboratory parameters
Dialysis therapy must be tailored to fit the patient. This a) Small solute clearance: urea, creatinine, potassium
requires a thorough exploration of the patient’s symptoms, b) Anemia: hemoglobin, ferritin, and iron-binding
nutritional status, cardiovascular risk, and fluid status, capacity
bone, and metabolic health and, equally important, their c) Volume status: sodium
goals, functional limitations and supports. Hence, assess- d) Nutrition: albumin, bicarbonate
ment of adequacy requires a thorough clinical assessment e) Bone and mineral disorder: phosphate, calcium,
and a frank discussion about patient expectations of parathyroid hormone, alkaline phosphatase
5) Medication review t = dialysis time (min)

a) Avoid overprescribing and consider pill burden (hence, Kt represents the volume (ml) of plasma
6) Cardiovascular assessment cleared of urea during the dialysis session)
a) Consider screening for left ventricular hypertrophy V = postdialysis volume of distribution of urea (ml) (equiv-
7) Functional status alent to total body water)
a) Multiple standardized measures of function and
The advantage of this formula is that it is not dependent
frailty exist, including the Karnovsky Performance
on absolute levels of blood urea, and therefore is independ-
Score or Clinical Frailty Scale [18]. Simple tests such
ent of nutrition. Rather, as dialyzer urea clearance is a gen-
as the Timed Up and Go may also be effective screen-
eral surrogate for dialytic clearance of many small
ing tools [13].
molecules, it represents a generalizable dialysis parameter.
8) Social environment and supports
In choosing a dialyzer with known characteristics, dialysate
a) Understand the patient’s social environment and the
and blood flow rates and treatment time, one can specify a
demands on their time that may conflict with dialy-
desired Kt/V. UKM computer programs are available that
sis (e.g. employment, training, caregiver). Consider
facilitate this calculation.
the needs of the patient’s caregivers. Patient and car-
In practice, owing to access recirculation and the error
egiver responsibilities may heavily influence the
involved in the multiple assumptions underlying the
choice of dialysis frequency and session length.
model, prescribed Kt/V often differs substantially from the
9) Goals and expectations
actual “delivered” Kt/V. Delivered Kt/V is therefore the key
a) What does the patient wish to achieve with dialysis
value to aid in adequacy assessments. Values below target
therapy? What are their goals and how do they prior-
should prompt titration of the variable parameters (for
itize those goals? For instance, do they wish to receive
instance an increase in dialyzer size or session time) to
a kidney transplant? Is dialysis aimed to extend life
increase clearance (consideration of access recirculation
expectancy? To maximize their function? To spend
may also be warranted), with improvement confirmed by
time with family or friends? To enable travel and flex-
subsequent testing of the delivered Kt/V.
ibility? Are their (and their family’s or carer’s) expec-
tations of hemodialysis realistic?
Single-pool Kt/V
In its simplest form, Kt/V is calculated assuming that urea
Urea Kinetic Modeling is distributed evenly through one fluid compartment and is
UKM refers to mathematical modeling that approximates the known as a single pool, or spKt/V. The most widely used
observed clearance of urea during one or more dialysis ses- equation to calculate delivered spKt/V is that of
sions by relating basic patient characteristics, such as urea Daugirdas [12]:
generation and total body water, with parameters of dialysis
therapy, such as time, dialysate flow rate, and dialyzer urea spKt / V ln U post / U pre 0.008 t 4 3.5 U post / U pre
clearance. The field developed with the understanding that Vuf / weight
simply targeting low blood urea levels did not always result in
better patient outcomes as low blood urea levels often where Upost/Upre is the ratio of postdialysis to predialysis
reflected poor nutrition rather than adequate dialysis [19]. urea, t is the treatment time in hours, Vuf is the ultrafiltra-
The details of modern UKM are beyond the scope of this tion volume, weight is bodyweight in kilograms, and ln is
chapter and quickly exhaust the patience of many physi- the natural logarithm.
cians. Fortunately, as many excellent resources exist for those This equation has the advantage of including ultrafiltra-
wanting a deeper understanding of the topic [7], we will tion volume (hence convective urea clearance) and intra-
focus attention mainly on the key concept of Kt/Vurea and the dialytic urea generation. It is validated between Kt/V values
evidence supporting its current role in dialysis prescription. of 0.7 and 2.1 [20].
Equilibrated Kt/V
Kt/V
The assumption of a single pool of urea is an over-
The most commonly used marker of small molecule clear-
simplification, as demonstrated by the fact that blood urea
ance is Kt/V, where:
levels rapidly rebound after dialysis is completed. This
K = dialyzer urea clearance (ml/min) (a function of dia- means that spKt/V overestimates urea clearance. To
lyzer surface area and permeability, and blood and account for this, equilibrated, or eKt/V can be calculated,
dialysate flow rates) either by taking a second urea sample 35 minutes after the
Diagnosi 25
completion of the dialysis session or by modifying spKt/V has been speculated to contribute to the observed survival
using a conversion equation shown to reliably approximate benefits of higher body fat in hemodialysis recipients [23].
measured eKt/V [8]: Conductivity-based measures of Kt/V are now available
on many dialysis machines. These operate by manipulating
eKt / V spKt / V t / t 35 the sodium concentration of the inflowing dialysate in a

controlled fashion and measuring the resulting differences
where t is dialysis time in minutes.
in electrical conductivity between the inflowing and out-
flowing dialysate. This provides a ready measure of small
Standard Kt/V
molecule clearance that can be used as a surrogate for urea
Standard Kt/V (stdKt/V) was devised to permit compari-
clearance and therefore Kt/V. While correlated with urea-
sons between dialysis regimens of different frequency or
based Kt/V, conductivity measures tend to underestimate
with peritoneal dialysis. The modeling equations utilize
Kt/V and within-patient variation may be high [24, 25].
the pre- and postdialysis blood urea, and dialysis and
Given these limitations and the lack the clear observational
patient parameters over a 1-week period to estimate the
and randomized controlled trial data to support their use,
urea clearance required to maintain those values in a
conductivity measures should be supplemented by peri-
steady state. A number of different modeling equations are
odic urea-based estimates of delivered Kt/V.
available to estimate stdKt/V, some of which incorporate
residual renal function and/or convective urea clear-
ance [9, 10]. StdKt/V is most important if patients receive Assessment of Residual Kidney Function
incremental hemodialysis (less than three times per week)
Residual kidney function (RKF) is associated with quality of
as an aid to the up-titration of dialysis in line with loss of
life [26] and lower mortality [27, 28]. This is thought to be due
residual renal function.
to the continuous provision of both small and middle molecu-
lar clearance and the beneficial impact on mineral metabo-
Urea Reduction Ratio
lism and smoothing of fluid status fluctuations. Assessment
URR is the simple ratio of the change in urea during dialy-
of RKF typically plays little part in the prescription of hemodi-
sis to the predialysis urea level (conventionally expressed
alysis, with most patients commencing a three times per week
as a percentage):
schedule from the initiation of dialysis [29]. This approach,
URR U pre U post / U pre 100 essentially treating RKF as “bonus” clearance that does not

require further consideration, has been challenged due to the
(This is sometimes written as the equivalent formula observation that more frequent dialysis may be associated
URR = 1 – (Upost/Upre).) with more rapid loss of renal function [30].
URR is related exponentially to Kt/V such that, as Kt/V Measurement of RKF is hampered by the fact that small
rises, further large increases result in only small changes in solutes fluctuate with each dialysis session, that glomerular
URR. For example moving Kt/V from 1.5 to 2.0 results in an filtration is not constant in the interdialytic interval [31],
increase in URR from 73% to 82%. While simpler, it does and that extrarenal clearance (e.g. gastrointestinal) is not
not account for urea clearance due to ultrafiltration, nor well-understood or quantified [32]. Optimal measurement
the generation of urea during the dialysis session. Despite demands urine collection over the entire (e.g. 44-hour)
these limitations, observational evidence suggests that interdialytic interval. Creatinine-based measurements tend
URR predicts outcomes similarly to Kt/V and it remains a to overestimate RKF owing to tubular creatinine secretion,
part of current treatment guidelines [21]. whereas urea-based measures underestimate it due to pas-
sive reabsorption of urea in the distal tubules.
Caveats in Interpreting Kt/V Multiple methods of assessing RKF exist. Residual urea
It is important to be aware of the risk of under-dosing by clearance (Kru) can be reasonably approximated in a patient
Kt/V in smaller patients, such as women and children, in performing three sessions per week using a 24-hour urine
whom the lower value of V may lead to the prescription of collection commencing the day prior to a mid-week dialy-
shorter dialysis sessions, smaller dialyzers, or slower blood sis session using the following formula:
or dialysate flows. The risks of this were emphasized by in
the HEMO trial, in which women randomized to the “stand- U urine
K ru urine flow rate ml / min
ard” Kt/V target had a higher mortality than those in the Userum 0.9
"high" Kt/V arm [22]. Conversely, patients with higher body
fat, in whom V overestimates total body water will tend to where 0.9 is a factor that adjusts the predialysis serum urea
require increased dialysis dosing to achieve target Kt/V. This value to approximate the mid-week mean [31].
An alternative is to estimate RKF by the average of Kru patients and the beneficial impact of improved fluid con-
and KrCr (residual creatinine clearance), utilizing the fact trol and increased dialysis frequency or duration on blood
that the former underestimates and latter overestimates pressure (see below) should be included in assessment of
true RKF. Some authors, however, recommend relying dialysis adequacy.
only on Kru on the understanding that the underestimate
results in an assumed “buffer” providing a modicum of
safety from under dialysis [33].
Treatment
Evidence Behind the Current Guidelines

