Report 2013

The Renal Association
UK Renal Registry
Southmead Hospital
Southmead Road
Bristol, BS10 5NB, UK
Telephone
+ 44 (0) 117 323 5665
Fax
+ 44 (0) 117 323 5664
Email
[email protected]
Web site
www.renalreg.org
Director
Ron Cullen
Medical Director
Fergus Caskey
Damian Fogarty
Medical Advisor
Terry Feest
Management Team
David Bull
Hilary Doxford
Retha Steenkamp
Project Management
Chrissie Jacobs
Sue Shaw
Clinical Informatics
Fiona Braddon
Shaun Mannings
Clinical Data Management
Fran Benoy-Deeney
Lynsey Billett
Paul Dawson
Jo Wilson
Programmers
Matthew Brealey
George Swinnerton
Business Support
Steph Shearn
Laura Woodward
Foreword
Welcome to the 2013 UK Renal Registry Report. It continues to be a tribute to the hard work of the renal commu-
nity and to the Renal Registry itself. Last year, David Wheeler referred to the achievements of the Registry but cau-
tioned against complacency. As the NHS in England goes through the biggest reorganisation since it’s formation so too
the Registry and the renal community need to consider and perhaps adjust and develop their remit. That is already
apparent as this report is the first to be published in the calendar year following data collection. For the teams that
supply the data to the Registry that does mean an increased responsibility in assuring data completeness and quality
freeing the Registry to devote energies to analysis and publication.
The NHS reorganisation also provides the stimulus to consider the role of the Registry. It certainly provides
the clinical teams with an insight into the quality of care delivered within renal units but as yet the message for
commissioners, patients and carers is not clear. The Francis report not only highlighted safety but also patient
experience as key areas to understand and monitor.
So it is important that the Registry understands the needs of commissioners to assure the quality of the service that
they commission. Aspects of those quality measures may be found within this report but there is a need to consider
system level measures that have a global reach to understand quality within provision. Safety clearly is a major element
within that and there is a need to develop safety measures within the dataset. It is gratifying to see a continued and
sustained fall in MRSA bacteraemia but there is a need to move beyond this single measure and document harm events
in more depth.
The Registry is now working closely with NHS England on completing pilots around patient experience and
outcome measures. These must be embedded into clinical practice in the way that systematic data collection has
been achieved for laboratory data. These need to then develop into patient centred outcome measures that provide
system level markers for both the users and the commissioners of the service to understand the quality of care.
Behind that, work continues to broaden data collection to include earlier stages of CKD into the remit of the
Registry and consolidate RaDaR, the rare disease registry. The Registry is also a key partner in the newly established
Acute Kidney Injury programme board, with the aim of significantly reducing the burden from AKI – a project set to
run over 3 years.
It is also a time for wider collaboration. The Registry is also an important component of the National cardio-
vascular intelligence network, a far reaching project linking the cardiovascular disease headings into a health atlas
for Public Health England.
Finally, there is the huge potential in the Registry to bridge the gap between randomised controlled trials and
observation – (NEJM 369;17: 1579 Lauer et al.) and use the Registry as a registry based randomised trial. This may
yield important benefit and be a valuable asset to the renal community.
Much has been delivered by the Registry over the twenty years since its inception. There is more to come and there
is the challenge to widen the horizon.
Well done to everyone involved in the production of this report.
Richard Fluck
National Clinical Director (Renal), NHS England
Chapters and appendices
UK Renal Registry 16th Annual Report: Introduction

Chapter 1 UK Renal Replacement Therapy Incidence in 2012: National and Centre-specific Analyses
Chapter 2 UK RRT Prevalence in 2012: National and Centre-specific Analyses
Chapter 3 Demographic and Biochemistry Profile of Kidney Transplant Recipients in the UK in 2012:
National and Centre-specific Analyses
Chapter 4 Demography of Patients Wait-listed for Renal Transplantation in the UK: National and Centre-
specific Analyses
Chapter 5 Comorbidities and Current Smoking Status amongst Patients starting Renal Replacement Therapy
in England, Wales and Northern Ireland from 2011 to 2012
Chapter 6 Demographics and Outcomes of Patients from Different Ethnic Groups on Renal Replacement
Therapy in the UK
Chapter 7 Demography of the UK Paediatric Renal Replacement Therapy population in 2012
Chapter 8 Survival and Cause of Death of UK Adult Patients on Renal Replacement Therapy in 2012:
Chapter 9 Adequacy of Haemodialysis in UK Adult Patients in 2012: National and Centre-specific Analyses
Chapter 10 Haemoglobin, Ferritin and Erythropoietin amongst UK Adult Dialysis Patients in 2012: National
and Centre-specific Analyses
Chapter 11 Blood Pressure Profile of Prevalent Patients receiving Renal Replacement Therapy in 2012:
Chapter 12 Biochemical Variables amongst UK Adult Dialysis patients in 2012: National and Centre-specific
Analyses
Chapter 13 Clinical, Haematological and Biochemical Parameters in Patients Receiving Renal Replacement
Therapy in Paediatric Centres in the UK in 2012: National and Centre-specific Analyses
Chapter 14 2012 Multisite Dialysis Access Audit in England, Northern Ireland and Wales and 2011 PD One
Year Follow-up: National and Centre-specific Analyses
Chapter 15 Epidemiology of Reported Infections amongst Patients Receiving Dialysis for Established Renal
Failure in England from May 2011 to April 2012: a Joint Report from Public Health England
and the UK Renal Registry
Appendix A The UK Renal Registry Statement of Purpose
Appendix B Definitions and Analysis Criteria
Appendix C Renal services described for non-physicians
Appendix D Methodology used for Analyses of PCT/HB Incidence and Prevalence Rates and of Standardised
Ratios
Appendix E Methodology for Estimating Catchment Populations of Renal Centres in England and
Wales for Dialysis Patients
Appendix F Additional Data Tables for 2012 new and existing patients
Appendix G UK Renal Registry dataset specification
Appendix H Coding: Ethnicity, EDTA Primary Renal Diagnoses, EDTA Causes of Death
Appendix I Acronyms and Abbreviations used in the Report
Appendix J Laboratory Conversion Factors
Appendix K Renal Centre Names and Abbreviations used in the Figures and Data Tables
Contents
UK Renal Registry 16th Annual Report: Introduction 1

Ron Cullen, Damian Fogarty 1
Incidence, prevalence and survival trends 1
Completeness of data returns from UK renal centres 2
Interpretation of centre-specific clinical measures and survival comparisons 2
Information governance 5
Paediatric data and summary 5
Vascular and peritoneal access and bacteraemia 6
Patient report 6
Peer-reviewed publications since the last annual Report 6
Chapter 1 UK Renal Replacement Therapy Incidence in 2012: National and Centre-specific Analyses 9
Julie Gilg, Anirudh Rao, Damian Fogarty 9
Introduction 10
Definitions 10
UK Renal Registry coverage 10
1. Geographical variation in incidence rates 10
2. Demographics and clinical characteristics of patients starting RRT 11
3. Late presentation and delayed referral of incident patients 29
Survival of incident patients 34
Summary 34
Chapter 2 UK RRT Prevalence in 2012: National and Centre-specific Analyses 37

Catriona Shaw, David Pitcher, Rishi Pruthi, Damian Fogarty 37
Introduction 38
Methods 38
Results 39
Prevalent patient numbers and changes in prevalence 39
Prevalent patients by RRT modality and centre 39
Changes in prevalence 39
Prevalence of RRT in Primary Care Trusts in England, Health and Social Care Areas in
Northern Ireland (HBs), Local Health Boards in Wales (HBs) and Health Boards in
Scotland (HBs) 41
Factors associated with variation in standardised prevalence ratios in Primary Care Trusts
in England, Health and Social Care Areas in Northern Ireland, Local Health Boards in
Wales and Health Boards in Scotland 44
Case mix in prevalent RRT patients 44
International comparisons 61
Summary 61
Chapter 3 Demographic and Biochemistry Profile of Kidney Transplant Recipients in the UK in 2012:
National and Centre-specific Analyses 63
Rishi Pruthi, Anna Casula, Iain MacPhee 63
Introduction 64
Transplant activity, waiting list activity and survival data 64
Transplant demographics 65
Clinical and laboratory outcomes 70
Analysis of prevalent patients by CKD stage 83
eGFR slope analysis 84
Causes of death in transplant recipients 86
Chapter 4 Demography of Patients Wait-listed for Renal Transplantation in the UK: National and
Centre-specific Analyses 89
Rishi Pruthi, Rachel Hilton, Laura Pankhurst, Nizam Mamode, Alex Hudson, Paul Roderick,
Rommel Ravanan 89
Introduction 90
Allocation policy 90
Methods 90
Results 91
Prevalent patient numbers listed for transplantation 91
Prevalent patients listed for transplantation by RRT modality and centre 91
Case mix in prevalent wait-listed patients 93
Median waiting times 102
Summary 105
Chapter 5 Comorbidities and Current Smoking Status amongst Patients starting Renal Replacement
Therapy in England, Wales and Northern Ireland from 2011 to 2012 107
Anirudh Rao, Retha Steenkamp, Fergus Caskey 107
Introduction 108
Methods 108
Study population 108
Results 109
Completeness of comorbidity returns from each participating centre 109
Prevalence of multiple comorbidity 110
Frequency of each comorbid condition 110
Prevalence of comorbidity by age group 110
Prevalence of comorbidity by ethnic origin 112
Prevalence of comorbidity amongst patients with diabetes mellitus 112
Late presentation and comorbidity 113
Age and comorbidity in patients by treatment modality at start of RRT 113
Comorbidity and survival within 90 days of starting RRT 115
Comorbidity and survival 1-year after 90 days of commencing RRT 116
Discussion 117
Chapter 6 Demographics and Outcomes of Patients from Different Ethnic Groups on Renal
Replacement Therapy in the UK 119
Udaya Udayaraj, Rishi Pruthi, Anna Casula, Paul Roderick 119
Introduction 120
Methods 120
Results 121
Regional variations in incidence of RRT 121
Age, gender and social deprivation 121
Primary renal disease causing ERF 121
Comorbidity 124
Late presentation 125
Treatment modality 125
eGFR at start of RRT 126
Haemoglobin prior to start of RRT 126
Patient outcome measures 126
Patient survival 128
Hospitalisation episodes 128
Discussion 128
Chapter 7 Demography of the UK Paediatric Renal Replacement Therapy population in 2012 135
Rishi Pruthi, Catherine O’Brien, Anna Casula, Fiona Braddon, Malcolm Lewis,
Heather Maxwell, Jelena Stojanovic, Yincent Tse, Carol Inward, Manish D Sinha 135
Introduction 136
Methods 136
Results 136
Accuracy and completeness of data returns 136
The UK paediatric prevalent ERF population in 2012 136
Modality of treatment 137
Cause of ERF 138
The UK incident paediatric ERF population in 2012 139
Trends in ERF demographics 139
Pre-emptive transplantation 142
Transfer of patients to adult renal services in 2012 143
Survival of children on RRT during childhood 143
Mortality data in 2012 144
Discussion 144
Data completeness 145
Incidence, prevalence and trends 145
Treatment modality of ERF 145
Pre-emptive transplantation 145
Comorbidities 145
Causes of ERF and observed trends 1998–2012 145
Transfer out and survival data 145
Chapter 8 Survival and Cause of Death of UK Adult Patients on Renal Replacement Therapy
in 2012: National and Centre-specific Analyses 147
Rishi Pruthi, Retha Steenkamp, Terry Feest 147
Introduction 148
Methods 148
Results of incident (new RRT) patient survival 150
Comparison of survival between UK countries 150
Modality 150
Age 151
Gender 154
Change in survival on renal replacement therapy by vintage 155
Analysis of centre variability in 1 year after 90 days survival 156
Analysis of the impact of adjustment for comorbidity on the 1 year after 90 day survival 158
Survival in patients with diabetes 159
Standard primary renal disease and survival 159
Results of prevalent patient survival analyses 161
One year survival of prevalent dialysis patients by centre 161
The one year death rate in prevalent dialysis patients in the 2011 cohort by age group 161
One year survival of prevalent dialysis patients by UK country, 2000 to 2011 cohort 161
One year survival of prevalent dialysis patients with a primary diagnosis of diabetes,
2002 to 2011 cohort years 162
Death rate on RRT compared with the UK general population 163
Results of analyses on causes of death 163
Median life expectancy on RRT 166
Appendix 1: Survival tables 172
Chapter 9 Adequacy of Haemodialysis in UK Adult Patients in 2012: National and Centre-specific

Analyses 179
Catriona Shaw, Retha Steenkamp, Andrew Davenport 179
Introduction 180
Methods 180
Results 181
Achieved URR 183
Changes in URR over time 183
Variation of achieved URR with time on dialysis 183
Discussion 185
Chapter 10 Haemoglobin, Ferritin and Erythropoietin amongst UK Adult Dialysis Patients in 2012:
National and Centre-specific Analyses 191
Anirudh Rao, Julie Gilg, Andrew Williams 191
Introduction 192
Methods 193
Results 194
Anaemia management in incident dialysis patients 194
Discussion 213
Chapter 11 Blood Pressure Profile of Prevalent Patients receiving Renal Replacement Therapy
in 2012: National and Centre-specific Analyses 217
Anirudh Rao, David Pitcher, Ken Farrington 217
Introduction 218
Methods 218
Results 219
BP on each modality 219
Relationship between the centre mean and the proportion above a threshold BP in that
centre 219
Centre-specific analyses of BP in haemodialysis patients 219
Adherence to guidelines 219
Centre-specific analyses of BP in peritoneal dialysis patients 219
Centre-specific analysis of BP in transplant patients 224
Discussion 224
Chapter 12 Biochemical Variables amongst UK Adult Dialysis patients in 2012: National and
Johann Nicholas, Catriona Shaw, David Pitcher, Anne Dawnay 227
Introduction 228
Methods 228
Results and discussions 229
Mineral and bone variables 229
Mineral and bone variables 254
Bicarbonate 255
Total cholesterol 262
Chapter 13 Clinical, Haematological and Biochemical Parameters in Patients Receiving Renal

Replacement Therapy in Paediatric Centres in the UK in 2012: National and
Rishi Pruthi, Heather Maxwell, Anna Casula, Fiona Braddon, Malcolm Lewis,
Catherine O’Brien, Jelena Stojanovic, Yincent Tse, Carol Inward, Manish D Sinha 267
Introduction 268
Methods 268
Statistical analyses 268
Standards 269
Results 269
Height, weight and BMI 269
Blood pressure 272
Haemoglobin 273
Phosphate, calcium, PTH and bicarbonate 276
Discussion 277
Summary 281
Chapter 14 2012 Multisite Dialysis Access Audit in England, Northern Ireland and Wales and
2011 PD One Year Follow-up: National and Centre-specific Analyses 283
Victoria Briggs, David Pitcher, Catriona Shaw, Richard Fluck, Martin Wilkie 283
Introduction 284
Methods 284
Results 285
Variations in first dialysis access 285
2011 PD access audit one-year follow-up 298
Discussion and recommendations 300
Acknowledgement 301
Chapter 15 Epidemiology of Reported Infections amongst Patients Receiving Dialysis for Established
Renal Failure in England from May 2011 to April 2012: a Joint Report from Public
Health England and the UK Renal Registry 303
Lisa Crowley, David Pitcher, Jennie Wilson, Rebecca Guy, Richard Fluck 303
Introduction 304
Methods 304
Results 305
Methicillin resistant Staphylococcus aureus 305
Clostridium difficile 305
Methicillin sensitive Staphylococcus aureus 305
Escherichia coli 307
Discussion 312
Summary 315
Acknowledgements 315
Appendix A The UK Renal Registry Statement of Purpose 317
This appendix is available on the web only and can be found at
http://www.karger.com/Journal/Home/228539 or www.renalreg.com
Appendix B Definitions and Analysis Criteria 317

Appendix C Renal services described for non-physicians 317

Appendix D Methodology used for Analyses of PCT/HB Incidence and Prevalence Rates and of
Standardised Ratios 317
Appendix E Methodology for Estimating Catchment Populations of Renal Centres in England for
Dialysis Patients 317
Appendix F Additional Data Tables for 2012 new and existing patients 318
Appendix G UK Renal Registry dataset specification 318

Appendix H Coding: Ethnicity, EDTA Primary Renal Diagnoses, EDTA Causes of Death 318
Appendix I Acronyms and Abbreviations used in the Report 319
Appendix J Laboratory Conversion Factors 323
Appendix K Renal Centre Names and Abbreviations used in the Figures and Data Tables 325
UK Renal Registry 16th Annual Report:
Introduction
Damian Fogarty, Ron Cullen

UK Renal Registry, Bristol, UK
The UK Renal Registry (UKRR) continues to provide a increased from 19.2% in 2000 to 24.9% in 2012. Greater
national source of NHS healthcare data on patients recognition that older patients can and do tolerate
dependent on renal replacement therapy (RRT) across dialysis and transplantation is accepted but there
the four nations. Using electronic reporting and substan- remains much variation in the prevalence and take on
tial integration across the 71 adult and 13 paediatric renal rates implying there is uncertainty about prognosis and
centres independent audit and analysis of dialysis and perhaps quality of life with dialysis in particular. This
transplant activity and care across the UK is provided. merits discussion, wider data collection and further
The UKRR is part of the UK Renal Association and is research.
funded directly by participating renal centres through Other notable points include the following:
an annual capitation fee per patient per annum, currently
£19 or 0.01% of annual RRT running costs. The UKRR . The number of patients receiving home haemo-
remains relatively unique amongst renal registries in dialysis (HD) increased by 19.3% from 905 patients
publishing both centre-specific analyses of indicators of in 2011 to 1,080 patients in 2012.
quality of care, such as haemoglobin and also age- . The median age of prevalent patients was 58 years
adjusted survival statistics for each renal centre [1]. (HD 66 years, peritoneal dialysis (PD) 63 years,
transplant 52 years). In 2000 the median age was
55 years (HD 63 years, PD 58 years, transplant 48
years).
Incidence, prevalence and survival trends . In 2012, 20.7% of the prevalent UK RRT population
(with ethnicity assigned) were from ethnic min-
This year 54,824 adult patients and 861 children and orities compared to 14.9% in 2007.
young people (,18 years) receiving RRT in the UK at . There were national, regional and dialysis centre
the end of December 2012 were analysed. This represents level variations in prevalence rates. A significant
an increase of 3.7% from the 2012 report. Incidence factor in this variation was the ethnic mix of local
remains stable at 108 new patients per million popu- populations, but a large amount of the variation
lation. The increase therefore in prevalence is attributable remains unexplained. Assessment of conservatively
to increasing survival of our patients despite the overall managed stage 5 CKD patients might explain
group becoming older. more of this variation.
Elderly patients aged over 85 accepted onto RRT . Unadjusted 1 year after 90 day survival for patients
nearly doubled between 2006 and 2011. The percentage starting RRT in 2011 increased to 87.5% from 87.3%
of RRT patients who are aged greater than 70 years has for those starting in 2010.
1
The UK Renal Registry The Sixteenth Annual Report
. In all incident RRT patients aged 565 years, unad- Interpretation of centre-specific clinical measures and
justed 1 year after 90 day survival increased from survival comparisons
63.9% in 1997 to 80.6% in the 2011 cohort.
. One year survival for prevalent diabetic patients The UKRR continues to advise caution in the inter-
increased from 81.6% in the 2002 cohort to 84.9% pretation of the comparisons of centre-specific attainment
in the 2011 cohort. of clinical performance measures provided in this report.
. In the prevalent RRT dialysis population, cardio- In general terms, the UKRR has not tested for ‘significant
vascular disease accounted for 22% of deaths and difference’ between the highest achiever of a standard and
treatment withdrawal 19%, whilst 21% were the lowest achiever, as centres were not identified in
recorded as other cause of death. advance of looking at the data and statistically this
. The median life years remaining for an incident approach can be invalid. As in previous reports, the arbi-
patient aged 25–29 years was 18.5 years and trary 95% confidence interval is shown for compliance
approximately 2.4 years for a 75+ year old. with a guideline. The calculation of this confidence interval
. There was a 5% increase in overall renal transplant (based on the binomial distribution) and the width of the
numbers in 2012, with a significant rise in kidney confidence interval depends on the number of values
donation from donors after circulatory death (19%). falling within the standard and the number of patients
. In 2012, approximately 1 in 50 transplant patients with reported data. However for many of these analyses
have graft failure per year and additionally about 1 no adjustment can be made for the range of factors
in 50 transplanted patients die each year. known to influence the measured variable as outlined
above. Some of the clinical measures are summarised here:
. In 2012, 88% of prevalent HD patients achieved a

Completeness of data returns from UK renal centres URR .65%. The between centre range of prevalent
patients achieving this target was wide (between
As stated in recent reports the UKRR continues to 69.7% and 100%). The median URR in 2012 was
review the processes used for collection and validation 75%. The UK Renal Registry (UKRR) will explore
of data and its communications with renal centres. It a possible move to reporting Kt/V combined with
remains our intention to publish data following initial residual renal function.
validation on the data portal section of the UKRR website . There was substantial variation in the average dose
(www.renalreg.com). of ESA prescription used (4,000 IU/week (Leeds,
Data completeness (table 1) has improved dramatically York) to 11,025 IU/week (Newcastle) with a median
over the last few years for returns on ethnic origin, primary Hb for these centres of 110 g/L (Leeds, York) and
renal diagnosis and date first seen by a nephrologist. Alas 116 g/L (Newcastle) respectively). This may reflect
comorbidity at the start of RRT remains poorly returned the adoption of a well validated and researched pro-
overall with 55% of patients having comorbidity data. tocol in the centres with lowest ESA requirements.
There are improvements at centre level: in the 2010 data . There continues to be poor correlation between
there were ten centres with an average completeness of median Hb achieved to median ferritin and ESA
,50% across the indicator areas; this reduced to five usage across the UK centres.
centres with respect to 2011 data and two centres for the . There was also a significant variation between the
2012 data. There are both in-centre process issues and centres in the percentages of patients treated with
also design issues with some of the electronic patient an ESA and having Hb .120 g/L (HD: 7–39%,
record systems. Clinical directors are encouraged to PD: 0–33%).
consider these aspects for their planning. These data . ESA resistance is quite rare with a prevalence of
deficiencies limit the UKRR’s ability to perform fully patients receiving high doses of ESA (HD
adjusted analyses and have been highlighted in a publi- .450 IU/kg/week, PD .300 IU/kg/week) of 1.0%
cation this year [1]. Linked data between the UKRR and and 2.2% for HD and PD patients respectively. Of
Hospital Episode Statistics enabled a dramatic reduction these half or less had failed to reach the target Hb
in missing data and showed that nearly all the variation .100 g/L and therefore have true ESA resistance.
between English renal centres in 3-year survival on RRT . There continues to be significant variation in the
was explained by demographic factors and by comorbidity. achievement of BP standards between UK renal
2
Introduction Introduction to the 16th UKRR Annual Report
Table 1. Percentage completeness of data returns for ethnicity, primary renal diagnosis, date first seen by a nephrologist, comorbidity at
the start of RRT (incident patients 2012) and cause of death (for deaths in 2012 amongst incident or existing patients)
Primary Date Cause Average

Centre Ethnicity diagnosis first seen Comorbidity of death completeness Country
Ulster 100.0 100.0 100.0 100.0 100.0 100.0 N Ireland

Wrexm 100.0 100.0 97.1 100.0 100.0 99.4 Wales
Sthend 96.2 100.0 100.0 100.0 100.0 99.2 England
Antrim 100.0 100.0 100.0 96.2 100.0 99.2 N Ireland
L Kings 99.2 100.0 96.0 100.0 100.0 99.0 England
Nottm 100.0 100.0 97.9 97.0 99.0 98.8 England
York 100.0 98.1 100.0 94.3 100.0 98.5 England
Bradfd 100.0 98.6 97.1 98.6 97.7 98.4 England
Leeds 99.4 98.7 98.0 98.1 97.7 98.4 England
Swanse 100.0 100.0 99.1 95.6 97.1 98.4 Wales
Exeter 99.3 100.0 97.1 100.0 95.1 98.3 England
Oxford 100.0 100.0 98.2 99.4 92.7 98.1 England
Kent 94.8 100.0 100.0 100.0 94.9 97.9 England
Sund 98.6 100.0 98.6 94.4 97.4 97.8 England
Hull 95.9 100.0 97.9 96.9 96.9 97.5 England
Newry 100.0 100.0 100.0 88.9 96.7 97.1 N Ireland
Middlbr 99.2 98.3 97.5 90.0 94.9 96.0 England
Wolve 100.0 100.0 100.0 88.1 90.9 95.8 England
B Heart 100.0 93.1 96.0 92.1 96.6 95.6 England
Dorset 100.0 100.0 95.8 91.7 88.9 95.3 England
Truro 100.0 92.0 98.0 100.0 78.8 93.7 England
Derby 93.8 97.5 100.0 91.4 85.2 93.6 England
Donc 100.0 97.5 95.0 82.5 92.6 93.5 England
Clwyd 100.0 100.0 95.5 81.8 89.5 93.3 Wales
Bangor 100.0 100.0 90.5 76.2 100.0 93.3 Wales
Stevng 98.2 100.0 99.1 100.0 67.7 93.0 England
West NI 100.0 95.2 100.0 66.7 100.0 92.4 N Ireland
Basldn 100.0 88.7 96.2 84.9 88.9 91.7 England
Redng 80.8 97.3 97.3 84.9 91.2 90.3 England
Carlis 100.0 100.0 94.7 52.6 94.7 88.4 England
Leic 97.4 83.8 97.0 64.3 94.1 87.3 England
Glouc 100.0 100.0 94.5 37.8 91.5 84.8 England
Belfast 100.0 95.6 89.0 50.6 79.7 83.0 N Ireland
Bristol 95.9 84.5 94.6 54.7 82.2 82.4 England
Prestn 100.0 98.6 95.8 9.5 97.6 80.3 England
Colchr 100.0 100.0 100.0 0.0 100.0 80.0 England
Dudley 100.0 98.2 98.2 0.0 90.9 77.5 England
Chelms 80.0 97.8 97.8 11.1 100.0 77.3 England
Sheff 98.1 99.4 98.7 83.5 0.8 76.1 England
Newc 98.1 98.1 89.4 77.9 16.9 76.1 England
Stoke 96.1 93.5 98.7 0.0 89.6 75.6 England
Shrew 96.5 68.4 98.3 100.0 7.9 74.2 England
B QEH 100.0 99.5 99.5 66.7 2.1 73.6 England
Norwch 94.6 91.9 64.9 37.8 76.1 73.1 England
L Barts 100.0 93.5 1.5 72.6 79.9 69.5 England
Carsh 86.0 77.3 88.0 53.3 40.8 69.1 England
Ports 94.4 98.1 96.9 33.5 19.8 68.6 England
Covnt 99.1 98.2 98.2 9.8 33.3 67.7 England
Camb∗ 99.2 32.3 100.0 2.4 94.1 65.6 England
M RI 100.0 96.3 92.4 26.3 9.9 65.0 England
L Rfree 87.9 99.6 99.2 29.6 7.0 64.7 England
Cardff 100.0 99.4 98.8 21.8 0.6 64.1 Wales
L St.G 89.0 81.3 65.9 36.3 42.4 63.0 England
Ipswi∗ 90.7 34.9 97.7 2.3 77.4 60.6 England
3
Table 1. Continued
Primary Date Cause Average

Centre Ethnicity diagnosis first seen Comorbidity of death completeness Country
Liv Ain 100.0 100.0 100.0 0.0 0.0 60.0 England

Brightn 94.9 97.8 91.8 14.0 1.1 59.9 England
Plymth 97.9 72.3 31.9 55.3 41.2 59.7 England
L West 100.0 99.7 0.3 0.9 96.8 59.5 England
Liv RI 95.5 81.8 99.1 0.9 2.8 56.0 England
L Guys 96.9 86.6 22.4 1.6 58.8 53.2 England
Wirral 98.0 38.0 97.9 2.0 2.7 47.7 England
Salford 89.6 21.6 10.6 0.0 0.0 24.4 England
Abrdn 100.0 65.7 Scotland

Airdrie 100.0 93.9 Scotland
D & Gall 100.0 81.3 Scotland
Dundee 100.0 62.2 Scotland
Dunfn 100.0 87.5 Scotland
Edinb 100.0 100.0 Scotland
Glasgw 100.0 96.0 Scotland
Inverns 100.0 95.7 Scotland
Klmarnk 100.0 96.8 Scotland
∗
These centres have an unrealistically high percentage of people with diagnosis ‘uncertain’. Therefore the primary diagnosis value given is the
percentage with a specific diagnosis
centres. Only 26% of PD patients and 27% of of their practice and perhaps offer RRT to sicker patients
transplant patients achieved the Renal Association than in previous times it is important to ensure that these
guideline of SBP ,130 mmHg and DBP ,80 mmHg. benchmarking activities do not create a negative pressure
. There was marked variation (45–80%) between such that centres do not offer treatment to patients if a
centres achieving their pre-dialysis SBP readings in local clinical decision has deemed this appropriate. So
the target range suggested by the RA guidelines of such differences between centres’ practices need to be
120–160 mmHg. interpreted in the light of measured and unmeasured
. 56% of HD patients and 61% of PD patients variables that may account for these differences, the clini-
achieved the audit measure for phosphate. 77% of cal impact of the differences and trend in these variables
HD and 78% of PD patients had adjusted calcium over time. For instance the one year survival of a centre
between 2.2–2.5 mmol/L. may be in the lowest quartile of centres but be improving
. 58% of HD and 65% of PD patients had a serum faster than others and may reflect excellent care given the
PTH between 16–72 pmol/L. case-mix and socio-demographic population base of the
region. Furthermore the interpretation of survival in
For a number of years de-anonymised centre specific RRT patients needs to be seen in the context of the
reports on survival of RRT patients have been published. total population with advanced CKD (symptomatic
This has taken on significant gravitas given the Francis, stage 5 CKD) that may merit renal replacement therapy.
Keogh Enquiries and the ongoing CQC inspections of Since conservative care is used for many patients in
patient care and outcomes at a number of hospital trusts. whom there is a choice not to start dialysis the selection
In 2011 (2010 data) the UKRR sent letters to six centres of sicker (and/or) older patients in one centre versus the
with lower than expected survival at one year after 90 practice in another centre may result in unmeasured
days for incident patients starting on RRT; in 2012 differences in survival due to this potential selection
(2011 data) this was required for only three centres and bias. For this important reason and the need to under-
for the 2012 data two centres were contacted. These stand the quality of conservative care it is planned to
centres are often managing cohorts of patients that may expand the UKRR remit (technically and with appropri-
be sicker than some of their benchmarked peers but ate information governance) to capture routine data on
this can only be assessed if the data to support this those patients with CKD stage 5. For the present centres
contention is returned. As centres push the boundaries are asked to report their outlying status internally at trust
4
level and follow up with robust mortality and morbidity ensure that there is an appropriate balance between the
meetings. protection of patient information and the use and sharing
The UKRR has no statutory powers. However, the fact of information to improve patient care. This review
that the UKRR provides centre-specific de-anonymised is available at https://www.gov.uk/government/uploads/
analyses of important clinical outcomes, including system/uploads/attachment_data/file/192572/2900774_
survival, makes it important to define how the UKRR InfoGovernance_accv2.pdf. This so called Caldicott 2
responds to apparent under-performance. The senior review is likely to shape data-sharing in many domains
management team of the UKRR (Director, Medical including healthcare registries.
Director and Head of Research and Audit) communicate
survival outlier status with the renal centres in advance of
publication of this finding. The centres are asked to pro-
vide evidence that the Clinical Governance department Paediatric data and summary
and Chief Executive of the Trust housing the service
have been informed. In the event that no such evidence The UKRR continues to provide a service for collect-
is provided, the Chair of the UKRR would inform the ing paediatric data. It is hoped that this task is getting
President of the Renal Association, who would then easier as the Hospital Trusts for those centres invest
take action to ensure that the findings were properly more resources into appropriate clinical information
investigated. These procedures are followed even if systems needed for day-to-day patient care and
there is evidence that further adjustment, for instance reporting structures. Notable aspects from the 2012
for comorbidity, might explain outlier status. Coupled data are:
with open publication of the analyses this should by itself
drive up the quality of care provided. . A total of 861 children and young people under
18 years with ERF were receiving treatment at
paediatric nephrology centres in 2012. 80.2% had
a functioning kidney transplant, 10.6% were receiv-
Information governance ing haemodialysis (HD) and 9.2% were receiving
peritoneal dialysis (PD).
At present the UKRR operates within a comprehensive . A third of children on RRT had one or more
governance framework which concerns data handling, reported comorbidities.
reporting and research, including data linkages and shar- . Median weight z-score for children on dialysis was
ing agreements. The Chair of the Renal Association Renal −1.1 whereas children with a functioning transplant
Information Governance Board is appointed as the Lead had a near normal weight (median z-score 0.1).
for Governance, with the UKRR Director responsible for . Median height z-score for children on dialysis was
day to day management of governance compliance and −2.0 and for children with a functioning transplant
the Head of Operations is the operational information −1.3.
governance lead. The Framework is based on good . 76% of transplant patients, 57% of haemodialysis
practice, as described in the Information Governance patients and 56% of peritoneal dialysis patients
Framework: (http://www.connectingforhealth.nhs.uk/ had a systolic blood pressure within the 90th per-
systemsandservices/infogov/igap/igaf ) and the Research centile standard.
Governance Framework for Health and Social Care . 92% of transplant patients, 74% of HD patients and
(2005). 83% of PD patients had a haemoglobin within or
The UKRR has temporary exemption, granted by above the age appropriate standard.
the Secretary of State under section 251 of The National . 50% of HD patients and 56% of PD patients
Health Service Act (2006), to hold patient identifiable achieved the audit standard for phosphate.
data. This exemption is reviewed annually. The UKRR . Over the past 15 years for those referred early there
has successfully completed the Connecting for Health has been a rise in pre-emptive transplantation rates,
information governance toolkit to a satisfactory standard. rising from 26.2% in 1998–2002 to 36.3% in 2008–
Recently following a request from the Secretary of 2012.
State for Health, Dame Fiona Caldicott carried out a . At transfer to adult services, 81.5% of patients had a
new independent review of information sharing to functioning kidney transplant.
5
Vascular and peritoneal access and bacteraemia have been falling since data collection began in 2007.
. In the same period there were 138 Clostridium
The Vascular Access Audit was funded by the Health- difficile infection episodes with a rate of 0.61 per
care Quality Improvement Partnership (HQIP) and run 100 prevalent dialysis patients per year.
by the NHS Information Centre from 2009–2012. The . Methicillin sensitive Staphylococcus aureus (MSSA)
expectation was that renal centres would have established bacteraemia rates were 1.15 per 100 prevalent
systems and processes that record dialysis access data for dialysis patients per year with 322 episodes of
all incident patients. The Renal Association and the blood stream infection reported.
UKRR always considered that this project should fall to . Eschericia coli data were available from June 2011
its systems and processes. Although all UK renal centres and showed a reported rate of 0.92 per 100 prevalent
have IT systems capable of collecting the �400 item dialysis patients per year.
UKRR dataset the additional items required for
paediatrics or detailed vascular access for instance are
not uniformly entered for a variety of reasons. Each
year the Renal Registry dataset is reviewed and the
Patient report
implications of any changes discussed with third party
suppliers of IT systems. The following are some key
It has been the intention of the UKRR to produce a
points:
patient report based on the data analysed for the main
annual report which is of particular interest and relevance
. In 2012, 51 centres in England, Wales and Northern
to patients. A patient leaflet will initially be produced in
Ireland (representing 82% of all centres) returned
conjunction with the National Kidney Federation early
data on first access from incident haemodialysis
in 2013 based on 2012 data. This leaflet will be issued
(HD) patients (n = 3,720) and peritoneal dialysis
via patient groups and patient charities. Additional
(PD) patients (n = 1,018).
patient leaflets will be produced over time with the aim
. For all incident HD patients, 38.3% started therapy
of producing an annual standalone patient focused
on arterio-venous fistula (AVF), 36.9% on tunnelled
chapter in the future.
line (TL), 23.5% on non-tunnelled line (NTL) and
1.2% by means of arterio-venous graft (AVG).
. Initial surgical assessment was a key determinant of
the likelihood of AVF formation; 70.4% of patients
assessed by a surgeon at least three months before Peer-reviewed publications since the last annual
commencing dialysis started on an AVF. Contrast- Report
ingly, only 9.7% of patients not surgically assessed
used an AVF as first dialysis access. The primary role of the UKRR is to use data to develop
. Length of time known to nephrology services and high-quality analyses to drive a cycle of continuous
likelihood of commencing dialysis using either an improvement in the care of patients with kidney disease
AVF or a PD catheter are strongly associated. For in the UK. Research is an important part of improving
patients presenting late, 84.6% started on a line the quality of existing analyses and developing new
(TL/NTL). Amongst patients known to the unit ones. Research from the Registry appears in peer-
for at least a year only 33.9% started via a line. reviewed journals [2–11] in addition to articles published
. For centres returning data on one year peritoneal in collaboration with the EDTA-ERA Registry [12–15],
dialysis outcomes, the majority of centres main- other reports published by the Registry [16] and posters
tained .50% of patients on peritoneal dialysis at presented at renal conferences [17–24]. A list of publi-
one year, however only five centres maintained cations involving analyses of UKRR data is available on
.80% on PD at one year. the UKRR website at www.renalreg.com.
. From May 1st 2011 to April 30th 2012 there were 49 With the progressive improvement in survival of
episodes of methicillin resistant Staphylococcus patients on RRT documented in this report it seems
aureus (MRSA) bacteraemia in end stage renal inevitable that the prevalence of RRT will continue to
failure patients on dialysis. This represents a further increase, even with continuing improvements in preven-
slight decline in MRSA bacteraemia rates which tive care, earlier referral of patients with advanced CKD
6
and where appropriate, provision of supportive care in on the outcomes of RRT and to develop reliable analyses
place of RRT for those who wish for it. RRT is a high of the epidemiology and outcomes of conservative man-
cost therapy and this will pose a challenge to the NHS agement of advanced CKD.
and to the UK renal community. This will make it
more important than ever to submit high quality data Conflicts of interest: none
References
1 Nephrol. Dial. Transplant. (2013) doi: 10.1093/ndt/gft363 European Best Practice Guidelines. Nephrology Dialysis Transplan-
2 Bartlett C, Simpson K, Turner AN. Patient access to complex chronic tation. 2009 Apr;24(4):1267–1274
disease records on the Internet. Bmc Medical Informatics and Decision 13 Jager KJ, Ocak G, Drechsler C, Caskey FJ, Evans M, Postorino M, et al.
Making. 2012 Aug 6;12 The EQUAL study: a European study in chronic kidney disease stage 4
3 Castledine CI, Gilg JA, Rogers C, Ben-Shlomo Y, Caskey FJ. How much patients. Nephrology Dialysis Transplantation. 2012 Oct;27:27–31
of the regional variation in RRT incidence rates within the UK is 14 Kramer A, Stel VS, Caskey FJ, Stengel B, Elliott RF, Covic A, et al.
explained by the health needs of the general population? Nephrology, Exploring the Association between Macroeconomic Indicators and
dialysis, transplantation: official publication of the European Dialysis Dialysis Mortality. Clinical journal of the American Society of
and Transplant Association–European Renal Association. 2012 Oct; Nephrology: CJASN. 2012 Oct;7(10):1655–1663
27(10):3943–3950 15 van de Luijtgaarden, M.W.M., et al., Factors Influencing the Decision to
4 Connor TMF, Oygar DD, Gale DP, Steenkamp R, Nitsch D, Neild GH, Start Renal Replacement Therapy: Results of a Survey Among European
et al. Incidence of End-Stage Renal Disease in the Turkish-Cypriot Nephrologists. American Journal of Kidney Diseases, 2012;60(6):940–
Population of Northern Cyprus: A Population Based Study. PloS one. 948
2013 Jan 17;8(1) 16 Fluck R, Pitcher D, Steenkamp R, Vascular Access Audit Report 2012:
5 Jones HE, Clothier JC, Casula A, Sinha MD, Hulton SA, Inward C. The UK Renal Registry and NHS Kidney Care 2012
UK experience of chronic dialysis in infants and children under two 17 Anwar S, Pruthi R, Kenchayikoppad S. Hypercalcemia a Risk Factor for
years of age: a post millennium perspective on behalf of the British Renal Transplant Dysfunction. Nephrology Dialysis Transplantation.
association for paediatric nephrology. Pediatric Nephrology. 2012 2012 May;27:156
Sep;27(9):1642–1643 18 Castledine C, Gilg J, Rogers C, Ben-Shlomo Y, Caskey F. Renal centre
6 Judge A, Caskey FJ, Welton NJ, Ansell D, Tomson CRV, Roderick PJ, characteristics and practice patterns associated with RRT incidence in
et al. Inequalities in rates of renal replacement therapy in England: the UK. Nephrology Dialysis Transplantation. 2012 May;27:72
does it matter who you are or where you live? Nephrology Dialysis 19 Castledine C, Gilg J, Rogers C, Ben-Shlomo Y, Caskey F. Factors
Transplantation. 2012 Apr;27(4):1598–1607 affecting the number of patients treated with either peritoneal or
7 Schaefer F, Jager K, Reid C, Simpson K. The new ERA-EDTA coding home haemodialysis in the UK. Nephrology Dialysis Transplantation.
system for coding causes of renal failure. Pediatric Nephrology. 2012 2012 May;27:274
Sep;27(9):1788–1789 20 Connor T, Oygar D, Nitsch D, Gale D, Steenkamp R, Neild GH, et al.
8 Sinha MD, Gilg JA, Kerecuk L, Reid CJD, British Assoc Paediat N. High incidence of end-stage renal disease in the Turkish-Cypriot
Progression to hypertension in non-hypertensive children following population of northern Cyprus: a population based study. Nephrology
renal transplantation. Nephrology Dialysis Transplantation. 2012 Dialysis Transplantation. 2012 May;27:384
Jul;27(7):2990–2996 21 Pruthi R, Ravanan R, Casula A, Roderick P. UK study of living kidney
9 Sinha MD, Kerecuk L, Gilg J, Reid CJD, British Assoc Paediat N. donor-recipient relationships: gender and ethnic variations. Nephrology
Systemic arterial hypertension in children following renal transplan- Dialysis Transplantation. 2012 May;27:69
tation: prevalence and risk factors. Nephrology Dialysis Transplantation. 22 Webb L, Casula A, Ben-Shlomo Y, Tomson C. The clinical character-
2012 Aug;27(8):3359–3368 istics of patients starting dialysis after renal transplant failure in the
10 Thomas HL, Banner NR, Murphy CL, Steenkamp R, Birch R, Fogarty UK 2000–2008: a UK Renal Registry analysis. Nephrology Dialysis
DG, et al. Incidence, Determinants, and Outcome of Chronic Kidney Transplantation. 2012 May;27:305
Disease After Adult Heart Transplantation in the United Kingdom. 23 Webb L, Casula A, Tomson C, Ben-Shlomo Y. Primary renal disease and
Transplantation. 2012 Jun 15;93(11):1151–1157 survival after transplant failure–is there a relationship? An analysis of
11 Udayaraj U, Ben-Shlomo Y, Roderick P, Casula A, Dudley C, Collett D, UK Renal Registry data. Nephrology Dialysis Transplantation. 2012
et al. Social Deprivation, Ethnicity, and Uptake of Living Kidney Donor May;27:304
Transplantation in the United Kingdom. Transplantation. 2012 Mar 24 Webb L, Casula A, Tomson C, Ben-Shlomo Y. Survival after renal
27;93(6):610–616 transplant failure: a UK Renal Registry analysis. Nephrology Dialysis
12 Couchoud C, Jager KJ, Tomson C, Cabanne J-F, Collart F, Finne P, et al. Transplantation. 2012 May;27:312
Assessment of urea removal in haemodialysis and the impact of the
7
Chapter 1 UK Renal Replacement Therapy
Incidence in 2012: National and
Centre-specific Analyses
Julie Gilga, Anirudh Raoa, Damian Fogartyab

a
UK Renal Registry, Bristol, UK; bCentre for Public Health, Queen’s University Belfast and Belfast Health and
Social Care Trust
Key Words
. The median age of all incident patients was 64.6
Acceptance rates . Comorbidity . Dialysis . End stage renal years but this is highly dependant on race (66.1
disease . End stage renal failure . Established renal failure . for White incident patients; 57.8 for non-White
Haemodialysis . Incidence . Peritoneal dialysis . Registries . patients).
Renal replacement therapy . Transplantation . Treatment . Diabetic renal disease remained the single most
modality common cause of renal failure (26%).
. By 90 days, 66.9% of patients were on haemo-
dialysis, 19.0% on peritoneal dialysis, 8.3% had
Summary had a transplant and 5.9% had died or stopped
treatment.
. In 2012 the incidence rate in the UK was stable at . The mean eGFR at the start of RRT was 8.5 ml/min/
108 per million population (pmp) reflecting renal 1.73 m2 similar to the previous four years.
replacement therapy (RRT) initiation for 6,891 . Late presentation (,90 days) fell from 23.9% in
new patients. 2006 to 19.3% in 2012.
. From 2006 to 2012 the incidence rate pmp was
stable for England but had increased from 95 pmp
in 2001.
9
Introduction social and demographic factors such as underlying

conditions, age, gender, social deprivation and ethnicity.
This chapter contains analyses of adult patients starting Thus, comparison of crude incidence rates by geographi-
renal replacement therapy (RRT) in the UK in 2012. It cal area can be misleading. This year’s report again uses
describes regional and national variations in incidence age and gender standardisation of PCT/HB rates as well
rates of RRT, the demographic and clinical characteristics as showing crude rates. It also gives the ethnic minority
of all patients starting RRT and analyses of late presen- percentage of each area as this influences incidence rates.
tation and delayed referral. The methodology and results The UKRR investigated the effect of socio-
for these analyses are in three separate sections. demographic, population health status and access to
care factors on RRT incidence. This work suggested
Definitions that population age, socio-economic deprivation and
The definition of incident patients is given in detail in the proportion of non-White residents were able to
appendix B: Definitions and Analysis Criteria (www. explain 22% of the observed variation in RRT incidence.
renalreg.com). In brief, it is all patients over 18 who The prevalence of diabetes in an area explained a further
commenced RRT in the UK in 2012 and who did not 4% of the variation and access to complex health
recover renal function within 90 days. Importantly this procedures (CABG/coronary angioplasty) a further 6%
does not include those with a failed renal transplant [1]. Much of the observed variation (about 2/3rds)
who returned to dialysis as they had already started RRT. remains unexplained and may be due to unmeasured
Differences may be seen in the 2007 to 2011 numbers elements of the above factors or be due to differences in
now quoted when compared with previous publications practice patterns at individual renal centres which have
because of retrospective updating of data in collaboration not yet been captured.
with renal centres, in particular for patients who were
initially thought to have acute renal failure. Where Methods
Crude incidence rates were calculated per million population
applicable and possible, pre-emptive transplant patients (pmp) and age/gender standardised incidence ratios were calcu-
were allocated to their work up centre rather than their lated as detailed in appendix D: Methodology used for Analyses
transplant centre. However, this was not possible for all (www.renalreg.com).
such patients and consequently some patients probably
remain incorrectly allocated to the transplanting centre. Results
The term established renal failure (ERF) as used within In 2012, the number of adult patients starting RRT in
this chapter is synonymous with the terms end stage the UK was 6,891 equating to an incidence rate of
renal failure/disease (ESRF or ESRD). 108 pmp (table 1.1), the same as in 2011. Wales remained
the country with the highest incidence rate (figure 1.1).
UK Renal Registry coverage For England, incidence rates have been stable for the
The UK Renal Registry (UKRR) received individual last seven years. There continued to be very marked
patient level data from all 71 adult renal centres in the gender differences in incidence rates which were
UK (five renal centres in Wales, five in Northern Ireland, 136 pmp (95% CI 132–140) in males and 80 pmp (95%
nine in Scotland, 52 in England). Data from centres in CI 77–83) in females. When incident patients aged
Scotland were obtained from the Scottish Renal Registry. under 18 were included, the UK rate was 110 pmp.
Data on children and young adults can be found in Table 1.2 shows incidence rates and standardised
chapter 7: Demography of the UK Paediatric Renal incidence ratios for PCT/HBs. The ratios calculated
Replacement Therapy population in 2012. using combined data from up to six years have been
used to determine areas with significantly high or low
incidence rates. Significantly high areas have been shaded
with bold text and significantly low areas shaded a lighter
1. Geographical variation in incidence rates grey with italicised text. There were wide variations
between areas, with 49 being significantly high and 48
Over the years, there have been wide variations in inci- being significantly low out of a total of 177 areas. Last
dence rates between renal centres. Equity of access to year these numbers were 53 and 48 areas respectively.
RRT is an important aim but hard to assess as the need The standardised incidence ratios ranged from 0.51 to
for RRT depends on many variables including medical, 2.37 (IQR 0.84, 1.18).
10
Chapter 1 UK Renal Replacement Therapy Incidence in 2012
Table 1.1. Number of new adult patients starting RRT in the UK in 2012
England N Ireland Scotland Wales UK
Number starting RRT 5,826 186 519 360 6,891

Total estimated population mid-2012 (millions)∗ 53.5 1.8 5.3 3.1 63.7
Incidence rate (pmp) 109 102 98 117 108
(95% CI) (106–112) (87–117) (89–106) (105–129) (106–111)
∗
Data from the Office for National Statistics, National Records of Scotland and the Northern Ireland Statistics and Research Agency – based on
the 2011 census
As would be expected, urban areas with high per- recognition of ERF), changes to treatment thresholds or
centages of non-White residents tended to have high the introduction of conservative care programmes.
incidence rates. Figure 1.2 shows the positive correlation Table 1.3 also shows centre level incidence rates (per
(r = 0.87, p , 0.001) between the standardised incidence million population). For the methodology used to
ratio and the percentage of the PCT/HB population that estimate catchment populations in England and Wales
was non-White. see appendix E: Methodology for Estimating Catchment
Confidence intervals are not presented for the crude Populations (www.renalreg.com). For Scotland, mid-
rates per million population but figures D1 and D2 in 2011 populations of Health Boards (from the General
appendix D can be used to determine if a PCT/HB falls Register Office for Scotland) were converted to centre
within the 95% confidence interval around the national level populations using an approximate mapping of
average rate. renal centres to HBs supplied by the Scottish Renal
The number of new patients starting RRT at each renal Registry. Estimates of the catchment populations in
centre from 2007 to 2012 is shown in table 1.3. For most Northern Ireland were supplied by personal communi-
centres there was a lot of variability in the numbers of cation from Dr D Fogarty.
incident patients from one year to the next making it There were falls of 8% and 17% respectively in the
hard to see any underlying trend. Some centres have number of new patients for Scotland and Wales between
had an increase in new patients over time and others 2007 and 2012. There was an increase of approximately
have fallen. The variation may reflect chance fluctuation, 6% in new patients for England between 2007 and
the introduction of new centres, changes in catchment 2012. Across all four countries the change between
populations or in completeness of reporting. Variation 2007 and 2012 was an increase of 3.3%.
over time may also be due to changing incidence of
established renal failure (increases in underlying disease
prevalence, survival from comorbid conditions and
2. Demographics and clinical characteristics of
patients starting RRT
150
Wales
140 N. Ireland Methods
Rate per million population
130 England
Age, gender, primary renal disease, ethnic origin and treatment
120 Scotland
modality were examined for patients starting RRT. Centre level
110 results are not shown for any centre with fewer than 10 incident
100 patients in the year. Individual EDTA codes for primary diagnoses
90 were grouped into eight categories, the details are given in
appendix H: Ethnicity and ERA-EDTA Coding (www.renalreg.
80
com).
70 Most centres electronically upload ethnicity coding to their
60 renal information technology (IT) system from the hospital
50 Patient Administration System (PAS). Ethnicity coding in these
PAS systems is based on self-reported ethnicity. For the remaining
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
centres, ethnicity coding is performed by clinical staff and

Year
recorded directly into the renal IT system (using a variety of
Fig. 1.1. RRT incidence rates in the countries of the UK 1990– coding systems). For all these analyses, data on ethnic origin
2012 were grouped into Whites, South Asians, Blacks, Chinese and
11
Table 1.2. Crude adult incidence rates (pmp) and age/gender standardised incidence ratios 2007–2012
PCT/HB – PCT in England, Health and Social Care Areas in Northern Ireland, Local Health Boards in Wales and Health Boards in Scotland
O/E – standardised incidence ratio
LCL – lower 95% confidence limit
UCL – upper 95% confidence limit
pmp – per million population
∗
– per year
Areas with significantly low incidence ratios over six years are italicised in greyed areas, those with significantly high incidence ratios over six
years are bold in greyed areas
Blank cells – no data returned to the UKRR for that year. For the one area not covered by the Registry for the entire period 2007–2012, the
combined years standardised incidence ratio and incidence rate are averages for the years covered by the Registry
Population data from the Office for National Statistics, National Records of Scotland and the Northern Ireland Statistics and Research Agency –
based on the 2011 census
% non-White – percentage of the PCT/HB population that is non-White, from 2011 census for E, W & NI (2001 for Scotland)
2012 2007–2012
Crude Crude %
Tot pop 2007 2008 2009 2010 2011 rate 95% 95% rate non-
UK Area PCT/HB (2011) O/E O/E O/E O/E O/E O/E pmp O/E LCL UCL pmp∗ White
North County Durham 513,000 0.69 0.69 0.76 0.78 0.84 1.05 123 0.80 0.71 0.90 92 1.8
East Darlington 105,600 1.13 1.04 0.94 0.96 0.93 1.26 142 1.04 0.83 1.31 115 3.8
Gateshead 200,300 0.81 0.54 0.80 0.77 0.75 0.88 100 0.76 0.62 0.92 85 3.7
Hartlepool 92,100 0.50 1.40 0.79 0.60 0.59 0.97 109 0.81 0.61 1.07 89 2.3
Middlesbrough 138,400 1.31 1.31 0.64 1.46 0.71 1.06 108 1.08 0.88 1.33 108 11.8
Newcastle 279,100 1.18 1.02 0.98 0.77 0.85 0.78 75 0.93 0.79 1.09 88 14.5
North Tyneside 201,200 0.76 0.54 0.92 0.99 0.61 0.87 99 0.78 0.65 0.95 88 3.4
Northumberland 316,300 0.75 0.65 0.59 0.63 0.84 0.78 98 0.71 0.61 0.82 87 1.6
Redcar and Cleveland 135,200 0.95 0.76 0.87 0.76 1.05 0.86 104 0.88 0.71 1.09 104 1.5
South Tyneside 148,200 1.20 0.54 1.42 0.72 1.00 0.52 61 0.90 0.73 1.10 102 4.1
Stockton-on-Tees Teaching 191,800 0.75 0.85 0.69 0.91 1.12 1.07 115 0.90 0.74 1.08 95 5.4
Sunderland Teaching 275,300 1.09 0.89 0.94 1.00 0.74 0.87 98 0.92 0.79 1.07 102 4.1
North Ashton, Leigh and Wigan 318,100 0.56 0.85 0.55 0.74 0.92 0.77 85 0.73 0.62 0.86 79 2.7
West Blackburn with Darwen 147,700 1.24 0.51 0.87 1.04 1.37 1.22 115 1.04 0.84 1.29 97 30.8
Teaching
Blackpool 142,100 0.98 0.92 1.03 0.55 0.78 1.43 169 0.95 0.78 1.16 110 3.3
Bolton Teaching 277,300 0.89 0.92 0.80 1.43 0.94 0.90 94 0.98 0.84 1.14 100 18.1
Bury 185,400 0.67 0.77 0.71 0.78 0.66 1.35 146 0.82 0.68 1.01 87 10.8
Central and Eastern Cheshire 462,800 0.66 0.67 0.68 0.75 0.77 0.74 89 0.71 0.63 0.81 83 3.1
Central Lancashire 467,400 0.78 0.90 0.94 0.62 0.78 0.89 98 0.82 0.72 0.93 89 7.8
Cumbria Teaching 499,800 0.64 0.74 0.61 0.69 0.59 0.60 76 0.64 0.57 0.73 80 1.5
East Lancashire Teaching 382,500 0.76 0.66 0.82 0.71 0.88 0.52 58 0.73 0.63 0.84 78 11.6
Halton and St Helens 301,100 0.94 0.52 0.81 0.89 1.11 0.92 103 0.87 0.75 1.01 95 2.0
Heywood, Middleton and 211,900 0.90 0.90 1.13 0.77 1.26 1.25 127 1.04 0.88 1.23 104 18.3
Rochdale
Knowsley 145,900 1.11 0.52 0.77 0.92 1.09 1.28 137 0.95 0.77 1.17 101 2.8
Liverpool 465,700 1.08 1.15 1.16 0.87 1.08 1.30 129 1.10 0.99 1.24 108 11.1
Manchester Teaching 502,900 1.29 1.31 1.42 1.31 1.24 1.41 109 1.33 1.19 1.49 103 33.4
North Lancashire Teaching 321,600 0.61 0.53 0.75 0.69 0.74 0.74 93 0.68 0.58 0.79 84 3.1
Oldham 225,200 0.91 1.09 0.90 0.97 0.98 0.71 71 0.93 0.78 1.11 91 22.5
Salford 234,500 0.62 1.02 1.01 1.39 0.65 0.87 85 0.92 0.78 1.10 90 9.9
Sefton 274,000 0.55 0.85 0.86 1.04 1.24 0.91 113 0.91 0.79 1.05 111 2.6
Stockport 283,300 0.82 0.79 0.62 0.89 0.83 0.64 74 0.76 0.65 0.90 86 7.9
Tameside and Glossop 252,900 1.33 0.76 0.90 0.96 0.93 0.59 63 0.91 0.77 1.07 96 8.2
Trafford 227,100 1.05 0.59 1.00 1.32 0.54 1.15 123 0.94 0.79 1.11 99 14.5
Warrington 202,700 0.74 0.61 1.10 0.61 0.50 0.86 94 0.74 0.61 0.90 79 4.1
Western Cheshire 237,400 0.90 0.54 0.85 1.26 1.05 0.87 105 0.91 0.78 1.07 108 2.8
Wirral 319,800 0.74 0.74 0.81 0.93 0.93 0.66 78 0.80 0.69 0.93 93 3.0
12
Table 1.2. Continued

2012 2007–2012
Crude Crude %
Yorkshire Barnsley 231,900 0.86 1.13 0.89 1.18 0.80 1.03 116 0.98 0.84 1.15 109 2.1
and the Bradford and Airedale Teaching 523,100 1.43 1.08 0.96 1.32 1.04 1.30 122 1.19 1.07 1.32 110 32.6
Humber Calderdale 204,200 0.84 0.88 1.01 0.61 0.59 0.77 83 0.78 0.65 0.95 83 10.3
Doncaster 302,500 0.64 0.76 1.02 0.95 1.05 0.80 89 0.87 0.75 1.01 95 4.7
East Riding of Yorkshire 334,700 0.67 0.98 0.89 0.72 0.77 0.83 108 0.81 0.71 0.93 103 1.9
Hull Teaching 256,100 1.09 1.05 0.99 0.93 0.71 0.79 78 0.92 0.78 1.09 90 5.9
Kirklees 423,000 0.72 0.74 1.03 0.94 1.05 0.86 90 0.89 0.78 1.02 91 20.9
Leeds 750,700 0.86 1.02 0.81 0.66 0.80 0.74 73 0.81 0.73 0.90 80 14.9
North East Lincolnshire 161,200 1.07 1.01 0.83 0.68 1.37 0.66 74 0.94 0.77 1.14 104 2.6
North Lincolnshire 163,600 0.70 0.81 0.75 0.71 1.49 1.16 134 0.94 0.78 1.14 107 4.1
North Yorkshire and York 799,000 0.83 0.71 0.80 0.64 0.87 0.92 111 0.80 0.73 0.87 94 3.4
Rotherham 257,700 1.02 1.27 0.91 1.07 0.73 0.82 93 0.97 0.83 1.13 107 6.4
Sheffield 551,800 1.17 1.15 1.30 1.07 0.98 1.25 127 1.15 1.04 1.27 115 16.3
Wakefield District 326,400 0.50 0.76 0.61 0.85 0.91 1.06 119 0.78 0.67 0.91 86 4.6
East Bassetlaw 113,000 1.68 0.61 0.68 0.84 0.82 1.04 124 0.94 0.75 1.18 111 2.6
Midlands Derby City 248,900 0.98 1.68 1.37 1.07 1.40 1.56 157 1.34 1.17 1.54 133 19.7
Derbyshire County 737,500 0.82 1.04 0.78 0.73 0.90 0.83 99 0.85 0.78 0.94 100 2.5
Leicester City 329,600 1.68 1.57 1.31 1.74 1.82 1.61 140 1.62 1.44 1.82 139 49.5
Leicestershire County and 688,800 0.86 0.71 0.80 0.93 0.83 0.71 83 0.81 0.73 0.89 92 8.3
Rutland
Lincolnshire Teaching 717,200 0.79 0.69 0.71 0.85 0.89 0.69 86 0.77 0.70 0.85 95 2.4
Northamptonshire Teaching 694,000 0.99 1.19 0.81 0.80 0.90 1.12 120 0.97 0.88 1.07 101 8.5
Nottingham City 303,900 0.97 1.31 1.46 1.49 1.06 1.18 102 1.24 1.08 1.43 106 28.5
Nottinghamshire County Teaching 673,800 1.06 0.91 1.01 0.90 0.90 0.82 95 0.93 0.85 1.02 106 4.8
West Birmingham East and North 421,400 1.45 1.73 1.45 1.38 1.86 1.61 154 1.58 1.43 1.75 149 36.1
Midlands Coventry Teaching 316,900 1.36 1.53 1.71 1.31 1.52 1.89 183 1.55 1.38 1.75 149 26.2
Dudley 313,300 0.96 0.82 1.40 0.80 0.80 1.19 137 1.00 0.87 1.14 113 10.0
Heart of Birmingham Teaching 299,200 2.47 2.83 2.68 2.19 1.89 2.14 160 2.37 2.12 2.64 177 70.5
Herefordshire 183,600 0.93 0.93 1.08 0.71 0.82 0.86 109 0.89 0.74 1.06 111 1.8
North Staffordshire 212,900 0.56 0.84 1.30 0.69 1.18 0.62 75 0.87 0.73 1.03 103 3.5
Sandwell 309,000 1.55 2.15 1.76 1.84 1.65 1.39 139 1.72 1.54 1.92 170 30.1
Shropshire County 307,100 0.78 1.00 0.71 0.92 0.92 0.73 91 0.84 0.73 0.97 103 2.0
Solihull 206,900 0.76 0.98 1.37 1.02 0.70 0.99 116 0.97 0.82 1.15 112 10.9
South Birmingham 353,700 1.26 1.53 1.39 1.09 1.26 1.09 105 1.27 1.12 1.43 121 25.3
South Staffordshire 628,500 0.95 0.88 0.77 1.00 0.97 0.76 89 0.89 0.81 0.98 102 4.7
Stoke on Trent 256,900 1.24 1.01 1.33 1.32 0.99 0.88 93 1.13 0.98 1.30 118 11.0
Telford and Wrekin 166,800 1.61 1.08 1.24 1.51 1.06 1.23 126 1.29 1.08 1.53 130 7.3
Walsall Teaching 269,500 1.13 1.37 1.01 1.84 1.10 1.34 145 1.30 1.14 1.48 138 21.1
Warwickshire 546,600 1.01 0.98 0.96 1.15 1.06 0.81 93 0.99 0.90 1.10 113 7.3
Wolverhampton City 249,900 1.01 1.44 1.11 1.45 1.18 1.41 148 1.27 1.10 1.46 131 32.0
Worcestershire 566,600 0.83 0.94 1.05 0.77 0.81 0.98 118 0.90 0.81 1.00 106 4.3
East of Bedfordshire 413,500 0.60 0.76 0.81 0.90 0.74 1.00 109 0.80 0.70 0.92 85 11.2
England Cambridgeshire 622,300 0.82 0.73 1.02 0.80 0.95 0.65 71 0.83 0.74 0.92 88 7.4
Hertfordshire 1,119,800 0.74 0.95 0.82 0.90 0.92 0.79 83 0.85 0.79 0.92 88 12.4
Great Yarmouth and Waveney 212,800 1.17 1.09 0.89 1.13 1.10 0.91 117 1.05 0.90 1.22 132 2.7
Luton 203,600 1.47 1.13 1.01 1.15 1.44 1.22 108 1.24 1.04 1.47 108 45.3
Mid Essex 375,200 0.92 0.84 0.93 0.90 0.94 0.75 85 0.88 0.77 1.00 98 4.4
Norfolk 762,000 1.07 0.88 0.69 0.81 0.81 0.77 97 0.84 0.76 0.91 104 3.5
13

2012 2007–2012
Crude Crude %
East of North East Essex 311,700 1.57 0.82 0.98 1.27 0.98 119 1.12 0.98 1.29 135 5.5
England Peterborough 184,500 1.09 1.03 1.19 0.70 0.96 0.62 60 0.93 0.76 1.13 89 17.5
South East Essex 345,600 1.03 0.91 0.62 0.78 0.79 0.81 95 0.82 0.71 0.94 95 5.7
South West Essex 407,100 0.92 1.11 0.69 0.85 1.02 1.11 115 0.95 0.84 1.08 97 9.8
Suffolk 614,800 0.93 0.72 0.86 0.74 0.63 0.88 104 0.79 0.71 0.88 92 5.3
West Essex 289,600 0.73 0.48 0.79 0.67 0.75 1.21 135 0.77 0.66 0.91 85 8.1
London Barking and Dagenham 187,000 1.15 1.56 1.48 1.45 1.67 2.20 176 1.59 1.35 1.87 126 41.7
Barnet 357,500 1.92 1.40 1.35 1.75 1.46 1.58 148 1.57 1.41 1.76 146 35.9
Bexley 232,800 1.09 1.17 1.28 1.39 1.19 0.86 90 1.16 1.00 1.35 120 18.1
Brent Teaching 312,200 1.99 1.92 2.17 2.72 2.19 2.49 215 2.25 2.03 2.49 192 63.7
Bromley 310,600 0.73 1.28 0.98 1.10 0.68 0.65 71 0.90 0.78 1.04 97 15.7
Camden 220,100 1.11 1.16 1.37 1.67 1.30 1.20 105 1.30 1.11 1.53 112 33.7
City and Hackney Teaching 254,600 1.35 1.24 1.68 1.67 1.87 2.04 149 1.64 1.42 1.90 119 44.6
Croydon 364,800 1.72 1.39 1.64 1.47 1.28 2.04 189 1.59 1.43 1.78 145 44.9
Ealing 339,300 1.95 1.54 2.27 2.05 1.85 2.26 197 1.99 1.79 2.21 172 51.0
Enfield 313,900 1.14 1.40 1.31 1.41 2.00 1.65 150 1.49 1.31 1.68 133 39.0
Greenwich Teaching 255,500 1.47 1.66 1.23 2.08 1.08 1.36 114 1.48 1.28 1.71 122 37.5
Hammersmith and Fulham 182,400 1.58 0.62 1.30 1.55 1.35 1.57 126 1.33 1.10 1.59 106 31.9
Haringey Teaching 255,500 1.13 1.58 1.08 1.41 1.90 2.39 192 1.59 1.38 1.83 126 39.5
Harrow 240,500 0.52 1.68 1.99 2.17 2.27 1.51 150 1.69 1.49 1.92 165 57.8
Havering 237,900 0.69 0.81 0.61 0.39 1.21 1.05 118 0.80 0.67 0.95 88 12.3
Hillingdon 275,500 0.91 1.46 1.33 1.40 1.59 1.47 138 1.36 1.19 1.56 126 39.4
Hounslow 254,900 1.47 1.19 1.59 1.92 1.85 1.85 161 1.64 1.44 1.88 141 48.6
Islington 206,300 1.22 0.92 1.59 1.50 1.63 2.31 184 1.53 1.30 1.79 121 31.8
Kensington and Chelsea 158,300 0.54 1.28 0.87 1.17 0.93 0.79 76 0.93 0.75 1.15 87 29.4
Kingston 160,400 0.88 1.49 0.74 0.89 1.06 1.13 106 1.03 0.84 1.27 96 25.5
Lambeth 304,500 1.95 1.61 1.96 1.52 1.85 1.82 138 1.78 1.57 2.02 135 42.9
Lewisham 276,900 1.83 1.61 2.31 1.46 1.90 1.99 162 1.85 1.64 2.10 150 46.5
Newham 310,500 1.65 1.78 2.03 2.52 2.27 2.02 139 2.05 1.81 2.31 140 71.0
Redbridge 281,400 1.38 1.54 1.81 1.56 1.39 2.15 192 1.64 1.45 1.86 145 57.5
Richmond and Twickenham 187,500 0.77 0.77 0.81 0.89 0.70 0.81 80 0.79 0.64 0.98 77 14.0
Southwark 288,700 2.33 2.10 1.51 1.87 2.03 1.86 142 1.95 1.72 2.20 148 45.8
Sutton and Merton 391,700 1.23 1.47 1.27 1.36 1.45 1.63 156 1.40 1.25 1.57 132 28.4
Tower Hamlets 256,000 1.77 2.00 1.90 1.46 1.81 2.02 133 1.83 1.58 2.11 120 54.8
Waltham Forest 259,700 2.41 1.32 1.64 1.15 1.86 1.17 96 1.59 1.39 1.83 130 47.8
Wandsworth 307,700 1.69 1.61 1.90 1.53 1.19 1.19 94 1.52 1.33 1.73 119 28.6
Westminster 219,600 0.71 1.46 1.71 1.29 1.49 1.35 123 1.34 1.14 1.56 121 38.3
South Brighton and Hove City 273,000 0.82 1.06 1.12 0.83 0.92 1.12 106 0.98 0.84 1.15 92 10.9
East East Sussex Downs and Weald 343,900 0.89 0.65 0.62 0.61 0.75 1.04 134 0.76 0.66 0.87 96 3.8
Coast
Eastern and Coastal Kent 759,600 1.31 1.19 1.04 1.04 0.90 0.88 103 1.06 0.97 1.15 121 5.0
Hastings and Rother 183,400 0.61 0.92 0.68 0.74 1.02 0.80 104 0.80 0.66 0.96 101 4.5
Medway 264,900 1.42 0.65 0.99 0.82 0.87 0.79 79 0.92 0.79 1.09 91 10.4
Surrey 1,124,800 0.80 0.93 0.97 1.04 0.97 0.97 108 0.95 0.88 1.02 104 9.5
West Kent 706,800 1.00 1.02 0.98 0.82 0.89 0.75 82 0.91 0.83 1.00 98 7.7
West Sussex 808,900 0.85 0.87 0.77 0.76 0.67 0.72 88 0.77 0.70 0.85 92 6.2
14

2012 2007–2012
Crude Crude %
South Berkshire East 410,100 1.34 1.23 1.32 1.25 1.36 0.85 80 1.22 1.09 1.38 115 26.6
Central Berkshire West 464,400 0.89 1.11 0.84 0.75 1.05 0.76 78 0.90 0.79 1.02 90 14.0
Buckinghamshire 521,000 0.77 0.84 0.93 0.75 0.79 0.75 83 0.80 0.71 0.90 87 13.3
Hampshire 1,322,100 0.77 0.83 0.83 0.76 0.74 0.69 81 0.77 0.71 0.83 88 5.0
Isle of Wight National Health Service 138,400 0.22 0.34 0.16 0.62 0.82 0.87 116 0.51 0.39 0.66 66 2.7
Milton Keynes 255,400 1.18 1.00 1.00 1.11 0.99 1.19 110 1.08 0.92 1.27 97 19.6
Oxfordshire 629,600 0.74 0.68 1.03 0.93 1.04 0.99 105 0.90 0.81 1.00 94 9.4
Portsmouth City Teaching 205,400 0.80 0.90 0.74 0.59 1.30 1.10 102 0.91 0.75 1.10 84 11.6
Southampton City 235,900 0.85 1.22 0.60 1.23 1.14 0.88 81 0.99 0.83 1.17 90 14.1
South Bath and North East Somerset 175,500 0.94 0.73 1.38 0.63 0.56 0.96 108 0.87 0.71 1.05 96 5.4
West Bournemouth and Poole Teaching 331,500 0.68 0.84 0.53 0.54 0.74 0.79 90 0.69 0.59 0.81 77 6.3
Bristol 428,100 1.05 1.56 1.19 1.45 1.38 1.26 117 1.31 1.18 1.47 121 16.0
Cornwall and Isles of Scilly 536,000 0.98 0.89 1.09 0.89 0.79 0.96 123 0.93 0.84 1.03 117 1.8
Devon 747,700 1.07 1.13 1.01 0.93 0.89 0.99 128 1.00 0.92 1.09 128 2.5
Dorset 413,800 0.72 0.92 0.69 0.61 0.70 0.65 89 0.71 0.63 0.81 97 2.1
Gloucestershire 598,300 0.88 0.68 1.13 0.87 0.92 1.17 137 0.94 0.85 1.04 108 4.6
North Somerset 203,100 0.82 1.19 0.88 0.99 0.84 0.99 123 0.95 0.81 1.12 116 2.7
Plymouth Teaching 256,600 1.73 1.05 1.15 1.29 1.10 0.95 101 1.21 1.05 1.39 127 3.9
Somerset 531,600 0.73 0.75 1.11 1.07 0.85 0.69 87 0.87 0.78 0.96 106 2.0
South Gloucestershire 263,400 0.88 0.98 0.69 1.17 0.58 0.82 91 0.85 0.73 1.00 93 5.0
Swindon 214,900 0.61 1.08 1.07 1.00 1.16 1.29 130 1.04 0.87 1.23 103 10.0
Torbay 131,200 0.90 1.62 0.70 1.50 0.87 1.10 145 1.11 0.92 1.34 144 2.5
Wiltshire 474,300 0.62 0.85 0.74 0.83 0.63 0.49 57 0.69 0.61 0.79 78 3.4
Wales Betsi Cadwaladr University 688,700 1.11 0.93 0.94 1.00 0.81 0.99 121 0.96 0.88 1.05 115 2.5
Powys Teaching 133,200 0.99 0.93 1.03 0.64 1.25 1.24 165 1.02 0.84 1.23 133 1.6
Hywel Dda 381,900 1.10 1.27 0.80 1.12 1.20 0.86 107 1.06 0.94 1.18 130 2.2
Abertawe Bro Morgannwg Univ. 517,700 1.51 1.20 1.52 1.47 1.14 1.35 155 1.36 1.25 1.49 153 3.9
Cwm Taf 293,500 1.61 1.07 1.31 0.99 1.45 0.86 95 1.21 1.07 1.38 132 2.6
Aneurin Bevan 577,000 1.34 0.95 0.95 1.30 1.17 1.16 132 1.14 1.04 1.26 127 3.9
Cardiff and Vale University 472,300 1.46 1.00 1.14 1.36 1.00 1.05 104 1.17 1.05 1.30 114 12.2
Scotland Ayrshire & Arran 373,800 0.85 0.82 0.88 1.08 0.81 0.89 107 0.89 0.78 1.01 105 0.7
Borders 113,900 1.20 1.13 0.97 1.06 0.55 0.48 61 0.89 0.71 1.12 113 0.6
Dumfries and Galloway 151,400 0.83 1.14 1.07 0.63 0.56 1.06 139 0.88 0.73 1.07 113 0.7
Fife 365,300 1.00 0.96 1.21 1.19 1.15 0.86 99 1.06 0.94 1.20 120 1.3
Forth Valley 298,100 1.33 0.77 1.07 1.03 0.79 0.84 94 0.97 0.84 1.12 106 1.1
Grampian 569,600 0.84 0.87 0.88 0.85 0.82 0.85 93 0.85 0.76 0.95 92 1.6
Greater Glasgow & Clyde 1,214,600 1.08 0.95 1.00 0.87 1.04 1.10 117 1.01 0.94 1.08 105 3.4
Highland 321,700 0.86 0.83 0.72 0.60 0.48 0.53 65 0.67 0.57 0.78 81 0.8
Lanarkshire 572,400 0.80 0.77 0.84 0.96 0.84 1.15 126 0.89 0.80 1.00 96 1.2
Lothian 836,600 0.88 0.97 0.85 0.62 0.72 0.73 75 0.79 0.72 0.87 81 2.8
Orkney 21,400 0.38 1.54 1.14 0.39 0.00 1.86 233 0.89 0.53 1.50 109 0.4
Shetland 23,200 1.58 0.00 0.39 0.40 0.78 0.00 0 0.52 0.26 1.04 57 1.1
Tayside 410,300 1.26 1.17 1.28 0.98 1.14 0.72 85 1.09 0.98 1.22 126 1.9
Western Isles 27,700 1.72 0.29 0.85 1.73 0.00 0.00 0 0.76 0.46 1.24 96 0.6
N Ireland Belfast 348,300 1.27 1.01 0.76 1.25 1.10 1.68 167 1.18 1.04 1.34 115 3.2
Northern 463,500 1.39 1.14 0.78 1.19 1.22 1.15 119 1.15 1.03 1.28 116 1.2
Southern 359,400 0.60 0.96 0.77 1.07 1.33 0.76 72 0.92 0.79 1.06 85 1.2
South Eastern 347,700 0.92 0.87 0.69 0.71 0.96 0.79 83 0.82 0.71 0.95 85 1.3
Western 295,300 1.04 0.90 1.21 0.84 1.10 0.56 54 0.94 0.81 1.10 89 1.0
15
2.8 ethnic minority populations in general compared with

the White population (5.1% of ethnic minorities were
2.4
over 65 years old compared to 16.9% of Whites) [3].
The median age of new patients with diabetes was similar
2.0
to the overall median and has not varied greatly over the
Standardised ratio
North East
1.6
North West last five years.
Yorkshire and the Humber
East Midlands There were large differences between centres in the
West Midlands
1.2 East of England
median age of incident patients (figure 1.6) reflecting
London
South East Coast
differences in the age and ethnic structure of the catch-
0.8 South Central ment populations and also, particularly in smaller
South West
Wales centres, chance fluctuations. The median age of patients
0.4 Scotland
Northern Ireland
starting treatment at transplant centres was 63.1 years
0.0 (IQR 49.8, 73.6) and at non-transplanting centres 65.9
0 10 20 30 40 50 60 70 80 90 years (IQR 52.7, 75.2) (p , 0.0001).
% non-White
Averaged over 2007–2012, crude PCT/HB incidence
Fig. 1.2. Age/gender standardised incidence ratio (2007–2012) by rates in the over 75 years age group varied from 0 per
percentage non-White
million age related population (pmarp) (Shetland) to
904 pmarp (Heart of Birmingham) (data not shown).
Excluding four areas which had much higher or
Others. The details of regrouping of the PAS codes into the above lower rates than the rest, there was 5.4-fold variation
ethnic categories are provided in appendix H: Ethnicity and ERA- (124 pmarp to 673 pmarp). The wide range of treatment
EDTA Coding (www.renalreg.com). Chi-squared, Fisher’s exact,
ANOVA and Kruskal Wallis tests were used as appropriate to
rates suggests that there was geographical variation in the
test for significant differences. prevalence of comorbid and predisposing renal con-
Estimated glomerular filtration rate (eGFR) at the start of RRT ditions as well as uncertainty within the renal community
was studied amongst patients with eGFR data within 14 days about the suitability of older patients for dialysis. The
before the start of RRT. The eGFR was calculated using the 5.4-fold variation between PCT/HBs seen in the over 75s
abbreviated 4 variable MDRD study equation [2]. For the purpose
of the eGFR calculation, patients who had missing ethnicity but a
was much greater than the 2.6-fold variation (66 pmp
valid serum creatinine measurement were classed as Whites. The to 172 pmp) seen in the overall analysis although a
eGFR values were log transformed in order to normalise the data. proportion of this difference is likely to be due to the
smaller numbers included in the over 75 analysis.
Results
Age Gender
Overall, incidence rates have levelled off in the last As in previous years, more men than women started
seven years (figure 1.3). Figure 1.4 shows RRT incidence RRT with 62.1% of new starters being male. This was a
rates for 2012 by age group and gender. For women, the slight fall from the 63.0% seen for 2011 and negates
peak rate was in the 75–79 age group and in men in the some of the increase seen in 2010 and 2011. The male
80–84 age group. Regarding numbers starting RRT percentage was above 50 for all age groups and above
(rather than rates), figure 1.5 shows that the 65–74 age 60 for over 55s (figure 1.7).
group contained the most patients starting on both HD
and PD. The pattern seen in this graph is very similar Ethnicity
to the pattern for 2011. As in previous reports, Scotland is not included in
In 2012, the median age of patients starting renal this section as ethnicity completeness was low. Across
replacement therapy was 64.6 years (table 1.4) and this English, Welsh and Northern Irish centres the average
has changed little over the last six years (data not completeness improved further this year up to 97.0%
shown). The median age at start was 66.9 years for (vs 92.9% for 2011). A large part of the improvement
patients starting on HD, 60.5 for patients starting on was due to three centres (Brighton, Reading, Liverpool
PD and 48.6 for those having a pre-emptive transplant RI) which improved from having data for 3%, 30% and
(table 1.5). The median age of non-White patients (57.8 40% of patients respectively to having data for 80% or
years) was considerably lower than for White patients more. Indeed, completeness was 80% or more for all
(66.1 years) reflecting the younger age distribution of centres for 2012 (table 1.6) and was over 90% for all
16
Table 1.3. Number of patients starting RRT by renal centre 2007–2012
Year Catchment 2012

population crude rate
Centre 2007 2008 2009 2010 2011 2012 (millions) pmp (95% CI)
England
B Heart 101 105 99 95 113 101 0.74 137 (110–164)
B QEH 222 268 255 197 215 216 1.70 127 (110–144)
Basldn 39 40 27 32 42 53 0.42 128 (93–162)
Bradfd 88 62 59 67 60 71 0.65 109 (84–134)
Brightn 120 119 117 106 119 136 1.30 105 (87–122)
Bristol 153 175 157 169 139 148 1.44 103 (86–119)
Camb 125 94 134 106 122 124 1.16 107 (88–126)
Carlis 26 30 28 22 28 19 0.32 59 (33–86)
Carsh 191 210 204 216 207 242 1.91 127 (111–142)
Chelms 51 36 51 45 47 45 0.51 88 (62–114)
Colchr n/a 58 21 32 44 29 0.30 97 (62–132)
Covnt 110 113 116 114 111 112 0.89 126 (102–149)
Derby 62 97 77 79 80 81 0.70 115 (90–140)
Donc 20 26 40 45 43 40 0.41 98 (67–128)
Dorset 62 82 74 71 79 72 0.86 84 (64–103)
Dudley 40 46 69 43 43 56 0.44 127 (94–160)
Exeter 126 135 145 139 112 138 1.09 127 (106–148)
Glouc 59 46 79 61 58 74 0.59 126 (97–155)
Hull 99 110 99 87 109 97 1.02 95 (76–114)
Ipswi 40 38 38 33 29 43 0.40 108 (76–140)
Kent 171 139 128 134 122 115 1.22 94 (77–111)
L Barts 215 206 237 203 249 263 1.83 144 (126–161)
L Guys 167 161 172 143 120 127 1.08 117 (97–138)
L Kings 122 151 126 144 140 125 1.17 107 (88–125)
L Rfree 185 172 169 204 223 240 1.52 158 (138–178)
L St.G 90 99 110 86 74 91 0.80 114 (91–137)
L West 273 317 357 365 365 352 2.40 147 (131–162)
Leeds 124 158 153 126 158 154 1.67 92 (78–107)
Leic 244 242 228 246 267 235 2.44 96 (84–109)
Liv Ain 34 42 38 50 61 63 0.48 130 (98–162)
Liv RI 112 102 110 99 114 110 1.00 110 (89–131)
M RI 159 131 146 161 156 160 1.53 104 (88–121)
Middlbr 100 95 96 101 100 120 1.00 120 (98–141)
Newc 106 99 97 91 98 104 1.12 93 (75–111)
Norwch 111 84 72 86 87 74 0.79 94 (73–116)
Nottm 129 115 133 116 116 99 1.09 91 (73–109)
Oxford 143 148 174 165 177 171 1.69 101 (86–116)
Plymtha 76 69 57 56 60 75 0.47 160 (124–196)
Ports 157 170 149 149 187 161 2.02 80 (67–92)
Prestn 132 113 146 124 140 147 1.49 98 (83–114)
Redng 92 103 94 89 103 73 0.91 80 (62–99)
Salford 110 139 125 149 126 134 1.49 90 (75–105)
Sheff 165 180 149 143 135 158 1.37 115 (97–133)
Shrew 58 59 48 58 61 57 0.50 114 (84–143)
Stevng 88 102 98 107 110 110 1.20 91 (74–108)
Sthend 34 36 23 28 29 26 0.32 82 (51–114)
Stoke 87 80 110 95 93 77 0.89 87 (67–106)
Sund 62 45 64 54 57 71 0.62 115 (88–142)
Truro 45 41 58 46 38 50 0.41 121 (87–155)
Wirral 53 39 63 62 62 50 0.57 87 (63–112)
Wolve 68 89 65 106 76 84 0.67 126 (99–152)
York 37 36 44 38 52 53 0.49 108 (79–137)
17
Year Catchment 2012

Centre 2007 2008 2009 2010 2011 2012 (millions) pmp (95% CI)
N Ireland
Antrim 37 41 21 41 30 26 0.30 87 (53–120)
Belfast 90 70 58 72 69 91 0.55 165 (131–199)
Newry 15 21 19 21 38 18 0.28 64 (35–94)
Ulster 18 14 13 20 35 30 0.30 100 (64–136)
West NI 29 31 37 26 38 21 0.35 60 (34–86)
Scotland
Abrdn 56 56 55 51 50 54 0.60 90 (66–114)
Airdrie 48 39 48 56 48 61 0.56 109 (82–136)
D & Gall 17 19 17 10 10 19 0.15 127 (70–184)
Dundee 62 64 69 50 58 41 0.41 100 (69–131)
Dunfn 37 30 33 45 43 29 0.37 78 (50–107)
Edinb 95 103 98 68 75 76 0.96 79 (61–97)
Glasgw 187 159 174 153 177 186 1.51 123 (105–141)
Inverns 26 25 21 27 12 13 0.34 38 (17–59)
Klmarnk 36 33 39 43 33 40 0.37 108 (75–142)
Wales
Bangor 36 40 30 26 20 21 0.22 96 (55–137)
Cardff 220 150 177 186 186 170 1.42 120 (102–138)
Clwyd 21 15 25 21 17 22 0.19 116 (68–164)
Swanse 128 125 116 135 118 113 0.89 128 (104–151)
Wrexm 27 21 19 25 26 34 0.24 142 (94–189)
% change
since 2007
England 5,483 5,652 5,728 5,583 5,756 5,826 6.3
N Ireland 189 177 148 180 210 186 −1.6
Scotland 564 528 554 503 506 519 −8.0
Wales 432 351 367 393 367 360 −16.7
UK 6,668 6,708 6,797 6,659 6,839 6,891 3.3
n/a – renal centre not yet operational
a
Plymouth had 75 incident patients in 2012 but only 47 of these were included in the data extract. The extra 28 patients have been included in
tables 1.1 and 1.3 but not in the remainder of this chapter. The estimated catchment population may be too low and hence the rate too high due
to the missing patients (an incident cohort 2008–2012 was used for this work)
400 700
0–44 Males
350 45–64 All
600 Females
65+
300 Total
500
250
400
200
300
150
200
100
50 100
0 0
20–24
25–29
30–34
35–39
40–44
45–49
50–54
55–59
60–64
65–69
70–74
75–79
80–84
85+
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
Year Age group (years)
Fig. 1.3. RRT incidence rates between 1980 and 2012 Fig. 1.4. RRT incidence rates in 2012 by age and gender
18
Table 1.4. Median, inter-quartile range and 90% range of the age Table 1.5. Median, inter-quartile range and 90% range of the age
of patients starting renal replacement therapy in 2012 by country of patients starting renal replacement therapy in 2012 by initial
treatment modality
Country Median IQR 90% range
Treatment Median IQR 90% range
England 64.5 (51.0–74.6) (31.2–83.7)
N Ireland 68.2 (52.0–76.0) (33.3–85.4) HD 66.9 (54.8–76.0) (34.7–84.4)
Scotland 63.9 (51.9–73.3) (35.2–82.7) PD 60.5 (47.0–71.2) (29.1–82.0)
Wales 67.1 (53.6–75.8) (34.1–83.8) Transplant 48.6 (38.4–58.3) (24.2–68.8)
UK 64.6 (51.3–74.5) (31.6–83.6)
but six centres. There was great variation between centres Primary renal diagnosis
in the percentage of incident patients who were non- The breakdown of primary renal disease (PRD) by
White ranging from zero in Antrim, Bangor, Colchester, centre is shown in table 1.7. The information was missing
Newry, Truro and Wrexham to over 50% in St Bartholo- for 6.3% of patients. Sixty-one centres provided data on
mew’s and London West. over 90% of incident patients and 33 of these centres
had 100% completeness. There was only a small amount
of missing data for Wales and none for Scotland, whilst
1,500 England had 7.4% missing (down from 12.0% for 2011)
HD
PD and Northern Ireland, 2.7% missing. The overall per-
1,250
centage missing was down on 2011 (6.3% from 10.2%)
and was slightly lower in under rather than over 65
Number of patients
1,000
year olds (5.3% and 7.3% respectively). As for 2011,
750 four centres had missing PRD for more than 25% of
incident patients and for these centres the percentages
500 in the diagnostic categories are not shown in table 1.7.
The UKRR continues to be concerned about centres
250
with apparently very high data completeness for PRD
0 but also very high rates of ‘uncertain’ diagnoses (EDTA
18–24 25–34 35–44 45–54 55–64 65–74 75–84 85+ code 00: Chronic renal failure; aetiology uncertain). It is
Age group (years) accepted that there will inevitably be a number of patients
Fig. 1.5. Number of incident dialysis patients in 2012, by age with uncertain aetiology and that the proportion of these
group and initial dialysis modality patients will vary between clinicians and centres as the
90
Upper quartile
Median
80 Lower quartile
70
Age (years)
60
50
40
Sthend
Ulster
Colchr
Truro
Antrim
Exeter
Newry
Covnt
Clwyd
Dorset
Dundee
Bangor
Glouc
Swanse
Brightn
Norwch
Dunfn
West NI
Inverns
Shrew
Wrexm
Sund
Liv Ain
Wirral
Chelms
Basldn
Carsh
Redng
Camb
D&Gall
L St.G
B Heart
Kent
Donc
Middlbr
Derby
Airdrie
Leic
Salford
Bristol
Sheff
Abrdn
Dudley
Hull
Cardff
Stoke
Klmarnk
Stevng
Ports
B QEH
Oxford
M RI
Leeds
Nottm
Prestn
Belfast
York
L West
L Rfree
Bradfd
Glasgw
L Kings
Wolve
Ipswi
Newc
Edinb
Liv RI
Carlis
L Guys
L Barts
England
N Ireland
Scotland
Wales
UK
Centre
Fig. 1.6. Median age of incident RRT patients by centre in 2012

White points indicate transplant centres
19
80 diagnoses (partly due to the reasons mentioned above).

For example, the percentage with diabetes as PRD varied
from about 10% to 44% of incident patients. The overall
Percentage male (95% CI)
70
percentage with uncertain aetiology was lower than last
year (15.9% vs. 17.3%). There were increases in the
60 percentages with diabetes, glomerulonephritis, hyper-
tension and ‘other’ and decreases in the percentages
with polycystic kidney disease, pyelonephritis and renal
50
vascular disease.
The overall UK distribution of PRDs is shown in
40 table 1.8. Diabetic nephropathy was the most common
18–24 25–34 35–44 45–54 55–64 65–74 75–84 85+
renal diagnosis in both the under and over 65 year age
Age group (years)
groups, accounting for 26% of all (non-missing) incident
Fig. 1.7. Percentage of patients starting RRT in 2012 who were diagnoses. Glomerulonephritis and autosomal dominant
male, by age group
polycystic kidney disease (ADPKD) made up higher
proportions of the younger than the older incident
definitions of e.g. renal vascular disease and hypertensive cohorts (17% vs. 10% and 10% vs. 3% respectively), whilst
renal disease remain relatively subjective. There was patients with renal vascular disease comprised a much
again a lot of variability between centres but, as in higher percentage of the older rather than the younger
previous years, a small number of centres had far higher patients (11% vs. 2%). Uncertainty about the underlying
percentages with ‘uncertain’ diagnosis than other centres. diagnosis was also much more likely in the older rather
This year, there were two centres with diagnosis than the younger cohort (20% vs. 12%).
‘uncertain’ for over 50% of their incident patients – For all primary renal diagnoses except ADPKD, the
Cambridge (68%) and Ipswich (65%). As the numbers male to female ratio was 1.3 or greater. This gender
with the specific PRDs are likely to be falsely low in difference may relate to factors such as smoking, hyper-
these centres, the breakdown into these categories has tension, atheroma and renal vascular disease which are
not been shown in table 1.7 or been used in the country more common in males and may influence the rate of
and UK averages. These centres have also been excluded progression of renal failure.
where PRD is used to stratify analyses. Table 1.9 shows the incidence rates for each PRD per
As in previous years, there was a lot of variability million population for the 2012 cohort. The incidence of
between centres in the percentages with the specific RRT due to diabetes as PRD was somewhat higher in
Table 1.6. Percentage of incident RRT patients (2012) in different ethnic groups by centre
Percentage in each ethnic group

% data not N with
Centre available data White South Asian Black Chinese Other
England
B Heart 0.0 101 70.3 24.8 5.0
B QEH 0.0 216 70.8 22.7 5.1 1.4
Basldn 0.0 53 79.2 3.8 11.3 5.7
Bradfd 0.0 71 57.7 42.3
Brightn 5.1 129 91.5 3.1 3.9 1.6
Bristol 4.1 142 90.8 4.9 4.2
Camb 0.8 123 96.7 0.8 0.8 0.8 0.8
Carlis 0.0 19 94.7 5.3
Carsh 14.1 208 72.6 13.0 10.1 0.5 3.8
Chelms 20.0 36 97.2 2.8
Colchr 0.0 29 100.0
Covnt 0.9 111 83.8 12.6 2.7 0.9
Derby 6.2 76 81.6 13.2 2.6 2.6
Donc 0.0 40 95.0 5.0
Dorset 0.0 72 98.6 1.4
20

% data not N with
Centre available data White South Asian Black Chinese Other
Dudley 0.0 56 85.7 10.7 3.6
Exeter 0.7 137 97.1 0.7 2.2
Glouc 0.0 74 95.9 2.7 1.4
Hull 4.1 93 96.8 3.2
Ipswi 9.3 39 97.4 2.6
Kent 5.2 109 95.4 1.8 2.8
L Barts 0.0 263 35.7 26.6 36.5 0.4 0.8
L Guys 3.2 123 62.6 6.5 23.6 0.8 6.5
L Kings 0.8 124 55.6 11.3 29.8 3.2
L Rfree 12.1 211 50.2 13.7 23.7 1.9 10.4
L St.G 11.0 81 56.8 19.8 16.0 1.2 6.2
L West 0.0 352 41.5 40.6 17.6 0.3
Leeds 0.6 153 83.7 11.1 4.6 0.7
Leic 2.6 229 79.5 16.6 2.2 1.7
Liv Ain 0.0 63 95.2 3.2 1.6
Liv RI 4.5 105 94.3 1.9 1.9 1.9
M RI 0.0 160 75.6 10.6 10.0 3.8
Middlbr 0.8 119 95.0 5.0
Newc 1.9 102 92.2 6.9 1.0
Norwch 5.4 70 87.1 12.9
Nottm 0.0 99 83.8 10.1 4.0 2.0
Oxford 0.0 171 78.9 10.5 4.1 6.4
Plymth 2.1 46 97.8 2.2
Ports 5.6 152 94.1 3.3 1.3 1.3
Prestn 0.0 147 88.4 10.2 1.4
Redng 19.2 59 72.9 16.9 6.8 1.7 1.7
Salford 10.4 120 82.5 15.8 0.8 0.8
Sheff 1.9 155 86.5 5.8 5.2 2.6
Shrew 3.5 55 96.4 1.8 1.8
Stevng 1.8 108 70.4 15.7 8.3 0.9 4.6
Sthend 3.8 25 96.0 4.0
Stoke 3.9 74 93.2 2.7 4.1
Sund 1.4 70 95.7 4.3
Truro 0.0 50 100.0
Wirral 2.0 49 98.0 2.0
Wolve 0.0 84 70.2 23.8 6.0
York 0.0 53 96.2 1.9 1.9
N Ireland 1.6
Antrim 0.0 26 100.0
Belfast 0.0 91 94.5 1.1 3.3 1.1
Newry 0.0 18 100.0
Ulster 0.0 30 96.7 3.3
West NI 0.0 21 95.2 4.8
Wales 2.2 0.3
Bangor 0.0 21 100.0
Cardff 0.0 170 95.3 3.5 0.6 0.6
Clwyd 0.0 22 90.9 9.1
Swanse 0.0 113 99.1 0.9
Wrexm 0.0 34 100.0
England 3.3 5,606 77.8 12.2 7.6 0.6 1.8
N Ireland 0.0 186 96.2 1.6 1.6 0.5
Wales 0.0 360 96.9 2.2 0.3 0.3 0.3
E, W & NI 3.0 6,152 79.4 11.3 7.0 0.6 1.7
Blank cells – no reported patients
21
Table 1.7. Distribution of primary renal diagnosis by centre in the 2012 incident RRT cohort
Percentage
% N Renal
data not with Uncertain Glomerulo- Hyper- Polycystic Pyelo- vascular
Centre available data aetiology Diabetes nephritis tension Other kidney nephritis disease
England
B Heart 6.9 94 20.2 34.0 7.5 7.5 16.0 4.3 6.4 4.3
B QEH 0.5 215 10.7 20.9 14.0 3.7 23.3 5.6 8.4 13.5
Basldn 11.3 47 2.1 27.7 21.3 14.9 12.8 6.4 4.3 10.6
Bradfd 1.4 70 24.3 27.1 17.1 7.1 10.0 8.6 1.4 4.3
Brightn 2.2 133 24.8 18.1 12.0 1.5 19.6 9.0 10.5 4.5
Bristol 15.5 125 13.6 23.2 17.6 4.0 18.4 8.0 9.6 5.6
Camba 0.0 124 67.7
Carlis 0.0 19 5.3 15.8 42.1 5.3 0.0 10.5 15.8 5.3
Carsh 22.7 187 24.1 19.8 10.2 6.4 18.2 9.1 7.5 4.8
Chelms 2.2 44 25.0 34.1 13.6 6.8 11.4 2.3 2.3 4.6
Colchr 2.2 29 44.8 24.1 3.5 3.5 10.3 3.5 6.9 3.5
Covnt 1.8 110 12.7 21.8 10.0 11.8 18.2 4.6 7.3 13.6
Derby 2.5 79 12.7 31.7 17.7 1.3 15.2 7.6 6.3 7.6
Donc 2.5 39 28.2 23.1 10.3 10.3 18.0 2.6 0.0 7.7
Dorset 0.0 72 5.6 23.6 11.1 9.7 26.4 9.7 8.3 5.6
Dudley 1.8 55 25.5 14.6 3.6 5.5 38.2 9.1 0.0 3.6
Exeter 0.0 138 8.7 26.1 15.9 8.7 16.7 5.1 7.3 11.6
Glouc 0.0 74 27.0 16.2 14.9 2.7 16.2 5.4 13.5 4.1
Hull 0.0 97 23.7 23.7 14.4 7.2 14.4 11.3 5.2 0.0
Ipswia 0.0 43 65.1
Kent 0.0 115 25.2 17.4 15.7 3.5 15.7 2.6 15.7 4.4
L Barts 6.5 246 14.2 31.3 11.8 14.6 15.5 4.5 6.5 1.6
L Guys 13.4 110 14.6 28.2 12.7 5.5 12.7 11.8 11.8 2.7
L Kings 0.0 125 13.6 39.2 13.6 12.0 8.8 7.2 4.8 0.8
L Rfree 0.4 239 6.7 26.8 15.1 11.3 27.2 3.4 2.5 7.1
L St.G 18.7 74 28.4 21.6 13.5 9.5 17.6 4.1 4.1 1.4
L West 0.3 351 13.7 35.3 14.5 2.9 18.8 5.4 4.0 5.4
Leeds 1.3 152 10.5 16.5 15.1 11.8 21.7 9.2 8.6 6.6
Leic 16.2 197 21.8 19.3 13.2 6.6 14.7 11.2 7.6 5.6
Liv Ain 0.0 63 22.2 17.5 17.5 14.3 11.1 3.2 6.4 7.9
Liv RI 0.0 90 10.0 20.0 11.1 21.1 23.3 6.7 7.8 0.0
M RI 3.8 154 15.6 29.2 9.1 17.5 14.3 7.1 4.6 2.6
Middlbr 1.7 118 19.5 24.6 11.0 3.4 18.6 6.8 6.8 9.3
Newc 1.9 102 16.7 19.6 23.5 4.9 19.6 3.9 4.9 6.9
Norwch 8.1 68 29.4 17.7 16.2 5.9 17.7 5.9 7.4 0.0
Nottm 0.0 99 13.1 26.3 15.2 4.0 23.2 6.1 5.1 7.1
Oxford 0.0 171 15.8 31.0 14.6 7.6 11.1 5.9 7.0 7.0
Plymthb 27.7 34
Ports 1.9 158 8.9 23.4 11.4 10.1 19.0 11.4 7.0 8.9
Prestn 1.4 145 13.1 28.3 13.1 11.0 13.8 6.2 7.6 6.9
Redng 2.7 71 11.3 33.8 12.7 5.6 18.3 4.2 8.5 5.6
Salfordb 78.4 29
Sheff 0.6 157 15.9 33.8 19.1 3.8 6.4 4.5 8.3 8.3
Shrewb 31.6 39
Stevng 0.0 110 14.6 15.5 6.4 1.8 52.7 4.6 3.6 0.9
Sthend 0.0 26 3.9 15.4 23.1 0.0 23.1 7.7 3.9 23.1
Stoke 6.5 72 8.3 27.8 18.1 9.7 15.3 13.9 5.6 1.4
Sund 0.0 71 5.6 23.9 5.6 22.5 16.9 11.3 5.6 8.5
Truro 8.0 46 10.9 10.9 23.9 15.2 19.6 4.4 6.5 8.7
Wirralb 62.0 19
Wolve 0.0 84 31.0 22.6 13.1 2.4 16.7 8.3 3.6 2.4
York 1.9 52 5.8 21.2 21.2 1.9 21.2 11.5 9.6 7.7
22
Percentage
% N Renal
data not with Uncertain Glomerulo- Hyper- Polycystic Pyelo- vascular
Centre available data aetiology Diabetes nephritis tension Other kidney nephritis disease
N Ireland
Antrim 0.0 26 42.3 30.8 7.7 3.9 11.5 0.0 3.9 0.0
Belfast 4.4 87 14.9 18.4 14.9 3.5 20.7 6.9 17.2 3.5
Newry 0.0 18 5.6 44.4 11.1 5.6 5.6 5.6 5.6 16.7
Ulster 0.0 30 10.0 20.0 10.0 30.0 10.0 3.3 3.3 13.3
West NI 4.8 20 5.0 15.0 20.0 15.0 30.0 0.0 10.0 5.0
Scotland
Abrdn 0.0 54 9.3 25.9 13.0 11.1 20.4 7.4 7.4 5.6
Airdrie 0.0 61 23.0 29.5 18.0 1.6 4.9 6.6 8.2 8.2
D & Gall 0.0 19 10.5 42.1 10.5 5.3 15.8 5.3 5.3 5.3
Dundee 0.0 41 17.1 14.6 26.8 2.4 24.4 4.9 4.9 4.9
Dunfn 0.0 29 20.7 31.0 10.3 6.9 17.2 0.0 6.9 6.9
Edinb 0.0 76 15.8 30.3 13.2 2.6 19.7 9.2 5.3 4.0
Glasgw 0.0 186 14.5 28.5 18.3 2.2 13.4 9.1 7.5 6.5
Inverns 0.0 13 46.2 15.4 7.7 0.0 7.7 15.4 0.0 7.7
Klmarnk 0.0 40 0.0 37.5 15.0 12.5 17.5 5.0 7.5 5.0
Wales
Bangor 0.0 21 9.5 38.1 19.1 9.5 14.3 0.0 0.0 9.5
Cardff 0.6 169 24.9 26.0 14.2 2.4 11.2 8.9 3.6 8.9
Clwyd 0.0 22 4.6 18.2 18.2 22.7 22.7 4.6 9.1 0.0
Swanse 0.0 113 15.9 29.2 14.2 4.4 18.6 2.7 3.5 11.5
Wrexm 0.0 34 11.8 26.5 14.7 0.0 20.6 8.8 5.9 11.8
England 7.4 5,381 15.7 25.3 13.7 7.9 18.1 6.7 6.7 5.9
N Ireland 2.7 181 16.0 22.7 13.3 9.4 17.1 4.4 11.1 6.1
Scotland 0.0 519 15.2 28.5 16.4 4.2 15.4 7.5 6.7 6.0
Wales 0.3 359 18.7 27.3 14.8 4.5 15.3 6.1 3.9 9.5
UK 6.3 6,440 15.9 25.6 14.0 7.4 17.7 6.7 6.6 6.1
The percentage in each category has been calculated after excluding those patients with data not available
a
For those centres judged to have high % uncertain aetiology, the percentages in the other diagnostic categories have not been calculated and
these centres have not been included in the country and UK averages
b
For those centres with .25% missing primary diagnoses, the percentages in the diagnostic categories have not been calculated
Wales than in the other countries. As there were some Table 1.8. Percentage distribution of primary renal diagnosis by
missing data, the rates for at least some of the diagnoses age in the 2012 incident RRT cohort
will be underestimates. Percentage with diagnosis
First established treatment modality Diagnosis Age ,65 Age 565 All patients
In 2012, the first treatment recorded, irrespective of
any later change, was haemodialysis in 73.0% of patients, Diabetes 28.6 22.3 25.6
Glomerulonephritis 17.3 10.4 14.0
peritoneal dialysis in 19.5% and pre-emptive transplant
Pyelonephritis 6.8 6.4 6.6
in 7.4%. The previous year on year fall in the proportion Hypertension 6.2 8.8 7.4
of patients starting on PD has now levelled off during the Polycystic kidney 10.1 3.1 6.7
last six years (table 1.10). The percentage having a pre- Renal vascular disease 1.7 10.9 6.1
emptive transplant has continued to rise. Table F.1.3 in Other 17.4 18.0 17.7
appendix F: Additional Data Tables for 2012 New and Uncertain aetiology 11.8 20.1 15.9
Existing Patients gives the treatment breakdown at start Percentages calculated after excluding those patients with data not
of RRT by centre. available
23
Table 1.9. Primary renal diagnosis RRT incidence rates (2012) per million population (unadjusted)
Diagnosis England N Ireland Scotland Wales UK
Diabetes 25.5 22.5 27.9 31.9 25.9

Glomerulonephritis 13.8 13.2 16.0 17.2 14.2
Pyelonephritis 6.7 11.0 6.6 4.6 6.7
Hypertension 8.0 9.3 4.1 5.2 7.5
Polycystic kidney 6.8 4.4 7.3 7.2 6.8
Renal vascular disease 6.0 6.0 5.8 11.1 6.2
Other 18.3 17.0 15.1 17.9 18.0
Uncertain aetiology 15.9 15.9 14.9 21.8 16.1
Data not available 8.1 2.7 0.0 0.3 6.8
All 109 102 98 117 108
The overall rates per country may be slightly different to those in table 1.1 as those centres whose PRD data has not been used have been excluded
from both the numerator and the denominator here
Many patients undergo a brief period of HD before percentages of those still receiving RRT at 90 days,
switches to other modalities are, or can be, considered. 71.0% were on HD, 20.2% on PD and 8.8% had received
Therefore, the established modality at 90 days is more a transplant. This small decrease for PD as a modality at
representative of the elective first modality and this 90 days (22.7%–20.2%) is similar in size to the increase
modality was used for the remainder of this section. for transplant patients (5.7%–8.8%) over the last 6 years.
For these analyses, the incident cohort from 1st October Figure 1.8 shows the modality breakdown with the
2011 to 30th September 2012 was used so that follow up HD patients further subdivided. Of those still on RRT
to 90 days was possible for all patients. By 90 days, 5.5% at 90 days, 43% were treated with hospital HD, 28%
of incident patients had died and a further 0.4% had with satellite HD, and only 0.2% were receiving home
stopped treatment, leaving 94.0% of the original cohort HD at this early stage.
still on RRT. Table 1.11 shows the percentages on each The percentage of incident patients who had died by
treatment modality at 90 days both as percentages of all 90 days varied considerably between centres (0% to
of those starting RRT and then of those still on treatment 23% although, as last year, the percentage was 12.5% or
at 90 days. Expressed as percentages of the whole incident less for all except one centre). Differences in the
cohort, 66.9% were on HD at 90 days, 19.0% were on PD definition of whether patients have acute or chronic
and 8.3% had received a transplant. Expressed as renal failure may be a factor in this apparent variation
along with possible differences in clinical practice.
The percentage of patients still on RRT at 90 days who
Table 1.10. Treatment at start and at 90 days by year of start had a functioning transplant at 90 days varied between
HD PD Transplant centres from 0% to 24%. The mean percentage of the
Start (%) (%) (%) incident cohort with a functioning transplant at 90 days
was significantly greater in transplanting compared to
Day 0 treatment non-transplanting centres (11.2% vs. 5.4%: p , 0.0001).
2007 74.7 20.5 4.8
2008 75.2 19.3 5.5 One possible reason could be that some patients trans-
2009 76.4 18.0 5.7 planted pre-emptively were attributed to the incident
2010 74.7 18.5 6.7 cohort of the transplanting centre rather than that of
2011 72.9 20.3 6.8 the referring centre (as mentioned earlier).
2012 73.1 19.5 7.4 Table 1.12 gives the HD/PD breakdown for those
Day 90 treatment
Oct 2006 to end Sept 2007 71.7 22.7 5.7 incident patients on dialysis at 90 days. The breakdown
Oct 2007 to end Sept 2008 72.0 21.5 6.5 is given by age group and overall. The percentage on
Oct 2008 to end Sept 2009 73.9 19.1 7.0 PD at 90 days was about 65% higher in patients aged
Oct 2009 to end Sept 2010 72.7 19.4 7.9 under 65 years than in older patients (27.6% vs.
Oct 2010 to end Sept 2011 71.0 20.5 8.5
Oct 2011 to end Sept 2012 71.0 20.2 8.8
16.7%). These percentages are similar to those for 2011.
There was a lot of variability in the percentage on PD
24
Table 1.11. RRT modality at 90 days by centre (incident cohort 1/10/2011 to 30/09/2012)
Status at 90 days of only those

Status at 90 days of all patients who started RRT (%) patients still on RRT (%)
Stopped
Centre N HD PD Tx treatment Died HD PD Tx
England
B Heart 105 78.1 17.1 1.0 0.0 3.8 81.2 17.8 1.0
B QEH 225 72.0 17.8 8.4 0.0 1.8 73.3 18.1 8.6
Basldn 51 72.6 19.6 3.9 0.0 3.9 75.5 20.4 4.1
Bradfd 73 74.0 11.0 9.6 0.0 5.5 78.3 11.6 10.1
Brightn 130 62.3 26.2 2.3 0.8 8.5 68.6 28.8 2.5
Bristol 140 72.1 16.4 5.7 0.0 5.7 76.5 17.4 6.1
Camb 125 62.4 10.4 20.8 0.0 6.4 66.7 11.1 22.2
Carlis 17 58.8 35.3 0.0 0.0 5.9 62.5 37.5 0.0
Carsh 228 70.2 15.4 8.8 0.4 5.3 74.4 16.3 9.3
Chelms 44 84.1 11.4 0.0 0.0 4.6 88.1 11.9 0.0
Colchr 36 91.7 2.8 0.0 0.0 5.6 97.1 2.9 0.0
Covnt 105 57.1 28.6 9.5 0.0 4.8 60.0 30.0 10.0
Derby 83 56.6 33.7 1.2 0.0 8.4 61.8 36.8 1.3
Donc 38 76.3 18.4 0.0 5.3 0.0 80.6 19.4 0.0
Dorset 76 60.5 27.6 4.0 5.3 2.6 65.7 30.0 4.3
Dudley 49 65.3 28.6 0.0 2.0 4.1 69.6 30.4 0.0
Exeter 126 72.2 19.1 4.0 0.8 4.0 75.8 20.0 4.2
Glouc 64 70.3 20.3 1.6 0.0 7.8 76.3 22.0 1.7
Hull 93 50.5 33.3 5.4 0.0 10.8 56.6 37.4 6.0
Ipswi 37 59.5 29.7 8.1 0.0 2.7 61.1 30.6 8.3
Kent 115 62.6 20.0 11.3 0.0 6.1 66.7 21.3 12.0
L Barts 274 63.9 24.8 6.2 0.0 5.1 67.3 26.2 6.5
L Guys 129 73.6 12.4 13.2 0.0 0.8 74.2 12.5 13.3
L Kings 130 69.2 26.9 2.3 0.0 1.5 70.3 27.3 2.3
L Rfree 240 63.3 19.6 12.9 0.4 3.8 66.1 20.4 13.5
L St.G 90 73.3 10.0 7.8 0.0 8.9 80.5 11.0 8.5
L West 365 78.6 5.2 12.6 0.0 3.6 81.5 5.4 13.1
Leeds 152 66.5 17.1 13.2 0.0 3.3 68.7 17.7 13.6
Leic 242 61.2 18.6 13.6 0.0 6.6 65.5 19.9 14.6
Liv Ain 69 72.5 18.8 1.5 0.0 7.3 78.1 20.3 1.6
Liv RI 111 53.2 25.2 10.8 0.9 9.9 59.6 28.3 12.1
M RI 170 52.4 25.9 20.0 0.0 1.8 53.3 26.4 20.4
Middlbr 127 74.8 3.2 12.6 0.0 9.5 82.6 3.5 13.9
Newc 107 60.8 15.9 12.2 0.0 11.2 68.4 17.9 13.7
Norwch 80 65.0 26.3 2.5 0.0 6.3 69.3 28.0 2.7
Nottm 98 43.9 38.8 7.1 0.0 10.2 48.9 43.2 8.0
Oxford 167 56.9 19.2 13.8 0.6 9.6 63.3 21.3 15.3
Plymtha 50
Ports 175 66.9 16.6 10.3 0.0 6.3 71.3 17.7 11.0
Prestn 133 66.9 15.0 11.3 0.8 6.0 71.8 16.1 12.1
Redng 84 61.9 32.1 3.6 0.0 2.4 63.4 32.9 3.7
Salford 119 70.6 25.2 2.5 0.8 0.8 71.8 25.6 2.6
Sheff 153 69.3 15.7 9.2 0.7 5.2 73.6 16.7 9.7
Shrew 56 66.1 25.0 0.0 0.0 8.9 72.6 27.5 0.0
Stevng 101 69.3 14.9 10.9 0.0 5.0 72.9 15.6 11.5
Sthend 22 81.8 18.2 0.0 0.0 0.0 81.8 18.2 0.0
Stoke 89 68.5 18.0 4.5 0.0 9.0 75.3 19.8 4.9
Sund 74 81.1 10.8 4.1 0.0 4.1 84.5 11.3 4.2
Truro 41 58.5 22.0 7.3 0.0 12.2 66.7 25.0 8.3
Wirral 47 61.7 29.8 2.1 0.0 6.4 65.9 31.8 2.3
Wolve 87 41.4 48.3 1.2 0.0 9.2 45.6 53.2 1.3
York 55 54.6 25.5 14.6 0.0 5.5 57.7 26.9 15.4
25
Status at 90 days of only those

Status at 90 days of all patients who started RRT (%) patients still on RRT (%)
Stopped
Centre N HD PD Tx treatment Died HD PD Tx
N Ireland
Antrim 31 74.2 16.1 6.5 3.2 0.0 76.7 16.7 6.7
Belfast 92 59.8 9.8 21.7 1.1 7.6 65.5 10.7 23.8
Newry 26 65.4 30.8 0.0 0.0 3.9 68.0 32.0 0.0
Ulster 26 69.2 7.7 0.0 0.0 23.1 90.0 10.0 0.0
West NI 30 76.7 10.0 3.3 6.7 3.3 85.2 11.1 3.7
Scotland
Abrdn 44 81.8 18.2 0.0 0.0 0.0 81.8 18.2 0.0
Airdrie 61 83.6 14.8 1.6 0.0 0.0 83.6 14.8 1.6
D & Gall 18 50.0 38.9 0.0 0.0 11.1 56.3 43.8 0.0
Dundee 41 75.6 19.5 0.0 0.0 4.9 79.5 20.5 0.0
Dunfn 31 80.7 12.9 0.0 0.0 6.5 86.2 13.8 0.0
Edinb 78 74.4 10.3 9.0 0.0 6.4 79.5 11.0 9.6
Glasgw 185 78.9 10.3 7.6 0.0 3.2 81.6 10.6 7.8
Inverns 12 75.0 25.0 0.0 0.0 0.0 75.0 25.0 0.0
Klmarnk 39 66.7 23.1 0.0 0.0 10.3 74.3 25.7 0.0
Wales
Bangor 16 68.8 18.8 0.0 0.0 12.5 78.6 21.4 0.0
Cardff 180 67.2 14.4 12.8 0.6 5.0 71.2 15.3 13.5
Clwyd 21 71.4 9.5 4.8 4.8 9.5 83.3 11.1 5.6
Swanse 128 67.2 23.4 0.8 1.6 7.0 73.5 25.6 0.9
Wrexm 32 43.8 28.1 12.5 3.1 12.5 51.9 33.3 14.8
England 5,797 66.1 19.6 8.6 0.3 5.5 70.1 20.8 9.1
N Ireland 205 66.3 13.2 11.2 2.0 7.3 73.1 14.5 12.4
Scotland 509 76.8 14.7 4.3 0.0 4.1 80.1 15.4 4.5
Wales 377 65.5 18.6 7.7 1.3 6.9 71.4 20.2 8.4
UK 6,888 66.9 19.0 8.3 0.4 5.5 71.0 20.2 8.8
a
Breakdown not shown for Plymouth as not all data was available (see table 1.3)
with some centres having over double the average The median age at start for those on HD at 90 days
percentage on PD for one or both of the age groups. was 66.3 years compared with 59.8 years for PD. There
Some centres had less than half the average percentage were 10 centres where the percentage of patients treated
on PD. with PD was the same as or higher in the over 65s than
the under 65s (a similar number to the 11 centres for
Transplant Home – HD 2011).
8.8% 0.2%
Modality change over time

PD
Hosp – HD Table 1.13 gives the breakdown of status/treatment
42.7%
20.2% modality at four subsequent time points by initial treat-
ment type for patients starting RRT in 2007. Fifty-three
percent of patients who started on HD had died within
five years of starting. This compared to 30% and 4% for
those starting on PD or transplant respectively. Of
those patients starting on PD, 92% were on PD at 90
Satellite HD days but this percentage dropped sharply at the later
28.1% time points. As expected and in contrast, 89% of patients
Fig. 1.8. RRT modality at 90 days (incident cohort 1/10/2011 to starting with a transplant were also transplant patients at
30/09/2012) the five year time point.
26
Table 1.12. Modality split of patients on dialysis at 90 days (incident cohort 1/10/2011 to 30/09/2012)
Age ,65 (%) Age 565 (%) All patients (%)
Centre N HD PD HD PD HD PD
England
B Heart 100 73.5 26.5 90.2 9.8 82.0 18.0
B QEH 202 73.0 27.0 89.7 10.3 80.2 19.8
Basldn 47 70.8 29.2 87.0 13.0 78.7 21.3
Bradfd 62 85.7 14.3 88.9 11.1 87.1 12.9
Brightn 115 60.8 39.2 78.1 21.9 70.4 29.6
Bristol 124 73.7 26.3 88.1 11.9 81.5 18.5
Camb 91 80.0 20.0 88.5 11.5 85.7 14.3
Carlis 16 54.5 45.5 80.0 20.0 62.5 37.5
Carsh 195 72.7 27.3 88.1 11.9 82.1 17.9
Chelms 42 81.0 19.0 95.2 4.8 88.1 11.9
Colchr 34 92.3 7.7 100.0 0.0 97.1 2.9
Covnt 90 56.4 43.6 74.5 25.5 66.7 33.3
Derby 75 60.5 39.5 64.9 35.1 62.7 37.3
Donc 36 81.3 18.8 80.0 20.0 80.6 19.4
Dorset 67 63.6 36.4 71.1 28.9 68.7 31.3
Dudley 46 54.2 45.8 86.4 13.6 69.6 30.4
Exeter 115 69.7 30.3 82.9 17.1 79.1 20.9
Glouc 58 79.3 20.7 75.9 24.1 77.6 22.4
Hull 78 50.0 50.0 75.0 25.0 60.3 39.7
Ipswi 33 70.6 29.4 62.5 37.5 66.7 33.3
Kent 95 68.6 31.4 80.0 20.0 75.8 24.2
L Barts 243 71.6 28.4 72.6 27.4 72.0 28.0
L Guys 111 80.3 19.7 93.3 6.7 85.6 14.4
L Kings 125 66.2 33.8 81.3 18.8 72.0 28.0
L Rfree 199 68.3 31.7 84.7 15.3 76.4 23.6
L St.G 75 84.6 15.4 91.7 8.3 88.0 12.0
L West 306 92.2 7.8 95.4 4.6 93.8 6.2
Leeds 127 69.6 30.4 91.4 8.6 79.5 20.5
Leic 193 74.7 25.3 78.4 21.6 76.7 23.3
Liv Ain 63 69.0 31.0 88.2 11.8 79.4 20.6
Liv RI 87 62.7 37.3 75.0 25.0 67.8 32.2
M RI 133 64.9 35.1 69.6 30.4 66.9 33.1
Middlbr 99 93.8 6.3 98.0 2.0 96.0 4.0
Newc 82 76.7 23.3 82.1 17.9 79.3 20.7
Norwch 73 58.1 41.9 81.0 19.0 71.2 28.8
Nottm 81 41.9 58.1 65.8 34.2 53.1 46.9
Oxford 127 77.0 23.0 71.7 28.3 74.8 25.2
Plymtha 44
Ports 146 76.4 23.6 83.8 16.2 80.1 19.9
Prestn 109 81.0 19.0 82.4 17.6 81.7 18.3
Redng 79 61.8 38.2 68.9 31.1 65.8 34.2
Salford 114 64.4 35.6 83.6 16.4 73.7 26.3
Sheff 130 74.6 25.4 88.9 11.1 81.5 18.5
Shrew 51 57.7 42.3 88.0 12.0 72.5 27.5
Stevng 85 76.3 23.7 87.2 12.8 82.4 17.6
Sthend 22 70.0 30.0 91.7 8.3 81.8 18.2
Stoke 77 80.6 19.4 78.0 22.0 79.2 20.8
Sund 68 82.9 17.1 93.9 6.1 88.2 11.8
Truro 33 55.6 44.4 79.2 20.8 72.7 27.3
Wirral 43 60.0 40.0 73.9 26.1 67.4 32.6
Wolve 78 45.8 54.2 46.7 53.3 46.2 53.8
York 44 55.0 45.0 79.2 20.8 68.2 31.8
27
Age ,65 (%) Age 565 (%) All patients (%)
Centre N HD PD HD PD HD PD
N Ireland
Antrim 28 64.3 35.7 100.0 0.0 82.1 17.9
Belfast 64 80.0 20.0 93.1 6.9 85.9 14.1
Newry 25 72.7 27.3 64.3 35.7 68.0 32.0
Ulster 20 85.7 14.3 92.3 7.7 90.0 10.0
West NI 26 83.3 16.7 92.9 7.1 88.5 11.5
Scotland
Abrdn 44 68.2 31.8 95.5 4.5 81.8 18.2
Airdrie 60 82.9 17.1 88.0 12.0 85.0 15.0
D & Gall 16 57.1 42.9 55.6 44.4 56.3 43.8
Dundee 39 73.3 26.7 83.3 16.7 79.5 20.5
Dunfn 29 82.4 17.6 91.7 8.3 86.2 13.8
Edinb 66 94.6 5.4 79.3 20.7 87.9 12.1
Glasgw 165 82.4 17.6 95.0 5.0 88.5 11.5
Inverns 12 80.0 20.0 71.4 28.6 75.0 25.0
Klmarnk 35 66.7 33.3 85.7 14.3 74.3 25.7
Wales
Bangor 14 80.0 20.0 77.8 22.2 78.6 21.4
Cardff 147 76.4 23.6 88.0 12.0 82.3 17.7
Clwyd 17 85.7 14.3 90.0 10.0 88.2 11.8
Swanse 116 59.1 40.9 83.3 16.7 74.1 25.9
Wrexm 23 44.4 55.6 71.4 28.6 60.9 39.1
England 4,968 71.6 28.4 82.6 17.4 77.1 22.9
N Ireland 163 77.2 22.8 89.3 10.7 83.4 16.6
Scotland 466 80.3 19.7 87.8 12.2 83.9 16.1
Wales 317 69.3 30.7 84.4 15.6 77.9 22.1
UK 5,914 72.4 27.6 83.3 16.7 77.9 22.1
a
Breakdown not shown for Plymouth as not all data was available (see table 1.3) and more PD than HD starters were missing
Table 1.13. Initial and subsequent modalities for patients starting RRT in 2007
Percentage
First treatment N Later modality 90 days 1 year 3 years 5 years
HD 4,981 HD 88 72 47 30
PD 3 3 2 1
Transplant 1 3 10 15
Other∗ 0 1 1 1
Died 7 20 40 53
PD 1,365 HD 4 13 20 18
PD 92 70 31 12
Other∗ 0 1 1 1
Died 1 5 19 30
Transplant 322 HD 1 1 3 5
PD 0 0 0 2
Died 1 2 3 4
∗
Other e.g. stopped treatment
28
10.0
Error bars = 95% CI
18% of patients may have had an incorrect date of
Geometric mean eGFR ml/min/1.73 m2
starting RRT allocated and thus, the eGFR used for

9.5
analysis may have been taken whilst they were already
9.0
receiving RRT. For details see the 12th Annual Report
chapter 13: The UK Renal Registry Advanced CKD
8.5 Study 2009 [5].
8.0
7.5
3. Late presentation and delayed referral of
7.0 incident patients
18–24 25–34 35–44 45–54 55–64 65–74 75–84 85+
Age group (years)
Introduction
Fig. 1.9. Geometric mean eGFR at start of RRT (2012) by age Late presentation to a nephrologist is regarded as a
group
negative aspect in renal care. It can be defined in a
number of ways as it has a range of possible causes.
Renal function at the time of starting RRT There are many patients with chronic kidney disease
The mean eGFR at initiation of RRT in 2012 was who are regularly monitored in primary or secondary
8.5 ml/min/1.73 m2. This increased with increasing age care and whose referral to nephrology services is delayed
after the 45–54 age group and was highest in the 85+ (delayed or late referral). In contrast, other patients
age group at about 9.1 ml/min/1.73 m2 (figure 1.9). By present late to medical services due to no particular
contrast, in the United States, 54% of patients starting deficiency in the service; those with either such slowly
RRT in 2009 had an eGFR greater than 10 ml/min/ progressive disease as to have remained asymptomatic
1.73 m2 [4]. for many years or the opposite with rapidly progressive
Figure 1.10 shows serial data from centres reporting glomerulonephritis. The main analyses presented here
annually to the UKRR since 2003. For both HD and do not differentiate between these groups and include
PD patients, average eGFR at start of RRT in 2012 was any patient first seen by renal services within 90 days of
slightly lower than for 2011. For the six years prior to starting RRT as ‘late presentation’.
2011 there was higher average eGFR at start of RRT for One analysis attempts to capture ‘late referrals’: it shows
PD than HD patients but the values were similar for the percentage presenting within 90 days of starting RRT
2011 and 2012. after excluding an acute renal disease group. This group is
Some caution should be applied to the analysis of made up of those people with conditions likely to present
eGFR at the start of RRT as a review of pre-RRT with rapidly deteriorating renal function: crescentic
biochemistry in nine renal centres revealed that up to glomerulonephritis (type I, II, III), renal vascular disease
due to malignant hypertension, renal vascular disease
due to polyarteritis, nephropathy (interstitial) due to cis-
9.5
Error bars = 95% CI HD platinum, Balkan nephropathy, Wegener’s granulomato-
Geometric mean eGFR ml/min/1.73m2
PD sis, cryoglobulinemic glomerulonephritis, myelomatosis/

9.0 light chain deposit disease, Goodpasture’s syndrome,
systemic sclerosis, haemolytic ureaemic syndrome
(including Moschcowitz syndrome), multi-system disease
8.5
– other, tubular necrosis (irreversible) or cortical necrosis,
kidney tumour(s) and surgical loss of kidney.
8.0
Methods
Date first seen by a nephrologist has not been collected from
7.5 the Scottish Renal Registry and so Scottish centres were excluded
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
from these analyses. Data were included from all incident patients
Year
in English, Welsh or Northern Irish centres in the years 2011 to
Fig. 1.10. eGFR on starting RRT 2003 to 2012, PD and HD 2012. This two year cohort is used for most of the analyses in
(restricted to centres reporting since 2003) order to make the late presentation percentages more reliably
29
estimated and to allow these to be shown for subgroups of patients. Results

The date first seen in a renal centre and the date of starting RRT Table 1.14 shows the percentage completeness of data
were used to define the late presenting cohort. A small amount
of data were excluded because of actual or potential inconsisten-
for 2011 and 2012. Average completeness for 2012 was
cies. Only data from those centres with 75% or more completeness similar to 2011 at just over 80%.
for the relevant year were used. Some data were excluded if 10% or
more of the patients were reported to have started RRT on the Late presentation by centre
same date as the first presentation. This was because investigation Figure 1.11 shows that late presentation varied
has shown that this is likely due to misunderstanding on the part between centres from 7–32% in patients starting RRT
of the renal centres resulting in incorrect recording of data. After in 2011 to 2012. The overall rate of late presentation
these exclusions, data on 9,937 patients were available for analysis.
Presentation times of 90 days or more were defined as early was 19.5% and was 14.2% once those people with diseases
presentation and times of less than 90 days were defined as late likely to present acutely were excluded. Table 1.15 shows
presentation. the overall percentage presenting late for the combined
Table 1.14. Percentage completeness of time of presentation data (2011 and 2012 incident RRT patients) by centre
N Percentage completeness N Percentage completeness
Centre 2011 2012 2011 2012 Centre 2011 2012 2011 2012
England Norwch 87 74 93.1 64.9

B Heart 113 101 97.3 96.0 Nottm 116 99 97.4 97.9
B QEH 215 216 97.7 99.5 Oxford 177 171 94.3 98.2
Basldn 42 53 100.0 96.2 Plymth 60 47 31.7 31.9
Bradfd 60 71 98.3 97.1 Ports 187 161 97.8 96.9
Brightn 119 136 17.1 91.8 Prestn 140 147 98.6 95.8
Bristol 139 148 88.2 94.6 Redng 103 73 63.1 97.3
Camb 122 124 98.4 100.0 Salford 126 134 0.8 10.6
Carlis 28 19 96.4 94.7 Sheff 135 158 100.0 98.7
Carsh 207 242 94.2 88.0 Shrew 61 57 100.0 98.3
Chelms 47 45 95.7 97.8 Stevng 110 110 97.3 99.1
Colchr 44 29 86.4 100.0 Sthend 29 26 100.0 100.0
Covnt 111 112 73.4 98.2 Stoke 93 77 100.0 98.7
Derby 80 81 95.0 100.0 Sund 57 71 94.7 98.6
Donc 43 40 100.0 95.0 Truro 38 50 97.4 98.0
Dorset 79 72 100.0 95.8 Wirrala 62 50 b
97.9
Dudley 43 56 97.7 98.2 Wolve 76 84 100.0 100.0
Exeter 112 138 99.1 97.1 York 52 53 100.0 100.0
Glouc 58 74 100.0 94.5 N Ireland
Hull 109 97 66.1 97.9 Antrim 30 26 96.7 100.0
Ipswi 29 43 92.9 97.7 Belfast 69 91 95.7 89.0
Kent 122 115 100.0 100.0 Newry 38 18 100.0 100.0
L Barts 249 263 2.0 1.5 Ulster 35 30 100.0 100.0
L Guys 120 127 94.1 22.4 West NI 38 21 94.7 100.0
L Kings 140 125 96.4 96.0 Wales
L Rfree 223 240 91.9 99.2 Bangor 20 21 100.0 90.5
L St.G 74 91 32.4 65.9 Cardff 186 170 97.3 98.8
b
L West 365 352 93.1 0.3 Clwyd 17 22 95.5
Leeds 158 154 98.1 98.0 Swanse 118 113 99.2 99.1
Leic 267 235 97.3 97.0 Wrexm 26 34 88.0 97.1
Liv Ain 61 63 58.3 100.0 England 5,756 5,798 81.8 81.6
Liv RI 114 110 7.1 99.1 N Ireland 210 186 97.1 94.6
M RI 156 160 81.2 92.4 Wales 367 360 92.9 98.0
Middlbr 100 120 98.0 97.5 E, W & NI 6,333 6,344 83.0 82.9
Newc 98 104 95.9 89.4
a
Although completeness was good for Wirral for 2012, the late presentation percentage was suspiciously high and is not shown in table 1.15 or
figure 1.11 due to concerns about data accuracy
b
Data not shown as .10% of patients reported as starting RRT on the same date as first presentation
30
60
Upper 95% Cl N = 9,937
% presenting late
50
Lower 95% Cl
Percentage of patients
40
30
20
10
0
B Heart
Sund
Carlis
Stevng
Exeter
L Guys
Nottm
Wrexm
Cardff
Bradfd
Antrim
Oxford
Shrew
Dorset
M RI
Leeds
Dudley
Liv Ain
Bristol
Glouc
Ports
L West
Prestn
L Kings
Hull
Sheff
Basldn
Covnt
Newry
Belfast
Leic
Bangor
Middlbr
West NI
Newc
Kent
Carsh
Truro
Redng
Wolve
Brightn
Derby
Donc
Sthend
Swanse
Clwyd
York
Ulster
Chelms
Camb
L Rfree
Colchr
Norwch
Stoke
Liv RI
B QEH
Ipswi
England
N Ireland
Wales
E, W & NI
Centre
Fig. 1.11. Percentage presenting late (2011/2012)
2011–2012 incident cohort, the percentages presenting initiative (www.dh.gov.uk) raising awareness of CKD
late amongst those patients defined as not having an amongst non-nephrologists and the introduction of
‘acute diagnosis’ and the percentages amongst non- estimated GFR reporting.
diabetics (as PRD). In 2012, 66.8% of incident patients presented over a
year before they needed to start RRT. There were 8.4%
Late presentation in 2012 and the trend over time of patients presenting within 6–12 months, 5.5% within
There has been a steady decline nationally in the 3–6 months and 19.3% within three months. Figure 1.12
proportion of patients presenting late to renal services, shows this breakdown by year for those 20 centres
with some centres achieving ,10% late presentation supplying data over 75% complete for each of the last
rates. This may be a consequence of the National CKD six years. The figure shows an increase over time in the
guidelines published by the Medical and GP Royal Col- percentage of patients presenting 12 months or more
leges [6], the Quality and Outcomes Framework (QOF) before starting RRT. As shown in previous reports this
Table 1.15. Percentage of patients presenting to a nephrologist less than 90 days before RRT initiation (2011–2012 incident patients) by
centre
Percentage presenting late
Centre N with data Overall (95% CI) Non-acute∗ Non-diab PRD
England
B Heart 204 7.4 (4.5–11.8) 4.5 10.5
B QEH 420 28.3 (24.2–32.8) 23.5 29.3
Basldn 93 19.4 (12.6–28.6) 12.2 23.2
Bradfd 126 13.5 (8.6–20.6) 11.9 15.6
Brightn 123 22.8 (16.2–31.0) 16.5 25.0
Bristol 260 17.7 (13.5–22.8) 11.4 20.4
Camb 243 25.9 (20.8–31.8)
Carlis 45 11.1 (4.7–24.1) 11.9 11.4
Carsh 406 21.9 (18.2–26.2) 17.9 23.3
Chelms 89 24.7 (16.9–34.7) 19.0 28.8
Colchr 67 26.9 (17.6–38.7) 18.5 22.7
Covnt 108 19.4 (13.0–28.0) 14.6 20.2
Derby 157 22.9 (17.0–30.2) 16.8 29.7
Donc 81 23.5 (15.5–33.9) 18.6 28.1
Dorset 148 15.5 (10.6–22.3) 14.0 17.5
Dudley 96 16.7 (10.5–25.5) 12.5 20.3
Exeter 243 11.9 (8.4–16.7) 10.0 14.6
31
Percentage presenting late
Centre N with data Overall (95% CI) Non-acute∗ Non-diab PRD
Glouc 124 17.7 (12.0–25.5) 13.9 19.0

Hull 95 19.0 (12.3–28.1) 14.9 23.0
Ipswi 68 32.4 (22.4–44.3) 34.6 44.4
Kent 237 21.5 (16.8–27.2) 16.8 25.0
L Guys 112 12.5 (7.5–20.0) 10.2 12.8
L Kings 255 18.8 (14.5–24.1) 14.0 27.5
L Rfree 443 26.2 (22.3–30.5) 21.5 28.7
L West 338 18.3 (14.6–22.8) 14.9 21.9
Leeds 299 16.4 (12.6–21.0) 8.7 18.8
Leic 482 19.9 (16.6–23.7) 11.7 23.5
Liv Ain 63 17.5 (9.9–28.9) 16.7 21.2
Liv RI 105 27.6 (19.9–36.9) 11.3 32.2
M RI 271 16.2 (12.3–21.1) 14.3 18.5
Middlbr 215 20.9 (16.0–26.9) 18.8 26.1
Newc 187 21.4 (16.1–27.9) 11.8 24.2
Norwch 81 27.2 (18.6–37.8) 17.7 30.4
Nottm 206 12.6 (8.7–17.9) 11.3 14.8
Oxford 332 15.1 (11.6–19.3) 11.3 19.0
Ports 336 18.2 (14.4–22.6) 9.4 20.3
Prestn 275 18.6 (14.4–23.6) 13.1 20.1
Redng 71 22.5 (14.3–33.7) 19.1 31.3
Sheff 289 19.0 (14.9–24.0) 13.5 23.8
Shrew 117 15.4 (9.9–23.1) 11.1 11.1
Stevng 215 11.6 (8.0–16.6) 9.8 12.1
Sthend 55 23.6 (14.3–36.6) 18.8 28.9
Stoke 169 27.2 (21.0–34.4) 19.0 32.3
Sund 124 8.9 (5.0–15.3) 5.6 10.0
Truro 85 22.4 (14.7–32.4) 18.7 25.0
Wolve 159 22.6 (16.8–29.8) 20.5 25.6
York 104 24.0 (16.8–33.2) 17.4 27.4
N Ireland
Antrim 55 14.6 (7.4–26.5) 13.2 18.4
Belfast 147 19.7 (14.1–27.0) 12.3 24.6
Newry 56 19.6 (11.2–32.1) 13.5 23.7
Ulster 65 24.6 (15.7–36.5) 21.0 22.5
West NI 57 21.1 (12.4–33.5) 16.3 22.9
Wales
Bangor 39 20.5 (10.6–36.0) 21.1 25.0
Cardff 347 13.3 (10.1–17.3) 10.3 16.3
Clwyd 21 23.8 (10.3–46.0) 22.2 17.7
Swanse 227 23.8 (18.7–29.8) 16.2 30.7
Wrexm 55 12.7 (6.2–24.4) 11.1 18.4
England 8,868 19.7 (18.9–20.5) 14.3 22.2
N Ireland 380 20.0 (16.3–24.3) 14.8 23.0
Wales 689 17.4 (14.8–20.4) 13.2 21.7
E, W & NI 9,937 19.5 (18.8–20.3) 14.2 22.2
(min, max) (7.4–42.6) (4.5–34.6) (10.0–44.4)
(IQR) (16.3–23.6) (11.7–18.5) (18.5–26.4)
Blank cells – data for PRD not used due to high % with uncertain aetiology
∗
Non-acute group excludes crescentic (extracapillary) glomerulonephritis (type I, II, III), nephropathy (interstitial) due to cis-platinum, renal
vascular disease due to malignant hypertension, renal vascular disease due to polyarteritis, Wegener’s granulomatosis, cryoglobulinemic
glomerulonephritis, myelomatosis/light chain deposit disease, Goodpasture’s syndrome, systemic sclerosis (scleroderma), haemolytic ureaemic
syndrome (including Moschcowitz syndrome), multi-system disease – other, tubular necrosis (irreversible) or cortical necrosis, Balkan
nephropathy, kidney tumour(s), and traumatic or surgical loss of kidney
32
100 Ethnicity and late presentation

% >12 months p for trend <0.0001
% 6–<12 months In the 2011 to 2012 cohort, the percentage of South
80
% 3–<6 months Asian and Black patients presenting late (,90 days)
% <3 months
was significantly lower than in Whites (16.4% vs.
60
19.8%: p = 0.002). The high incidence of diabetes in
non-Whites (as discussed below, patients with diabetes
tended to present earlier) explains some of the difference
40
in presentation time between the ethnic groups. When
patients with diabetes were excluded, the percentages
20 presenting late (,90 days) became 20.0% in South
Asian and Black patients vs. 22.6% in Whites (p = 0.1).
0
2007 2008 2009 2010 2011 2012
Year
Primary renal disease and late presentation
In the 2011 to 2012 cohort, late presentation differed
Fig. 1.12. Late presentation rate by year (2007–2012)
Restricted to centres reporting continuous data for 2007–2012 significantly between primary renal diagnoses (Chi-
squared test p , 0.0001) (table 1.16). Patients in the
increase was most marked in the years just before those acute group or with data not available had high rates of
shown in the figure. In 2005, only 52.6% of incident late presentation. Those with diabetes and pyelonephritis
patients presented over a year before they needed to or adult polycystic kidney disease had low rates. There
start RRT compared with the 66.8% seen for 2012. was a notable decline in the proportion of diabetics
presenting late up until 2007. Since then the proportion
Age and late presentation has been stable. The decline seen earlier likely reflects
In the 2011 to 2012 cohort, patients who presented late national initiatives to screen patients with diabetes for
were not significantly older or younger than patients who proteinuria and falling GFR.
presented earlier (590 days before RRT initiation)
(median age 66.1 vs. 64.7 years: p = 0.1). Except for
the two youngest age groups, the median duration of
pre-RRT care did not vary greatly with age group Table 1.16. Late presentation by primary renal diagnosis (2011–
(figure 1.13). 2012 incident patients)
Late presentation
Gender and late presentation
In the 2011 and 2012 cohort, there was no significant Diagnosis N N %
difference in the ratio of males to females by time of Uncertain aetiology 1,407 294 20.9
presentation (male : female ratio 1.68 in early presen- Diabetes 2,251 204 9.1
tation, 1.84 in late presentation, p = 0.08). Glomerulonephritis 1,160 179 15.4
Other identified category 893 167 18.7
7
Polycystic kidney or 1,270 130 10.2
pyelonephritis
6 Renal vascular disease 1,140 179 15.7
Acute group 889 488 54.9
5 Data not available 296 127 42.9
Years (IQR)
4 Unlike elsewhere in the report, the RVD group includes hypertension

3
Polycystic and pyelonephritis are grouped together
Acute group includes crescentic (extracapillary) glomerulonephritis
2 (type I, II, III), nephropathy (interstitial) due to cis-platinum, renal
vascular disease due to malignant hypertension, renal vascular disease
1 due to polyarteritis, Wegener’s granulomatosis, cryoglobulinemic
glomerulonephritis, myelomatosis/light chain deposit disease, Good-
0
18–24 25–34 35–44 45–54 55–64 65–74 75–84 85+
pasture’s syndrome, systemic sclerosis (scleroderma), haemolytic
Age group (years)
ureaemic syndrome (including Moschcowitz syndrome), multi-system
disease – other, tubular necrosis (irreversible) or cortical necrosis,
Fig. 1.13. Median duration of pre-RRT care by age group (incident Balkan nephropathy, kidney tumour(s), and traumatic or surgical
patients 2011–2012) loss of kidney
33
Table 1.17. Percentage prevalence of specific comorbidities RRT initiation than patients presenting earlier (92 vs.
amongst patients presenting late (,90 days) compared with those 102 g/L: p , 0.0001). This may reflect inadequate pre-
presenting early (590 days) (2011–2012 incident patients)
dialysis care with limited anaemia management, but
Comorbidity ,90 days 590 days p-value alternatively those presenting late may be more likely to
have anaemia because of multisystem disease or inter-
Ischaemic heart disease 16.3 19.5 0.02 current illness. More detailed analyses of haemoglobin
Cerebrovascular disease 9.7 10.0 0.7
Peripheral vascular disease 8.8 12.7 0.0003 at start of RRT and late presentation can be found in
Diabetes (not a cause of ERF) 10.2 9.6 0.5 chapter 10: Haemoglobin, Ferritin and Erythropoietin
Liver disease 4.2 2.8 0.02 amongst UK Adult Dialysis Patients in 2012: National
Malignancy 19.9 11.1 ,0.0001 and Centre-specific Analyses.
COPD 9.0 7.3 0.1
Smoking 15.4 14.2 0.3
eGFR at start of RRT and late presentation
In the 2011 to 2012 cohort, eGFR at start of RRT
Modality and late presentation was significantly lower in patients presenting late than
In the 2011 to 2012 cohort, late presentation was those presenting earlier (7.9 vs. 8.7 ml/min/1.73 m2:
associated with initial modality. The percentage of p , 0.0001). These findings are in contrast to some of
patients whose first modality was PD was significantly the studies in the literature which have found the
lower in the late presentation group than in those opposite [7, 8].
presenting earlier (9.3% vs. 22.7%: p , 0.0001). By 90
days after RRT initiation this difference was reduced,
although it was still highly significant (12.5% vs. 22.0%:
p , 0.0001). Survival of incident patients
Comorbidity and late presentation See chapter 8: Survival and Causes of Death of UK
In the 2011 to 2012 cohort, the percentage of patients Adult Patients on Renal Replacement Therapy in 2012.
who were assessed as having no comorbidity was slightly
lower in those who presented late than those presenting
earlier (43.3% vs. 47.0%: p = 0.03). Ischaemic heart
disease and peripheral vascular disease were significantly Summary
less common in the group presenting late (table 1.17).
Liver disease was significantly more common in those RRT incidence rates for 2012 were similar to 2011 for
presenting late as was malignancy; perhaps because of England and for the UK as a whole. At least partly
the potential for rapid decline in renal function in this because of the smaller numbers involved, rates have
group. The evidence in the literature is in keeping with been more variable over the last few years for Northern
these findings with subtle variation between the individ- Ireland, Scotland and Wales. Wales continues to have
ual comorbidities [7–9]. the highest incidence rate. There remain large centre
variations in incidence rates for RRT. Significant num-
Haemoglobin and late presentation bers of patients continue to present late to renal centres.
In the 2011 to 2012 cohort, patients presenting late had
a significantly lower average haemoglobin concentration at Conflicts of interest: none
References
1 Castledine CI et al.: How much of the regional variation in RRT incidence 3 http://www.ons.gov.uk/ons/rel/ethnicity/focus-on-ethnicity-and-identity/
rates within the UK is explained by the health needs of the general popu- focus-on-ethnicity-and-identity-summary-report/focus-on—ethnicity-
lation? Nephrology Dialysis Transplantation 2012;27(10):3943/–3950 and-identity-summary-report.pdf.
2 Kuan Y et al.: GFR prediction using the MDRD and Cockcroft and Gault 4 US Renal Data System: USRDS 2011 Annual Data Report: Atlas of
equations in patients with end-stage renal disease. Nephrology Dialysis Chronic Kidney Disease and End-Stage Renal Disease in the United
Transplantation 2005;20(11):2394–2401 States, National Institutes of Health, National Institute of Diabetes and
34
Digestive and Kidney Diseases, Bethesda, MD, 2011. Publications based 7 Kazmi WHet al.: Late nephrology referral and mortality among patients
upon USRDS data reported here or supplied upon request must include with end-stage renal disease: a propensity score analysis. Nephrology
this citation and the following notice: The data reported here have been Dialysis Transplantation 2004;19(7):1808–1814
supplied by the United States Renal Data System (USRDS) 8 Roubicek C et al.: Timing of nephrology referral: Influence on mortality and
5 Ford DJ, Fogarty DG, Steenkamp R, Tomson CRV, Ben-Shlomo Y, Ansell morbidity. American journal of kidney diseases: the official journal of the
D. Chapter 13: The UK Renal Registry Advanced CKD Study: frequency National Kidney Foundation 2000;36(1):35–41
of incorrect reporting of date of start of RRT. Nephron Clinical Practice; 9 Winkelmayer WC et al.: A Propensity Analysis of Late Versus Early
115(suppl 1):c271–c278 Nephrologist Referral and Mortality on Dialysis. Journal of the American
6 http://www.renal.org/CKDguide/full/UKCKDfull.pdf. Society of Nephrology 2003;14(2):486–492.
35
Chapter 2 UK RRT Prevalence in 2012:
National and Centre-specific Analyses
Catriona Shawa, David Pitchera, Rishi Pruthia, Damian Fogartyab

a
UK Renal Registry, Bristol, UK; bBelfast Health and Social Care Trust, Belfast, UK
Key Words
. For all ages, the prevalence rate in men exceeded
Chronic kidney disease . Comorbidity . Diabetes . Dialysis . that in women, peaking in age group 80–84 years
End stage renal disease . Established renal failure . Ethnicity at 2,973 pmp and for females in age group 75–79
. Haemodialysis . Peritoneal dialysis . Prevalence . Primary years at 1,528 pmp.
Care Trust . Renal replacement therapy . Transplantation . . The most common identifiable renal diagnosis was
Treatment modality glomerulonephritis (18.8%), followed by uncertain
aetiology (16.7%) and diabetes (15.5%).
. Transplantation continued as the most common
Summary treatment modality (50.4%), HD was used in
42.7% and PD in 6.9% of RRT patients.
. There were 54,824 adult patients receiving renal . Prevalence rates in patients aged .85 years contin-
replacement therapy (RRT) in the UK on 31st ued to increase between 2011 and 2012 (952 pmp to
December 2012, an absolute increase of 3.7% from 983 pmp). There was 20 fold variation between
2011. The actual number of patients increased PCT/HBs in prevalence rates in patients aged .80
across all modalities: 2.3% increase haemodialysis years suggesting there was uncertainty regarding
(HD), 0.3% peritoneal dialysis (PD) and 5.6% for the risks and benefits of RRT in the elderly.
those with a functioning transplant. . In 2012, 20.7% of the prevalent UK RRT population
. The UK adult prevalence of RRT was 861 per (with ethnicity assigned) were from ethnic min-
million population (pmp). The reported prevalence orities compared to 14.9% in 2007.
in 2000 was 523 pmp. . There were national, regional and dialysis centre
. The number of patients receiving home HD level variations in prevalence rates. A significant fac-
increased by 19.3% from 905 patients in 2011 to tor in this variation was the ethnic mix of local
1,080 patients in 2012. populations, but a large amount of the variation
. The median age of prevalent patients was 58 years remains unexplained. Assessment of conservatively
(HD 66 years, PD 63 years, transplant 52 years). managed stage 5 CKD patients might explain
In 2000 the median age was 55 years (HD 63 more of this variation.
years, PD 58 years, transplant 48 years). The percen-
tage of RRT patients aged greater than 70 years
increased from 19.2% in 2000 to 24.9% in 2012.
37
Introduction standardised prevalence ratio was the observed prevalence number

divided by the expected prevalence number. The expected number
This chapter presents data on all adult patients on RRT of prevalent patients in a specific age/sex group (e.g. females
70–74) for a PCT is found by multiplying the total number of
in the UK at the end of 2012. The UK Renal Registry people (from the census) in that age/sex group in that PCT by
(UKRR) received data returns for 2012 from all five the overall rate in the whole of the UK for that same age/sex
renal centres in Wales, all five in Northern Ireland and group. Summing together the expected numbers in each of the
all 52 in England. Data from all nine centres in Scotland age/sex groups gives the overall expected number of prevalent
were obtained from the Scottish Renal Registry. Demo- patients for that PCT. A ratio below 1 indicates that the observed
number was less than expected given the area’s population
graphic data on children and young adults can be structure. This was statistically significant at the 5% level if the
found in chapter 7. upper confidence limit was less than 1. To enable assessment of
These analyses of prevalent RRT patients are per- whether a centre was an outlier in this regard, funnel plots for
formed annually to aid clinicians and policy makers in smaller and larger populations have been included (appendix D:
figures D3, D4) which show the 95% confidence intervals around
planning future RRT requirements in the UK. It is
the national average prevalence. The proportion of non-Whites in
important to understand national, regional and centre each PCT/HB was obtained from the ONS [1], the NRS [2] and the
level variation in numbers of prevalent patients as part NISRA [3].
of the planning process. In addition, knowledge about The prevalence rate per million population for each centre
variation in case mix is also reported to improve under- was calculated using a derived catchment population. For a full
description of the methodology used to estimate the catchment
standing of where resources should be focussed to
populations see appendix E: Methodology for Estimating Catch-
improve equity of provision of RRT in the UK. ment Populations Analyses (www.renalreg.com). For Scotland,
The term established renal failure (ERF) used within mid-2011 populations of Health Boards (from the General
this chapter is synonymous with the terms end stage Register Office for Scotland) were converted to centre level popu-
renal failure and end stage renal disease, which are in lations using an approximate mapping of renal centres to HBs
supplied by the Scottish Renal Registry. Estimates of the catch-
more widespread international usage. Patients have ment populations in Northern Ireland were supplied by personal
disliked the term ‘end stage’ which reflects the inevitable communication from Dr D Fogarty.
outcome of this disease. Throughout this chapter, haemodialysis refers to all modes of
HD treatment, including haemodiafiltration (HDF). Several
centres reported significant numbers of patients on HDF, but
other centres did not differentiate this treatment type in their
UKRR returns. Where joint care of renal transplant recipients
Methods between the referring centre and the transplant centre occurred,
the patient was allocated to the centre which saw the patient
These analyses relate to the prevalent RRT cohort in the UK in most frequently, usually the referring centre. Thus the number
2012. The cohort was defined as all adult patients receiving RRT of patients allocated to a transplant centre is often lower than
on the UKRR database on 31st December 2012. Population esti- that recorded by the centre itself and as a converse pre-emptively
mates were obtained from the UK Office for National Statistics transplanted patients are sometimes allocated to the transplanting
(ONS) [1], the National Records of Scotland (NRS) [2] and the centre rather than the referring centre if no transfer out code
Northern Ireland Statistic and Research Agency (NISRA) [3]. had been sent through. Queries and updated information are
The number of adult prevalent RRT patients was calculated for welcomed by the UKRR at any point during the year if this has
the UK as a whole and for each UK country, using UKRR data occurred.
from all renal centres. Crude prevalence rates were calculated Prevalent patients on RRT in 2012 were examined by time on
per million population (pmp) and standardised prevalence ratios RRT, age group, gender, ethnic origin, primary renal disease,
were calculated as detailed in appendix D: Methodology used for presence of diabetes and treatment modality (see appendix H:
Analyses (www.renalreg.com) for Primary Care Trusts (PCTs) in Coding (www.renalreg.com)). In this year’s analysis of prevalence,
England, Health & Social Care Areas in Northern Ireland, Local only adult patients on RRT contributed to the numerator. In
Health Boards in Wales and Health Boards in Scotland. These previous years, children have also been included in the numerator.
areas will be referred to in this report as ‘PCT/HBs’ reflecting Data on the paediatric population are presented in chapter 7.
the period of time before re-organisation of PCTs in England. Some centres electronically upload ethnicity coding to their
Briefly, data from all areas were used to calculate overall age and renal information technology (IT) system from the hospital
gender specific prevalence rates. The age and gender breakdown Patient Administration System (PAS). Ethnicity coding in these
of the population in each PCT/HB were obtained from the PAS systems is based on self-reported ethnicity and uses a differ-
mid-2011 population estimate based on 2011 Census data from ent coding system to those centres not linked to PAS [4]. For the
the ONS [1], the NRS [2] and the NISRA [3]. The population remaining centres, ethnicity coding is performed by clinical staff
breakdown and the overall prevalence rates were used to calculate and recorded directly into the renal IT system (using a variety of
the expected age and gender specific prevalence numbers for each coding systems). For all these analyses, data on ethnic origin
PCT/HB for each of the last six years. The age and gender were grouped into Whites, South Asians, Blacks, Chinese and
38
Chapter 2 UK RRT prevalence in 2012
Table 2.1. Prevalence of adult RRT in the UK on 31/12/2012
All UK centres 46,076 1,520 4,492 2,736 54,824