Prognosis
Current guidelines are derived from the results of two piv-
Longer treatment times are associated with longer life otal clinical trials: the National Co-operative Dialysis Study
expectancy. Using data from 12 countries, Tentori et al. (NCDS) and the HEMO study (Box 41.1). The NCDS, pub-
demonstrated an average reduction in risk of mortality of lished in 1981, sought to determine if lower time-averaged
6% for every 30 minute increase in session length [34]. The urea levels were superior to higher, and if 3 vs. 4.5 hours
reasons for this finding are not clearly understood. Higher was superior [42]. The time averaged urea (TACurea) targets
Kt/V may be one factor, but owing to the kinetics of small of 50 vs. 100 mg/dl (18 vs. 36 mmol/l) were approximately
solute clearance the increase in Kt/V with increased time equivalent to predialysis urea levels of 70 and 120 mg/dl
steadily diminishes as session length increases (c.f. (25 vs. 43 mmol/l). Patients with residual creatinine clear-
Figure 41.2). Indeed, the observed mortality relationship is ance greater than 3 ml/min were excluded. The study was
lost above Kt/V of approximately 1.3 [35]. Other factors are terminated early as participants in the high TACurea arm
clearly important. had a higher incidence of death, withdrawal from dialysis,
Observational studies have shown a dose-related increase and hospitalization. There were no significant differences
in mortality with increasing ultrafiltration rates [16, 17]. between those assigned to 3 vs. 4.5 hours of dialysis.
For example, a 22% increase in mortality in those with rates Subsequent criticism of the study focused on the fact that
>10 ml/h/kg (vs. 10 ml/h/kg) and a further 31% increase the use of urea concentration as the treatment target was
in those with rates >13 ml/h/kg (vs. 10–13 ml/h/kg) [36]. likely to have led to participants with a higher protein
The benefits of slower ultrafiltration are hypothesized to intake to be prescribed higher dialytic clearance compared
stem from lower incidence of intradialytic hypotension to participants in the same arm of the study with a lower
and the myocardial strain associated with dialysis [37]. protein intake. This resulted in divergent outcomes
Other potential benefits include preservation of RKF, depending on nutritional status, with little between-group
which in itself is clearly associated with improved progno- separation in delivered dialysis clearance in those with low
sis [28]. Furthermore, increased session length is associ- protein intake and a large difference for those with a higher
ated with lower phosphate levels, which are consistently protein intake. To account for this, Gotch and Sargent
associated with lower mortality [38]. developed Kt/V as a way to measure dialysis dose indepen-
Other prognostic factors that might impact on adequacy dently of protein catabolic rate [19]. Their reanalysis of the
assessment are hypertension and fluid overload. The asso- NCDS data established that good clinical outcomes were
ciation with fluid overload is consistent. Studies utilizing associated with Kt/V > 0.8. Further work increased this
bioimpedance spectroscopy have shown fluid overload cut-off value to 1.0 and observational data suggested that
(both pre-and postdialysis) to be positively associated with outcomes continued to improve above this value [43, 44].
mortality [39, 40]. The relationship between hypertension The resulting uncertainty as to the optimal dose (based on
and mortality in the ESKD population is the subject of sub- Kt/V) led to the development of the HEMO study.
stantial debate. Observational studies suggest a U-shaped The HEMO study, published in 2002, randomized partici-
association, with marked increases in mortality with low pants in a factorial design to high or low flux membrane
(<130/70 mmHg) predialysis systolic and diastolic blood and a higher or lower dialysis dose by Kt/V [22]. The low
pressure, and less pronounced increases with systolic blood and high Kt/V arms achieved an spKt/V of 1.32 ± 0.09
pressure >180 mmHg [41]. Confounding by cardiovascular (eKt/V 1.16 ± 0.08) and 1.71 ± 0.11 (eKt/V 1.53 ± 0.09),
disease is likely to contribute to this phenomenon. respectively; approximately equal to a URR of 65–75%. The
Interdialytic blood pressure is poorly correlated with predi- study found no difference in mortality with dialysis dose (or
alysis values and may display a different (more traditional) flux): relative risk (high vs. standard) 0.96 (0.84–1.10,
relationship with mortality. This uncertainty aside, current P = 0.53) and, for high vs. low flux, 0.92 (0.81–1.05, P = 0.23).
practice remains to treat hypertension in hemodialysis Nor were there differences in hospital admissions. The
Treatmen 27
higher Kt/V was achieved not only by increasing treatment The observed benefits of longer dialysis times may be in
time (average 219 vs. 190 minutes) but also by a 20% increase part due to slower ultrafiltration, with ultrafiltration rates
in blood flow rates and small increases in dialyzer size and <10 ml/kg/h being associated with lower mortality and
dialysate flow rates. As such, the results cannot be general- lower cardiovascular morbidity, presumably due in part to
ized to include increasing Kt/V while holding dialyzer size the deleterious effects of intradialytic hypotension [16, 36,
and flow rates the same, either by increased dialysis time or 47]. Despite this, the possibility of residual confounding
by increasing standardized Kt/V by increasing dialysis fre- limits any conclusions regarding mortality based on obser-
quency. During screening, patients had to be able to achieve vational data. Indeed, concern has been raised regarding
an eKt/V of at least 1.30 during a 4.5 hour session in order to frequent short daily in-center dialysis with an increased
enter the study. Due to the difficulty of achieving high Kt/V mortality seen in a propensity matched cohort study
in large patients, only 3% of participants were greater than (HR 1.6, 95% CI 1.1–2.3) [49].
100 kg, thus also limiting the generalizability to larger Between 2007 and 2017, a series of small, but important
patients. The study also excluded those with a residual urea randomized trials of more frequent or extended hours dial-
clearance of greater than 1.5 ml/min per 35 l urea volume ysis (the findings of which are summarized in Table 41.2)
and 67% of participants were anuric. Compared to the have been published. The first three trials focused on
standard-dose group, women in the high-dose group had a dialysis frequency and compared five or six times per
19% lower relative risk of mortality, but men had a 16% week to thrice weekly hemodialysis, with the Frequent
higher relative risk of mortality.1 Differences in body size Hemodialysis Network (FHN) Daily study utilizing short
and composition have been postulated to explain these find- dialysis sessions (1.5–2.75 hours) [50] and the Culleton
ings and no further randomized studies are available to et al. [51] and the FHN Nocturnal [52] studies utilizing ses-
clarify them. sions of at least 6 hours in length. In contrast, the ACTIVE
Dialysis trial compared total weekly dialysis hours of
24 hours (mostly delivered in three 8-hour sessions) to a
standard regimen of 18 hours per week (delivered in three
Increased Dialysis Frequency and Session
sessions) [53]. All studies were small (totaling 584 partici-
Length
pants) and lasted no longer than 12 months. All demon-
The “un-physiological” nature of thrice weekly intermit- strated a reduction in left ventricular mass, although this
tent hemodialysis has always been recognized and is ele- was significant only in the Culleton et al. and FHN Daily
gantly demonstrated by the clear peak in mortality at the studies. There were clear reductions in two important sur-
end of the long 2-day break in patients on standard thrice rogate endpoints: serum phosphate and systolic blood pres-
weekly dialysis routines [45]. Yet, the randomized trials of sure, with associated reductions in the need for phosphate
different urea clearance targets have been limited to thrice binders and antihypertensives. Quality of life measure-
weekly dialysis with session times reaching a maximum of ments were also stable or improved in the frequent or
4.5 hours (the mean value in the high dose HEMO arm was extended hours dialysis groups [50–52, 54].
4 hours, 9 minutes) [22, 42]. Long term follow-up data is available for the FHN trials
Observational evidence clearly favors longer sessions. and provides conflicting signals. After a median follow-up
Dialysis times 4.5 hours and Kt/V 1.3 are associated with of 3.6 years, the risk for mortality in the frequent in-center
lower mortality [46]. Saran et al. demonstrated that (after hemodialysis group of the FHN Daily trial was almost half
adjustment for confounders including Kt/V measured urea that in the conventional group (HR 0.54 [95% CI 0.31 to
clearance) each 30 minute increase in dialysis time was 0.93, P = 0.024]) [55]. In contrast, in long-term follow-up of
associated with a 7% relative risk reduction in all-cause the 87 participants of the FHN Nocturnal trial, an increased
mortality [47]. They also observed a synergistic association risk of mortality was observed in the frequent home dialy-
between Kt/V and dialysis time, where the mortality bene- sis group (HR 3.88 [95% CI 1.27–11.79, P = 0.01]) [56]. Both
fit of increased Kt/V was amplified at longer dialysis times, results, the latter in particular, should be interpreted with
although this interaction has not been observed in all stud- caution owing to the small number of events in each arm.
ies [46]. A meta-analysis of observational studies con- The potential harms of frequent dialysis could include an
cluded that, compared with conventional hemodialysis, increase in dialysis-related myocardial stunning [37] or
home nocturnal dialysis (hazard ratio [HR] 0.46, 95% CI cerebral ischemia (which is not simply predictable from
0.38–0.55), in-center nocturnal dialysis (HR 0.73, 95% blood pressure and is associated with cognitive dysfunc-
CI 0.60–0.90), and home short daily dialysis (HR 0.54, 95% tion) [57]. As such, the long-term impact of frequent dialy-
CI 0.31–0.95) all provide substantial mortality benefits [48]. sis remains unclear.
Table 41.2 Summary of findings: individual randomized trials of increased dialysis time or frequency.
Culleton et al. FHN Nocturnal FHN Daily ACTIVE Dialysis

N = 51 N = 87 N = 245 N = 200
6 m 12 m 12 m 12 m
6×/week
5-6×/week ≥6 h/session 5-6×/week ≥24 h/week
≥6 h/session (home) 1.5–2.75 h/session (3×/week, 8 h)
(home, nocturnal) vs. (in-center) (in-center, home)
vs. 3×/week vs. vs.
3×/week <5 h 3×/week ≤18 h/week Certainty of
spKt/V ≥ 1.2 eKt/V ≥ 1.1 eKt/V ≥ 1.1 (3×/week, 4–5 h) evidence
Outcomes (in-center) (home) (in-center) (in-center, home) (GRADE) Conclusion
Mortality – HR 3.88 (1.27, 11.79) HR 0.54 (0.31, 0.93) – Low Widely divergent estimates of the
●○○○ mortality impact of frequent nocturnal
and daily dialysis do not permit
definitive answer
Left ventricular −8.1 (−16.2, −0.1) −4.4 (−10.8, 2.1) −7.1 (−12.0, −2.2) −6.0 (−14.8, 2.7)2 Moderate More frequent or extended dialysis may
mass index (g/m2) ●●○○ reduce left ventricular mass index
Phosphate −0.48 (−0.81, −0.16) −0.45 (−0.68, −0.23) −0.18 (−0.29, −0.07) −0.25 (−0.35, −0.16) High Frequent or extended dialysis reduces
(mmol/l) ●●●● phosphate
Systolic blood −11 (−24, 2) −9.73 (−16.9, −2.5) −10.1 (−14.3, −6.0) −2.9 (−6.8, 1.1) Moderate Frequent dialysis lowers blood pressure,
pressure (mmHg) ●●○○ the impact of extended dialysis is not
clear and may be affected by changes in
antihypertensive agents
Parathyroid −8.9 vs. +1.64 −4.0 (−6.7, 0.4) 2.8 (−0.3, 6.8) −1.3 (−14.4, 11.9) High Frequent or extended dialysis is unlikely
hormone (pmol/l) ●●●● to change parathyroid hormone
Calcium (mmol/l) 0.025 (−0.075, 0.100) No change No change 0.05 (0.01, 0.09) High Frequent or extended dialysis does not
●●●● have an important effect on calcium
Physical quality 1.5 (−2.2, 5.2) 0.6 (−3.4, 4.7) 3.2 (1.0, 5.4) 2.3 (0.5, 4.1) Low Physical and mental quality of life may
of life (score ●○○○ improve with frequent daily or extended
range 0–100) dialysis; neither appear to be affected by
Mental quality of 2.5 (−3.0, 7.9) 3.7 ± 0.9 vs. 0.2 ± 1.0; 2.5 (0.4, 4.7) Low frequent nocturnal dialysis
life (score range P < 0.01 ●○○○
0–100)
c41.indd 28 09-12-2022 16:19:04