Total estimated population, mid-2012 (millions)∗ 53.5 1.8 5.3 3.1 63.7
Prevalence rate HD (pmp) 369 381 361 351 367
Prevalence rate PD (pmp) 61 46 44 65 60
Prevalence rate dialysis (pmp) 430 427 405 416 427
Prevalence rate transplant (pmp) 432 407 440 474 434
Prevalence rate total (pmp) 861 834 845 890 861
95% confidence intervals total (pmp) 853–869 792–875 821–870 857–923 853–868
∗
Data from the Office for National Statistics, National Records of Scotland and the Northern Ireland Statistics and Research Agency – based on
the 2011 census
Others as described in appendix H: Coding (www.renalreg.com). 983 pmarp in 2012. It is likely that this ageing of the preva-
Time on RRT was defined as median time on treatment and was lent population was due to an increasing number of older
calculated from the most recent start date. Patients without an
accurate start date were excluded from this calculation. Analyses
patients starting RRT, although improving patient survival
were done for the UK as a whole, by UK country, at centre level will also contribute.
and split by treatment modality when appropriate.
Chi-squared test, Fisher’s exact test, linear regression and Prevalent patients by RRT modality and centre
Kruskal Wallis tests were used as appropriate to test for significant The number of prevalent patients in each renal centre
differences between groups. The data were analysed using SAS 9.3.
and the distribution of their treatment modalities varied
widely (table 2.2). Many factors including geography,
local population density, age distribution, ethnic com-
position, prevalence of diseases predisposing to kidney
Results disease and the social deprivation index of that popu-
lation may contribute to this.
Prevalent patient numbers and changes in prevalence
The number of patients for each country (table 2.1) Changes in prevalence
was calculated by adding the patient numbers in each Overall growth in the prevalent UK RRT population
renal centre and these differ marginally from those from 2011 to 2012 was 3.7% (table 2.3), an annual growth
quoted elsewhere when patients are allocated to geo- rate which has been fairly consistent over the last 10–15
graphical areas by their individual postcodes, as some years (figure 2.2). Most of the growth in the prevalent
centres treat patients across national boundaries. RRT population was due to a continued increase in the
There were 54,824 adult patients receiving RRT in the size of the prevalent RRT population in England, Wales
UK at the end of 2012, giving an adult UK population
prevalence of 861 pmp (table 2.1) compared with 3,000
Wales
841 pmp in 2011. Prevalence rates increased in all of the N Ireland
UK countries in 2012. PD prevalence increased in North- 2,500
England
Prevalence rate pmp
ern Ireland but remained static or decreased in the other 2,000

Scotland
three countries compared with 2011. The decline in PD
prevalence in the UK overall noted since 1997 seems to 1,500
have plateaud in 2011 and 2012 with a static overall
1,000
prevalence of 60 pmp. Once more, the prevalence of trans-
planted patients increased in the UK. Northern Ireland had 500
a higher RRT prevalence rate for patients aged 65 and older
compared with the other UK countries (figure 2.1). In the 0
20–24
25–29
30–34
35–39
40–44
45–49
50–54
55–59
60–64
65–69
70–74
75–79
80–84
85+
UK, the RRT prevalence rate in patients aged 80–84 contin-

ued to rise over time from 1,824 per million age related Age group
population (pmarp) in 2011 to 1,896 pmarp in 2012 and Fig. 2.1. Prevalence rates per million population by age group and
in patients aged .85 years from 952 pmarp in 2011 to UK country on 31/12/2012
39
Table 2.2. Number of prevalent RRT patients by treatment modality and centre on 31/12/2012
Catchment 2012
Centre HD PD Dialysis Transplant RRT (millions) pmp (95% CI)
England
B Heart 435 47 482 188 670 0.74 908 (839–977)
B QEHa 926 159 1,085 886 1,971 1.70 1,160 (1,109–1,211)
Basldn 164 32 196 68 264 0.42 636 (559–713)
Bradfd 208 29 237 271 508 0.65 779 (711–847)
Brightn 371 85 456 375 831 1.30 641 (597–684)
Bristola 494 66 560 777 1,337 1.44 929 (879–979)
Camba 350 35 385 728 1,113 1.16 961 (905–1,018)
Carlis 61 27 88 128 216 0.32 673 (584–763)
Carsh 764 112 876 599 1,475 1.91 771 (732–811)
Chelms 129 26 155 69 224 0.51 439 (381–496)
Colchr 117 117 117 0.30 391 (320–462)
Covnta 363 100 463 437 900 0.89 1,009 (943–1,075)
Derby 220 89 309 168 477 0.70 679 (618–740)
Donc 172 29 201 60 261 0.41 636 (559–714)
Dorset 260 48 308 302 610 0.86 708 (652–764)
Dudley 169 63 232 84 316 0.44 715 (636–794)
Exeter 397 77 474 372 846 1.09 777 (724–829)
Glouc 219 36 255 162 417 0.59 710 (642–778)
Hull 334 90 424 365 789 1.02 773 (719–827)
Ipswi 129 31 160 179 339 0.40 850 (759–940)
Kent 384 62 446 476 922 1.22 753 (704–802)
L Bartsa 895 195 1,090 865 1,955 1.83 1,068 (1,021–1,116)
L Guysa 626 31 657 1,088 1,745 1.08 1,612 (1,537–1,688)
L Kings 492 86 578 340 918 1.17 784 (733–834)
L Rfreea 714 120 834 1,031 1,865 1.52 1,228 (1,173–1,284)
L St.Ga 284 54 338 386 724 0.80 907 (841–974)
L Westa 1,426 52 1,478 1,626 3,104 2.40 1,294 (1,248–1,340)
Leedsa 495 87 582 834 1,416 1.67 848 (804–892)
Leica 872 160 1,032 950 1,982 2.44 814 (778–849)
Liv Ain 175 20 195 195 0.48 403 (346–459)
Liv RIa 366 63 429 812 1,241 1.00 1,241 (1,172–1,310)
M RIa 507 82 589 1,121 1,710 1.53 1,117 (1,064–1,170)
Middlbr 339 11 350 439 789 1.00 786 (731–841)
Newca 285 47 332 614 946 1.12 844 (790–898)
Norwch 318 55 373 239 612 0.79 778 (716–840)
Nottma 376 81 457 549 1,006 1.09 925 (868–982)
Oxforda 423 86 509 1,026 1,535 1.69 908 (863–953)
Plymthab 131 35 166 293 459 0.47 977 (888–1,067)
Portsa 555 83 638 809 1,447 2.02 715 (678–752)
Prestn 536 69 605 476 1,081 1.49 724 (681–767)
Redng 271 72 343 328 671 0.91 737 (682–793)
Salford 380 104 484 398 882 1.49 592 (553–631)
Sheff a 588 69 657 650 1,307 1.37 953 (901–1,005)
Shrew 195 41 236 118 354 0.50 707 (633–781)
Stevng 409 32 441 224 665 1.20 552 (510–594)
Sthend 118 14 132 81 213 0.32 672 (582–763)
Stoke 305 79 384 311 695 0.89 781 (723–839)
Sund 198 22 220 201 421 0.62 681 (616–746)
Truro 154 23 177 200 377 0.41 913 (820–1,005)
Wirral 202 32 234 234 0.57 409 (357–462)
Wolve 285 92 377 151 528 0.67 790 (722–857)
York 135 32 167 229 396 0.49 805 (725–884)
40
Catchment 2012
Centre HD PD Dialysis Transplant RRT (millions) pmp (95% CI)
Northern Ireland
Antrim 132 13 145 80 225 0.30 750 (652–848)
Belfasta 228 28 256 445 701 0.55 1,275 (1,180–1,369)
Newry 91 16 107 81 188 0.28 671 (575–767)
Ulster 108 8 116 32 148 0.30 493 (414–573)
West NI 135 19 154 104 258 0.35 737 (647–827)
Scotland
Abrdn 230 25 255 249 504 0.60 840 (767–913)
Airdrie 194 11 205 183 388 0.56 693 (624–762)
D & Gall 51 16 67 61 128 0.15 853 (706–1,001)
Dundee 181 21 202 201 403 0.41 983 (887–1,079)
Dunfn 147 20 167 111 278 0.37 751 (663–840)
Edinba 265 37 302 420 722 0.96 752 (697–807)
Glasgwa 624 47 671 878 1,549 1.51 1,026 (975–1,077)
Inverns 74 18 92 126 218 0.34 641 (556–726)
Klmarnk 150 41 191 111 302 0.37 816 (724–908)
Wales
Bangor 90 15 105 105 0.22 481 (389–573)
Cardffa 482 77 559 989 1,548 1.42 1,090 (1,036–1,144)
Clwyd 84 18 102 70 172 0.19 907 (771–1,042)
Swanse 328 68 396 266 662 0.89 748 (691–805)
Wrexm 96 22 118 131 249 0.24 1,036 (908–1,165)
England 19,721 3,272 22,993 23,083 46,076
N Ireland 694 84 778 742 1,520
Scotland 1,916 236 2,152 2,340 4,492
Wales 1,080 200 1,280 1,456 2,736
UK 23,411 3,792 27,203 27,621 54,824
Blank cells indicate no patients on that treatment type attending that centre when data were collected
Centres prefixed ‘L’ are London centres
The numbers of patients calculated for each country quoted above differ marginally from those quoted elsewhere when patients are allocated to
areas by their individual post codes, as some centres treat patients from across national boundaries
a
Transplant centres
b
The catchment population for Plymouth may be too low, see appendix E
and Scotland, with slower growth in the prevalent RRT prevalence pattern by treatment modality is shown in
population in Northern Ireland. The increases in preva- figure 2.2. The steady growth in transplant numbers
lence across Scotland and England were similar at 4%. was maintained in 2012. The increase in haemodialysis
The increase in prevalence in Wales was 2.4%. In Northern patient numbers has been associated with an increase in
Ireland the increase in the prevalent RRT population was home haemodialysis, from 2.0% of the dialysis popu-
lower in magnitude at 1.5% between 2011 and 2012. lation in 2007 to 4.0% in 2012. The slow contraction in
From 2011 to 2012, there was a 0.7% pmp growth in PD observed in more recent years may have started to
prevalent HD patients, a 4.3% pmp growth in those plateau in 2012, with only a small reduction in the
with a functioning transplant and a 1.5% pmp decline prevalent PD population from 7.2% in 2011 to 6.9% in
in patients on PD. Between 2007 and 2012 there was an 2012.
average annual 2.6% pmp growth in HD, 4.8% pmp fall
in PD, and 4.6% pmp growth in prevalent transplant Prevalence of RRT in Primary Care Trusts in
patients in the UK (table 2.4). In the same period there England, Health and Social Care Areas in Northern
was an average annual 16.8% pmp growth in the use of Ireland (HBs), Local Health Boards in Wales (HBs)
home haemodialysis (data not shown). and Health Boards in Scotland (HBs)
Prevalence rates between centres showed marked The need for RRT depends on many factors such
variation (table 2.2); the long-term (1997–2012) UK as predisposing conditions but also on social and
41
Table 2.3. Number of prevalent patients on RRT by centre at year end 2008–2012
Date
% change % annual change
Centre 31/12/2008 31/12/2009 31/12/2010 31/12/2011 31/12/2012 2011–2012 2008–2012
England
B Heart 598 624 633 664 670 0.9 2.9
B QEH 1,714 1,821 1,844 1,912 1,971 3.1 3.6
Basldn 217 214 214 233 264 13.3 5.0
Bradfd 414 422 455 467 508 8.8 5.2
Brightn 722 737 770 775 831 7.2 3.6
Bristol 1,247 1,232 1,261 1,315 1,337 1.7 1.8
Camb 927 940 1,004 1,074 1,113 3.6 4.7
Carlis 205 205 206 215 216 0.5 1.3
Carsh 1,249 1,302 1,377 1,380 1,475 6.9 4.2
Chelms 207 225 238 216 224 3.7 2.0
Colchr 118 116 120 119 117 −1.7 −0.2
Covnt 745 794 844 874 900 3.0 4.8
Derby 389 419 459 448 477 6.5 5.2
Donc 154 196 222 248 261 5.2 14.1
Dorset 515 553 585 586 610 4.1 4.3
Dudley 275 292 303 284 316 11.3 3.5
Exeter 708 731 785 796 846 6.3 4.6
Glouc 325 366 377 381 417 9.4 6.4
Hull 696 725 725 757 789 4.2 3.2
Ipswi 294 312 316 340 339 −0.3 3.6
Kent 714 744 797 864 922 6.7 6.6
L Barts 1,526 1,638 1,778 1,872 1,955 4.4 6.4
L Guys 1,447 1,613 1,625 1,681 1,745 3.8 4.8
L Kings 784 786 837 858 918 7.0 4.0
L Rfree 1,510 1,546 1,639 1,727 1,865 8.0 5.4
L St.G 624 663 684 716 724 1.1 3.8
L West 2,576 2,734 2,879 3,020 3,104 2.8 4.8
Leeds 1,342 1,348 1,383 1,425 1,416 −0.6 1.4
Leic 1,660 1,737 1,809 1,927 1,982 2.9 4.5
Liv Ain 130 147 159 189 195 3.2 10.7
Liv RI 1,200 1,223 1,238 1,250 1,241 −0.7 0.8
M RI 1,424 1,450 1,552 1,646 1,710 3.9 4.7
Middlbr 682 707 711 752 789 4.9 3.7
Newc 901 898 900 917 946 3.2 1.2
Norwch 567 591 615 611 612 0.2 1.9
Nottm 955 975 1,008 1,019 1,006 −1.3 1.3
Oxford 1,318 1,343 1,421 1,446 1,535 6.2 3.9
Plymth 443 456 461 465 459 −1.3 0.9
Ports 1,268 1,301 1,333 1,394 1,447 3.8 3.4
Prestn 876 940 970 1,011 1,081 6.9 5.4
Redng 578 618 636 688 671 −2.5 3.8
Salford 758 784 837 820 882 7.6 3.9
Sheff 1,216 1,216 1,254 1,260 1,307 3.7 1.8
Shrew 325 337 343 342 354 3.5 2.2
Stevng 580 582 606 639 665 4.1 3.5
Sthend 204 207 212 207 213 2.9 1.1
Stoke 603 643 658 696 695 −0.1 3.6
Sund 343 368 369 388 421 8.5 5.3
Truro 297 320 335 352 377 7.1 6.1
Wirral 216 223 223 234 234 0.0 2.0
Wolve 490 490 531 513 528 2.9 1.9
York 276 321 338 340 396 16.5 9.4
42
Date
% change % annual change
Centre 31/12/2008 31/12/2009 31/12/2010 31/12/2011 31/12/2012 2011–2012 2008–2012
N Ireland
Antrim 220 215 217 220 225 2.3 0.6
Belfast 726 680 682 685 701 2.3 −0.9
Newry 163 170 177 190 188 −1.1 3.6
Ulster 97 114 115 137 148 8.0 11.1
West NI 236 258 256 266 258 −3.0 2.3
Scotland
Abrdn 456 452 462 478 504 5.4 2.5
Airdrie 245 310 326 344 388 12.8 12.2
D & Gall 113 118 118 122 128 4.9 3.2
Dundee 370 395 385 400 403 0.8 2.2
Dunfn 220 241 263 278 278 0.0 6.0
Edinb 695 721 730 700 722 3.1 1.0
Glasgw 1,568 1,469 1,505 1,477 1,549 4.9 −0.3
Inverns 212 228 230 223 218 −2.2 0.7
Klmarnk 263 273 284 299 302 1.0 3.5
Wales
Bangor 112 110 113 108 105 −2.8 −1.6
Cardff 1,374 1,426 1,517 1,534 1,548 0.9 3.0
Clwyd 146 144 142 136 172 26.5 4.2
Swanse 602 598 624 656 662 0.9 2.4
Wrexm 223 219 223 237 249 5.1 2.8
England 39,552 41,175 42,879 44,353 46,076 3.9 3.9
N Ireland 1,442 1,437 1,447 1,498 1,520 1.5 1.3
Scotland 4,142 4,207 4,303 4,321 4,492 4.0 2.0
Wales 2,457 2,497 2,619 2,671 2,736 2.4 2.7
UK 47,593 49,316 51,248 52,843 54,824 3.7 3.6
demographic factors such as age, gender, social depri- dardisation to compare RRT prevalence rates. The ethnic
vation and ethnicity. Hence, comparison of crude preva- minority profile is also provided to help understand the
lence rates by geographical area can be misleading. This differences in standardised prevalence ratios (SPRs).
section, as in previous reports, uses age and gender stan- The impact of social deprivation was reported in the
2003 UKRR Report [4].
There were substantial variations in the crude PCT/
60,000
PD HB prevalence rates pmp, from 430 pmp (Shetland,
50,000 Home HD population 23,200) to 1,630 pmp (Brent, population
HD 312,200). There were similar variations in the standar-
Number of patients
40,000 Transplant
dised prevalence ratios (ratio of observed:expected preva-
30,000
lence rate given the age/gender breakdown of the PCT/
HB) from 0.48 (Shetland) to 2.23 (Brent) (table 2.5).
20,000 Confidence intervals are not presented for the rates per
10,000
million population for 2012 but figures D3 and D4 in
appendix D (www.renalreg.com) can be used to deter-
0 mine if a PCT/HB falls within the range representing
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
the 95% confidence limit of the national average preva-

Year lence rate. The annual standardised prevalence ratios
Fig. 2.2. Growth in prevalent patients by treatment modality at were inherently more stable than the annual standardised
the end of each year 1997–2012 incidence ratios (chapter 1).
43
Table 2.4. Change in RRT prevalence rates pmp 2007–2012 by modality∗
Prevalence % growth in prevalence pmp
Year HD pmp PD pmp Dialysis pmp Transplant pmp RRT pmp HD PD Dialysis Tx RRT
2007 323 76 399 346 746
2008 342 69 411 363 774 5.8 −9.0 2.9 4.9 3.8
2009 354 64 417 377 794 3.5 −7.8 1.6 3.7 2.6
2010 359 62 421 397 818 1.5 −3.2 0.8 5.4 3.0
2011 365 60 426 416 841 1.7 −2.2 1.1 4.7 2.9
2012 367 60 427 434 861 0.7 −1.5 0.4 4.3 2.3
Average annual growth 2007–2012 2.6 −4.8 1.4 4.6 2.9
∗
Differences in the figures for dialysis and RRT prevalence and the sum of the separate modalities are due to rounding
Tx = transplant
Factors associated with variation in standardised standardised prevalence ratios in each region of England
prevalence ratios in Primary Care Trusts in England, and in Wales, Northern Ireland and Scotland are pre-
Health and Social Care Areas in Northern Ireland, sented in table 2.6. These calculations have not taken
Local Health Boards in Wales and Health Boards in into account variation in ethnicity between areas. Wales
Scotland and Northern Ireland previously had higher than
In 2012, there were 57 PCT/HBs with a significantly expected prevalence rates but in more recent years were
low SPR, 73 with a ‘normal’ SPR and 47 with a signifi- similar to their expected rates. Scotland had lower than
cantly high SPR (table 2.5). The areas with high and expected prevalence rates of RRT. There was marked
low SPRs have been fairly consistent over the last few variation (20–fold) in prevalence rates in over 80 year
years. They tend to reflect the demographics of the olds between PCT/HBs (data not shown).
regions in question such that urban, ethnically diverse
populations in areas of high social deprivation have the Case mix in prevalent RRT patients
highest prevalence rates of renal replacement therapy. Time on RRT (vintage)
Mean SPRs were significantly higher in the 75 PCT/ Table 2.7 shows the median time, in years, since start-
HBs with an ethnic minority population greater than ing RRT of prevalent RRT patients on 31st December
10% than in those with lower ethnic minority populations 2012. Median time on RRT for all prevalent patients
(p , 0.001). The SPR was positively correlated with the remained fairly static at 5.9 years. Patients with function-
percentage of the population that are non-White ing transplants had survived a median of 10.2 years on
(r = 0.69 p , 0.001). In 2012 for each 10% increase in RRT whilst the median time on RRT of HD and PD
ethnic minority population, the standardised prevalence patients was significantly less (3.4 and 1.7 years respect-
ratio increased by 0.16 (equates to �16%). In figure 2.3, ively, p , 0.001).
the relationship between the ethnic composition of a
PCT/HB and its SPR is demonstrated. Age
Only five of the 102 PCT/HBs with ethnic minority The median age of prevalent UK patients on RRT at
populations of less than 10% had high SPRs: Abertawe 31st December 2012 was static (58.3 years) compared
Bro Morgannwg University, Aneurin Bevan, Belfast, with 2011 (58.2 years) (table 2.8) and significantly higher
Cwm Taf, and Greater Glasgow & Clyde. Forty-two than in 2005 when it was 55 years. There were marked
(56%) of the 75 PCT/HBs with ethnic minority populations differences between modalities; the median age of HD
greater than 10% had high SPRs, whereas seven (9%) patients (66.4 years) was greater than that of those on
(Bedfordshire, Brighton and Hove City, Buckingham- PD (63.4 years) and substantially higher than that of
shire, Hertfordshire, Leeds, Richmond & Twickenham transplanted patients (52.3 years). Half of the UK
and Trafford) had low SPRs. However, not all PCT/HBs prevalent RRT population was in the 40–64 years age
with a high (.15%) ethnic minority population also group (table 2.9). The proportion of patients aged
had higher than expected RRT prevalence rates (e.g. 75 years and older was 17.1% in Wales, 16.1% in North-
Bromley, Oldham, Kensington). The age and gender ern Ireland, 15.7% in England and 13.4% in Scotland
44
Table 2.5. Prevalence of RRT and standardised prevalence ratios in PCT/HB areas
PCT/HB – PCT in England, Health and Social Care Areas in Northern Ireland, Local Health Boards in Wales and Health Boards in Scotland
O/E – standardised prevalence ratio. Ratio of observed:expected rate of RRT given the age and gender breakdown of the area
Blank cells – no data returned to the UKRR for that year
Areas with significantly low prevalence ratios in 2012 are italicised in greyed areas, those with significantly high prevalence ratios in 2012 are
bold in greyed areas
Population data from the Office for National Statistics, National Records of Scotland and the Northern Ireland Statistics and Research Agency –
based on the 2011 Census
% non-White – percentage of the PCT/HB population that is non-White, from 2011 Census for E, W & NI (2001 for Scotland)
ONS specifies that the populations should be rounded to the nearest 100 when being presented
2012 %
Total 2007 2008 2009 2010 2011 95% 95% Crude rate non-
UK area Name population O/E O/E O/E O/E O/E O/E LCL UCL pmp White
North East County Durham 513,000 0.90 0.87 0.86 0.85 0.87 0.87 0.79 0.96 801 1.8
Darlington 105,600 0.86 0.89 0.91 0.85 0.79 0.86 0.69 1.07 767 3.8
Gateshead 200,300 0.87 0.82 0.86 0.84 0.81 0.84 0.72 0.99 759 3.7
Hartlepool 92,100 0.88 0.94 0.92 0.86 0.85 0.91 0.72 1.14 804 2.3
Middlesbrough 138,400 1.07 1.10 1.10 1.10 1.10 1.14 0.96 1.35 932 11.8
Newcastle 279,100 1.00 1.01 0.97 0.91 0.90 0.89 0.77 1.02 692 14.5
North Tyneside 201,200 0.98 0.93 0.95 0.97 0.91 0.92 0.79 1.07 835 3.4
Northumberland 316,300 0.85 0.82 0.77 0.74 0.75 0.74 0.65 0.84 724 1.6
Redcar and Cleveland 135,200 1.04 1.02 1.01 0.97 1.01 0.95 0.79 1.13 888 1.5
South Tyneside 148,200 1.05 0.98 1.06 0.96 0.98 0.93 0.78 1.11 850 4.1
Stockton-on-Tees Teaching 191,800 0.84 0.84 0.81 0.80 0.85 0.87 0.74 1.02 751 5.4
Sunderland Teaching 275,300 0.96 0.99 0.97 0.97 0.93 0.93 0.82 1.06 839 4.1
North West Ashton, Leigh and Wigan 318,100 0.86 0.80 0.80 0.81 0.86 0.91 0.81 1.03 811 2.7
Blackburn with Darwen Teaching 147,700 1.31 1.23 1.23 1.21 1.23 1.21 1.03 1.43 941 30.8
Blackpool 142,100 0.77 0.80 0.86 0.81 0.81 0.93 0.78 1.11 859 3.3
Bolton Teaching 277,300 1.05 1.02 0.94 1.02 1.05 1.05 0.92 1.18 880 18.1
Bury 185,400 0.88 0.85 0.92 0.89 0.90 0.92 0.78 1.08 793 10.8
Central and Eastern Cheshire 462,800 0.82 0.79 0.79 0.79 0.79 0.79 0.71 0.88 741 3.1
Central Lancashire 467,400 0.80 0.83 0.86 0.85 0.85 0.88 0.79 0.97 774 7.8
Cumbria Teaching 499,800 0.76 0.75 0.73 0.72 0.70 0.70 0.63 0.78 688 1.5
East Lancashire Teaching 382,500 1.09 1.04 0.99 0.97 0.97 0.92 0.82 1.03 810 11.6
Halton and St Helens 301,100 0.93 0.86 0.88 0.90 0.94 0.94 0.83 1.06 837 2.0
Heywood, Middleton and Rochdale 211,900 0.99 0.99 1.02 0.95 1.00 1.01 0.87 1.17 830 18.3
Knowsley 145,900 1.14 1.08 1.05 0.96 0.96 0.97 0.81 1.16 836 2.8
Liverpool 465,700 1.06 1.07 1.08 1.04 1.04 1.01 0.91 1.12 816 11.1
Manchester Teaching 502,900 1.05 1.12 1.15 1.18 1.16 1.21 1.09 1.33 799 33.4
North Lancashire Teaching 321,600 0.78 0.75 0.76 0.74 0.75 0.76 0.67 0.86 731 3.1
Oldham 225,200 0.93 0.93 0.95 0.92 0.91 0.93 0.80 1.08 755 22.5
Salford 234,500 0.79 0.84 0.83 0.84 0.84 0.87 0.75 1.01 695 9.9
Sefton 274,000 0.87 0.83 0.83 0.86 0.92 0.89 0.78 1.01 850 2.6
Stockport 283,300 0.87 0.88 0.83 0.86 0.87 0.85 0.75 0.97 777 7.9
Tameside and Glossop 252,900 1.03 0.99 0.98 0.99 0.98 0.97 0.85 1.11 842 8.2
Trafford 227,100 0.76 0.72 0.75 0.86 0.82 0.85 0.73 0.99 735 14.5
Warrington 202,700 0.91 0.88 0.96 0.87 0.85 0.84 0.71 0.98 735 4.1
Western Cheshire 237,400 0.93 0.92 0.95 0.97 0.99 0.96 0.84 1.10 906 2.8
Wirral 319,800 0.93 0.86 0.83 0.80 0.80 0.80 0.70 0.91 738 3.0
Yorkshire Barnsley 231,900 1.05 1.06 1.08 1.12 1.10 1.06 0.93 1.21 957 2.1
and the Bradford and Airedale Teaching 523,100 1.13 1.12 1.10 1.17 1.16 1.22 1.12 1.34 941 32.6
Humber Calderdale 204,200 1.08 1.06 1.07 1.09 1.02 0.96 0.82 1.11 838 10.3
Doncaster 302,500 0.97 0.96 0.97 0.95 0.99 0.97 0.86 1.10 860 4.7
East Riding of Yorkshire 334,700 0.81 0.83 0.84 0.81 0.81 0.81 0.71 0.91 804 1.9
45

2012 %
Yorkshire Hull Teaching 256,100 1.08 1.00 1.05 1.03 0.99 0.96 0.84 1.11 777 5.9
and the Kirklees 423,000 1.08 1.01 1.01 1.02 1.01 1.02 0.92 1.13 856 20.9
Humber Leeds 750,700 0.98 0.93 0.91 0.91 0.90 0.87 0.80 0.95 701 14.9
North East Lincolnshire 161,200 0.99 1.00 0.98 0.97 1.03 1.01 0.86 1.18 900 2.6
North Lincolnshire 163,600 0.88 0.85 0.75 0.71 0.80 0.85 0.72 1.01 783 4.1
North Yorkshire and York 799,000 0.80 0.79 0.81 0.81 0.81 0.84 0.78 0.91 794 3.4
Rotherham 257,700 1.11 1.13 1.10 1.13 1.06 1.05 0.92 1.19 939 6.4
Sheffield 551,800 1.11 1.10 1.10 1.13 1.10 1.11 1.02 1.21 901 16.3
Wakefield District 326,400 0.84 0.81 0.81 0.81 0.84 0.86 0.76 0.97 772 4.6
East Bassetlaw 113,000 0.97 0.90 0.80 0.78 0.78 0.84 0.68 1.03 788 2.6
Midlands Derby City 248,900 0.99 1.05 1.14 1.11 1.10 1.15 1.01 1.31 928 19.7
Derbyshire County 737,500 0.87 0.87 0.85 0.84 0.85 0.82 0.76 0.89 773 2.5
Leicester City 329,600 1.66 1.68 1.69 1.73 1.75 1.76 1.60 1.93 1265 49.5
Leicestershire County and Rutland 688,800 0.90 0.89 0.88 0.89 0.87 0.86 0.79 0.94 790 8.3
Lincolnshire Teaching 717,200 0.78 0.77 0.76 0.77 0.80 0.79 0.73 0.86 761 2.4
Northamptonshire Teaching 694,000 0.91 0.92 0.91 0.90 0.91 0.91 0.83 0.99 784 8.5
Nottingham City 303,900 1.15 1.16 1.20 1.27 1.21 1.19 1.05 1.34 849 28.5
Nottinghamshire County Teaching 673,800 1.00 0.99 0.96 0.95 0.95 0.91 0.84 0.99 842 4.8
West Birmingham East and North 421,400 1.52 1.57 1.54 1.49 1.52 1.53 1.40 1.67 1179 36.1
Midlands Coventry Teaching 316,900 1.19 1.21 1.25 1.30 1.33 1.38 1.24 1.54 1079 26.2
Dudley 313,300 0.91 0.88 0.92 0.90 0.84 0.91 0.81 1.03 827 10.0
Heart of Birmingham Teaching 299,200 2.25 2.28 2.30 2.28 2.18 2.18 1.98 2.40 1380 70.5
Herefordshire 183,600 0.88 0.80 0.84 0.78 0.79 0.79 0.67 0.93 773 1.8
North Staffordshire 212,900 0.88 0.88 0.92 0.88 0.93 0.88 0.76 1.02 827 3.5
Sandwell 309,000 1.44 1.52 1.58 1.55 1.55 1.53 1.39 1.69 1233 30.1
Shropshire County 307,100 0.90 0.93 0.90 0.87 0.85 0.84 0.74 0.95 808 2.0
Solihull 206,900 0.97 0.92 0.98 0.95 0.92 0.87 0.75 1.02 802 10.9
South Birmingham 353,700 1.31 1.32 1.34 1.30 1.30 1.28 1.16 1.42 1001 25.3
South Staffordshire 628,500 0.91 0.91 0.87 0.87 0.90 0.84 0.77 0.92 781 4.7
Stoke on Trent 256,900 1.11 1.07 1.11 1.12 1.12 1.07 0.94 1.22 911 11.0
Telford and Wrekin 166,800 1.01 1.02 1.07 1.06 1.05 1.01 0.86 1.19 845 7.3
Walsall Teaching 269,500 1.22 1.28 1.24 1.32 1.30 1.26 1.12 1.42 1076 21.1
Warwickshire 546,600 1.02 0.98 1.00 1.01 1.01 0.98 0.90 1.07 900 7.3
Wolverhampton City 249,900 1.22 1.23 1.25 1.19 1.10 1.11 0.98 1.26 929 32.0
Worcestershire 566,600 0.84 0.84 0.85 0.85 0.86 0.87 0.79 0.95 819 4.3
East of Bedfordshire 413,500 0.85 0.87 0.86 0.88 0.86 0.88 0.79 0.99 771 11.2
England Cambridgeshire 622,300 0.87 0.83 0.85 0.88 0.91 0.87 0.80 0.96 758 7.4
Hertfordshire 1,119,800 0.82 0.91 0.91 0.93 0.93 0.92 0.86 0.98 780 12.4
Great Yarmouth and Waveney 212,800 0.54 0.82 0.91 0.97 0.97 0.93 0.80 1.07 902 2.7
Luton 203,600 1.21 1.27 1.26 1.28 1.35 1.36 1.18 1.56 992 45.3
Mid Essex 375,200 0.86 0.85 0.86 0.83 0.83 0.79 0.70 0.89 714 4.4
Norfolk 762,000 0.94 0.92 0.89 0.85 0.82 0.79 0.73 0.85 761 3.5
North East Essex 311,700 0.88 0.86 0.87 0.89 0.87 0.77 0.99 815 5.5
Peterborough 184,500 1.03 0.96 1.02 1.00 1.03 1.00 0.85 1.18 792 17.5
South East Essex 345,600 0.91 0.90 0.89 0.86 0.84 0.84 0.74 0.94 776 5.7
South West Essex 407,100 0.94 0.97 0.95 0.96 0.98 0.97 0.87 1.08 818 9.8
Suffolk 614,800 0.84 0.82 0.83 0.83 0.82 0.80 0.73 0.88 742 5.3
West Essex 289,600 0.75 0.69 0.71 0.75 0.74 0.82 0.72 0.94 732 8.1
London Barking and Dagenham 187,000 1.18 1.16 1.24 1.33 1.44 1.51 1.31 1.74 1027 41.7
Barnet 357,500 1.40 1.45 1.41 1.48 1.47 1.51 1.38 1.67 1172 35.9
Bexley 232,800 1.15 1.15 1.19 1.23 1.23 1.24 1.09 1.41 1044 18.1
Brent Teaching 312,200 1.79 2.01 2.10 2.20 2.20 2.23 2.04 2.43 1630 63.7
46

2012 %
London Bromley 310,600 1.00 1.04 1.00 1.03 1.02 0.99 0.88 1.12 866 15.7
Camden 220,100 1.14 1.21 1.24 1.27 1.29 1.30 1.13 1.48 954 33.7
City and Hackney Teaching 254,600 1.35 1.28 1.35 1.44 1.51 1.56 1.39 1.77 1017 44.6
Croydon 364,800 1.31 1.31 1.38 1.37 1.42 1.45 1.31 1.59 1118 44.9
Ealing 339,300 1.56 1.84 1.86 1.88 1.86 1.92 1.75 2.10 1426 51.0
Enfield 313,900 1.40 1.41 1.39 1.40 1.51 1.52 1.37 1.68 1147 39.0
Greenwich Teaching 255,500 1.06 1.15 1.18 1.30 1.34 1.34 1.19 1.52 959 37.5
Hammersmith and Fulham 182,400 1.19 1.24 1.31 1.31 1.34 1.38 1.19 1.60 970 31.9
Haringey Teaching 255,500 1.36 1.41 1.41 1.42 1.56 1.69 1.51 1.89 1186 39.5
Harrow 240,500 1.45 1.67 1.76 1.82 1.87 1.86 1.67 2.06 1497 57.8
Havering 237,900 0.82 0.82 0.84 0.82 0.87 0.88 0.76 1.02 778 12.3
Hillingdon 275,500 0.92 1.31 1.32 1.33 1.41 1.44 1.29 1.61 1111 39.4
Hounslow 254,900 1.20 1.41 1.44 1.51 1.58 1.62 1.44 1.81 1192 48.6
Islington 206,300 1.35 1.27 1.29 1.37 1.42 1.58 1.38 1.79 1100 31.8
Kensington and Chelsea 158,300 0.88 1.09 1.08 1.12 1.11 1.08 0.91 1.27 872 29.4
Kingston 160,400 1.08 1.19 1.16 1.14 1.16 1.17 1.00 1.38 916 25.5
Lambeth 304,500 1.56 1.55 1.61 1.58 1.67 1.73 1.56 1.92 1176 42.9
Lewisham 276,900 1.65 1.62 1.71 1.65 1.71 1.75 1.58 1.95 1246 46.5
Newham 310,500 1.48 1.52 1.57 1.77 1.87 1.90 1.72 2.11 1166 71.0
Redbridge 281,400 1.22 1.34 1.39 1.47 1.43 1.50 1.34 1.67 1112 57.5
Richmond and Twickenham 187,500 0.64 0.70 0.76 0.77 0.77 0.77 0.65 0.93 640 14.0
Southwark 288,700 1.65 1.68 1.71 1.76 1.85 1.89 1.71 2.09 1288 45.8
Sutton and Merton 391,700 1.17 1.20 1.24 1.26 1.27 1.33 1.21 1.46 1054 28.4
Tower Hamlets 256,000 1.22 1.26 1.39 1.44 1.47 1.54 1.36 1.76 914 54.8
Waltham Forest 259,700 1.44 1.42 1.40 1.46 1.56 1.52 1.35 1.71 1078 47.8
Wandsworth 307,700 1.33 1.33 1.40 1.38 1.34 1.29 1.15 1.45 894 28.6
Westminster 219,600 0.98 1.12 1.20 1.22 1.30 1.31 1.15 1.50 1016 38.3
South East Brighton and Hove City 273,000 0.83 0.84 0.83 0.81 0.81 0.83 0.72 0.96 656 10.9
Coast East Sussex Downs and Weald 343,900 0.80 0.75 0.71 0.71 0.69 0.76 0.67 0.85 739 3.8
Eastern and Coastal Kent 759,600 0.87 0.92 0.93 0.96 0.96 0.97 0.90 1.05 878 5.0
Hastings and Rother 183,400 0.76 0.79 0.74 0.78 0.76 0.74 0.62 0.88 725 4.5
Medway 264,900 0.86 0.90 0.90 0.90 0.91 0.92 0.80 1.05 751 10.4
Surrey 1,124,800 0.85 0.87 0.88 0.88 0.87 0.89 0.84 0.95 794 9.5
West Kent 706,800 0.85 0.88 0.89 0.86 0.85 0.87 0.80 0.95 768 7.7
West Sussex 808,900 0.82 0.83 0.83 0.84 0.82 0.80 0.74 0.87 753 6.2
South Berkshire East 410,100 1.16 1.16 1.19 1.22 1.25 1.25 1.13 1.37 983 26.6
Central Berkshire West 464,400 1.11 1.10 1.11 1.04 1.05 1.00 0.91 1.11 835 14.0
Buckinghamshire 521,000 0.95 0.94 0.92 0.91 0.87 0.86 0.78 0.95 764 13.3
Hampshire 1,322,100 0.77 0.79 0.81 0.80 0.79 0.78 0.73 0.83 719 5.0
Isle of Wight National Health Service 138,400 0.59 0.59 0.55 0.56 0.60 0.64 0.52 0.79 650 2.7
Milton Keynes 255,400 0.93 0.95 0.93 0.95 0.96 0.97 0.84 1.11 752 19.6
Oxfordshire 629,600 0.96 0.92 0.89 0.90 0.92 0.93 0.85 1.02 789 9.4
Portsmouth City Teaching 205,400 0.97 0.97 0.93 0.91 0.97 0.99 0.84 1.15 755 11.6
Southampton City 235,900 0.91 0.95 0.94 0.99 1.02 1.04 0.90 1.20 784 14.1
South West Bath and North East Somerset 175,500 0.92 0.84 0.87 0.85 0.79 0.79 0.66 0.94 695 5.4
Bournemouth and Poole Teaching 331,500 0.86 0.84 0.81 0.79 0.79 0.77 0.68 0.88 688 6.3
Bristol 428,100 1.25 1.30 1.26 1.22 1.23 1.27 1.16 1.40 972 16.0
Cornwall and Isles of Scilly 536,000 1.03 1.00 1.00 0.97 0.93 0.93 0.85 1.01 910 1.8
Devon 747,700 0.85 0.87 0.87 0.86 0.85 0.87 0.81 0.94 860 2.5
Dorset 413,800 0.83 0.86 0.86 0.84 0.79 0.80 0.72 0.89 826 2.1
Gloucestershire 598,300 0.88 0.82 0.85 0.85 0.86 0.88 0.80 0.96 807 4.6
North Somerset 203,100 0.96 0.97 0.90 0.88 0.89 0.91 0.79 1.06 876 2.7
47

2012 %
South West Plymouth Teaching 256,600 1.14 1.12 1.11 1.15 1.15 1.12 0.99 1.27 951 3.9
Somerset 531,600 0.84 0.81 0.82 0.84 0.87 0.84 0.76 0.92 805 2.0
South Gloucestershire 263,400 1.01 1.00 0.94 1.00 0.97 0.95 0.83 1.08 839 5.0
Swindon 214,900 0.86 0.85 0.87 0.90 0.94 0.96 0.82 1.11 796 10.0
Torbay 131,200 0.84 0.94 0.88 0.93 0.96 0.98 0.82 1.16 976 2.5
Wiltshire 474,300 0.72 0.75 0.73 0.73 0.74 0.72 0.64 0.80 651 3.4
Wales Betsi Cadwaladr University 688,700 0.96 0.94 0.91 0.89 0.84 0.86 0.79 0.93 807 2.5
Powys Teaching 133,200 0.89 0.89 0.94 0.88 0.86 0.87 0.73 1.05 886 1.6
Hywel Dda 381,900 0.96 1.00 0.96 0.91 0.94 0.88 0.79 0.98 843 2.2
Abertawe Bro Morgannwg University 517,700 1.25 1.17 1.20 1.24 1.23 1.20 1.11 1.31 1084 3.9
Cwm Taf 293,500 1.51 1.43 1.39 1.31 1.35 1.27 1.14 1.41 1118 2.6
Aneurin Bevan 577,000 1.16 1.09 1.08 1.11 1.09 1.09 1.00 1.18 974 3.9
Cardiff and Vale University 472,300 1.16 1.06 1.07 1.06 1.05 1.02 0.93 1.13 822 12.2
Scotland Ayrshire & Arran 373,800 1.13 1.14 1.08 1.07 1.01 0.99 0.89 1.10 939 0.7
Borders 113,900 0.97 1.01 1.03 1.08 0.97 0.91 0.75 1.11 913 0.6
Dumfries and Galloway 151,400 0.95 0.96 0.93 0.90 0.87 0.87 0.74 1.03 878 0.7
Fife 365,300 0.97 0.96 0.95 0.95 0.99 0.97 0.87 1.08 881 1.3
Forth Valley 298,100 0.96 0.93 0.92 0.93 0.89 0.87 0.76 0.99 778 1.1
Grampian 569,600 1.01 0.98 0.95 0.95 0.94 0.96 0.88 1.05 853 1.6
Greater Glasgow & Clyde 1,214,600 1.17 1.13 1.09 1.06 1.06 1.08 1.02 1.14 925 3.4
Highland 321,700 1.11 1.05 1.03 0.99 0.90 0.86 0.76 0.97 833 0.8
Lanarkshire 572,400 0.98 0.96 0.95 0.96 0.93 0.98 0.89 1.07 865 1.2
Lothian 836,600 0.96 0.93 0.90 0.86 0.82 0.82 0.76 0.89 694 2.8
Orkney 21,400 0.89 1.07 1.02 0.93 0.79 0.76 0.47 1.24 747 0.4
Shetland 23,200 0.71 0.50 0.54 0.57 0.49 0.48 0.26 0.89 430 1.1
Tayside 410,300 1.13 1.05 1.07 1.03 1.02 0.97 0.88 1.08 897 1.9
Western Isles 27,700 0.81 0.72 0.69 0.82 0.67 0.58 0.35 0.94 578 0.6
Northern Belfast 348,300 1.34 1.28 1.18 1.18 1.15 1.16 1.04 1.30 933 3.2
Ireland Northern 463,500 1.14 1.10 1.05 1.01 1.05 1.05 0.96 1.16 878 1.2
Southern 359,400 1.00 1.00 0.98 1.01 1.05 0.98 0.87 1.10 765 1.2
South Eastern 347,700 1.00 0.99 0.96 0.89 0.92 0.89 0.79 1.01 759 1.3
Western 295,300 1.15 1.12 1.15 1.14 1.08 1.00 0.88 1.13 792 1.0
(table 2.9). Furthermore, there existed a wide range In the UK on 31st December 2012, 63.5% of patients
between centres in the proportion of patients aged over aged less than 65 years on RRT had a functioning
75 (9.2% in Liverpool RI to 36.8% in Colchester). transplant (table 2.15), compared with only 26.9% aged
Colchester had the highest median age (70.4 years), 65 years and over. There was a similar pattern in all
whilst Belfast the lowest (53.8 years) (table 2.8). This four UK countries.
could reflect either variation in the demography of the
catchment populations or follow-up of younger trans- Gender
plant patients (as above in the case of Belfast). The Standardising the age of the UK RRT prevalent patients,
median age of the non-White dialysis population was by using the age and gender distribution of the UK popu-
lower than the overall dialysis population (60.9 vs. 66.1 lation by PCT/HB (from mid-2011 population estimates),
years, data not shown). The differing age distributions allowed estimation of crude prevalence rates by age and
of the transplant and dialysis populations are illustrated gender (figure 2.5). This shows a progressive increase in
in figure 2.4, demonstrating that the age peak for preva- prevalence rate with age, peaking at 2,138 pmp (a slight
lent dialysis patients is 24 years later than for prevalent increase from 2,099 pmp in 2011) in the age group 75–
transplant patients. 79 years before showing a reducing prevalence rate in
48
2.4 Table 2.7. Median time on RRT of prevalent patients on

31/12/2012
2.0 Median time treated

Modality N (years)
Standardised prevalence ratio
1.6 Haemodialysis 23,034 3.4

Peritoneal dialysis 3,752 1.7
Transplant 26,365 10.2
North East
1.2 North West All RRT 53,151 5.9
Yorkshire and the Humber
East Midlands All patients without a treatment modality were excluded
West Midlands Median time on RRT was calculated from the most recent start date.
0.8 East of England
London For patients who recovered for .90 days and then subsequently
South East Coast restarted RRT the median time from the start of RRT was calculated
South Central from the most recent start date
0.4 South West
Wales Patients with an initial treatment modality of transferred in or trans-
Scotland ferred out were excluded from the calculation of median time on RRT
Northern Ireland since their treatment start date was not accurately known
0.0
0 20 40 60 80
% non-White
improvement compared with 51 of 71 (71.8%) in 2011
Fig. 2.3. Standardised prevalence ratios for all PCT/HB areas by and 36 centres in 2006. Ethnicity completeness for preva-
percentage non-White on 31/12/2012 (excluding areas with <5% lent RRT patients improved in the UK from 88.6% in
ethnic minorities)
2011 to 92.0% in 2012, with 97.9% ethnicity completeness
in England, 99.9% completeness in Wales and 100% in
Northern Ireland. Completeness of ethnicity data was
age groups over 80 years. Crude prevalence rates in males highest in prevalent transplant patients. This may relate
exceeded those of females for all age groups, peaking in age to the fact that the intensive work-up for transplantation
group 80–84 years at 2,973 pmp and for females in age may increase the recording of data. Completeness of
group 75–79 years at 1,528 pmp. Survival on RRT is ethnicity data from Scotland was low at 33.6%.
described in chapter 8. In 2012, 20.7% of the prevalent UK RRT population
(with ethnicity assigned) were from ethnic minorities
Ethnicity (22.7% in England). The proportion of the prevalent
Fifty-nine of the 71 centres (83.1%) provided ethnicity UK RRT population (with ethnicity assigned) from
data that were at least 90% complete (table 2.10), an ethnic minorities in Wales, Scotland and Northern Ire-
Table 2.6. Standardised prevalence rate ratio of RRT for each Strategic Health Authority in England and for Wales, Scotland and
Northern Ireland in 2012
UK Area Total population O/E 95% LCL 95% UCL Crude rate pmp
North East 2,596,400 0.88 0.85 0.92 792.6
North West 7,089,100 0.91 0.88 0.93 790.2
Yorkshire and the Humber 5,285,700 0.96 0.93 0.99 832.2
East Midlands 4,506,800 0.94 0.91 0.97 835.6
West Midlands 5,608,700 1.10 1.07 1.13 948.9
East of England 5,862,400 0.88 0.85 0.90 780.6
London 8,204,400 1.49 1.46 1.52 1,101.8
South East Coast 4,465,200 0.87 0.84 0.89 778.7
South Central 4,182,300 0.91 0.88 0.94 779.0
South West 5,306,100 0.89 0.87 0.92 829.4
Wales 3,064,300 1.02 0.99 1.06 925.2
Scotland 5,299,900 0.95 0.92 0.98 850.2
Northern Ireland 1,814,300 1.02 0.97 1.07 829.5
O/E – observed/expected prevalence rate ratio given the age/gender breakdown of each region
Bold – higher than expected prevalence rate ratio
49
Table 2.8. Median age of prevalent RRT patients by treatment modality in renal centres on 31/12/2012
Median age Median age
Centre HD PD Transplant RRT Centre HD PD Transplant RRT
England Redng 69.4 62.4 56.5 60.3
B Heart 66.6 53.9 50.8 62.5 Salford 64.2 59.7 51.2 57.7
B QEH 64.7 58.0 51.3 57.1 Sheff 65.5 64.2 52.0 58.4
Basldn 67.8 65.3 50.8 64.2 Shrew 67.6 61.9 53.9 62.2
Bradfd 61.8 56.6 50.6 54.3 Stevng 67.1 66.2 51.8 60.6
Brightn 69.2 66.8 53.8 62.3 Sthend 72.1 65.1 54.9 65.6
Bristol 68.9 56.0 53.4 58.2 Stoke 66.3 68.6 50.8 59.3
Camb 72.1 71.3 52.6 58.3 Sund 65.5 60.4 53.3 58.1
Carlis 67.2 62.7 52.7 58.4 Truro 71.9 67.0 57.5 63.9
Carsh 68.9 66.4 52.3 62.0 Wirral 65.0 60.2 64.9
Chelms 68.0 65.8 59.3 65.3 Wolve 66.7 59.0 51.7 59.8
Colchr 70.4 70.4 York 66.4 56.8 52.0 57.4
Covnt 68.0 66.6 50.9 58.8 N Ireland
Derby 66.9 64.3 54.2 61.7 Antrim 70.9 60.4 51.1 64.7
Donc 66.3 62.6 56.1 64.0 Belfast 64.5 60.8 50.8 53.8
Dorset 71.5 69.8 57.8 64.7 Newry 65.3 69.7 52.5 60.4
Dudley 69.0 61.9 56.9 63.0 Ulster 71.7 64.9 56.7 69.1
Exeter 72.2 68.3 53.2 62.9 West NI 66.8 46.7 50.7 59.5
Glouc 71.5 68.5 55.5 64.5
Scotland
Hull 66.9 62.1 51.5 58.8
Abrdn 66.2 57.0 52.5 57.2
Ipswi 66.3 66.3 54.0 59.3
Airdrie 62.6 51.5 51.6 56.6
Kent 69.6 64.5 53.4 60.4
L Barts 60.1 60.3 50.5 55.1 D & Gall 64.7 69.8 51.5 60.7
L Guys 62.6 58.8 49.8 54.0 Dundee 69.7 65.5 52.7 61.3
L Kings 63.2 60.8 52.8 58.1 Dunfn 66.5 62.0 51.3 60.0
L Rfree 67.6 63.0 51.1 57.2 Edinb 60.1 69.6 51.5 54.8
L St.G 66.7 62.2 53.9 59.9 Glasgw 65.0 63.9 52.9 56.9
L West 65.8 62.1 53.5 58.8 Inverns 68.9 65.2 47.9 54.8
Leeds 66.8 55.1 52.0 56.8 Klmarnk 66.2 59.1 50.4 57.9
Leic 66.2 66.0 52.4 59.5 Wales
Liv Ain 67.1 59.9 65.9 Bangor 66.0 67.1 66.1
Liv RI 61.8 58.1 51.8 54.9 Cardff 68.3 68.1 52.1 57.1
M RI 62.8 61.8 50.3 54.0 Clwyd 65.5 71.0 57.0 62.2
Middlbr 67.3 55.5 52.6 57.6 Swanse 71.1 62.9 56.5 63.7
Newc 62.5 64.1 54.3 57.0 Wrexm 71.5 62.9 52.6 57.9
Norwch 71.7 65.1 53.9 63.4 England 66.5 63.3 52.3 58.4
Nottm 68.7 62.7 51.1 57.4 N Ireland 67.8 64.1 51.1 58.2
Oxford 66.3 64.6 51.1 55.7 Scotland 65.0 63.3 51.8 57.2
Plymth 68.4 67.4 54.5 59.0 Wales 68.6 66.1 52.9 59.7
Ports 66.2 63.8 53.1 58.3 UK 66.4 63.4 52.3 58.3
Prestn 63.9 65.9 52.7 58.5
Blank cells indicate no patients on that treatment modality attending that centre when data were collected
land were very small, although it should be noted that coding and reporting of ethnicity data as well as an
there was a high level of missing ethnicity data in increasing incidence of ERF and increased referral rates
Scotland. The ONS estimates that approximately 14% in these populations.
of the UK general population are designated as belonging Amongst the centres with more than 50% returns
to an ethnic minority [1]. The relative proportion of there was wide variation in the proportion of patients
patients reported to the UKRR as receiving RRT and from ethnic minorities, ranging from 0.5% in two centres
belonging to an ethnic minority has increased from (Truro and Newry) to over 50% in 3 centres: London
14.9% in 2007 which may be due to improvements in Barts (60.2%), London West (55.5%) and London Royal
50
Table 2.9. Percentage of prevalent RRT patients in each age group by centre on 31/12/2012
Centre N 18–39 years 40–64 years 65–74 years 75+ years

England
B Heart 670 14.0 42.8 22.8 20.3
B QEH 1,971 14.9 52.0 17.7 15.4
Basldn 264 12.5 39.4 22.0 26.1
Bradfd 508 20.7 48.8 19.1 11.4
Brightn 831 11.4 44.8 22.4 21.4
Bristol 1,337 16.1 47.9 20.0 15.9
Camb 1,113 14.0 50.7 20.0 15.3
Carlis 216 13.9 53.2 19.9 13.0
Carsh 1,475 10.6 45.9 22.6 20.8
Chelms 224 7.6 41.1 25.4 25.9
Colchr 117 5.1 27.4 30.8 36.8
Covnt 900 12.9 48.7 19.8 18.7
Derby 477 11.7 45.5 24.3 18.4
Donc 261 11.5 42.9 21.5 24.1
Dorset 610 9.8 41.1 28.7 20.3
Dudley 316 7.3 48.7 20.9 23.1
Exeter 846 10.0 44.4 23.6 21.9
Glouc 417 10.1 42.2 23.0 24.7
Hull 789 13.6 50.6 20.2 15.7
Ipswi 339 10.3 54.9 21.8 13.0
Kent 922 12.9 46.3 23.8 17.0
L Barts 1,955 17.3 55.1 16.6 11.0
L Guys 1,745 19.7 53.6 15.6 11.0
L Kings 918 12.3 51.7 20.4 15.6
L Rfree 1,865 17.8 48.3 18.4 15.5
L St.G 724 13.7 49.9 19.6 16.9
L West 3,104 12.0 52.8 21.3 13.8
Leeds 1,416 17.6 50.0 19.8 12.6
Leic 1,982 13.6 49.1 22.4 14.9
Liv Ain 195 8.7 38.5 24.1 28.7
Liv RI 1,241 16.0 57.7 17.2 9.2
M RI 1,710 18.4 55.6 16.4 9.6
Middlbr 789 13.7 50.4 19.0 16.9
Newc 946 14.4 53.7 21.5 10.5
Norwch 612 11.3 41.8 22.5 24.3
Nottm 1,006 16.3 48.8 19.1 15.8
Oxford 1,535 16.3 53.0 17.6 13.1
Plymth 459 13.5 49.5 24.4 12.6
Ports 1,447 14.0 50.9 20.6 14.5
Prestn 1,081 12.4 53.4 20.5 13.7
Redng 671 10.1 49.2 22.5 18.2
Salford 882 13.8 52.3 20.6 13.3
Sheff 1,307 13.8 51.6 19.2 15.3
Shrew 354 12.1 44.4 21.2 22.3
Stevng 665 12.2 46.6 20.5 20.8
Sthend 213 13.6 34.7 24.4 27.2
Stoke 695 14.8 46.5 20.0 18.7
Sund 421 12.8 52.7 21.6 12.8
Truro 377 12.2 40.3 24.4 23.1
Wirral 234 7.7 43.2 21.8 27.4
Wolve 528 10.8 49.8 20.8 18.6
York 396 19.2 46.2 21.7 12.9
51
Centre N 18–39 years 40–64 years 65–74 years 75+ years

N Ireland
Antrim 225 10.2 40.4 25.3 24.0
Belfast 701 17.4 54.6 17.1 10.8
Newry 188 14.9 47.9 22.9 14.4
Ulster 148 9.5 33.1 26.4 31.1
West NI 258 17.1 45.3 21.7 15.9
Scotland
Abrdn 504 19.0 50.2 17.5 13.3
Airdrie 388 15.5 52.1 18.0 14.4
D & Gall 128 12.5 47.7 22.7 17.2
Dundee 403 12.2 46.2 22.1 19.6
Dunfn 278 13.3 46.8 24.1 15.8
Edinb 722 15.5 56.6 18.1 9.7
Glasgw 1,549 13.6 55.5 18.7 12.2
Inverns 218 15.1 56.4 13.8 14.7
Klmarnk 302 10.6 52.6 21.9 14.9
Wales
Bangor 105 8.6 37.1 30.5 23.8
Cardff 1,548 15.1 51.6 19.6 13.7
Clwyd 172 12.2 45.3 26.7 15.7
Swanse 662 10.7 42.6 24.0 22.7
Wrexm 249 16.5 44.6 17.7 21.3
England 46,076 14.2 49.8 20.3 15.7
N Ireland 1,520 15.2 48.0 20.7 16.1
Scotland 4,492 14.4 53.0 19.1 13.4
Wales 2,736 13.7 47.8 21.4 17.1
UK 54,824 14.2 50.0 20.2 15.6
(min : max) (5.1 : 20.7) (27.4 : 57.7) (13.8 : 30.8) (9.2 : 36.8)
Free (50.9%). Three additional centres had over 40% Primary renal diagnosis
of prevalent patients from ethnic minorities: Bradford Data for primary renal diagnosis (PRD) were not com-
(42.3%), London Kings (48.5%) and London St Georges plete for 3.6% of patients (table 2.11) and there remained
(44.6%). a marked inter-centre difference in completeness of data
3.0
Transplant
3,000
Dialysis Males
2.5 Females
2,500
All UK
Prevalence rate pmp
2.0 2,000
1.5 1,500
1,000
1.0
500
0.5
0
20–24
25–29
30–34
35–39
40–44
45–49
50–54
55–59
60–64
65–69
70–74
75–79
80–84
85+
0.0
15 25 35 45 55 65 75 85 95
Age (years) Age group
Fig. 2.4. Age profile of prevalent RRT patients by modality on Fig. 2.5. Prevalence rate of RRT patients per million population
31/12/2012 by age and gender on 31/12/2012
52
Table 2.10. Ethnicity of prevalent RRT patients by centre on 31/12/2012

Data not N
Centre available with data White Black S Asian Chinese Other
England
B Heart 0.0 670 61.9 7.2 29.6 0.6 0.7
B QEH 0.0 1,971 64.3 9.0 23.4 0.9 2.4
Basldn 0.0 264 85.2 8.3 4.2 0.8 1.5
Bradfd 1.4 501 57.7 1.8 39.7 0.0 0.8
Brightn 3.7 800 92.1 2.9 3.5 0.3 1.3
Bristol 0.4 1,331 89.9 5.0 3.6 0.4 1.1
Camb 1.2 1,100 93.1 1.9 4.0 0.2 0.8
Carlis 0.0 216 99.1 0.0 0.9 0.0 0.0
Carsh 6.8 1,374 72.9 9.6 12.7 1.5 3.3
Chelms 5.4 212 92.5 2.8 1.9 0.9 1.9
Colchr 0.0 117 95.7 0.9 0.9 0.9 1.7
Covnt 0.3 897 81.7 4.0 13.5 0.7 0.1
Derby 1.3 471 82.0 3.8 13.2 0.6 0.4
Donc 0.0 261 96.6 1.1 1.1 0.8 0.4
Dorset 0.0 610 97.5 0.2 0.7 0.5 1.1
Dudley 0.0 316 86.4 2.8 8.9 0.6 1.3
Exeter 0.2 844 98.3 0.6 0.4 0.2 0.5
Glouc 0.0 417 94.2 1.7 2.9 0.0 1.2
Hull 37.3 495 97.2 0.6 1.6 0.2 0.4
Ipswi 1.2 335 94.0 3.0 2.7 0.3 0.0
Kent 0.7 916 94.9 0.7 3.2 0.1 1.2
L Barts 0.0 1,955 39.8 32.4 25.8 1.5 0.4
L Guys 0.9 1,730 67.0 22.0 6.4 1.2 3.4
L Kings 1.9 901 51.5 34.4 11.0 1.7 1.4
L Rfree 3.1 1,807 49.1 21.7 19.0 1.6 8.5
L St.G 11.7 639 55.4 22.4 12.8 2.2 7.2
L West 0.0 3,104 44.5 17.9 33.5 1.0 3.1
Leeds 0.7 1,406 81.2 4.3 13.2 0.1 1.1
Leic 1.8 1,946 77.3 3.6 17.6 0.3 1.2
Liv Ain 2.1 191 95.8 0.5 2.1 1.0 0.5
Liv RI 1.4 1,224 93.5 2.0 1.6 1.4 1.6
M RI 0.6 1,699 79.1 6.2 11.8 0.9 2.0
Middlbr 0.3 787 94.0 0.6 5.0 0.3 0.1
Newc 0.1 945 93.4 0.7 4.2 0.7 0.8
Norwch 0.5 609 96.1 0.5 1.0 2.1 0.3
Nottm 0.0 1,006 87.0 5.1 6.5 0.0 1.5
Oxford 2.0 1,505 83.7 3.7 9.3 0.6 2.7
Plymth 0.0 459 97.4 0.7 0.4 0.7 0.9
Ports 1.1 1,431 94.3 0.9 3.1 0.0 1.7
Prestn 0.0 1,081 85.8 0.8 12.8 0.0 0.6
Redng 4.9 638 71.2 6.7 20.4 0.2 1.6
Salford 1.6 868 82.1 1.7 14.2 0.5 1.5
Sheff 0.5 1,301 91.8 2.2 3.8 0.7 1.5
Shrew 0.6 352 96.0 1.4 2.3 0.0 0.3
Stevng 0.3 663 69.4 9.5 17.6 0.6 2.9
Sthend 0.0 213 84.5 2.3 4.2 2.3 6.6
Stoke 15.5 587 93.4 0.3 4.3 0.3 1.7
Sund 0.2 420 96.9 0.7 2.1 0.2 0.0
Truro 0.0 377 99.5 0.0 0.3 0.0 0.3
Wirral 1.3 231 95.2 0.4 1.7 1.3 1.3
Wolve 0.0 528 71.4 9.1 19.3 0.2 0.0
York 3.3 383 97.4 0.5 1.6 0.0 0.5
53

Data not N
Centre available with data White Black S Asian Chinese Other
N Ireland
Antrim 0.0 225 99.1 0.0 0.9 0.0 0.0
Belfast 0.0 701 98.4 0.1 1.0 0.3 0.1
Newry 0.0 188 99.5 0.0 0.0 0.5 0.0
Ulster 0.0 148 97.3 0.0 2.0 0.7 0.0
West NI 0.0 258 98.4 0.4 0.8 0.4 0.0
Scotland
Abrdn 60.7 198
Airdrie 66.2 131
D & Gall 88.3 15
Dundee 55.8 178
Dunfn 82.4 49
Edinb 93.1 50
Glasgw 92.5 116
Inverns 14.7 186 98.9 0.0 1.1 0.0 0.0
Klmarnk 55.0 136
Wales
Bangor 0.0 105 96.2 1.0 1.0 0.0 1.9
Cardff 0.2 1,545 94.2 0.9 4.0 0.5 0.4
Clwyd 0.6 171 98.8 0.0 0.6 0.6 0.0
Swanse 0.0 662 97.9 0.3 1.5 0.0 0.3
Wrexm 0.0 249 99.2 0.4 0.4 0.0 0.0
England 2.1 45,104 77.3 8.1 12.1 0.7 1.9
N Ireland 0.0 1,520 98.6 0.1 0.9 0.3 0.1
Scotland 76.4 1,059
Wales 0.1 2,732 95.9 0.7 2.7 0.3 0.4
UK 8.0 50,415 79.3 7.3 11.0 0.7 1.7
Percentage breakdown is not shown for centres with less than 50% data completeness, but these centres are included in national averages
Blank cells – less than 50% data completeness
Appendix H ethnicity coding
returns. Only one centre had 540% primary renal diag- chester, 48% uncertain PRD); the UK and national totals
nosis data coded as uncertain and has been excluded have been appropriately adjusted. The range for the
from the between centre analysis and other analyses remaining 70 centres was between 5.0% and 34.5%, and
where PRD is included in the case-mix adjustment (Col- has shown improvement over time. Completeness of
Table 2.11. Primary renal diagnosis in prevalent RRT patients by age and gender on 31/12/2012
Inter- Age ,65 Age 565

% all centre M:F
Primary diagnosis∗ N patients range % N % N % ratio
Aetiology uncertain 9,154 16.7 5.0–34.5 5,092 14.5 4,062 20.7 1.6
Glomerulonephritis 10,289 18.8 8.5–28.6 7,523 21.4 2,766 14.1 2.1
Pyelonephritis 6,008 11.0 3.9–18.5 4,473 12.7 1,535 7.8 1.1
Diabetes 8,456 15.5 9.6–24.9 5,064 14.4 3,392 17.3 1.6
Polycystic kidney 5,286 9.7 4.1–16.7 3,510 10.0 1,776 9.1 1.1
Hypertension 3,249 5.9 1.5–15.4 1,773 5.0 1,476 7.5 2.4
Renal vascular disease 1,743 3.2 0.6–9.1 354 1.0 1,389 7.1 2.0
Other 8,568 15.7 9.5–25.3 6,071 17.3 2,497 12.8 1.3
Not sent 1,954 3.6 0.2–37.5 1,266 3.6 688 3.5 1.6
∗
Appendix H: ERA-EDTA coding
Excluded centre: 540% primary renal diagnosis aetiology uncertain (Colchr)
54
PRD data has also continued to improve and no centres 2012, from 7,798 in 2011, representing 15.5% of all preva-
had .50% missing data in 2012. lent patients (compared with 13.5% in 2006) (table 2.13).
Glomerulonephritis (GN) remained the most common The median age at start of RRT for patients with diabetes
primary renal diagnosis in the 2012 prevalent cohort at (56 years) was nine years higher compared with patients
18.8% (table 2.11). Diabetes accounted for 15.5% of without diabetes (47 years), although the median age at
renal disease in prevalent patients on RRT, although it the end of 2012 for prevalent diabetic patients was only
was more common in the 565 year age group compared three years higher than for individuals without diabetes.
to the younger group (17.3% vs. 14.4%). This contrasted This reflects reduced survival for patients with diabetes
with incident patients where diabetes was the pre- compared with patients without diabetes on RRT.
dominant diagnostic code in 25.6% of new RRT patients. Median time on RRT for patients with diabetes was less
Younger patients (age ,65 years) are more likely to have when compared with patients without diabetes (3.5
GN or pyelonephritis and less likely to have renal vascu- years vs. 6.7 years) and this difference in survival has
lar disease or hypertension as the cause of their renal not changed over the last five years. Patients with diabetes
failure. starting RRT in Scotland were three years younger and in
As described before, the male:female ratio was greater Northern Ireland three years older compared with the
than unity for all primary renal diagnoses (table 2.11). UK average age of patients with diabetes starting RRT
In individuals aged less than 65 years, renal trans- (data not shown).
plantation to dialysis ratio was greater than 1 in all Sixty percent of patients with diabetes as primary renal
PRD groups except diabetes and renovascular disease. diagnosis were undergoing HD. In patients with a differ-
In those aged .65 years, dialysis was more prevalent ent primary renal diagnosis 39% were undergoing HD
than renal transplantation in all PRD groups except (table 2.13). The percentage of patients with a functioning
polycystic kidney disease (PKD) (table 2.12). transplant was much lower in prevalent patients with
diabetes than in prevalent patients without diabetes
Diabetes (32% vs. 54%). However, the proportion of patients
Diabetes included all prevalent patients with type 1 or with diabetes as PRD with a functioning transplant has
type 2 diabetes as the primary renal diagnosis (ERA-
EDTA coding) and did not include patients with diabetes Table 2.13. Age relationships in patients with diabetes and
as a comorbidity. This analysis did not differentiate patients without diabetes and modality in prevalent RRT patients
between type 1 and type 2 diabetes as this distinction on 31/12/2012
was not made in the data submitted by most centres.
Patients with Patients without
The number of prevalent patients with diabetes as a diabetesa diabetesb
primary renal diagnosis increased 8.4% to 8,456 in
N 8,456 44,297
M : F ratio 1.59 1.54
Median age on 31/12/12 61 58
Table 2.12. Transplant : dialysis ratios by age and primary renal Median age at start of RRTcd 56 47
disease in the prevalent RRT population on 31/12/2012 Median years on RRTd 3.5 6.7
% HDe 60 39
Transplant : dialysis ratio % PDe 9 6
% transplante 32 54
Primary diagnosis∗ ,65 565
Excluded centre: 540% primary renal diagnosis aetiology uncertain
Aetiology uncertain 1.8 0.3 (Colchr)
Glomerulonephritis 2.2 0.7 a
Patients with diabetes: patients with a primary renal disease code of
Pyelonephritis 2.5 0.4 diabetes
b
Diabetes 0.8 0.1 Patients without diabetes: all patients excluding patients with
Polycystic kidney 2.3 1.4 diabetes and patients with a missing primary renal disease code
c
Hypertension 1.1 0.3 Median age at start of RRT was calculated from the most recent RRT
Renal vascular disease 0.9 0.1 start date
d
Other 1.9 0.3 Patients with an initial treatment modality of transferred in or
Not sent 2.1 0.3 transferred out were excluded from the calculation of median age at
start of RRT and median years on RRT, since their treatment start
∗
Appendix H ERA-EDTA coding date was not accurately known
e
Excluded centre: 540% primary renal diagnosis aetiology uncertain Patients without a treatment modality code were excluded from
(Colchr) calculating the % per treatment modality
55
Table 2.14. Treatment modalities by age and diabetes status on Home – HD

31/12/2012 2.0%
Hosp – HD
,65 years 565 years 19.4%
All other All other Transplant

Diabetesa causesb Diabetesa causesb 50.4%
N 5,064 28,796 3,392 15,501

% HD 46.8 28.0 78.9 60.7
% PD 8.2 5.3 9.7 8.6
% transplant 45.0 66.7 11.4 30.6
Excludes all patients without a treatment modality code Satellite – HD
Excluded centre with 540% PRD aetiology uncertain (Colchr) 21.3%
a
Patients with diabetes are patients with a primary renal disease code
of diabetes
b
Patients without diabetes are calculated as all patients excluding
patients with diabetes and patients with a missing primary renal APD CAPD
3.5% 3.4%
disease code
Fig. 2.6. Treatment modality in prevalent RRT patients on 31/12/
2012
increased since 2004 when only 26% of patients with
diabetes had a functioning transplant. For older patients
with diabetes (age 565 years), 11.4% had a functioning As mentioned earlier, treatment modality was related
transplant compared with 30.6% of their peers without to patient age. Younger patients (age ,65 years), were
diabetes (table 2.14). In Northern Ireland, 23.6% of more likely to have a functioning transplant (63.5%)
prevalent patients with diabetes had a functioning trans- when compared with patients aged over 65 years
plant compared with the UK average of 31.5% although (26.9%) (table 2.15). HD was the principal modality in
on average the Northern Ireland patients with diabetes the older patients (64.1%). However, in the elderly, inter-
were older by three years (data not shown). A higher preting the proportion of patients on renal replacement
proportion of prevalent patients without diabetes therapy who are transplanted is not straight forward as
(18.7%) were on home dialysis therapies (home HD this depends on approaches to dialysis and conservative
and PD) compared with prevalent patients with diabetes care in this age group.
(14.8%). Figure 2.7 shows the association between age and RRT
modality. Beyond 54 years of age, transplant prevalence
Modalities of treatment declined, whilst HD prevalence increased. The pro-
Transplantation was the most common treatment portion of each age group treated by PD remained
modality (50.4%) for prevalent RRT patients in 2012, more stable across the age spectrum.
followed closely by centre-based HD (40.7%) in either The proportion of prevalent dialysis patients receiving
hospital centre (19.4%) or satellite unit (21.3%) (figure 2.6). HD, ranged from 69.3% in Carlisle to 100% in Colchester
Satellite based haemodialysis was more prevalent than (table 2.16).
hospital centre haemodialysis for the first time in 2012. Overall, the proportion of dialysis patients treated in a
Home therapies made up the remaining 8.9% of treat- satellite haemodialysis unit has increased to 42.9% this
ment therapies, largely PD in its different formats year compared to 41.5% in 2011, and 39.9% in 2010.
(6.9%) which was similar to 2011. The proportion on Although there are satellite units in Scotland, the data
continuous ambulatory peritoneal dialysis (CAPD) and provided for 2012 did not distinguish between main
automated PD (APD) was 3.4% and 3.5% respectively, centre and satellite unit haemodialysis. In 2012, the
although the proportion on APD may be an under- number of centres that had more than 50% of their
estimate due to centre level coding issues which mean haemodialysis activity taking place in satellite units was
the UKRR cannot always distinguish between these 28, an increase from 2011 (table 2.16 and figure 2.8).
therapies. The term CAPD has been used for patients There was also wide variation between centres in the
receiving non-disconnect as well as disconnect CAPD proportion of dialysis patients on APD treatment, ranging
systems, because the proportion of patients using non- from 0% to 19.4% (table 2.16). Twelve of the 70 centres
disconnect systems was very small. with a PD programme did not report having any patients
56
Table 2.15. Percentage of prevalent RRT patients by dialysis and transplant modality by UK country on 31/12/2012
,65 years 565 years
UK country N % HD % PD % transplant N % HD % PD % transplant

England 29,491 30.9 6.0 63.1 16,585 64.0 9.1 26.9
N Ireland 961 30.6 4.5 64.9 559 71.6 7.3 21.1
Scotland 3,028 31.6 4.2 64.3 1,464 65.6 7.5 26.9
Wales 1,684 26.5 5.6 67.9 1,052 60.2 10.1 29.8
UK 35,164 30.7 5.8 63.5 19,660 64.1 9.0 26.9
All patients without a treatment modality code were excluded
on APD, whilst in the Northern Ireland centres almost all In 2012, the percentage of dialysis patients receiving
PD patients were on this form of the modality. home HD varied from 0% in eight centres, to greater
than 5% in 23 centres (table 2.16). In the UK, the overall
Home haemodialysis percentage of dialysis patients receiving home haemo-
The use of home HD as a RRT peaked in 1982 when dialysis has increased from 3.4% in 2011 to 4.0% in 2012.
almost 2,200 patients were estimated to be on this The proportion of dialysis patients receiving home
modality, representing 61% of HD patients reported to haemodialysis was greatest in Wales at 5.9%, compared
the ERA-EDTA Registry at that time. The fall in the with 4.9% in Northern Ireland, 3.9% in England and
use of this modality to just 445 patients (2.4% of HD 2.9% in Scotland (figure 2.8, table 2.16). The proportion
patients) in 2006 was probably due to an increase in on home haemodialysis has increased in each of the
availability and uptake of renal transplantation, and four countries since 2011. Forty-seven renal centres
also the similar expansion of hospital HD provision across the UK had an increase in the proportion of
with the introduction of satellite units. In the last seven individuals on home haemodialysis compared with
years there has been renewed interest in home HD and 2011. In 2007, for comparison, the proportion of patients
a target of 15% of HD patients on this modality has receiving home haemodialysis was 2% in each of the four
been suggested [6]. Equipment changes and patient UK countries.
choice has helped drive this change. Since 2006 there
has been a gradual increase in the proportion of prevalent Change in modality
patients receiving haemodialysis in their own homes so The relative proportion of RRT modalities in prevalent
that in 2012 it reached 4.6% of HD patients (n = 1,080, patients has changed dramatically over the past decade.
figure 2.2). These numbers may be an underestimate as The main features are depicted in figure 2.9, which
some centres have been unable to submit data for patients describes a decline in the proportion of patients treated
coded as home HD and work is ongoing to address this. by PD after 2000. This may however have started to
100 *
961
116
90
80 681
856
2,634
4,895
7,403
6,540
4,307
70
60
Percentage
967
50
40 768 Transplant
Peritoneal dialysis
30 624
223 340 Haemodialysis
73
20
1,853
3,358
4,408
5,824
5,533
1,246
276
913
10 Fig. 2.7. Treatment modality distribution

0 by age in prevalent RRT patients on 31/12/
18–24 25–34 35–44 45–54 55–64 65–74 75–84 85+ 2012
Age group ∗
N = 25
57
Table 2.16. Percentage of prevalent dialysis patients by dialysis modality by centre on 31/12/2012
Haemodialysis Peritoneal dialysis
Centre N Total Home Hospital Satellite CAPD APD
England
B Heart 482 90.3 3.7 79.9 6.6 7.3 2.5
B QEH 1,085 85.3 4.9 10.5 70.0 5.5 9.1
Basldn 196 83.7 0.0 83.2 0.5 8.2 7.7
Bradfd 237 87.8 0.8 71.7 15.2 1.7 10.6
Brightn 456 81.4 7.9 43.0 30.5 11.8 6.6
Bristol 560 88.2 5.5 15.4 67.3 5.7 6.1
Camb 385 90.9 3.4 37.4 50.1 0.0 0.0
Carlis 88 69.3 0.0 51.1 18.2 12.5 18.2
Carsh 876 87.2 2.5 19.1 65.6 3.7 9.1
Chelms 155 83.2 0.0 83.2 0.0 11.0 5.8
Colchr 117 100.0 0.0 100.0 0.0 0.0 0.0
Covnt 463 78.4 4.3 74.1 0.0 21.6 0.0
Derby 309 71.2 8.1 63.1 0.0 18.8 10.0
Donc 201 85.6 0.0 45.3 40.3 1.5 12.9
Dorset 308 84.4 0.7 20.8 63.0 6.5 8.4
Dudley 232 72.8 5.2 50.9 16.8 16.0 11.2
Exeter 474 83.8 0.6 12.5 70.7 7.8 8.4
Glouc 255 85.9 1.2 76.1 8.6 2.8 11.4
Hull 424 78.8 2.4 36.6 39.9 10.4 10.4
Ipswi 160 80.6 3.1 66.3 11.3 10.0 9.4
Kent 446 86.1 4.0 22.9 59.2 13.9 0.0
L Barts 1,090 82.1 1.7 35.1 45.4 5.6 12.3
L Guys 657 95.3 6.1 16.7 72.5 2.0 2.7
L Kings 578 85.1 1.2 20.6 63.3 6.8 8.1
L Rfree 834 85.6 2.3 3.7 79.6 3.8 10.4
L St.G 338 84.0 1.5 41.7 40.8 4.1 11.8
L West 1,478 96.5 1.0 22.2 73.3 1.6 2.0
Leeds 582 85.1 2.1 19.1 63.9 3.6 11.3
Leic 1,032 84.5 6.0 16.6 61.9 4.6 11.0
Liv Ain 195 89.7 2.6 9.2 78.0 2.6 7.7
Liv RI 429 85.3 8.2 37.1 40.1 9.8 4.9
M RI 589 86.1 11.5 30.9 43.6 2.6 11.4
Middlbr 350 96.9 3.4 30.9 62.6 3.1 0.0
Newc 332 85.8 7.5 78.3 0.0 1.5 12.7
Norwch 373 85.3 5.1 49.3 30.8 11.3 3.5
Nottm 457 82.3 7.7 39.0 35.7 8.1 9.6
Oxford 509 83.1 3.9 32.4 46.8 4.3 12.6
Plymth 166 78.9 4.2 74.7 0.0 17.5 3.6
Ports 638 87.0 1.4 18.7 66.9 13.0 0.0
Prestn 605 88.6 6.6 19.2 62.8 2.6 8.8
Redng 343 79.0 1.8 34.4 42.9 13.7 7.3
Salford 484 78.5 4.6 33.5 40.5 10.7 9.5
Sheff 657 89.5 6.1 39.0 44.4 10.5 0.0
Shrew 236 82.6 6.8 45.8 30.1 17.4 0.0
Stevng 441 92.8 6.4 33.1 53.3 7.3 0.0
Sthend 132 89.4 2.3 87.1 0.0 10.6 0.0
Stoke 384 79.4 6.8 46.9 25.8 3.7 16.9
Sund 220 90.0 1.4 54.6 34.1 6.4 3.6
Truro 177 87.0 2.8 45.2 39.0 6.2 6.8
Wirral 234 86.3 1.7 42.3 42.3 3.0 10.7
Wolve 377 75.6 3.7 23.6 48.3 24.4 0.0
York 167 80.9 7.2 31.7 41.9 18.6 0.6
58
Centre N Total Home Hospital Satellite CAPD APD
N Ireland
Antrim 145 91.0 2.8 88.3 0.0 1.4 7.6
Belfast 256 89.1 8.2 80.9 0.0 0.8 9.8
Newry 107 85.0 2.8 82.2 0.0 0.0 15.0
Ulster 116 93.1 3.5 89.7 0.0 0.0 6.9
West NI 154 87.7 3.9 83.8 0.0 0.0 12.3
Scotland
Abrdn 255 90.2 2.0 88.2 0.0 5.5 4.3
Airdrie 205 94.6 0.0 94.6 0.0 3.4 2.0
D & Gall 67 76.1 1.5 74.6 0.0 11.9 11.9
Dundee 202 89.6 0.0 89.6 0.0 2.5 7.9
Dunfn 167 88.0 0.0 88.0 0.0 0.0 12.0
Edinb 302 87.8 2.0 85.8 0.0 4.0 8.3
Glasgw 671 93.0 5.2 87.8 0.0 2.4 4.6
Inverns 92 80.4 7.6 72.8 0.0 7.6 12.0
Klmarnk 191 78.5 4.2 74.4 0.0 2.1 19.4
Wales
Bangor 105 85.7 13.3 54.3 18.1 5.7 8.6
Cardff 559 86.2 5.4 12.7 68.2 9.3 4.5
Clwyd 102 82.4 2.9 79.4 0.0 6.9 0.0
Swanse 396 82.8 7.1 47.5 28.3 14.1 3.0
Wrexm 118 81.4 0.9 67.0 13.6 18.6 0.0
England 22,993 85.8 3.9 33.4 48.5 7.1 7.0
N Irelanda 778 89.2 4.9 84.3 0.0 0.5 10.2
Scotlandb 2,152 89.0 2.9 86.2 0.0 3.4 7.6
Wales 1,280 84.4 5.9 37.2 41.3 11.2 3.6
UK 27,203 86.1 4.0 39.2 42.9 6.8 6.9
a
There are no satellite units in Northern Ireland
b
All haemodialysis patients in Scotland are shown as receiving treatment at home or in centre as no data is available regarding satellite dialysis
100
% home HD
90 % sat HD
80
Percentage of dialysis patients
70
60
50
40
30
20
10
0
L Rfree
Liv Ain
Exeter
B QEH
Cardff
Ports
Bristol
L Guys
L West
Carsh
Leeds
Dorset
L Kings
Leic
Prestn
Kent
Middlbr
Wolve
Stevng
Oxford
L Barts
Camb
Redng
York
Salford
M RI
Hull
Sheff
Wirral
L St.G
Donc
Liv RI
Truro
Nottm
Sund
Brightn
Shrew
Norwch
Swanse
Stoke
Carlis
Dudley
Bangor
Bradfd
Wrexm
Ipswi
Glouc
B Heart
Basldn
Derby
Belfast
Newc
Covnt
Plymth
West NI
Ulster
Clwyd
Newry
Antrim
Sthend
Colchr
Chelms
England
N Ireland
Wales
UK
Centre
Fig. 2.8. Percentage of prevalent haemodialysis patients treated with satellite or home haemodialysis by centre on 31/12/2012
∗
Scottish centres excluded as information on satellite HD was not available. No centres in Northern Ireland have satellite dialysis units
59
55
50
45
40
Percentage on modality
35
30
25
20
15
10 % transplant
% HD
5 % PD
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Fig. 2.9. Modality changes in prevalent
Year RRT patients from 1997–2012
plateau, with only a minor reduction from 7.2% of the tube insertion and acute PD. The introduction of dialysis
RRT population in 2011 to 6.9% in 2012. For the first best practice tariffs in England may result in further
time since 2007, the absolute number of patients on PD changes to the types of treatment patients receive in
increased from 3,780 patients in 2011 to 3,792 patients England.
in 2012. Time on PD has decreased marginally over The proportion of patients treated with HD has
that last six years, from a median of 2.0 years in 2007 stabilised in the last three years. The proportion of
to 1.7 years in 2012 probably reflecting increased trans- patients with a functioning transplant which had been
plantation rates in this largely younger patient group. on a slight downward trend has reversed since 2007,
Since 2009 there have been small increases in the size probably due to continued increases in living organ and
of the incident population commencing PD as the first non-heart beating donation [7].
established modality. The determinants of this are likely Figure 2.10 depicts in more detail the modality
to be multi-factorial and include the effect of patient or changes in the prevalent dialysis population during this
physician choice regarding the treatment modality at time and highlights a sustained reduction in the
start of RRT, the general health and fitness of patients proportion of patients treated by CAPD. There was a
starting RRT, organisational level flexibility around PD sustained increase in the proportion of prevalent HD
30
25
20
15
% hosp HD
% CAPD
10 % sat HD
% APD
% home HD
5
Fig. 2.10. Detailed dialysis modality
changes in prevalent RRT patients from
0 1997–2012
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 ∗
Scottish centres excluded as information on
Year satellite HD was not available
60
patients treated at satellite units with a steady decline in Summary

hospital centre haemodialysis since 2004.
There continues to be growth across the UK in
prevalent patients on RRT with regional and centre
level variation. There was no real difference in prevalence
International comparisons rates between the four nations of the UK once adjusted
for background population characteristics. In general,
At the time of writing this report, prevalence rate data areas with large ethnic minority populations had higher
were not yet available for 2012 from other countries. standardised prevalence ratios. There were increasing
Therefore international comparisons of prevalence rates numbers of patients on HD and those with a functioning
are not presented. This data will be added to the UKRR transplant. There was an absolute increase in patient
data portal when it is available. numbers on PD in 2012, with only a minor reduction
in the relative proportion on PD between 2011 and
2012. The prevalence rate in the over 80 year age group
continues to increase. There have been substantial
increases in home HD use in some areas although several
centres are still unable to offer this modality.
Conflicts of interest: none
References
1 Office for National Statistics. www.statistics.gov.uk 6 NICE 2002. Technology appraisal No 48. National Institute Clinical
2 National Records of Scotland. http://www.nrscotland.gov.uk/ Excellence. www.nice.org.uk
3 Northern Ireland Statistics and Research Agency. http://www.nisra.gov. 7 NHS Blood and Transplant activity report 2009/2010. Transplant activity
uk/ in the UK. http://www.organdonation.nhs.uk/ukt/statistics/transplant_
4 Office for National Statistics. The classification of ethnic groups. (www. activity_report/current_activity_reports/ukt/activity_report_2009_10.pdf
statistics.gov.uk)
5 Ansell D, Feest T: The sixth annual report. Chapter 17: Social deprivation
on renal replacement therapy. Bristol, UK Renal Registry, 2003
61
Chapter 3 Demographic and
Biochemistry Profile of Kidney Transplant
Recipients in the UK in 2012: National and
Rishi Pruthia, Anna Casulaa, Iain MacPheeb

a
UK Renal Registry, Bristol, UK; bSt George’s, University of London, UK
Key Words
. The median eGFR of prevalent renal transplant
Blood pressure . Bone metabolism . Chronic kidney disease . recipients was 51.3 ml/min/1.73 m2.
Deceased donor . eGFR . Epidemiology . Ethnicity . Graft
. The median eGFR of patients one year after
function . Haemoglobin . Live donor . Outcomes . Renal transplantation was 56.4 ml/min/1.73 m2 post live
transplantation . Survival transplant, 52.7 ml/min/1.73 m2 post brainstem
death transplant and 49.4 ml/min/1.73 m2 post
circulatory death transplant.
Summary . 13.7% of prevalent transplant patients had eGFR
,30 ml/min/1.73 m2.
. There was a 5% increase in overall renal transplant . The median decline in eGFR slope beyond the
numbers in 2012, with a significant rise in kidney first year after transplantation was −0.53 ml/min/
donation from donors after circulatory death (19%). 1.73 m2/year.
. In 2012, death-censored renal transplant failure . In 2012, infection (23%) and malignancy (20%)
rates in prevalent patients were similar to previous remained amongst the commonest causes of death
years at 2.2% per annum. Transplant patient death in patients with a functioning renal transplant.
rates remained stable at 2.3 per 100 patient years.
. The median age of incident and prevalent renal
transplant patients in the UK was 50.5 and 52.2
years respectively.
63
Introduction This enables ODT to generate comprehensive analyses of

renal transplant activity and graft survival statistics.
This chapter includes independent analyses regarding NHSBT attributes a patient to the centre that per-
renal transplant activity and survival data from the formed the transplant operation irrespective of where
UK Transplant Registry, held by the Organ Donation the patient was cared for before or after the procedure
and Transplantation Directorate (ODT) of NHS Blood and hence only reports on transplant centre performance.
and Transplant (NHSBT). The UK Renal Registry
(UKRR) has performed additional analyses of renal Methods
transplant recipient follow-up data examining demo- In 2012, there were 23 UK adult renal transplant centres, 19 in
graphics, clinical and biochemical variables. NHSBT England, 2 in Scotland and 1 each in Northern Ireland and Wales.
records all the information regarding the episode of Comprehensive information from 1999 onwards concerning
the number of patients on the transplant waiting list, the number
transplantation (donor and recipient details) and the of transplants performed, the number of deceased kidney donors
UKRR holds additional information on key clinical and (donor after brainstem death and donor after circulatory death),
biochemical variables in renal transplant recipients. The living kidney donors, patient survival and graft survival is available
co-operation between these two organisations results in on the NHSBT website (http://www.organdonation.nhs.uk/ukt/
a comprehensive database describing the clinical care statistics/statistics.asp).
delivered to renal transplant patients within the UK.
This further allows for the comparison of key outcomes Results
between centres and provides insight into the processes During 2012, 2,901 kidney or kidney plus other organ
involved in the care of such patients in the UK. transplants were performed. The absolute number of
This chapter is divided into six sections: (1) transplant living kidney donors showed a 1% rise in 2012 represent-
activity, waiting list and survival data; (2) transplant ing 35.6% of all transplants performed whilst donor after
demographics; (3) clinical and laboratory outcomes; (4) circulatory death transplants continued to increase and
analysis of prevalent patients by chronic kidney disease comprised 24.4% of all kidney transplants performed. A
(CKD) stage; (5) eGFR slope analysis; and (6) causes of small rise in the number of transplants from donors
death in transplant recipients. Methodology, results and after brainstem death was also noted in 2012 partially
conclusions of these analyses are discussed in detail for reversing the small decline noted in 2011 (table 3.1).
all six sections separately. There were small differences in one and five year risk-
The UK Renal Registry methodology is described adjusted patient and graft survival rates amongst UK
elsewhere [1]. The UKRR collects quarterly clinical data renal transplant centres (table 3.2). These graft survival
via an electronic data extraction process from hospital rates include grafts with primary non-function (which
based renal IT systems on all patients receiving renal are excluded from analysis by some countries).
replacement therapy. Throughout the chapter, the
number preceding the centre name in each figure indi-
cates the percentage of missing data for that centre for Table 3.1. Kidney and kidney plus other organ transplant num-
that variable. bers in the UK, 1/1/2010–31/12/2012
Unless otherwise specified, prevalent transplant
% change
patients were defined as patients with a functioning Organ 2010 2011 2012 2011–2012
renal transplant on the 31st December 2012.
Donor after brainstem deatha 989 951 967 2
Donor after circulatory deathb 549 594 708 19
Living donor kidney 1,027 1,026 1,034 1
Kidney and liver 9 16 17 6
Transplant activity, waiting list activity and Kidney and heart 0 0 3
survival data Kidney and pancreasc 150 163 172 6
Small bowel (inc kidney) 1 2 0
Total kidney transplants 2,725 2,752 2,901 5
Introduction
a
NHSBT prospectively collects donor and recipient data Includes en bloc kidney transplants (7 in 2010, 7 in 2011, 4 in 2012)
around the episode of transplantation. They also request and double kidney transplants (6 in 2010, 5 in 2011, 7 in 2012)
b
Includes en bloc kidney transplants (2 in 2010, 2 in 2011, 4 in 2012)
that transplant centres provide an annual paper based and double kidney transplants (16 in 2010, 32 in 2011, 52 in 2012)
data return on the status of the recipient’s graft function. c
Includes DCD transplants (29 in 2010, 28 in 2011, 35 in 2012)
64
Chapter 3 Outcomes in UK renal transplant recipients in 2012
Table 3.2. Risk-adjusted first adult kidney transplant only, graft and patient survival percentage rates for UK centres∗
Deceased donor Deceased donor Living kidney donor Living kidney donor
1 year survival 5 year survival 1 year survival 5 year survival
Centre Graft Patient Graft Patient Graft Patient Graft Patient

B QEH 88 96 83 89 96 99 88 96
Belfast 93 95 91 92 94 100 92 93
Bristol 94 96 84 85 98 99 95 98
Camb 92 97 85 90 99 99 96 100
Cardff 96 98 85 88 95 98 88 96
Covnt 88 94 89 91 96 100 88 96
Edin 90 95 83 85 95 98 91 97
Glasgw 93 97 84 84 96 96 96 97
L Barts 91 91 88 90 95 98 92 93
L Guys 93 96 81 90 97 98 92 95
L Rfree 94 97 88 93 98 100 93 95
L St.G 96 99 85 92 99 100 92 95
L West 94 98 88 92 96 99 83 95
Leeds 92 95 86 90 95 100 92 98
Leic 92 93 82 79 96 98 91 93
Liv RI 93 95 81 94 95 100 92 92
M RI 94 96 84 88 98 98 93 98
Newc 93 95 83 89 99 99 93 98
Nottm 94 95 80 86 95 99 91 94
Oxford 94 96 89 87 96 96 98 94
Plymth 88 97 86 89 95 99 88 93
Ports 95 95 80 88 94 99 82 91
Sheff 91 98 81 92 98 100 89 100
All centres 93 96 84 88 96 96 96 96
∗
Information courtesy of NHSBT; statistical methodology for computing risk-adjusted estimates can be obtained from the NHSBT website (see
http://www.organdonation.nhs.uk/ukt/statistics/statistics.asp)
Cohorts for survival rate estimation: 1 year survival: 1/1/2007–31/12/2011; 5 year survival: 1/1/2003–31/12/2007; first grafts only – re-grafts
excluded for patient survival estimation. Since the cohorts to estimate 1- and 5-year survival are different, some centres may appear to have
5 year survival better than 1 year survival
Using data from the UKRR on prevalent renal only rate of 2.2% per annum and patient death rate of 2.3
transplant patients on 1st January 2012, the death rate per 100 patient years were similar to those noted in 2011.
during 2012 was 2.3/100 patient years (CI 2.1–2.5)
when censored for return to dialysis and 2.4/100 patient
years (CI 2.2–2.6) without censoring for dialysis. These
death rates are similar to those observed over the last Transplant demographics
few years.
During 2012, 2.2% of prevalent transplant patients Introduction
experienced graft failure (excluding death as a cause of Since 2008, all UK renal centres have established
graft failure) maintaining the fall in graft failure rates electronic linkage to the UKRR or Scottish Renal
noted over the last couple of years. Whilst it might be pre- Registry, giving the UKRR complete coverage of
mature to assume that graft failure rates are falling in the individual patient level data across the UK.
UK the 0.5% fall noted in the last five years is certainly The following sections need to be interpreted in the
encouraging. context of variable repatriation policies; some transplant
centres continue to follow up and report on all patients
Conclusions they transplant, whereas others refer patients back to
In 2012, the increased number of kidney transplants non-transplant centres for most or all ongoing post-
performed was mostly due to the growing use of organs transplant care. Some transplant centres only refer
from donors after circulatory death. The graft failure back patients when their graft is failing. The time
65
post-transplantation that a patient is referred back to renal transplant recipients, with some areas having higher
their local centre varies between transplant centres. The than the predicted number of prevalent transplant
UKRR is able to detect duplicate patients (being reported patients per million population and others lower. There
from both transplant and referring centres) and in such are a number of potential explanations for these incon-
situations care is attributed to the referring centre. This sistencies, including geographical differences in access
process may result in some discrepancies in transplant to renal transplantation in the UK. This has previously
numbers particularly in Oxford/Reading and Clywd/ been analysed in detail by the UKRR [2] and is currently
Liverpool RI. the focus of a large national study (access to Transplant
and Transplant Outcome Measures (ATTOM)).
Methods The proportion of prevalent RRT patients with a
Four centres (Bangor, Colchester, Liverpool Aintree and transplant relative to the number on dialysis has been
Wirral) did not have any transplant patients and were excluded relatively stable over the last decade.
from some of the analyses. Their dialysis patients were included
in the relevant dialysis population denominators.
For the analysis of primary renal diagnosis (PRD) in transplant Age and gender
recipients, a few centres were excluded from some of the take-on The gender ratio amongst incident and prevalent trans-
years because of concerns relating to the reliability of PRD coding plant patients has remained stable for at least the last ten
(with these centres submitting a high percentage of uncertain or
missing aetiology codes).
years (table 3.6, figure 3.1). Note absolute patient numbers
Information on patient demographics (age, gender, ethnicity differ from those published in previous reports as a result
and PRD) for patients in a given renal centre was obtained from of additional data validation and reallocation of patients.
UKRR patient registration data fields. Individual patients were The average age of incident transplant patients has steadily
assigned to the centre that returned data for them during 2012. increased during the same time period. There has also been
The prevalence of transplant patients in areas covered by individ-
ual primary care trusts (PCT) or Health Boards/Social Care Areas
a gradual increase in the average age of prevalent trans-
(HB) was estimated based on the post code of the registered plant patients, which could reflect the increasing age at
address for patients on renal replacement therapy (RRT). Data which patients are transplanted and/or improved survival
on ethnic origin, supplied as Patient Administration System after renal transplantation over the last few years. The
(PAS) codes, were retrieved from fields within renal centre IT prevalent transplant patient workload across the UK
systems. For the purpose of this analysis, patients were grouped
into Whites, South Asians, Blacks, Others and Unknown. The
increased to 27,621 patients at the end of 2012. The con-
details of ethnicity regrouping into the above categories are tinued expansion of this patient group means there is a
provided in appendix H: Coding http://www.renalreg.com. need for careful planning by renal centres for future service
provision and resource allocation.
Results and discussion
Prevalent transplant numbers across the UK are Primary renal diagnosis
described in table 3.3. The primary renal diagnosis of patients receiving
The prevalence of renal transplant recipients in each kidney transplants in the UK has remained relatively
PCT/HB in England, Northern Ireland (Health and stable over the last five years (table 3.7).
Social Care Trust Areas), Scotland (Health Boards) and
Wales (Local Health Boards) and the proportion of Ethnicity
prevalent patients according to modality in the renal It was difficult to compare the proportion of patients
centres across the UK is described in tables 3.4 and 3.5 within each ethnic group receiving a transplant to those
respectively. After standardisation for age and gender, commencing dialysis from the same group because data
unexplained variability was evident in the prevalence of on ethnicity were missing in a considerable number of
Table 3.3. The prevalence per million population (pmp) of renal transplants in adults in the UK on 31/12/2012
All UK centres 23,083 742 2,340 1,456 27,621

Total population, mid-2012 estimates from ONS∗ (millions) 53.5 1.8 5.3 3.1 63.7
Prevalence pmp transplant 432 407 440 474 434
∗
Office of National Statistics, UK
66
Table 3.4. The prevalence per million population (pmp) of patients with a renal transplant and standardised rate ratio in the UK, as on
31st December 2008–2012
a
PCT/HB – Primary Care Trust (England); Health and Social Care Trust Areas (Northern Ireland); Health Board (Scotland) and Local Health
Board (Wales)
b
Population numbers based on the 2011 mid-year estimates by age group and gender (data obtained from the Office of National Statistics)
c
O/E – age and gender standardised prevalence rate ratio
PCTs with significantly high average rate ratios are bold in greyed areas
PCTs with significantly low average rate ratios are italicised in greyed areas
Age and gender
Rate pmp standardised rate ratio 2012
Population
UK Area PCT/HBa coveredb 2008 2009 2010 2011 2012 O/Ec LCL UCL
North East County Durham 513,000 390 398 411 433 439 0.96 0.84 1.09
Darlington 105,600 369 331 360 407 407 0.92 0.68 1.24
Gateshead 200,300 374 389 389 404 449 1.00 0.82 1.24
Hartlepool 92,100 369 358 402 413 456 1.03 0.76 1.40
Middlesbrough 138,400 434 470 477 520 549 1.32 1.06 1.66
Newcastle 279,100 373 376 380 408 398 0.99 0.82 1.20
North Tyneside 201,200 482 507 557 577 581 1.28 1.07 1.53
Northumberland 316,300 398 398 383 436 436 0.92 0.77 1.08
Redcar and Cleveland 135,200 525 540 547 562 570 1.25 1.00 1.56
South Tyneside 148,200 439 445 432 472 479 1.06 0.84 1.33
Stockton-on-Tees Teaching 191,800 391 401 391 381 407 0.93 0.74 1.16
Sunderland Teaching 275,300 418 403 421 469 487 1.08 0.91 1.28
North West Ashton, Leigh and Wigan 318,100 358 339 387 446 490 1.08 0.92 1.270
Blackburn with Darwen Teaching 147,700 305 312 312 339 379 0.93 0.72 1.21
Blackpool 142,100 338 345 345 338 408 0.90 0.70 1.16
Bolton Teaching 277,300 408 418 433 483 516 1.20 1.02 1.41
Bury 185,400 351 410 410 421 448 1.02 0.82 1.26
Central and Eastern Cheshire 462,800 302 305 341 359 378 0.82 0.70 0.95
Central Lancashire 467,400 300 312 347 370 396 0.89 0.77 1.03
Cumbria Teaching 499,800 330 370 390 394 412 0.87 0.76 0.99
East Lancashire Teaching 382,500 405 405 403 429 437 0.99 0.85 1.15
Halton and St Helens 301,100 299 312 345 365 385 0.85 0.71 1.02
Heywood, Middleton and Rochdale 211,900 382 396 406 439 453 1.07 0.87 1.30
Knowsley 145,900 329 363 377 377 398 0.91 0.71 1.18
Liverpool 465,700 305 320 344 376 391 0.94 0.81 1.09
Manchester Teaching 502,900 247 251 296 328 364 1.00 0.87 1.16
North Lancashire Teaching 321,600 320 317 311 320 342 0.75 0.62 0.90
Oldham 225,200 351 378 395 409 426 1.03 0.84 1.25
Salford 234,500 290 316 345 371 426 1.03 0.85 1.26
Sefton 274,000 296 310 347 361 376 0.82 0.68 1.00
Stockport 283,300 342 371 395 413 431 0.96 0.80 1.14
Tameside and Glossop 252,900 411 419 455 490 498 1.12 0.94 1.33
Trafford 227,100 282 277 317 343 374 0.86 0.69 1.06
Warrington 202,700 385 419 390 405 439 0.98 0.79 1.20
Western Cheshire 237,400 320 358 379 400 425 0.92 0.76 1.12
Wirral 319,800 313 331 338 353 356 0.79 0.66 0.95
Yorkshire and the Barnsley 231,900 367 371 392 405 423 0.93 0.76 1.13
Humber Bradford and Airedale Teaching 523,100 390 419 447 449 499 1.25 1.11 1.42
Calderdale 204,200 431 441 475 509 544 1.21 1.01 1.46
Doncaster 302,500 321 344 350 380 410 0.93 0.78 1.10
East Riding of Yorkshire 334,700 338 362 371 382 412 0.86 0.73 1.02
Hull Teaching 256,100 351 375 387 398 429 1.03 0.86 1.24
67

Age and gender
Population
Yorkshire and the Kirklees 423,000 390 400 416 437 454 1.06 0.92 1.22
Humber Leeds 750,700 320 338 360 384 412 1.00 0.90 1.12
North East Lincolnshire 161,200 323 347 366 409 434 0.98 0.78 1.24
North Lincolnshire 163,600 269 251 257 263 275 0.60 0.45 0.81
North Yorkshire and York 799,000 362 388 412 439 469 1.02 0.93 1.13
Rotherham 257,700 357 376 415 450 466 1.04 0.87 1.24
Sheffield 551,800 301 321 359 382 391 0.95 0.84 1.09
Wakefield District 326,400 319 316 343 361 386 0.85 0.71 1.01
East Midlands Bassetlaw 113,000 283 274 301 301 327 0.70 0.51 0.97
Derby City 248,900 257 309 362 370 418 1.02 0.84 1.24
Derbyshire County 737,500 290 294 315 347 370 0.79 0.70 0.89
Leicester City 329,600 458 525 525 561 586 1.53 1.33 1.76
Leicestershire County and Rutland 688,800 382 389 417 433 457 1.00 0.90 1.12
Lincolnshire Teaching 717,200 283 289 303 322 343 0.74 0.65 0.84
Northamptonshire Teaching 694,000 346 362 386 406 406 0.92 0.82 1.03
Nottingham City 303,900 230 244 319 339 362 0.96 0.80 1.16
Nottinghamshire County Teaching 673,800 327 346 389 420 453 0.99 0.88 1.11
West Midlands Birmingham East and North 421,400 344 356 373 399 420 1.08 0.93 1.25
Coventry Teaching 316,900 350 363 385 410 429 1.08 0.91 1.28
Dudley 313,300 271 287 300 310 290 0.65 0.53 0.80
Heart of Birmingham Teaching 299,200 381 384 398 398 414 1.21 1.01 1.44
Herefordshire 183,600 278 300 300 310 332 0.71 0.55 0.91
North Staffordshire 212,900 319 343 348 371 399 0.86 0.70 1.06
Sandwell 309,000 337 353 353 359 398 0.97 0.82 1.16
Shropshire County 307,100 293 329 339 355 342 0.73 0.60 0.89
Solihull 206,900 285 290 305 319 358 0.80 0.63 1.00
South Birmingham 353,700 328 328 362 373 382 0.95 0.81 1.13
South Staffordshire 628,500 309 318 333 344 344 0.74 0.65 0.85
Stoke on Trent 256,900 362 389 417 413 440 1.02 0.85 1.23
Telford and Wrekin 166,800 234 276 288 294 282 0.65 0.49 0.86
Walsall Teaching 269,500 338 360 378 401 416 0.98 0.81 1.18
Warwickshire 546,600 353 373 412 443 468 1.03 0.91 1.16
Wolverhampton City 249,900 284 304 300 300 308 0.74 0.59 0.92
Worcestershire 566,600 286 312 337 344 365 0.79 0.69 0.91
East of England Bedfordshire 413,500 343 370 387 397 450 1.01 0.88 1.17
Cambridgeshire 622,300 320 358 394 407 423 0.96 0.85 1.08
Hertfordshire 1,119,800 328 346 391 412 438 1.01 0.93 1.10
Great Yarmouth and Waveney 212,800 230 291 301 315 334 0.73 0.58 0.92
Luton 203,600 344 354 388 437 481 1.25 1.03 1.53
Mid Essex 375,200 320 360 376 424 410 0.90 0.77 1.06
Norfolk 762,000 310 329 339 349 349 0.76 0.68 0.86
North East Essex 311,700 308 321 343 372 395 0.89 0.74 1.06
Peterborough 184,500 249 287 298 347 352 0.86 0.68 1.10
South East Essex 345,600 298 330 336 339 368 0.81 0.68 0.97
South West Essex 407,100 290 317 341 366 383 0.89 0.76 1.04
Suffolk 614,800 290 320 342 372 386 0.86 0.75 0.97
West Essex 289,600 273 321 363 363 390 0.88 0.73 1.05
London Barking and Dagenham 187,000 267 326 348 412 428 1.17 0.94 1.46
Barnet 357,500 406 467 503 559 632 1.57 1.37 1.78
Bexley 232,800 438 455 498 511 524 1.24 1.04 1.48
Brent Teaching 312,200 522 573 605 612 657 1.65 1.44 1.89
68

Age and gender
Population
London Bromley 310,600 441 454 483 493 512 1.17 1.00 1.37
Camden 220,100 363 404 427 477 504 1.26 1.05 1.52
City and Hackney Teaching 254,600 275 299 322 322 342 0.92 0.74 1.13
Croydon 364,800 310 345 356 386 395 0.96 0.82 1.13
Ealing 339,300 525 545 584 598 634 1.57 1.37 1.79
Enfield 313,900 433 440 468 535 583 1.46 1.27 1.69
Greenwich Teaching 255,500 305 360 391 423 458 1.17 0.98 1.41
Hammersmith and Fulham 182,400 323 400 438 444 477 1.21 0.98 1.49
Haringey Teaching 255,500 352 399 438 470 520 1.32 1.12 1.57
Harrow 240,500 570 640 690 699 723 1.74 1.50 2.02
Havering 237,900 282 303 315 336 336 0.77 0.62 0.96
Hillingdon 275,500 417 468 512 563 592 1.47 1.26 1.71
Hounslow 254,900 408 475 526 537 557 1.39 1.18 1.63
Islington 206,300 431 475 499 528 572 1.46 1.22 1.74
Kensington and Chelsea 158,300 367 385 461 474 474 1.09 0.87 1.36
Kingston 160,400 380 393 399 418 461 1.12 0.89 1.40
Lambeth 304,500 292 325 325 365 414 1.06 0.89 1.27
Lewisham 276,900 368 394 412 426 455 1.15 0.97 1.37
Newham 310,500 232 293 332 354 396 1.12 0.94 1.34
Redbridge 281,400 355 380 455 476 526 1.34 1.14 1.57
Richmond and Twickenham 187,500 261 299 315 347 379 0.87 0.69 1.10
Southwark 288,700 405 461 492 526 571 1.46 1.26 1.71
Sutton and Merton 391,700 378 411 431 452 500 1.20 1.04 1.38
Tower Hamlets 256,000 223 258 309 316 355 1.03 0.84 1.27
Waltham Forest 259,700 354 377 412 439 450 1.16 0.96 1.39
Wandsworth 307,700 338 338 357 390 435 1.12 0.94 1.32
Westminster 219,600 355 437 483 474 501 1.19 0.99 1.44
South East Coast Brighton and Hove City 273,000 275 289 319 333 337 0.81 0.66 0.99
East Sussex Downs and Weald 343,900 294 311 320 334 372 0.81 0.68 0.97
Eastern and Coastal Kent 759,600 340 374 404 440 483 1.09 0.99 1.21
Hastings and Rother 183,400 305 305 322 349 338 0.74 0.57 0.94
Medway 264,900 366 393 415 415 442 1.03 0.86 1.24
Surrey 1,124,800 351 369 380 386 413 0.93 0.85 1.02
West Kent 706,800 364 386 390 399 426 0.97 0.86 1.08
West Sussex 808,900 344 352 368 386 383 0.85 0.76 0.95
South Central Berkshire East 410,100 407 444 502 527 568 1.37 1.20 1.56
Berkshire West 464,400 418 450 459 484 493 1.14 1.00 1.30
Buckinghamshire 521,000 407 415 441 453 489 1.10 0.97 1.24
Hampshire 1,322,100 348 364 382 396 414 0.91 0.84 0.99
Isle of Wight National Health Service 138,400 303 318 332 332 347 0.73 0.55 0.97
Milton Keynes 255,400 329 352 392 423 458 1.09 0.91 1.31
Oxfordshire 629,600 405 410 429 442 480 1.12 1.00 1.25
Portsmouth City Teaching 205,400 355 355 399 399 419 1.05 0.85 1.30
Southampton City 235,900 343 356 352 399 428 1.09 0.90 1.33
South West Bath and North East Somerset 175,500 291 325 308 302 308 0.71 0.55 0.93
Bournemouth and Poole Teaching 331,500 335 332 341 365 353 0.81 0.68 0.97
Bristol 428,100 432 446 474 488 516 1.30 1.14 1.48
Cornwall and Isles of Scilly 536,000 416 437 446 465 511 1.09 0.97 1.23
Devon 747,700 354 388 399 400 419 0.90 0.81 1.01
Dorset 413,800 418 428 445 442 442 0.94 0.81 1.08
Gloucestershire 598,300 328 329 341 374 371 0.82 0.72 0.93
69

Age and gender
Population
South West North Somerset 203,100 384 409 433 443 483 1.05 0.86 1.28
Plymouth Teaching 256,600 468 503 511 546 573 1.35 1.14 1.58
Somerset 531,600 348 367 386 420 421 0.92 0.80 1.04
South Gloucestershire 263,400 444 448 475 490 509 1.14 0.96 1.35
Swindon 214,900 335 349 409 428 437 1.01 0.82 1.23
Torbay 131,200 404 450 473 495 495 1.07 0.84 1.36
Wiltshire 474,300 310 314 346 371 386 0.86 0.74 0.99
Wales Betsi Cadwaladr University 688,700 327 338 354 351 348 0.77 0.67 0.87
Powys Teaching 133,200 360 375 413 405 375 0.78 0.59 1.04
Hywel Dda 381,900 380 401 398 424 424 0.93 0.79 1.08
Abertawe Bro Morgannwg University 517,700 433 454 487 547 579 1.30 1.16 1.46
Cwm Taf 293,500 535 569 630 664 685 1.55 1.35 1.78
Aneurin Bevan 577,000 437 458 501 520 584 1.31 1.18 1.46
Cardiff and Vale University 472,300 394 404 436 464 502 1.22 1.07 1.39
Scotland Ayrshire & Arran 373,800 399 396 393 388 415 0.89 0.76 1.04
Borders 113,900 378 386 448 448 509 1.05 0.81 1.36
Dumfries and Galloway 151,400 363 383 390 409 409 0.85 0.66 1.09
Fife 365,300 315 323 342 367 389 0.85 0.72 1.00
Forth Valley 298,100 295 295 315 339 369 0.81 0.67 0.98
Grampian 569,600 348 377 393 404 421 0.93 0.82 1.06
Greater Glasgow & Clyde 1,214,600 424 431 444 460 510 1.16 1.07 1.25
Highland 321,700 423 476 504 494 497 1.04 0.89 1.22
Lanarkshire 572,400 383 404 416 440 479 1.05 0.94 1.19
Lothian 836,600 326 339 357 377 390 0.89 0.80 1.00
Orkney 21,400 514 420 373 373 373 0.77 0.39 1.54
Shetland 23,200 215 258 258 215 258 0.56 0.25 1.25
Tayside 410,300 422 417 419 429 441 0.98 0.84 1.13
Western Isles 27,700 289 289 289 325 325 0.67 0.35 1.29
Northern Ireland Belfast 348,300 362 379 422 431 459 1.12 0.96 1.31
Northern 463,500 339 356 371 390 406 0.95 0.82 1.10
Southern 359,400 298 300 320 359 403 0.99 0.84 1.16
South Eastern 347,700 348 359 359 394 403 0.93 0.78 1.09
Western 295,300 305 322 342 359 369 0.89 0.73 1.07
patients who were classified as ethnicity ‘unknown’ each renal centre, particularly for blood pressure. Better
(table 3.8). The percentages of patients with unknown eth- data records (or possibly better extraction of data held
nicity between 2007 and 2012 provided in this year’s chap- within renal IT systems) would facilitate more meaning-
ter are different from those in last year’s chapter [3]; this ful comparisons between centres and help to determine
reflects retrospective input of ethnicity data, improving the causes of inter-centre differences in outcomes. For
data completeness. this reason, along with differences in repatriation policies
of prevalent transplant patients between centres as high-
lighted previously, caution needs to be exercised when
comparing centre performance.
Clinical and laboratory outcomes The 71 renal centres in the UK comprise 52 centres in
England, five in Wales, five in Northern Ireland and nine
Introduction in Scotland. Four centres (Bangor, Colchester, Liverpool
There continued to be marked variation in the Aintree and Wirral) were reported as having no trans-
completeness of data (tables 3.9a, 3.9b) reported by planted patients and were therefore excluded. After
70
Table 3.5. Distribution of prevalent patients on RRT by centre and modality on 31/12/2012
Centre Total % HD % PD % Transplant
Transplant centres
B QEH 1,971 47 8 45
Belfast 701 33 4 63
Bristol 1,337 37 5 58
Camb 1,113 31 3 65
Cardff 1,548 31 5 64
Covnt 900 40 11 49
Edinb 722 37 5 58
Glasgw 1,549 40 3 57
L Barts 1,955 46 10 44
L Guys 1,745 36 2 62
L Rfree 1,865 38 6 55
L St George’s 724 39 7 53
L West 3,104 46 2 52
Leeds 1,416 35 6 59
Leic 1,982 44 8 48
Livrpl RI 1,241 29 5 65
Man RI 1,710 30 5 66
Newc 946 30 5 65
Nottm 1,006 37 8 55
Oxford 1,535 28 6 67
Plymth 459 29 8 64
Ports 1,447 38 6 56
Sheff 1,307 45 5 50
Dialysis centres
Abrdn 504 46 5 49
Airdrie 388 50 3 47
Antrim 225 59 6 36
B Heart 670 65 7 28
Bangor 105 86 14
Basldn 264 62 12 26
Bradfd 508 41 6 53
Brightn 831 45 10 45
Carlis 216 28 13 59
Carsh 1,475 52 8 41
Chelms 224 58 12 31
Clwyd 172 49 10 41
Colchester 117 100
D & Gall 128 40 13 48
Derby 477 46 19 35
Doncaster 261 66 11 23
Dorset 610 43 8 50
Dudley 316 53 20 27
Dundee 403 45 5 50
Dunfn 278 53 7 40
Exeter 846 47 9 44
Glouc 417 53 9 39
Hull 789 42 11 46
Inverns 218 34 8 58
Ipswi 339 38 9 53
Kent & Canterbury 922 42 7 52
Klmarnk 302 50 14 37
L Kings 918 54 9 37
Livrpl Ain 195 90 10
Middlbr 789 43 1 56
Newry 188 48 9 43
71
Centre Total % HD % PD % Transplant
Norwch 612 52 9 39
Prestn 1,081 50 6 44
Redng 671 40 11 49
Salford 882 43 12 45
Shrew 354 55 12 33
Stevng 665 62 5 34
Sthend 213 55 7 38
Stoke 695 44 11 45
Sund 421 47 5 48
Swanse 662 50 10 40
Truro 377 41 6 53
Ulster 148 73 5 22
West NI 258 52 7 40
Wirral 234 86 14
Wolve 528 54 17 29
Wrexm 249 39 9 53
York 396 34 8 58
England 46,076 43 7 50
Northern Ireland 1,520 46 6 49
Scotland 4,492 43 5 52
Wales 2,736 39 7 53
UK 54,824 43 7 50
Blank cells denote no patients on that modality
Table 3.6. Median age and gender ratio of incident and prevalent transplant patients 2007–2012
Incident transplants Prevalent transplants∗
Year N Median age M : F ratio N Median age M : F ratio
2007 2,133 45.6 1.6 20,744 50.2 1.5

2008 2,343 46.4 1.5 22,229 50.4 1.5
2009 2,493 48.3 1.6 23,480 50.8 1.5
2010 2,581 49.6 1.7 24,876 51.2 1.6
2011 2,625 49.1 1.7 26,168 51.7 1.6
2012 2,782 50.5 1.6 27,621 52.2 1.6
∗
As on 31st December for given year
1,300
1,200 Males
1,100 Females
All UK
1,000
900
800
700
600
500
400
300
200
100
0
20–24
25–29
30–34
35–39
40–44
45–49
50–54
55–59
60–64
65–69
70–74
75–79
80–84
85+
Fig. 3.1. Transplant prevalence rate per

million population by age and gender on
Age group
31/12/2012
72
Table 3.7. Primary renal diagnosis in renal transplant recipients 2007–2012
New transplants by year Established transplants on 01/01/2012
2007 2008 2009 2010 2011 2012

Primary diagnosis % % % % % % N % N
Aetiology uncertain 15.2 14.5 14.0 13.8 14.4 11.9 322 15.8 4,140
Diabetes 14.9 12.9 12.8 11.8 12.5 14.8 399 9.3 2,428
Glomerulonephritis 23.2 21.9 23.3 19.4 22.6 22.5 609 23.1 6,050
Polycystic kidney disease 13.4 13.4 13.1 13.3 12.3 13.3 359 12.6 3,294
Pyelonephritis 11.7 12.1 11.2 9.3 10.1 9.8 265 13.6 3,555
Reno-vascular disease 5.4 6.7 5.9 6.8 6.5 6.8 184 5.6 1,471
Other 15.0 16.5 15.2 15.6 16.4 17.4 470 16.5 4,311
Not available 1.0 1.9 4.5 10.1 5.3 3.5 95 3.5 919
exclusion of these four centres, prevalent patient data Centres with ,20 patients or ,50% data completeness have
from 67 renal centres across the UK were analysed. been excluded from the figures. Scottish centres were also excluded
from blood pressure analyses as data not provided.
For the one year post-transplant analyses, in which
patients were assigned to the centre that performed Prevalent patient data
their transplant, all 23 transplant centres across the UK Biochemical and clinical data for patients with a functioning
were included in the analysis for the first time this year. transplant followed in either a transplanting or non-transplanting
centre were included in the analyses. The cohort consisted of
Methods prevalent patients as on 31st December 2012. Patients were con-
Data for key laboratory variables are reported for all prevalent sidered as having a functioning transplant if ‘transplant’ was listed
patients with valid data returns for a given renal centre (both as the last mode of RRT in the last quarter of 2012. Patients were
transplanting and non-transplanting centres) and for one year assigned to the renal centre that sent the data to the UKRR but
post-transplant results for patients transplanted 2005–2011, with some patients will have received care in more than one centre. If
patients attributed to the transplant centre that performed the data for the same transplant patient were received from both the
procedure. transplant centre and non-transplant centre, care was allocated
Time since transplantation may have a significant effect on to the non-transplant centre. Patients with a functioning trans-
key biochemical and clinical variables and this is likely to be plant of less than three months duration were excluded from
independent of a centre’s clinical practices. Therefore, inter-centre analyses. For haemoglobin, estimated glomerular filtration rate
comparison of data on prevalent transplant patients is open to (eGFR), corrected calcium, phosphate and blood pressure (BP),
bias. To minimise bias relating to fluctuations in biochemical the latest value in quarter 3 or quarter 4 of 2012 was used.
and clinical parameters occurring in the initial post-transplant
period, one year post-transplantation outcomes are also reported. Estimated glomerular filtration rate (eGFR)
It is presumed that patient selection policies and local clinical For the purpose of eGFR calculation, the original 4–variable
practices are more likely to be relevant in influencing outcomes MDRD formula was used (with a constant of 186) to calculate
12 months post-transplant and therefore comparison of outcomes eGFR from the serum creatinine concentration as reported by
between centres is more robust. However, even the 12 months the centre (unless otherwise stated). A wide variety of creati-
post-transplant comparisons could be biased by the fact that in nine assays are in use in clinical biochemistry laboratories in
some centres, repatriation of patients only occurs if the graft is the UK, and it is not possible to ensure that all measurements
failing whereas in others it only occurs if the graft function is of creatinine concentration collected by the UKRR are harmo-
stable. nised. Although many laboratories are now reporting assay
Table 3.8. Ethnicity of patients who received a transplant in the years 2007–2012
Year % White % S Asian % Black % Other % Unknown
2007 76.9 8.2 5.5 2.1 7.3

2008 74.8 8.9 6.3 1.8 8.2
2009 73.5 10.3 6.7 2.4 7.1
2010 74.4 10.4 5.9 2.3 6.9
2011 74.1 9.5 6.2 2.5 7.7
2012 71.8 9.8 7.2 2.9 8.2
73
Table 3.9a. Percentage completeness by centre for prevalent transplant patients on 31/12/2012a
Blood Blood
Centre N Ethnicity eGFRb pressure Centre N Ethnicity eGFRb pressure
England Prestn 466 100 98 0

B Heart 181 100 97 3 Redngc 326 100 98 0
B QEH 858 100 94 93 Salford 384 100 97 0
Basldn 66 100 100 2 Sheff 627 100 99 97
Bradfd 261 98 86 69 Shrew 116 100 63 1
Brightn 363 97 88 0 Stevng 216 100 67 23
Bristol 755 100 99 72 Sthend 79 100 99 61
Camb 686 98 99 97 Stoke 309 65 98 0
Carlis 123 100 96 0 Sund 194 100 100 0
Carsh 565 97 90 0 Truro 191 100 98 19
Chelms 67 97 96 94 Wolve 146 100 98 95
Covnt 423 100 95 81 York 221 95 99 53
Derby 159 100 96 83 N Ireland
Donc 59 100 100 100 Antrim 79 100 99 65
Dorset 293 100 89 81 Belfast 432 100 98 45
Dudley 83 100 96 16 Newry 78 100 100 86
Exeter 365 100 99 92 Ulster 31 100 97 90
Glouc 156 100 100 89 West NI 101 100 98 93
Hull 348 61 97 25 Scotland
Ipswi 178 100 98 0 Abrdn 245 62 98 n/a
Kent 454 100 59 85 Airdrie 178 41 63 n/a
L Barts 836 100 99 0 D & Gall 61 13 95 n/a
L Guys 1,054 99 94 0 Dundee 196 72 98 n/a
L Kings 329 98 99 0 Dunfn 107 28 96 n/a
L RFree 1,002 98 98 77 Edinb 401 10 96 n/a
L St.G 378 88 96 0 Glasgw 841 10 82 n/a
L West 1,590 100 96 0 Inverns 121 94 13 n/a
Leeds 810 99 97 96 Klmarnk 110 76 65 n/a
Leic 930 97 97 48 Wales
Liv RI 794 100 89 2 Cardff 964 100 99 98
M RI 1,080 99 98 0 Clwyd 68 99 0 0
Middlbr 423 100 96 46 Swanse 251 100 100 100
Newc 603 100 99 0 Wrexm 129 100 78 0
Norwch 232 100 97 41 England 22,356 98 95 37
Nottm 535 100 100 87 N Ireland 721 100 98 60
Oxford 978 97 99 16 Scotland 2,260 32 83 n/a
Plymth 280 100 97 85 Wales 1,412 100 92 84
Ports 784 100 95 19 UK 26,749 93 94 41
a
Scottish centres excluded from blood pressure analysis as data not provided by the Scottish Renal Registry
b
Patients with missing ethnicity were classed as White for eGFR calculation
c
Data relating to blood pressure could not be extracted from this centre due to technical problems
results that have been aligned to the isotope dilution-mass a patient’s record is from a timeline entry in data returned from
spectrometry standard (which would necessitate use of the a non-transplant centre, in these instances the patient was re-
modified MDRD formula), this was not the case at the end of assigned to the nearest transplant centre (table 3.10).
2012. Patients with valid serum creatinine results but no ethni- Patients who had died or experienced graft failure within 12
city data were classed as White for the purpose of the eGFR months of transplantation were excluded from the analyses.
calculation. Patients with more than one transplant during 2005–2011 were
included as separate episodes provided each of the transplants
One year post-transplant data functioned for a year.
Patients who received a renal transplant between 1st January For each patient, the most recent laboratory or blood pressure
2005 and 31st December 2011 were assigned according to the result for the relevant 4th/5th quarter (10–15 months) after renal
renal centre in which they were transplanted. In a small number transplantation was taken to be representative of the one year
of instances, the first documented evidence of transplantation in post-transplant outcome. Again, for the purpose of the eGFR
74
Table 3.9b. Percentage completeness by centre for prevalent transplant patients on 31/12/2012a
Total serum Adjusted serum Serum Serum

Centre N Haemoglobin cholesterol calciumb phosphate PTH
England
B Heart 181 95 41 92 92 2
B QEH 858 94 73 94 93 0
Basldn 66 98 47 98 62 32
Bradfd 261 82 47 83 75 55
Brightn 363 88 20 79 79 20
Bristol 755 99 70 99 99 98
Camb 686 99 77 99 99 93
Carlis 123 94 65 93 88 19
Carsh 565 90 47 88 88 0
Chelms 67 94 66 96 81 25
Covnt 423 95 0 92 75 38
Derby 159 94 75 92 89 79
Donc 59 100 86 100 100 22
Dorset 293 89 55 85 60 20
Dudley 83 96 63 98 98 41
Exeter 365 99 71 98 97 21
Glouc 156 100 43 97 97 40
Hull 348 97 21 97 97 18
Ipswi 178 98 38 98 98 62
Kent 454 95 45 93 93 0
L Barts 836 98 98 99 99 67
L Guys 1,054 94 33 89 89 33
L Kings 329 99 41 99 99 22
L RFree 1,002 98 67 97 97 71
L St.G 378 96 16 96 96 16
L West 1,590 96 20 96 96 17
Leeds 810 97 85 97 97 49
Leic 930 96 88 96 96 56
Liv RI 794 89 57 85 87 68
M RI 1,080 99 43 98 98 59
Middlbr 423 95 31 92 91 12
Newc 603 98 69 98 98 45
Norwch 232 98 93 94 94 24
Nottm 535 100 55 97 92 78
Oxford 978 99 55 98 98 29
Plymth 280 97 41 95 94 42
Ports 784 94 35 92 88 17
Prestn 466 98 41 95 92 2
Redng 326 98 76 97 80 40
Salford 384 91 76 94 94 82
Sheff 627 99 41 99 99 25
Shrew 116 91 67 77 78 7
Stevng 216 96 70 91 88 54
Sthend 79 99 29 96 96 13
Stoke 309 98 98 98 98 39
Sund 194 100 85 100 100 88
Truro 191 98 60 96 96 57
Wolve 146 97 60 94 82 37
York 221 85 55 98 95 21
75
Table 3.9b. Continued
Total serum Adjusted serum Serum Serum

Centre N Haemoglobin cholesterol calciumb phosphate PTH
N Ireland
Antrim 79 96 99 96 99 97
Belfast 432 97 97 97 97 24
Newry 78 99 99 99 99 83
Ulster 31 97 97 97 97 55
West NI 101 97 96 92 93 60
Scotland
Abrdn 245 98 n/a n/a 96 n/a
Airdrie 178 98 n/a n/a 98 n/a
D & Gall 61 100 n/a n/a 95 n/a
Dundee 196 98 n/a n/a 97 n/a
Dunfn 107 96 n/a n/a 95 n/a
Edinb 401 95 n/a n/a 94 n/a
Glasgw 841 99 n/a n/a 98 n/a
Inverns 121 4 n/a n/a 2 n/a
Klmarnk 110 96 n/a n/a 95 n/a
Wales
Cardff 964 99 74 99 98 12
Clwyd 68 94 94 94 94 59
Swanse 251 98 68 98 98 57
Wrexm 129 97 89 97 97 95
England 22,356 96 55 95 93 40
N Ireland 721 97 97 97 97 45
Scotlanda 2,260 93 n/a n/a 91 n/a
Wales 1,412 98 75 98 98 30
UK 26,749 96 57c 95c 93 40c
a
Limited dataset provided by the Scottish Renal Registry for Scottish centres shown and included in corresponding UK analyses
b
Serum calcium corrected for serum albumin
c
Excluding Scotland
calculation patients with valid serum creatinine results but missing in estimating GFR 560 ml/min/1.73 m2 is questionable
ethnicity data were classed as White. [5], therefore a figure describing this is not included in
this chapter.
Results and discussion Figure 3.4 shows the percentage of prevalent patients by
Post-transplant eGFR in prevalent transplant patients centre with eGFR ,30 ml/min/1.73 m2 as a funnel plot,
When interpreting eGFR post-transplantation, it is enabling a more reliable comparison of outcomes between
important to remember that estimated GFR formulae centres across the UK. The solid lines show the 2 standard
only have a modest predictive performance in the trans- deviation limits (95%) and the dotted lines the limits for 3
plant population [4]. Median eGFR in each centre and standard deviations (99.9%). With 65 centres included and
percentage of patients with eGFR ,30 ml/min/1.73 m2 a normal distribution, 3–4 centres would be expected to
are shown in figures 3.2 and 3.3. The median eGFR fall between the 95–99% CI (1 in 20) and no centres should
was 51.3 ml/min/1.73 m2, with 13.7% of prevalent trans- fall outside the 99.9% limits.
plant recipients having an eGFR ,30 ml/min/1.73 m2. There continued to be variation between centres; these
Table 3.11 summarises the proportion of transplant data show over-dispersion with 17 centres falling outside
patients with an eGFR ,30 ml/min/1.73 m2 by centre. the 95% CI of which eight centres were outside the
Whilst local repatriation policies on timing of transfer 99.9% CI. Four centres (Newry, London St Georges,
of care for patients with failing transplants from trans- London West, Nottingham) fell outside the lower 99.9%
plant centres to referring centres might explain some of CI suggesting a lower than expected proportion of
the differences, it is notable that both transplanting and patients with eGFR ,30 ml/min/1.73 m2. Liverpool RI,
non-transplanting centres feature at both ends of the Portsmouth, Manchester RI and London Barts fell out-
scale. The accuracy of the 4–variable MDRD equation side the upper 99.9% CI suggesting a higher than
76
Table 3.10. Number of patients per transplant centre after allocation of patients in non-transplant centres (transplanted between
2005–2011)
Total patients per Patients reallocated to

transplant centre a transplant centre
Transplant centre N Non-transplant centre N
B QEH 877 Stoke 2

Belfast 331 Antrim 2
Newry 7
Ulster 1
West NI 7
Bristol 687 Dorset 2
Camb 1,029 Stevng 1
Cardff 731 Swansea 2
Covnt 357 n/a
Edinb 606 Abrdn 5
Dundee 8
Inverns 2
Glasgw 570 Airdrie 1
L Barts 678 n/a
L Guys 1,156 Basldn 1
Kent 1
L Kings 2
L Rfree 578 n/a
L St.G 455 Carsh 2
L West 1,075 n/a
Leeds 903 n/a
Leic 526 n/a
Liv RI 559 Prestn 1
M RI 866 Salford 2
Newc 778 Middlbr 2
Nottm 377 n/a
Oxford 1,063 n/a
Plymth 416 n/a
Ports 424 n/a
Sheff 377 n/a
Total 15,419 51
80
70
Median eGFR ml/min/1.73 m2
60
50
40
Upper quartile
30 Median eGFR
Lower quartile N = 25,151
20
14 Bradfd
0 Newry
2 Stoke
1 Cardff
2 Abrdn
3 B Heart
1 Antrim
4 Edinb
1 York
4 L St.G
1 Sheff
2 Belfast
5 D&Gall
3 Ulster
10 Carsh
4 L West
2 West NI
2 Dundee
5 Covnt
35 Klmarnk
3 Leeds
18 Glasgw
4 Carlis
0 Donc
0 Nottm
6 L Guys
4 Chelms
4 Middlbr
1 Exeter
2 Wolve
6 B QEH
1 L Kings
4 Derby
37 Airdrie
1 Newc
1 Oxford
3 Leic
33 Stevng
1 Bristol
12 Brightn
2 L Rfree
0 Glouc
2 Redng
3 Hull
11 Dorset
2 Truro
1 Sthend
4 Dudley
0 Sund
4 Dunfn
3 Plymth
0 Swanse
2 Prestn
37 Shrew
22 Wrexm
3 Salford
0 Basldn
2 M RI
1 L Barts
2 Ipswi
3 Norwch
5 Ports
1 Camb
41 Kent
11 Liv RI
5 England
2 N Ireland
17 Scotland
8 Wales
6 UK
Centre
Fig. 3.2. Median eGFR in prevalent transplant patients by centre on 31/12/2012
77
35
N = 25,151 Upper 95% Cl
30 % with eGFR 0–29
Lower 95% Cl
25
20
15
10
0
0 Newry
0 Donc
5 D&Gall
4 L St.G
2 Stoke
0 Nottm
2 Belfast
2 Wolve
2 Dundee
3 Ulster
2 West NI
5 Covnt
1 Sthend
4 L West
1 York
4 Dudley
3 B Heart
4 Carlis
1 Bristol
4 Derby
4 Edinb
0 Glouc
1 Antrim
6 L Guys
2 Redng
1 Exeter
1 Cardff
6 B QEH
1 Sheff
3 Leeds
2 Abrdn
4 Middlbr
2 Ipswi
11 Dorset
1 L Kings
3 Leic
10 Carsh
33 Stevng
37 Airdrie
3 Plymth
2 L Rfree
1 Newc
3 Norwch
1 Oxford
12 Brightn
37 Shrew
18 Glasgw
0 Sund
35 Klmarnk
2 Truro
4 Dunfn
3 Hull
14 Bradfd
0 Swanse
1 Camb
0 Basldn
4 Chelms
22 Wrexm
3 Salford
2 M RI
1 L Barts
41 Kent
2 Prestn
11 Liv RI
5 Ports
5 England
2 N Ireland
17 Scotland
8 Wales
6 UK
Centre
Fig. 3.3. Percentage of prevalent transplant patients by centre on 31/12/2012 with eGFR ,30 ml/min/1.73 m2
Table 3.11. Proportion of prevalent transplant patients with eGFR ,30 ml/min/1.73 m2 on 31/12/2012
Patients with eGFR data Percentage with Patients with eGFR data Percentage with
Centre N eGFR ,30 Centre N eGFR ,30
Ulster 30 10.0 Stoke 304 8.6

D & Gall 58 6.9 Brightn 319 15.0
Donc 59 6.8 Redng 319 11.6
Chelms 64 17.2 L Kings 326 12.9
Basldn 66 16.7 Hull 337 16.0
Klmarnk 71 15.5 Exeter 361 11.6
Shrew 73 15.1 L St.G 364 8.2
Dudley 75 10.7 Salford 371 18.1
Antrim 78 11.5 Edinb 383 11.5
Newry 78 3.8 Covnt 403 10.2
Sthend 78 10.3 Middlbr 406 12.6
West NI 99 10.1 Belfast 424 9.2
Wrexm 100 18.0 Prestn 458 19.4
Dunfn 103 15.5 Carsh 509 13.0
Airdrie 113 13.3 Nottm 533 9.0
Carlis 118 11.0 Newc 596 14.1
Wolve 143 9.8 Sheff 621 12.2
Stevng 144 13.2 Camb 682 16.4
Derby 152 11.2 Glasgw 689 15.1
Glouc 156 11.5 Liv RI 705 20.4
Ipswi 175 12.6 Ports 739 22.1
B Heart 175 10.9 Bristol 745 11.1
Truro 187 15.5 Leeds 789 12.4
Dundee 192 9.9 B QEH 806 12.2
Sund 194 15.5 L Barts 824 18.2
York 218 10.6 Leic 899 12.9
Bradfd 224 16.1 Cardff 954 11.9
Norwch 226 14.2 Oxford 965 14.5
Abrdn 240 12.5 L Rfree 980 14.1
Swanse 250 16.4 L Guys 994 11.6
Dorset 262 12.6 M RI 1,060 18.1
Kent 267 18.4 L West 1,530 10.5
Plymth 272 14.0
78
40 A number of factors including comorbidity, immuno-

Dotted lines show 99.9% limits
35 Solid lines show 95% limits suppressive medication, graft function, ACE inhibitor use,
erythropoietin (EPO) use, intravenous or oral iron use, as
30
25 well as centre practices and protocols for management of
20 anaemia, affect haemoglobin concentrations in transplant
15
patients. Most of these data are not collected by the
10
UKRR and therefore caution must be used when interpret-
ing analyses of haemoglobin attainment. Figures 3.7a and
5
3.7b report centre results stratified according to graft func-
0
0 200 400 600 800 1,000 1,200 1,400 1,600 tion as estimated by eGFR. The percentage of prevalent
Number of patients with data in centre transplant patients achieving Hb 5100 g/L in each centre,
Fig. 3.4. Funnel plot of percentage of prevalent transplant stratified by eGFR, is displayed in figures 3.8a and 3.8b.
patients with eGFR ,30 ml/min/1.73 m2 by centre size on Figure 3.9 describes the percentage of prevalent
31/12/2012 patients by centre with haemoglobin ,100 g/L as a fun-
nel plot enabling more reliable comparison of outcomes
expected proportion of patients with eGFR ,30 ml/min/ between centres across the UK. With 65 centres included
1.73 m2. and a normal distribution, 3–4 centres would be expected
to fall between the 95%–99.9% CI (1 in 20) and no centres
eGFR in patients one year after transplantation should fall outside the 99.9% CI purely as a chance event.
Graft function at one year post-transplantation may One centre (London Barts) fell outside the upper
predict subsequent long term graft outcome [6]. 99.9% CI and three further centres (London Royal Free,
Figures 3.5a, 3.5b, and 3.5c show the median one year Norwich and Oxford) fell outside the upper 95% CI indi-
post-transplant eGFR for patients transplanted between cating a higher than predicted proportion of transplant
2005–2011, by transplant type. Living kidney donation patients not achieving the haemoglobin target. Six centres
had the highest median eGFR at one year (56.4 ml/min/ fell outside the lower 99.9% CI, indicating they performed
1.73 m2), followed by donation after brainstem death better than expected with fewer than predicted patients
(52.7 ml/min/1.73 m2) and donation after circulatory having a haemoglobin ,100 g/L.
death (49.4 ml/min/1.73 m2).
Figures 3.6a, 3.6b and 3.6c show one year post-trans- Blood pressure in prevalent transplant patients
plant eGFR by donor type and year of transplantation. In the absence of controlled trial data, the opinion
An upward trend in eGFR (p , 0.001) over the time based recommendation of the UK Renal Association
period was noticed with both live and donation after (RA) published in the 2010 guideline for the care of
brainstem death transplant, but not with donation after kidney transplant recipients is that ‘Blood pressure
circulatory death (p = 0.5). should be <130/80 mmHg (or <125/75 mmHg if protein-
uria)’ [9]. This blood pressure target is the same as that
Haemoglobin in prevalent transplant patients used in previous annual reports [10].
Transplant patients have previously fallen under the As indicated in table 3.9a, completeness for blood
remit of the UK Renal Association Complications of pressure data returns was variable and only centres
Chronic Kidney Disease (CKD) guidelines. Updated with .50% data returns were included for consideration.
guidelines regarding the management of anaemia in Despite this restriction, caution needs to be exercised in
CKD were published by the association in November interpretation of these results because of the volume of
2010 [7] which have now been adopted for this report. missing data and potential bias, (e.g. a centre may be
These guidelines recommend achieving a population more likely to record and report blood pressure data
distribution centred on a mean of 11 g/dl with a range electronically in patients with poor BP control).
of 10–12 g/dl [8] (equivalent to 110 g/L, range 100– Figures 3.10a and 3.10b show the percentage of patients
120 g/L). However, many transplant patients with good with a blood pressure of ,130/80 mmHg, by eGFR.
transplant function will have haemoglobin con- The percentage of patients with BP ,130/80 (systolic
centrations .120 g/L without the use of erythopoiesis BP ,130 and diastolic BP ,80 mmHg) was higher
stimulating agents, and so it is inappropriate to audit (27.6% vs. 24.4%) in those with better renal function
performance using the higher limit. (eGFR 530 ml/min/1.73 m2).
79
80
Median eGFR ml/min/1.73 m2 Median eGFR ml/min/1.73 m2 Median eGFR ml/min/1.73 m2
30
40
50
60
70
80
30
40
50
60
70
80
35
40
45
50
55
60
65
70
75
5 L West 0 Belfast 80
1 L St.G
11 Glasgw 5 L Guys 11 Glasgw
4 L Barts 1 L St.G 2 L Rfree

The UK Renal Registry
0 Belfast 2 M RI
2 L Rfree
1 Oxford 2 Cardff
2 Cardff
5 Edinb 2 Ports
1 Oxford
11 Glasgw 1 Newc
5 L Guys
2 M RI 1 L St.G
0 Sheff
2 Plymth 0 Sheff
5 Covnt
2 Cardff 5 L West
2 Plymth
4 L Barts 2 Leeds
7 B QEH 2 L Rfree 2 Plymth
1 Newc 2 Leeds 7 B QEH
5 Edinb 1 Bristol 3 Liv RI
3 Liv RI 5 Camb 1 Oxford
Transplant centre
Transplant centre
Transplant centre
1 Bristol 0 Sheff 5 Covnt
5 L West 5 Covnt 5 Edinb
0 Leic 5 L Guys
2 Leeds
N = 2,370
N = 5,765
N = 4,725
1 Newc 1 Bristol
3 Nottm
Fig. 3.5a. Median eGFR one year post-live donor transplant by transplant centre 2005–2011
7 B QEH 0 Leic
5 Camb
3 Nottm 4 L Barts
2 M RI
3 Liv RI 3 Nottm
0 Leic
Median eGFR
Median eGFR
Median eGFR
Lower quartile
Lower quartile
Lower quartile
2 Ports 5 Camb
Upper quartile
Upper quartile
Upper quartile
Fig. 3.5b. Median eGFR one year post-brainstem death donor transplant by transplant centre 2005–2011
Fig. 3.5c. Median eGFR one year post-circulatory death donor transplant by transplant centre 2005–2011
2 Ports
3 England 3 England
3 England
0 N Ireland 0 N Ireland
8 Scotland 8 Scotland 8 Scotland
2 Wales 2 Wales 2 Wales
3 UK 3 UK 3 UK
The Sixteenth Annual Report
80
N = 329 N = 439 N = 668 N = 756 N = 828 N = 853 N = 852
75
70
65
60
55
50
45
40 Upper quartile
Median eGFR
35
Lower quartile
30
2005 2006 2007 2008 2009 2010 2011
Year
Fig. 3.6a. Median eGFR one year post-live donor transplant by year of transplantation 2005–2011
80
N = 666 N = 720 N = 869 N = 871 N = 885 N = 893 N = 861
75
70
65
60
55
50
45
40
35 Upper quartile
Median eGFR
30 Lower quartile
25
2005 2006 2007 2008 2009 2010 2011
Year
Fig. 3.6b. Median eGFR one year post-brainstem death donor transplant by year of transplantation 2005–2011
80
N = 139 N = 167 N = 262 N = 380 N = 431 N = 483 N = 508
75
70
65
60
55
50
45
40
35
Upper quartile
30 Median eGFR
25 Lower quartile
20
2005 2006 2007 2008 2009 2010 2011
Year
Fig. 3.6c. Median eGFR one year post-circulatory death donor transplant by year of transplantation 2005–2011
81
82
Percentage of patients Median Hb g/dl Median Hb g/dl
31/12/2012
75
80
85
90
95
100
11.0
11.5
12.0
12.5
13.0
13.5
14.0
14.5
15.0
15.5
16.0
9.0
9.5
10.0
10.5
11.0
11.5
12.0
12.5
13.0
13.5
14.0
14.5
15.0
2 Wolve 3 Carlis
1 West NI 0 Truro
1 West NI
0 Dunfn 4 Stoke 0 Carsh
1 Dudley 0 Sheff 0 Hull
2 Chelms 2 Chelms
0 Exeter 0 Sund 0 Bristol
0 Plymth 0 Bristol 1 Middlbr
1 Dorset 0 Stoke
0 Sund 0 Covnt 1 Dorset
Upper 95% Cl
Lower 95% Cl
3 Carlis 2 Swanse 2 Swanse
2 Belfast 1 Newry
0 Stoke 0 Liv RI 1 Antrim
% with Hb >100 g/L

1 Prestn 0 B QEH 0 Donc
0 Dundee 0 Cardff
1 Wrexm 1 Newc 0 Sheff
0 Truro 0 Exeter 1 Prestn
0 Sheff 0 Exeter
0 Ipswi 0 Belfast
0 Dundee
1 Newry 0 Plymth 0 Truro
0 Carsh 0 Shrew
N = 21,535
2 Swanse 6 Dorset
3 Klmarnk
0 B QEH 1 Ports 0 Ulster
1 Newc 0 Cardff 2 Belfast
0 Stevng 0 Glasgw
3 Klmarnk 0 Carsh 1 Newc
0 Redng 0 Hull 1 Covnt
2 Leic 1 Edinb
0 D&Gall 0 Edinb 1 Brightn
0 Airdrie 13 York 0 Plymth
0 Bristol 7 Bradfd
0 Glouc 0 M RI 0 Sund
1 Covnt 0 Prestn 2 B Heart
1 Ports 0 Glouc
3 Derby 0 Middlbr 3 Derby
Centre
Centre
0 Abrdn 2 Brightn 2 Wolve
0 Camb 0 Stevng
Centre
6 Salford 1 Oxford
0 Dunfn
0 Hull 0 Leic 0 D&Gall
1 Middlbr 0 Camb 1 Liv RI
1 Brightn 0 Nottm
2 B Heart 0 Leeds 0 Redng
1 Leeds 6 Salford 0 Abrdn
1 Edinb 15 York
0 Glasgw 0 Redng 0 M RI
0 Cardff 1 L West 0 Oxford
0 L Guys 0 B QEH
0 Sthend 0 Kent 1 Leeds
1 Liv RI 3 L St.G 0 Camb
0 M RI 6 Salford
0 Nottm 0 Glasgw 0 Ipswi
1 Antrim 0 Bradfd 2 Leic
0 L West 1 Ports
0 Shrew 0 Ipswi
0 L St.G
0 L St.G 0 L Rfree 0 Airdrie
15 York 1 Dudley
0 Donc 0 L Guys
2 Basldn
0 Kent 0 Nottm 0 L Guys
0 Ulster 0 Norwch 0 Norwch
N = 3,418
0 L Rfree
N = 21,532
0 Sthend
7 Bradfd 0 Abrdn 0 L Kings
0 L Kings 1 L Barts 0 Kent
0 Oxford 1 Wrexm
2 Basldn 0 L Kings 0 L West
0 L Barts 1 England 0 L Rfree
0 Norwch 0 L Barts
2 N Ireland
Median
Median
1 England 1 England
1 N Ireland 0 Scotland 1 N Ireland
Fig. 3.7b. Median haemoglobin for prevalent transplant patients with eGFR ,30 ml/min/1.73 m2 by centre on 31/12/2012
Fig. 3.7a. Median haemoglobin for prevalent transplant patients with eGFR 530 ml/min/1.73 m2 by centre on 31/12/2012
1 Scotland 1 Scotland
1 Wales 1 Wales
Lower quartile
1 Wales
Lower quartile
Upper quartile
Upper quartile
1 UK 1 UK 1 UK
Fig. 3.8a. Percentage of prevalent transplant patients with eGFR 530 ml/min/1.73 m2 achieving haemoglobin 5100 g/L by centre on
100
90
80
70
Upper 95% Cl
60 % with Hb >100 g/L
Lower 95% Cl N = 3,418
50
0 Truro
4 Stoke
2 Brightn
0 Ipswi
0 Exeter
0 Covnt
1 Ports
0 Belfast
0 Sund
0 Bradfd
0 Bristol
0 Liv RI
0 Plymth
0 Hull
0 Sheff
2 Swanse
1 Oxford
1 L West
0 Prestn
0 Carsh
0 Edinb
0 Cardff
0 L Kings
13 York
0 M RI
0 Leeds
0 Glasgw
0 Middlbr
0 Redng
1 Newc
0 Camb
0 B QEH
0 Leic
0 Norwch
6 Salford
0 L Rfree
0 Kent
3 L St.G
0 Nottm
0 L Guys
1 L Barts
6 Dorset
0 Abrdn
1 England
2 N Ireland
0 Scotland
1 Wales
1 UK
Centre
Fig. 3.8b. Percentage of prevalent transplant patients with eGFR ,30 ml/min/1.73 m2 achieving haemoglobin 5100 g/L by centre on
31/12/2012
15
Dotted lines show 99.9% limits Analysis of prevalent patients by CKD stage
Solid lines show 95% limits
10
Introduction
Approximately 2.2% of prevalent transplant patients
returned to dialysis in 2012, a similar percentage to that
seen over the last few years. Amongst patients with native
5
chronic kidney disease, late presentation is associated with
poor outcomes, largely attributable to lack of specialist
management of anaemia, acidosis, hyperphosphataemia
0 and to inadequate advance preparation for dialysis. Trans-
0 200 400 600 800 1,000 1,200 1,400 1,600
Number of patients with data in centre plant recipients on the other hand, are almost always
Fig. 3.9. Funnel plot of percentage of prevalent transplant patients followed up regularly in specialist transplant or renal
with haemoglobin ,100 g/L by centre size on 31/12/2012 clinics and it would be reasonable to expect patients with
60
N = 8,405 Upper 95% Cl
% with SBP/DBP <130/80
50 Lower 95% Cl
40
30
20
10
0
27 Bristol
6 West NI
15 Newry
12 Glouc
0 Swanse
47 York
22 L Rfree
10 Dorset
7 Exeter
3 Leeds
4 Wolve
1 Cardff
28 Bradfd
12 Nottm
7 Ulster
0 Donc
12 Kent
1 Camb
16 Covnt
1 B QEH
2 Chelms
13 Plymth
16 Derby
39 Sthend
38 Antrim
2 Sheff
62 England
39 N Ireland
8 Wales
58 E, W & NI
Centre
Fig. 3.10a. Percentage of prevalent transplant patients with eGFR 530 ml/min/1.73 m2 achieving blood pressure of ,130/80 mmHg
by centre on 31/12/2012
83
60
N = 1,318 Upper 95% Cl
% with SBP/DBP <130/80
50 Lower 95% Cl
40
30
20
10
0
7 Exeter
25 Bristol
2 Cardff
19 L Rfree
15 Nottm
12 Covnt
1 Leeds
8 Plymth
0 Swanse
48 Leic
12 Kent
6 Dorset
14 Bradfd
2 B QEH
11 Camb
31 Norwch
0 Sheff
62 England
37 N Ireland
11 Wales
59 E, W & NI
Centre
Fig. 3.10b. Percentage of prevalent transplant patients with eGFR ,30 ml/min/1.73 m2 achieving blood pressure of ,130/80 mmHg by
centre on 31/12/2012
failing grafts to receive appropriate care and therefore have eGFR slope analysis
many of their modifiable risk factors addressed before
complete graft failure and return to dialysis. Introduction
The gradient of deterioration in eGFR (slope) may
Methods predict patients likely to have early graft failure. The
The transplant cohort consisted of prevalent transplant recipi- eGFR slope and its relationship to specific patient
ents as on 31st December 2012 (N = 25,166) and were classified characteristics are presented here.
according to the KDIGO staging criteria with the suffix of ‘T’ to
represent their transplant status. Patients with missing ethnicity
information were classified as White for the purpose of calculating Methods
eGFR. Prevalent dialysis patients, except those who commenced All UK patients aged 518 years receiving a renal transplant
dialysis in 2012, comprised the comparison dialysis cohort between 1st January 2001 and 31st December 2010, were con-
(N = 21,242) including 2,467 peritoneal dialysis patients. Only sidered for inclusion. A minimum duration of 18 months graft
patients on peritoneal dialysis were considered when examining function was required and three or more creatinine measurements
differences in serum phosphate between transplant recipients from the second year of graft function onwards were used to plot
and dialysis patients. For both the transplant and dialysis cohorts, eGFR slope. If a transplant failed but there were at least three
the analysis used the most recent available value from the last two creatinine measurements between 18 months post-transplant
quarters of the 2012 laboratory data. Scottish centres were and graft failure, the patient was included but no creatinine
excluded from blood pressure, calcium, cholesterol and PTH measurements after the quarter preceding the recorded date of
analyses as corresponding data was not provided. transplant failure were analysed.
Slopes were calculated using linear regression, assuming linear-
ity, and the effect of age, ethnicity, gender, diabetes, donor type,
Results and discussion year of transplant and current transplant status were analysed. P
Table 3.12 shows that 13.7% of the prevalent trans- values were calculated using the Kruskal-Wallis test. eGFR was
plant population (3,442 patients), had moderate to calculated using the CKD-EPI equation and results expressed
advanced renal impairment of eGFR ,30 ml/min/ as ml/min/1.73 m2/year. The CKD-EPI equation was used in
1.73 m2. The table also demonstrates that patients with preference to the MDRD formula as it is thought to have a greater
degree of accuracy at higher levels of eGFR [11].
failing grafts achieved UK Renal Association standards
for some key biochemical and clinical outcome variables
less often than dialysis patients. This substantial group of Results and discussion
patients represents a considerable challenge, as resources The study cohort consisted of 14,783 patients. The
need to be channelled to improve key outcome variables median GFR slope was −0.53 ml/min/1.73 m2/year
and achieve a safe and timely modality switch to another (table 3.13). The gradient was steeper for Black recipients
form of renal replacement therapy. (−1.23 ml/min/1.73 m2/year), in keeping with previously
84
Table 3.12. Analysis by CKD stage for prevalent transplant patients compared with prevalent dialysis patients on 31/12/2012
Stage 1–2T Stage 3T Stage 4T Stage 5T

(560) (30–59) (15–29) (,15) Stage 5D
Patients N 8,713 13,011 3,020 422 21,242

% of patients 34.6 51.7 12.0 1.7
eGFR ml/min/1.73 m2a
mean + SD 77.1 + 15.0 45.5 + 8.3 23.8 + 4.1 11.8 + 2.4
median 73.1 45.6 24.3 12.1
Systolic BP mmHg
mean + SD 133.7 + 17.1 136.1 + 17.9 139.5 + 20.2 143.1 + 22.6 130.9 + 25.1
% 5130 58.7 63.6 69.1 72.6 49.3
Diastolic BP mmHg
mean + SD 78.2 + 10.0 78.0 + 10.4 78.0 + 11.6 79.4 + 11.8 68.4 + 14.6
% 580 46.8 46.9 46.7 49.0 21.6
Cholesterol mmol/L
mean + SD 4.5 + 1.0 4.6 + 1.1 4.7 + 1.2 4.8 + 1.3 4.0 + 1.1
% 54 70.0 72.7 72.7 72.6 46.0
Haemoglobin g/L
mean + SD 136 + 16 128 + 16 116 + 15 106 + 15 112 + 14
% ,100 1.3 3.4 11.6 33.3 16.7
Phosphate mmol/Lb
mean + SD 0.9 + 0.2 1.0 + 0.2 1.1 + 0.3 1.5 + 0.4 1.6 + 0.4
% .1.7 0.2 0.4 2.0 27.6 35.6
Corrected calcium mmol/L
mean + SD 2.4 + 0.2 2.4 + 0.2 2.4 + 0.2 2.4 + 0.2 2.4 + 0.2
% .2.5 27.9 27.4 20.2 20.5 18.4
% ,2.2 5.3 6.2 9.8 15.7 16.2
PTH pmol/L
median 8.5 9.5 16.3 32.1 30.0
% .72 0.4 1.0 3.2 17.9 16.2
a
Prevalent transplant patients with no ethnicity data were classed as White
b
Only PD patients included in stage 5D, N = 2,467
published data suggesting poorer outcomes for this group underway to characterise the patterns of progression
[12, 13]. There was no statistically significant difference more precisely.
in eGFR slope in recipients of deceased donor kidneys The findings in this study differ slightly from previous
(−0.56 ml/min/1.73 m2/year) compared to patients UKRR work exploring eGFR changes in transplant
who received organs from live donors (−0.48 ml/min/ recipients [14]. This identified that male donor to female
1.73 m2/year). Female patients had a steeper slope recipient transplantation, younger recipients, diabetes,
(−0.82 ml/min/1.73 m2/year) than males (−0.36 ml/ white ethnicity, and human leukocyte antigen (HLA) mis-
min/1.73 m2/year), as did diabetic patients (−1.02 ml/ match were associated with faster decline in eGFR. These
min/1.73 m2/year) compared to non-diabetic patients differences may be explained by patients with eGFR
(−0.45 ml/min/1.73 m2/year). The slope was steeper in .60 ml/min/1.73 m2 at one year post-transplantation
younger recipients, possibly reflecting increased risk of being excluded and the more complex multivariable
immunological damage. As might be expected, the model used in the previous work. Udayaraj and colleagues
steepest slope was in patients where the transplant [14] also adjusted for factors such as HLA mismatch and
subsequently failed. This analysis has assumed linearity donor age, which were not available for the patients
of progression of fall in GFR and further work is studied in this chapter.
85
Table 3.13. Differences in median eGFR slope between prevalent transplant patients
Median Lower Upper

Patient characteristic N slope quartile quartile p-value
Age at transplant <40 4,808 −0.93 −3.89 1.14 ,0.0001

40–55 5,795 −0.38 −2.64 1.58
>55 4,180 −0.34 −2.60 1.57
Ethnicity S Asian 1,236 −1.01 −3.78 1.53 ,0.0001

Black 783 −1.23 −4.43 1.02
Other 271 −1.26 −4.61 1.53
White 11,495 −0.47 −2.84 1.41
Gender Male 9,024 −0.36 −2.69 1.56 ,0.0001

Female 5,759 −0.82 −3.56 1.30
Diabetes Non-diabetic 12,531 −0.45 −2.88 1.49 ,0.0001

Diabetic 1,816 −1.02 −3.75 1.17
Donor Deceased 9,855 −0.56 −2.99 1.39 n.s.