EQ-5D health 0.05 (−0.07, 0.17) 0.04 (−0.03, 0.11) Low Increased dialysis does not improve
utility ●○○○ health utility
Other quality of Improved effects of No change in Beck No change in Beck Kidney Disease –
life kidney disease (8.6 /100 depression inventory depression inventory Composite Score (3.48
[2.0, 15.2]) and burden of [1.44, 5.51]; P < 0.001)
kidney disease (9.4 /100
[1.3, 17.5]); no change in
symptoms/problems,
sleep
Medication use Reduction or cessation of Fewer antihypertensives Fewer antihypertensives Fewer phosphate High Frequent or extended hours dialysis
antihypertensives and (−0.97 vs. 0.26; (−0.87 vs. −0.23; binders (−0.83 [−1.61, ●●●● reduces the need for antihypertensives
(16/26 vs. 3/25; P < 0.001); fewer P < 0.001); lower −0.04]; P = 0.04); and phosphate binders
P < 0.001) phosphate phosphate binders (no equivalent dose of fewer
binders (19/26 vs. 3/25; binders 73% vs. 8%; phosphate binders antihypertensives
P < 0.001) more common P < 0.001) (−1.35 g/day [−2.50, (−0.29 [−0.53, −0.06];
−0.20]) P = 0.01)
Incident anuria Trend to increase (67% No change (53% vs. 48%; Low Loss of residual renal function may be
vs. 36%; NS) NS) ●○○○ more rapid with frequent dialysis
Adverse events No difference in vascular Less intradialytic No difference in No difference in Moderate An increase in vascular access events
access complications hypotension (3.1% vs. hospitalizations or access access interventions ●●○○ with frequent dialysis cannot be
9.5% of sessions; interventions or adverse events excluded
P < 0.001); trend toward
more vascular access
complications (HR 1.88
[0.97, 3.64])
CI, confidence interval. GRADE assessment of the certainty of the evidence [1]: High: This research provides a very good indication of the likely effect. The likelihood that the effect will be
substantially different* is low. Moderate: This research provides a good indication of the likely effect. The likelihood that the effect will be substantially different* is moderate. Low: This
research provides some indication of the likely effect. However, the likelihood that it will be substantially different* is high. Very low: This research does not provide a reliable indication of the
likely effect. The likelihood that the effect will be substantially different* is very high. *Substantially different = a large enough difference that it might affect decision-making.
c41.indd 29 09-12-2022 16:19:04

The FHN Daily trial identified a number of negative When these observations are coupled to the well-described
effects of frequent dialysis. A nonsignificant increase in the association between RKF and survival in hemodialysis
number of vascular access interventions (HR 1.35 [95% CI patients [27, 28], the argument for incremental exposure to
0.84–2.18, P = 0.22]) was noted, which was significant dialysis appears compelling.
when time to first access intervention was examined (HR However, the observational evidence (including both
1.71, 95%CI 1.08–2.73) [50]. A similarly significant increase prospective and retrospective studies) has proven conflict-
in the risk of vascular access failure or intervention was ing. While similar survival between twice weekly/incre-
seen in the FHN Nocturnal trial (HR 1.88 [95% CI 0.97– mental dialysis and standard thrice weekly has been
3.64, P = 0.06]) [52]. No significant differences in vascular demonstrated in multiple studies [60–63], residual con-
access complications were noted in the Culleton et al. or founding may explain many of these findings and an
ACTIVE trials [51, 53]. The impact of frequent dialysis on increase in mortality has also been observed in other stud-
residual renal function was examined in both the FHN ies. For example, a prospective study involving 685 partici-
Daily and Nocturnal trials with an increase in incident pants found a fourfold increase in mortality in adjusted
anuria (i.e. loss of residual renal function) seen in the analysis among patients with RKF undergoing twice
Nocturnal trial (relative risk [RR] 1.83, P = 0.06) but not in weekly dialysis compared to those undergoing three-times
the Daily trial (RR 1.11, P = 0.99) [30]. Caregiver burden weekly dialysis (HR 4.20 [95% CI 1.02–17.32, P = 0.04]) [64].
also emerged as a consideration in the FHN Nocturnal A larger retrospective study in the United States compared
trial, where a trend toward increased perceived caregiver 351 incremental dialysis patients to 8068 matched patients
burden (as measured by the Cousineau scale) was identi- (including matching by RKF) and found equivalent mor-
fied (6.1 [95% CI −0.8 to +13.1, P = 0.08]). No such trend tality in those with baseline RKF > 3.0 ml/min/1.73 m2 (HR
was evident in the FHN Daily trial [58]. 0.99, 95% CI 0.76–1.28) but increased mortality in those
with 3.0 ml/min/1.73 m2 (HR 1.61, 95% CI 1.07–2.44) [65].
These studies highlight the potential risks involved with
Incremental Dialysis
incremental dialysis, especially in patients with minimal
There is growing interest in a gradual approach to dialysis residual renal function (e.g. urea clearance <3 ml/
initiation. This is informed by the knowledge that RKF min/1.73m2, urine volume <600 ml/day) and emphasize
typically remains substantial when dialysis is initiated and that a well-structured program with clear patient under-
the obvious corollary that patients might therefore require standing is necessary for such a program to be imple-
less dialysis in the early months and years after commenc- mented. Future randomized controlled trials are required
ing dialysis. Moreover, observational studies suggest that to ascertain the benefits and safety of this approach.
twice-weekly dialysis is associated with preservation of
RKF [59]. The hypothesis that more dialysis accelerates
Summary of Findings
loss of RKF was strengthened by the results of the FHN
trials in which participants assigned to frequent dialysis The evidence base for dialysis dosing is limited, with no
suffered a greater decline in RKF and/or urine volume [30]. high-quality systematic reviews published that include all
Box 41.1 Current Guideline Positions

The most recent Kidney Disease Outcomes Quality be reduced in the presence of significant residual renal
Initiative (KDOQI) guidelines for Hemodialysis Adequacy: function, provided this is measured regularly (ungraded).
2015 Update [21] incorporate the long-term follow-up They suggest that patients be offered short frequent
data from the FHN studies and the findings of the ACTIVE dialysis (<3 hours, five to seven sessions per week) or long
Dialysis trial. duration dialysis (defined as 6–8 hours, three to six ses-
KDOQI guidelines recommend a per session spKt/V sions per week) after considering individual patient pref-
target of 1.4 with a minimum delivered Kt/V of 1.2. For erences and the potential benefits of these therapies. It is
patients on more or less frequent dialysis the recom- recommended that the potential risks be discussed prior
mended target stdKt/V is 2.3 with a delivered stdKt/V of to commencing such therapies. Increased dialysis should
2.1 (using an equation that considers ultrafiltration and be considered in the presence of difficulty controlling
residual renal function). A minimum time per session of interdialytic weight gain, blood pressure, volume over-
3 hours is recommended for those with minimal residual load, or metabolic parameters such as hyperkalemia,
renal function. They also suggest that dialysis dose may hyperphosphatemia, or metabolic acidosis (ungraded).
Reference 31
recent studies. Observational evidence suggests that longer Conclusion

sessions and slower rates of ultrafiltration are associated
with lower mortality. A single randomized trial has demon- Dialysis adequacy must be assessed on an individual basis
strated that targeting a higher urea clearance (Kt/V) does with patient needs being defined in line with patient goals
not improve mortality, but this trial does not address the and circumstances. We suggest that the KDOQI guidelines
question of longer hours or more frequent sessions (than form part of an adequacy assessment, but emphasize that
the standard 3.5–4 hours three times per week). A number they are insufficient alone. We suggest, based on observa-
of small studies demonstrate clear benefits of more fre- tional evidence, a minimum dialysis time of 4 hours for
quent or extended hours on the surrogate endpoints of patients dialyzing three times per week. This is likely to be
blood pressure and serum phosphate. They also suggest insufficient for many patients and increasing dialysis hours
that such regimes may reduce left ventricular hypertrophy. (or frequency) will typically be required to achieve patient
Importantly, these gains were achieved while maintaining needs. We recognize that important questions, such as the
or improving quality of life. However, conflicting results level of phosphate or blood pressure required to avoid
have been observed on long-term mortality. Concern has death or morbidity, remain unanswered. Further research
been raised about loss of residual renal function with fre- is required to define optimal surrogate endpoints (with ref-
quent dialysis and a possible increase in vascular access erence to clinical events and patient-centered outcomes) as
complications. No randomized trials are available to inform this will allow more informed goal-setting on the part of
the use of RKF in dialysis prescription. patients and clinicians.
Notes
1 This interaction was not significant after Bonferroni 3 Effect estimates are adjusted for baseline characteristics
correction for multiple testing. (chosen as primary analysis by study authors).
2 N = 95. 4 Median change in frequent vs. standard groups (P = 0.05).
References
Hultcrantz, M., Rind, D., Akl, E.A. et al. (2017). The