Live 4,928 −0.48 −3.10 1.60
Year of transplant 2001 942 −0.54 −2.22 0.68 0.0003

2002 896 −0.58 −2.30 0.64
2003 1,103 −0.54 −2.26 0.89
2004 1,281 −0.36 −2.14 1.20
2005 1,253 −0.14 −2.10 1.50
2006 1,610 −0.50 −2.72 1.29
2007 1,750 −0.57 −2.72 1.50
2008 1,951 −0.53 −3.17 1.81
2009 2,011 −0.90 −4.43 1.95
2010 1,986 −0.86 −5.62 3.24
Status of transplant Died 955 −0.94 −3.95 1.74 ,0.0001
at end of follow-up Failed 1,048 −5.88 −10.75 −2.83
Re-transplanted 65 −4.20 −6.69 −1.62
Functioning 12,715 −0.24 −2.36 1.63
All 14,783 −0.53 −3.02 1.46

n.s. – not significant
Causes of death in transplant recipients Analysis of prevalent patients included all those aged over 18
years and receiving RRT on 31st December 2012.
Introduction
Differences in causes of death between dialysis and Results and discussion
transplant patients may be expected due to selection for Tables 3.14, 3.15 and figure 3.11 show the differences
transplantation and use of immunosuppression. Chapter in the causes of death between prevalent dialysis and
8 includes a more detailed discussion on causes of death transplant patients. Death due to cardiovascular disease
in dialysis patients. was less common in transplanted patients than in dialysis
patients, perhaps reflecting the cardiovascular screening
Methods undertaken during transplant work-up; transplant
The cause of death is sent by renal centres as an ERA-EDTA recipients are a pre-selected lower risk group of patients.
registry code. These have been grouped into the following The leading causes of death amongst transplant patients
categories: cardiac disease, cerebrovascular disease, infection, were infection (23%), other (23%) and malignancy
malignancy, treatment withdrawal, other and uncertain.
Some centres have high data returns to the UKRR regarding (20%). There has been a reduction over time in the
cause of death, whilst others return no information. Provision of proportion of deaths in transplant patients attributed to
this information is not mandatory. cardiovascular or stroke disease (43% in 2003 compared
86
Table 3.14. Cause of death by modality in prevalent RRT patients on 1/1/2012
All modalities Dialysis Transplant
Cause of death N % N % N %
Cardiac disease 647 22 575 22 72 18

Cerebrovascular disease 135 5 118 5 17 4
Infection 532 18 437 17 95 23
Malignancy 292 10 208 8 84 20
Treatment withdrawal 511 17 498 19 13 3
Other 624 21 528 20 96 23
Uncertain 245 8 212 8 33 8
Total 2,986 2,576 410
No cause of death data 1,414 32 1,160 31 254 38
Table 3.15. Cause of death in prevalent transplant patients on 1/1/2012 by age
All age groups ,65 years 565 years
Infection 95 23 48 24 47 22
Malignancy 84 20 42 21 42 20
Other 96 23 43 22 53 25
Uncertain 33 8 16 8 17 8
Total 410 198 212
No cause of death data 254 38 126 39 128 38
25
Dialysis
Transplant
20
15
Percentage
10
0
Cardiac disease
Cerebrovascular
disease
Infection
Malignancy
Treatment
withdrawal
Other
Uncertain
Fig. 3.11. Cause of death by modality for

Cause of death prevalent patients on 1/1/2012
87
to 22% in 2012) with an increase in the proportion the rising death rate secondary to malignancy may
ascribed to infection or malignancy (30% in 2003 com- include the increasing age of transplant recipients and
pared to 43% in 2011). This change has also been the increased intensity of immunosuppressive regimens
reported in other registries, e.g. ANZDATA (http:// leading to complications of over-immunosuppression.
www.anzdata.org.au) and may reflect better management
of cardiovascular risk (although table 3.12 shows BP Conflicts of interest: Dr I MacPhee has received research fund-
management remained suboptimal). Explanations for ing and speaker honoraria from Astellas.
References
1 Ansell D, Tomson CRV: UK Renal Registry 11th Annual Report 8 UK Renal Association Clinical Practice Guidelines Committee: Guide-
(December 2008) Chapter 15 The UK Renal Registry, UKRR database, line 3.7: Target haemoglobin. 2007 RA Guidelines – Complications of
validation and methodology. Nephron Clin Pract 2009;111(suppl 1): CKD, 4th Edition. 2007. http://www.renal.org/Clinical/GuidelinesSection/
c277–c285 ComplicationsofCKD.aspx
2 Pruthi R, Ravanan R, O’Neill J, Roderick P, Pankhurst L, Udayaraj U: 9 UK Renal Association Clinical Practice Guidelines Committee: Guide-
15th annual report: Chapter 9 centre variation in access to renal line: Post-operative Care of the Kidney Transplant Recipient, 5th
transplantation in the UK (2006–2008). Nephron Clin Pract. 2013; Edition. 2011. http://www.renal.org/Clinical/GuidelinesSection/Post-
123(suppl 1):183–93. doi: 10.1159/000353328 operative-Care-Kidney-Transplant-Recipient.aspx
3 Pruthi R, Casula A, MacPhee I: UK Renal Registry 15th annual report: 10 UK Renal Association Clinical Practice Guidelines Committee: Guide-
Chapter 3 demographic and biochemistry profile of kidney transplant line 2.1: Treatment of patients with CKD. 2007 RA Guidelines – CKD,
recipients in the UK in 2011: national and centre-specific analyses. 4th Edition. 2007. http://www.renal.org/Clinical/GuidelinesSection/
Nephron Clin Pract. 2013;123(suppl 1):55–80. doi: 10.1159/000353322 CKD.aspx
4 Bosma RJ, Doorenbos CRC, Stegeman CA, Homan van der Heide JJ, 11 White CA, Akbari A, Doucette S, Fergusson D, Knoll GA: Estimating
Navis G: Predictive Performance of Renal Function Equations in Glomerular Filtration Rate in Kidney Transplantation: Is the New
Renal Transplant Recipients: An analysis of Patient Factors in Bias. Chronic Kidney Disease Epidemiology Collaboration Equation Any
Am J Transplant 2005;5:2183–2203 Better? Clin Chem 2010;56:3:474–477
5 Froissart M, Rossert J, Jacquot C, Paillard M, Houillier P: Predictive 12 Ng FL, Holt DW, Chang RWS, MacPhee IAM: Black renal transplant
Performance of the Modification of Diet in Renal Disease and recipients have poorer long-term graft survival than CYP3A5 expressers
Cockcroft-Gault Equations for Estimating Renal Function. J Am Soc from other ethnic groups. Nephrol Dial Transplant 2010;25:628–634
Nephrol. 2005;16:763–773 13 Isaacs RB, Nock SL, Spencer CE, Connors AF Jr, Wang XQ, Sawyer R,
6 Hariharan, S, McBride MA, Cherikh WS, Tolleris CB, Bresnahan BA, Lobo PI: Racial disparities in renal transplant outcomes. Am J Kidney
Johnson CP: Post-transplant renal function in the first year predicts Dis 1999;34:4:706–712
long-term kidney transplant survival. Kidney Int 2002;62:1:311–318 14 Udayaraj U, Casula A, Ansell D, Dudley CRK, Ravanan R: Chronic
7 UK Renal Association Clinical Practice Guidelines Committee: Anaemia of Kidney Disease in Transplant Recipients – Is It Different From Chronic
CKD, 5th Edition. 2010. http://www.renal.org/clinical/GuidelinesSection/ Native Kidney Disease? Transplantation 2010;90:7:765–770
AnaemiaInCKD.aspx
88
Chapter 4 Demography of Patients Wait-
listed for Renal Transplantation in the UK:
Rishi Pruthia, Rachel Hiltonb, Laura Pankhurstc, Nizam Mamodeb, Alex Hudsonc,
Paul Roderickd, Rommel Ravanane
a
UK Renal Registry, Bristol, UK; bGuys & St Thomas’ Hospital, London, UK; cOrgan Donation and Transplantation Directorate,
NHS Blood and Transplant, Bristol, UK; dSouthampton University, Southampton, UK; eSouthmead Hospital, Bristol, UK
. The proportion of patients listed aged 70 or more

Key Words
was 8% in England, 11% in Wales, 7% in Northern
Blood group . Calculated reaction frequency . Demography .
Ireland and 6% in Scotland, with wide variation
End stage renal disease . Established renal failure . Ethnicity . between centres.
Kidney allocation . Match grade . Prevalence . Renal replace- . Of patients listed, 50% had blood group type O,
ment therapy . Transplantation . Transplant waiting list . Wait whilst blood group AB was the least common
listing times accounting for just 3% of listed patients. The
percentage of patients listed with blood group B
showed inter-centre variation with some centres
Summary having more than a quarter of patients listed with
blood group B.
. There were 6,699 patients registered on the active . Of all patients listed for kidney transplantation, 43%
transplant list for kidney only transplantation at were sensitised (cRF 510), with nearly a quarter
the beginning of 2011. (23%) of all patients listed being highly sensitised
. The UK population prevalence rate for listing for (cRF 585). Patients listed on haemodialysis
kidney transplantation was 107 pmp compared had the largest proportion of highly sensitised
with a dialysis prevalence rate of 424 pmp, with patients with 30% having a cRF 585, whilst only
wide inter-centre variation. 8% of patients listed pre-emptively were highly
. A quarter of the patients listed (25%) were from sensitised.
ethnic minority groups (Black or South Asian). Only . Adult White patients had significantly shorter wait-
10% (61/593) of Black patients were pre-emptively ing times (1098 days, CI: 1071–1125) as compared
listed compared to 16% of Asian and 17% of White to Black patients (1,396 days, CI: 1,301–1,491) or
patients. Asian patients (1411 days, CI: 1,334–1,488).
. The median age of prevalent listed patients on . Median waiting times in highly sensitised patients
dialysis was 53 years, which was significantly lower (2,218 days CI: 1,958–2,478) was more than twice
than the median age of the prevalent haemodialysis that seen in patients who were not sensitised
(HD) patients (66.3 years) and those on peritoneal (1,063 days CI: 1,039–1,087).
dialysis (PD) (61.7 years), p , 0.0001.
89
Introduction These are: waiting time, HLA match and age combined,
donor-recipient age difference, geographical location of
For suitable patients with established renal failure patient relative to donor, HLA-DR homozygosity,
(ERF), renal transplantation is accepted as the optimal HLA-B homozygosity and blood group match (figure 4.1).
modality of renal replacement therapy, conferring both Full details of the allocation policy can be accessed at:
better quality of life and better life expectancy than http://www.odt.nhs.uk/pdf/kidney_allocation_policy.pdf.
dialysis. In the UK, after completing necessary medical Whilst the analysis of these variables at a centre level
and surgical assessment (guided by national guidelines is beyond the scope of a UK Renal Registry (UKRR)
[1]), ‘suitable’ patients are listed for transplantation on report, this report aims to provide clinicians with a better
the UK Transplant Registry at NHSBT (National Health understanding of the ‘make-up’ of the UK Transplant
Service Blood and Transplant). The number of people Registry by:
registered on this database however are far greater than
the number of donor organs available in the UK which (i) Defining the prevalence rates of listing, for
has led to the development and implementation of an individual UK countries and by age group
allocation policy for deceased donor kidneys. This policy (ii) Providing centre level analysis of listing patterns
aims to ensure equity of allocation whilst taking into by age group, ethnicity, gender, calculated HLA
account the importance of achieving a good match antibody reaction frequency (cRF), matchability
between donor and recipient. score, blood group and primary renal disease
(PRD)
Allocation policy (iii) Providing median waiting times by ethnicity,
All kidneys from deceased donors whose death has blood group and calculated HLA antibody
been defined by brain-stem death criteria are allocated reaction frequency (cRF).
through the national allocation scheme managed by Clinicians may find these analyses provide a better
NHSBT. The current scheme was implemented in 2006 understanding of their practice patterns and service
to meet agreed objectives and address issues of inequity needs.
of access to transplantation and utilises an evidence-
based computer algorithm [2, 3]. This is based on a tier
system, with all patients listed for kidney transplantation
being allocated into one of five tiers (figure 4.1). Paedia-
Methods
tric patients are prioritised within Tiers A and B
according to waiting time, whilst within tiers C, D and These analyses relate to the prevalent patients active on the
E patients are prioritised according to a points based transplant waiting list in the UK at the beginning of 2011. The
system (highest score first), based on seven elements. cohort was defined as all patients listed for renal transplantation
Summary of 2006 Scheme

All patients are allocated into one of the following tiers:
Tier A 000 mismatched children (DR homozygous or HSP)
Tier B 000 mismatched children (all others)
Tier C 000 mismatched adults (DR homozygous or HSP)
Tier D 000 mismatched adults & favourably mismatched children
Tier E All other eligible patients
Within tiers A and B: patients are prioritised by waiting time only
Within tiers C to E: patients are prioritised by point score
Waiting time points: 1 point for each day on list
HLA match & age points combined: max 3,500
Age difference points: –0.5*(donor-recipient age diff)2
Location points: 900 same centre, 750 local area
HLA homozygous points: HLA-B 100, HLA-DR 500
Blood group points: –1000 for B patients when donor is O
Fig. 4.1. Summary of national allocation
scheme
90
Chapter 4 Demography of patients wait-listed for renal transplantation
Table 4.1. Prevalence of registration for kidney transplantation and dialysis in the UK on 01/01/2011 (including children ,18 years)
Total estimated population, mid-2010 (millions)∗ 52.2 1.8 5.2 3.0 62.3
Total number registered for transplantation 5,748 178 533 240 6,699
Prevalence rate registration for transplantation (pmp) 110 98 102 79 107
Prevalence rate dialysis (pmp) 424 440 415 436 424
∗
Data from the Office for National Statistics, National Records of Scotland and the Northern Ireland Statistics and Research Agency
pmp = per million population
on the UK Transplant Registry at NHSBT on 1st January 2011. and the resulting HLA antibody reaction frequency (cRF) was
Prevalent listed patients were extracted from the NHSBT database. expressed as a percentage of HLA incompatible donors. These
Patients that had commenced dialysis were matched to the UKRR were then categorised into five groups: ‘0–9%’, ‘10–29%’, ‘30–
database. Patients were allocated to renal centres based on the 84%’, and ‘585%’; ‘0–9%’ was classed as being un-sensitised,
origin of their data returns to the UKRR as opposed to their post- and ‘585%’ was classed as being highly sensitised.
code. Population estimates were obtained from the UK Office of Chi-squared test, Fisher’s exact test and Kruskal Wallis tests
National Statistics (ONS) [4], the National Records of Scotland were used as appropriate to test for significant differences between
(NRS) [5] and the Northern Ireland Statistic and Research Agency groups. The data were analysed using SAS 9.3.
(NISRA) [6]. Crude prevalence rates were calculated per million
population (pmp) and centre level analyses were performed fol-
lowing a merge of data between NHSBT and the UKRR allowing
listed patients to be re-allocated to their main renal centre. Results
The prevalence rate per million population for each centre was
calculated using a derived catchment population. For a full
description of the methodology used to estimate the catchment Prevalent patient numbers listed for transplantation
populations see appendix E: Methodology for Estimating Catch- There were 6,699 patients registered on the active
ment Populations (www.renalreg.com). For Scotland, mid-2010 transplant list for kidney only transplantation at the
populations of Health Boards (HBs) (from the General Register beginning of 2011, giving a UK population prevalence
Office for Scotland) were converted to centre level populations
using an approximate mapping of renal centres to HBs supplied
rate for listing for kidney transplantation of 107 pmp
by the Scottish Renal Registry. Estimates of the catchment compared with a dialysis prevalence rate of 424 pmp
populations in Northern Ireland were supplied by personal com- (table 4.1). There were no significant differences in
munication from Dr D Fogarty. prevalence rates for dialysis in all four of the UK
Throughout this chapter, haemodialysis refers to all modes of countries; however prevalence rates for listing were
HD treatment, including haemodiafiltration (HDF). Several
centres reported significant numbers of patients on HDF, but
significantly lower in Wales at 79 pmp. This may be
other centres did not differentiate this treatment type in their explained by the higher prevalence rate of dialysis for
UKRR returns. Prevalent patients listed for transplantation were patients aged .80 seen in Wales who are less likely to
examined by gender, ethnicity, age group, primary renal disease, be listed. Figure 4.2 shows that Northern Ireland had a
blood group, match grade and calculated HLA antibody reaction higher prevalence rate for listing patients aged 65+
frequency (Report appendix H: Coding (www.renalreg.com).
Analyses were done for the UK as a whole, by UK country, at compared with the other UK countries, mirroring the
centre level and split by treatment modality as appropriate. trend seen in prevalence of dialysis patients in UK
Match grade was calculated for each listed patient by NHSBT countries (chapter 2).
using a pool of 10,000 donors that were blood group identical,
HLA compatible and 000 or favourably (100, 010, 110) HLA mis- Prevalent patients listed for transplantation by RRT
matched. The match count was then converted into a standardised
score, and categorised as: easy to match (1–3), moderate to match modality and centre
(4–7) and difficult to match (8–10). UK and centre analyses were The number of prevalent patients listed for trans-
performed using the three generated categories. plantation in each renal centre and the distribution of
Calculated HLA antibody reaction frequency (cRF) for each their treatment modalities varied widely (table 4.2).
patient was determined by NHS Blood & Transplant-Organ Many factors including geography, local population
Donation and Transplantation Directorate (NHSBT–ODT) from
the unacceptable HLA specificities reported for each patient. The density, age distribution, ethnic composition, prevalence
unacceptable specificities were compared with the HLA types of of diseases predisposing to kidney disease and the social
blood group identical donors from a pool of 10,000 UK donors deprivation index of that population may contribute to
91
Table 4.2. Number of prevalent listed patients by treatment modality and centre on 01/01/2011
Rate of patients listed on dialysis

Total number listed Catchment population
Centre HD PD on dialysis (millions) pmp 95% CI
England
B Heart 94 13 107 0.74 145 (118–172)
B QEHa 208 72 280 1.70 165 (145–184)
Basldn 12 3 15 0.42 36 (18–54)
Bradfd 30 17 47 0.65 72 (51–93)
Brightn 45 21 66 1.30 51 (39–63)
Bristola 83 26 109 1.44 76 (62–90)
Camba 45 6 51 1.16 44 (32–56)
Carlis 13 4 17 0.32 53 (28–78)
Carsh 93 31 124 1.91 65 (53–76)
Chelms 15 13 28 0.51 55 (35–75)
Colchr 14 0 14 0.30 47 (22–71)
Covnta 64 18 82 0.89 92 (72–112)
Derby 36 26 62 0.70 88 (66–110)
Donc 34 9 43 0.41 105 (74–136)
Dorset 59 19 78 0.86 91 (70–111)
Dudley 25 23 48 0.44 109 (78–139)
Exeter 38 22 60 1.09 55 (41–69)
Glouc 23 15 38 0.59 65 (44–85)
Hull 45 17 62 1.02 61 (46–76)
Ipswi 8 10 18 0.40 45 (24–66)
Kent 60 25 85 1.22 69 (55–84)
L Bartsa 134 61 195 1.83 107 (92–122)
L Guysa 100 16 116 1.08 107 (88–127)
L Kings 72 30 102 1.17 87 (70–104)
L Rfreea 166 26 192 1.52 126 (109–144)
L St.Ga 48 13 61 0.80 76 (57–96)
L Westa 330 14 344 2.40 143 (128–159)
Leedsa 111 41 152 1.67 91 (77–105)
Leica 235 71 306 2.44 126 (112–140)
Liv Ain 19 1 20 0.48 41 (23–59)
Liv RIa 82 25 107 1.00 107 (87–127)
M RIa 115 35 150 1.53 98 (82–114)
Middlbr 58 9 67 1.00 67 (51–83)
Newca 41 25 66 1.12 59 (45–73)
Norwch 40 13 53 0.79 67 (49–86)
Nottma 80 48 128 1.09 118 (97–138)
Oxforda 81 43 124 1.69 73 (60–86)
Plymthab 20 13 33 0.47 70 (46–94)
Portsa 143 44 187 2.02 92 (79–106)
Prestn 94 30 124 1.49 83 (68–98)
Redng 64 37 101 0.91 111 (89–133)
Salford 99 49 148 1.49 99 (83–115)
Sheff a 114 19 133 1.37 97 (80–113)
Shrew 26 8 34 0.50 68 (45–91)
Stevng 83 14 97 1.20 81 (65–97)
Sthend 11 8 19 0.32 60 (33–87)
Stoke 54 20 74 0.89 83 (64–102)
Sund 34 11 45 0.62 73 (52–94)
Truro 28 8 36 0.41 87 (59–116)
Wirral 29 10 39 0.57 68 (47–90)
Wolve 36 19 55 0.67 82 (61–104)
York 28 5 33 0.49 67 (44–90)
92
Rate of patients listed on dialysis

Total number listed Catchment population
Centre HD PD on dialysis (millions) pmp 95% CI
Northern Ireland
Antrim 11 3 14 0.30 47 (22–71)
Belfasta 50 12 62 0.55 113 (85–141)
Newry 20 3 23 0.28 82 (49–116)
Ulster 11 0 11 0.30 37 (15–58)
West NI 36 5 41 0.35 117 (81–153)
Scotland
Abrdn 37 11 48 0.60 80 (57–103)
Airdrie 27 3 30 0.56 54 (34–73)
D & Gall 10 2 12 0.15 80 (35–125)
Dundee 16 7 23 0.41 56 (33–79)
Dunfn 19 7 26 0.37 70 (43–97)
Edinba 69 22 91 0.96 95 (75–114)
Glasgwa 186 24 210 1.51 139 (120–158)
Inverns 14 7 21 0.34 62 (35–88)
Klmarnk 24 11 35 0.37 95 (63–126)
Wales
Bangor 14 4 18 0.22 83 (44–121)
Cardff a 64 29 93 1.42 65 (52–79)
Clwyd 11 2 13 0.19 69 (31–106)
Swanse 45 12 57 0.89 64 (48–81)
Wrexm 8 6 14 0.24 58 (28–89)
England 3,619 1,156 4,775
N Ireland 128 23 151
Scotland 402 94 496
Wales 142 53 195
UK 4,291 1,326 5,617
Centres prefixed ‘L’ are London centres
areas by their individual postcodes, as some centres treat patients from across national boundaries
a
Transplant centres
b
The catchment population for Plymouth may be too low, see appendix E
this. Many of these factors are also likely to be the Case mix in prevalent wait-listed patients
cause behind the wide inter-centre variation seen in Gender
listing patients pre-emptively between transplant centres Table 4.4 shows that the gender distribution of patients
with a range of 11 to 125 patients listed across 24 listed for transplantation was similar to that seen in the
transplanting centres (table 4.3). prevalent dialysis population with 59% of patients listed
400
England
350 Scotland
Wales
Prevalent rate pmp
300
NI
250
200
150
100
50
0 Fig. 4.2. Prevalence rates of registration
0–14
15–19
20–24
25–29
30–34
35–39
40–44
45–49
50–54
55–59
60–64
65–69
70–74
75–79
80–84
85+
for kidney transplantation in the UK per

million population by age group and UK
Age group country on 01/01/2011
93
Table 4.3. Number of prevalent listed patients pre-emptively Ethnicity

listed by transplant centre on 01/01/2011 Ethnicity completeness for prevalent listed patients in
Transplant centre Number of pre-emptive listed patients
the UK was 100% at the beginning of 2011 across all UK
countries. Table 4.4 shows that a quarter of the patients
M RI 125 listed (25%) were from ethnic minority groups (Black
B QEH 112 or South Asian) which compared to 12% of the UK
Leic 97
L Guys 71
general population who were designated as belonging to
Bristol 67 an ethnic minority. Whilst there was little difference
L Rfree 61 across modalities, Black patients were seen to have the
L St.G 56 lowest proportion of pre-emptively listed patients, with
L West 56 only 10% (61/593) of listed Black patients being pre-
Leeds 50
Oxford 49 emptively listed compared to 17% (817/4,835) and 16%
Camb 34 (175/1,089) of White and South Asian listed patients
Liv RI 33 respectively. Amongst renal centres there was wide
Nottm 31 variation between centres with respect to the proportion
Newc 30 of patients listed from ethnic minorities (table 4.5,
Sheff 30
Ports 30 figure 4.4), ranging from zero percent (0%) in 12 centres
Cardiff 29 to over 50% in London Barts (72%), London West (70%),
Belfast 27 London St Georges (69%), London Kings (69%), London
Glasgw 19 Royal Free (65%), Birmingham Heartlands (61%) and
L Barts 18
Edin 16
London Guys (53%).
Plymth 15
L GOSH 15
Covnt 11 Age
The median age of prevalent listed patients on dialysis
UK 1,082
at 1st January 2011 was 53 years, which was significantly
lower than the median age of the prevalent HD patients
being male. There was wide inter-centre variation with a (66.3 years) and those on PD (61.7 years), p , 0.0001.
range of 37–91%, and only 11 centres had a preponder- As for those listed pre-emptively the median age was
ance of women listed (figure 4.3). Sub-analysis by slightly lower than those on dialysis at 52 years.
modality did not show any significant gender differences. Table 4.4 shows that 79% of the UK prevalent listed
Table 4.4. Number and percentage of prevalent listed patients and their modalities by gender, ethnicity and age group on 01/01/2011
Modality
HD PD Pre-emptive Total
N % N % N % N %
Gender Male 2,595 60 724 55 614 57 3,933 59
Female 1,696 40 602 45 468 43 2,766 41
Ethnicity White 2,968 69 1,050 79 817 76 4,835 72

Asian 738 17 176 13 175 16 1,089 16
Black 461 11 71 5 61 6 593 9
Other 124 3 29 2 29 3 182 3
Age group 0–17 20 0 24 2 52 5 96 1
18–34 511 12 148 11 111 10 770 11
35–49 1,265 29 380 29 303 28 1,948 29
50–59 1,098 26 356 27 261 24 1,715 26
60–69 1,024 24 334 25 300 28 1,658 25
70+ 373 9 84 6 55 5 512 8
94
100
Female
90 Male
80
70
60
Percentage
50
40
30
20
10
0
Ulster
Sund
Brightn
Bangor
L St.G
Wolve
Shrew
Kent
Inverns
Ports
Derby
Liv RI
Dundee
Donc
Camb
Chelms
Colchr
Wirral
Bradfd
L Kings
B Heart
West NI
Swanse
Stevng
Glasgw
Sheff
Clwyd
Dunfn
Edinb
York
Leeds
L West
Leic
Abrdn
Belfast
Hull
Stoke
Redng
Dorset
Oxford
Carlis
Dudley
Exeter
Middlbr
M RI
Covnt
L Barts
Nottm
Newc
Bristol
Glouc
L Guys
B QEH
Plymth
Prestn
Cardff
Truro
Antrim
D&Gall
Salford
Carsh
L Rfree
Newry
Norwch
Basldn
Klmarnk
Liv Ain
Ipswi
Airdrie
Wrexm
Sthend
Centre
Fig. 4.3. Percentage of prevalent listed patients by gender and centre on 01/01/2011
Table 4.5. Ethnicity of prevalent listed patients by centre on 01/01/2011
Ethnicity
White Asian Black Other
Centre N N % N % N % N %
England
Basldn 15 13 87 1 7 1 7 0 0
B Heart 107 42 39 54 50 10 9 1 1
B QEH 280 151 54 91 33 30 11 8 3
Bradfd 47 25 53 21 45 1 2 0 0
Brightn 66 54 82 4 6 4 6 4 6
Bristol 109 86 79 6 6 8 7 9 8
Camb 51 44 86 4 8 2 4 1 2
Carlis 17 17 100 0 0 0 0 0 0
Carsh 124 74 60 18 15 18 15 14 11
Chelms 28 23 82 1 4 1 4 3 11
Colchr 14 13 93 0 0 0 0 1 7
Covnt 82 52 63 22 27 5 6 3 4
Derby 62 48 77 11 18 3 5 0 0
Donc 43 42 98 1 2 0 0 0 0
Dorset 78 76 97 2 3 0 0 0 0
Dudley 48 38 79 7 15 3 6 0 0
Exeter 60 60 100 0 0 0 0 0 0
Glouc 38 35 92 2 5 1 3 0 0
Hull 62 56 90 2 3 2 3 2 3
Ipswi 18 16 89 0 0 1 6 1 6
Kent 85 84 99 0 0 1 1 0 0
Leeds 152 97 64 38 25 9 6 8 5
Leic 306 209 68 79 26 16 5 2 1
Liv Ain 20 19 95 1 5 0 0 0 0
Liv RI 107 95 89 1 1 5 5 6 6
L Barts 195 55 28 78 40 54 28 8 4
L Guys 116 55 47 4 3 52 45 5 4
L Kings 102 32 31 14 14 51 50 5 5
L Rfree 192 68 35 48 25 70 36 6 3
95
Ethnicity
White Asian Black Other
Centre N N % N % N % N %
L St.G 61 19 31 18 30 18 30 6 10
L West 344 104 30 143 42 77 22 20 6
M RI 150 103 69 33 22 11 7 3 2
Middlbr 67 64 96 2 3 1 1 0 0
Newc 66 61 92 4 6 0 0 1 2
Norwch 53 50 94 2 4 0 0 1 2
Nottm 128 106 83 7 5 12 9 3 2
Oxford 124 89 72 21 17 10 8 4 3
Plymth 33 32 97 0 0 0 0 1 3
Ports 187 161 86 10 5 10 5 6 3
Prestn 124 100 81 21 17 2 2 1 1
Redng 101 60 59 32 32 8 8 1 1
Salford 148 111 75 31 21 4 3 2 1
Sheff 133 119 89 8 6 5 4 1 1
Shrew 34 31 91 1 3 2 6 0 0
Sthend 19 15 79 1 5 2 11 1 5
Stevng 97 69 71 16 16 10 10 2 2
Stoke 74 65 88 6 8 2 3 1 1
Sund 45 43 96 1 2 0 0 1 2
Truro 36 35 97 0 0 0 0 1 3
Wirral 39 33 85 3 8 1 3 2 5
Wolve 55 37 67 16 29 2 4 0 0
York 33 32 97 0 0 0 0 1 3
N Ireland
Antrim 14 14 100 0 0 0 0 0 0
Belfast 62 60 97 1 2 0 0 1 2
Newry 23 22 96 0 0 0 0 1 4
Ulster 11 11 100 0 0 0 0 0 0
West NI 41 41 100 0 0 0 0 0 0
Scotland
Abrdn 48 45 94 2 4 1 2 0 0
Airdrie 30 30 100 0 0 0 0 0 0
D & Gall 12 12 100 0 0 0 0 0 0
Dundee 23 22 96 1 4 0 0 0 0
Dunfn 26 26 100 0 0 0 0 0 0
Edinb 91 88 97 2 2 0 0 1 1
Glasgw 210 193 92 12 6 4 2 1 0
Inverns 21 21 100 0 0 0 0 0 0
Klmarnk 35 33 94 1 3 0 0 1 3
Wales
Bangor 18 18 100 0 0 0 0 0 0
Cardff 93 83 89 7 8 1 1 2 2
Clwyd 13 13 100 0 0 0 0 0 0
Swanse 57 54 95 2 4 1 2 0 0
Wrexm 14 14 100 0 0 0 0 0 0
England 4,775 3,218 67 886 19 525 11 146 3
Northern Ireland 151 148 98 1 1 0 0 2 1
Scotland 496 470 95 18 4 5 1 3 1
Wales 195 182 93 9 5 2 1 2 1
UK 5,617 4,018 72 914 16 532 9 153 3
96
100
90 Other
Black
80 Asian
White
70
60
Percentage
50
40
30
20
10
0
L Barts
L West
L St.G
L Kings
L Rfree
B Heart
L Guys
Bradfd
B QEH
Redng
Carsh
Covnt
Leeds
Wolve
Leic
M RI
Stevng
Oxford
Salford
Derby
Bristol
Sthend
Dudley
Prestn
Brightn
Chelms
Nottm
Wirral
Ports
Camb
Basldn
Stoke
Liv RI
Ipswi
Cardff
Sheff
Hull
Shrew
Glasgw
Glouc
Newc
Colchr
Abrdn
Klmarnk
Norwch
Swanse
Liv Ain
Middlbr
Sund
Dundee
Newry
Edinb
Belfast
Plymth
York
Truro
Dorset
Donc
Kent
Airdrie
Antrim
Bangor
Carlis
Clwyd
D&Gall
Dunfn
Exeter
Inverns
Ulster
West NI
Wrexm
Centre
Fig. 4.4. Ethnicity of prevalent listed patients by centre on 01/01/2011
population was aged between 35–69 years, with only 8% (table 4.6). Analysis by centre (table 4.6) showed wide
of patients aged 70 or above. The proportion of patients variation in the proportion of patients listed aged 70 or
listed aged 70 or more was 8% in England, 11% in above by centre with four centres (Basildon, Colchester,
Wales, 7% in Northern Ireland and 6% in Scotland Ipswich and London Barts) listing no patients, compared
Table 4.6. Number and percentage of prevalent listed patients in each age group by centre on 01/01/2011
Age group (years)
0–17 18–34 35–49 50–59 60–69 70+
Centre N % N % N % N % N % N %
England
Basldn 1 7 5 33 6 40 3 20
B Heart 17 16 27 25 27 25 24 22 12 11
B QEH 4 1 38 14 73 26 90 32 60 21 15 5
Bradfd 11 23 15 32 10 21 8 17 3 6
Brightn 1 2 7 11 16 24 16 24 17 26 9 14
Bristol 3 3 12 11 35 32 23 21 29 27 7 6
Camb 5 10 17 33 16 31 8 16 5 10
Carlis 2 12 5 29 4 24 5 29 1 6
Carsh 12 10 37 30 28 23 37 30 10 8
Chelms 3 11 8 29 9 32 7 25 1 4
Colchr 3 21 2 14 9 64
Covnt 6 7 24 29 27 33 19 23 6 7
Derby 8 13 15 24 15 24 20 32 4 6
Donc 6 14 10 23 10 23 13 30 4 9
Dorset 7 9 17 22 12 15 26 33 16 21
Dudley 5 10 15 31 15 31 11 23 2 4
Exeter 4 7 16 27 15 25 22 37 3 5
Glouc 5 13 10 26 9 24 9 24 5 13
Hull 8 13 21 34 16 26 15 24 2 3
Ipswi 4 22 8 44 5 28 1 6
Kent 8 9 17 20 23 27 31 36 6 7
Leeds 10 7 22 14 47 31 36 24 26 17 11 7
Leic 31 10 71 23 67 22 95 31 42 14
Liv Ain 4 20 5 25 4 20 4 20 3 15
Liv RI 14 13 39 36 30 28 19 18 5 5
L Barts 30 15 63 32 71 36 31 16
97
Age group (years)
0–17 18–34 35–49 50–59 60–69 70+
Centre N % N % N % N % N % N %
L Guys 1 1 14 12 42 36 32 28 19 16 8 7
L Kings 11 11 37 36 30 29 22 22 2 2
L Rfree 20 10 68 35 44 23 40 21 20 10
L St.G 7 11 19 31 9 15 17 28 9 15
L West 2 1 25 7 86 25 102 30 82 24 47 14
M RI 13 9 57 38 43 29 26 17 11 7
Middlbr 10 15 19 28 18 27 14 21 6 9
Newc 2 3 9 14 8 12 19 29 23 35 5 8
Norwch 7 13 13 25 12 23 18 34 3 6
Nottm 14 11 16 13 38 30 26 20 28 22 6 5
Oxford 12 10 36 29 39 31 30 24 7 6
Plymth 6 18 5 15 9 27 12 36 1 3
Ports 18 10 43 23 38 20 54 29 34 18
Prestn 18 15 34 27 38 31 29 23 5 4
Redng 8 8 35 35 28 28 23 23 7 7
Salford 1 1 19 13 42 28 40 27 38 26 8 5
Sheff 18 14 42 32 39 29 27 20 7 5
Shrew 1 3 7 21 13 38 5 15 7 21 1 3
Sthend 1 5 10 53 3 16 4 21 1 5
Stevng 12 12 35 36 20 21 20 21 10 10
Stoke 10 14 21 28 21 28 16 22 6 8
Sund 7 16 19 42 8 18 6 13 5 11
Truro 2 6 6 17 8 22 14 39 6 17
Wirral 6 15 9 23 14 36 7 18 3 8
Wolve 6 11 16 29 14 25 16 29 3 5
York 2 6 12 36 10 30 5 15 4 12
Northern Ireland
Antrim 3 21 1 7 8 57 2 14
Belfast 12 19 19 31 12 19 18 29 1 2
Newry 5 22 6 26 2 9 9 39 1 4
Ulster 2 18 3 27 2 18 3 27 1 9
West NI 5 12 10 24 8 20 13 32 5 12
Scotland
Abrdn 8 17 15 31 14 29 8 17 3 6
Airdrie 4 13 11 37 7 23 6 20 2 7
D & Gall 5 42 3 25 3 25 1 8
Dundee 1 4 9 39 5 22 5 22 3 13
Dunfn 2 8 6 23 10 38 6 23 2 8
Edinb 1 1 9 10 34 37 20 22 21 23 6 7
Glasgw 3 1 26 12 71 34 62 30 38 18 10 5
Inverns 3 14 3 14 6 29 8 38 1 5
Klmarnk 6 17 8 23 6 17 14 40 1 3
Wales
Bangor 2 11 6 33 1 6 6 33 3 17
Cardff 1 1 12 13 31 33 21 23 20 22 8 9
Clwyd 2 15 5 38 2 15 3 23 1 8
Swanse 5 9 12 21 13 23 19 33 8 14
Wrexm 1 7 4 29 4 29 4 29 1 7
England 39 1 554 12 1,384 29 1,255 26 1,146 24 397 8
N Ireland 0 0 24 16 41 27 25 17 51 34 10 7
Scotland 4 1 59 12 162 33 133 27 109 22 29 6
Wales 1 1 22 11 58 30 41 21 52 27 21 11
UK 44 1 659 12 1,645 29 1,454 26 1,358 24 457 8
areas by their individual postcodes, as some centres treat patients from across national boundaries
Blank cells denote no patients listed for that age group within corresponding centre
98
100
90 0–17
18–34
80 35–49
70 50–59
60–69
60 70+
Percentage
50
40
30
20
10
0
Dorset
Ports
Truro
Bangor
Liv Ain
L St.G
Antrim
Swanse
Leic
L West
Brightn
Glouc
Dundee
West NI
York
B Heart
Sund
L Rfree
Stevng
Camb
Donc
Ulster
Middlbr
Cardff
D&Gall
Stoke
Carsh
Clwyd
Dunfn
Wirral
Newc
M RI
Covnt
Leeds
Wrexm
Kent
Redng
L Guys
Airdrie
Edinb
Derby
Bristol
Bradfd
Abrdn
Carlis
Norwch
Oxford
Wolve
Salford
B QEH
Sthend
Sheff
Exeter
Inverns
Glasgw
Nottm
Liv RI
Newry
Dudley
Prestn
Chelms
Hull
Plymth
Shrew
Klmarnk
L Kings
Belfast
Colchr
Basldn
L Barts
Ipswi
Centre
Fig. 4.5. Percentage of listed patients in each age group on 01/01/2011 by centre
to Dorset, Portsmouth, Truro and Bangor, where more patients listed for transplantation on 1st January 2011
than a sixth of their listed patients were aged 70 or at 22% (table 4.7), whilst hypertension only accounted
more (figure 4.5). These differences may be due to for 7% and renovascular disease only 2%. This may be
variation in local listing practices, although could also explained by the fact that younger patients (age ,65
reflect variation in the ethnic make-up of the catchment years) who are more likely to be listed are more likely
population and the social deprivation index of the local to have GN or pyelonephritis and less likely to have
population. renovascular disease or hypertension as the cause of
their renal failure which are more prominent in older age.
Primary renal diagnosis Diabetes accounted for just 10% of listed patients,
Data for primary renal diagnosis (PRD) were not lower than the 15% seen in prevalent patients.
complete for 3% of patients (table 4.7) and there Amongst patients pre-emptively listed the most com-
remained a marked inter-centre difference in complete- mon diagnosis was polycystic kidney disease (PKD),
ness of data returns for PRD to the UKRR. Glomerulone- which is probably a reflection of the fact that these
phritis (GN) was the most common PRD amongst patients are often known to renal services for many
Table 4.7. Number and percentage of prevalent listed patients and their modalities by primary renal diagnosis on 01/01/2011
Modality
Primary renal diagnosis N % N % N % N %

Diabetes 463 11 114 9 41 6 618 10
Glomerulonephritis 926 22 323 24 124 20 1,373 22
Hypertension 311 7 83 6 26 4 420 7
Missing 127 3 40 3 47 7 214 3
Other 709 17 212 16 84 13 1,005 16
Polycystic kidney disease 493 11 189 14 131 21 813 13
Pyelonephritis 489 11 126 10 72 11 687 11
Renovascular 89 2 21 2 8 1 118 2
Uncertain 684 16 218 16 103 16 1,005 16
99
Table 4.8. Number and percentage of prevalent listed patients and their modalities by blood group, match grade and cRF group on
01/01/2011
Modality
N % N % N % N %
Blood group O 2,189 51 639 48 517 48 3,345 50
A 1,290 30 475 36 373 35 2,138 32
B 684 16 181 14 154 14 1,019 15
AB 128 3 31 2 37 3 196 3
Match grade Easy 1,175 27 482 36 422 39 2,079 31
Moderate 1,684 39 601 45 492 46 2,777 41
Difficult 1,432 33 243 18 167 15 1,842 28
cRF group 0 to ,10 2,191 51 833 63 767 71 3,791 57
10 to ,30 172 4 75 6 57 5 304 5
30 to ,85 644 15 229 17 174 16 1,047 16
85 to 100 1,284 30 189 14 83 8 1,556 23
years prior to starting dialysis allowing their timely work pre-emptively were sensitised. This is likely a reflection
up to be pre-emptively listed. of haemodialysis patients having an increased risk of
exposure to sensitising events (e.g. blood transfusions)
Blood group relating to dialysis complications and access procedures
Table 4.8 shows that 50% of patients listed had blood as compared to those listed pre-emptively and also
group type O, whilst blood group AB was the least selective enrichment of the HD population with patients
common accounting for just 3% of listed patients. The with previous failed transplants (due to longer RRT
percentage of patients listed with blood group B (who vintage). Similar reasons are also likely to account for
are known to have the longest median waiting times) the disparity seen in distribution of highly sensitised
showed inter-centre variation (see table 4.9, figure 4.6) patients (cRF 585) which constitute nearly a quarter
with some centres having more than a quarter of patients (23%) of all patients listed for transplantation. Patients
listed with blood group B (London St George’s 31% and listed on haemodialysis had the largest proportion of
London West 26%) whilst four centres had none highly sensitised patients with 30% having a cRF 585,
(Antrim, Basildon, Colchester, Truro). This may partly whilst only 8% of patients listed pre-emptively were
be due to the ethnic make-up of the catchment popu- highly sensitised.
lation with both London West and St George’s having Centre analysis highlighted wide variation in the
a large non-White prevalent dialysis population. proportion of highly sensitised patients listed (table 4.10,
Additionally the actual number of patients listed in figure 4.7) ranging from 50% of patients or more in
Antrim, Basildon, Colchester and Truro were quite Ipswich and Liverpool Aintree, to only 9% in
small, which may explain why all blood groups were Wolverhampton.
not represented in their listed patients. Similar trends were also noted when analysing match
scores by modality (table 4.8) with those listed on haemo-
Calculated HLA antibody reaction frequency (cRF) and dialysis having the greatest proportion of patients that
match grade were difficult to match (33%) as compared to those
Table 4.8 shows that 43% of all patients listed for who were pre-emptively listed (15%). Centre variation
kidney transplantation on the 1st January 2011 were was also seen in the proportion of patients that were
sensitised (cRF 510). Patients on haemodialysis had difficult to match ranging from 48% of patients at
the largest proportion of sensitised patients with 49% London Royal Free, to only 13% at Wolverhampton
having a cRF 510, whilst only 29% of patients listed (table 4.10, figure 4.8).
100
Table 4.9. Number and percentage of prevalent listed patients in each blood group by centre on 01/01/2011
Blood group
O A B AB
Centre N % N % N % N %
England
Basldn 9 60 6 40
B Heart 44 41 33 31 24 22 6 6
B QEH 116 41 94 34 63 23 7 3
Bradfd 26 55 11 23 10 21
Brightn 31 47 24 36 9 14 2 3
Bristol 54 50 37 34 16 15 2 2
Camb 29 57 16 31 4 8 2 4
Carlis 11 65 3 18 3 18
Carsh 73 59 31 25 18 15 2 2
Chelms 13 46 13 46 2 7
Colchr 7 50 7 50
Covnt 36 44 28 34 13 16 5 6
Derby 29 47 20 32 13 21
Donc 22 51 17 40 4 9
Dorset 48 62 27 35 2 3 1 1
Dudley 25 52 15 31 8 17
Exeter 27 45 28 47 4 7 1 2
Glouc 18 47 18 47 2 5
Hull 30 48 23 37 3 5 6 10
Ipswi 11 61 5 28 2 11
Kent 47 55 25 29 12 14 1 1
Leeds 82 54 42 28 23 15 5 3
Leic 148 48 89 29 53 17 16 5
Liv Ain 13 65 5 25 1 5 1 5
Liv RI 55 51 40 37 8 7 4 4
L Barts 90 46 58 30 44 23 3 2
L Guys 58 50 40 34 13 11 5 4
L Kings 48 47 30 29 17 17 7 7
L Rfree 92 48 49 26 46 24 5 3
L St.G 23 38 17 28 19 31 2 3
L West 171 50 71 21 89 26 13 4
M RI 80 53 46 31 21 14 3 2
Middlbr 39 58 23 34 2 3 3 4
Newc 32 48 18 27 15 23 1 2
Norwch 28 53 22 42 3 6
Nottm 80 63 38 30 10 8
Oxford 54 44 47 38 19 15 4 3
Plymth 21 64 10 30 2 6
Ports 80 43 79 42 20 11 8 4
Prestn 67 54 28 23 23 19 6 5
Redng 49 49 36 36 13 13 3 3
Salford 71 48 49 33 25 17 3 2
Sheff 60 45 59 44 10 8 4 3
Shrew 17 50 13 38 3 9 1 3
Sthend 11 58 4 21 4 21
Stevng 50 52 29 30 16 16 2 2
Stoke 34 46 29 39 8 11 3 4
Sund 31 69 10 22 4 9
Truro 17 47 19 53
Wirral 16 41 16 41 7 18
Wolve 31 56 17 31 7 13
York 17 52 9 27 5 15 2 6
101

Blood group
O A B AB
Centre N % N % N % N %
N Ireland
Antrim 6 43 8 57
Belfast 34 55 17 27 10 16 1 2
Newry 15 65 4 17 2 9 2 9
Ulster 5 45 5 45 1 9
West NI 23 56 15 37 3 7
Scotland
Abrdn 29 60 12 25 7 15
Airdrie 17 57 8 27 5 17
D&Gall 6 50 2 17 3 25 1 8
Dundee 11 48 7 30 4 17 1 4
Dunfn 21 81 4 15 1 4
Edinb 51 56 23 25 16 18 1 1
Glasgw 116 55 48 23 39 19 7 3
Inverns 15 71 4 19 2 10
Klmarnk 19 54 10 29 4 11 2 6
Wales
Bangor 10 56 7 39 1 6
Cardff 38 41 38 41 13 14 4 4
Clwyd 6 46 4 31 3 23
Swanse 28 49 20 35 8 14 1 2
Wrexm 7 50 6 43 1 7
England 2,371 50 1,523 32 742 16 139 3
Northern Ireland 83 55 49 32 16 11 3 2
Scotland 285 57 118 24 81 16 12 2
Wales 89 46 75 38 26 13 5 3
UK 2,828 50 1,765 31 865 15 159 3
Blank cells denote no patients listed for that blood group within corresponding centre
100
90
80
70
60
Percentage
50
40
30
AB
20 B
A
10 O
0
Dunfn
Inverns
Sund
Newry
Liv Ain
Carlis
Plymth
Nottm
Dorset
Ipswi
Abrdn
Basldn
Carsh
Middlbr
Sthend
Camb
Airdrie
Wolve
West NI
Edinb
Bangor
Bradfd
Kent
Glasgw
Belfast
Klmarnk
Prestn
Leeds
M RI
Norwch
Dudley
Stevng
York
Liv RI
Donc
Colchr
D&Gall
L Guys
Shrew
Wrexm
L West
Bristol
Swanse
Redng
Newc
Hull
Leic
Salford
L Rfree
Dundee
Glouc
Truro
L Kings
Brightn
Derby
Chelms
Clwyd
L Barts
Stoke
Ulster
Sheff
Exeter
Covnt
Oxford
Antrim
Ports
B QEH
B Heart
Wirral
Cardff
L St.G
Centre
Fig. 4.6. Percentage of listed patients by blood group on 01/01/2011 by centre
102
Table 4.10. Centre analysis of number and percentage of prevalent listed patients by cRF and match score on 01/01/2011
cRF Group Match score
0 to ,10 10 to ,30 30 to ,85 85 to 100 Easy Moderate Difficult
Centre N % N % N % N % N % N % N %
England
Basldn 9 60 4 27 2 13 5 33 7 47 3 20
B Heart 63 59 8 7 13 12 23 22 23 22 54 50 30 28
B QEH 137 49 12 4 46 16 85 30 70 25 119 43 91 33
Bradfd 23 49 2 4 10 21 12 26 11 23 23 49 13 28
Brightn 47 71 2 3 9 14 8 12 24 36 27 41 15 23
Bristol 66 61 3 3 14 13 26 24 29 27 50 46 30 28
Camb 20 39 7 14 7 14 17 33 12 24 20 39 19 37
Carlis 7 41 7 41 3 18 7 41 5 29 5 29
Carsh 60 48 6 5 20 16 38 31 32 26 46 37 46 37
Chelms 14 50 2 7 6 21 6 21 7 25 13 46 8 29
Colchr 8 57 4 29 2 14 4 29 6 43 4 29
Covnt 41 50 2 2 15 18 24 29 28 34 23 28 31 38
Derby 33 53 7 11 10 16 12 19 18 29 29 47 15 24
Donc 27 63 3 7 4 9 9 21 18 42 16 37 9 21
Dorset 40 51 6 8 10 13 22 28 36 46 22 28 20 26
Dudley 25 52 2 4 8 17 13 27 16 33 19 40 13 27
Exeter 30 50 1 2 11 18 18 30 20 33 22 37 18 30
Glouc 23 61 1 3 6 16 8 21 16 42 16 42 6 16
Hull 33 53 2 3 11 18 16 26 20 32 20 32 22 35
Ipswi 5 28 2 11 2 11 9 50 6 33 6 33 6 33
Kent 53 62 3 4 13 15 16 19 31 36 37 44 17 20
Leeds 66 43 6 4 23 15 57 38 42 28 58 38 52 34
Leic 201 66 2 1 66 22 37 12 102 33 136 44 68 22
Liv Ain 7 35 1 5 1 5 11 55 7 35 5 25 8 40
Liv RI 52 49 3 3 22 21 30 28 38 36 37 35 32 30
L Barts 122 63 7 4 30 15 36 18 36 18 105 54 54 28
L Guys 57 49 10 9 14 12 35 30 19 16 53 46 44 38
L Kings 61 60 2 2 19 19 20 20 22 22 49 48 31 30
L Rfree 80 42 12 6 33 17 67 35 25 13 75 39 92 48
L St.G 35 57 6 10 7 11 13 21 10 16 24 39 27 44
L West 264 77 5 1 28 8 47 14 83 24 172 50 89 26
M RI 63 42 5 3 32 21 50 33 33 22 61 41 56 37
Middlbr 30 45 6 9 11 16 20 30 17 25 26 39 24 36
Newc 31 47 4 6 5 8 26 39 23 35 20 30 23 35
Norwch 23 43 6 11 9 17 15 28 22 42 12 23 19 36
Nottm 68 53 5 4 25 20 30 23 40 31 61 48 27 21
Oxford 58 47 8 6 14 11 44 35 32 26 50 40 42 34
Plymth 17 52 1 3 6 18 9 27 15 45 10 30 8 24
Ports 109 58 2 1 29 16 47 25 64 34 64 34 59 32
Prestn 52 42 7 6 29 23 36 29 40 32 41 33 43 35
Redng 53 52 7 7 12 12 29 29 29 29 40 40 32 32
Salford 59 40 3 2 39 26 47 32 42 28 56 38 50 34
Sheff 58 44 9 7 24 18 42 32 41 31 54 41 38 29
Shrew 16 47 7 21 11 32 10 29 13 38 11 32
Sthend 12 63 2 11 5 26 6 32 8 42 5 26
Stevng 54 56 4 4 12 12 27 28 29 30 40 41 28 29
Stoke 36 49 8 11 10 14 20 27 27 36 25 34 22 30
Sund 20 44 2 4 10 22 13 29 15 33 17 38 13 29
Truro 16 44 1 3 6 17 13 36 13 36 10 28 13 36
Wirral 20 51 4 10 4 10 11 28 10 26 17 44 12 31
Wolve 37 67 6 11 7 13 5 9 27 49 21 38 7 13
York 15 45 2 6 3 9 13 39 7 21 16 48 10 30
103
cRF Group Match score
0 to ,10 10 to ,30 30 to ,85 85 to 100 Easy Moderate Difficult
Centre N % N % N % N % N % N % N %
N Ireland
Antrim 9 64 1 7 4 29 6 43 5 36 3 21
Belfast 28 45 11 18 23 37 23 37 18 29 21 34
Newry 11 48 5 22 7 30 4 17 13 57 6 26
Ulster 7 64 2 18 2 18 6 55 3 27 2 18
West NI 25 61 3 7 8 20 5 12 13 32 21 51 7 17
Scotland
Abrdn 33 69 2 4 3 6 10 21 15 31 21 44 12 25
Airdrie 20 67 1 3 4 13 5 17 10 33 12 40 8 27
D&Gall 6 50 1 8 5 42 4 33 4 33 4 33
Dundee 15 65 2 9 6 26 8 35 11 48 4 17
Dunfn 16 62 3 12 7 27 12 46 6 23 8 31
Edinb 46 51 5 5 9 10 31 34 33 36 31 34 27 30
Glasgw 112 53 6 3 27 13 65 31 71 34 86 41 53 25
Inverns 13 62 2 10 6 29 8 38 9 43 4 19
Klmarnk 15 43 2 6 2 6 16 46 9 26 12 34 14 40
Wales
Bangor 9 50 2 11 4 22 3 17 8 44 7 39 3 17
Cardff 53 57 3 3 13 14 24 26 35 38 38 41 20 22
Clwyd 4 31 4 31 5 38 4 31 4 31 5 38
Swanse 41 72 2 4 5 9 9 16 26 46 23 40 8 14
Wrexm 5 36 2 14 2 14 5 36 3 21 5 36 6 43
England 2,556 54 215 5 769 16 1,235 26 1,359 28 1,956 41 1,460 31
Northern Ireland 80 53 5 3 25 17 41 27 52 34 60 40 39 26
Scotland 276 56 18 4 51 10 151 30 170 34 192 39 134 27
Wales 112 57 9 5 28 14 46 24 76 39 77 39 42 22
UK 3,024 54 247 4 873 16 1,473 26 1,657 29 2,285 41 1,675 30
Blank cells denote no patients listed for that category within corresponding centre
100
90
80
70
60
Percentage
50
40
30
85 to 100
20
30 to <85
10 10 to <30
0 to <10
0
Liv Ain
Ipswi
Klmarnk
D&Gall
Newc
York
Clwyd
Leeds
Belfast
Truro
Wrexm
Oxford
L Rfree
Edinb
M RI
Camb
Shrew
Salford
Sheff
Glasgw
Carsh
Newry
B QEH
L Guys
Exeter
Middlbr
Covnt
Prestn
Sund
Redng
Antrim
Inverns
Norwch
Dorset
Wirral
Liv RI
Stevng
Plymth
Dudley
Stoke
Dunfn
Sthend
Dundee
Cardff
Hull
Bradfd
Ports
Bristol
Nottm
B Heart
Chelms
L St.G
Glouc
Donc
Abrdn
L Kings
Derby
Kent
L Barts
Ulster
Carlis
Airdrie
Bangor
Swanse
Colchr
L West
Basldn
West NI
Brightn
Leic
Wolve
Centre
Fig. 4.7. Centre analysis of the percentage of patients listed by calculated reaction frequency group (cRF) on 01/01/2011
104
100
Difficult
90 Moderate
Easy
80
70
60
Percentage
50
40
30
20
10
0
L Rfree
L St.G
Wrexm
Liv Ain
Klmarnk
Clwyd
L Guys
Covnt
M RI
Camb
Carsh
Truro
Norwch
Middlbr
Hull
Newc
Prestn
Leeds
Belfast
Oxford
Salford
D&Gall
Ipswi
B QEH
Shrew
Redng
Ports
Dunfn
Wirral
L Kings
York
Exeter
Liv RI
Stoke
Edinb
Carlis
Sund
Stevng
Sheff
Colchr
Chelms
B Heart
L Barts
Bradfd
Bristol
Dudley
Airdrie
Sthend
Newry
L West
Dorset
Glasgw
Abrdn
Plymth
Derby
Brightn
Leic
Cardff
Antrim
Nottm
Donc
Kent
Basldn
Inverns
Ulster
Dundee
West NI
Bangor
Glouc
Swanse
Wolve
Centre
Fig. 4.8. Centre analysis of the percentage of patients listed by match score on 01/01/2011
Median waiting times need to match donor and recipient blood groups waiting
times are seen to differ across ethnicity, blood groups and
The median waiting times for receiving a deceased level of sensitisation. Adult White patients were seen to
DBD kidney via the national allocation scheme are have significantly shorter waiting times (1,098 days, CI:
shown by ethnicity, blood group and cRF in tables 4.11, 1,071–1,125) as compared to Black patients (1,396 days,
4.12 and 4.13 respectively. These times were calculated CI: 1,301–1,491) or Asian patients (1,411 days, CI:
using patients registered for kidney only transplants in 1,334–1,488) with similar trends seen across paediatric
the UK between 1st January 2006 and 31st December ethnic groups (table 4.11).
2009. The overall median waiting time was 1,160 days Across blood groups, adult patients with blood group O
for an adult (aged 518 years at time of registration) (1,373 days) and B (1,343 days) were seen to have signifi-
and 339 days for a paediatric patient (aged ,18 years cantly longer waiting times than those with blood group A
at time of registration). Due to the allocation algorithm (931 days) or AB (607 days). These differences were not
stratifying patients on level of sensitisation and the seen to be significant across paediatric patients (table 4.12).
Table 4.11. Median waiting time to kidney only transplant in Table 4.12. Median waiting time to kidney only transplant in the
the UK by ethnicity, for patients registered 1st January 2006 to UK by blood group, for patients registered 1st January 2006 to
31st December 2009 31st December 2009
Patients Waiting time (days) Patients Waiting time (days)

registered registered
Ethnicity N Median 95% CI Blood group N Median 95% CI
Adult Adult
White 6,899 1,098 (1,071–1,125) O 4,066 1,373 (1,335–1,411)
South Asian 1,252 1,411 (1,334–1,488) A 3,364 931 (899–963)
Black 667 1,396 (1,301–1,491) B 1,259 1,343 (1,287–1,399)
Other 236 1,209 (1,046–1,372) AB 365 607 (521–693)
Total 9,054 1,160 (1,136–1,184) Total 9,054 1,160 (1,136–1,184)
Paediatric Paediatric
White 248 266 (212–320) O 168 410 (294–526)
South Asian 73 542 (458–626) A 121 269 (161–377)
Black 18 623 (361–885) B 48 241 (128–354)
Other 11 276 (33–519) AB 13 504 (0–1,101)
Total 350 339 (263–415) Total 350 339 (263–415)
105
Table 4.13. Median waiting time to kidney only transplant in the Summary
UK by sensitisation at registration, for patients registered 1st Jan-
uary 2006 to 31st December 2009 Inter-centre variation exists in the number of patients
Patients Waiting time (days) wait-listed (both pre-emptively and after commencing
Level of registered dialysis) and in the proportion listed across different
sensitisation N Median 95% CI ethnic groups, age and blood groups. This may reflect
Adult differences in geography, local population density, age
0–9 6,731 1,063 (1,039–1,087) distribution, ethnic composition, prevalence of diseases
10–29 308 1,148 (1,014–1,282) predisposing to kidney disease and the social deprivation
30–84 1,297 1,475 (1,400–1,550) index of that population as well as individual centre
85+ 718 2,218 (1,958–2,478) practice patterns. Significant unexplained inter-centre
Total 9,054 1,160 (1,136–1,184)
variation was also seen in the proportion of patients listed
Paediatric that were highly sensitised.
0–9 217 299 (212–386)
10–29 15 138 (2–274) Median waiting times are seen to differ significantly
30–84 91 312 (215–409) across blood groups, degree of sensitisation and ethnic
85+ 27 1,241 (836–1,646) groups, with differences in blood group being one
Total 350 339 (263–415) probable factor in explaining the differences in median
waiting times seen amongst the major ethnic groups.