1 7 Jain, A.K., Blake, P., Cordy, P. et al. (2012). Global trends
GRADE Working Group clarifies the construct of certainty in rates of peritoneal dialysis. J. Am. Soc. Nephrol. 23 (3):
of evidence. J. Clin. Epidemiol. 87: 4–13. 533–544.
2 Scribner, B.H., Buri, R., Caner, J.E. et al. (1960). The 8 Rocco, M.V. (2015). Chronic hemodialysis therapy in the
treatment of chronic uremia by means of intermittent west. Kidney Dis. (Basel) 1 (3): 178–186.
hemodialysis: a preliminary report. Trans. Am. Soc. Artif. 9 Bieber, B., Qian, J., Anand, S. et al. (2014). Two-times
Intern. Organs 6: 114–122. weekly hemodialysis in China: frequency, associated
3 Maslow, A.H. (1943). A theory of human motivation. patient and treatment characteristics and quality of life in
Psychol. Rev. 50 (4): 370–396. the China Dialysis Outcomes and Practice Patterns study.
4 Liyanage, T., Ninomiya, T., Jha, V. et al. (2015). Worldwide Nephrol. Dial. Transplant. 29 (9): 1770–1777.
access to treatment for end-stage kidney disease: a 10 Hyodo, T., Hirawa, N., Hayashi, M. et al. (2017). Present
systematic review. Lancet 385 (9981): 1975–1982. status of renal replacement therapy at 2015 in Asian
5 Bello, A., Levin, A., Tonelli, M. et al. (2017). Global Kidney countries (Myanmar, Vietnam, Thailand, China, and
Health Atlas: A Report by the International Society of Japan). Ren Replace Ther 3 (11) https://doi.org/10.1186/
Nephrology on the Current State of Organization and s41100-016-0082-7.
Structures for Kidney Care Across the Globe. Brussels, 11 Prasad, N. and Jha, V. (2015). Hemodialysis in Asia.
Belgium: International Society of Nephrology. Kidney Dis. (Basel) 1 (3): 165–177.
6 Fresenius Medical Care (2013). Fresenius medical care 12 Daugirdas, J.T., Blake, P.G., and Ing, T.S. (2015).
annual report 2013: ESRD patients in 2013: a global Handbook of Dialysis, 5e. Wolters Kluwer.
perspective. In: (ed. Care FM). Bad Homburg: Fresenius 13 Ortega-Perez de Villar, L., Martinez-Olmos, F.J., Junque-
Medical Care. Jimenez, A. et al. (PLoS One, 2018). Test-retest reliability
and minimal detectable change scores for the short in dialysis patients; findings from a marginal structural
physical performance battery, one-legged standing test model. Nephrol. Dial. Transplant. 26 (9): 2978–2983.
and timed up and go test in patients undergoing 28 Obi, Y., Rhee, C.M., Mathew, A.T. et al. (2016). Residual
hemodialysis. 13 (8): e0201035. kidney function decline and mortality in incident
14 Spalding, E.M., Chamney, P.W., and Farrington, K. hemodialysis patients. J. Am. Soc. Nephrol. 27 (12):
(2002). Phosphate kinetics during hemodialysis: evidence 3758–3768.
for biphasic regulation. Kidney Int. 61 (2): 655–667. 29 Casino, F.G. and Basile, C. (2018). How to set the stage
15 Laursen, S.H., Vestergaard, P., and Hejlesen, O.K. (2018). for a full-fledged clinical trial testing ’incremental
Phosphate kinetic models in hemodialysis: a systematic haemodialysis’. Nephrol. Dial. Transplant. 33: 1103–1109.
review. Am. J. Kidney Dis. 71 (1): 75–90. 30 Daugirdas, J.T., Greene, T., Rocco, M.V. et al. (2013).
16 Flythe, J.E., Kimmel, S.E., and Brunelli, S.M. (2011). Effect of frequent hemodialysis on residual kidney
Rapid fluid removal during dialysis is associated with function. Kidney Int. 83 (5): 949–958.
cardiovascular morbidity and mortality. Kidney Int. 79 (2): 31 Daugirdas, J.T. (2016). Estimating weekly urine flow rate
250–257. and residual kidney urea clearance: a method to deal
17 Movilli, E., Gaggia, P., Zubani, R. et al. (2007). with interdialytic variability. Semin. Dial. 29 (6): 510–514.
Association between high ultrafiltration rates and 32 Shafi, T. and Levey, A.S. (2017). Measurement and
mortality in uraemic patients on regular haemodialysis. A estimation of residual kidney function in patients on
5-year prospective observational multicentre study. dialysis. Adv. Chronic Kidney Dis. 25 (1): 93–104.
Nephrol. Dial. Transplant. 22 (12): 3547–3552. 33 Mathew, A.T., Fishbane, S., Obi, Y. et al. (2016).
18 Alfaadhel, T.A., Soroka, S.D., Kiberd, B.A. et al. (2015). Preservation of residual kidney function in hemodialysis
Frailty and mortality in dialysis: evaluation of a clinical patients: reviving an old concept. Kidney Int. 90 (2):
frailty scale. Clin. J. Am. Soc. Nephrol. 10 (5): 832–840. 262–271.
19 Gotch, F.A. and Sargent, J.A. (1985). A mechanistic 34 Tentori, F., Zhang, J., Li, Y. et al. (2012). Longer dialysis
analysis of the National Cooperative Dialysis Study session length is associated with better intermediate
(NCDS). Kidney Int. 28 (3): 526–534. outcomes and survival among patients on in-center three
20 Daugirdas, J.T. (1993). Second generation logarithmic times per week hemodialysis: results from the Dialysis
estimates of single-pool variable volume Kt/V: an analysis Outcomes and Practice Patterns Study (DOPPS). Nephrol.
of error. J. Am. Soc. Nephrol. 4 (5): 1205–1213. Dial. Transplant. 27 (11): 4180–4188.
21 National Kidney Foundation (2015). KDOQI clinical 35 Held, P.J., Port, F.K., Wolfe, R.A. et al. (1996). The dose of
practice guideline for hemodialysis adequacy: 2015 hemodialysis and patient mortality. Kidney Int. 50 (2):
update. Am. J. Kidney Dis. 66 (5): 884–930. 550–556.
22 Eknoyan, G., Beck, G.J., Cheung, A.K. et al. (2002). Effect 36 Assimon, M.M., Wenger, J.B., Wang, L. et al. (2016).
of dialysis dose and membrane flux in maintenance Ultrafiltration rate and mortality in maintenance
hemodialysis. N. Engl. J. Med. 347 (25): 2010–2019. hemodialysis patients. Am. J. Kidney Dis. 68 (6):
23 Davenport, A. (2013). Differences in prescribed Kt/V and 911–922.
delivered haemodialysis dose--why obesity makes a 37 Burton, J.O., Jefferies, H.J., Selby, N.M. et al. (2009).
difference to survival for haemodialysis patients when Hemodialysis-induced cardiac injury: determinants and
using a ’one size fits all’ Kt/V target. Nephrol. Dial. associated outcomes. Clin. J. Am. Soc. Nephrol. 4 (5):
Transplant. 28 (Suppl 4): iv219–iv223. 914–920.
24 Racki, S., Zaputovic, L., Maleta, I. et al. (2005). 38 Palmer, S.C., Hayen, A., Macaskill, P. et al. (2011). Serum
Assessment of hemodialysis adequacy by ionic levels of phosphorus, parathyroid hormone, and calcium
dialysance: comparison to standard method of urea and risks of death and cardiovascular disease in
removal. Ren. Fail. 27 (5): 601–604. individuals with chronic kidney disease: a systematic
25 Grzegorzewska, A.E. and Banachowicz, W. (2006). review and meta-analysis. JAMA 305 (11): 1119–1127.
Comparisons of Kt/V evaluated using an online method 39 Hecking, M., Moissl, U., Genser, B. et al. (2018). Greater
and calculated from urea measurements in patients on fluid overload and lower interdialytic weight gain are
intermittent hemodialysis. Hemodial. Int. 10 (Suppl 2): independently associated with mortality in a large
S5–S9. international hemodialysis population. Nephrol. Dial.
26 Rhee, C.M., Unruh, M., Chen, J. et al. (2013). Infrequent Transplant. 33 (10): 1832–1842.
dialysis: a new paradigm for hemodialysis initiation. 40 Tabinor, M., Elphick, E., Dudson, M. et al. (2018).
Semin. Dial. 26 (6): 720–727. Bioimpedance-defined overhydration predicts survival in
27 van der Wal, W.M., Noordzij, M., Dekker, F.W. et al. (2011). end stage kidney failure (ESKF): systematic review and
Full loss of residual renal function causes higher mortality subgroup meta-analysis. Sci. Rep. 8 (1): 4441.
Reference 33
1 Han, Y.C., Tu, Y., Zhou, L.T. et al. (2018). Peridialysis BP

4 54 Unruh, M.L., Larive, B., Chertow, G.M. et al. (2013).
levels and risk of all-cause mortality: a dose-response Effects of 6-times-weekly versus 3-times-weekly
meta-analysis. J. Hum. Hypertens 33 (1): 41–49. hemodialysis on depressive symptoms and self-reported
42 Lowrie, E.G., Laird, N.M., Parker, T.F. et al. (1981). Effect mental health: Frequent Hemodialysis Network (FHN)
of the hemodialysis prescription of patient morbidity: Trials. Am. J. Kidney Dis. 61 (5): 748–758.
report from the National Cooperative Dialysis Study. N. 55 Chertow, G.M., Levin, N.W., Beck, G.J. et al. (2016).
Engl. J. Med. 305 (20): 1176–1181. Long-term effects of frequent in-center hemodialysis.
43 Shinzato, T., Nakai, S., Akiba, T. et al. (1997). Survival in J. Am. Soc. Nephrol. 27 (6): 1830–1836.
long-term haemodialysis patients: results from the annual 56 Rocco, M.V., Daugirdas, J.T., Greene, T. et al. (2015).
survey of the Japanese Society for Dialysis Therapy. Long-term effects of frequent nocturnal hemodialysis on
Nephrol. Dial. Transplant. 12 (5): 884–888. mortality: The Frequent hemodialysis Network (FHN)
44 Port, F.K., Ashby, V.B., Dhingra, R.K. et al. (2002). nocturnal trial. Am. J. Kidney Dis. 66 (3): 459–468.
Dialysis dose and body mass index are strongly associated 57 MacEwen, C., Sutherland, S., Daly, J. et al. (2017).
with survival in hemodialysis patients. J. Am. Soc. Relationship between hypotension and cerebral ischemia
Nephrol. 13 (4): 1061–1066. during hemodialysis. J. Am. Soc. Nephrol. 28 (8):
45 Foley, R.N., Gilbertson, D.T., Murray, T. et al. (2011). Long 2511–2520.
interdialytic interval and mortality among patients 58 Suri, R.S., Larive, B., Garg, A.X. et al. (2011). Burden on
receiving hemodialysis. N. Engl. J. Med. 365 (12): caregivers as perceived by hemodialysis patients in the
1099–1107. Frequent Hemodialysis Network (FHN) trials. Nephrol.
46 Marshall, M.R., Byrne, B.G., Kerr, P.G. et al. (2006). Dial. Transplant. 26 (7): 2316–2322.
Associations of hemodialysis dose and session length 59 Mathew, A.T., Obi, Y., Rhee, C.M. et al. (2018).
with mortality risk in Australian and New Zealand Incremental dialysis for preserving residual kidney
patients. Kidney Int. 69 (7): 1229–1236. function-does one size fit all when initiating dialysis?
47 Saran, R., Bragg-Gresham, J.L., Levin, N.W. et al. (2006). Semin. Dial. 31 (4): 343–352.
Longer treatment time and slower ultrafiltration in 60 Hanson, J.A., Hulbert-Shearon, T.E., Ojo, A.O. et al.
hemodialysis: associations with reduced mortality in the (1999). Prescription of twice-weekly hemodialysis in the
DOPPS. Kidney Int. 69 (7): 1222–1228. USA. Am. J. Nephrol. 19 (6): 625–633.
48 Mathew, A., McLeggon, J.A., Mehta, N. et al. (2018). 61 Lin, X., Yan, Y., Ni, Z. et al. (2012). Clinical outcome of
Mortality and hospitalizations in intensive dialysis: a twice-weekly hemodialysis patients in shanghai. Blood
systematic review and meta-analysis. Can. J. Kidney Purif. 33 (1–3): 66–72.
Health Dis. 5: 2054358117749531. 62 Elamin, S. and Abu-Aisha, H. (2012). Reaching target
49 Suri, R.S., Lindsay, R.M., Bieber, B.A. et al. (2013). A hemoglobin level and having a functioning arteriovenous
multinational cohort study of in-center daily hemodialysis fistula significantly improve one year survival in twice
and patient survival. Kidney Int. 83 (2): 300–307. weekly hemodialysis. Arab. J. Nephrol. Transplant. 5 (2):
50 Chertow, G.M., Levin, N.W., Beck, G.J. et al. (2010). 81–86.
In-center hemodialysis six times per week versus three 63 Mathew, A., Obi, Y., Rhee, C.M. et al. (2016). Treatment
times per week. N. Engl. J. Med. 363 (24): 2287–2300. frequency and mortality among incident hemodialysis
51 Culleton, B.F., Walsh, M., Klarenbach, S.W. et al. (2007). patients in the United States comparing incremental with
Effect of frequent nocturnal hemodialysis vs conventional standard and more frequent dialysis. Kidney Int. 90 (5):
hemodialysis on left ventricular mass and quality of life: 1071–1079.
a randomized controlled trial. JAMA 298 (11): 1291–1299. 64 Hwang, H.S., Hong, Y.A., Yoon, H.E. et al. (2016).
52 Rocco, M.V., Lockridge, R.S. Jr., Beck, G.J. et al. (2011). Comparison of clinical outcome between twice-weekly
The effects of frequent nocturnal home hemodialysis: the and thrice-weekly hemodialysis in patients with residual
Frequent Hemodialysis Network Nocturnal Trial. Kidney kidney function. Medicine (Baltimore) 95 (7): e2767.
Int. 80 (10): 1080–1091. 65 Obi, Y., Streja, E., Rhee, C.M. et al. (2016). Incremental
53 Jardine, M.J., Zuo, L., Gray, N.A. et al. (2017). A trial of hemodialysis, residual kidney function, and mortality
extending Hemodialysis hours and quality of life. J. Am. risk in incident dialysis patients: A Cohort Study. Am. J.
Soc. Nephrol. 28 (6): 1898–1911. Kidney Dis. 68 (2): 256–265.
34
42
General Management of the Hemodialysis Patient

David Collister1, Lonnie Pyne1, and Michael Walsh1, 2, 3
1
Department of Medicine, Division of Nephrology, McMaster University, Hamilton, Canada
2
Department of Health Research Methods, Evaluation and Impact, McMaster University, Hamilton, Canada
3
Population Health Research Institute, Hamilton Health Sciences/McMaster University, Hamilton, Canada
Introduction more prevalent on the day of the long interdialytic interval

compared to short intervals due to dietary intake and
The care of a hemodialysis (HD) patient is typically pro- impaired renal clearance despite enhanced gastrointestinal
vided in a multidisciplinary setting that includes nurses, and salivary clearances in HD patients [1]. Higher dietary
dieticians, pharmacists, social workers, and nephrologists. potassium intake is associated with increased mortality in
It traditionally focuses on the adequacy of small solute dialysis patients even when adjusted for its association with
clearance (Chapter 41), vascular access (Chapter 44), increased energy and protein intake as well as other nutri-
anemia (Chapter 33), and mineral bone disorders tional parameters. However, higher vegetable intake, often
(Chapter 36). This chapter is dedicated to other important considered the primary source of potassium, was associated
issues related to the optimal care of hemodialysis patients, with lower mortality in a more recent, large study [2, 3].
including potassium, acidosis, blood pressure control Other contributors to hyperkalemia in HD include acidosis
(hypertension [HTN] and intradialytic hypotension [IDH]), as well as several medications including renin-angiotensin-
nutrition, and cardiovascular (CV) disease. It summarizes aldosterone system (RAAS) inhibitors, beta-blockers, hepa-
the available evidence relevant to these general aspects of rin, and nonsteroidal anti-inflammatory drugs.
the management of the HD patient. In chronic kidney disease (CKD), there is a U-shaped
association between serum potassium and mortality [4]. In
HD, hyperkalemia, defined as a serum potassium of
Potassium Metabolism 6–6.5 mmol/l compared to normal values, is associated with
an increase in hospitalization and mortality regardless of the
The general principles of the assessment and treatment of day of the week, while a serum potassium of 5.5–6 mmol/l is
disorders of potassium metabolism (hypokalemia and associated with adverse outcomes following the long dialytic
hyperkalemia) are detailed in Chapter 60. The specific interval [5]. In the analyses of the Dialysis Outcomes and
problem of potassium as it relates to patients on HD is dis- Practice Patterns Study (DOPPS) data, sudden death due to
cussed in this chapter. arrhythmia, cardiac arrest or hyperkalemia was associated
with lower dialysate potassium, including both <1.5 and
2–2.5 meq/l compared to >3 meq/l [6], suggesting that large
Prevalence and Outcomes of Hyperkalemia
intradialytic potassium shifts may be harmful from a CV
Hyperkalemia is common among HD patients. In a pro- perspective. However, a dialysate potassium of 1 meq/l in
spective observational study from France, the proportion of individuals with predialysis potassium >5.5 mmol/l was pro-
patients with a serum potassium >5 mmol/l, >5.5 mmol/l, tective against mortality in an adjusted Cox proportional
or >6 mmol/l at any time was 73.8%, 57.9%, and 34.5%, hazards model [7]. The heterogeneity of results across large
respectively. Not surprisingly, hyperkalemia is 2.0–2.4 times observational studies challenges any determination of the
Potassium Metabolis 35
Table 42.1 General management of the hemodialysis patient.
Strength of
Management of We suggest adherence to a low potassium diet, ensuring adequate Very low Weak
hyperkalaemia dialysis prescription, prescribing a dialysate potassium of
1–4 mmol/l depending on predialysis potassium levels, and
correcting acidosis
Management of We suggest using a dialysate bicarbonate of 22–24 mmol/l and Very low (bicarbonate Weak (bicarbonate
metabolic using bicarbonate rather than acetate as the form of base delivery dialysate concentration) concentration) and
acidosis and moderate (use of weak (bicarbonate
bicarbonate vs. acetate) vs acetate)
Avoidance of We suggest that increasing dialysis session time may reduce the Moderate (increasing Weak
intradialytic risk of intradialytic hypotension. In addition, cooled dialysate, session time)
hypotension prescription of midodrine, sodium profiling and the use of Low (cooled dialysate,
biofeedback may also be used. sodium profiling, use of
biofeedback)
Prescription of
midodrine (very low)
Hypertension We suggest using self-home recorded BP or ambulatory BP Low Weak
measurement monitoring to guide treatment rather than measures pre or post
and targets dialysis BP where possible, but it is difficult to recommend target
BP in this population based on current evidence
Hypertension We suggest reducing target weight and using antihypertensive Low Weak
treatment agents, and that the selection of antihypertensive agents should
be guided by other comorbidities and tolerance
Lipid-lowering We recommend not initiating lipid-lowering therapy with a statin High Strong
agents or ezetimibe in patients receiving HD
Nutrition We suggest assessing nutritional status by using a measure such Low Weak
as subjective global assessment and managing malnutrition by
treating the underlying cause, ensuring adequate dialysis and
increasing nutritional intake. We suggest generally using oral
nutrition with or without supplements, but parenteral or
intradialytic therapies may be considered.
degree to which hyperkalemia itself, rather than the numer- and how urgently to institute treatment is largely guided by
ous other behavioral and health issues correlated with the degree of hyperkalemia and whether there are associ-
hyperkalemia, is causative of increased mortality. ated symptoms or signs (e.g. muscle weakness, paralysis,
arrhythmias). It should be emphasized that although
arrhythmias are considered the primary mechanism by
Management of Hyperkalemia
which hyperkalemia causes death, the typical electrocar-
There is very little evidence guiding when and how to treat diogram (ECG) changes of hyperkalemia do not correlate
hyperkalemia to reduce the risk of death and the strength strongly with serum potassium concentration nor do ECG
of current recommendations is weak (Table 42.1). findings associate strongly with the risk of death [8–10].
Randomized controlled trials (RCTs) testing treatment A 2005 systematic review for emergency interventions
strategies vary in terms of how they define hyperkalemia for hyperkalemia summarized the treatment for patients
and rarely measure the degree to which treatments reduce with hyperkalemia in general. For HD patients, the major
the risk of mortality or other patient-important outcomes. difference is the availability and timing of hemodialysis. In
Rather, this body of literature examines the degree and symptomatic patients, maneuvers to stabilize the myocar-
rapidity with which serum potassium is reduced. Whether dium and shift serum potassium intracellularly should be
36 General Management of the Hemodialysis Patient
Table 42.2 Summary of findings: midodrine, cooled dialysate, sodium profiling, and biofeedback approaches to intradialytic
hypotension (IDH).
Midodrine vs. placebo [43]
Absolute number
(95% CI) of
No. of patients affected Quality of
participants per 1000 treated Relative effect the evidence
Outcomes (studies) patients for 1 year (95% CI) (GRADE) Conclusion
Intradialytic hypotension Not Not estimable

estimated
Nadir intradialytic 79(6) 13.3 mmHg higher Very low It is uncertain whether midodrine
systolic BP (95% CI 8.6–18.0) ●○○○ improves intradialytic systolic BP
in hemodialysis
Nadir intradialytic 79(6) 5.9 mmHg higher Very low It is uncertain whether midodrine
diastolic BP (95% CI 2.7–9.1) ●○○○ improves intradialytic diastolic BP
in hemodialysis
Post dialysis systolic BP 101(8) 12.4 mmHg higher Very low It is uncertain whether midodrine
(95% CI 7.5–17.7) ●○○○ improves post dialysis systolic BP
in hemodialysis
Post-dialysis diastolic 101(8) 7.3 mmHg higher Very low It is uncertain whether midodrine
BP (95% CI 3.7–10.9) ●○○○ improves post dialysis diastolic BP
in hemodialysis
Symptoms of IDH Not Not estimable: Very low It is uncertain whether midodrine
estimated: generally, report ●○○○ improves symptoms related to IDH
(six studies) improvement in
symptoms
Adverse events and Not Not estimable: no Very low It is uncertain whether midodrine
safety estimated serious adverse ●○○○ is safe in patients receiving
events reported hemodialysis
Cooled dialysate vs. standard temperature dialysate [54]
Absolute number (95% CI) of Relative Quality of the

No. of participants patients affected per 1000 effect evidence
Outcomes (studies) treated patients for 1 year (95% CI) (GRADE) Conclusion
Intradialytic 120 (11); 552 HD Not estimated 0.32 (0.18 to Low It is unclear whether or not using
hypotension sessions 0.56) ●●○○ a lower dialysate temperature
will reduce the frequency of
intradialytic hypotension
BP changes (10)
during and after
dialysis
Dialysis adequacy (seven studies) Not estimated Kt/V 0.-5 Low It is unclear whether cooled
(−0.09 to ●●○○ dialysate reduces small solute
−0.01) clearance compared to standard
temperature dialysate
Symptoms of 2547 dialysis Not estimated 2.95 (0.88 to Low Cooled dialysate may increase
discomfort sessions (nine 9.82) ●●○○ symptoms compared to standard
trials) temperature dialysate
Potassium Metabolis 37
Cooled dialysate vs. standard temperature dialysate [54]
Absolute number (95%

No. of CI) of patients affected Quality of the
participants per 1000 treated Relative effect evidence
Outcomes (studies) patients for 1 year (95% CI) (GRADE) Conclusion
Sodium profiling vs. standard dialysis [56]

Outcomes No. of Absolute number (95% Relative effect (95% CI) Quality of the Conclusion
participants CI) of patients affected evidence
(studies) per 1000 treated (GRADE)
patients for 1 year
Intradialytic 3600 dialysis Not estimated Pooled OR for all Very low It is uncertain whether or not
Hypotension sessions (13) profiles ●○○○ using sodium profiling
(IDH) 0.71 (0.60–0.85) compared to standard dialysis
Benefit varied with reduces the risk of IDH
profile; benefit only
evidence with stepwise
profiling
Hemodialysis with biofeedback vs. standard dialysis [61]

Outcomes No. of Absolute number (95% Relative effect Quality of the Conclusion
participants CI) of patients affected (95% CI) evidence
(studies) per 1000 treated patients (GRADE)
for 1 year
Intradialytic 266 (6) 257 (from 92 fewer to 0.61 (0.44–0.86) Low Biofeedback may reduce the
hypotension 370 fewer episodes) ●●○○ frequency of IDH compared
(IDH) with standard hemodialysis
Predialysis 266 (6) Not estimable 3 mmHg higher Very low Biofeedback makes no
systolic BP (SBP) (2 lower to 7 ●○○○ difference to pre dialysis SBP
higher) compared with standard
hemodialysis
Post-dialysis 266 (6) Not estimable 7 mmHg higher Very low Biofeedback may result in
systolic BP (5 to 19 mmHg ●○○○ higher post dialysis DBP
(DBP) higher) compared with standard
hemodialysis
Quality of life Uncertain (3) Not estimable No difference Very low It is uncertain whether
outcomes seen, not ●○○○ biofeedback improves quality of
quantified life compared with standard
hemodialysis
Mortality 104(2) Not estimated 0.37 (0.07–2.01) Very low It is uncertain whether
●○○○ biofeedback improves mortality
compared with standard
hemodialysis
considered while waiting to initiate hemodialysis. 3a trial has shown SZC to be safe, well tolerated, and effec-
Additionally, the use of binding resins to enhance gastroin- tive in controlling serum potassium compared to placebo
testinal elimination of potassium may be considered, in the hemodialysis cohort in addition to usual hemodialy-
although the risks and benefits are poorly delineated, par- sis. It is too early to propose how these new, expensive
ticularly in dialysis patients. agents should be used in the hemodialysis cohort [11].
New potassium binding resins are being investigated in
addition to the widely available sodium polystyrene
Dialytic Potassium Removal
sulfonate. These include patiromer and sodium zirconium
cyclosilicate (SZC). Although trials of the new agents have Potassium removal during dialysis occurs by diffusion and
primarily involved nondialysis participants, a recent phase convection; for diffusive therapies it is proportional to the
gradient between serum and dialysate potassium concentra- ●● hypoalbuminemia [20]