Table 4.13 shows that the level of sensitisation also has
an impact on median waiting times with waiting times in
highly sensitised patients (2,218 days CI: 1,958–2,478)
being more than twice that seen in patients who were References
not sensitised (1,063 days CI: 1,039–1,087), which
1 Renal Association Clinical Practice Guidelines Committee: Assessment of
was highly significant p 4 0.0001. This trend was also the Potential Kidney Transplant Recipient, 5th Edition. 2011. http://www.
seen in paediatric listed patients with highly sensitised renal.org/Clinical/GuidelinesSection/AssessmentforRenalTransplantation.
paediatric patients having a significantly longer median aspx
2 UK Kidney Transplantation: organ allocation policy. http://www.odt.nhs.
waiting time of 1,241 days as compared to 299 days in uk/pdf/kidney_allocation_policy.pdf
paediatric patients who were not sensitised. 3 http://www.odt.nhs.uk/pdf/kidney_allocation_policy.pdf
4 Office for National Statistics. www.statistics.gov.uk
5 National Records of Scotland. http://www.nrscotland.gov.uk/
6 Northern Ireland Statistics and Research Agency. http://www.nisra.gov.
uk/
106
Chapter 5 Comorbidities and Current
Smoking Status amongst Patients
starting Renal Replacement Therapy in
England, Wales and Northern Ireland
from 2011 to 2012
Anirudh Raoa, Retha Steenkampa, Fergus Caskeyabc

a
UK Renal Registry, Bristol, UK; bSouthmead Hospital, Bristol, UK; cSchool of Social and Community Medicine,
University of Bristol, UK
Key Words
. Diabetes mellitus (primary renal disease and co-
Comorbidity . Diabetes . Dialysis . eGFR . Ethnicity . Haemo- morbidity) and ischaemic heart disease were the
globin . Mortality . Renal replacement therapy . Smoking . most common conditions, observed in 35% and
Survival analysis 19% of patients respectively. Ischaemic heart disease,
cerebrovascular disease, chronic obstructive pulmon-
ary disease (COPD), claudication and malignancy
were more prevalent in patients aged .65 years.
Summary . In 2011–2012, 14% of incident RRT patients were
recorded as being smokers at the initiation of dialysis.
. Data on comorbidity at the time of start of renal . There was a higher prevalence of ischaemic heart
replacement therapy (RRT) were submitted for only disease (p , 0.02) and peripheral vascular disease
7,085 (55.9%) of the incident adult (518 years) (p , 0.0003) in patients presenting early to a neph-
RRT patients reported to the UK Renal Registry rologist than amongst those referred late. Malig-
(UKRR) between 2011 and 2012. In 2012, nine nancy (p , 0.0001) was more common in patients
centres provided data on 100% of new patients and who were referred late.
11 centres provided data for less than 5% of new . In the multivariable survival analysis (incident
patients. patients in 2007–2012), malignancy (hazard ratio
. In patients with comorbidity data, more than half (HR) 2.9) and liver disease (HR 2.2) were strongly
had one or more comorbidities (52.9%). In the sub- associated with reduced survival at 1-year in indi-
group of patients aged 565 years, 64% had one or viduals aged ,65 years at start of RRT who survived
more comorbidities. more than 90 days.
107
Introduction Table 5.1. Comorbid conditions listed in the UKRR dataset

. Angina
The number and extent of comorbid illnesses in . Previous myocardial infarction (MI) within 3 months prior
patients initiating dialysis is increasing [1–3]. These to start of RRT
comorbidities are significant predictors of mortality and . Previous MI more than 3 months prior to start of RRT
.
other adverse outcomes [4]. It is therefore imperative to Previous coronary artery bypass graft (CABG) or coronary
account for differences in the comorbid illness burden angioplasty
(in some analyses the above four variables are combined
amongst the groups of dialysis patients being compared. under the term ‘ischaemic heart disease’)
The importance of adjusting for comorbidity when . Cerebrovascular disease
undertaking centre [5–7] and international survival . Diabetes (when not listed as the primary renal disease)
. Chronic obstructive pulmonary disease (COPD)
comparisons [8] is well recognised. This also allows
. Liver disease
for fair comparisons to be made between treatment . Claudication
modalities and costs. . Ischaemic or neuropathic ulcers
However, an important consideration in applying . Non-coronary angioplasty, vascular graft, or aneurysm
case-mix adjustment to analyses is data completeness. If . Amputation for peripheral vascular disease
individuals with comorbidity data differ systematically (in some analyses these four variables are combined under
the term ‘peripheral vascular disease’)
from those without data, entering variables into statistical . Smoking
models can further bias outcome measures and provide . Malignancy
invalid associations [9, 10].
The aim of this work is to describe the completeness of
comorbidity data submitted to the UK Renal Registry comorbid conditions. Comorbidities were grouped into broader
(UKRR), the prevalence of comorbid conditions and categories for some analyses:
current smoking status in incident renal replacement . ‘Ischaemic heart disease’ was defined as the presence of one
therapy (RRT) patients and to examine the association or more of the following conditions: angina, myocardial
between these comorbidities and early mortality. infarction (MI) in the three months prior to starting
RRT, MI more than three months prior to starting RRT or
coronary artery bypass grafting (CABG)/angioplasty.
. ‘Peripheral vascular disease’ was defined as the presence of
one or more of the following conditions: claudication,
Methods ischaemic or neuropathic ulcers, non-coronary angioplasty,
vascular graft, aneurysm or amputation for peripheral
Study population vascular disease.
Incident adult (518 years) RRT patients during 2011 and 2012 . ‘Non-coronary vascular disease’ was defined as the presence
in the centres submitting data to the UKRR were considered. Of of cerebrovascular disease or any of the data items that
these, patients who had data recorded on comorbid conditions comprise ‘peripheral vascular disease’.
were included in statistical analyses. Data on completeness of
comorbidity returns from each centre and overall may differ Specific consideration needs to be made regarding diabetes
from those in previous UKRR reports due to some centres coding. The UKRR also collect data on primary renal disease
retrospectively entering previously missing comorbidity data. (PRD), and have used these data alongside the comorbidity data
to determine which people had diabetes mellitus. The comorbidity
Centre exclusions screen is intended to capture those patients who have diabetes only
The nine centres in Scotland do not provide comorbidity data when it is not the PRD, however some clinicians do enter ‘yes’ in
to the UKRR and are not included in these analyses. There was the comorbidity field in such cases. Prior to statistical analyses,
concern that data extraction in four centres was inaccurate and these fields were examined together to identify these cases and
these centres were excluded from this year’s comorbidity analyses. ensure diabetes is only counted as either the PRD or a comorbid
condition for a certain individual.
Definition of comorbidity and method of data collection
Clinical staff in each centre are responsible for recording in Ethnicity data reporting
yes/no format the presence or absence of 13 comorbid conditions Some centres electronically upload ethnicity coding to their
and information on current tobacco smoking (table 5.1) for each renal IT system from the hospital Patient Administration System
patient at the time of starting RRT on their renal information (PAS) [11]. Ethnicity coding in PAS is based on self-reported
technology (IT) system. Definitions of each of these conditions ethnicity and uses a different system [11] to the remaining centres
are given in appendix B (www.renalreg.com). where coding of ethnicity is performed by clinical staff and
Patients were classified as having complete comorbidity data if recorded directly into the renal IT system (using a variety of
there was at least one entry (yes/no) for any one or more of the coding systems). For all these analyses, data on ethnic origin
108
Chapter 5 Comorbidity in UK RRT patients
were grouped into Whites, South Asians, Blacks and Others. particular comorbidity with those who did not have the co-
Appendix H (www.renalreg.com) details the regrouping of the morbidity (reference group). For both the univariable and multi-
PAS codes into the above ethnic categories. variable Cox models, patients were first stratified by age group
(,65 years and 565 years) to account for the increasing incidence
Statistical methods of certain comorbidities with age, which may otherwise confound
The statistical methods for the three individual sections of this the analyses. The multivariable models used an automatic
chapter are described separately. selection procedure to identify the variables most strongly related
to survival. The potential variables to be included were: age (per 10
1) Patient demographics year increase), smoking status, diabetes (listed as PRD or not listed
The proportion of patients starting RRT with various co- as PRD) and the other 12 comorbidities listed in table 5.1. The
morbidities was examined by age group (18–34, 35–44, 45–54, automatic procedure starts by including only the variable most
55–64, 65–74 and 575 years), primary renal disease, ethnic strongly related to survival. Then, with that variable included, it
origin and first modality of RRT. Chi-squared, Fischer’s exact fits models adding each of the remaining variables in turn (singly)
and Kruskal-Wallis tests were used as appropriate to test for and chooses the variable that adds most to the model (in addition
statistically significant differences between groups. to the contribution made by the first variable included). The
process continues in this way, adding variables that make a further
2) Late presentation (referral) and start of RRT
significant contribution to the model, and removing any whose
Referral time was defined as the number of days between the
contribution becomes non-significant once other variables have
date first seen by a nephrologist and the date of starting RRT.
been added. The final model only includes those variables selected
Referral times of 90 or more days and less than 90 days define
by the process. These automatic methods have been used to give an
early and late presentation, respectively. Data on referral time
indication of the variables most strongly related to survival but
were incomplete and therefore only patients with data on co-
caution is needed in interpreting these because, amongst other
morbidity and referral time from centres with .75% data
factors, when using correlated variables, a slight difference in the
completeness for referral time were included in this analysis.
data (or in the algorithm chosen) could result in different variables
Many UKRR analyses, including those presented here, rely on
being included in the final models. A more robust analysis would
the accuracy of the date of start of RRT. A discussion of the issues
make a considered judgement of which variables should be
around definition of the start date is included in chapter 13 of the
included (rather than an automatic one) and may require
2009 Report [12].
additional interaction terms.
3) Patient survival For each model, a R2 value was calculated using the Royston
The UKRR collected data with a ‘timeline’ entry on all patients and Sauerbrei method [13]. The R2 value is the percentage of
who had started RRT for established renal failure (ERF). Patients the variation in mortality which is explained by the variables
presenting acutely and initially classified as acute renal failure included in the final model.
requiring dialysis who continued to require long-term dialysis, All statistical analyses were performed using SAS version 9.3.
can subsequently be re-classified by clinicians as having had ERF
from the date of their first RRT. The death rate is high in the
first 90 days of commencing RRT with variability observed
between centres. This between centre variation may in part be
due to clinician variation in the classification of patients who Results
present acutely requiring RRT and who may be deemed from
the start to be unlikely to recover renal function. As mortality
rate varies with time on RRT and to remove the influence of Completeness of comorbidity returns from each
between centre variation in the classification of patients, the participating centre
survival analysis was stratified into two time frames. This also The number of patients with data on comorbidity and
enables comparison with results from other national registries. other variables included in the analyses are summarised
The association of comorbid conditions and survival within the
first 90 days was analysed and subsequently the association of
in figure 5.1.
comorbid conditions and 1-year survival in the cohort who Of the 37,285 incident RRT patients starting RRT
survived after 90 days from the start of RRT was also analysed. between 2007–2012, only 20,916 individuals had comor-
For each of the follow up periods, the association of baseline bidity reported to the UKRR. Of 12,677 incident RRT
comorbidity with survival was analysed using univariable and patients in 2011 and 2012, 7,085 individuals (55.8%)
multivariable Cox regression models. For analyses of survival
within the first 90 days, the cohort included patients starting
from 58 centres had data on comorbidity reported. In
RRT between 1st January 2007 and 30th September 2012 to 2012, 6,344 patients commenced RRT in centres in
allow a minimum of three months follow-up from the start of England, Wales and Northern Ireland. Comorbidity
RRT. For the 1-year survival analyses on individual patients who data were provided for 3,479 (54.8%) of those patients
survived at least 90 days after the start of RRT, the cohort included (tables 5.2, 5.3). Table 5.2 highlights the continued wide
data on individuals who started RRT between 1st January 2007
and 30th September 2011.
variation in the completeness of data returns with nine
For each variable, the models were used to estimate the hazard centres providing data on 100% of patients, but 11 centres
ratio of death, comparing the survival experience of patients with a providing data for less than 5% of new patients in 2012.
109
Incident RRT patients

(most recent start) in
England, Wales and
Northern Ireland
(2007–2012) N = 37,285
Comorbidity reported
(2007–2012)
N = 20,916
Incident RRT patients 1 year after 90 days

90 day survival
(2011–2012) with survival
(2007–2012)
comorbidity reported (2007–2012)
N = 20,117
N = 7,085 N = 15,964
Referral date reported

Primary renal
Ethnicity data (centres with >75%
diagnosis
N = 6,979 completeness)
N = 6,985
N = 6,112
Fig. 5.1. Flow chart showing number of patients included in the various analyses
Limiting the comparison to the centres that reported Diabetes mellitus (either listed as the cause of PRD or
in 2007, data completeness for comorbidity has dropped as a comorbidity) was present in 35% of all patients.
slightly. Completeness was 56.4% in 2007 and 55.1% in This is different to the sum of diabetes (not listed as
2012 (table 5.3). When centres with 0% completeness PRD) and diabetes listed as PRD in table 5.5 and reflects
for comorbidity were excluded, the median percentage some patients having both an entry in the comorbidity
of comorbidity returns in 2012 was 81.8%. For centres field for diabetes and having it recorded as their PRD
returning comorbidity data there has been an annual as described in the methods section.
improvement in completeness since 2007 of 10%
(table 5.3), albeit with a small decline in the most recent Prevalence of comorbidity by age group
year. Ischaemic heart disease, cerebrovascular disease,
COPD, claudication, malignancy and non-coronary
Prevalence of multiple comorbidity angioplasty were more prevalent in patients 65 years
Including all incident patients from the years 2011– and over. Liver disease, ischaemic/neuropathic ulcers
2012 (n = 12,677), comorbidity data were available for and prior amputation were more frequently observed in
7,085 (55.8%). More than half of these patients had one younger patients; actual percentages, nevertheless, were
or more comorbidities (52.9%) (table 5.4), but in the quite small (table 5.5). Smoking was also more common
subgroup of patients aged 565 years, this increased to amongst patients under 65 years. With age categorised in
64% for patients with one or more comorbidities 10 year age groups, prevalence of most comorbidities was
recorded (table 5.5). seen to increase markedly from 18–65 years with some
appearing to plateau beyond this (figures 5.2, 5.3). In
Frequency of each comorbid condition those patients aged .75 years there was a slight reduction
Table 5.5 lists the prevalence of specific comorbidities in several reported comorbidities apart from ischaemic
and the percentage of the total number of incident heart disease (angina, MI, CABG), non-coronary angio-
patients for whom data were available for that item. plasty and cerebrovascular accidents.
110
Table 5.2. Percentage completeness of comorbidity data returns on incident patients from individual renal centres 2007–2012
Percentage completeness of comorbidity data
Centre 2007 2008 2009 2010 2011 2012
England
B Heart 34.7 37.1 61.6 75.8 94.7 92.1
B QEH 33.3 32.8 39.6 39.1 50.7 66.7
Basldn 76.9 87.5 88.9 90.6 95.2 84.9
Bradfd 100.0 91.9 93.2 91.0 100.0 98.6
Brightn 36.7 34.5 12.0 6.6 9.2 14.0
Bristol 85.6 77.1 86.6 96.5 89.2 54.7
Camb 2.4 0.0 3.7 0.0 0.8 2.4
Carlis 92.3 96.7 85.7 63.6 67.9 52.6
Carsh 77.0 83.3 77.9 72.7 80.7 53.3
Chelms 54.9 36.1 35.3 26.7 19.2 11.1
Colchr 0.0 0.0 0.0 0.0 0.0
Covnt 0.0 0.9 0.9 3.5 2.7 9.8
Derby 85.5 91.8 93.5 84.8 82.5 91.4
Donc 90.0 26.9 42.5 60.0 62.8 82.5
Dorset 91.9 84.2 90.5 95.8 100.0 91.7
Dudley 0.0 2.2 4.4 0.0 2.3 0.0
Exeter 32.5 29.6 48.3 69.8 88.4 100.0
Glouc 94.9 89.1 67.1 44.3 51.7 37.8
Hull 98.0 92.7 84.9 87.4 97.3 96.9
Ipswi 50.0 34.2 10.5 12.1 0.0 2.3
Kent 76.0 81.3 89.1 100.0 100.0 100.0
L Barts 85.1 84.0 86.1 76.4 74.7 72.6
L Guys 7.2 3.1 3.5 2.8 5.0 1.6
L Kings 100.0 99.3 98.4 100.0 98.6 100.0
L Rfree 11.4 14.5 11.2 19.1 28.7 29.6
L St.G 68.9 70.7 60.0 59.3 51.4 36.3
L West 53.5 45.4 2.8 1.9 2.2 0.9
Leeds 82.3 79.1 90.2 91.3 98.1 98.1
Leic 77.1 76.9 69.7 65.5 49.1 64.3
Liv Ain 47.1 66.7 71.1 0.0 0.0 0.0
Liv RI 72.3 66.7 71.8 2.0 0.0 0.9
M RI 35.9 41.2 64.4 41.6 37.8 26.3
Middlbr 79.0 90.5 91.7 94.1 97.0 90.0
Newc 23.6 34.3 35.1 69.2 84.7 77.9
Norwch 18.0 21.4 23.6 41.9 46.0 37.8
Nottm 93.8 88.7 97.7 96.6 98.3 97.0
Oxford 86.7 82.4 92.5 96.4 98.9 99.4
Plymth 79.0 75.4 84.2 76.8 70.0 55.3
Ports 70.1 61.8 67.1 53.7 41.2 33.5
Prestn 43.9 42.5 50.0 44.4 20.0 9.5
Redng 57.6 66.0 66.0 66.3 78.6 84.9
Salford 10.9 2.2 0.8 0.7 0.0 0.0
Sheff 58.2 51.7 55.0 78.3 77.8 83.5
Shrew 67.2 88.1 85.4 100.0 100.0 100.0
Stevng 73.9 78.4 94.9 98.1 100.0 100.0
Sthend 88.2 80.6 95.7 75.0 86.2 100.0
Stoke 0.0 0.0 0.0 0.0 0.0 0.0
Sund 100.0 97.8 98.4 92.6 100.0 94.4
Truro 95.6 73.2 87.9 84.8 92.1 100.0
Wirral 15.1 15.4 17.5 11.3 6.5 2.0
Wolve 92.7 96.6 100.0 99.1 94.7 88.1
York 86.5 80.6 75.0 97.4 98.1 94.3
111
Percentage completeness of comorbidity data
Centre 2007 2008 2009 2010 2011 2012
N Ireland
Antrim 13.5 31.7 33.3 95.1 73.3 96.2
Belfast 33.3 32.9 46.6 52.8 42.0 50.6
Newry 26.7 90.5 100.0 95.2 100.0 88.9
Ulster 94.4 100.0 100.0 95.0 97.1 100.0
West NI 75.9 71.0 83.8 84.6 86.8 66.7
Wales
Bangor 69.4 67.5 86.7 96.2 95.0 76.2
Cardff 10.9 16.0 23.2 28.5 32.8 21.8
Clwyd 47.6 53.3 60.0 57.1 76.5 81.8
Swanse 96.9 96.0 97.4 88.2 92.4 95.6
Wrexm 66.7 81.0 94.7 100.0 100.0 100.0
England 57.7 56.7 55.6 55.1 56.0 54.1
N Ireland 41.3 51.4 65.5 76.7 74.3 70.4
Wales 46.5 55.8 58.0 59.5 62.1 59.2
UK 56.4 56.5 56.0 56.0 56.9 54.8
Blank cell denotes no data returned for that year
Table 5.3. Summary of completeness of incident patient comorbidity returns (2007–2012)
Years
Combined
2007 2008 2009 2010 2011 2012 years
Renal centres included N 61 62 62 62 62 62
New patients N 6,104 6,180 6,243 6,156 6,333 6,344 37,360
Patients with comorbid data entries N 3,445 3,490 3,493 3,450 3,606 3,479 20,963
Percentage of patients with comorbid data entries 56.4 56.5 56.0 56.0 56.9 54.8 56.1
Percentage restricted to centres reporting since 2007 56.4 57.0 56.1 56.3 57.3 55.1 56.4
Median percentage amongst only centres returning .0% comorbidity 71.2 71.0 71.4 75.8 81.6 81.8 75.8
Prevalence of comorbidity by ethnic origin specifically was much more frequently reported in
Figures 5.4 and 5.5 illustrate the presence of comorbid- South Asian patients (51.1%) than in White individuals
ity by ethnic origin and age group. Figure 5.4 shows the (32.1%) (table 5.6). The reported prevalence of smoking
prevalence of having at least one comorbidity recorded was highest in individuals of White ethnicity (15%).
amongst patients of White origin was nearly 10% higher
compared to incident patients from an ethnic minority. Prevalence of comorbidity amongst patients with
Figure 5.5 shows that this higher trend was observed diabetes mellitus
across most age groups. However, diabetes mellitus Table 5.7 describes comorbidity amongst patients with
and without diabetes (as either primary renal disease or
comorbidity). As would be expected, patients with
Table 5.4. Number of reported comorbidities in patients starting diabetes mellitus had higher prevalence of peripheral vas-
RRT, as a percentage of those for whom comorbidity data were cular disease (20.9% compared to 7.0% in non-diabetic
available 2010–2012 patients). Similarly, there was a statistically significant
Number of higher prevalence of ischaemic heart disease (27.7% and
comorbidities 0 1 2 3 4 5+ 14.4% respectively) and cerebrovascular disease (14.1%
and 8.3% respectively) in the diabetic patients. Similar
Percentage 47.1 27.1 13.1 7.0 3.4 2.3
proportions of patients with diabetes and non-diabetic
112
Table 5.5. Frequency with which each condition was reported in incident RRT patients 2011–2012
Age ,65 years Age 565 years

% overall
Comorbidity N (%) N (%) p value∗ prevalence
Any comorbidity present 1,459 (41.6) 2,291 (64.0) ,0.0001 52.9
Angina 194 (5.6) 536 (15.2) ,0.0001 10.4
MI in past 3 months 42 (1.2) 99 (2.8) ,0.0001 2.0
MI .3 months ago 208 (6.0) 467 (13.2) ,0.0001 9.7
CABG/angioplasty 176 (5.1) 385 (10.9) ,0.0001 8.0
Cerebrovascular disease 231 (6.7) 496 (14.0) ,0.0001 10.4
Diabetes (not listed as PRD) 182 (5.2) 476 (13.5) ,0.0001 9.4
Diabetes listed as PRD 1,008 (29.1) 765 (21.7) ,0.0001 25.4
COPD 155 (4.5) 345 (9.8) ,0.0001 7.1
Liver disease 154 (4.4) 68 (1.9) ,0.0001 3.2
Claudication 142 (4.1) 277 (7.9) ,0.0001 6.0
Ischaemic/neuropathic ulcers 147 (4.2) 123 (3.5) 0.0989 3.9
Angioplasty/vascular graft 77 (2.2) 208 (5.9) ,0.0001 4.1
Amputation 110 (3.2) 86 (2.4) 0.06 2.8
Smoking 516 (15.4) 431 (12.6) 0.0008 14.0
Malignancy 234 (6.8) 659 (18.6) ,0.0001 12.7
∗
p values from Chi-squared tests for differences between age groups in the percentage with the comorbidity
patients were smokers at the time of initiation of RRT uted data to this analysis. Patients referred to a nephrol-
(14.0% and 13.8% respectively). Malignancy was more ogist early had a higher prevalence of peripheral vascular
common in non-diabetic patients (p , 0.0001) and disease, and ischaemic heart disease. There was a much
may reflect ‘competing risks’, with diabetic patients higher proportion of patients with malignancy in the
tending to die at a younger age with cardiovascular late referral group.
disease, rather than developing malignancy in older age.
Age and comorbidity in patients by treatment
Late presentation and comorbidity modality at start of RRT
Table 5.8 shows the presentation time for patients with All comorbidities were more prevalent in patients
various comorbidities. In total, 6,112 individuals contrib- receiving haemodialysis as their initial modality of
35 Smoking Ischaemic ulcers

Total IHD
All PVD Non-coronary angioplasty
Angina 20 CVA Amputee
30 MI >3 months ago Claudication
CABG 18
MI in past 3 months
25 16
14
20
12
15 10
8
10 6
4
5
2
0 0
18–34 35–44 45–54 55–64 65–74 75+ 18–34 35–44 45–54 55–64 65–74 75+
Age group Age group
Fig. 5.2. Prevalence of ischaemic heart disease amongst incident Fig. 5.3. Prevalence of non-coronary vascular disease amongst
patients 2011–2012 by age at start of RRT incident patients 2011–2012 by age at start of RRT
113
60 80
White
South Asian
50 Black
Percentage of patients (95% CI)
Percentage of patients (95% CI)

60
40
30
40
20
10
20
0
White South Asian Black Other
Ethnic origin
0
Fig. 5.4. Presence of comorbid conditions at the start of RRT by 18–34 35–44 45–54 55–64 65–74 75+
ethnic origin amongst patients starting RRT 2011–2012 Age group
Fig. 5.5. Percentage of patients with comorbidity by ethnic origin

in each age group at the start of RRT 2011–2012
Table 5.6. Prevalence of comorbidities amongst incident patients starting RRT 2011–2012 by ethnic group, as percentages of the total
number of patients in that ethnic group for whom comorbidity data were available
White South Asian Black Other
Comorbidity N (%) N (%) N (%) N (%) p value∗

Ischaemic heart disease 1,082 (19.4) 166 (22.9) 42 (9.2) 15 (12.8) ,0.0001
Cerebrovascular disease 562 (10.0) 80 (11.0) 64 (14.0) 6 (5.1) 0.01
Diabetes (not listed as PRD) 538 (9.6) 70 (9.6) 29 (6.3) 8 (6.8) 0.09
Diabetes listed as PRD 1,257 (22.5) 302 (41.5) 147 (32.2) 43 (36.1) ,0.0001
COPD 456 (8.2) 30 (4.1) 8 (1.7) 3 (2.6) ,0.0001
Liver disease 155 (2.8) 23 (3.2) 26 (5.7) 11 (9.4) ,0.0001
Peripheral vascular disease 723 (13.0) 43 (5.9) 28 (6.2) 8 (6.8) ,0.0001
Smoking 817 (15.0) 59 (8.4) 41 (9.4) 13 (11.5) ,0.0001
Malignancy 806 (14.4) 40 (5.5) 31 (6.8) 8 (6.8) ,0.0001
∗
p values from Chi-squared tests for differences between ethnic groups in the percentage with the comorbidities
Table 5.7. Number and percentage of patients with and without diabetes (either as primary diagnosis or comorbidity) who have other
comorbid conditions
Non-diabetic patients Diabetic patients

Comorbidity N (%) N (%) p value∗
Ischaemic heart disease 635 (14.4) 650 (27.7) ,0.0001

Cerebrovascular disease 366 (8.3) 331 (14.1) ,0.0001
COPD 312 (7.1) 170 (7.2) 0.79
Liver disease 135 (3.1) 79 (3.4) 0.50
Peripheral vascular disease 309 (7.0) 489 (20.9) ,0.0001
Smoking 592 (13.8) 320 (14.0) 0.78
Malignancy 635 (14.4) 218 (9.3) ,0.0001
∗
p values from Chi-squared tests for differences in the percentage with the comorbidities between diabetic and non-diabetic patients
114
Table 5.8. Percentage prevalence of specific comorbidities amongst patients presenting late (,90 days) compared with those presenting
early (590 days) (2011–2012 incident patients)
Late referral Early referral

Ischaemic heart disease 176 (16.3) 970 (19.5) 0.02

Cerebrovascular disease 105 (9.7) 500 (10.0) 0.7
Diabetes (not listed as PRD) 110 (10.2) 477 (9.6) 0.5
COPD 97 (9.0) 365 (7.3) 0.1
Liver disease 45 (4.2) 138 (2.8) 0.02
Peripheral vascular disease 95 (8.8) 633 (12.7) 0.0003
Malignancy 217 (19.9) 554 (11.1) ,0.0001
Smoking 158 (15.4) 690 (14.2) 0.3
∗
p values from Chi-squared tests for differences between referral groups in the percentage with the comorbidities
treatment than in those starting on peritoneal dialysis non-transplanted patients (77.4% vs. 43.4% respectively)
(table 5.9). The median age for all patients starting (table 5.10).
dialysis in England, Wales and Northern Ireland in
2011–2012 was 67.3 years (IQR 54.5–76.4) for haemo- Comorbidity and survival within 90 days of
dialysis and 60.5 years (IQR 46.6–71.8) for peritoneal starting RRT
dialysis. In comparison, the median age of patients with In univariable analysis stratified by age, most comor-
comorbidity data starting RRT on HD was 67.6 years bidity was associated with an increased risk of death in
compared with 60.6 years for those starting on PD. For the first 90 days after starting RRT when compared
patients with pre-emptive transplant the median age of with a patient in the same age group without that comor-
patients with comorbidity data was 49.5 years. For most bidity. This was true amongst patients aged ,65 years
of the comorbid conditions, the median age of patients and those aged 565 years, the associations being more
on HD was higher than for patients on PD (table 5.9). profound for those aged ,65 years (data not shown).
As it would be expected a greater percentage of the trans- Results of the multivariable stepwise Cox regression
planted patients had no comorbidities when compared to analyses stratified by age group (,65 and 565) are
Table 5.9. Number (and percentage) of incident patients with comorbid conditions starting PD and HD 2011–2012
HD PD
Comorbidity N (%) Median age N (%) Median age p value∗
Angina 625 (12.3) 72.7 100 (6.5) 70.0 ,0.0001

MI in past 3 months 132 (2.6) 71.3 9 (0.6) 73.2 ,0.0001
MI .3 months ago 556 (10.9) 72.3 114 (7.4) 69.4 ,0.0001
CABG/angioplasty 426 (8.4) 71.1 121 (7.8) 70.7 0.483
Cerebrovascular disease 613 (12.0) 72.2 108 (7.0) 69.6 ,0.0001
Diabetes (not listed as PRD) 558 (10.9) 72.9 87 (5.6) 68.9 ,0.0001
COPD 439 (8.6) 71.5 56 (3.6) 67.6 ,0.0001
Liver disease 187 (3.7) 58.8 29 (1.9) 58.4 0.001
Claudication 354 (6.9) 71.1 62 (4.0) 64.7 ,0.0001
Ischaemic/neuropathic ulcers 223 (4.4) 64.6 37 (2.4) 60.2 0.0004
Angioplasty/vascular graft 241 (4.7) 72.6 41 (2.7) 66.3 0.0004
Amputation 163 (3.2) 64.0 30 (1.9) 61.0 0.01
Smoking 715 (14.5) 65.4 210 (13.8) 55.5 0.47
Malignancy 757 (14.8) 73.4 126 (8.1) 71.3 ,0.0001
∗
p values from Chi-squared tests for differences between modalities in the percentage with the comorbidities
115
Table 5.10. Comorbidity amongst incident patients (2011–2012) who underwent transplantation (by the end of 2012) compared to
those who remained on dialysis or died
Not transplanted Transplanted

Patients with comorbidity data 6,315 770

No comorbidity present 2,739 43.4 596 77.4 ,0.0001
Ischaemic heart disease 1,288 20.7 36 4.7 ,0.0001
Cerebrovascular disease 711 11.4 16 2.1 ,0.0001
Diabetes (not cause of ERF) 638 10.2 20 2.6 ,0.0001
COPD 488 7.8 12 1.6 ,0.0001
Liver disease 206 3.3 16 2.1 0.0775
Peripheral vascular disease 790 12.7 28 3.7 ,0.0001
Smoking 885 14.7 62 8.3 ,0.0001
Malignancy 875 14.0 18 2.4 ,0.0001
shown in tables 5.11 and 5.12. As identified in the uni- Comorbidity and survival 1-year after 90 days of
variable models, the relative magnitude of the hazard commencing RRT
ratios associated with comorbidity in younger patients Age, smoking and four other comorbidities were
tended to be greater than in the older patient group. independently associated with an increased hazard of
Diabetes did not emerge as an independent predictor of death within the first year after 90 days of commencing
death, perhaps explained by its close association with, RRT for patients aged ,65 years and three of these
and mediation in the causal pathway by, cardiovascular (age, malignancy and ischaemic/neuropathic ulcers)
diseases. Some comorbidities may appear not to be were among the nine variables independently associated
associated with an increased risk of death in this analysis with mortality beyond day 90 in patients 565 years
because of the low number of patients in these groups or (tables 5.12, 5.13 and 5.14). Diabetes mellitus was
because of selection within the cohort. For example, independently associated with increased mortality in
individuals with severe comorbid disease, and whose patients ,65 years but not in those aged 565 years.
prognosis on RRT was considered very poor, may not Overall the final six variables in the model exploring
have been started on RRT (for instance, liver disease in death in the year after the first 90 days of starting RRT
those aged 565 years). in patients ,65 years explain 26% of the variation in
The final four variables in the model examining death survival. For patients 565 years, only 10% of the vari-
within the first 90 days of starting RRT in patients aged ation in survival was explained by the nine variables
,65 (table 5.11) explain 31% of the variation in survival. included in the final model.
For patients aged 565, the final eight variables in the
Table 5.12. Multivariable Cox proportional hazards model∗ for
model explain 12% of the variation in survival predictors of death within the first 90 days of starting RRT during
(table 5.12). 01/01/2007–30/09/2012: patients aged 565 years
Table 5.11. Multivariable Cox proportional hazards model∗ for Hazard

predictors of death within the first 90 days of starting RRT during Comorbidity ratio 95% CI p value
01/01/2007–30/09/2012: patients aged ,65 years
MI in past 3 months 2.0 1.4–2.9 0.000
Comorbidity Hazard ratio 95% CI p value Amputation 1.8 1.1–2.9 0.030
Ischaemic/neuropathic ulcers 1.7 1.1–2.6 0.012
Malignancy 4.3 3.0–6.3 ,0.0001 Malignancy 1.6 1.3–1.9 ,0.0001
Ischaemic/ 2.3 1.3–4.1 0.004 Angina 1.5 1.2–1.9 ,0.0001
neuropathic ulcers Age (per 10 years) 1.5 1.3–1.7 ,0.0001
Angina 1.9 1.2–3.0 0.004 COPD 1.4 1.0–1.8 0.022
Age (per 10 years) 1.6 1.3–1.9 ,0.0001 Diabetes of either category 0.8 0.6–1.0 0.018
∗ ∗
This is the result of a stepwise procedure. The variables considered in This is the result of a stepwise procedure. The variables considered
the model were: age (in 10 year units) and the 14 comorbidity in the model were: age (in 10 year units), and the 14 comorbidity
variables except that ‘diabetes (not listed as PRD)’ was replaced by variables except that ‘diabetes (not listed as PRD)’ was replaced by
‘diabetes of either category’ which included ‘diabetes listed as PRD’ ‘diabetes of either category’ which included ‘diabetes listed as PRD’
116
Table 5.13. Multivariable Cox proportional hazards model∗ for account for 31% and 12% of the variation in survival in
predictors of death in the year after the first 90 days of starting patients aged ,65 and 565 years respectively. Whereas
RRT during 01/01/2007–30/09/2012: patients aged ,65 years
for predictors of death in the year after the first 90 days
Hazard of starting RRT the model accounted for 26% and 10%
Comorbidity ratio 95% CI p value of the variation in survival in patients aged ,65 and
565 years respectively. It is noteworthy that even in
Malignancy 2.9 2.2–3.8 ,0.0001
Liver disease 2.2 1.6–3.1 ,0.0001
analyses with 100% comorbidity completeness, the pro-
Ischaemic/neuropathic ulcers 2.1 1.5–3.0 ,0.0001 portion of variance in survival that can be explained by
Diabetes of either category 1.8 1.5–2.2 ,0.0001 these major medical disorders generally remains below
Smoking 1.5 1.2–1.9 0.001 50% when age, primary renal disease, ethnicity and comor-
Age (per 10 years) 1.5 1.3–1.6 ,0.0001 bidities are included in the statistical model.
∗
This is the result of a stepwise procedure. The variables considered A number of studies have demonstrated the associ-
in the model were: age (in 10 year units) and the 14 comorbidity ation of various laboratory and physiological parameters,
variables except that ‘diabetes (not listed as PRD)’ was replaced by
‘diabetes of either category’ which included ‘diabetes listed as PRD’
for example serum albumin, systolic blood pressure, body
mass index, serum phosphate, and parathyroid hormone,
with mortality and other outcomes in dialysis patients
Table 5.14. Multivariable Cox proportional hazards model∗ for [19–22]. Future studies of survival should also consider
predictors of death in the year after the first 90 days of starting other factors such as nutrition, mobility, cognition and
RRT during 01/01/2007–30/09/2012: patients aged 565 years socio-economic status in addition to centre level factors
Hazard at the start of dialysis to better assess the risk factors
Comorbidity ratio 95% CI p value and outcomes for RRT patients. Data completeness
permitting, the UKRR is in a unique position to test the
Malignancy 1.7 1.5–1.9 ,0.0001 association of these parameters and account for the
COPD 1.6 1.4–1.9 ,0.0001 variation in survival.
Age (per 10 years) 1.5 1.4–1.7 ,0.0001
Amputation 1.5 1.0–2.2 0.047 A number of approaches are currently being explored
Ischaemic/neuropathic ulcers 1.5 1.1–2.0 0.024 by the UKRR to improve comorbidity data completeness,
MI in past 3 months 1.4 1.0–1.9 0.027 including collaboration with renal IT suppliers, linkage
Cerebrovascular disease 1.4 1.2–1.7 ,0.0001 with other secondary data sources (e.g. Hospital Episode
Angioplasty/vascular graft 1.3 1.1–1.7 0.016
Angina 1.3 1.1–1.5 0.004
Statistics dataset) and statistical imputation techniques.
Multiple imputation [23] is a statistical technique for
∗
This is the result of a stepwise procedure. The variables considered estimating missing data. In multiple imputation, missing
in the model were: age (in 10 year units) and the 14 comorbidity
variables except that ‘diabetes (not listed as PRD)’ was replaced by comorbidities for an individual patient are estimated
‘diabetes of either category’ which included ‘diabetes listed as PRD’ dependent on available information that is correlated to
the missing comorbidities or explains the reason for the
missing data. In the future the UKRR is likely to use this
Discussion combination of approaches to adjust for case-mix when
exploring the variation in outcome between centres.
Case-mix adjustment is integral to quality reporting
[14, 15], risk adjustment in clinical research [16, 17], Conflicts of interest: none
resource allocation and management of patients with
comorbid conditions in day to day practice [18].
Comorbidity data completeness continues to be a References
cause for concern with overall completeness of comor-
bidity reporting to the UKRR being fairly static. Missing 1 U.S. Renal Data System. USRDS 2007 Annual data report: Atlas of
data may hamper case-mix adjustment but also intro- chronic kidney disease and end-stage renal disease in the United States.
National Institutes of Health, National Institute of Diabetes and Diges-
duces the risk of selection bias, so caution must be used tive and Kidney Diseases, Bethesda, MD, 2007
in interpreting the influence of comorbidity on patient 2 Oreopoulos DG, Dimkovic N. Geriatric nephrology is coming of age.
outcomes. J Am Soc Nephrol 2003;14:1099–1101
3 van Manen JG, Korevaar JC, Dekker FW, Boeschoten EW, Bossuyt PM,
The Multivariable Cox proportional hazards model for et al. How to adjust for comorbidity in survival studies in ESRD patients:
predictors of death within the first 90 days of starting RRT a comparison of different indices. Am J Kidney Dis 2002;40:82–89
117
4 Miskulin D. Characterizing comorbidity in dialysis patients: principles 13 Royston P, Sauerbrei W: A new measure of prognostic separation in
of measurement and applications in risk adjustment and patient care. survival data. Statistics in medicine 2004;23:723–748
Peritoneal Dialysis International 2005;25(4):320–332 14 Lacson Jr E, Teng M, Lazarus J, Lew N, Lowrie E, Owen W. Limitations
5 Hodsman A, Ben-Shlomo Y, Roderick P, Tomson CRV: The ‘centre of the facility-specific standardization mortality ratio for profiling health
effect’ in nephrology: What do differences between nephrology centres care quality in dialysis units. Am J Kidney Dis 2001;37:267–75
tell us about clinical performance in patient management? Nephron 15 Lowrie EG, Lew NL. Death risk in hemodialysis patients:the predictive
Clin Pract 2011;119:1: c10–c17 value of commonly measured variables and an evaluation of death rate
6 Ansell D, Roderick P, Hodsman A, Steenkamp R, Tomson C: Chapter 6: differences between facilities.Am J Kidney Dis 1990;15:458–82
Survival of Incident and Prevalent patients; in Ansell D, Feehally J, 16 Greene T, Beck GJ, Gassman JJ, Gotch FA, Kusek JW, Levey AS, et al.
Feest TG, Tomson C, Williams AJ, Warwick G: UK Renal Registry Design and statistical issues of the Hemodialysis (HEMO) Study.
Report 2007, UK Renal Registry, Bristol, UK, 2007 Control Clin Trials 2000;21:502–25
7 Khan IH, Campbell MK, Cantarovich D, Catto GR, Delcroix C, Edward 17 Liu Y, Coresh J, Eustace J, Longenecker J, Jaar B, Fink N, et al. Associ-
N, Fontenaille C, Fleming LW, Gerlag PG, van Hamersvelt HW, ation between cholesterol level and mortality in dialysis patients.
Henderson IS, Koene RA, Papadimitriou M, Ritz E, Russell IT, Stier E, JAMA 2004;291:451–9
Tsakiris D, MacLeod AM: Survival on renal replacement therapy in 18 Nissenson A, Collins J, Dickmeyer J, Litchfield T, Mattern W,
Europe: Is there a ‘centre effect’? Nephrol Dial Transplant 1996;11: McMahill C, et al. Evaluation of disease-state management of dialysis
300–307 patients. Am J Kidney Dis 2001;37:938–44
8 Marcelli D, Stannard D, Conte F, et al. ESRD patient mortality with 19 Zager P, Nikolic J, Brown R, Campbell M, Hunt W, Peterson D, et al.
adjustment for comorbid conditions in Lombardy (Italy) versus the ‘‘U’’ curve association of blood pressure and mortality in hemodialysis
United States. Kidney Int 1996;50(3):1013–1018 patients. Kidney Int 1998;54:561–9
9 Jager KJ, Zoccali C. Comorbidity data collection by renal registries – a 20 Fleischmann E, Teal N, Dudley J, May W, Bower J, Salahudeen A.
remaining challenge. Nephrol Dial Transplant 2009;24: 2311–2313 Influence of excess weight on mortality and hospital stay in 1346 hemo-
10 Collier T, Steenkamp R, Tomson C, Caskey F, Ansell D, Roderick P, dialysis patients. Kidney Int 1999;55:1560–7
Nitsch D: Patterns and effects if missing comorbidity data for patients 21 Block G, Hulbert-Shearon T, Levin N, Port F. Association of serum
starting renal replacement therapy in England, Wales and Northern phosphorus and calcium x phosphate product with mortality risk in
Ireland. Nephrol Dial Transplant 2011;26:3651–3658 chronic hemodialysis patients: a national study. Am J Kidney Dis
11 Office for National Statistics: The classification of ethnic groups (www. 1998;31:604–17
statistics.gov.uk). 2005 22 Leavey S, Strawderman R, Jones C, Port F, Held P. Simple nutritional
12 Ford DJ, Fogarty DG, Steenkamp R, Tomson CR, Ben-Shlomo Y, indicators as independent predictors of mortality in hemodialysis
Ansell D: UK Renal Registry 12th Annual Report (December 2009): patients. Am J Kidney Dis 1998;31:997–1006
Chapter 13: the UK Renal Registry advanced CKD study: frequency of 23 Sterne JAC, White IR, Carlin JB, et al. Multiple imputation for missing
incorrect reporting of date of start of RRT. Nephron Clin Pract 2010; data in epidemiological and clinical research: potential and pitfalls. BMJ
115(Suppl.1):c271–278 2009;338
118
Chapter 6 Demographics and Outcomes
of Patients from Different Ethnic Groups
on Renal Replacement Therapy in the UK
Udaya Udayaraja, Rishi Pruthib, Anna Casulab, Paul Roderickc

a
Southmead Hospital, Bristol, UK; bUK Renal Registry, Bristol, UK; cSouthampton University, Southampton, UK
Key Words
. The proportion of patients with at least one co-
Access . Demography . Ethnic group . Ethnicity . Hospitalis- morbidity was greater amongst White patients
ation . Incidence . Outcome . Survival . Transplantation compared to South Asian and Black patients
(55.5%, 45.8%, 37.1% respectively, p , 0.0001).
. South Asian and Black patients were referred earlier
Summary to renal centres; started RRT at a lower eGFR and
had a lower Hb at the start of RRT compared to
. Data returns on ethnicity have significantly White patients. The proportion of patients starting
improved over the years to approximately 97% PD and having pre-emptive transplantation was
completeness in 2012. lower amongst both ethnic minorities.
. There was considerable variation in ethnicity break- . The attainment of various laboratory standards was
down between centres; at the London Barts and comparable or better for the ethnic minorities
London West centres only 38% and 45% respect- compared to White patients except for calcium
ively of incident patients were White compared standard attainment (for South Asians) and haemo-
with 99% at some of the South West centres. dialysis dose attainment (for Black patients).
. The age–gender standardized incidence ratio of . Compared to White patients, both ethnic minorities
renal replacement therapy (RRT) was higher (2–3 had similar rates of listing for deceased donor
times) in regions with a high ethnic minority popu- kidney transplantation but had lower rates of
lation compared to those with a low ethnic minority deceased donor transplantation once wait-listed,
population. and lower rates of living kidney donor transplan-
. South Asian and Black patients were significantly tation.
younger than Whites (with median ages of 58.7, . One and five year kidney allograft adjusted survival
54.4, 65.5 years respectively, p , 0.0001); had more was poorer for Black patients but similar for South
diabetes causing established renal failure (ERF) Asians compared to White patients.
(40.2%, 31.0%, 25.0% respectively, p , 0.0001) and . Black and South Asian patients had a better survival
lived in more deprived areas. on dialysis compared to White patients.
119
Introduction their regrouping are described elsewhere [2]. Social deprivation

was measured at super output area level using the adjusted
The ethnic minority population in the UK has Index of Multiple Deprivation (IMD) [3]. The super output
areas were sorted by their IMD score and divided into quintiles,
increased from 9.7 % in the 2001 Census to 13% in the a high IMD quintile indicating a higher level of deprivation.
most recent 2011 Census [1]. Although the ethnic Each patient was allocated an IMD score and a quintile by match-
make-up of the UK is increasingly diverse, this chapter ing their postcode of residence to the 2001 Census lower layer
mainly reports on the characteristics and comparisons super output area.
of patient level outcomes of those on renal replacement eGFR was calculated using the 4 variable MDRD study
equation [4] using the most recent creatinine data that was
therapy (RRT) from the three main ethnic groups: available within 14 days before start of RRT. Similarly, the most
White, South Asian and Black. Patients from other ethnic recent Hb data within 14 days before start of RRT only were
groups were a heterogeneous population and accounted included in the analyses.
for a small proportion of all patients on RRT and there- Chi-squared and Kruskal Wallis tests were used to compare
groups where appropriate.
fore are not discussed in detail in this chapter.
Patient outcome measures
1) Attainment of laboratory standards on dialysis
Only patients who started RRT from 2003 to 2011 and were on
dialysis at the end of their first year of RRT were included in the
Methods analyses. Values from the 4th quarter (or 3rd quarter if 4th quarter
reading not available) in the first year of RRT for each of the fol-
Data on patients (.18 years old) from all 71 UK adult renal lowing variables were used to ascertain achievement of standards
centres starting RRT between 2003 and 2012 and who did not set by the UK Renal Association: Hb 100–120 g/L; phosphate
recover renal function within 90 days and who had data on ethni- (PO4) 1.1–1.7 mmol/L; corrected calcium (Ca) 2.2–2.5 mmol/L;
city were considered. Centres in Scotland were excluded from parathyroid hormone (PTH) 16–72 pmol/L; urea reduction ratio
further analysis due to poor ethnicity data completeness (15.3%). (URR) .65%. Patients who did not have a recorded value for a
The patient cohort used for the various analyses differ slightly, laboratory variable either in the 3rd or 4th quarter in their first
and these variations are described in the individual sections. year of RRT were excluded from the analysis for that laboratory
Details of ethnicity coding used by centres and regrouping of standard. For patients on HD, all the variables were measured
these codes by the UK Renal Registry (UKRR) can be found in pre-dialysis. For the analysis on URR in HD patients, patients
appendix H at www.renalreg.com. on home dialysis and those who received less than three dialysis
sessions per week were not included. Logistic regression analyses
Regional variations in RRT incidence rates by ethnic group were performed to compare attainment of standards between
Data completeness for ethnicity for patients on RRT has ethnic groups adjusting for age (,35, 35–44, 45–54, 55–64, 65–
improved over the years. The proportion of patients with missing 74, 75+ years), gender, primary renal diagnosis, year of start of
ethnicity data has decreased from 13.3 % in 2003 to 3% in 2012. As RRT, dialysis modality at one year, IMD quintile and centre as
missing ethnicity data would bias the estimates of incidence rates fixed effect. Adjustments for comorbidity were not performed
in a population, only patients starting RRT in the years 2010–2012 due to incomplete data.
(�98 % ethnicity data completeness) were included in this analy-
sis. Details of methods used to calculate age–gender standardized 2) Access to kidney transplantation
incidence rates can be found in appendix D at www.renalreg.com. The UKRR has previously reported on access to transplantation
As census data for Northern Ireland population by ethnic group for the various ethnic groups in the UK and the detailed method-
was not available for all age groups above 65 years, Northern ology is described in those reports [5, 6, 7].
Ireland centres were excluded from analyses that required
age–gender standardization. 3) Kidney transplant outcomes
Kidney allograft survival and allograft function amongst those
Demographics and clinical characteristics with a functioning graft at one year and five years were compared
All patients starting RRT between 2003 and 2012 with data on between the ethnic groups. For those who had more than one
ethnicity from centres in England, Wales and Northern Ireland kidney transplant during the study period, only the first transplant
were included for these analyses. The following patient episode was included in the analyses. For the one year graft out-
characteristics at start of RRT were studied: age, gender, social comes analyses, patients who had a kidney only transplant and
deprivation, primary renal diagnosis, comorbidity, estimated who had data on ethnicity, IMD score, primary renal diagnosis
glomerular filtration rate (eGFR), haemoglobin (Hb), time and donor type between 2003 and 2011 were included. For the
between first seen at renal centre and start of RRT (,90 days, five year graft outcome analyses, patients who had a kidney only
90–365 days, .365 days), and treatment modality at start, transplant between 2003 and 2007 were included in the analyses
90 days and at 1 year from start of RRT. to allow five year follow up for all patients. Kaplan-Meier analyses
Details of EDTA coding for primary renal diagnosis used by with and without censoring for death were performed to compare
renal centres can be found in appendix H at www.renalreg.com. unadjusted graft survival between ethnic groups. Cox proportional
Details of comorbid conditions listed in the UKRR dataset and hazards model censoring for death, and death with functioning
120
Chapter 6 Demographics and outcomes of patients by ethnic group
graft as a competing event were performed adjusting for age at RRT in each centre, from 62% in London Barts and
transplant as a continuous variable, gender, primary renal diagno- 55% in London West to 1% in some of the South West
sis, IMD quintile, year of transplant, time on RRT prior to trans-
plantation and type of donor (post brain stem death donor versus
centres, with an overall median of 6%. Ethnic distribution
post circulatory death donor versus live donor). Other donor of the population accounted for some of the regional vari-
details such as donor age, cold ischaemia time, human leucocyte ations in RRT incidence. The age–gender standardized
antigen (HLA) mismatch were not available to be included in incidence ratio of RRT was higher (2–3 times) in regions
the adjusted analyses. with a high ethnic minority population compared to
Graft function amongst those with a functioning graft at one
year and five years was estimated using the CKD-EPI equation
those with a low ethnic minority population (figure 6.1).
[8] from the most recent serum creatinine available in the last However previous work by the UKRR has shown that
quarter of the first and fifth years post kidney transplantation only 31% of this regional variation in RRT incidence in
respectively. the UK could be explained by demographics, health
and access to health service factors [10].
4) Patient survival on dialysis
Patients who started RRT between 2003 and 2012 (excluding
patients in the last quarter of 2012 to allow at least 90 days of Age, gender and social deprivation
RRT) and who had data on ethnicity were considered. Unadjusted 62.2% of patients were male; this degree of male pre-
survival at 90 days from start of RRT, one year from start of RRT ponderance was observed for White and South Asian
and one year after 90 days from start of RRT is reported. For the patients although to a lesser extent with Black patients
one year after 90 day survival analyses, patients who started RRT
from the last quarter of 2011 were not included to allow adequate (58.0%, p , 0.0001) (table 6.2). The proportion of male
follow up. Kaplan Meier analyses and a Cox proportional hazards patients amongst Black patients has however increased
model adjusting for age as a continuous variable, gender, centre as from the 48% observed in the 1997–2003 cohort.
random effect, year of RRT start and IMD quintile were used with Of all patients starting RRT in 2012, 49% were aged
and without censoring for transplantation to compare survival 565 years compared to only 16% aged 565 years in
after 90 days from RRT start between the ethnic groups. Due to
non-proportionality, stratified analyses were performed by the general UK population [1]. The higher incidence of
primary renal disease (diabetic, non-diabetics), age group (,45, RRT amongst older people was more pronounced for
45–64, 565 years) and dialysis modality at day 90 from RRT Black and South Asian patients compared to White
start. Patients were followed up until 31st December 2012 or patients (incidence rate of 1,191, 1,133 and 283 per
death if earlier. million population respectively), (table 6.3, figure 6.2).
The EDTA codes for causes of death were used by centres and
these can be found in appendix H at www.renalreg.com Amongst all patients starting RRT, Black and South
There was no significant difference between those who were Asians were younger compared to White patients, with
included and excluded due to missing ethnicity data except that median ages of 54.4, 58.7 and 65.5 years respectively
the cohort without an ethnic code was older (median age 71.0 (p , 0.0001) (table 6.2).
years vs. 64.2 years, p , 0.0001). Data on residence postcode to calculate IMD score was
5) Hospitalisation episodes not available for 250 (0.5%) patients and this was not differ-
The UKRR has done collaborative work using Hospital Episode ent between the ethnic groups. Black and South Asian
Statistics (HES) data. This cohort included all RRT patients over patients predominantly lived in socially deprived areas.
the age of 18 years who started RRT for ERF in English renal The proportion of patients living in IMD quintile 5 areas
centres between 1 January 2002 and 31 December 2006. Detailed
was greater for Black and South Asians than White patients
methodology for this has been previously published [9]. This
cohort was used to calculate unadjusted hospitalisation rates and (45.7%, 38.7%, 20.9% respectively, p , 0.0001) (table 6.4).
cause of hospitalisation by ethnic group.
Primary renal disease causing ERF
Data on primary renal disease was missing for 2,473
(4.9%) of all patients and this was equally distributed
Results between the ethnic groups.
Diabetes was the leading cause of ERF in all ethnic
Regional variations in incidence of RRT groups. However, the proportion of patients with dia-
Data completeness and ethnic composition by centre betes as cause of ERF was greater amongst South Asian
in the incident population 2003–2012 is shown in and Black patients compared to White patients (40.2%,
table 6.1. Overall completeness was 92%, excluding 31.0%, 20.5% respectively) (table 6.5).
Scottish centres. There was huge variation between Amongst Black and South Asian patients diabetes was
centres in the proportion of non-White patients on more common in those aged 565 years, as compared to
121
Table 6.1. Percentage of incident RRT patients (2003–2012) in different ethnic groups by centre

N with %
Centre White Asian Black Chinese Other data completeness
England
B Heart 69.4 23.9 5.8 0.3 0.6 1,040 99.6
B QEH 68.2 19.6 8.8 0.6 2.9 1,932 99.6
Basldn 88.8 2.3 6.5 1.5 1.0 400 99.0
Bradfd 59.1 38.5 1.9 0.0 0.5 624 95.4
Brightn 93.6 3.6 2.1 0.0 0.8 899 83.4
Bristol 91.7 3.2 3.7 0.6 0.8 1,562 97.4
Camb 95.6 2.0 1.2 0.6 0.6 1,054 90.6
Carlis 98.5 0.8 0.0 0.4 0.4 267 98.5
Carsh 75.7 11.0 8.8 1.3 3.3 1,623 82.0
Chelms 94.0 3.5 1.6 0.6 0.3 319 76.9
Colchr 95.4 2.0 1.3 0.0 1.3 152 82.6
Covnt 82.1 13.1 4.1 0.7 0.0 970 96.6
Derby 85.5 8.9 3.9 1.2 0.5 662 88.9
Donc 96.7 1.4 1.4 0.0 0.5 211 98.6
Dorset 97.8 0.9 0.3 0.2 0.9 667 99.4
Dudley 88.7 7.4 3.0 0.2 0.7 462 97.5
Exeter 99.0 0.3 0.2 0.1 0.3 934 77.0
Glouc 96.0 1.7 1.0 0.2 1.0 577 93.2
Hull 405 39.9
Ipswi 96.1 0.8 3.1 0.0 0.0 382 94.3
Kent 96.1 1.9 0.6 0.3 1.1 787 97.3
L Barts 38.0 28.1 29.5 1.0 3.4 1,908 98.7
L Guys 62.9 5.7 27.9 0.9 2.6 1,183 87.7
L Kings 56.5 10.0 31.1 1.2 1.2 1,167 92.3
L Rfree 51.0 17.5 21.8 0.9 8.9 1,444 95.1
L St.G 57.9 14.3 21.3 1.2 5.3 489 89.1
L West 45.2 32.2 16.8 0.7 5.2 3,038 95.7
Leeds 82.4 12.7 3.8 0.1 1.0 1,388 89.3
Leic 80.0 15.7 3.0 0.3 1.0 2,213 98.3
Liv Ain 95.7 1.4 0.7 1.4 0.7 277 78.3
Liv RI 93.3 1.2 1.6 1.7 2.3 1,018 87.6
M RI 77.0 11.9 8.0 0.9 2.3 890 98.0
Middlbr 95.9 3.7 0.2 0.2 0.0 979 97.1
Newc 93.8 4.2 0.6 0.4 1.1 1,009 99.2
Norwch 95.8 0.8 0.3 2.5 0.6 649 77.6
Nottm 89.2 4.9 4.7 0.0 1.3 1,202 99.9
Oxford 85.6 7.5 4.0 0.6 2.2 1,570 96.6
Plymth 98.0 0.5 0.2 0.7 0.7 608 95.5
Ports 94.4 2.9 1.5 0.0 1.2 1,448 93.7
Prestn 85.9 12.9 0.9 0.0 0.3 1,175 97.1
Redng 74.9 17.9 5.4 0.4 1.5 822 96.3
Salford 83.4 13.5 1.5 0.3 1.3 1,235 97.6
Sheff 91.0 5.0 2.3 0.5 1.2 1,471 93.2
Shrew 95.8 2.7 0.8 0.2 0.4 474 97.3
Stevng 76.6 14.3 6.5 0.6 2.0 1,012 98.5
Sthend 88.7 2.8 3.5 2.1 2.8 283 83.2
Stoke 94.5 2.6 0.4 0.2 2.2 458 84.5
Sund 97.0 2.1 0.7 0.2 0.0 560 97.6
Truro 98.5 0.0 0.8 0.8 0.0 386 80.1
Wirral 97.1 1.2 0.0 1.2 0.6 523 94.6
Wolve 76.1 16.3 7.1 0.5 0.0 854 99.7
York 97.5 0.9 0.7 0.0 0.9 447 97.2
122

N with %
Centre White Asian Black Chinese Other data completeness
N Ireland
Antrim 98.8 1.2 0.0 0.0 0.0 259 95.6
Belfast 98.4 0.8 0.2 0.5 0.3 665 94.9
Newry 99.4 0.0 0.0 0.0 0.6 166 96.0
Ulster 96.5 2.8 0.0 0.7 0.0 144 98.0
West NI 98.7 0.8 0.0 0.4 0.0 237 99.2
Scotland
Abrdn 128 23.0
Airdrie 206 42.0
D & Gall 10 5.9
Dundee 203 34.3
Dunfn 9 2.6
Edinb 26 2.9
Glasgw 51 2.8
Inverns 99.4 0.0 0.0 0.6 0.0 170 65.1
Klmarnk 44 11.2
Wales
Bangor 97.6 0.7 0.7 0.4 0.7 286 89.1
Cardff 94.1 4.5 0.9 0.4 0.2 1,344 74.1
Clwyd 98.3 1.7 0.0 0.0 0.0 117 61.9
Swanse 98.1 1.1 0.7 0.0 0.1 1,161 98.9
Wrexm 98.3 0.9 0.4 0.0 0.4 230 83.0
England 79.5 11.0 7.1 0.6 1.8 48,109 92.5
N Ireland 98.4 1.0 0.1 0.3 0.2 1,471 96.1
Scotland 847 15.3
Wales 96.3 2.6 0.7 0.2 0.2 3,138 83.1
E, W & NI 81.0 10.2 6.5 0.6 1.7 52,718 91.9
UK 81.3 10.1 6.4 0.6 1.6 53,565 85.2
Blank cells denote ,20 patients or ,50% data completeness
Table 6.2. Percentage distribution of gender and age at start of

RRT by ethnic group in the incident population 2003–2012
2.8 Asian Black White
N 5,383 3,442 42,723

2.4
% male 61.3 58.3 62.6
Age at RRT start
2.0 % ,65 64.9 67.6 48.8
Standardised rate ratio
% 65+ 35.1 32.4 51.2

1.6 median 58.7 54.4 65.5
North East
North West
IQR 47.0–69.0 42.6–68.9 51.6–75.2
1.2 Yorkshire and the Humber
East Midlands
West Midlands
East of England
0.8
London
Table 6.3. Incidence rate by ethnic group in under 65 and over
South East Coast 65 year age groups at RRT start (2010–2012)
0.4 South Central
South West
Wales
Incidence rate (pmp) Asian Black White
0.0
0 20 40 60 80 ,65 years 121 160 56
% non-White 565 years 1,133 1,191 283
Overall 179 224 97
Fig. 6.1. Age/gender standardized incidence ratio (2010–2012) by
percentage non-White Pmp = per million population
123
1,500
Asian
Black
White
1,250
1,000
750
500
250
0
20–24
25–29
30–34
35–39
40–44
45–49
50–54
55–59
60–64
65–69
70–74
75–79
80–84
>85
All ages
Fig. 6.2. Age profile of incident RRT
patients (2010–2012), by ethnicity, in
Age group (years) England and Wales
White patients where it was proportionally higher in with White patients whilst hypertensive renal disease was
those aged ,65 years. This may reflect a difference more common amongst Black patients.
between ethnic groups in the underlying type of diabetes
leading to ERF. Adult polycystic kidney disease and Comorbidity
renovascular disease accounted for a lower proportion Patients with missing data on comorbidity
of renal disease in the ethnic minority groups compared (N = 21,896, 42%) were excluded from the analyses.
Data incompleteness was comparable between ethnic
Table 6.4. Percentage distribution of deprivation by ethnic group groups (p = 0.5). The results presented here should be
in incident patients 2003–2012 interpreted with caution due to significant missing data.
There was a wide variation in data completeness on
Deprivation quintile∗ (%) Asian Black White comorbidity between centres. Results from analyses
1 7.8 4.2 17.3 including only centres with data completeness of
2 10.1 6.4 19.9 550% were similar.
3 17.0 12.5 21.0 Overall, the proportion of patients with at least one
4 26.5 31.3 21.0 comorbidity was greater amongst White patients com-
5∗ 38.7 45.7 20.9
pared to South Asians and Black patients (55.5%,
∗
Quintile 5 most deprived 45.8%, 37.1% respectively, p , 0.0001). However
Table 6.5. Percentage distribution of primary renal diagnosis, by ethnic group, in the incident cohort 2003–2012
Asian Black White
Diagnosis ,65 65+ All ages ,65 65+ All ages ,65 65+ All ages
Diabetes 36.1 47.9 40.2 23.8 46.2 31.0 24.1 17.0 20.5
Uncertain aetiology 19.7 23.9 21.2 15.6 14.2 15.2 12.6 24.6 18.7
Other 13.1 7.3 11.0 18.4 10.5 15.9 17.8 16.4 17.1
Glomerulonephritis 14.4 4.8 11.0 15.1 5.6 12.0 17.3 10.0 13.6
Pyelonephritis 6.2 4.2 5.5 2.9 4.2 3.3 8.9 7.9 8.4
Polycystic kidney 3.8 1.6 3.1 4.9 1.4 3.8 12.3 3.6 7.9
Renovascular disease 1.7 4.6 2.7 1.5 4.6 2.5 2.4 13.2 7.9
Hypertension 5.1 5.7 5.3 17.7 13.2 16.3 4.6 7.3 6.0
N with data 3,279 1,765 5,044 2,186 1,036 3,222 19,201 19,965 39,166
% data not available∗ 4.9 5.3 5.1 5.4 6.6 5.8 4.2 5.5 4.9
∗
This includes data not sent and data from centres excluded from analysis because 550% PRD of uncertain aetiology
124
Table 6.6. Percentage of patients with comorbidity at start of years was observed in all ethnic groups. However, late
RRT (2003–2012) by ethnic origin referral was more common amongst White patients
Comorbidity Asian Black White compared to Black and South Asian patients (21.3%,
19.9%, 17.6% respectively, p , 0.0001). There was an age
Coronary heart disease 25.1 9.8 22.0 interaction with referral pattern between ethnic groups
Diabetes (not listed as PRD) 9.5 6.4 8.4
in that late referral was more common amongst White
Diabetes (as PRD or comorbidity) 40.0 30.9 20.0
COPD∗ 3.7 2.2 7.6 patients but only in those aged 565 (table 6.8). When
Malignancy 4.1 6.2 13.5 stratified by diabetic status, there was no difference in
Liver disease 4.0 4.7 2.6 late referral between ethnic groups (table 6.9). This
Smoking 6.6 7.3 15.2 suggests that the early referral patterns observed in Black
Vascular disease 15.2 14.8 20.1
and South Asian patients was probably due to higher
One or more comorbidities present 45.8 37.1 55.5 incidence of diabetes in these groups.
∗
Chronic obstructive pulmonary disease
Treatment modality
diabetes (both as primary renal disease and as a comor- Haemodialysis (HD) was the commonest starting
bidity not causing renal disease) was more common RRT modality in all ethnic groups (73.3%) followed by
amongst the two ethnic minorities. Coronary heart peritoneal dialysis (PD) (21.2%) and pre-emptive trans-
disease was more common in South Asian and White plantation (5.6%). The proportion of patients starting
patients compared to Black patients. Vascular disease, PD was lower amongst Black and South Asians compared
malignancy and smoking were more common amongst to White patients (16.4%, 18.4%, 21.9% respectively,
White patients (table 6.6). These trends were seen in p , 0.0001). Similarly, pre-emptive transplantation
both those aged ,65 and 565, although the magnitude rates were lower amongst South Asian and Black patients
of difference between the ethnic groups for the two age compared to White patients (3.1%, 4.2%, 6.0% respect-
groups varied depending on the comorbidity (table 6.7). ively, p , 0.0001). There was no difference (p = 0.6) in
the type of kidney donor (post circulatory death donor,
Late presentation post brain stem death donor, live donor) between the
19,817 (38.4%) patients were excluded from the analysis ethnic groups amongst those who had a pre-emptive
due to not having data on the date first seen by a nephrol- kidney transplant. Compared to those referred late
ogist. Overall, late referral has decreased over the years (,90 days of RRT start), patients who were referred
with the majority (64%) of patients being referred at earlier were more likely to start on PD (25.0% vs.
least a year or more prior to start of RRT compared to 11.2%, p , 0.0001) and had more pre-emptively trans-
only 46% in 1997–2003, although one should interpret plantation (6.9% vs. 1.0%, p , 0.0001). This trend was
this with caution due to potential bias introduced by the seen in all ethnic groups except in Black patients where
significant proportion of missing data. This overall the pre-emptive transplantation rate was similar amongst
decrease in late referral compared to the previous cohort those referred early and late (data not shown).
Table 6.7. Percentage of patients with comorbidity at start of RRT (2003–2012) by age and ethnic origin
Asian Black White
Comorbidity ,65 65+ ,65 65+ ,65 65+

Coronary heart disease 19.1 37.1 5.7 18.9 13.4 30.2
Diabetes (not listed as PRD) 7.4 13.9 5.1 9.1 5.0 11.7
Diabetes (as PRD or comorbidity) 35.9 47.6 23.7 46.2 23.7 16.6
COPD∗ 3.0 5.1 1.4 4.0 4.8 10.2
Malignancy 2.4 7.6 3.9 11.4 7.5 19.2
Liver disease 3.9 4.1 5.3 3.2 3.4 1.7
Smoking 7.3 5.3 8.2 5.2 18.1 12.5
Vascular disease 11.6 22.7 10.1 25.1 14.7 25.2
One or more comorbidities present 37.4 62.8 30.0 52.4 44.4 66.2
∗
Chronic obstructive pulmonary disease
125
Table 6.8. Presentation in incident patients 2003–2012, by ethnicity and age
Asian Black White
Presentation ,65 65+ ,65 65+ ,65 65+

N 1,822 935 955 471 13,425 14,123
% ,90 days 18.8 15.7 22.1 14.7 19.9 22.6
% 90–365 days 16.7 12.8 16.7 11.9 15.9 14.3
% .365 days 64.5 71.4 61.3 73.5 64.2 63.1
eGFR at start of RRT become more anaemic earlier in their CKD course
The eGFR at start of RRT has increased over the years compared to non-diabetics [11], a stratified analysis by
indicating patients are being started on RRT earlier in the diabetes status was performed but the results were similar
course of their chronic kidney disease stage (CKD) (data not shown).
(figure 6.3). This trend was observed in all ethnic groups.
White patients started at a higher eGFR compared to Patient outcome measures
Black and South Asian patients. The median eGFR at Attainment of laboratory standards on dialysis
RRT start in 2003–2012 for White patients was The proportion of patients in each ethnic group
8.5 ml/min/1.73 m2 compared to 8.0 ml/min/1.73 m2 who achieved the Renal Association standard varied
for Black and 7.8 ml/min/1.73 m2 for South Asian depending on the outcome measure studied. Table 6.10
patients (p , 0.0001). As missing data accounted for shows the multivariate logistic regression model with
49% of this cohort, caution should be taken in interpret- and without adjustments for various confounding factors.
ing this result. Compared to White patients, South Asian patients had
Preliminary work undertaken by the UKRR on a similar attainment of the Hb and PTH standards; better
cohort of CKD stage 5 patients in the UK has shown attainment for the URR and phosphate standards; and
that Black and South Asian patients had a much more lower attainment of the calcium standard. Black patients
rapid decline in their eGFR in the year preceding RRT had similar attainment of the Hb, calcium and PTH
compared to White patients despite adjustments for standards; lower attainment of the URR standard but
age, gender and primary renal disease (unpublished better attainment of the phosphate standard.
data).
Access to kidney transplantation
Haemoglobin prior to start of RRT The UKRR in collaboration with the Organ Donation
Due to missing data, 25,134 (49%) patients were Transplantation Directorate of NHS Blood and Trans-
excluded. White patients had higher mean Hb (102.3 g/L) plant (ODT) previously reported on access to kidney
prior to start of RRT compared to South Asian patients transplantation for the ethnic minority patients starting
(99.9 g/L, p , 0.0001) and Black patients (95.7 g/L, RRT in the years 1997–2004 [5, 6, 7]. Compared to the
p , 0.0001). Data on erythropoietin use prior to start of White patients, South Asian (hazard ratio (HR) 1.10,
RRT was not available to further explore the reasons for 95% CI 0.97–1.24) and Black patients (HR 0.95, 95%
the differences in Hb at start of RRT between ethnic CI. 0.79–1.14) had similar rates of being listed for a
groups. As it is well known that diabetic patients (more kidney transplant once adjusted for various patient
common amongst Black and South Asian patients) characteristics including social deprivation and centre
Table 6.9. Presentation in incident patients 2003–2012, by ethnicity, stratified by diabetes
Asian Black White
Presentation Diabetic Non-diabetic Diabetic Non-diabetic Diabetic Non-diabetic

N 1,024 1,584 403 937 5,335 20,394
% ,90 days 12.1 21.3 11.9 21.7 10.3 23.4
% 90–365 days 18.4 13.1 18.1 13.5 17.7 14.2
% .365 days 69.5 65.7 70.0 64.9 72.0 62.5
126
9.5
9.0
8.5
8.0
Asian
7.5 Black
White
Linear (Asian)
7.0 Linear (Black)
Linear (White)
6.5
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Fig. 6.3. Median eGFR at start of RRT by
Year year of start and ethnic group
effects. However, once on the waiting list, South Asian Graft failure (excluding deaths with functioning grafts)
(HR 0.74, 95% CI 0.65–0.85) and Black patients (HR in the first year following kidney transplantation was
0.66, 95% CI 0.49–0.87) had lower rates of deceased greater for Black patients (7.5%) and South Asian
donor kidney transplantation. Similarly the likelihood (6.1%) patients compared to White patients (4.2%)
of living donor kidney transplantation in the fully (p = 0.0001). However, in the multivariate Cox regression
adjusted analyses was lower for South Asian patients analyses censoring for death, South Asian patients had a
(odds ratio (OR) 0.66, 95% CI 0.45–0.96) and Black similar graft survival but Black patients a lower graft
patients (OR 0.40, 95% CI 0.21–0.73) compared to survival compared to White patients (table 6.11). Results
White patients. A more recent analysis of patients were similar when analyses were repeated with death
starting RRT between 2006 and 2008 confirmed no ethnic as a competing risk event. Amongst those who had a
disparities in access to waiting list but the lower rates of functioning graft at one year post kidney transplantation
deceased donor transplantation once waitlisted, and for (N = 8,479), the median eGFR was better for Black
live donor transplantation persisted for the ethnic (57.2 ml/min/1.73 m2, interquartile range (IQR) 42.9–
minorities [12]. 71.5) and South Asian (58.5 ml/min/1.73 m2, IQR 45.2,
73.3) patients compared to White (51.5 ml/min/1.73 m2,
Kidney transplant outcomes IQR 40.0, 64.1, p , 0.0001) patients.
One year graft outcomes
The analyses included 9,091 kidney only transplants. Of Five year graft outcomes
these kidney only transplants, 237 (2.5%) were excluded For the analyses, 2,912 kidney only transplants were
either due to lack of matching between the UKRR and included. Of these kidney only transplants, 126 (4.1%)
ODT databases (N = 159) or lost to follow up (N = 78). were excluded either due to lack of matching between
Table 6.10. Odds ratio (OR) (95% confidence interval) of attainment of RA standards at one year after starting RRT in dialysis patients,
in Asian and Black patients compared to White patients
White Asian Black
OR N Unadjusted Adjusted N Unadjusted Adjusted N

Haemoglobin 1 23,982 1.01 (0.94–1.09) 1.03 (0.94–1.11) 3,255 0.98 (0.90–1.07) 0.94 (0.85–1.04) 2,135
Calcium 1 21,375 0.85 (0.79–0.92) 0.89 (0.81–0.97) 3,018 0.86 (0.78–0.94) 0.95 (0.85–1.06) 2,023
Phosphate 1 23,559 1.03 (0.96–1.11) 1.15 (1.06–1.25) 3,221 1.05 (0.96–1.15) 1.21 (1.09–1.34) 2,114
PTH 1 20,553 1.12 (1.04–1.21) 1.05 (0.96–1.15) 2,685 1.05 (0.95–1.15) 0.97 (0.87–1.09) 1,737
URR 1 14,393 1.62 (1.42–1.84) 1.73 (1.49–2.00) 1,961 0.81 (0.70–0.93) 0.77 (0.65–0.91) 1,011
127
Table 6.11. Cox-regression analysis of one year graft failure by ethnicity of kidney-only transplants between 2003 and 2011
Unadjusted Cox-regression Adjusted Cox-regression
HR (95% CI) p-value HR (95% CI) p-value

Asian 1.4 (1.1–1.9) 0.01 1.3 (0.9–1.9) 0.1
Black 1.8 (1.3–2.5) 0.0004 1.7 (1.2–2.3) 0.0007
White 1 (reference) 1 (reference)
the UKRR and ODT databases (N = 101) or lost to follow Deaths due to cerebrovascular disease, ischemic heart
up (N = 25). Graft failure (excluding deaths with func- disease and infection were more common for South
tioning grafts) at five years following kidney transplan- Asian and Black patients, whilst deaths due to malig-
tation was greater for Black patients (17.2%) compared nancy, withdrawal from RRT and other causes were
to South Asian (9.2%) and White (9.8%) (p = 0.03) more common in White patients. These trends were
patients. In the multivariate Cox regression analyses cen- seen both in those aged ,65 and 565 years (table 6.14).
soring for death, White and South Asian patients had a
similar graft survival but Black patients had lower graft Hospitalisation episodes
survival (table 6.12). Results were similar when analyses The number of admissions and the number of
were repeated with death as a competing risk event. admitted days per year was greater for HD patients com-
Amongst those who had a functioning graft at five years pared to PD patients. Amongst HD patients, the number
post kidney transplantation (N = 2,482), the median of admissions and the number of admitted days per year
eGFR was better for Black (60.4 ml/min/1.73 m2, IQR was greater for White patients compared to South Asian
42.8–75.7) and South Asian (58.1 ml/min/1.73 m2, IQR and Black patients (p , 0.001); for PD patients, there was
44.7, 71.3) patients compared to White patients (50.3 ml/ no major difference seen between the ethnic groups
min/1.73 m2, IQR 38.0, 64.2, p , 0.0001). (unpublished data). The reasons for admission for the
ethnic groups are shown in table 6.15. Cautious inter-
Patient survival pretation from these data is required as a significant
Figure 6.4 shows the unadjusted survival in the first proportion of patients had ‘CKD not otherwise specified’
year of RRT for the different age groups. Overall, South coded as a reason for the hospitalisation.
Asian and Black patients have better survival than
White patients and this is more apparent in the 55–75
age groups. The survival of patients on RRT in the first
year has improved over the years 2003–2011 for both Discussion
South Asian and White patients but there appears to be
a declining trend for Black patients (figure 6.5). In the Data completeness on ethnicity has improved over the
multivariate adjusted Cox regression analysis including most recent years reducing the probability of selection
41920 patients, survival after 90 days of starting RRT bias that might have occurred due to missing ethnicity
without censoring for transplantation was better for data in the previous years’ reports. Therefore, one should
South Asian and Black patients compared to White interpret with caution any perceived time trends in
patients (table 6.13). Results were similar when censored incidence rates or patient demographics between ethnic
for transplantation (data not shown). groups.
Table 6.12. Cox-regression analysis of five year graft failure by ethnicity of kidney-only transplants between 2003 and 2007

Asian 0.9 (0.6–1.5) 0.8 0.9 (0.6–1.4) 0.6
Black 1.8 (1.1–2.8) 0.01 1.5 (1.1–2.1) 0.02
128
100
90
80
Percentage survival
70
60
50
Asian
Black
White
40
90 day 1 year after 1 year
survival 90 day survival survival
30
18–34
35–44
45–54
55–64
65–74
75–84
85+
18–34
35–44
45–54
55–64
65–74
75–84
85+
18–34
35–44
45–54
55–64
65–74
75–84
85+
Fig. 6.4. Unadjusted survival by age group
and ethnicity in patients starting RRT
Age group between 2003 and 2012
100
95
Percentage survival
90
85 Asian
Black
White
80
Fig. 6.5. Age-60 adjusted survival one year
2003
2004
2005
2006
2007
2008
2009
2010
2011
after 90 days of incident patients by year of

Year of RRT start RRT start and ethnic group
Table 6.13. Cox-regression analysis of patient survival after 90 days from RRT start, by ethnic group, incident cohort 2003–2012

Asian 0.63 (0.59–0.67) ,0.0001 0.68 (0.60–0.77) ,0.0001
Black 0.5 (0.46–0.54) ,0.0001 0.58 (0.52–0.64) ,0.0001
129
Table 6.14. Cause of deaths for incident patients 2003–2012 that died by the end of 2012, by ethnic group
All ages Age ,65 Age 565
Asian Black White Asian Black White Asian Black White