tions. Once HD is initiated, serum potassium decreases ●● impaired physical function [21]
approximately 1 mmol/l during the first hour of treatment ●● thyroid and growth hormone abnormalities [22]
and by a further 1 mmol/l over the next 2 hours and typically ●● insulin resistance [23]
removes more than 100 mmol per 4-hour session but varies ●● hypertriglyceridemia [24]
depending on the HD prescription [12]. Dialysate potassium ●● reduced respiratory reserve
concentrations typically range from 1 to 4 mmol/l based on ●● increased mortality [25, 26].
the predialysis serum potassium values. The safety of using
very low dialysate potassium concentrations, particularly
potassium-free dialysate, in hemodialysis is questioned, Base Delivery in Dialysis
although there is little evidence that the risk of serious
The preferred form of base delivery to dialysis patients in
hypokalemia is substantially higher with very low potassium
the dialysate is in the form of bicarbonate (HCO3) and
dialysate. High bicarbonate dialysate may enhance the likeli-
potentially other anions, rather than acetate, as acetate is
hood of restoring normokalemia by reducing acidosis and
associated with hemodynamic instability [27] (Table 42.1).
promoting an extracellular shift of potassium whereas high
Observational studies suggest high bicarbonate dialysate
glucose dialysate can impair potassium removal due to insu-
increases predialysis serum bicarbonate, hyperparathy-
lin secretion and intracellular shifting [13].
roidism, serum albumin, and triceps skin-fold thick-
For HD patients with recurrent hyperkalemia, adherence
ness [28, 29]. The International Society of Renal Nutrition
to dietary potassium restriction is the first step, and the dial-
and Metabolism recommends targeting a predialysis HCO3
ysis prescription should be reviewed paying particular atten-
of 22–24 mmol/l to avoid protein energy wasting and the
tion to adequacy, recirculation, dialysate potassium, and
Kidney Disease Outcome Quality Initiative (KDOQI) rec-
dialysate bicarbonate and performing a careful review and
ommends maintaining a serum CO2 > 22 meq/l in HD
modification if possible of any contributing medications.
patients, but there are no RCTs to support alterations in
dialysate or that adding oral bicarbonate improves patient-
Management of Metabolic Acidosis important outcomes [30, 31] (Table 42.1).
Metabolic acidosis is the most common acid-base abnor-

mality in patients with CKD and in patients on dialysis but Intradialytic Hypotension
this abnormality may coexist with other acid-base disor-
ders, including respiratory disorders [14]. Acid-base bal- Introduction
ance requires bicarbonate (HCO3) reclamation in the IDH is a reduction in blood pressure during a dialysis
proximal convoluted tubule and the renal excretion of non- treatment. The variety of definitions of IDH has resulted
volatile acids (approximately 1 meq/kg of net acid produc- in broad estimates of frequency ranging between 10%
tion per day from dietary protein metabolism [15]) in the and 55% of dialysis treatments [34, 36]. The potential
form of titratable acids (TA) and ammonium (NH4). These importance of even IDH is due to its association with
mechanisms are dysfunctional in advanced CKD, leading to poor outcomes, particularly cardiovascular morbidity
both a nonanion gap metabolic acidosis and an anion gap and mortality. Presumably this is due to repeated
metabolic acidosis due the accumulation of phosphate, sul- ischemic insults to vulnerable vascular beds. However,
fate, urate, and hippurate. Metabolic acidosis increases as the association between IDH and clinical outcomes may
dietary protein intake increases acid generation, but is also primarily be due to residual confounding rather than a
affected by interdialytic weight gain (IDWG), dialysis sched- causal relationship.
ule, dialysate bicarbonate concentration, and non-HCO3 Proposed causes of IDH include ultrafiltration (UF) rates
anion concentrations and adequacy. that exceed intravascular refilling, an inappropriate target
weight loss during HD treatment, intravascular osmolality
Consequences of Acidosis reduction, autonomic dysfunction, and decreased cardiac
reserve. Risk factors for IDH include dialysis vintage,
Metabolic acidosis in a hemodialysis patient may be associ-
female gender, lower predialysis blood pressure (BP), and
ated with several consequences including:
hypoalbuminemia [32–34]. Studies of interventions to
●● increased bone resorption, reduced bone mineral den- prevent IDH typically focus on changes in blood pressure
sity, and hyperparathyroidism [16, 17] and have insufficient power and duration to assess whether
●● increased muscle protein catabolism and protein degra- preventing IDH reduces patient-important outcomes such
dation [18, 19] as cardiovascular events or mortality.
Intradialytic Hypotension 39
Definition midodrine with cooled dialysate showed no effect compared

to cooled dialysate alone [44, 45] (Table 42.2).
There is no consensus definition of IDH although a number
Dialysis-based interventions include longer HD session
of definitions are used in the literature [35]. These include
time, cooled dialysate, changes in dialysate sodium, and
definitions based on nadir BP, drop in BP from pre-HD
biofeedback dialysis.
level, symptoms associated with a drop in BP, a drop in BP
Shorter weekly HD time, increased IDWG, and conse-
prompting intervention, and combinations of the above.
quently higher ultrafiltration rates consistently show an
Systolic BP (SBP) nadir less than 90 mmHg in at least 30%
association with poor outcomes, including cardiovascular
of sessions has the most evidence showing a consistent
morbidity and mortality [46–48]. Higher UF volume is
association with mortality. There appeared to be a dose–
an independent risk factor for myocardial stunning [38].
response relationship, with more frequent IDH increasing
More frequent dialysis is associated with improved hemo-
risk for death. The addition of symptoms or intervention
dynamic tolerability and less dialysis-induced myocardial
criteria did not increase the strength of this association. In
stunning [42]. Both frequent hemodialysis network tri-
the subset of patients with predialysis BP 160 mmHg,
als (frequent in-center HD and frequent nocturnal HD)
nadir BP less than 100 mmHg was strongly associated with
demonstrated a decrease in the rate of hypotensive events
mortality [36].
requiring intervention in the frequent hemodialysis groups
as compared to routine HD [49, 50]. The extended follow-up
Prognosis/Pathogenesis of the in-center HD trial showed decreased mortality in the
more frequent HD group [51]. However, the extended fol-
IDH is associated with poor outcomes, particularly cardio- low-up of the nocturnal HD trial showed increased mortal-
vascular morbidity and mortality [32, 36]. Repeated intra- ity in the more frequent HD group [52]. Whether increased
dialytic ischemic insults to susceptible vascular beds dialysis frequency reduces mortality and whether any ben-
induced by IDH are felt to be the mechanism by which efits are due to reducing IDH is uncertain.
IDH may cause cardiovascular events [37, 38]. Intradialytic Cooled dialysate compared to standard temperatures has
myocardial stunning has been one marker of hemodynamic been shown to decrease IDH, LV mass index, and myocardial
stress that has been studied. Higher ultrafiltration volumes stunning and protect against brain white matter changes [37,
and greater decreases in intradialytic systolic blood 39, 40, 53]. The reduction in IDH is mediated primarily
pressure are independent risk factors for the development through increased peripheral vascular resistance [40, 53]. A
of intradialytic myocardial stunning (Burton 2009). After recent systematic review and meta-analysis of randomized
1 year of follow-up patients with myocardial stunning at trials on the topic of reduced temperature dialysate demon-
baseline had a greater decrease in left ventricular ejection strated a decrease in the rate of IDH by 70% (Table 42.2).
fraction and increased mortality [38]. Several interventions There was a trend toward increased symptoms of discomfort
that lead to improvement in intradialytic hemodynamics in the cooled dialysate group that was not statistically signifi-
have shown decreases in surrogate outcomes such as left cant. The trials were limited by short-term follow-up, lack of
ventricular mass index, intradialytic myocardial stunning, data on cardiovascular events and mortality, and low to very
and brain white matter changes [37, 39–42]. low quality of evidence [54]. A cluster-RCT of cooled vs.
standard dialysate temperature and its effect on mortality
Management and cardiovascular events will help to answer these questions
(Major Cardiovascular and Other Patient-Important
Several potential interventions exist to prevent IDH, Outcomes With Personalized dialYsate TEMPerature, MY
including pharmacologic and dialysis-based interventions. TEMP, http://ClinicalTrials.gov Identifier: NCT0262836).
Pharmacologic interventions are not a well-studied area in Dialysate sodium concentrations affect the hemody-
the prevention of IDH. The most studied pharmacologic namic tolerability of HD. Increased dialysate sodium levels
agent in the prevention of IDH is midodrine. Midodrine is a are felt to protect against a drop in serum osmolality and
prodrug of desglymidodrine, an alpha-1 adrenergic receptor resultant fluid shifts out of the vascular compartment. The
agonist. Midodrine is one of the most commonly used phar- desire to improve the tolerability of HD by increasing
macologic agents to prevent IDH. A systematic review con- dialysate sodium is tempered by the concerns over a
ducted in 2004 found that midodrine as compared to control positive sodium balance. The net positive sodium balance
resulted in significant improvements in post-HD and nadir is thought to lead to increased thirst and IDWG, requiring
BP [43]. The studies were limited by small sample size, short increased UF rates and potentially increased IDH as a
duration, and high risk of bias. There were no RCTs of consequence. Data from the DOPPS showed an association
midodrine for IDH prevention. The two studies that assessed between higher dialysate sodium and higher IDWG [55].
The same analysis of DOPPS data showed an inverse volume above an individualized critical value associated
relationship between dialysate sodium and hospitalization with the development of IDH.
as well as mortality. Studies on the effect of lowering Trials of biofeedback compared to dialysis without
dialysate sodium on IDH have had mixed results. Some biofeedback demonstrated less intradialytic myocardial
have shown no effect and others an increase in the rate of stunning and decreased rates of IDH [58–60]. A meta-
IDH when dialysate sodium has been lowered (Hussein analysis conducted in 2013 found biofeedback dialysis was
2017). Thus far, there are no RCTs demonstrating an effect associated with a reduced rate of IDH as compared to
of dialysate sodium concentration on outcomes such as standard HD. The majority of included studies (6/8) used
mortality, cardiovascular mortality, heart failure or other biofeedback dialysis with alterations in both UF rate and
patient-important outcomes. A large cluster-RCT is being dialysate conductivity to maintain hemodynamic stability.
conducted to establish whether dialysate sodium of The studies were limited by small sample size, short
137 mmol/l reduces important outcomes compared to a duration of follow-up, and low quality of evidence due to
dialysate sodium of 140 mmol/l (Randomized Evaluation risk of bias from lack of blinding and potential for
of Sodium Dialysate Levels on Vascular Events, RESOLVE, publication bias [61]. There is no high-quality evidence
http://ClinicalTrials.gov Identifier: NCT02823821). that biofeedback improves patient-important outcomes.
Sodium profiling, where dialysate sodium concentration Trials of blood volume guided biofeedback dialysis
is adjusted throughout the dialysis treatment, may decrease where only UF rate is adjusted have not shown similar
the risk of IDH. The dialysate sodium concentration is benefits. An early trial of blood volume guided HD with
initially elevated, frequently to the supraphysiologic range, operator adjustments of UF rate showed an increased rate
and is reduced based on any one of several possible profiles. of hospitalization and mortality in the blood volume
A meta-analysis compared the effects of various sodium guided HD group and no decrease in the rate of IDH [62].
profiles on IDH [56]. A stepwise decrease in dialysate A recent crossover trial of biofeedback dialysis with UF
sodium was associated with a decrease in IDH. Linear rate adjustments alone in IDH prone patients did not find a
profiling was not associated with a decrease in the incidence significant decrease in the rate of IDH as compared to
of IDH. Many of the included studies were limited by standard of care [63].
unclear allocation concealment and blinding [56].
Sodium profiles may result in a positive, negative, or
neutral net sodium balance. There are concerns about net Summary
sodium gain and its associated consequences with sodium IDH is a common issue in chronic hemodialysis patients. It
profiling. A small crossover study of IDH prone patients is associated with poor outcomes, including increased
evaluated sodium balance positive or neutral profiles cardiovascular morbidity and mortality. While many
with or without simultaneous UF profiling [57]. Positive interventions have been proposed to prevent IDH few have
sodium balance profiles with or without UF profiling and a solid evidence base. The literature is complicated by the
neutral sodium balance profiles with UF profiling reduced fact that there have been a number of different definitions
the incidence of intradialytic discomfort, early session of IDH in use. There is a theoretical framework and
termination, UF failure, and inadequate dose (Kt/V) as biological plausibility potentially causally linking IDH to
compared to control and neutral sodium profiles without cardiovascular morbidity and mortality. However, critical
UF profiling (Table 42.2). Positive sodium balance sodium evidence demonstrating that decreasing the incidence of
profiling was associated with increased IDWG and UF IDH improves patient important outcomes such as cardio-
requirements. Sodium balance neutral sodium profiling vascular events and mortality is lacking (Table 42.1).
combined with UF profiling was more likely to achieve
target weight than positive sodium balance profiles.
However, only positive sodium balance profiles signifi-
Hypertension
cantly increased the nadir of systolic BP compared to
control.
Introduction
Biofeedback dialysis is an additional dialysis-based inter-
vention that may involve adjustments in dialysate sodium Blood pressures that meet the definition of hypertension
and/or UF to reduce the incidence of IDH (Table 42.2). The for nondialysis patients are common in HD patients.
most commonly used form of biofeedback dialysis is blood However, whether similar definitions of hypertension
volume guided biofeedback dialysis. This method uses are relevant to the hemodialysis patient population is
measures of relative blood volume to adjust UF rate and/or uncertain and studies use heterogeneous definitions in
dialysate conductivity/sodium to maintain relative blood hemodialysis. Differences arise between studies based
Hypertensio 41
on definitions that vary in terms of blood pressure Blood Pressure Targets

thresholds, timing of measurements (predialysis, during
The optimal BP target is uncertain in HD, with no large
dialysis, post-dialysis, or on nondialysis days) and meth-
RCTs evaluating intensive versus liberal BP targets in
ods of measurement (e.g. at dialysis, home monitoring,
contrast to CKD and the general population. A pilot trial of
ambulatory monitoring). For example, in study of 355
dialysis patients with a predialysis SBP > 155 mmHg
HD patients using thresholds of 140/90 for midweek,
randomized participants to a target predialysis SBP 110–
immediately predialysis BP, and 135/85 for 44-hour
140 mmHg (n = 62) or 155–165 mmHg (n = 64) for 1 year.
ambulatory blood pressure monitoring (ABPM), 35% or
Dry weight probing, beta-blockers, and inhibitors of the
47% had HTN which was associated with a hazard ratios
RAAS were allowed [73]. SBP between the two groups
for mortality of 1.87 (95% confidence interval [CI]) [64].
differed by 12.9 mmHg and there was an excess of
KDOQI guidelines define HTN in HD patients as
hospitalizations, vascular access thrombosis, and intra-
predialysis BP >140/90 mmHg or postdialysis BP
dialytic hypotension in the lower target group.
>130/80 mmHg [65]. European Renal Association–
European Dialysis and Transplant Association (ERA-
EDTA) guidelines use home BP >135/85 mmHg, Blood Pressuring Lowering Therapies
ABPM>130/80, and midweek office BP >140/90 mmHg
with no recommendation for blood pressure measured Nonpharmacologic therapies to lower BP include gradu-
at dialysis treatments [66]. ally reducing the dry weight to achieve euvolemia (e.g. no
symptoms of signs of hypovolemia or hypervolemia) and
sodium restriction to reduce interdialytic weight gain.
Pathophysiology Additional hemodialysis treatment time or sessions may be
HTN often reflects inadequate volume control in HD necessary to safely achieve euvolemia. Reducing the
patients. The removal of excess sodium by reducing dry dialysate sodium may also reduce IDWG, volume status,
weight can result in normalization of BP in many cases and BP control [74]. In the DRIP trial, a RCT comparing
with a lag phenomenon supported by some [67] but not all ultrafiltration (weight loss of 0.1 kg/10 kg per dialysis until
studies [68]. BP is related to many factors, including signs or symptoms of hypotension) (n = 100) to physician
volume status, which increases postdialysis linearly but visits only (n = 50) without any changes in hypertensive
with fluctuations [69], but also sympathetic nervous system therapy in HD patients with a mean interdialytic ABPM of
activity, the RAAS, arteriosclerosis, and drugs (e.g. 135/75 and a postdialysis weight reduction of 0.9 kg at
erythropoietin-stimulating agents). 4 weeks resulted in SBP −6.9 mmHg (95% CI −12.4 to −1.3,
P = 0.016) and DBP −3.3 mmHg (95% CI −12.2 to 1.0,
P = 0.048) with similar changes at 8 weeks, which does not
Measurement and Outcomes support the lag phenomenon [68].
Options for monitoring BP include predialysis and postdi- The treatment of HTN with antihypertensive agents to
alysis BP measurements, self-recorded home BP (SHBP) lower BP decreased cardiovascular events and mortal-
monitoring, and ABPM (Table 42.1). ABPM and SHBP ity [75, 76] (Table 42.1). In a meta-analysis of eight RCTs
monitoring correlate with outcomes while dialysis meas- (1679 patients including both HD and peritoneal dialysis
urements do not appear to be as reliable with predialysis with varying inclusion criteria) for BP lowering in dialysis
overestimating and postdialysis underestimating BP read- patients, the weighted mean SBP was 4.5 mmHg lower and
ing related to changes in volume status [70]. There is a J- diastolic blood pressure (DBP) 2.3 mmHg lower in actively
or U-shaped curve for BP and mortality in the HD treated patients compared to controls. Any BP lowering
population but most studies are based on predialysis BP treatment (which included carvedilol, rampiril, fosinopril,
measurements [71] and do not distinguish between BP telmisartan, candesartan, losartan, and amlodipine)
lowering by therapies (e.g. probing dry weight, antihyper- reduced the risk of CV events (risk ratio [RR] 0.71, 95% CI
tensives) or confounders (e.g. infection, CV disease). For 0.55–0.92, P = 0.009), CV mortality (RR 0.71, 95% CI 0.50–
example, in a Cox regression model of 24 525 HD patients 0.99, P = 0.044) and all-cause mortality (RR 0.80, 95% CI
from DOPPS (excluding predialysis SBP < 110 and 0.66–0.96, P = 0.014) compared to control regimens [75].
DBP < 50 suggestive of comorbid illness) adjusted for con- Although the above meta-analysis suggests benefits to
founders, all-cause mortality was elevated with SBP 110– treatment with antihypertensives, whether classes of drugs
129 mmHg and lower with SBP 150–159 mmHg as differ importantly from one another was not clear. Small
compared to 130–139 mmHg. Mortality was elevated with trials comparing classes of drugs do suggest differences in
DBP 50–59 mmHg compared to 80–90 mmHg [72]. tolerability and adverse events. For example a RCT
compared lisonopril (n = 100) to atenolol (n = 100) in HD trials. Neither 4D nor ARORA demonstrated a statistical
patients with HTN defined by ambulatory blood benefit of LDL lowering with statin-based regimens on
pressure > 135/85 and left ventricular hypertrophy (LVH). their composite endpoint of cardiovascular death, nonfatal
The trial was stopped early for potential excess harms in myocardial infraction (MI) or stroke. The 4D trial did show
lisonopril-treated patients, including serious adverse a benefit on the secondary endpoint of combined cardiac
events, hospitalization, HTN, and hyperkalemia [77]. events (RR 0.82; 95% CI 0.68–0.99, P = 0.03).
However, excess harm was not seen in trials comparing The Study of Heart and Renal Protection (SHARP) trial
other angiotensin blocking drugs to controls [78]. examined the role of lipid lowering with a statin and
Selection of antihypertensives may be based on extrapo- ezetimibe regimen in patients with CKD. One-third of the
lating evidence from the nondialysis population with spe- patients with CKD enrolled in the SHARP trial were on
cific indications (e.g. beta-blockers or RAAS inhibition for dialysis at the outset of the trial [86]. The SHARP trial
patients with reduced left ventricular function [79]). demonstrated a significant benefit of simvastatin plus
Options include beta-blockers with consideration of dialyz- ezetimibe in patients with CKD on the outcome of first
ability [80], angiotensin converting enzyme (ACE) inhibi- major atherosclerotic event. This benefit was consistent
tors or angiotensin receptor blocker (ARB) (reduction in LV with the previous trials of statins in patients with normal
mass but not CV events in a meta-analysis of five RCTs with renal function when adjusted for LDL lowering achieved.
BP 120–170/70–80), dihydropyridine calcium channel Within the subgroup of patients on dialysis in the SHARP
blockers, alpha-blockers, clonidine, mineralocorticoid trial there was no statistically significant benefit of LDL
receptor antagonists (BP lowering ranging from 1.7–11 vs. lowering therapy. SHARP was not powered to assess the
2–5.2 in controls, RR CV mortality 0.40, 95% 0.15–0.75, RR subgroup of patients on dialysis. There was no statistical
mortality 0.40, 95% CI 0.23–0.69 [81]) and minoxidil. The evidence that the effects in patients on dialysis differed
use of RAAS inhibition was associated with less mortality significantly from the beneficial effect seen in patients with
in DOPPS with a hazard ratio of 0.89 (95% CI 0.80–0.99) and CKD not on dialysis but, as in most trials, there was limited
0.94 (95% CI 0.90–0.99) in 11 241 incident and 37 124 preva- power to detect such difference in effect.
lent HD patients [82]. However, in an RCT comparing olm- The Cholesterol Treatment Trialists (CTT) conducted a
esartan (n = 235) to control (n = 234) in HD patients with meta-analysis on the topic of the effect of CKD on LDL low-
average pre-HD BP >140/90 targeting BP <140/90 did not ering with statin-based regimens using individual partici-
show any difference in between the composite of death, pant level data [87]. There was evidence that statin therapy
nonfatal stroke, nonfatal myocardial infarction, and coro- lowered the risk of vascular events in patients with CKD
nary vascularization. Thus, ideal targets and therapeutic not on dialysis. Even when accounting for the smaller
agents remain uncertain. decreases in LDL seen in CKD this relative benefit declined
proportionately with estimated glomerular filtration rate
(eGFR). There was little evidence of any benefit in patients
Resistant Hypertension
on dialysis [87]. A Cochrane Review on the topic of statins
Resistant hypertension in HD may be due to inadequate in patients on dialysis found similar results showing no sig-
volume control or nonadherence in addition to other sec- nificant benefit of statin therapy on mortality or cardiovas-
ondary causes of HTN. Paradoxical hypertension can occur cular events [88].
during hemodialysis, possibly related the dialyzability of On the basis of these studies, there is no compelling evi-
antihypertensives, neurohormonal activation, an altered dence that initiation of hemodialysis is an indication for sta-
balance of vasoactive substances, endothelial dysfunction, tin therapy to prevent cardiovascular events (Table 42.1). By
and hypervolemia [83]. The approach to these conditions is the same token, there is no compelling evidence that patients
similar to that already outlined, including reducing dry already receiving a statin at the time of hemodialysis initia-
weight, lowering sodium concentration in dialysate, and tion should have the statin discontinued since patients with a
preferentially utilizing nondialyzable antihypertensives. historical indication for statin therapy were not studied.
Statin Therapy in Patients on Hemodialysis