N deaths 1,477 788 17,476 661 359 4,993 816 429 12,483
% of incident patients 27.4 22.9 40.9 18.9 15.4 24.0 43.2 38.5 57.0
COD (%)
Cerebrovascular disease 6.7 8.4 3.9 7.2 8.3 3.4 6.3 8.5 4.0
Cardiac disease 29.8 26.2 22.0 32.0 25.4 25.0 27.9 27.0 20.9
Infection 22.2 19.1 17.3 21.9 23.3 18.0 22.5 15.2 17.0
Malignancy 6.4 7.2 9.5 5.9 6.2 10.7 6.9 8.1 9.1
Other 17.4 16.3 24.6 18.7 21.2 27.5 16.3 11.9 23.4
Treatment withdrawal 9.4 12.1 17.1 5.3 7.3 10.0 12.7 16.6 19.8
Uncertain 8.1 10.6 5.6 9.1 8.3 5.4 7.4 12.8 5.7
N with no COD data 654 384 7,984 286 166 2,348 368 218 5,636
COD = cause of death
Black and South Asian patients were younger com- 6% of Black and South Asian patients being aged 565
pared to White patients. This, to a certain extent, was years compared to 18% of White patients [1]. It is well
probably a reflection of the younger age distribution for established that the progression to ERF and the incidence
ethnic minorities in the general population with only of RRT is much greater amongst ethnic minorities
compared to Whites [13–18]. However, these analyses
Table 6.15. Cause of hospitalisation from 90 days to one year showed that the disparity in incidence rates was more
following the start of dialysis amongst incident patients between pronounced amongst those aged 565 years and the
2002–2006, by ethnic group reasons for this are not obvious.
Percentage Life expectancy estimates for ethnic minorities in the
general population are lower than for the White popu-
Cause of hospitalisation Asian Black White lation [19] and therefore the higher incidence amongst
Abdominal pain 2.7 1.9 1.7 the elderly ethnic minority patients cannot be attributed
Access 19.6 23.3 17.9 to the possibility of them living longer to reach ERF. It is
Biochemistry 1.2 2.4 1.5 also not known if there are variations in the uptake of
Bronchitis 4.7 3.2 3.6 conservative management of ERF between the ethnic
Cancer 0.8 1.2 2.2
groups. Although the incidence of RRT (supply) is higher
Catheter 1.0 1.3 1.6
Chest pain 2.7 1.4 1.6 in the ethnic minorities, population estimates of CKD stage
CKD codes 32.5 33.3 34.1 5 (demand) are needed to ensure that there is no ethnic
CVA 0.7 0.7 0.7 disparity in access to RRT (demand–supply mismatch).
Fracture 1.8 1.3 2.5 The proportion of patients starting RRT who had at
Gastroenteritis 3.7 2.6 3.4
GI bleed 0.3 0.4 0.8
least one comorbidity was greater amongst White
Hernia 0.4 0.6 0.9 patients although ill-health is generally more frequently
High risk sepsis 3.3 3.2 3.3 reported by ethnic minorities in the general population
Ischaemic heart disease 6.3 3.7 5.9 [20]. However, the comorbidity patterns in the RRT
Low risk sepsis 2.9 2.0 1.8 population are consistent with greater incidence of
Miscellaneous 6.7 8.9 7.3
Neuro 1.9 2.2 1.9 coronary heart disease in South Asian patients, cerebro-
Overload 2.5 2.6 2.4 vascular accidents in Black patients and lower cancer
Peritonitis 1.1 1.5 1.2 rates seen in ethnic minorities in the general population
Syncope 1.6 1.4 2.0 [20].
UTI 1.7 0.8 1.7
Early referral to a renal centre was associated with
Total numbers 1,989 1,802 28,104 better uptake of PD. However despite being referred
CVA = cerebrovascular accident earlier, ethnic minorities had lower uptake of PD and
UTI = urinary tract infection lower Hb at start of RRT. They also started RRT at a
130
lower eGFR compared to White patients. The lower approaches or less active encouragement by nephrolo-
uptake of PD seen in ethnic minorities may however be gists to seek living related donors [22]; lack of suitable
as a consequence of confounding by differing centre donors with family members living outside the UK who
practices of PD use. It is also possible that the unexpected are therefore unable to be assessed or complete donor
rapid decline in kidney function in the preceding year of work up; and high prevalence of diabetes in the immedi-
RRT (unpublished work by UKRR) could have resulted ate family [23]. It has also been observed that Black
in insufficient time for adequate education about dialysis patients on dialysis had more positive coping strategies
modalities to enable patients to choose PD, or the than Whites and this may affect their perception of the
appropriate management of anaemia prior to the need need for transplant [24].
for RRT. The poor graft survival for Black patients reported in
However, once established on dialysis, the attainment this cohort is consistent with previous reports from the
of laboratory standards was better or similar for the UK [25, 26] and the USA [27, 28]. However a study
ethnic minorities for most standards except calcium for from France suggested that compared to White patients,
South Asian and URR for Black patients. Importantly, graft survival was similar for Black patients with a genetic
the attainment of the Hb standard (which was lower at pool similar to African Americans suggesting the possible
start of RRT) was no longer different between the ethnic role of social deprivation and health care access in poor
groups at one year from start of RRT. Data on use of outcomes for Black patients in the USA [29]. In the
calcium containing phosphate binders, vitamin D analyses, these disparities were observed despite adjust-
analogues, duration of HD session and type of vascular ments for area level deprivation. Black and Indo Asian
access are not available to explore the reasons for these patients have a greater likelihood of receiving kidneys
differences. These results are slightly different from at higher risk of delayed or inferior outcomes, i.e.
those previously reported [21] on a cohort of patients expanded criteria donor (ECD) kidneys, compared to
starting RRT between 1997–2004 in which attainment White patients in the USA [30]. Previous UKRR work
of Hb 5100 g/L was lower amongst Black patients and in collaboration with ODT has shown that Black and
attainment of PTH 432 pmol/L was lower for South South Asian patients were more likely to receive kidneys
Asian and Black patients. These differences were prob- with longer cold ischaemic time and HLA mismatches
ably due to the different range used for each of the both of which could influence graft survival [7]. Donor
laboratory measures analysed in this report to comply information for this cohort was not available to explore
with current UK guidelines. When analyses were the reasons for the apparent persistent inferior graft
repeated using the previous RA standards, results were survival for Black patients in the UK.
similar to the earlier report. There was a paradox in that Black and South Asian
It is reassuring to note equitable access to the trans- patients despite having reduced life expectancy in the
plant waiting list for ethnic minorities but there continues general population [19] appeared to have better survival
to be a disparity in access to deceased donor transplan- on dialysis. No adjustment for baseline comorbidity was
tation once on the waiting list. It is well acknowledged made in this report due to incomplete data but these results
that this is due to blood group and HLA disparity are consistent with previous studies from North America
compared with the predominantly White donor pool in and the UK that have adjusted for baseline comorbidity
the UK. The new UK organ allocation scheme introduced although residual confounding from missing comorbidity
in 2006 gave a greater emphasis in the points scoring data could not be excluded in these studies [31, 32, 33].
system to patients waiting longer for a transplant. The Hospitalisation rates were higher for White patients on
lack of observed impact in this report following the dialysis compared to South Asian and Black patients. Due
introduction of the new scheme may be due to the fact to several of these episodes being coded as ‘CKD not
that the majority of patients included in this report otherwise specified’, it was not possible to determine if
irrespective of their ethnicity would have waited for a the increased hospitalisation rates amongst White
similar duration of time on the waiting list, whereas the dialysis patients was due to newly acquired comorbidity
new allocation scheme would have improved access to whilst on RRT that could account for the increased
a small proportion who were on the waiting list well mortality. Several mechanisms including better adap-
before 2006. Living donor transplantation rates were tation on dialysis, better social support, less withdrawal
lower for ethnic minorities and several recipient and from dialysis and greater use of Vitamin D analogues
donor factors have been suggested including fewer amongst ethnic minorities have also been suggested for
131
better survival amongst ethnic minority dialysis patients Further studies examining survival from a predefined
[34, 35, 36, 37]. Another possible mechanism suggested eGFR early in the course of CKD stage 4–5 are needed
for this paradox is survivor bias i.e. ethnic minority patients to explore this hypothesis with more detailed assessment
with CKD and significant comorbidity are more likely to of CVD (e.g. LVEF, ABPI etc.).
die prematurely before reaching ERF or possibly less likely There are other patient outcome measures that merit
to be referred or accepted onto RRT [38]. However a more comparison between ethnic groups on RRT in the UK
recent study from the USA has shown that mortality is such as quality of life and mental health. This is currently
similar between Black and White patients with CKD stages within the remit of collaborative work being considered
3–4 questioning this hypothesis [39]. by the UKRR. Data on cause of hospitalisation episodes
Another possible mechanism is lead time bias. White for dialysis patients are required to help understand the
patients started RRT at a slightly higher eGFR compared differences in survival between the ethnic groups.
to ethnic minorities in this study. However, this differ- This report confirms the persistent high incidence of
ence was clinically very small to entirely account for the RRT, the better survival on dialysis and the poor access
ethnic differences in mortality observed in this study. It to kidney transplantation for South Asian and Black
is well established that Black and South Asian patients patients and early allograft loss for Black patients.
have rapid progression from their underlying CKD to This, in the context of increasing ethnic diversity of
ERF. It is therefore possible that they have less ‘CKD the general population and ageing of ethnic minorities
vintage’ compared to the White patients i.e may therefore will have a significant impact on the prevalence of ethnic
start RRT early with a reduced arteriosclerotic load minority patients on dialysis and impose a disproportion-
when compared with the White population. Although ate demand on dialysis provision in those areas with a high
ischaemic heart disease was more common amongst ethnic minority population. More effort is needed to
South Asian patients, the proportion of patients with at reduce progression of CKD to ERF in ethnic minorities.
least one comorbidity and those with vascular disease
and smoking were more prevalent in White patients. Conflicts of interest: none
References
1 Office for National Statistics, 2011 Census: Aggregate data (England and 11 Anemia management in people with chronic kidney disease. National
Wales). UK Data Service Census Support Institute for Health and Care Excellence 2011. Downloaded from
2 Shaw C et al. UK Renal Registry 14th annual report: Chapter 4 co- www.nice.org.uk
morbidities and current smoking status amongst patients starting 12 Pruthi R et al. UK Renal Registry 15th annual report: Chapter 9 centre
renal replacement therapy in England, Wales and Northern Ireland variation in access to renal transplantation in the UK (2006–2008).
from 2009 to 2010. Nephron Clin Pract 2012;120(suppl 1):c81–c91 Nephron Clin Pract 2013:123(suppl 1):183–193
3 Payne RA and GA Abel. UK indices of multiple deprivation – a way to 13 Department of Health: Saving lives: Our healthier nation. London:
make comparisons across constituent countries easier. Health Statistics HMSO. 1999
Quarterly 53, Spring 2012. Office of National statistics. London 14 Bryson CL, Ross HJ, Boyko EJ, Young BA. Racial and ethnic variations in
4 Levey AS et al. A simplified equation to predict glomerular filtration rate albuminuria in the US Third National Health and Nutrition Examin-
from serum creatinine [abstract]. J Am Soc Nephrol 2000;11:A0828 ation Survey (NHANES III) population: associations with diabetes and
5 Udayaraj et al. Social deprivation, ethnicity, and uptake of living kidney level of CKD. Am J Kidney Dis 2006;48(5):720–726
donor transplantation in the United Kingdom. Transplantation 2012; 15 Krop JS, Coresh J, Chambless LE, Shahar E, Watson RL, Szklo M et al.
Mar 27;93(6) A community-based study of explanatory factors for the excess risk
6 Udayaraj et al. Social deprivation, ethnicity, and access to the deceased for early renal function decline in blacks vs. whites with diabetes: the
donor kidney transplant waiting list in England and Wales. Transplan- Atherosclerosis Risk in Communities study. Arch Intern Med 1999;
tation 2010 Aug 15;90(3):279–285 159(15):1777–1783
7 Udayaraj et al. Social deprivation, ethnicity and renal replacement 16 Tarver-Carr ME, Powe NR, Eberhardt MS, LaVeist TA, Kington RS,
therapy in England and Wales: Equity of access and outcomes (Dis- Coresh J et al. Excess risk of chronic kidney disease among African-
sertation). University of Bristol, England, 2009 American versus white subjects in the United States: a population-
8 Levey A, Stevens L, Schimd C, et al. A new equation to estimate based study of potential explanatory factors. J Am Soc Nephrol 2002;
glomerular filtration rate. Ann Intern Med 2009;150(9):604–612 13(9):2363–2370
9 Fotheringham J et al. Variation in centre-specific survival in patients 17 Hsu CY, Lin F, Vittinghoff E, Shlipak MG. Racial differences in the
starting renal replacement therapy in England is explained by enhanced progression from chronic renal insufficiency to end-stage renal disease
comorbidity information from hospitalization data. Nephrol Dial in the United States. J Am Soc Nephrol 2003;14(11):2902–2907
Transplant (2013) 0:doi: 10.1093/ndt/gft363 18 Peralta CA, Ziv E, Katz R, Reiner A, Burchard EG, Fried L et al. African
10 Castledine et al. How much of the regional variation in RRT incidence ancestry, socioeconomic status, and kidney function in elderly African
rates within the UK is explained by the health needs of the general Americans: a genetic admixture analysis. J Am Soc Nephrol 2006;
population? Nephrol Dial Transplant 2012 Oct;27(10):3943–3950 17(12):3491–3496
132
19 Wohland P et al. Life Expectancy variation across England’s local areas 30 Schold JD, Kaplan B, Chumbler NR, Howard RJ, Srinivas TR, Ma L et al.
by ethnic group in 2001. Journal of Maps; 6(1):354–359 Access to quality: evaluation of the allocation of deceased donor kidneys
20 Ethnicity and ill health. Postnote (276). Parlimetnary Office of Science for transplantation. J Am Soc Nephrol 2005;16(10):3121–3127
and Technology. London, 2007 31 Pei YP, Greenwood CM, Chery AL, Wu GG. Racial differences in
21 Udayaraj et al. Ethnicity, socioeconomic status, and attainment of survival of patients on dialysis. Kidney Int 2000;58(3):1293–1299
clinical practice guideline standards in dialysis patients in the United 32 Yan G et al. The relationship of age, race, and ethnicity with survival in
Kingdom. Clin J Am Soc Nephrol 2009;4(5):979–987 dialysis patients. Clin J Am Soc Nephrol 2013;8(6):953–961
22 Navaneethan SD, Singh S. A systematic review of barriers in access to 33 Roderick, P et al. Survival of patients from South Asian and Black
renal transplantation among African Americans in the United States. populations starting renal replacement therapy in England and Wales.
Clin Transplant 2006;20:769–775 Nephrol Dial Transplant 2009 Dec;24(12):3774–3782
23 Tankersley MR, Gaston RS, Curtis JJ et al. The living donor process in 34 Lunsford SL, Simpson KS, Chavin KD, Hildebrand LG, Miles LG,
kidney transplantation: influence of race and comorbidity. Transplant Shilling LM et al. Racial differences in coping with the need for kidney
Proc 1997;29:3722–3723 transplantation and willingness to ask for live organ donation. Am J
24 Lunsford SL, Simpson KS, Chavin KD et al. Racial differences in coping Kidney Dis 2006;47(2):324–331
with the need for kidney transplantation and willingness to ask for live 35 Powe NR. Let’s get serious about racial and ethnic disparities. J Am Soc
organ donation. Am J Kidney Dis 2006;47:324–331 Nephrol 2008;19(7):1271–1275
25 Rudge C et al. Renal transplantation in the United Kingdom for patients 36 Wolfe M et al. Impact of activated vitamin D and race on survival among
from ethnic minorities Transplantation 2007;83(9):1169–1173 hemodialysis patients. J Am Soc Nephrol 2008 Jul;19(7):1379–1388
26 Loucaidou M et al. Outcome of renal transplantation in south asian 37 Nicholas SB et al. Racial disparities in kidney disease outcomes. Semin
recipients is similar to that in non-asians. Transplantation 2004;78: Nephrol 2013;33(5):409–415
1021–1024 38 Mehrotra R, Kermah D, Fried L, Adler S, Norris K. Racial differences
27 Press R, Carrasquillo O, Nickolas T, Radhakrishnan J, Shea S, Barr RG. in mortality among those with CKD. J Am Soc Nephrol 2008;19(7):
Race/ethnicity, poverty status, and renal transplant outcomes. Trans- 1403–1410
plantation 2005;80(7):917–924 39 Babayev R et al. Association of race and body mass index with ESRD and
28 Isaacs RB, Nock SL, Spencer CE, Connors AF, Jr., Wang XQ, Sawyer R mortality in CKD stages 3–4: results from the Kidney Early Evaluation
et al. Racial disparities in renal transplant outcomes. Am J Kidney Dis Program (KEEP). Am J Kidney Dis 2013;61(3):404–412
1999;34(4):706–712
29 Pallet N et al. Kidney transplant in black recipients: are African
Europeans different from African Americans? Am J Transplant 2005;
5(11):2682–2687
133
Chapter 7 Demography of the UK
Paediatric Renal Replacement Therapy
population in 2012
Rishi Pruthia, Catherine O’Brienb, Anna Casulaa, Fiona Braddona, Malcolm Lewisc,
Heather Maxwelld, Jelena Stojanovice, Yincent Tsef, Carol Inwardg, Manish D Sinhae
a
UK Renal Registry, Bristol, UK; bBirmingham Children‘s Hospital, Birmingham, UK; cManchester Children’s Hospital,
Manchester, UK; dRoyal Hospital for Sick Children (Yorkhill), Glasgow, UK; eEvelina Childrens Hospital, London, UK;
f
Royal Victoria Infirmary, Newcastle, UK; gBristol Royal Hospital for Children, Bristol, UK
Key Words (HD) and 9.2% were receiving peritoneal dialysis

Aetiology . Children . Demography . End stage renal disease . (PD).
Established renal failure . Incidence . Prevalence . Ethnicity
. In patients aged ,16 years the prevalence of ERF
. Renal replacement therapy . Survival was 56.7 pmarp and the incidence 9.0 pmarp.
. A third of patients had one or more reported
comorbidities.
. Over the past 15 years for those referred early, there
Summary has been a rise in pre-emptive transplantation rates,
rising from 26.2% in 1998–2002 to 36.3% in 2008–
. A total of 861 children and young people under 18 2012.
years with established renal failure (ERF) were . At transfer to adult services, 81.5% of patients had a
receiving treatment at paediatric nephrology centres functioning kidney transplant.
in 2012. . Being on dialysis was seen to lower survival signifi-
. At the census date, 80.2% had a functioning kidney cantly compared to having a functioning transplant
transplant, 10.6% were receiving haemodialysis with a hazard ratio of 6.3 (CI: 3.4–11.7).
135
Introduction ensure a minimum of 1 year follow up at the census date, and

were followed up to a maximum age of 16 years.
Established renal failure (ERF) requiring renal repla-
Statistical analyses
cement therapy (RRT) is a rare but significant cause of Statistical analyses were performed using SAS 9.3, with group
long term morbidity and mortality during childhood, analyses using Chi-square test and median analyses using
with specialist care being provided in 13 paediatric Kruskal-Wallis test. A Cox regression model was used in calculat-
nephrology centres in the UK. All centres are equipped ing hazard ratios for patient survival, adjusting for gender, age at
start of RRT, and RRT modality as a time dependent variable.
to provide peritoneal dialysis and haemodialysis, with
Survival probabilities were calculated using univariate Kaplan
ten centres also undertaking kidney transplantation for Meier curves.
children. In the United Kingdom (UK) in 2011, the
prevalence rate of treated ERF in children aged under
16 years was 56.8 and the incidence rate was 8.3 per
million age related population (pmarp). Results
The objectives of this report are:
Accuracy and completeness of data returns
(i) To describe the UK prevalence, incidence, causes
Efforts to improve the overall accuracy of the entire
of ERF and modality of treatment of children on
paediatric dataset by clinical teams, data managers and
RRT on 31st December 2012
statisticians over these past few years have resulted in
(ii) To describe trends of the same over the past
improved accuracy of the database, analyses and con-
15 years, and
clusions. The data returns, now showing near 100%
(iii) To describe pre-emptive transplantation rates and
data completeness being achieved by all centres for a
survival of children on RRT aged ,16 years old in
range of data items including, gender, ethnicity, treat-
the UK.
ment modality and age at start of RRT. Data complete-
ness for other core items was better than previous
reports and is shown in table 7.1 [2].
Methods
The UK paediatric prevalent ERF population in 2012
Data collection was performed by all 13 paediatric nephrology
A total of 861 children and young people under
centres managing children on RRT in the UK in 2012. Most 18 years with ERF were receiving treatment at paediatric
centres submitted data electronically to the UK Renal Registry nephrology centres in 2012. At the census date, 80.2%
(UKRR) with only two centres submitting data using paper- had a functioning kidney transplant, 10.6% were
based data returns this year. These data items were then manually receiving haemodialysis (HD) and 9.2% were receiving
entered into the current paediatric UKRR database. Thus 92% of
data returns including 791 of 861 children were performed peritoneal dialysis (PD).
electronically in 2012. Patients aged 16–18 years may receive their medical
In this report, patient groups are described as: (i) ‘prevalent’ care either in a paediatric or in an adult nephrology
group: patients who were receiving RRT on the 31st December centre. As data were incomplete for the 16–18 year old
2012; (ii) ‘incident’ group: patients who started RRT between adolescent patients, they have been excluded from the
1st January and 31st December 2012; and (iii) ‘5 year’ groups:
patients who started RRT in the periods of 1998–2002, 2003– majority of subsequent analyses (particularly when
2007 and 2008–2012. describing incidence and prevalence rates).
The populations used to calculate the incidence and preva- There were 679 children under 16 years of age receiv-
lence rates were obtained from the Office for National Statistics ing RRT in the UK in 2012. Table 7.2 shows the number
(ONS) [1]. The mid-2012 population estimate produced by the of patients receiving RRT by age group and gender plus
ONS, based on the 2011 Census, was used for calculating the
2012 incident and prevalent group rates; the 2001 Census data rate of RRT pmarp. The prevalence of RRT increased
was used for the 1998–2002, 2003–2007 and 2008–2012 ‘5 year’ with age and was higher in males across all age groups
groups. with an overall male to female prevalence ratio of 1.5.
Infants under the age of three months and ‘late presenters’ The reported prevalence rate in under 16 year olds was
(defined as children commencing dialysis within three months 56.7 pmarp.
following review by a paediatric nephrologist) were excluded
from analyses when calculating pre-emptive transplantation Table 7.3 shows the ethnic origin of current RRT
rates. For survival analysis, only patients starting RRT between patients and their prevalence rates. Children from
1st January 1998 and 31st December 2011 were included to ethnic minorities displayed higher prevalent rates of
136
Chapter 7 Demography of renal replacement therapy in children
Table 7.1. Data completeness for paediatric prevalent ERF population in 2012
Percentage completeness
First seen Height at Weight at Creatinine at Primary renal

Centre N date RRT start RRT start RRT start diagnosis
Blfst_P 36 94.4 86.1 88.9 94.4 100.0

Bham_P 89 100.0 95.5 96.6 97.8 97.8
Brstl_P 54 100.0 98.2 98.2 100.0 100.0
Cardf_P 26 96.2 100.0 100.0 100.0 100.0
Glasg_P 53 96.2 90.6 94.3 96.2 98.1
L Eve_P 96 100.0 64.6 70.8 71.9 100.0
L GOSH_P 172 98.3 88.4 95.9 95.4 100.0
Leeds_P 77 100.0 85.7 98.7 98.7 100.0
Livpl_P 34 97.1 79.4 85.3 88.2 97.1
Manch_P 73 98.6 93.2 98.6 98.6 100.0
Newc_P 35 100.0 82.9 85.7 88.6 100.0
Nottm_P 90 96.7 71.1 85.6 98.9 100.0
Soton_P 26 100.0 76.9 76.9 92.3 96.2
UK 861 98.5 84.9 91.1 93.7 99.4
Table 7.2. The UK paediatric prevalent ERF population in 2012, by age group and gender
All patients Males Females
Age group N pmarp N pmarp N pmarp Ratio M: F
0–1.99 years 21 12.9 16 19.3 5 6.3 3.1

2–3.99 years 46 29.1 35 43.2 11 14.2 3.0
4–7.99 years 140 46.1 86 55.4 54 36.5 1.5
8–11.99 years 186 67.1 115 81.0 71 52.5 1.5
12–15.99 years 286 96.3 164 107.8 122 84.2 1.3
Under 16 years 679 56.7 416 67.8 263 45.0 1.5
pmarp – per million age related population
RRT when compared with White children, with South in figure 7.1. Of the 79% with a functioning transplant,
Asian children displaying the highest rates. 52% received a deceased donor transplantation and
48% a living donor transplantation.
Modality of treatment The treatment modality in use at the start of RRT is
Current treatment modality in the prevalent paedia- displayed in figure 7.2. This shows that 48% of patients
tric population less than 16 years old in 2012 is displayed were treated with PD at the start of RRT whilst 29% of
Table 7.3. The UK paediatric prevalent ERF population by age and ethnic group in 2012a
White South Asian Black Otherb
Age group N pmarp N pmarp N pmarp N
0–3.99 years 47 18.2 13 61.6 0 0.0 3

4–7.99 years 97 40.5 25 128.2 5 64.1 4
8–11.99 years 140 54.7 29 139.1 8 95.9 13
12–15.99 years 211 78.3 45 204.9 9 102.5 8
Under 16 years 495 48.4 112 134.3 22 65.9 28
a
ethnicity data missing in two children who are excluded from this table
b
pmarp not expressed for group ‘Other’, as heterogeneous group
137
HD Live transplant
10% 14%
HD
Live transplant PD
29%
38% 11% Deceased donor
transplant
9%
Deceased donor
transplant PD
41% 48%
Fig. 7.1. RRT treatment used by prevalent paediatric patients Fig. 7.2. Treatment modality at start of RRT in prevalent paedia-
,16 years old in 2012 tric patients under 16 years of age in 2012
patients were treated with HD. Twenty-three percent of 2012. There has been a marked improvement in data
children under 16 were reported to have received a completeness in this category over the last few years
pre-emptive transplant. with missing data falling to only 0.7% which was similar
Further treatment modality analysis by age is shown in to that seen in the 2011 report [2]. Of the 679 patients,
table 7.4 which demonstrates that in the under two year renal dysplasia + reflux remained the commonest
old age group no children received a transplant and that condition causing ERF (33%), whilst there were no
the majority of patients were being treated with PD documented patients with drug nephrotoxicity.
(57.1%). This contrasts with older children in the 12 to As for associated comorbidities at the onset of RRT,
15.99 year age group where 85% had a functioning table 7.6 shows that congenital abnormalities were the
graft and where similar proportions were on HD and commonest, reported in 9.4% of patients, followed by
PD. Subsequent analysis of RRT modality by gender syndromic diagnosis at 8.8%. Overall 65.5% of patients
and ethnicity showed no difference. However as absolute had no registered comorbidities, with 23% having one
sub-group numbers are small, caution is needed in comorbidity listed, and 11.5% having two or more co-
conducting any comparative analyses. morbidities. Centre analysis showed significant variation
in reporting of registered comorbidities, with some
Cause of ERF centres, Cardiff (90%), Birmingham (84%), Glasgow
Table 7.5 and figure 7.3 show the diagnostic categories (80%) and GOSH (79%) reporting no comorbidity in
for the prevalent ERF population under 16 years in the majority of their patients, as compared to other centres
Table 7.4. Current treatment modality by age in the prevalent paediatric ERF population in 2012
Current treatment
HD PD Live transplant Deceased donor transplant
Age group N % N % N % N %
0–1.99 years 9 42.9 12 57.1 0 0.0 0 0.0

2–3.99 years 9 19.6 11 23.9 21 45.7 5 10.9
4–7.99 years 14 10.0 18 12.9 62 44.3 46 32.9
8–11.99 years 17 9.1 12 6.5 69 37.1 88 47.3
12–15.99 years 21 7.3 22 7.7 105 36.7 138 48.3
16–17.99 years 21 11.5 4 2.2 67 36.8 90 49.5
Under 16 years 70 10.3 75 11.0 257 37.8 277 40.8
Under 18 years 91 10.6 79 9.2 324 37.6 367 42.6
138
Table 7.5. Number, percentage and gender by primary renal disease as cause of ERF in the prevalent paediatric ERF population under
16 years in 2012∗
Diagnostic group Total % Male Female M: F ratio
Renal dysplasia + reflux 224 33.0 140 84 1.7

Obstructive uropathy 126 18.6 118 8 14.8
Glomerular disease 81 11.9 38 43 0.9
Congenital nephrotic syndrome 66 9.7 37 29 1.3
Tubulo-interstitial diseases 50 7.4 23 27 0.9
Uncertain aetiology 33 4.9 16 17 0.9
Renovascular disease 31 4.6 18 13 1.4
Polycystic kidney disease 28 4.1 11 17 0.6
Metabolic 20 2.9 8 12 0.7
Malignancy & associated disease 15 2.2 6 9 0.7
Missing 5 0.7 1 4 0.3
Total 679 100 416 263 1.6
∗
In 2012 there were no patients with ERF secondary to ‘drug nephrotoxicity’
which reported no comorbidity in a smaller proportion the 12–15.99 year age group, with the 0–1.99 year age
of patients, Bristol (27%) and Manchester (42%). This group having the next highest rates.
variation in reporting needs further investigation.
Trends in ERF demographics
The UK incident paediatric ERF population in 2012 There were 1,656 children under 16 years of age who
There were 117 patients under 18 years of age who had received RRT in the UK over the 15-year period
commenced RRT at paediatric renal centres in 2012. As between 1998–2012. Analysis of ERF demographics for
previously, the following analyses are restricted to the children less than 16 years of age over this period
108 patients who were under 16 years of age. included 547 patients reported to the paediatric registry
The incidence rate of RRT was 9.0 pmarp in 2012. between 1998–2002, 536 between 2003–2007 and 573
Patients commencing RRT in 2012 are displayed by age between 2008–2012. Comparing the current 5-year
and gender in table 7.7.
Table 7.8 shows that the reported incidence of RRT has Table 7.6. Registered comorbidities at onset of RRT in prevalent
been rising since 1998, the highest incidence rates seen in paediatric patients aged ,16 years with ERF in 2012
Percentage of all
Malignancy & Prevalent Comorbidity N RRT patients
associated disease Incident
Metabolic
Cerebral palsy 7 1.0
Polycystic kidney disease Chromosomal abnormality 17 2.5
Congenital abnormality 64 9.4
Primary renal disease
Renovascular disease
Congenital heart disease 14 2.1
Uncertain aetiology Consanguinity 27 4.0
Developmental delay 54 8.0
Tubulo-interstitial diseases
Diabetes 3 0.4
Congenital nephrotic
syndrome
Family member with ERF 19 2.8
Glomerular disease
Liver disease 12 1.8
Malignancy 4 0.6
Obstructive uropathy Neural tube defect 4 0.6
Renal dysplasia + reflux
Prematurity 54 8.0
Psychological disorder 6 0.9
0 5 10 15 20 25 30 35 40 Syndromic diagnosis 60 8.8
No reported comorbidity 445 65.5
Fig. 7.3. Primary renal disease percentage in incident and One reported comorbidity 156 23.0
prevalent paediatric ERF patients in 2012 for whom a causative Two or more comorbidities 78 11.5
diagnosis was reported
139
Table 7.7. The incident paediatric ERF population in the UK in 2012, by age group and gender
All patients Male Female
Age group N pmarp N pmarp N pmarp M: F ratio
0–1.99 years 20 12.3 16 19.3 4 5.1 3.8

2–3.99 years 12 7.6 8 9.9 4 5.2 1.9
4–7.99 years 19 6.3 10 6.4 9 6.1 1.1
8–11.99 years 19 6.9 10 7.0 9 6.7 1.1
12–15.99 years 38 12.8 3 15.1 15 10.4 1.5
Under 16 years 108 9.0 67 10.9 41 7.0 1.6
period with the two previous 5-year periods there has comorbidities over the last 15 years. Syndromic diagnoses
been an overall increase in the number of children treated (8.6%), congenital abnormalities (8.0%), developmental
with RRT, particularly in children aged under two years delay (7.9%) were the most common reported co-
(table 7.9). The percentage of children on RRT who morbidities in 2008–2012, with little change in the
were from South Asian or Black ethnic backgrounds also percentage of children receiving RRT with a reported
increased during this period (table 7.10). The reported comorbidity over the last 15 years.
patient population at most paediatric renal centres has As for changes in modality at the start of RRT,
similarly grown in size since 1998–2002 (table 7.11). figure 7.4 shows that the percentage of children who
Table 7.12 shows the number and percentage of were using PD at the start of RRT has fallen from
children receiving RRT with each of the major reported 54.7% in 1998–2002 to 43.7% in 2008–2012, whilst the
percentage commencing RRT on HD increased from
Table 7.8. Reported average incident rate by age group, in 5-year 23.1% in 1998–2002 to 29.1% in 2008–2012. During
time periods, of children under 16 years of age commencing RRT this period the overall percentage receiving a transplant
at the start of RRT remained largely unchanged although
Per million age related population
living donation has risen from 7.1% in 1998–2002 to
Age group 1998–2002 2003–2007 2008–2012 18.0% in 2008–2012, with a corresponding fall in
deceased donor transplantation from 15.1% to 9.3%
0–1.99 years 11.3 12.7 12.5 for the same time period.
2–3.99 years 6.7 5.2 7.6 Table 7.13 shows the diagnostic categories for 540 of
4–7.99 years 5.5 6.3 6.5
8–11.99 years 8.9 7.7 8.8 the 546 (98.9%) patients in 1998–2002, for 525 of the
12–15.99 years 13.2 13.5 13.9 536 (97.9%) patients in 2003–2007 and 564 of the 573
Under 16 years 9.1 9.3 9.9 (98.4%) patients in 2008–2012 aged ,16 years for
whom a causative diagnosis was reported.
Table 7.9. Number and percentage of children who commenced RRT, by age group and 5-year period, at start of RRT
1998–2002 2003–2007 2008–2012 1998–2012
Age group N % N % N % % change
0–1.99 years 78 14.3 90 16.8 99 17.3 3.0

2–3.99 years 48 8.8 35 6.5 59 10.3 1.5
4–7.99 years 81 14.8 87 16.2 92 16.1 1.2
8–11.99 years 139 25.4 113 21.1 120 20.9 −4.5
12–15.99 years 201 36.7 211 39.4 203 35.4 −1.3
Under 16 years 547 536 573
140
Table 7.10. Number and percentage of children under 16 years who commenced RRT, by ethnicity and 5-year period of starting RRT∗
1998–2002 2003–2007 2008–2012 1998–2012
Ethnic group N % N % N % % change
White 428 79.0 407 76.9 407 71.9 −7.1

S Asian 84 15.5 82 15.5 98 17.3 1.8
Black 13 2.4 18 3.4 21 3.7 1.3
Other 17 3.1 22 4.2 40 7.1 3.9
Under 16 years 542 529 566
∗
Five children in 1998–2002, seven in 2003–2007 and seven in 2008–2012 with no ethnicity recorded are excluded from this table
Table 7.11. Number and percentage of children under 16 years reported to the UKRR, by renal centre and 5-year period of starting RRT∗
1998–2002 2003–2007 2008–2012 1998–2012
Centre N % N % N % % change
Blfst_P 17 3.1 15 2.8 26 4.5 1.4

Bham_P 51 9.3 55 10.3 66 11.5 2.2
Brstl_P 36 6.6 40 7.5 29 5.1 −1.5
Cardf_P 17 3.1 24 4.5 17 3.0 −0.1
Glasg_P 40 7.3 36 6.7 43 7.5 0.2
L Eve_P 61 11.2 45 8.4 62 10.8 −0.4
L GOSH_P 87 15.9 102 19.0 115 20.1 4.1
Leeds_P 46 8.4 55 10.3 44 7.7 −0.7
Livpl_P 23 4.2 28 5.2 16 2.8 −1.4
Manch_P 58 10.6 44 8.2 54 9.4 −1.2
Newc_P 28 5.1 28 5.2 23 4.0 −1.1
Nottm_P 60 11.0 51 9.5 57 9.9 −1.0
Soton_P 22 4.0 13 2.4 21 3.7 −0.4
Total <16 546 536 573
∗
One child in 1998–2002 with unknown centre of RRT start was excluded from this table
Table 7.12. Trends in comorbidity at the start of RRT in the paediatric population under 16 years, by 5-year period
1998–2002 2003–2007 2008–2012 1998–2012
Comorbidity N % N % N % % change
Cerebral palsy 6 1.1 12 2.2 7 1.2 0.1

Chromosomal abnormality 18 3.3 17 3.2 9 1.6 −1.7
Congenital abnormality 43 7.9 49 9.1 46 8.0 0.2
Congenital heart disease 15 2.7 13 2.4 13 2.3 −0.5
Consanguinity 29 5.3 15 2.8 19 3.3 −2.0
Developmental delay 47 8.6 39 7.3 45 7.9 −0.7
Diabetes 2 0.4 6 1.1 2 0.3 0.0
Family member with ERF 21 3.8 16 3.0 12 2.1 −1.7
Liver disease 3 0.5 11 2.1 10 1.7 1.2
Malignancy 7 1.3 5 0.9 2 0.3 −0.9
Neural tube defect 2 0.4 5 0.9 4 0.7 0.3
Prematurity 31 5.7 24 4.5 37 6.5 0.8
Psychological disorder 11 2.0 6 1.1 10 1.7 −0.3
Syndromic diagnosis 30 5.5 52 9.7 49 8.6 3.1
No reported comorbidity 369 67.5 347 64.7 399 69.6 2.2
One reported comorbidity 118 21.6 133 24.8 115 20.1 −1.5
Two or more comorbidities 60 11.0 56 10.4 59 10.3 −0.7
141
60
1998–2002
2003–2007
2008–2012
50
40
30
20
10
0
HD PD Deceased donor Live transplant
transplant Fig. 7.4. Treatment modality at start of
Modality RRT by 5-year time period
Overall there has been an increase in the percentage of started RRT between 1998–2012, pre-emptive trans-
children receiving RRT with renal dysplasia + reflux and plantation was seen to occur in 32.5% of patients and
interestingly also those with an uncertain aetiology was significantly higher in males (35.4%) than females
between 1998–2002 and 2008–2012 although absolute (27.8%), p = 0.006 (table 7.14). Ethnicity was also seen
numbers are very small (table 7.13). to be a significant factor, with children from Black
(14.7%) and South Asian (19.3%) ethnicity having
Pre-emptive transplantation significantly lower rates of transplantation than their
Of a total of 1,656 patients who started RRT between White counterparts (35.8%), p , 0.0001. Analysis by
1998–2012, 460 patients were excluded from this age at start of RRT showed that as expected, the lowest
analysis (94 patients were excluded due to being aged rate of pre-emptive transplantation was in the three
,3 months, and a further 366 patients were excluded months to two year group (5.1%), whilst children aged
due to being late presenters). Of 1,196 patients identified four to sixteen years had similar rates of pre-emptive
as being aged three months to ,16 years and having transplantation. As for PRD, children with polycystic
Table 7.13. Number and percentage of children under 16 years for whom a primary renal diagnosis had been reported as a cause of
ERF, by 5 year time period and observed change in proportion of patients in each diagnostic group∗
1998–2002 2003–2007 2008–2012 1998–2012
Primary renal diagnosis N % N % N % % change
Renal dysplasia + reflux 149 27.6 182 34.7 181 32.1 4.5
Obstructive uropathy 84 15.6 75 14.3 95 16.8 1.3
Glomerular disease 130 24.1 105 20.0 96 17.0 −7.1
Congenital nephrotic syndrome 30 5.6 26 5.0 37 6.6 1.0
Tubulo-interstitial diseases 38 7.0 48 9.1 44 7.8 0.8
Uncertain aetiology 11 2.0 28 5.3 34 6.0 4.0
Renovascular disease 26 4.8 13 2.5 20 3.5 −1.3
Polycystic kidney disease 15 2.8 17 3.2 20 3.5 0.8
Metabolic 34 6.3 18 3.4 30 5.3 −1.0
Malignancy & associated disease 7 1.3 9 1.7 6 1.1 −0.2
Drug nephrotoxicity 16 3.0 4 0.8 1 0.2 −2.8
∗
Six children in 1998–2002, eleven in 2003–2007 and nine in 2008–2012 with no PRD recorded are excluded from this table
142
Table 7.14. Demographics of pre-emptive transplantation in Table 7.15. Modality, gender, ethnicity and primary renal
children aged 3 months to 16 years in the UK between 1998– diagnosis of patients transferred out of paediatric nephrology
2012, analysed by 5-year time period, gender, ethnicity, age at centres in 2012
start of RRT and primary renal diagnosis
%
N (%) N distribution
pre-emptively
N transplanted Modality
HD 11 13.6
Total cohort analysed (1998–2012) 1,196 389 (32.5) PD 4 4.9
Transplant 66 81.5
Time period
1998–2002 408 107 (26.2) Gender
2002–2007 388 137 (35.3) Female 27 33.3
2008–2012 400 145 (36.3) Male 54 66.7
Gender Ethnicity∗
Male 742 263 (35.4) Black 3 3.7
Female 454 126 (27.8) Other 2 2.5
South Asian 13 16.3
Ethnicity
White 62 77.5
Black 34 5 (14.7)
Other 49 16 (32.7) Primary Renal Diagnosis∗
South Asian 197 38 (19.3) Glomerular disease 22 27.5
White 899 322 (35.8) Renal dysplasia + reflux 21 26.3
Obstructive uropathy 12 15.0
Age at start of RRT
Congenital nephrotic syndrome 6 7.4
3 months–1.99 years 117 6 (5.1)
Uncertain aetiology 6 7.5
2–3.99 years 118 32 (27.1)
Metabolic 4 5.0
4–7.99 years 211 75 (35.6)
Tubulo-interstitial diseases 3 3.8
8–11.99 years 288 101 (35.1)
Drug nephrotoxicity 2 2.4
12–15.99 years 462 175 (37.9)
Polycystic kidney disease 2 2.5
Primary renal diagnosis Malignancy & associated disease 1 1.3
Renal dysplasia + reflux 387 161 (41.6) Renovascular disease 1 1.3
Glomerular disease 223 26 (11.7) ∗
Obstructive uropathy 219 94 (42.9) Ethnicity missing in 1 patient, and PRD missing in 1 patient
Congenital nephrotic syndrome 78 5 (6.4)
Tubulo-interstitial diseases 73 19 (26.0)
Metabolic 69 29 (42.0) transferred out was 18.1 years with an inter-quartile
Polycystic kidney disease 40 18 (45.0)
Renovascular disease 37 15 (40.5)
range of 17.7 years to 18.5 years.
Uncertain aetiology 25 8 (32.0) Table 7.15 shows that of the transferred patients 66.7%
Malignancy & associated disease 13 1 (7.7) were male, with ethnic minorities constituting 22.5% of
Drug nephrotoxicity 12 3 (25.0) patients. The vast majority (81.5%) had a functioning
renal transplant at the time of transfer to an adult
renal centre. Glomerular disease and renal dysplasia +
reflux accounted for the primary renal diagnosis in
kidney disease (45%) and obstructive uropathy (42.9%) over 50% of patients.
had the highest rates of pre-emptive transplantation,
whilst those with congenital nephrotic syndrome Survival of children on RRT during childhood
(6.4%) had the lowest rate. Over time there appears to Of patients under the age of 16, 1,548 were identified
have been a rise in pre-emptive transplantation rates, as starting RRT between 1998 and 2011 at paediatric
rising from 26.2% in 1998–2002 to 36.3% in 2008– centres in the UK and were included in the survival
2012, p = 0.004 (table 7.14). analyses. At the census date (31st December 2012)
there were a total of 103 deaths within the cohort on
Transfer of patients to adult renal services in 2012 RRT at age ,16, with a median follow up time of
A total of 81 patients were reported by paediatric 3.6 years (range of one day to 15 years). Table 7.16
nephrology centres to have been transferred to adult shows the survival hazard ratios after adjustment for
renal services in 2012. The median age of patients age at start of RRT, gender and RRT modality, and
143
Table 7.16. Survival hazard ratio during childhood for paediatric starting RRT aged .8 years, or 5 year survival probability
RRT patients aged ,16 years in the UK adjusted for age at start of for children starting RRTaged .12 years. This figure again
RRT, gender and RRT modality
highlights worse outcomes for those aged 0–1.99 years.
Hazard Confidence
ratio interval p-value Mortality data in 2012
There were nine deaths in renal paediatric centres in
Age
0–1.99 years 4.7 2.4–9.3 ,0.0001 2012, eight children were aged ,16 and one between
2–3.99 years 2.4 1.1–5.5 0.03 16–18 years at the time of death. In children aged
4–7.99 years 1.6 0.7–3.7 0.23 ,16 years with treated ERF, the reported mortality in
8–11.99 years 1.3 0.6–3.0 0.48 2012 in the UK at paediatric centres was 1.2% (8/679).
12–16 years 1.0 – – The median age at death was 10.8 years with a range of
Gender 0.2 years to 17.1 years. At the time of death, three chil-
Female 1.3 0.9–1.9 0.19 dren had received a kidney transplant and three were
Male 1.0 –
on dialysis (one haemodialysis and two PD).
RRT modality Septicaemia was cited as a cause of death in three
Dialysis 6.3 3.4–11.7 ,0.0001
Transplant 1.0 – patients, two of which were associated with trans-
plantation and one with peritoneal dialysis. One patient
died as a result of chest complications during PD
catheter insertion. Three further patients were receiving
highlights that children starting RRT at 0–1.99 years had active palliative care at the time of death. A clear cause
the worst survival outcomes with a hazard ratio of 4.7 of death could not be identified in the two remaining
(CI 2.4–9.3, p , 0.0001) when compared to 12–16 year patients who died in 2012.
olds. Outcomes in the 2–3.99 age group were also
significantly worse with a hazard ratio of 2.4 (CI 1.1–
5.5, p = 0.03). Being on dialysis, as expected, was seen
to lower survival significantly compared to having a Discussion
functioning transplant with a hazard ratio of 6.3 (3.4–
11.7, p , 0.0001). Figure 7.5 shows unadjusted Kaplan This report from the Paediatric Renal Registry has
Meier survival probabilities. As the maximum age of focussed on the current demography and the demo-
follow up was restricted to 16 years, it was not possible graphic trends over the past 15 years of the UK paediatric
to calculate 10 year survival probabilities for patients ERF population.
1.00
0.95
0.90
Survival
0.85
0.80 Age 0–2

Age 2–4
Age 4–8
0.75 Age 8–12
Age 12–16
0.70 Fig. 7.5. Unadjusted KM in paediatric

0 2 4 6 8 10 patients starting RRT between 1998 and
Years 2011, by age at start
144
This report includes 679 children and adolescents increased over the past 15 years (26.2% in 1998–2002
under 16 years of age, who were receiving RRT in 2012. to 36.3% in 2008–2012). There were significantly lower
The sub-section on the trends in demographics includes rates of pre-emptive transplantation in girls and ethnic
children and adolescents under 16 years of age on RRT; minorities and further detailed studies investigating
546 from 1998–2002, 536 from 2003–2007 and 573 these would be important.
from 2008–2012.
Comorbidities
Data completeness At the onset of RRT, 34.5% of patients had one or
The ongoing sustained effort to improve data accuracy more associated comorbidities. This overall proportion
must continue and the aim to move to full electronic of children with associated comorbidities has shown
annual returns from all centres remains. A revised data little change over the past 15 years. There continues to
set (The NEW Paediatric Dataset) will be used in the be significant variation in registered comorbidity rates
near future to improve registry returns. These ongoing between centres (10–73%, data not shown); it is likely
efforts to improve the quality and consistency of the that this is influenced by different reporting practices
data received will be rewarded by enabling enhanced between centres. This remains an area for further work
interpretation of centre specific measures of clinical from the registry and individual centres.
performance. Nearly 92% of data was submitted elec-
tronically from 11 of 13 paediatric nephrology centres Causes of ERF and observed trends 1998–2012
in the UK. Data returns were complete for key data As previously, renal dysplasia + reflux (33%), glomeru-
items and this together with improved checking and lar disease (11.9%) and obstructive uropathy (18.6%) were
validation procedures within the registry contributed to the commonest listed aetiologies for children with ERF.
continuing quality improvement. These accounted for 63.5% of all patients for whom a
primary diagnosis had been reported. Observation of
Incidence, prevalence and trends trends over the 15-year period showed an increase in
The incidence rate of RRT in the less than 16 year age the percentage of children receiving RRT with renal
group was 9.0 pmarp in 2012; this rate has been rising dysplasia + reflux and those with unknown aetiology.
since 1998. The overall prevalence rate of RRT in the
less than 16 year age group was 56.7 pmarp. The preva- Transfer out and survival data
lence of RRT increased with age and was higher in Data relating to transfer to adult renal services is
males across all age groups. Overall, there was a continu- included in the current report. The median age of trans-
ing trend of increased prevalence of children on RRT fer was 18.1 years. Of patients receiving RRT, 81.5%
with increased age, in keeping with improved survival transferred with a functioning renal transplant. There
with increasing age. This coupled with an increase in appears to be variation in practice between centres
the number of children receiving RRT over the past regarding transition and transfer out arrangements; it
15 years has led to a steady increase in the prevalent is also likely that variability exists in reporting of ‘transfer
ERF population. out’ timelines to the registry for patients being transi-
tioned to adult centres. Unpublished results from a
Treatment modality of ERF survey conducted by the paediatric subcommittee of
Peritoneal dialysis was the initial treatment modality the registry earlier this year highlighted that transition
for 48% of children at the start of treatment, 29% com- practices varied as to when children began the process
menced HD and 23% received a pre-emptive transplant. (range: 15–16 years); and when they were expecting to
Age influenced the modality of RRT with the majority of have successfully ‘transitioned children’ and transferred
the under two’s (57%) receiving PD. Overall the majority them out into adult services with some centres aiming
of prevalent children (79%) on RRT had a functioning for 16 years whilst others for 18 years. Consensus regard-
transplant. ing terminology and process will facilitate future com-
parative interpretation.
Pre-emptive transplantation Survival data of children on ERF during childhood
Over the last 15 years, pre-emptive transplantation who commenced RRT between 1998 and 2011 highlights
was seen to occur in 32.5% of children under 16 years the less favourable outcome for children less than
age. The rate of pre-emptive transplantation has two years of age. The data also highlights the significantly
145
better survival of children with functioning transplants long term survival data needs follow up into young
when compared to those on dialysis. Longer term adulthood. This is the focus of an ongoing project of
survival data up to five years was available for those the UK Renal Registry.
aged ,12 years and 10 year survival data for those
,8 years only. For the majority of children on RRT Conflicts of interest: none
References
1 http://www.Ons.Gov.Uk/census
2 Pruthi R, O‘Brien C, Casula A, Braddon F, Lewis M, Maxwell H, Tse Y,
Inward C, Sinha MD. UK Renal Registry 15th Annual Report (December
2011): Chapter 4 Demography of the UK Paediatric Renal Replacement
Therapy Population in 2011. Nephron Clin Pract. 2013;123(suppl 1):
81–92. doi: 10.1159/000353323
146
Chapter 8 Survival and Cause of Death
of UK Adult Patients on Renal
Replacement Therapy in 2012:
Rishi Pruthi, Retha Steenkamp, Terry Feest

UK Renal Registry, Bristol, UK
Key Words
. One year age adjusted survival for prevalent dialysis
Cause of death . Comorbidity . Dialysis . End stage renal patients remained relatively unchanged at 89.7% in
disease . Established renal failure . Haemodialysis . Median the 2011 cohort compared to 89.8% in the 2010
life expectancy . Outcome . Peritoneal dialysis . Renal cohort.
replacement therapy (RRT) . Survival . Transplant . Vintage . One year survival for prevalent diabetic patients
increased from 81.6% in the 2002 cohort to 84.9%
in the 2011 cohort. An increase in survival was
Summary also observed between the 2010 and 2011 cohorts.
. RRT patients aged 35–39 had a mortality rate 16.6
. Unadjusted 1 year after 90 day survival for patients times higher than the age matched general popu-
starting renal replacement therapy (RRT) in 2011 lation, whereas RRT patients aged 85+ had a
increased to 87.5% from 87.3% for those starting mortality rate only 2.7 times higher. The overall
in 2010. relative risk of death improved across most age
. In incident patients aged 565 years, unadjusted 1 groups in the 2011 cohort.
year after 90 day survival increased from 63.9% in . In the prevalent RRT dialysis population, cardio-
1997 to 80.6% in the 2011 cohort. An increase in vascular disease accounted for 22% of deaths and
survival was also observed between the 2010 and treatment withdrawal 19%, whilst 21% were
2011 cohorts. recorded as other cause of death.
. In incident patients aged 565 years the one year . The median life years remaining for an incident
survival of diabetic patients was better than that of patient aged 25–29 years was 18.5 years and
non-diabetic patients, and three year survival was approximately 2.4 years for a 75+ year old.
similar.
147
Introduction about any apparent differences in outcome, for centres

which appear to be outliers the UKRR will follow the
The analyses presented in this chapter examine: a) clinical governance procedures as set out in chapter 2
survival from the start of renal replacement therapy of the 2009 UKRR Report [5].
(RRT) of adults; b) survival amongst all prevalent adult
dialysis patients alive on 31st December 2011; c) the
cause of death for incident and prevalent adult patients
and d) the projected life years remaining for adult Methods
patients starting RRT. They encompass the outcomes
from the total incident adult UK dialysis population The unadjusted survival probabilities (with 95% confidence
reported to the UK Renal Registry (UKRR), including intervals) were calculated using the Kaplan–Meier method, in
the 19.5% who started on peritoneal dialysis and the which the probability of surviving more than a given time can
7% who received a pre-emptive renal transplant. These be estimated for members of a cohort of patients, without any
adjustment for age or other factors that affect the chances of
results are therefore a true reflection of the outcomes in survival. Where centres are small, or the survival probabilities are
the whole UK adult RRT population. Analyses of survival greater than 90%, the confidence intervals are only approximate.
within the first year of starting RRT include patients who In order to estimate the difference in survival of different sub-
were recorded as having started RRT for established renal groups of patients within the cohort, a stratified proportional
failure (as opposed to acute kidney injury) but who had hazards model (Cox) was used where appropriate. The results
from the Cox model were interpreted using a hazard ratio.
died within the first 90 days of starting RRT, a group When comparing two groups, the hazard ratio is the ratio of the
excluded from most other countries’ registry data. As is estimated hazard for group A relative to group B, where the
common in other countries, survival analyses are also hazard is the risk of dying at time t given that the individual
presented for the first year after 90 days. has survived until this time. The underlying assumption of a
The term established renal failure (ERF) used through- proportional hazards model is that the hazard ratio remains
constant throughout the period under consideration. Whenever
out this chapter is synonymous with the terms end stage used, the assumptions of the proportional hazards model were
renal failure (ESRF) and end stage renal disease (ESRD) tested.
which are in more widespread international usage. To allow comparisons between centres with differing age distri-
Within the UK, patients have disliked the term ‘end butions, survival analyses were statistically adjusted for age and
stage’; the term ERF was endorsed by the English reported as survival adjusted to age 60. This gives an estimate of
what the survival would have been if all patients in that centre
National Service Framework for Renal Services, published had been aged 60 at the start of RRT. This age was chosen because
in 2004. it was approximately the average age of patients starting RRT
The prevalent dialysis patient group was defined as all 15 years ago at the start of the UKRR’s data collection. The average
patients over 18 years old, alive and receiving dialysis on age of patients commencing RRT in the UK has been stable
31st December 2011 who had been on dialysis for at least recently around an age of 62 years, but the UKRR has maintained
age adjustment to 60 years for comparability with all previous
90 days at one of the UK adult renal centres. years’ analyses. Diabetic patients were included in all analyses
Since 2006, the UKRR has openly reported and unless stated otherwise, and in many analyses diabetic patients
published centre attributable RRT survival data. It is were also analysed separately and compared to non-diabetic
again stressed that these are raw data which continue to patients. All analyses were undertaken using SAS 9.3.
require very cautious interpretation. The UKRR can Definition of renal replacement therapy start date
adjust for the effects of the different age distributions of The incident survival figures quoted in this chapter are from
patients in different centres, but lacks sufficient data the first day of renal replacement therapy whether with dialysis
from many participating centres to enable adjustment or a pre-emptive transplant. In the UKRR all patients starting
for primary renal diagnosis, other comorbidities at start RRT for ERF are included from the date of the first RRT treatment
wherever it took place (a date currently defined by the clinician) if
of RRT (age and comorbidity, especially diabetes, are the clinician considered the renal failure irreversible. Should a
major factors associated with survival [1–3]) and ethnic patient recover renal function within 90 days they were then
origin, which have been shown to have an impact on excluded. These UK data therefore may include some patients
outcome (for instance, better survival is expected in who died within 90 days who had developed acute potentially
centres with a higher proportion of Black and South reversible renal failure but were recorded by the clinician as
being in irreversible established renal failure.
Asian patients) [4]. This lack of information on case- Previously, the UKRR asked clinicians to re-enter a code for
mix makes interpretation of any apparent difference in established renal failure in patients initially coded as having
survival between centres difficult. Despite the uncertainty acute renal failure once it had become clear that there was no
148
Chapter 8 Survival in UK RRT patients in 2012
recovery of kidney function. However, adherence to this require- The one year incident survival is for patients who started RRT
ment was very variable, with some clinicians entering a code for from 1st October 2010 until the 30th September 2011 and followed
established renal failure only once a decision had been made to up for one full year (e.g. patients starting RRT on 1st December
plan for long-term RRT [6]. All UK nephrologists have now 2010 were followed through to 30th November 2011). The 2012
been asked to record the date of the first haemodialysis session incident patients could not be analysed as they had not yet been
and to record whether the patient was considered to have acute followed for a sufficient length of time.
kidney injury (acute renal failure) or to be in ERF at the time. For analysis of 1 year after 90 day survival, patients who started
For patients initially categorised as ‘acute’, but who were sub- RRT from 1st October 2010 until 30th September 2011 were
sequently categorised as ERF, the UKRR assigns the date of this included in the cohort and they were followed up for a full one
first ‘acute’ session as the date of start of RRT. year after 90 days.
UKRR analyses of electronic data extracted for the immediate To help identify any centre differences in survival from the
month prior to the start date of RRT provided by clinicians small centres (where confidence intervals are large), an analysis
highlighted additional inconsistencies in the definition of this of 1 year after 90 day survival using a rolling four year combined
first date when patients started on peritoneal dialysis, with the incident cohort from 2008 to 2011 was also undertaken. For those
date of start reported to the UKRR being later than the actual centres which had joined the UKRR after 2008, data were not
date of start. These findings are described in detail in chapter 13 available for all the years but the available data were included.
of the 2009 Report [6]. This concern is unlikely to be unique to The death rate per 1,000 patient years was calculated by
the UK, but will be common to analyses from all renal centres dividing the number of deaths by the person years exposed. Person
and registries. years exposed are the total days at risk for each patient (until
In addition to these problems of defining day 0 within one death, recovery or lost to follow-up) expressed as years. All
country, there is international variability on when patient data patients, even those who died within the first 90 days of RRT,
are collected by national registries with some countries (often were included in the death rate calculation.
for financial re-imbursement or administrative reasons) defining Adjustment of 1 year after 90 day survival for the effect of
the 90th day after starting RRT as day 0, whilst others collect comorbidity was undertaken using a rolling five year combined
data only on those who have survived 90 days and report as incident cohort from 2007 to 2011. Twenty-one centres returned
zero the number of patients dying within the first 90 days. .85% of comorbidity data for patients in the combined cohort.
Thus as many other national registries do not include reports Adjustment was first performed to a mean age of 60 years, then
on patients who do not survive the first 90 days, survival from to the average distribution of primary diagnoses for all 21
90 days onwards is also reported to allow international com- centres. The individual centre data were then further adjusted
parisons. This distinction is important, as there is a much higher for average distribution of comorbidity present at these centres.
death rate in the first 90 days, which would distort comparisons. The survival hazard function was calculated as the probability
of dying in a short time interval considering survival to that
Methodology for incident patient survival interval.
Patients were considered ‘incident’ at the time of their first
RRT, thus patients re-starting dialysis after a failed transplant Methodology for prevalent dialysis patient survival
were not included. For prevalent dialysis patients, all patients who had been
Some patients recover renal function after more than 90 days established on dialysis for at least 90 days on 31st December
but subsequently returned to RRT. If recovery was for less than 2011 were included in these analyses. Prevalent dialysis patients
90 days, the start of renal replacement therapy was calculated on 31st December 2011 were followed up in 2012 and were
from the date of the first episode and the recovery period censored at transplantation. When a patient is censored at
ignored. If recovery was for 90 days or more, the length of time transplantation, this means that the patient is considered as alive
on RRT was calculated from the day on which the patient restarted up to the point of transplantation, but the patient’s status post-
RRT. transplant is not considered.
The incident survival cohort was NOT censored at the time of As discussed in previous reports, comparison of survival of
transplantation and therefore included the survival of the 7% who prevalent dialysis patients between centres is complex. Survival
received a pre-emptive transplant. An additional reason for not of prevalent dialysis patients can be studied with or without
censoring was to facilitate comparison between centres. Centres censoring at transplantation and it is common practice in some
with a high proportion of patients of South Asian and Black origin registries to censor at transplantation. Censoring could cause
are likely to have a healthier dialysis population, because South apparent differences in survival between those renal centres with
Asian and Black patients are less likely to undergo early transplan- a high transplant rate and those with a low transplant rate,
tation [7], and centres with a high pre-emptive transplant rate are especially in younger patients where the transplant rate is high-
likely to have a less healthy dialysis population as transplantation est. Censoring at transplantation systematically removes younger
selectively removes fit patients only. fitter patients from the survival data. The differences are likely
The incident (‘take-on’) population in any specific year to be small due to the relatively small proportion of patients
excludes those who recovered within 90 days from the start of being transplanted in a given year compared to the whole
RRT, but includes patients who recovered from ERF after 90 dialysis population (about 22% of the dialysis population aged
days. For survival analyses, patients newly transferred into a under 65 and 3% of the population aged 65 years and over). To
centre who were already on RRT were excluded from the incident allow comparisons with other registries the survival results for
population for that centre and were counted at the centre at which prevalent dialysis patients CENSORED for transplantation have
they started RRT. been quoted. To understand survival of patients, including
149
survival following transplantation, the incident patient analyses Results of incident (new RRT) patient survival
should be viewed.
The 2011 incident cohort included 6,750 patients who
Methodology of cause of death
The EDTA-ERA Registry codes for cause of death were used. started RRT, without any period of renal function recov-
These have been grouped into the following categories: ery lasting more than 90 days. The unadjusted 1 year after
90 day survival for incident patients starting RRT in 2011
. Cardiac disease (table 8.1) has increased to 87.5% compared to 87.3% in
. Cerebrovascular disease the 2010 cohort.
. Infection
. Malignancy
. Treatment withdrawal Comparison of survival between UK countries
. Other Two years incident data have been combined to
. Uncertain increase the size of the patient cohort, so that any differ-
ences between the four UK countries are more likely to be
Some centres had high completeness of data returns to the
UKRR for cause of death, whilst others returned no information. reliably identified (table 8.2). These data have not been
Completeness of cause of death data was calculated for all adjusted for differences in primary renal diagnosis,
prevalent patients on RRT that died in a specific year with cause ethnicity, socio-economic status or comorbidity, nor for
of death data completed for that year. differences in life expectancy in the general populations
Adult patients aged 18 years and over from England, Wales,
Scotland and Northern Ireland were included in the analyses of
of the four UK countries. There was no significant differ-
cause of death. The incident patient analysis included all patients ence in the 90 day survival between the UK countries. One
starting RRT in the years 2000–2011. Analysis of prevalent year after 90 day survival was significantly lower in Wales
patients included all those aged over 18 years and receiving RRT compared to England. It has been postulated that a greater
on 31st December 2011. The death rate was calculated for the prevalence of cardiovascular disease in Wales compared to
UK general population (data from the Office of National Statistics)
by age group and compared with the same age group for prevalent
England may account for the difference.
patients on RRT on 31st December 2011. There are known regional differences in the life
expectancy of the general population within the UK.
Methodology of median life expectancy (life table calculations) Table 8.3 shows differences in life expectancy between
Kaplan Meier survival analyses were used to calculate the the UK countries. These differences in life expectancy
hazard of death by age group (18–34, 35–44, 45–54, 55–64,
65–74, 75+) for incident patients starting RRT from 2000–2009,
are not accounted for in these analyses and are likely to
with at least three years follow-up from 2010 to 2012. The patient be one of the reasons behind the variation in survival
inclusion criteria are the same to that of the incident patient cohort between renal centres and UK countries.
described above. Patients were followed until death, censoring
(recovery or lost to follow-up) or the end of the study period. Modality
Life expectancy which gives the probability of surviving until the
next time period was calculated as: 1 – hazard of death. Median
It is impossible to obtain truly valid comparisons of
life years remaining is then the difference between the age when survival of patients starting RRT on different treatment
reaching the 50% probability of survival and the age of starting modalities, as modality selection is not random. In the
RRT. UK, patients starting peritoneal dialysis as a group were
younger and fitter than those starting haemodialysis
Methodology for comparing mortality in prevalent RRT
patients with the mortality in the general population and were transplanted more quickly. The age adjusted
Data on the UK population in mid-2012 and the number of 1 year survival estimates for incident patients starting
deaths in each age group in 2012 were obtained from the Office RRT on HD and PD were 89.3% and 92.9% respectively,
of National Statistics. The age specific UK death rate was calcu- both showing a slight increase from the previous year
lated as the number of deaths in the UK per thousand people in (figure 8.1, table 8.4). Over the last six years the one
the population. The age specific expected number of deaths in
the RRT population was calculated by applying the UK age specific
death rate to the total of years exposed for RRT patients in that age
group. This is expressed as deaths per 1,000 patient years. The age Table 8.1. Unadjusted survival of incident patients, 2011 cohort
specific number of RRT deaths is the actual number of deaths
observed in 2012 in RRT patients. The RRT observed death rate Survival
was calculated as number of deaths observed in 2012 per 1,000 Interval (%) 95% CI N
patient years exposed. Relative risk of death was calculated as Survival at 90 day 94.5 93.9–95.0 6,750
the ratio of the observed and expected death rates for RRT Survival 1 year after 90 days 87.5 86.7–88.3 6,359
patients.
150
Table 8.2. Incident patient survival across the UK countries, combined 2 year cohort (2010–2011), adjusted to age 60
Interval England N Ireland Scotland Wales UK

Survival at 90 day (%) 96.2 96.0 95.8 96.6 96.2
95% CI 95.8–96.6 94.4–97.7 94.7–96.8 95.6–97.7 95.8–96.5
Survival 1 year after 90 days (%) 90.5 90.4 88.9 88.2 90.3
95% CI 89.9–91.1 87.7–93.1 87.1–90.7 86.1–90.3 89.7–90.9
Table 8.3. Life expectancy in years in UK countries, 2008–2010 Table 8.4. One year after 90 day incident patient survival by first
(source ONS [8]) established modality 2005–2011 cohort (adjusted to age 60)
(excluding patients whose first modality was transplantation)
At birth At age 65
Age adjusted 1 year after 90 days % survival
Country Male Female Male Female 95% CI
England 78.6 82.6 18.2 20.8 Year HD PD
Northern Ireland∗ 77.1 81.5 17.4 20.2
Scotland 75.8 80.4 16.8 19.3 2011 89.3 92.9
Wales 77.6 81.8 17.7 20.3 88.3–90.3 91.6–94.3
UK 78.2 82.3 18.0 20.6 2010 87.7 93.3
∗ 86.6–88.8 91.9–94.7
Provisional data for Northern Ireland 2009 87.6 93.1
86.5–88.7 91.6–94.6
year after 90 days survival has progressively improved 2008 87.1 93.1
86.0–88.2 91.7–94.4
in HD patients, but remained static in PD patients
2007 87.8 94.5
(table 8.4). 86.7–88.9 93.3–95.7
2006 86.5 94.1
Age 85.4–87.7 92.8–95.4
Tables 8.5 to 8.10 show survival of all incident patients, 2005 85.3 92.5
84.0–86.5 91.1–94.0
those aged 65 and above and those aged below 65 years,
for up to ten years after start of renal replacement
therapy. In the UK, short term survival (survival at 90
days) increased to 94.5% (94.2% for patients starting There is a steep decline in survival with advancing age
RRT in 2010) (table 8.5). Survival 1 year after 90 days (figures 8.2 and 8.3).
also increased compared to last year and this was mainly There was a curvilinear increase in death rate per 1,000
due to an increase in survival for patients aged younger patient years with age, shown in figure 8.3 for the period
than 65 years (table 8.6). Longer term survival of patients one year after 90 days. There were differences between
on RRT continued to improve (tables 8.8, 8.9, 8.10). the overall death rates across all age groups with the
100
98
96
94
Percentage survival
92
90
88
86
84 Fig. 8.1. Trend in 1 year after 90 day
82 incident patient survival by first modality,
2005–2011 cohort (adjusted to age 60)
80
2005 2006 2007 2008 2009 2010 2011 2005 2006 2007 2008 2009 2010 2011 (excluding patients whose first modality
Year was transplantation)
151
Table 8.5. Unadjusted 90 day survival of incident patients, 2011 Table 8.6. Unadjusted 1 year after day 90 survival of incident
cohort, by age patients, 2011 cohort, by age
Age Survival (%) 95% CI N Age Survival (%) 95% CI N
18–64 97.8 97.2–98.2 3,370 18–64 94.1 93.2–94.8 3,284

565 91.2 90.2–92.1 3,380 565 80.6 79.1–82.0 3,075
All ages 94.5 93.9–95.0 6,750 All ages 87.5 86.7–88.3 6,359
death rate in Scotland and Wales significantly higher than Table 8.7. Increase in proportional hazard of death for each 10
in England. year increase in age, 2011 incident cohort
Hazard of death for
The effect of censoring age related survival at the time of Interval 10 year age increase 95% CI
transplantation
The current method for calculating survival for incident First 90 days 1.70 1.56–1.85
1 year after first 90 days 1.64 1.55–1.73
patients does not censor at transplantation. From
Table 8.8. Unadjusted survival of incident patients, 1997–2011 cohort for patients aged 18–64
95% CI for
Cohort 1 year 2 year 3 year 4 year 5 year 6 year 7 year 8 year 9 year 10 year latest year N
2011 93.4 92.5–94.2 3,370

2010 92.2 86.6 85.4–87.7 3,375
2009 91.3 85.4 80.8 79.4–82.2 3,160
2008 91.6 86.2 81.4 77.3 75.8–78.6 3,481
2007 92.7 87.2 82.0 77.1 73.4 71.8–74.8 3,347
2006 90.8 85.2 80.3 76.0 72.4 68.5 66.8–70.1 3,182
2005 89.7 83.6 78.6 73.8 69.3 65.7 62.6 60.8–64.4 2,828
2004 89.7 83.6 78.2 72.8 68.2 64.5 61.5 57.7 55.7–59.6 2,571
2003 89.5 82.8 77.5 72.6 67.6 63.5 59.8 57.0 54.4 52.3–56.5 2,271
2002 88.7 80.9 74.9 69.4 65.3 61.4 58.0 55.1 52.0 49.9 47.7–52.1 2,034
2001 88.3 81.3 75.5 70.5 65.3 60.6 56.5 53.0 50.2 48.2 45.7–50.7 1,611
2000 89.2 81.4 74.6 69.3 64.0 59.4 55.9 52.7 50.3 47.6 45.0–50.1 1,533
1999 87.0 81.1 73.4 67.6 62.2 58.1 54.0 51.1 48.7 47.1 44.4–49.8 1,349
1998 87.6 80.3 74.5 69.6 64.2 59.2 55.4 53.4 50.2 47.9 45.0–50.8 1,172
1997 85.3 77.5 69.6 63.7 58.8 54.5 51.5 49.1 47.9 44.1 40.0–48.1 589
Table 8.9. Unadjusted survival of incident patients, 1997–2011 cohort for patients aged 565
95% CI for
2011 77.5 76.1–78.9 3,380

2010 76.4 63.6 61.9–65.2 3,287
2009 76.9 63.8 52.8 51.1–54.6 3,147
2008 74.8 61.5 50.3 40.9 39.2–42.6 3,184
2007 75.3 61.5 50.2 41.0 32.5 30.9–34.1 3,221
2006 72.1 58.6 47.4 37.8 29.6 23.8 22.3–25.3 3,139
2005 71.3 57.4 45.5 36.4 28.2 21.5 16.9 15.6–18.3 2,946
2004 69.3 54.4 43.0 34.6 27.4 21.6 17.0 13.5 12.2–14.9 2,633
2003 68.4 53.9 42.1 32.2 24.7 18.5 14.6 11.5 8.9 7.8–10.1 2,317
2002 66.1 50.9 40.6 32.2 24.3 18.7 14.1 11.3 8.7 6.9 5.9–8.1 2,090
2001 66.6 52.0 38.1 28.9 21.7 16.3 12.2 9.6 8.1 6.2 5.1–7.6 1,557
2000 66.0 52.4 39.6 28.6 22.3 17.4 13.4 10.0 7.8 6.0 4.8–7.2 1,497
1999 68.4 51.7 39.3 30.1 22.5 16.6 12.0 9.0 6.9 5.5 4.3–6.9 1,218
1998 62.7 45.6 36.3 26.6 20.2 14.1 10.7 7.7 5.7 4.6 3.5–6.1 1,017
1997 63.3 46.5 31.7 22.8 14.6 9.9 5.9 4.5 2.7 2.0 0.9–3.7 412
152
Table 8.10. Unadjusted survival of incident patients, 1997–2011 cohort for patients of all ages
95% CI for
2011 85.5 84.6–86.3 6,750

2010 84.4 75.2 74.2–76.3 6,662
2009 84.1 74.6 66.9 65.7–68.0 6,307
2008 83.6 74.4 66.5 59.9 58.7–61.0 6,665
2007 84.1 74.6 66.4 59.3 53.3 52.0–54.5 6,568
2006 81.5 72.0 64.0 57.1 51.2 46.3 45.1–47.6 6,321
2005 80.3 70.3 61.7 54.8 48.3 43.2 39.3 38.1–40.6 5,774
2004 79.4 68.8 60.4 53.5 47.6 42.9 39.0 35.4 34.1–36.7 5,204
2003 78.8 68.2 59.7 52.3 46.1 40.9 37.1 34.2 31.6 30.2–33.0 4,588
2002 77.2 65.7 57.5 50.6 44.6 39.8 35.8 32.9 30.0 28.2 26.8–29.6 4,124
2001 77.7 66.9 57.2 50.1 44.0 38.9 34.9 31.8 29.6 27.7 26.1–29.3 3,168
2000 77.7 67.1 57.3 49.3 43.5 38.7 35.0 31.7 29.4 27.1 25.5–28.7 3,030
1999 78.2 67.2 57.2 49.8 43.4 38.5 34.1 31.2 28.9 27.4 25.7–29.2 2,567
1998 76.0 64.3 56.8 49.7 43.8 38.3 34.7 32.2 29.6 27.9 26.0–29.8 2,189
1997 76.3 64.8 54.0 46.9 40.7 36.2 32.9 30.8 29.4 26.8 24.1–29.6 1,001
800
figure 8.4, it can be seen that 50% of patients starting RRT Wales
700 Scotland
aged between 45–54 survived for over 10 years, 50% of N Ireland
patients starting RRT aged between 55–64 survived for 600 England
5.75 years and 50% of patients starting RRT aged between 500
Death rate
65–74 survived for 3.3 years. 400

Figure 8.5 shows the survival of incident patients, 300
excluding those who died within the first 90 days and
200
shows that 50% of patients aged between 55–64 years
100
survived for 6 years and 50% of patients aged between
65–74 years survived for 3.6 years. 0
18–34 35–44 45–54 55–64 65–74 75–84 85+
Censoring at transplantation would make the longer Age group
term outcomes of younger patients (who were more likely Fig. 8.3. One year after 90 days death rate per 1,000 patient years
to have undergone transplantation) appear worse than by UK country and age group for incident patients, 2008–2011
they actually were. Without censoring, the 10 year cohort
100
95
90
Percentage survival
85
80
75
70
65 90 day 1 year after 1 year

survival 90 day survival survival
60
55
18–34
35–44
45–54
55–64
65–74
75–84
85+
18–34
35–44
45–54
55–64
65–74
75–84
85+
18–34
35–44
45–54
55–64
65–74
75–84
85+
Age group
Fig. 8.2. Unadjusted survival of incident patients by age group, 2011 cohort
153
100
90
80
Percentage survival
70
60
50 18–34
40 35–44
45–54
30
55–64
20 65–74
10 75+
0 Fig. 8.4. Survival of incident patients
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 (unadjusted), 1997–2011 cohort (from day
Years 0), without censoring at transplantation
survival for patients aged 18–34 years was 83.6% It is important to note that these death rates are not
(figure 8.4), which contrasts with a 57.5% survival if cen- directly comparable with those produced by the USRDS
soring at the time of transplantation (data not shown). Registry, as the UK data include the first 90 day period
For more detailed information on this effect, refer to when death rates are higher than subsequent time periods.
the 2008 Report [9]. The unadjusted survival analyses (tables 8.8, 8.9, 8.10,
figures 8.8, 8.9) and annual death rates (figure 8.7) show
Age and hazard of death by age in the first 12 months a large improvement in 1 to 10 year survival across
Figure 8.6 shows the monthly hazard of death from the the years for both those aged under and those over
first day of starting RRT by age group, which falls sharply 65 years. One year survival amongst patients aged less
during the first 4–5 months, particularly for older patients. than 65 years at start of RRT has improved from 85.3%
A 10 year increase in patient age was associated with a in the 1997 cohort to 93.4% in the 2011 cohort.
1.70 times increased risk of death within 90 days and Similarly, for patients aged 65 years and over there has
a 1.64 times increased risk of death within 1 year after been a 14.2% absolute improvement in one year survival
90 days (table 8.7). from the 1997 to 2011 cohorts. As these are observational
data it remains difficult to attribute this reduction in risk
Changes in survival in the 2000–2011 cohort of death to any specific improvements in care.
The death rate per 1,000 patient years in the first year
of starting RRT from 2000 to 2011 is shown in figure 8.7. Gender
There was a declining trend in the overall death rate, There were no survival differences between genders in
although this appears to have levelled off during the an incident cohort of patients starting RRT from 2000 to
last four years. There has been a steeper rate of decline 2009 and followed up for a minimum of three years until
in the older age group (aged 65 years and older). 2012 (figure 8.10). Gender differences were investigated
100
90
80
Percentage survival
70
60
50 18–34
40 35–44
30 45–54
55–64
20 65–74
10 75+
0 Fig. 8.5. Survival of incident patients
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 (unadjusted), 1997–2011 cohort (from day
Years 90), without censoring at transplantation
154
0.07 90
18–34
0.06 35–44 80
0.05 45–54 70
Hazard of death
Percentage survival
55–64 60 1 year
0.04 65–74 2 year
75+ 50 3 year
0.03
40 4 year
0.02 5 year
30 6 year
0.01 7 year
20 8 year
0.00 9 year
0 1 2 3 4 5 6 7 8 9 10 11 12 10 10 year
Months 0
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Fig. 8.6. First year monthly hazard of death, by age group 1997– Year
2011 combined incident cohort Fig. 8.9. Change in long term survival by year of starting RRT, for
incident patients aged 565 years
550
65+
500 All ages
RRT (vintage) when comparing survival without
450 18–64 censoring for transplantation. Figure 8.11 shows the
400 instantaneous hazard of death by age group. The
350
apparent vintage effect when censoring for trans-
Death rate
300
250
plantation (data not shown) is at least in part because
200 these younger and healthier patients are only included
150 in the survival calculation up to the date of trans-
100 plantation. In the older age groups there were decreasing
50 numbers remaining alive beyond seven years accounting
0 for the increased variability seen. Figures 8.12 and 8.13
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Year show these data for the non-diabetic and diabetic patients
Fig. 8.7. One-year incident death rate per 1,000 patient years by respectively. Non-diabetic patients were defined as all
age group, 2000–2011 cohort incident patients excluding patients with diabetes as the
primary renal disease.
in the first 90 days and 1 year after the first 90 days and
there was also no evidence of a survival difference (data Time trend changes in incident patient survival, 2000–2011
not shown). cohort
The time trend changes are shown in figure 8.14. The
Change in survival on renal replacement therapy left hand plot, which includes only those centres that
by vintage have been sending data continuously since 2000, shows
Incident RRT patients in the UK continued to show a similar improvement in survival to the plot in which
little evidence of a worsening prognosis with time on data from all renal centres are analysed.
100
95 Male
90 Female
85 80
Percentage survival
80
1 year
75 60
Survival
2 year
70 3 year
65 4 year
5 year 40
60 6 year
55 7 year
8 year 20
50 9 year
45 10 year
40 0
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 0 12 24 36 48 60 72 84 96 108 120 132
Year Months
Fig. 8.8. Change in long term survival by year of starting RRT, for Fig. 8.10. Long term survival of incident patients by gender,
incident patients aged 18–64 years 2000–2009 combined cohort, adjusted to age 60
155
0.5
18–34
35–44
45–54
0.4 55–64
65–74
75+
Hazard of death
0.3
0.2
0.1
Fig. 8.11. Six monthly hazard of death, by