There have been a number of RCTs evaluating low-density
Nutrition
lipoprotein (LDL) lowering with statin-based regimens in
Introduction
patients with end-stage renal disease (ESRD) to prevent
vascular events. The major trials of statin use exclusively in Malnutrition is reported in up to 72% of adult patients
patients with ESRD were the 4D [84] and AURORA [85] receiving hemodialysis. Chronic inflammation and
Nutritio 43
cardiovascular disease are commonly associated with the protein (prealbumin or more correctly transthyretin).
development of malnutrition, leading to the concept of the Plasma creatinine concentrations reflect muscle mass,
malnutrition–inflammation–atherosclerosis syndrome. dietary protein, residual kidney function, and dialysis
The International Society of Renal Nutrition recommends clearance, in addition to nutritional status. Blood
the term protein-energy wasting syndrome, rather than transthyretin concentrations change rapidly in response to
malnutrition, to better characterize the loss of body protein changes in hepatic protein production and may add
and fuel reserves in patients receiving hemodialysis, which prognostic information independent of albumin
may be independent of nutritional intake. While some concentrations. Overall, the degree to which these are
studies demonstrated declining nutritional status was markers of malnutrition or are prognostic markers
associated with adverse outcomes in hemodialysis patients, independent of nutrition is unclear.
this finding was not universal. Whether discrepancies
between results reflect the use of different nutritional
Dietary Protein Intake
markers, measurement error inherent in assessing
nutritional status, or whether poor nutrition is not truly A measure of dietary protein intake is commonly
causative of poor outcome but rather reflects declining recommended and may best reflect the role of nutritional
health for other reasons, remains uncertain. However, intervention. Food diaries, including mobile applications
since poor nutrition is reasonably likely to be at least or food intake recall instruments, are the most common
partially causative of malnutrition/protein-energy wasting way to measure dietary protein intake. A simple method
and its associated poor outcomes, it is commonly perceived free of recall bias is to measure protein nitrogen appearance
as a therapeutic target. (PNA), a proxy for protein catabolic rate and roughly
equivalent to protein intake. PNA can be calculated from
total nitrogen appearance (the sum of dialysis, urine, and
Assessment of Nutritional Status
fecal nitrogen losses) and should be normalized to fat-free,
Clinical practice guidelines developed in Australasia, edema-free standardized body weight. Since measuring
Canada, Europe, the UK, and the USA made specific nitrogen in stool is difficult, equations estimate PNA from
recommendations to assess nutritional status in nitrogen in serum, urine, and dialysate. These equations
hemodialysis patients. Although these guidelines differed are more accurate in peritoneal dialysis patients, in whom
in terms of the method of assessing nutritional status, they dialysate urea concentrations can be measured directly
all suggest a combination of measures should be used rather than estimated.
(Table 42.1). The most commonly recommended assess-
ment methods follow.
Assessment of Changes in Lean Body Mass
Lean body mass may be estimated using anthropometrics,
Subjective Global Assessment
multifrequency bioimpedance assay (BIA), and dual-
The Subjective Global Assessment (SGA) is the most energy X-ray absorptiometry (DEXA). BIA parameters
commonly recommended method and includes an indicative of poor nutritional status predicts mortality in
evaluation of gastrointestinal symptoms (appetite, dialysis patients. DEXA measurements are reproducible
anorexia, nausea, vomiting, diarrhea), weight change over but the degree to which they associate with mortality in
time, functional impairment, and a visual examination of dialysis patients is uncertain. Hand grip strength is a sim-
subcutaneous tissue and muscle mass. While the SGA is ple marker of muscle mass and predicted clinical events in
inexpensive and easy to perform, it has not been rigorously men but not women in one study of patients receiving
validated as a measure of malnutrition and may be insensi- hemodialysis.
tive to nonsevere forms of malnutrition.
Management of Malnutrition
Markers of Visceral Protein Stores
The general principles of treating a patient receiving hemo-
Like SGA, markers of visceral protein stores, for example dialysis who is malnourished are (i) to treat any underlying
serum albumin, are commonly recommended to assess cause (e.g. infection, heart failure, depression etc.), (ii) to
malnutrition. However, serum albumin decreases with ensure the patient is receiving adequate dialysis, and (iii) to
systemic inflammation and other factors not associated increase nutritional intake. While the first of these princi-
with nutrition. Other potentially useful markers of ples does not have any evidence to guide it, it seems logical
nutrition include serum creatinine and thyroid binding and at low risk of causing harm. The second principle is
guided by the observations that some individuals with uncertain if they improve quality of life, physical or mental
lower small solute clearance are more likely to have function, or survival. These trials tend to be at high risk of
protein-energy wasting. However, in the HEMO trial that bias (unblinded, unclear allocation methods) and have
compared a higher dialysis dose (average delivered Kt/V of small sample sizes.
1.53) to a lower dose (average delivered Kt/V 1.10) there If oral nutrition is not feasible, tolerated or does not
was no evidence that a higher dose improved markers of improve markers of nutrition, enteral feeding is an option
nutritional status [89] (HEMO). Similarly, in the Frequent for some. Similar to supplements, enteral feeding appears
Hemodialysis Network trial, daily dialysis did not improve to improve markers of nutritional status but its effect on
markers of nutrition [90]. patient-important outcomes is unclear.
Although intensive nutritional support may improve Finally, one may consider parenteral nutrition, either
markers of poor nutritional status, it is uncertain whether intradialytic or total parenteral nutrition. While these may
this translates to improved survival, function or quality of also improve markers of nutritional status in small, rand-
life for patients receiving dialysis. Having said that, increas- omized trials at high risk of bias, they have also not demon-
ing nutritional intake may start with dietary advice/coun- strated benefits in patient important outcomes. For both
seling and encouraging intake of food with or without enteral and parenteral feeding, understanding the risk of
supplements (Table 42.1). RCTs of nutritional supplements harm is less well delineated than estimating benefits. Risks
compared to standard diets appear to increase markers of are often extrapolated from other populations which may
nutrition (e.g. serum albumin and lean body mass) but it is or may not be broadly similar to hemodialysis patients.
References
Yusuf, A.A., Hu, Y., Singh, B. et al. (2016). Serum

1 8 Aslam, S., Friedman, E.A., and Ifudu, O. (2002).
potass

(Evidence-Based Medicine) Donald A. Molony, Jonathan C. Craig, Giovanni Strippoli - Evidence-Based Nephrology, 2 Volume Set-Wiley-Blackwell (2022)

Uploaded by

Copyright:

Available Formats

(Evidence-Based Medicine) Donald A. Molony, Jonathan C. Craig, Giovanni Strippoli - Evidence-Based Nephrology, 2 Volume Set-Wiley-Blackwell (2022)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Evidence-Based Medicine) Donald A. Molony, Jonathan C. Craig, Giovanni Strippoli - Evidence-Based Nephrology, 2 Volume Set-Wiley-Blackwell (2022)

Uploaded by

Copyright:

Available Formats

09-12-2022 15:09:44

Jonathan C. Craig, MBChB, DipCH, MMed(Clin Epi), PhD, FAHMS

Giovanni F.M. Strippoli, MD, PhD, MPH, MM (Epi)

With section editors

Aminu Bello Liz Lightstone

ffirs_Vol1.indd 3 09-12-2022 15:09:44

Limit of Liability/Disclaimer of Warranty

Library of Congress Cataloging-in-Publication Data

Cover Design: Wiley

Set in 9.5/12.5pt STIXTwoText by Straive, Pondicherry, India

ffirs_Vol1.indd 4 09-12-2022 15:09:44

1 An Introduction to the Epidemiology of Chronic Kidney Disease 3

2 The Surveillance and Burden of Chronic Kidney Disease 14

3 Progression of Chronic Kidney Disease: An Evidence-based Approach to Risk Stratification 26

4 Screening for Chronic Kidney Disease 44

5 Risk Prediction in Chronic Kidney Disease 60

6 Chronic Kidney Disease in Disadvantaged Populations 72

Part 2 Acute Kidney Injury 85

7 Overview / Definition, Classification, and Epidemiology of Acute Kidney Disease 87

8 Pre-Renal Failure and Obstructive Disease 96

9 Hepatorenal Syndrome 107

10 Acute Tubular Necrosis 123

11 Iodinated Contrast and Acute Kidney Injury 145

12 Miscellaneous Etiologies of Acute Kidney Injury 163

13 Renal Replacement Therapy in Acute Kidney Injury 185

Part 3 Primary Diseases of the Kidney 203

14 Renal Biopsy 205

15 Minimal Change Disease and Focal Segmental Glomerulosclerosis in Adults 214

16 Membranous Nephropathy 235

ftoc_V1.indd 5 09-12-2022 15:12:29

17 IgA Nephropathy in Adults and Children 254

18 Membranoproliferative Glomerulonephritis 272

19 Autosomal Dominant Polycystic Kidney Disease 288

20 Urinary Tract Infections 305

21a Toxic Nephropathies: Environmental Agents and Metals 311

21b Toxic Nephropathies: Nonsteroidal Anti-Inflammatory Drugs 319

Part 4 Secondary Diseases of the Kidney 329

23 Renovascular Disease 354

24 Secondary Diseases of the Kidney: Diabetic Nephropathy 366

25 Lupus Nephritis 400

26 Hemolytic Uremic Syndrome 425

28 ANCA-associated Vasculitis 461

29 Paraprotein-associated Kidney Disorders 478

30 Primary Immune Complex mediated Membranoproliferative Glomerulonephritis and

Part 5 Chronic Kidney Disease and Complications 505

31 Cardiovascular Disease and CKD 507

32 Infection and CKD 526

33 Treatment of Anemia in Chronic Kidney Disease 542

34 Dyslipidemia in Chronic Kidney Disease 549

35 Chronic Kidney Disease and Hypertension 573

36 Chronic Kidney Disease-Mineral and Bone Disorder 589

38 Preparation for Dialysis 658

39 Choice of Hemodialysis or Peritoneal Dialysis for Kidney Replacement Therapy 671

ftoc_V1.indd 6 09-12-2022 15:12:29

40 Modalities of Extracorporeal Therapy 3

41 Dialysis Dose and Adequacy for Hemodialysis 20

42 General Management of the Hemodialysis Patient 34

43 Infections in Hemodialysis Patients 48

44 Vascular Access for Hemodialysis 66

49 Urgent-start Peritoneal Dialysis 156

Part 10 Electrolytes and Acid-Base Disorders 301

70 Shigatoxin-related Hemolytic-uremic Syndrome 463

Part 12 Patient-centered Care and Outcomes 473

71 Shared Decision-making 475

utcomes are now being developed, with advancement of