0.0 vintage and age group, 1997–2011 incident
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 cohort after day 90 (not censored at
Years on RRT transplantation)
0.5
18–34
35–44
0.4 45–54
55–64
65–74
Hazard of death
0.3 75+
0.2
0.1
0.0 vintage and age group, 1997–2011 non-
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 diabetic incident cohort after day 90 (not
Years on RRT censored at transplantation)
Analysis of centre variability in 1 year after 90 days cautiously as they were not adjusted for comorbidity,
survival ethnicity or primary renal disease and patient numbers
The one year after 90 day survival for the 2011 incident were small in many centres. Survival results for centres
cohort is shown in figure 8.15 for each renal centre. The with less than 20 incident patients in 2011 (Clwyd,
tables for these data and for 90 day survival are given in Dumfries & Galloway, Inverness) are not shown in
appendix 1 at the end of this chapter (tables 8.25 and figure 8.15, although they were included in the national
8.26). The age adjusted individual centre survival for and UK survival calculations.
each of the last nine years can also be found in appendix In the analysis of 2011 incident cohort survival data,
1, table 8.27. There was much variability in survival some of the smaller centres had wide confidence intervals
between centres, but these results have to be interpreted (figure 8.15) due to small numbers of patients. This was
0.5
18–34
35–44
0.4 45–54
55–64
Hazard of death
65–74
0.3 75+
0.2
0.1
0.0 vintage and age group, 1997–2011 diabetic
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 incident cohort after day 90 (not censored
Years on RRT at transplantation)
156
92
90
Percentage survival
88
86
84
All UKRR sites

82 (England, Northern Ireland,
2000 UKRR sites only Scotland Wales) Fig. 8.14. Change in one-year after 90 day
80 survival, 2000–2011 incident cohort
00 01 02 03 04 05 06 07 08 09 10 11 00 01 02 03 04 05 06 07 08 09 10 11 (adjusted to age 60)
Year Showing 95% confidence intervals
100
95
Percentage survival
90
85
80
Upper 95% Cl
75 Survival
Lower 95% Cl
70
L St.G
West NI
Ipswi
L Guys
Bristol
B Heart
Carsh
Bangor
Sthend
L Barts
Dudley
York
Hull
M RI
B QEH
Stoke
Truro
Redng
Belfast
Abrdn
Nottm
Dunfn
Shrew
Prestn
Salford
Camb
Carlis
Plymth
Derby
Leic
Ports
Stevng
Klmarnk
Basldn
Brightn
L Rfree
L Kings
L West
Covnt
Dundee
Edinb
Glouc
Norwch
Wolve
Liv RI
Bradfd
Donc
Middlbr
Wrexm
Oxford
Sund
Glasgw
Kent
Exeter
Dorset
Leeds
Cardff
Newry
Wirral
Sheff
Liv Ain
Ulster
Antrim
Newc
Swanse
Colchr
Airdrie
Chelms
England
N Ireland
Wales
Scotland
UK
Centre
Fig. 8.15. Survival one-year after 90 days, adjusted to age 60, 2011 incident cohort
addressed by including a larger cohort across several consistently had survival above the 95% upper limit for
years, which will also assess sustained performance. the last few years. With 71 centres it would be expected
Similar to previous years, this is shown as a rolling four that only three centres would be outside these limits by
year cohort from 2008 to 2011. These data are presented chance. It is important to acknowledge that these data
as a funnel plot in figure 8.16. For any number of patients have not been adjusted for any patient related factor
in the incident cohort (x-axis) one can identify whether
any given survival rate (y-axis) falls within, plus or 100
98 Dotted lines show 99.9% limits
minus 2 standard deviations (SDs) from the national Solid lines show 95% limits
96
mean (solid lines, 95% limits) or 3 SDs (dotted lines,
Percentage survival
94
99.9% limits). Table 8.11 allows centres to be identified 92
90
on this graph by finding the number of patients treated 88
by the centre and then looking up this number on the 86
84
x-axis. Two centres (Swansea, Glasgow) had survival 82
below the 95% lower limit whilst seven centres (Ipswich, 80
London St. George’s, Stevenage, London Guys, London 78
0 100 200 300 400 500 600 700 800 900 1,000 1,100 1,200 1,300
Barts, London West, Western Trust Northern Ireland) Number of incident patients
had survival above the 95% upper limit. Amongst these, Fig. 8.16. Funnel plot for age adjusted 1 year after 90 days
St George’s was above the 99% upper limit having survival, 2008–2011 incident cohort
157
Table 8.11. Age adjusted (to age 60) 1 year after 90 day survival, 2008–2011 incident cohort
1 year after 90 day 1 year after 90 day

Centre N survival % Centre N survival %
D & Gall 58 87.2 Redng 347 92.5

Clwyd 75 91.0 Middlbr 357 86.4
Ulster 77 83.6 L St.G 366 94.1
Wrexm 81 87.3 Hull 370 89.2
Inverns 82 89.8 Newc 372 88.3
Newry 86 88.3 B Heart 393 91.2
Carlis 105 85.1 Liv RI 395 91.5
Bangor 109 89.5 Stevng 400 92.6
Sthend 112 89.2 Covnt 410 90.2
Antrim 115 91.6 Camb 416 89.7
West NI 126 94.6 Brightn 425 88.6
Basldn 129 88.3 Nottm 445 91.5
Colchr 131 87.1 Swanse 460 85.0
Donc 132 89.5 Exeter 492 90.2
Klmarnk 138 88.3 Prestn 499 87.5
York 141 90.3 Kent 499 89.7
Dunfn 144 90.9 Salford 517 88.4
Ipswi 154 94.1 Leeds 533 89.8
Liv Ain 158 84.6 L Kings 546 89.1
Truro 170 91.6 M RI 556 89.6
Airdrie 173 86.0 L Guys 581 92.5
Dudley 179 85.1 Oxford 590 89.5
Chelms 185 88.2 Ports 598 89.6
Wirral 205 88.9 Sheff 601 91.7
Sund 207 85.1 Bristol 607 89.1
Abrdn 209 88.6 Glasgw 608 87.0
Dundee 216 88.6 Cardff 662 87.8
Shrew 219 89.4 L Rfree 731 91.3
Bradfd 227 88.0 Carsh 760 89.8
Glouc 233 90.7 L Barts 845 92.4
Plymth 233 90.4 B QEH 882 90.9
Belfast 256 90.5 Leic 903 90.9
Dorset 280 90.4 L West 1,307 91.5
Derby 310 89.4 England 21,226 90.1
Norwch 310 89.9 N Ireland 660 90.4
Edinb 317 86.1 Scotland 1,945 87.8
Wolve 320 88.6 Wales 1,387 87.2
Stoke 343 88.7 UK 25,218 89.8
except age (i.e. not comorbidity, primary renal disease or Analysis of the impact of adjustment for comorbidity
ethnicity) and have not been censored at transplantation, on the 1 year after 90 day survival
so the effect of differing centre rates of transplantation Although comorbidity returns to the UKRR have
was not taken into account. Variation in the proportion remained poor, there was an increase in the number of
of patients with terminal illness receiving RRT between centres returning more than 85% of comorbidity data
centres could also contribute to variations in survival to the UKRR for patients starting RRT in 2011. Using
and provide a possible explanation for lower survival the combined incident cohort from 2007–2011, it was
than expected for some centres. In addition, another found that 21 centres had returned comorbidity data
possible reason why several of the best performing for more than 85% of patients and these centres were
centres are London based could be that they serve large included in this analysis. Adjustment was first performed
ethnic minority populations which are known to have to age 60, then to the average distribution of primary
better survival on dialysis [4]. diagnoses for all 21 centres. Further adjustment was
158
Table 8.12. The effect of adjustment for age, PRD and comorbidity on survival, 2007–2011 incident cohort, % survival 1 year after
90 days
Age, PRD and

Centre∗ Unadjusted Age adjusted Age, PRD adjusted comorbidity adjusted
Ulster 78.9 85.2 86.7 87.3

Swanse 80.9 86.8 88.4 90.0
Sund 84.6 86.6 87.4 88.2
Bradfd 84.8 86.9 87.6 88.9
Basldn 84.9 89.6 90.4 91.3
Middlbr 85.5 88.7 89.4 90.2
Dorset 85.9 90.7 90.8 91.2
Wolve 85.9 89.0 89.8 90.0
Derby 86.4 90.3 91.2 91.4
Wrexm 86.5 90.1 90.9 90.5
Leeds 86.6 89.4 90.2 91.0
L Kings 86.6 88.8 89.9 90.0
Hull 86.9 90.0 90.5 91.0
Bristol 87.8 90.8 91.3 91.7
Oxford 88.3 90.5 90.9 90.9
Shrew 89.2 92.3 92.8 90.8
Nottm 89.2 91.7 92.4 92.8
Truro 90.4 93.2 93.7 93.3
Kent 90.8 93.0 93.3 93.0
York 91.8 93.9 94.3 93.9
Stevng 93.8 94.8 95.4 94.9
All 21 centres 87.3 90.3 91.0 91.3
∗
Centre included if .85% comorbidity data available
then made to the average distribution of comorbidities during the first 90 days of starting RRT (2011 cohort)
present at those centres. (figure 8.18) and in the subsequent year (figure 8.19);
Research has suggested that adjustment for comorbid- this might be due to patient selection.
ity explains a modest part of the variance in ERF patient Long term survival for diabetic and non-diabetic
outcomes [10]. At centre level however, the prevalence of patients was evaluated in a cohort of patients starting
comorbidities could vary substantially between patient RRT from 2000 to 2009 with a minimum of three years
populations of different centres and it could be expected follow-up until 2012. These data show large differences
that adjustment for comorbidity may explain an in the 18–44 year and 45–64 year age groups between
increased amount of the variance in outcome. It can be diabetic and non-diabetic patient survival, but there
seen that adjustment for age has the largest effect, most was very little difference in three year survival between
notably in those centres with the lower unadjusted diabetics and non-diabetics in the older age group. In
survival figures. There were only minor differences for the age group 18–44, 89% of non-diabetic patients were
most centres after adjustment for primary renal diagnosis. alive five years after start of RRT compared to 70% for
In four centres (Swansea, Bradford, Basildon, Middles- diabetic patients. In the age group 45–64, 66% of non-
brough) adjustment for comorbidity had a noticeable diabetic patients were alive 5 years after start of RRT
effect on adjusted survival (table 8.12, figure 8.17) helping compared to 49% for diabetic patients (figure 8.20).
explain the lower survival noted in figure 8.15.
Standard primary renal disease and survival
Survival in patients with diabetes It is hard to set survival standards because these should
Although it has previously been shown that diabetic be age, gender, ethnicity and comorbidity adjusted and
patients have worse long term survival compared to this is not yet possible from UKRR data. The current
non-diabetic patients [3], non-diabetic patient survival 5th edition of the Renal Association Clinical Practice
in the older age group (65 years and older) was worse Guidelines [11] does not set any standards for audit of
compared to diabetic patients in the same age group patient survival.
159
100
Unadjusted
Adjusted age
Adjusted age & PRD
95 Adjusted age, PRD & comorb
90
Percentage survival
85
80
75
70
Ulster
Swanse
Sund
Bradfd
Basldn
Middlbr
Dorset
Wolve
Derby
Wrexm
Leeds
L Kings
Hull
Bristol
Oxford
Shrew
Nottm
Truro
Kent
York
Stevng
All 21 centres
Fig. 8.17. The effect on survival after
sequential adjustment for age, PRD and
Centre
comorbidity, 2007–2011 incident cohort
100
The 3rd Renal Standards document defined standard
98 primary renal disease using the EDTA-ERA diagnosis
Percentage survival
96 codes (including only codes 00–49); this excluded

patients with renal disease due to diabetes and other sys-
94 Non-diabetic 18–44
Diabetic 18–44
temic diseases. It is more widespread practice to simply
92 Non-diabetic 45–64 exclude patients with diabetes, so these analyses are also
Diabetic 45–64 included in this report to allow comparison with reports
90 Diabetic 65+
Non-diabetic 65+ from other registries. The survival for patients starting
88 RRT in the 2011 cohort in younger age groups (aged
1 11 21 31 41 51 61 71 81 91
Days
18–54) and followed up for a maximum of one year is
shown in table 8.13. For a longer term comparison, the
Fig. 8.18. Survival at 90 days for incident diabetic and non-
diabetic patients by age group for patients starting RRT, 2011
2002 cohort is also included (table 8.13).
cohort
100 100
90
80
95
Percentage survival
Percentage survival
70
60
90 Non-diabetic 18–44 50
Diabetic 18–44 40 Non-diabetic 18–44
Non-diabetic 45–64 Diabetic 18–44
30
85 Diabetic 45–64 Non-diabetic 45–64
Diabetic 65+ 20 Diabetic 45–64
Diabetic 65+
Non-diabetic 65+ 10 Non-diabetic 65+
80 0
0 30 60 90 120 150 180 210 240 270 300 330 360 0 12 24 36 48 60 72 84 96 108 120
Days Months
Fig. 8.19. Survival at 1 year after 90 days for incident diabetic and Fig. 8.20. Long term survival for incident diabetic and non-
non-diabetic patients by age group for patients starting RRT, 2011 diabetic patients by age group, 2000–2009 cohort, followed up
cohort for a minimum of 3 years
160
Table 8.13. One-year incident dialysis patient survival (from day 0–365), patients aged 18–54, 2011 and 2002 cohort (excludes patients
whose first modality was transplantation)
2011 cohort 2002 cohort
Standard primary All primary renal diseases Standard primary All primary renal diseases
First treatment renal diseasea except diabetesb renal diseasea except diabetesb
All dialysis % 97.1 95.3 95.4 93.9
95% CI 95.8–98.0 94.0–96.3 93.7–97.1 92.2–95.5
HD % 96.5 94.3 93.4 91.6
95% CI 94.7–97.7 92.7–95.6 90.7–96.0 89.2–94.0
PD % 98.3 97.4 98.6 97.9
95% CI 96.0–99.3 95.4–98.6 71.1–100 96.3–99.6
a
Includes patients with EDTA diagnostic codes 00–49
b
Excludes patients with diabetes as primary renal disease
Results of prevalent patient survival analyses prevalent dialysis patient survival compared to the 2010
cohort when three centres were outliers below the 95%
Tables 8.14 and 8.16 show the one year survival on lower limits compared to two centres in this most recent
dialysis, after censoring at the time of transplantation. analysis. The number of centres that were outliers above
Patients who have been on dialysis for less than 90 days the 95% upper limit decreased from five in the 2010
were excluded. One year survival for prevalent dialysis cohort to two in this most recent analysis.
patients remained relatively unchanged at 89.7% in the The effect of censoring at transplantation on survival
2011 cohort compared to 89.8% in the 2010 cohort. was investigated in the 2011 prevalent dialysis cohort.
Table 8.15 gives the 2011 cohort one year death rate Results show that this had a minimal effect on prevalent
for prevalent dialysis patients in each UK country. The dialysis patient 1 year survival and outlier status (data
one-year death rate in Wales was significantly higher not shown). Table 8.14 allows centres in figure 8.25 to
than in the three other UK countries: the higher median be identified by finding the number of patients treated
age in Wales together with socio-economic reasons by the centre and the corresponding survival and then
probably explains this. looking this up on the axes of the funnel plot.
Figure 8.21 shows the one year survival of dialysis
patients who were alive and receiving dialysis on 31st The one year death rate in prevalent dialysis patients
December 2011, stratified by age group. in the 2011 cohort by age group
The death rates for prevalent patients on dialysis by
One year survival of prevalent dialysis patients by age group are shown in figure 8.26. The younger patients
centre included in this analysis are a selected higher risk group,
The age-adjusted one year survival of dialysis patients as the similar aged transplanted patients have been
in each centre is shown in table 8.14 and is illustrated excluded. The increase in the death rate was not linear
in figures 8.22 and 8.23; the data for those patients aged with age; with a 10 year increase in age in the younger
,65 years and those aged 65 years and over are separ- patients, the death rate increased by about 10 deaths
ated. Figure 8.24 shows the age adjusted (adjusted to per 1,000 patient years compared with an increase of
age 60) data and in figure 8.25 as a funnel plot. The 160 deaths per 1,000 patient years in the older age groups.
solid lines show the 2 standard deviation limits (95% The apparent differences between the countries were not
limits) and the dotted lines the limits for 3 standard statistically significant except for Wales where the death
deviations (99.9% limits). With over 70 centres included, rate was significantly higher compared to England.
it would be expected by chance that three centres would
fall outside the 95% (1 in 20) confidence limits. The One year survival of prevalent dialysis patients by UK
survival for two centres (Leeds, Cardiff) was below the country, 2000 to 2011 cohort
95% confidence limits and for two centres (London One year survival for prevalent patients seemed to be
West, Birmingham QEH) was above the 95% confidence improving in most of the UK countries (figure 8.27). In
limits. The funnel plot analysis shows an improvement in Northern Ireland and Wales numbers were much
161
Table 8.14. One year survival of prevalent dialysis patients in each centre (adjusted to age 60), 2011 cohort
Adjusted Lower Upper Adjusted Lower Upper

Centre N 1 year survival 95% CI 95% CI Centre N 1 year survival 95% CI 95% CI
England Prestn 555 90.6 88.4 92.8

B Heart 466 88.3 85.8 91.0 Redng 318 90.8 88.0 93.7
B QEH 1,037 91.7 90.1 93.2 Salford 469 88.9 86.2 91.6
Basldn 181 88.4 84.5 92.5 Sheff 632 88.8 86.7 91.0
Bradfd 218 87.7 83.7 91.8 Shrew 212 89.9 86.5 93.4
Brightn 415 89.4 86.9 92.0 Stevng 505 91.9 89.9 94.0
Bristol 524 90.6 88.5 92.8 Sthend 135 87.8 83.3 92.5
Camb 460 88.9 86.5 91.3 Stoke 379 90.6 88.1 93.2
Carlis 82 88.8 83.0 95.0 Sund 179 86.4 81.8 91.2
Carsh 809 91.2 89.6 92.9 Truro 166 89.6 85.8 93.5
Chelms 148 90.7 86.8 94.7 Wirral 236 90.4 87.1 93.8
Colchr 106 89.1 84.3 94.2 Wolve 367 88.6 85.8 91.5
Covnt 424 91.7 89.4 94.0 York 146 88.6 84.2 93.2
Derby 327 90.1 87.3 93.0 N Ireland
Donc 180 91.1 87.6 94.7 Antrim 160 91.5 88.1 95.1
Dorset 293 90.4 87.7 93.2 Belfast 288 89.8 86.8 92.9
Dudley 202 91.4 88.1 94.9 Newry 125 84.1 78.7 90.0
Exeter 426 88.0 85.5 90.6 Ulster 116 91.6 87.7 95.6
Glouc 227 90.6 87.6 93.7 West NI 181 92.3 89.0 95.7
Hull 399 91.1 88.8 93.6 Scotland
Ipswi 157 90.4 86.5 94.5 Abrdn 230 90.9 87.6 94.3
Kent 441 89.3 86.8 91.8 Airdrie 168 86.4 81.6 91.4
L Barts 994 90.0 88.2 91.8 D & Gall 65 87.4 80.9 94.3
L Guys 634 91.1 89.1 93.1 Dundee 214 92.0 89.1 95.0
L Kings 563 89.9 87.6 92.2 Dunfn 180 88.2 84.3 92.4
L Rfree 740 90.2 88.3 92.1 Edinb 313 90.7 87.8 93.8
L St.G 340 88.5 85.6 91.5 Glasgw 657 88.5 86.4 90.7
L West 1,383 91.5 90.2 92.8 Inverns 98 88.0 82.9 93.4
Leeds 569 86.7 84.3 89.2 Klmarnk 185 89.8 86.0 93.7
Leic 954 90.2 88.6 91.9 Wales
Liv Ain 153 83.8 78.7 89.2 Bangor 107 89.8 84.9 95.0
Liv RI 499 89.0 86.5 91.6 Cardff 574 86.3 83.9 88.8
M RI 537 90.5 88.2 92.8 Clwyd 95 90.8 86.0 95.9
Middlbr 304 89.0 86.0 92.0 Swanse 404 86.6 83.8 89.5
Newc 304 89.4 86.2 92.7 Wrexm 107 88.1 83.0 93.4
Norwch 350 91.3 88.9 93.7 England 21,851 89.9 89.4 90.3
Nottm 477 88.9 86.5 91.4 N Ireland 870 90.1 88.4 91.9
Oxford 498 88.1 85.6 90.6 Scotland 2,110 89.3 88.1 90.5
Plymth 167 84.2 79.5 89.2 Wales 1,287 87.1 85.5 88.8
Ports 564 89.9 87.7 92.1 UK 26,118 89.7 89.3 90.1
smaller, the death rate was therefore more variable draw conclusions on trends in these countries. The
with very wide confidence intervals and it is difficult to change in prevalent survival by centre over the cohort
years 2002 to 2011 is shown in this chapter, appendix
Table 8.15. One-year death rate per 1,000 prevalent dialysis 1, table 8.28.
patient years in the 2011 cohort and median age of prevalent
patients by country One year survival of prevalent dialysis patients with a
England N Ireland Scotland Wales primary diagnosis of diabetes, 2002 to 2011 cohort
years
Death rate 149 155 156 207 The age-adjusted survival for patients with diabetic
95% CI 144–155 129–185 139–175 181–235
renal disease in the UK has increased slightly in the
Median age 66.1 68.6 66.1 68.1
2011 cohort year to 84.9% (table 8.17).
162
Table 8.16. One-year survival of prevalent RRT patients in the UK (unadjusted unless indicated otherwise)
Patient group Patients Deaths Survival 95% CI
Dialysis patients 2011 cohort

All 26,118 3,555 85.8 85.4–86.2
All – adjusted to age 60 26,118 3,555 89.7 89.3–90.1
2 year survival – dialysis patients
All patients alive on 31/12/2010 25,567 6,171 73.9 73.3–74.5
Dialysis patients 2011 cohort
All age ,65 12,293 897 92.2 91.6–92.6
All age 65+ 13,825 2,658 80.5 79.8–81.2
Non-diabetic ,55 6,095 246 95.6 95.1–96.1
Non-diabetic 55–64 3,673 315 90.9 89.9–91.8
Non-diabetic 65–74 4,757 650 86.0 84.9–86.9
Non-diabetic 75+ 6,265 1,454 76.7 75.6–77.7
Non-diabetic ,65 9,768 561 93.8 93.3–94.3
Diabetic ,65 2,525 336 85.9 84.4–87.2
Non-Diabetic 65+ 11,022 2,104 80.6 79.9–81.4
Diabetic 65+ 2,803 554 80.0 78.5–81.4
Cohorts of patients alive on 31/12/2011 unless indicated otherwise
Death rate on RRT compared with the UK general was in the 2010 cohort. With the reduction in rates of
population death on RRT over the last 10 years, the relative risk of
The death rate compared to the general population is death is falling (7.7 in 1998–2001 cohort, 6.1 in 2011
shown in table 8.18. Figure 8.28 shows that the relative cohort).
risk of death on RRT decreased with age from 16.6
times that of the general population at age 35–39 years
to 2.7 times the general population at age 85 and over.
Figure 8.28 also shows that the relative risk of death Results of analyses on causes of death
has decreased substantially for the younger age groups
(,50 years of age) compared to the relative risk of Data completeness
death in the 1998–2001 cohort. The relative risk of Having increased significantly in recent years, data
death was unchanged at 6.1, in the 2011 cohort as it completeness for cause of death data in the UK showed
only a marginal rise of 0.2% (table 8.19) with both
100
Northern Ireland and Scotland recording more than
Upper 95% Cl 85% of cause of death data. Northern Ireland centres
95 Unadj 1yr survival overall had the highest rate of data return for cause of
Lower 95% Cl
death (92.3%) and their cause of death completeness
90
improved by about 3% compared with the previous
Percentage survival
85 year. Patterns of cause of death must be cautiously inter-

preted, as there are significant differences between the
80
cause of death for centres with a high proportion of
75 non-returns when compared to centres with good returns
(570%). Some centres consistently achieve a very high
70
rate of data return for cause of death because a process
65 is in place to ensure that these data were entered. Several
centres have shown significant improvement in data
60
18–34 35–44 45–54 55–64 65–74 75–84 85+ returns, but unfortunately some centres that were report-
Age group ing these data in previous years have stopped reporting
Fig. 8.21. One year survival of prevalent dialysis patients by age cause of death data. There is still much variability
group, 2011 cohort between the centres regarding the completeness of
163
164
Percentage survival Percentage survival Percentage survival
75
80
85
90
95
100
60
65
70
75
80
85
90
95
100
50
55
60
65
70
75
80
85
90
95
100
West NI West NI Norwch
Dundee Stevng Ulster
Stevng Edinb B QEH
Covnt Covnt B Heart
B QEH Dudley York
Ulster Clwyd Stoke
Antrim L Rfree Middlbr
Survival
L West Abrdn Leic
Dudley Dundee Prestn
Norwch L West Derby
Lower 95% Cl
Upper 95% Cl
Carsh Chelms Kent
Hull Antrim Redng
L Guys L Guys L Barts
Donc Salford Hull
Abrdn Ipswi L West
Clwyd Carsh Sund
Redng Hull Bangor
Edinb Newc M RI
Chelms Basldn Antrim
Stoke L Kings Donc
Glouc Ulster Carlis
Bristol B QEH Dorset
Prestn M RI Bristol
M RI Liv RI L Guys
Dorset Wirral Carsh
Ipswi Donc Belfast
Wirral Colchr Dundee
Leic Bristol Wirral
L Rfree Shrew Abrdn
Derby Glouc Covnt
L Barts Bradfd Exeter
L Kings Prestn Dudley
Shrew Nottm Klmarnk
Ports Redng Ports
Bangor Ports Shrew
Belfast Dorset West NI
Klmarnk Derby Stevng
Truro Klmarnk L Kings
Brightn Stoke Sheff
Centre
Centre
Centre
Newc Norwch Glouc
Kent Leic Edinb
Colchr L Barts Sthend
Liv RI Bangor Ipswi
Middlbr Wolve Oxford
Nottm Brightn Brightn
Salford Glasgw
Camb Belfast Leeds
Sheff Inverns Truro
Carlis Wrexm Newc
Wolve L St.G Clwyd
York Carlis Wolve
Glasgw Dunfn Airdrie
L St.G Camb Glasgw
Basldn Truro L St.G
B Heart Sheff Liv RI
Fig. 8.22. One year survival of prevalent dialysis patients aged under 65 by centre, 2011 cohort
Dunfn D&Gall Swanse

Wrexm Kent Dunfn
Fig. 8.24. One year survival of prevalent dialysis patients by centre adjusted to age 60, 2011 cohort
Oxford Oxford Camb
Exeter Middlbr Chelms
Inverns Cardff Salford
Nottm
Fig. 8.23. One year survival of prevalent dialysis patients aged 65 years and over by centre, 2011 cohort
Sthend B Heart
Bradfd Sthend L Rfree
D&Gall Exeter Plymth
Leeds Leeds Bradfd
Swanse Airdrie Wrexm
Sund York Cardff
Airdrie Swanse Colchr
Cardff Sund D&Gall
Plymth Liv Ain Basldn
Newry Plymth Liv Ain
Liv Ain Newry Inverns
England England England
Survival
Survival
N Ireland N Ireland N Ireland

Scotland Scotland Scotland
Wales Wales Wales
Lower 95% Cl
Upper 95% Cl
Lower 95% Cl
Upper 95% Cl
UK UK UK
100 500
Dotted lines show 99.9% limits England
Solid lines show 95% limits N Ireland
95 400
Scotland
Percentage survival
Wales
Death rate
300
90
200
85
100
80
0
18–34 35–44 45–54 55–64 65–74 75–84 85+
75
Age group
0 150 300 450 600 750 900 1,050 1,200 1,350
Number of prevalent patients Fig. 8.26. One year death rate per 1,000 patient years by UK
Fig. 8.25. One year survival funnel plot of prevalent dialysis country and age group for prevalent dialysis patients, 2011 cohort
patients by centre adjusted to age 60, 2011 cohort
cause of death with some centres returning no data and Cause of death in prevalent RRT patients in the 2011 cohort
other centres having 100% completeness (table 8.19). Table 8.22, figures 8.29 and 8.30 show the cause of
death for both prevalent dialysis and transplant patients
Causes of death in incident RRT patients in the 2011 cohort. These data are neither age adjusted
Causes of death within the first 90 days nor adjusted for differences in the comorbidity between
See table 8.20. the two groups. Cardiac disease as a cause of death was
less common in transplanted patients as these were a
Cause of death within one year after 90 days pre-selected low risk group of patients. Malignancy and
Treatment withdrawal as a cause of death (tables 8.20, infection were both responsible for a greater percentage
8.21) in incident patients in the first 90 days and one of deaths in prevalent transplanted patients, with treat-
year after 90 days was more common in older (aged ment withdrawal a common cause of death in the
65+) patients and malignancy more common in younger prevalent dialysis population.
patients (,65 years old). Infection within the first 90 days Table 8.23 shows that malignancy and infection were
as the cause of death was more common in older patients. slightly more common in younger (,65 years) prevalent
Cardiac disease remained the leading cause of death both transplanted patients as the cause of death than in older
in the first 90 days and one year after 90 days. (565 years old) transplanted patients.
94
92
90
88
Percentage survival
86
84
82
80
78
76
England N Ireland Scotland Wales
74
72
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Cohort year
Fig. 8.27. Serial 1 year survival for prevalent dialysis patients by UK country, 2000 to 2011 cohort years, adjusted to age 60
165
Table 8.17. Serial 1 year survival of prevalent dialysis patients with a primary diagnosis of diabetes, 2002–2011 cohort years
Year 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
1 year survival % 81.6 81.7 82.8 82.4 84.7 83.5 83.9 83.3 84.8 84.9
Table 8.18. Death rate by age group for all prevalent RRT patients, 2011 cohort, compared with the general population and with
previous analyses in the 1998–2001 cohort
UK Expected UKRR death Relative Relative

population UK Death rate number of UKRR rate per 1,000 risk of risk of death
Age mid 2012 deaths in per 1,000 deaths in UKRR deaths in prevalent death in 1998–2001
group (thousands) 2012 population population 2012 RRT patients 2012 cohort
20–24 4,332 1,550 0.4 0 10 10 28.8 41.1

25–29 4,318 1,982 0.5 1 18 12 25.1 41.8
30–34 4,240 2,661 0.6 1 18 9 13.7 31.2
35–39 4,036 3,690 0.9 3 43 15 16.6 26.0
40–44 4,567 6,315 1.4 6 89 21 15.5 22.6
45–49 4,686 9,690 2.1 11 141 27 13.0 19.0
50–54 4,236 13,384 3.2 17 226 41 13.0 12.8
55–59 3,684 18,736 5.1 27 284 53 10.4 10.1
60–64 3,624 29,012 8.0 44 437 79 9.8 10.4
65–69 3,345 41,101 12.3 64 553 107 8.7 7.9
70–74 2,476 51,932 21.0 96 682 149 7.1 7.2
75–79 2,047 71,835 35.1 132 792 211 6.0 5.3
80–84 1,534 96,291 62.8 149 652 275 4.4 4.0
85+ 1,439 215,351 149.7 166 452 408 2.7 3.0
Total 48,564 563,530 11.6 717 4,397 87 6.1 7.7
Table 8.24 shows the cause of death for prevalent ‘other’ cause of death has increased, as has treatment
dialysis patients in the 2011 cohort. Prevalent dialysis withdrawal (19% in 2011 cohort). Infection as cause of
patients aged 65 years and over were substantially more death remained at a similar level to the 2000 cohort
likely to withdraw from treatment than younger patients (figure 8.31).
and cardiac disease was much more common as a cause
of death in younger (,65 years) dialysis patients.
Figure 8.31 shows cause of death for prevalent patients
in the 2000 to 2011 cohort. Over time, cardiac disease Median life expectancy on RRT
as cause of death has decreased markedly and there has
been a gradual decline in cerebrovascular disease as a The statistical methodology for this analysis is
cause of death. The proportion of patients coded with described in the methodology section at the start of this
50
Relative risk 1998–2001
Relative risk in 2011
40
Risk of death
30
20
10
0
20–24
25–29
30–34
35–39
40–44
45–49
50–54
55–59
60–64
65–69
70–74
75–79
80–84
85+
Fig. 8.28. Relative risk of death in all

prevalent RRT patients in the 2011 cohort
Age group compared with the UK general population
166
Table 8.19. Percentage completeness of EDTA cause of death for prevalent patients by centre and year
Centre 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
England
B Heart 76.3 76.4 68.1 85.7 84.5 93.9 100.0 96.6 96.1 96.6
B QEH 0.0 60.2 4.8 5.1 3.5 0.7 1.2 2.0 2.1
Basldn 92.3 84.0 45.0 22.7 45.5 47.6 80.0 68.8 84.6 88.9
Bradfd 88.1 83.3 87.8 90.0 88.2 92.5 79.5 97.0 97.6 97.7
Brightn 0.0 0.0 0.0 12.0 0.0 1.1 2.4 1.1 1.1
Bristol 85.0 89.9 76.7 60.2 58.7 65.8 70.0 89.4 95.2 82.2
Camb 0.0 1.6 1.5 1.3 0.0 0.0 5.0 10.3 62.0 94.1
Carlis 60.0 77.3 87.0 91.3 73.9 47.6 80.6 100.0 92.9 94.7
Carsh 0.0 0.0 0.0 0.0 0.8 0.8 0.8 6.7 25.0 40.8
Chelms 35.0 69.7 64.0 76.5 71.4 86.7 86.7 87.0 100.0
Colchr 0.0 50.0 77.3 82.6 100.0
Covnt 3.0 1.7 0.0 0.0 0.0 1.2 0.0 0.0 1.4 33.3
Derby 11.1 69.0 77.6 75.6 83.3 97.8 73.5 91.2 88.5 85.2
Donc 100.0 94.3 90.9 91.7 92.6
Dorset 0.0 30.6 61.5 66.7 87.2 88.9 85.2 95.7 94.9 88.9
Dudley 3.4 31.7 14.3 5.9 6.3 5.3 0.0 94.3 88.1 90.9
Exeter 35.1 40.8 34.7 17.5 4.7 2.1 3.0 89.5 84.6 95.1
Glouc 63.0 43.2 51.6 44.4 55.6 60.4 65.8 97.2 93.6 91.5
Hull 38.9 83.6 81.5 76.0 76.5 51.6 17.3 90.8 93.5 96.9
Ipswi 47.1 30.4 10.3 21.9 35.5 13.6 18.8 70.0 77.8 77.4
Kent 56.8 89.2 89.0 96.2 94.9
L Barts 86.5 83.3 87.4 74.6 77.0 70.1 74.6 82.6 79.9
L Guys 1.2 0.0 0.0 0.0 3.5 0.0 0.0 69.5 84.2 58.8
L Kings 31.5 66.7 85.7 90.6 75.6 88.2 67.1 96.1 97.6 100.0
L Rfree 0.0 0.0 0.0 0.9 1.7 0.0 7.0
L St.G 16.7 17.9 21.4 77.6 47.9 42.4
L West 79.1 67.5 79.8 31.3 18.9 5.8 2.2 2.2 95.0 96.8
Leeds 58.6 73.8 67.2 66.7 29.6 27.9 33.6 99.0 99.1 97.7
Leic 77.0 88.2 71.5 77.0 65.5 69.5 69.3 74.5 60.9 94.1
Liv Ain 100.0 66.7 50.0 81.3 73.3 66.7 100.0 85.0 95.7 0.0
Liv RI 74.1 69.9 39.8 65.5 76.8 75.6 79.2 71.6 76.4 2.8
M RI 4.0 0.9 1.0 4.7 3.1 9.9
Middlbr 66.7 42.0 77.6 63.5 54.8 23.4 46.7 88.2 97.5 94.9
Newc 29.9 27.1 19.4 29.8 48.7 35.7 40.8 14.0 45.0 16.9
Norwch 30.8 21.0 21.4 18.2 21.2 44.4 75.8 70.3 76.1
Nottm 90.6 94.4 97.0 87.5 87.0 98.8 97.1 98.8 100.0 99.0
Oxford 8.7 1.9 2.8 0.0 0.0 1.0 0.0 84.6 97.4 92.7
Plymth 52.8 46.9 43.2 39.6 56.7 70.7 47.5 78.7 43.6 41.2
Ports 32.7 55.1 21.5 7.3 17.5 5.9 43.6 67.0 23.3 19.8
Prestn 73.8 75.9 50.0 55.4 47.8 38.1 17.9 95.7 98.9 97.6
Redng 86.0 77.1 81.5 77.1 97.8 89.6 83.0 100.0 96.7 91.2
Salford 1.7 1.3 0.0 0.0 1.3 0.0 1.3 0.0 0.0 0.0
Sheff 98.8 19.6 3.1 5.5 8.1 0.9 1.9 3.0 0.8 0.8
Shrew 25.0 66.7 53.1 85.7 62.5 20.5 46.0 0.0 7.9
Stevng 71.0 66.2 75.0 57.5 52.2 60.3 70.0 86.3 86.8 67.7
Sthend 66.7 25.0 41.2 9.4 3.2 57.7 75.0 92.3 90.0 100.0
Stoke 16.1 21.0 28.6 53.9 57.9 89.6
Sund 53.1 54.8 56.3 60.0 60.5 50.0 78.9 93.5 95.1 97.4
Truro 80.6 57.1 2.3 6.9 0.0 18.4 26.3 93.3 94.9 78.8
Wirral 85.7 64.5 31.3 88.2 68.4 87.5 24.2 62.2 0.0 2.7
Wolve 98.5 96.6 92.2 48.5 52.3 65.8 76.4 96.9 94.1 90.9
York 82.5 67.6 41.4 83.3 38.5 62.1 64.3 96.6 97.3 100.0
N Ireland
Antrim 4.3 10.0 8.8 3.8 26.9 100.0 100.0 100.0
Belfast 17.2 33.8 38.3 20.0 26.2 81.4 80.0 79.7
Newry 0.0 42.9 16.7 15.4 85.7 95.2 100.0 96.7
Ulster 100.0 85.7 92.9 90.0 75.0 95.0 95.2 100.0
West NI 46.2 57.7 38.9 25.0 45.8 100.0 87.0 100.0
167

Centre 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Scotland
Abrdn 47.7 31.7 2.8 0.0 0.0 82.9 97.6 92.1 97.6 65.7
Airdrie 26.7 10.3 40.0 26.3 26.8 79.3 100.0 96.8 97.0 93.9
D & Gall 69.2 76.9 80.0 76.9 100.0 93.3 94.1 100.0 100.0 81.3
Dundee 92.1 92.1 86.1 2.8 0.0 50.0 90.6 85.7 59.5 62.2
Dunfn 80.0 66.7 81.3 50.0 53.8 61.9 89.3 78.6 90.0 87.5
Edinb 60.4 44.2 50.9 29.3 45.0 85.9 96.2 98.3 95.1 100.0
Glasgw 49.6 41.9 40.2 53.2 55.3 75.4 88.0 66.9 98.5 96.0
Inverns 0.0 0.0 0.0 0.0 0.0 65.2 90.0 91.7 100.0 95.7
Klmarnk 4.0 10.0 0.0 11.1 9.4 95.8 93.3 93.9 94.4 96.8
Wales 34.1 30.7 28.6 30.0 43.4 36.4 47.2 53.0 48.6 50.3
Bangor 39.1 42.1 66.7 35.0 86.2 52.4 76.9 73.9 90.0 100.0
Cardff 3.5 2.6 3.5 2.2 4.1 0.0 1.6 6.0 7.9 0.6
Clwyd 22.2 0.0 0.0 11.1 45.5 84.2 83.3 100.0 85.7 89.5
Swanse 92.0 89.2 85.7 92.4 97.3 94.8 89.8 98.0 87.5 97.1
Wrexm 10.7 3.7 3.7 0.0 22.7 69.2 100.0 95.7 92.6 100.0
England 52.3 51.8 46.8 40.8 36.8 36.0 37.8 58.3 63.4 64.3
N Ireland 20.4 38.7 33.6 22.4 42.1 91.5 89.0 92.3
Scotland 50.5 42.5 40.3 32.3 33.5 75.2 92.5 83.8 93.1 89.1
Wales 34.1 30.7 28.6 30.0 43.4 36.4 47.2 53.0 48.6 50.3
UK 50.5 49.2 44.2 39.2 36.8 39.3 43.4 61.2 66.1 66.3
Blank cells denote data not available for that year
Table 8.20. Cause of death in the first 90 days for incident patients by age group, 2000–2011 cohort
Cardiac disease 644 27 152 29 492 26
Infection 416 17 76 14 340 18
Malignancy 216 9 65 12 151 8
Other 554 23 138 26 416 22
Uncertain 95 4 16 3 79 4
Total 2,412 525 1,887
No cause of death data 2,537 51 555 51 1,982 51
Table 8.21. Cause of death in 1 year after 90 days for incident patients by age group, 2000–2011 cohort
Cardiac disease 1,000 23 316 26 684 22
Infection 804 18 226 18 578 18
Malignancy 460 10 155 13 305 10
Other 934 21 291 24 643 20
Uncertain 232 5 73 6 159 5
Total 4,390 1,225 3,165
No cause of death data 4,430 50.2 1,255 50.6 3,175 50.1
168
Table 8.22. Cause of death in prevalent RRT patients by modality, 2011 cohort
All modalities Dialysis Transplant
Infection 532 18 437 17 95 23
Malignancy 292 10 208 8 84 20
Other 624 21 528 21 96 23
Uncertain 245 8 212 8 33 8
Total 2,986 2,576 410
No cause of death data 1,414 32 1,160 31 254 38
Uncertain Uncertain
8% 8% Cardiac disease
Cardiac disease 18%
23%
Other Other
20% 23% Cerebrovascular
disease
4%
Cerebrovascular
disease
5%
Treatment withdrawal
3%
Infection Infection
Treatment withdrawal 17% 24%
19%
Malignancy Malignancy
8% 20%
Fig. 8.29. Percentage contribution to cause of death for prevalent Fig. 8.30. Percentage contribution to cause of death for prevalent
dialysis patients, 2011 cohort transplant patients, 2011 cohort
chapter. Figure 8.32 shows median life expectancy on be different for low risk patients (e.g. polycystic kidney
RRT after 90 days by age group. All incident patients disease with a transplant) vs. high risk patients (diabetes
starting RRT from 2000 to 2009 have been included in with previous myocardial infarction on dialysis) even
this analysis and patients were followed up for a mini- within the same age group. Median life years remaining
mum of three years. The estimated median survival will for non-diabetic and diabetic patients (figure 8.33) were
Table 8.23. Cause of death in prevalent transplanted patients by age group, 2011 cohort
Infection 95 23 48 24 47 22
Malignancy 84 20 42 21 42 20
Other 96 23 43 22 53 25
Uncertain 33 8 16 8 17 8
Total 410 198 212
No cause of death data 254 38 126 39 128 38
169
Table 8.24. Cause of death in prevalent dialysis patients by age group, 2011 cohort
Cardiac disease 575 22 172 28 403 21
Infection 437 17 105 17 332 17
Malignancy 208 8 45 7 163 8
Other 528 21 143 23 385 20
Uncertain 212 8 58 9 154 8
Total 2,576 614 1,962
No cause of death data 1,160 31 331 35 829 30
Cardiac disease
35
Infection
Uncertain
Treatment withdrawal
30 Other
Malignancy
Cerebrovascular disease
25
Percentage
20
15
10
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Cause of death
Fig. 8.31. Cause of death in prevalent RRT patients by cohort year
25 25
20 20
Years remaining
Years remaining
15 15
10 10
5 5
0 0
20–24
25–29
30–34
35–39
40–44
45–49
50–54
55–59
60–64
65–69
70–74
75+
20–24
25–29
30–34
35–39
40–44
45–49
50–54
55–59
60–64
65–69
70–74
75+
Age group Age group
Fig. 8.32. Median life expectancy on RRT after 90 days, by age Fig. 8.33. Median life expectancy on RRT after 90 days by age
group, incident patients starting RRT from 2000–2009 group, incident diabetic patients starting RRT from 2000–2009
170
also calculated and show that median life expectancy for group (565 years) the median life years remaining
patients younger than 45 was on average nine years more were similar between diabetic and non-diabetic patients.
for non-diabetic patients (data not shown) compared
with age matched diabetic patients. In the older age Conflicts of interest: none
References
1 Plantinga LC, Fink NE, Levin NW, et al. Early, Intermediate, and Long- frequency of incorrect reporting of date of start of RRT. Nephron Clini-
Term Risk Factors for Mortality in Incident Dialysis Patients: The cal Practice;115(suppl 1):c271–c78
Choices for Healthy Outcomes in Caring for ESRD (CHOICE) Study. 7 Malek SK, Keys BJ, Kumar S, Milford E, Tullius SG. Racial and
American journal of kidney diseases: the official journal of the National ethnic disparities in kidney transplantation. Transplant International
Kidney Foundation 2007;49(6):831–40 2011;24(5):419–24 doi: 10.1111/j.1432–2277.2010.01205.x[published
2 Miskulin DC, Meyer KB, Martin AA, et al. Comorbidity and its change Online First: Epub Date]
predict survival in incident dialysis patients. American journal of kidney 8 Office for National Statistics. www.ons.gov.uk, http://www.ons.gov.uk/
diseases: the official journal of the National Kidney Foundation ons/dcp171778_238743.pdf
2003;41(1):149–61 9 Ansell D, Roderick P, Hodsman A, Ford D, Steenkamp R, Tomson C.
3 Nitsch D, Burden R, Steenkamp R, Ansell D, Byrne C, Caskey F, et al. UK Renal Registry 11th Annual Report (December 2008): Chapter 7
Patients with diabetic nephropathy on renal replacement therapy in Survival and cause of death of UK adult patients on renal replacement
England and Wales. Qjm-an International Journal of Medicine. 2007 therapy in 2007: national and centre-specific analyses. Nephron Clin
Sep;100(9):551–60 Pract. 2009;111(suppl 1):c113–39
4 Roderick P, Byrne C, Casula A, Steenkamp R, Ansell D, Burden R, et al. 10 van Manen JG, van Dijk PCW, Stel VS, Dekker FW, Cleries M, Conte F,
Survival of patients from South Asian and Black populations starting et al. Confounding effect of comorbidity in survival studies in patients on
renal replacement therapy in England and Wales. Nephrology Dialysis renal replacement therapy. Nephrology Dialysis Transplantation. 2007;
Transplantation. 2009 Dec;24(12):3774–82 22(1):187–95.
5 Tomson C, Maggs C. UK Renal Registry 12th Annual Report (December 11 Renal Association. Clinical Practice Guidelines. 5th edition. 2010; http://
2009): Chapter 2: introduction. Nephron Clin Pract. 2010;115(suppl 1): www.renal.org/Clinical/GuidelinesSection/Guidelines.aspx
c3–8
6 Ford DJ, Fogarty DG, Steenkamp R, Tomson CRV, Ben-Shlomo Y,
Ansell D. Chapter 13: The UK Renal Registry Advanced CKD Study:
171
Appendix 1: Survival tables

Table 8.25. One-year after 90–day incident survival percentage by centre, 2011 cohort, unadjusted and adjusted to age 60
Unadjusted Adjusted Adjusted Unadjusted Adjusted Adjusted

1 year after 1 year after 1 year after 1 year after 1 year after 1 year after
90 days 90 days 90 days 90 days 90 days 90 days
Centre survival survival 95% CI Centre survival survival 95% CI
England Ports 87.7 91.2 87.6–95.0

B Heart 92.0 94.4 90.7–98.2 Prestn 89.9 91.8 87.8–96.0
B QEH 91.2 93.3 90.1–96.5 Redng 90.4 93.0 88.5–97.8
Basldn 85.7 91.0 83.9–98.7 Salford 90.1 91.7 87.4–96.1
Bradfd 87.0 88.9 81.6–96.9 Sheff 83.8 87.6 82.8–92.6
Brightn 86.7 91.0 86.4–95.8 Shrew 86.7 91.9 86.6–97.4
Bristol 92.0 94.5 91.4–97.7 Stevng 88.5 91.1 86.6–95.8
Camb 86.4 91.6 87.6–95.8 Sthend 89.3 94.3 88.3–100.0
Carlis 88.5 91.5 82.9–100.0 Stoke 88.9 93.1 88.6–97.8
Carsh 90.8 94.3 91.7–97.0 Sund 88.2 88.7 79.9–98.4
Chelms 75.6 80.8 71.4–91.6 Truro 89.9 93.0 86.7–99.8
Colchr 72.5 84.1 75.5–93.6 Wirral 83.5 87.9 81.2–95.2
Covnt 88.4 90.4 85.1–95.9 Wolve 83.7 89.3 84.2–94.8
Derby 87.8 91.3 86.1–96.9 York 92.5 93.6 87.0–100.0
Donc 83.8 88.9 81.1–97.5 N Ireland
Dorset 82.4 88.2 82.4–94.4 Antrim 78.3 86.3 76.1–97.9
Dudley 90.0 93.7 88.1–99.7 Belfast 89.9 92.8 87.9–98.1
Exeter 82.9 88.5 83.7–93.5 Newry 83.3 88.1 79.1–98.2
Glouc 80.6 89.6 84.2–95.4 Ulster 79.4 86.3 77.5–96.1
Hull 89.7 93.3 89.4–97.4 West NI 94.3 95.9 90.7–100.0
Ipswi 94.6 95.5 89.7–100.0 Scotland
Kent 84.3 88.5 83.6–93.7 Abrdn 89.8 92.8 87.0–99.0
L Barts 92.9 93.7 90.7–96.9 Airdrie 81.3 84.1 74.1–95.4
L Guys 93.2 94.7 91.2–98.3 Dundee 83.8 90.3 84.5–96.5
L Kings 88.0 90.9 86.9–95.1 Dunfn 90.2 92.4 85.6–99.7
L Rfree 89.1 91.0 87.4–94.6 Edinb 89.3 90.2 83.6–97.3
L St.G 95.9 96.8 93.3–100.0 Glasgw 85.1 88.6 84.3–93.1
L West 88.5 90.7 87.9–93.5 Klmarnk 88.0 91.1 82.1–100.0
Leeds 85.0 88.2 83.6–93.0 Wales
Leic 88.1 91.3 88.3–94.3 Bangor 92.1 94.3 87.1–100.0
Liv Ain 81.8 86.8 79.5–94.7 Cardff 83.3 88.1 84.2–92.2
Liv RI 87.8 89.0 83.3–95.0 Swanse 79.2 85.4 80.0–91.2
M RI 92.1 93.3 89.6–97.2 Wrexm 83.3 88.8 79.3–99.6
Middlbr 84.6 88.9 83.4–94.8 England 88.0 91.1 90.3–92.0
Newc 84.0 86.0 79.2–93.3 N Ireland 86.4 90.5 87.1–94.1
Norwch 84.0 89.4 83.9–95.3 Scotland 86.8 90.1 87.7–92.6
Nottm 87.9 92.8 89.1–96.7 Wales 82.6 87.7 84.8–90.7
Oxford 85.9 88.8 84.4–93.3 UK 87.5 90.9 90.0–91.7
Plymth 88.2 91.3 85.0–98.1
Excluded: centres with less than 20 patients (Clwyd, Dumfries & Galloway, Inverness)
172
Table 8.26. Ninety day incident survival percentage by centre, 2011 cohort, unadjusted and adjusted to age 60
Unadjusted Adjusted Adjusted Unadjusted Adjusted Adjusted

Centre 90 day survival 90 day survival 90 day 95% CI Centre 90 day survival 90 day survival 90 day 95% CI
England Ports 95.4 97.0 94.9–99.1

B Heart 97.1 98.2 96.1–100.0 Prestn 97.9 98.4 96.7–100.0
B QEH 95.9 97.0 95.0–99.1 Redng 91.3 94.2 90.4–98.2
Basldn 97.2 98.4 95.4–100.0 Salford 94.3 95.5 92.6–98.6
Bradfd 93.2 94.5 89.4–99.9 Sheff 93.8 95.8 93.1–98.5
Brightn 89.2 93.4 89.8–97.1 Shrew 90.9 95.0 91.1–99.0
Bristol 96.5 97.8 95.8–99.7 Stevng 99.1 99.4 98.1–100.0
Camb 96.0 97.7 95.6–99.7 Sthend 96.6 98.3 95.0–100.0
Carlis 96.4 97.5 92.8–100.0 Stoke 92.3 95.5 92.1–99.1
Carsh 93.6 96.3 94.4–98.4 Sund 95.7 96.0 90.9–100.0
Chelms 90.0 93.6 88.3–99.1 Truro 91.1 94.3 89.2–99.8
Colchr 97.4 98.7 96.2–100.0 Wirral 91.0 94.0 89.4–98.8
Covnt 88.9 91.9 87.5–96.4 Wolve 93.5 96.1 93.2–99.2
Derby 89.2 92.8 88.3–97.5 York 95.2 96.3 91.4–100.0
Donc 92.5 95.5 90.8–100.0 N Ireland
Dorset 93.7 96.2 93.0–99.5 Antrim 92.0 95.4 89.6–100.0
Dudley 93.0 96.2 92.0–100.0 Belfast 98.6 99.1 97.4–100.0
Exeter 94.9 96.9 94.5–99.4 Newry 96.8 98.0 94.4–100.0
Glouc 95.4 97.8 95.4–100.0 Ulster 94.4 96.6 92.1–100.0
Hull 92.4 95.4 92.3–98.6 West NI 97.2 98.2 94.8–100.0
Kent 94.3 96.1 93.3–99.0 Scotland
L Barts 97.7 98.1 96.5–99.8 Abrdn 94.2 96.2 92.0–100.0
L Guys 97.6 98.2 96.3–100.0 Airdrie 97.6 98.0 94.4–100.0
L Kings 97.3 98.1 96.2–100.0 Dundee 93.3 96.3 92.8–99.9
L Rfree 96.2 97.2 95.3–99.1 Dunfn 89.1 91.9 85.5–98.9
L St.G 96.1 97.0 93.8–100.0 Edinb 95.7 96.4 92.4–100.0
L West 96.6 97.5 96.1–98.9 Glasgw 93.4 95.3 92.7–98.1
Leeds 93.6 95.4 92.6–98.2 Klmarnk 78.1 85.5 76.3–95.9
Leic 94.6 96.3 94.5–98.2 Wales
Liv Ain 87.5 92.3 87.3–97.6 Cardff 96.9 98.0 96.4–99.6
Liv RI 91.7 93.3 89.2–97.6 Swanse 94.2 96.6 94.1–99.1
M RI 94.6 95.8 93.0–98.7 Wrexm 92.3 95.1 88.7–100.0
Middlbr 93.4 95.8 92.5–99.1 England 94.4 96.3 95.7–96.8
Newc 90.2 92.2 87.4–97.2 N Ireland 96.5 97.8 96.2–99.4
Norwch 89.4 93.6 89.6–97.8 Scotland 93.1 95.3 93.7–96.9
Nottm 90.2 94.5 91.4–97.6 Wales 95.8 97.4 96.1–98.7
Oxford 94.6 96.1 93.6–98.6 UK 94.5 96.3 95.8–96.8
Plymth 96.2 97.4 94.0–100.0
Excluded: centres with less than 20 patients (Clwyd, Dumfries & Galloway, Inverness) and centres with no deaths recorded in the first 90 days of
RRT (Ipswich and Bangor)
173
Table 8.27. One year after 90-day incident survival by centre for incident cohort years 2003–2011 adjusted to age 60
Centre 2003 2004 2005 2006 2007 2008 2009 2010 2011
England
B Heart 88.2 86.4 83.6 88.5 93.5 93.6 84.3 92.0 94.4
B QEH 88.0 90.4 86.9 92.9 89.8 92.2 88.3 93.3
Basldn 92.6 92.3 92.8 90.9 89.9 89.3 88.5 84.8 91.0
Bradfd 88.3 80.9 86.1 80.8 84.2 84.4 92.4 87.6 88.9
Brightn 90.6 84.3 87.2 94.2 89.3 84.7 88.3 91.0
Bristol 85.7 88.0 82.8 92.6 91.4 84.0 88.7 88.9 94.5
Camb 89.4 86.9 89.8 90.9 93.4 91.2 87.7 89.5 91.6
Carlis 82.5 86.9 79.5 89.9 96.5 87.8 71.5 86.3 91.5
Carsh 89.4 85.8 90.2 88.7 87.2 86.6 88.0 89.8 94.3
Chelms 82.2 82.8 94.3 86.6 90.8 93.4 85.6 80.8
Colchr 86.6 84.6 96.8 84.1
Covnt 81.8 87.6 82.5 88.6 90.4 86.9 94.2 89.0 90.4
Derby 86.5 83.7 87.9 93.1 96.6 90.5 87.6 87.4 91.3
Donc 89.8 84.6 91.5 88.9
Dorset 85.9 91.3 82.5 86.3 90.4 93.5 92.7 87.4 88.2
Dudley 90.5 81.3 97.3 92.7 85.6 70.3 84.6 87.8 93.7
Exeter 82.3 88.5 86.1 88.9 86.4 87.0 88.5 95.3 88.5
Glouc 82.9 83.4 95.1 89.7 87.0 94.3 90.1 92.3 89.6
Hull 89.3 88.8 85.7 93.6 89.8 85.4 88.9 88.0 93.3
Ipswi 93.2 97.4 84.4 93.9 96.0 95.8 91.3 93.2 95.5
Kent 91.8 90.0 89.3 90.6 88.5
L Barts 87.1 91.0 94.0 86.5 93.1 90.1 91.9 93.7
L Guys 94.8 91.6 90.4 92.9 92.0 90.5 95.0 91.4 94.7
L Kings 88.0 86.9 91.8 86.5 87.9 89.7 86.3 89.7 90.9
L Rfree 93.3 89.8 94.4 95.2 88.6 90.3 91.0
L St.G 92.1 94.0 92.2 93.7 96.8
L West 95.9 92.4 94.4 92.8 92.9 94.5 93.8 88.8 90.7
Leeds 87.1 89.6 89.9 85.7 87.4 88.7 89.9 92.7 88.2
Leic 89.0 87.5 84.6 87.9 89.8 90.5 90.2 91.6 91.3
Liv Ain 87.0 82.9 78.6 82.5 89.1 86.8
Liv RI 90.2 80.9 90.1 86.7 86.2 94.1 94.4 88.5 89.0
M RI 90.2 87.8 87.6 89.5 93.3
Middlbr 82.4 85.3 83.3 91.5 87.8 82.3 87.9 88.1 88.9
Newc 87.2 85.4 82.1 86.3 85.8 91.5 84.5 88.8 86.0
Norwch 84.0 90.7 86.4 91.1 89.0 92.0 92.1 89.4
Nottm 85.9 85.6 86.9 92.0 90.0 91.1 88.6 93.5 92.8
Oxford 89.4 87.8 87.8 90.2 89.3 87.1 91.0 90.6 88.8
Plymth 84.0 77.7 84.5 81.2 90.1 87.8 89.9 93.8 91.3
Ports 89.8 88.4 82.4 87.6 88.7 88.8 88.9 88.1 91.2
Prestn 85.2 87.2 88.5 83.7 91.4 82.1 86.8 87.6 91.8
Redng 92.1 90.7 90.5 91.3 90.7 95.2 89.5 92.9 93.0
Salford 88.4 85.1 89.0 90.6 89.2 86.0 88.3 86.7 91.7
Sheff 87.5 91.7 90.6 88.7 90.9 92.5 93.7 92.2 87.6
Shrew 87.4 86.2 87.8 91.8 93.0 83.6 86.9 91.9
Stevng 93.8 93.3 76.7 85.4 90.7 90.2 96.3 93.8 91.1
Sthend 91.8 90.4 91.1 94.9 91.8 86.5 91.2 83.0 94.3
Stoke 87.4 89.9 85.5 87.0 93.1
Sund 80.6 86.7 80.5 83.6 88.7 85.3 79.9 84.1 88.7
Truro 86.9 92.7 90.6 89.6 90.2 89.2 93.9 90.8 93.0
Wirral 96.6 85.5 86.9 86.0 88.9 90.4 83.9 93.0 87.9
Wolve 83.6 88.0 84.1 89.3 89.5 89.1 90.3 87.5 89.3
York 76.1 91.2 83.9 82.5 95.1 86.2 93.9 86.3 93.6
174

Centre 2003 2004 2005 2006 2007 2008 2009 2010 2011
N Ireland
Antrim 87.3 94.0 86.9 92.2 97.2 90.1 86.3
Belfast 86.8 93.2 91.0 88.4 90.4 89.3 92.8
Newry 90.1 92.0 88.1
Ulster 90.9 86.3
West NI 90.2 97.3 93.1 97.5 91.3 95.9
Scotland
Abrdn 86.0 88.7 84.1 82.7 86.0 86.4 89.2 85.4 92.8
Airdrie 74.6 86.3 75.1 80.7 76.7 88.3 94.0 81.9 84.1
D & Gall 84.5 87.5
Dundee 86.9 85.7 84.8 89.5 82.0 86.2 87.4 90.2 90.3
Dunfn 88.2 89.8 78.2 80.3 87.4 87.0 89.9 93.5 92.4
Edinb 86.7 79.4 83.2 88.8 90.0 84.2 84.2 86.3 90.2
Glasgw 87.4 80.9 86.2 83.4 88.0 84.2 87.8 86.8 88.6
Inverns 87.6 89.2 84.2 83.9 90.6 87.2 96.7
Klmarnk 83.7 87.4 96.3 82.8 87.6 90.1 82.9 88.3 91.1
Wales
Bangor 91.1 80.8 82.2 81.5 92.3 87.6 87.1 89.1 94.3
Cardff 87.2 85.6 87.2 87.5 84.5 83.6 89.6 89.7 88.1
Clwyd 75.3 96.9 92.1
Swanse 84.6 78.0 83.0 84.3 89.0 85.2 83.5 86.9 85.4
Wrexm 93.4 79.7 97.7 85.6 89.9 82.0 88.8
England 88.4 87.8 87.9 89.1 90.3 89.6 89.6 89.9 91.1
N Ireland 88.9 91.7 90.9 88.9 92.0 90.3 90.5
Scotland 86.0 84.7 84.5 84.6 86.6 86.0 86.7 87.8 90.1
Wales 87.0 82.8 86.0 86.4 86.8 84.6 87.9 88.7 87.7
UK 88.0 87.2 87.4 88.6 89.7 89.0 89.4 89.7 90.9
Blank cells: centres with less than 20 patients for that year or centres with no data available for that year
175
Table 8.28. One year prevalent patient survival by centre for prevalent cohort years 2002–2011, adjusted to age 60
Centre 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
England
B Heart 87.9 86.8 88.1 86.5 87.1 90.1 90.7 87.4 89.4 88.3
B QEH 99.7 89.1 89.1 88.4 88.5 88.4 90.2 89.5 91.2 91.7
Basldn 84.6 87.9 90.4 90.2 90.5 92.7 91.8 88.8 91.1 88.4
Bradfd 83.2 88.2 86.3 82.8 84.2 87.8 84.6 89.3 88.0 87.7
Brightn 99.7 87.1 84.3 87.6 87.4 89.0 87.5 90.1 88.4 89.4
Bristol 89.0 86.8 87.4 87.6 89.1 87.3 84.9 85.7 89.6 90.6
Camb 87.3 88.1 87.4 89.4 88.0 92.6 90.0 91.4 93.1 88.9
Carlis 83.4 82.9 83.7 83.8 85.7 86.9 80.2 80.4 93.2 88.8
Carsh 84.6 87.4 86.3 89.4 88.7 90.1 89.0 89.5 89.8 91.2
Chelms 98.4 86.4 82.9 85.6 87.5 85.0 86.0 89.5 84.1 90.7
Colchr 91.0 86.5 88.9 89.1
Covnt 87.0 89.0 89.1 85.1 87.0 87.1 90.8 90.0 90.9 91.7
Derby 86.7 88.7 87.9 88.8 87.2 90.7 90.8 90.3 90.2 90.1
Donc 88.7 83.8 88.8 91.8 91.1
Dorset 90.3 88.3 89.4 87.0 87.7 89.8 90.0 93.0 89.9 90.4
Dudley 85.0 86.4 85.9 87.2 87.2 88.7 88.6 90.7 87.6 91.4
Exeter 86.9 86.2 83.7 90.9 87.1 85.3 85.3 86.5 88.2 88.0
Glouc 83.5 88.8 88.1 91.1 88.2 86.1 91.7 92.1 89.5 90.6
Hull 86.0 86.2 84.6 85.9 90.0 86.9 88.0 87.5 90.0 91.1
Ipswi 84.8 90.2 86.0 84.5 86.5 92.7 84.8 87.8 92.0 90.4
Kent 86.2 87.9 90.5 89.8 89.3
L Barts 83.8 85.7 88.3 89.2 88.8 90.9 92.9 91.7 90.0
L Guys 88.8 88.5 89.3 87.4 90.5 90.3 91.3 91.0 93.9 91.1
L Kings 77.7 81.1 86.7 89.2 84.9 88.0 88.0 89.4 90.1 89.9
L Rfree 90.2 90.4 90.3 91.3 89.8 90.3 91.6 90.2
L St.G 95.8 94.3 89.2 90.8 91.9 88.5
L West 91.3 91.0 91.1 91.1 91.4 90.1 91.9 90.3 90.4 91.5
Leeds 86.3 85.9 89.1 88.7 88.3 87.4 88.9 90.9 88.8 86.7
Leic 83.8 85.2 86.7 84.4 89.7 89.6 88.6 90.4 89.8 90.2
Liv Ain 91.5 88.0 97.2 87.2 90.7 88.5 92.0 89.9 89.7 83.8
Liv RI 84.4 85.7 84.2 88.0 85.0 86.9 89.5 89.3 90.8 89.0
M RI 86.3 86.4 87.5 86.8 88.4 90.5
Middlbr 84.6 83.6 86.2 85.4 87.4 87.0 86.6 83.7 93.1 89.0
Newc 81.0 81.0 86.1 83.9 86.1 86.4 87.2 86.3 85.2 89.4
Norwch 87.3 88.3 90.2 87.5 91.0 89.4 89.8 91.2 91.3
Nottm 85.3 86.7 84.7 83.4 89.5 88.4 87.9 89.7 90.1 88.9
Oxford 87.0 88.3 87.3 87.2 86.8 87.8 88.6 87.4 88.0 88.1
Plymth 84.7 85.7 87.6 83.5 82.5 87.8 85.6 85.0 89.7 84.2
Ports 82.1 89.1 85.9 85.2 89.8 88.4 89.2 88.3 88.2 89.9
Prestn 84.8 85.6 85.8 86.3 90.7 90.1 89.7 90.1 88.1 90.6
Redng 82.7 89.2 86.2 89.0 90.6 88.8 92.3 88.8 89.3 90.8
Salford 84.4 81.8 83.6 85.9 88.0 86.5 87.9 85.2 87.7 88.9
Sheff 91.1 87.8 87.0 89.2 88.8 88.8 89.7 89.6 88.7 88.8
Shrew 94.5 84.7 86.3 86.6 89.1 88.9 87.9 85.9 87.4 89.9
Stevng 88.6 89.5 88.7 89.5 89.7 92.4 90.4 89.9 92.7 91.9
Sthend 87.3 88.5 87.0 83.4 86.3 90.2 91.0 92.4 90.3 87.8
Stoke 84.5 87.3 88.5 86.8 90.9 90.6
Sund 75.5 81.8 86.4 79.4 83.7 87.5 85.2 84.7 83.7 86.4
Truro 90.3 89.9 85.1 91.8 89.3 89.4 89.0 90.7 89.0 89.6
Wirral 83.5 87.4 89.4 88.5 88.1 89.6 90.2 88.6 90.7 90.4
Wolve 85.0 87.6 86.8 89.3 87.8 92.8 89.4 87.4 89.3 88.6
York 81.1 83.0 89.4 84.0 88.5 87.9 88.8 90.0 84.1 88.6
176

Centre 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
N Ireland
Antrim 83.5 92.2 86.0 89.5 90.7 89.8 92.8 91.5
Belfast 85.8 86.4 90.9 88.9 88.8 88.8 89.9 89.8
Newry 87.2 87.5 87.4 90.9 94.3 88.2 92.1 84.1
Ulster 86.1 91.6 89.4 92.6 88.2 90.6 90.5 91.6
West NI 88.9 83.7 91.5 93.0 89.7 91.8 91.5 92.3
Scotland
Abrdn 80.1 85.4 87.8 86.3 87.3 89.6 89.4 89.4 89.0 90.9
Airdrie 84.5 84.2 83.0 79.9 79.5 86.1 85.6 89.4 88.5 86.4
D & Gall 85.1 83.1 92.1 82.1 90.6 84.6 88.4 87.3 91.3 87.4
Dundee 83.5 86.0 87.4 87.6 84.1 84.2 93.8 87.9 88.4 92.0
Dunfn 84.2 88.9 91.0 88.7 88.8 91.0 87.9 88.0 90.2 88.2
Edinb 83.2 86.4 86.4 87.4 88.5 88.9 86.8 89.6 83.3 90.7
Glasgw 84.1 85.6 87.5 86.4 88.1 88.3 88.5 88.7 88.1 88.5
Inverns 87.6 86.9 87.2 86.5 93.8 89.2 92.2 89.0 86.8 88.0
Klmarnk 82.8 87.6 85.2 92.2 87.3 89.3 88.4 88.4 89.1 89.8
Wales
Bangor 81.2 89.8 86.6 88.5 81.4 88.7 85.0 85.4 86.8 89.8
Cardff 80.7 84.7 84.2 84.0 88.8 82.6 86.6 86.0 88.4 86.3
Clwyd 90.0 76.5 83.6 79.2 91.3 88.0 89.6 80.0 93.7 90.8
Swanse 82.0 87.2 89.2 85.9 88.2 89.5 87.4 87.7 89.2 86.6
Wrexm 86.0 85.9 83.6 85.8 88.2 85.9 89.6 87.5 86.1 88.1
England 88.7 88.0 87.9 88.4 88.6 88.9 89.1 89.2 89.9 89.9
N Ireland 86.1 87.6 89.4 90.4 89.9 89.7 91.2 90.1
Scotland 84.0 86.0 87.2 86.6 87.4 88.1 88.8 88.7 87.8 89.3
Wales 82.8 85.6 85.8 84.9 88.0 85.7 87.1 86.2 88.7 87.1
UK 88.2 88.0 87.7 88.0 88.5 88.7 89.0 89.0 89.7 89.7
Blank cells: data not reported for that year or less than 20 patients in the year
177
Chapter 9 Adequacy of Haemodialysis in
UK Adult Patients in 2012: National and
Catriona Shawa, Retha Steenkampa, Andrew Davenportb

a
UK Renal Registry, Bristol, UK; bRoyal Free Hospital, London, UK
Key Words prevalent patients achieving this target was wide

Adequacy . Haemodialysis . Urea reduction ratio (69.7–100%).
. The median URR in 2012 was 75%.
. URR was greater in those with longer dialysis
vintage. Ninety one percent of patients who had
survived on renal replacement therapy (RRT) for
Summary more than two years achieved a URR .65%
compared with only 74% of those on RRT for
. Data suitable for urea reduction ratio (URR) analyses only six months.
were available in 15,286 (75.2%) of the 20,332 . Large variation between centres in the percentage of
patients receiving haemodialysis (HD) in the UK patients achieving the UK Renal Association’s (RA)
on the 30/9/2012. URR guideline persists. The UK Renal Registry
. In 2012, 88% of prevalent HD patients achieved (UKRR) will explore a possible move to reporting
a URR .65%. The between centre range of Kt/V combined with residual renal function.
179
Introduction from the 4th quarter of the year to the 3rd quarter was because
many centres reported a dialysis frequency of less than 3 times a
Amongst patients with established renal failure (ERF), week in the 4th quarter. This could be due to changes in dialysis
patterns during the December holiday season, or due to some
the delivered dose of HD is an important predictor of inaccuracy in the data on the part of some renal centres. Data
outcome [1] and has been shown to influence survival from those patients who had died before that date have not
[2–4]. The delivered dose of HD depends on treatment been included in the analysis. The second analysis involved
(duration and frequency of dialysis, dialyser size, dialy- adult incident patients who had commenced treatment with HD
sate and blood flow rate) and patient characteristics during 2011. For these patients, analysis was undertaken using
the last recorded URR in the quarter in which the patient had
(size, weight, haematocrit and vascular access) [5]. The started dialysis. The incident HD patient cohort was followed up
two widely accepted measures of urea clearance are for one year and the last recorded URR in the quarter after one
Kt/V, the ratio between the product of urea clearance year follow-up was used for this analysis.
(K, in ml/min) and dialysis session duration (t, in Data from patients known to be receiving more or less than
thrice weekly HD were omitted from analysis for both the incident
minutes) divided by the volume of distribution of urea
and prevalent population. Patients whose data recording for the
in the body (V, in ml) and URR which is derived solely number of dialysis sessions per week were missing, were assumed
from the percentage fall in serum urea (URR) during a to be dialysing thrice weekly. Home HD patients were excluded
dialysis treatment. Whilst Kt/V is a more accurate from the analysis.
descriptor of urea clearance, its calculation is more com- Analyses of the data from both groups of patients included
calculation of the median URR and of the proportion of patients
plex and requires additional data items not commonly
who had achieved the RA guideline (as outlined below) in each
reported by most UK renal centres [6, 7]. The UKRR of the renal centres as well as for the country as a whole. This
has historically presented analyses based on URR rather year the median URR and proportion of patients who achieved
than Kt/V for comparative audit of haemodialysis the RA guideline were also calculated separately for males and
adequacy as these data are more widely available. females. The number of dialysis sessions per week and the time
per dialysis session is new in this year’s report and is shown by
Based on published evidence, clinical practice guide- renal centre. The nine centres in Scotland do not provide data
lines have been developed by various national and on number of dialysis sessions per week and the time per dialysis
regional organisations [8–11]. There is considerable session to the UKRR and are not included in these analyses.
uniformity between them with regard to the recom- All patients with data were included in the statistical analyses at
mendations for minimum dose of dialysis although a national level, although centres with fewer than 20 patients, or
providing less than 50% data completeness were excluded from
there are differences in the methodology advised. The the comparison between centres. The number preceding the centre
main objective of this chapter is to determine the extent name in each figure indicates the percentage of missing data for
to which patients undergoing HD treatment for estab- that centre.
lished renal failure in the UK received the dose of HD, The UK RA clinical practice guidelines [9] in operation at the
as measured by URR, recommended in the UK RA cur- time these data were collected were as follows:
rent clinical practice guidelines [9]. HD should take place at least three times per week
in nearly all patients. Reduction of dialysis frequency to
twice per week because of insufficient dialysis facilities is
unacceptable.
Methods Every patient receiving thrice weekly HD should have
consistently:
Seventy-one renal centres in the UK submitted data electroni- . either URR >65%
cally to the UKRR on a quarterly basis [12]. The majority of these . or equilibrated Kt/V (eKt/V) of >1.2 (or single pool Kt/V
centres have satellite units but for the purposes of this study the of >1.3) calculated from pre- and post-dialysis urea
data from the renal centres and their associated satellite units values, duration of dialysis and weight loss during
were amalgamated. However, because not all centres report fre- dialysis).
quency of HD, it is possible that data from a small number of
patients receiving HD at a different frequency were included in To achieve a URR above 65% or eKt/V above 1.2 consist-
the analyses. Data from two groups of patients were analysed. ently in the vast majority of the HD population clinicians
Firstly, analysis was undertaken using data from the prevalent should aim for a minimum target URR of 70% or minimum
adult HD patient population as of the 30th September 2012. For eKt/V of 1.4 in individual patients.
this analysis, data for URR were taken from the 3rd quarter of The duration of thrice weekly HD in adult patients with
2012 unless that data point was missing in which case data from minimal residual renal function should not be reduced
the 2nd quarter were taken. The prevalent population only below 4 hours without careful consideration.
included patients receiving HD who were alive on September Patients receiving HD twice weekly for reasons of geogra-
30th 2012. This change in the methodology from using data phy should receive a higher sessional dose of HD. If this
180
Chapter 9 UK haemodialysis dose
cannot be achieved, then it should be recognised that there is a Table 9.1. Percentage completeness of URR data returns for
compromise between the practicalities of HD and the patient’s prevalent patients on HD by centre, on 30/9/2012
long-term health.
Measurement of the ‘dose’ or ‘adequacy’ of HD should be Centre % completeness Centre % completeness
performed monthly in all hospital HD patients and may be
performed less frequently in home HD patients. All dialysis Abrdn 99.5 L Rfree 0.0
units should collect and report this data to their regional Airdrie 100.0 L St.G 0.0
network and the UKRR. Antrim 99.2 L West 94.7
Post-dialysis blood samples should be collected either by B Heart 99.5 Leeds 99.6
the slow-flow method, the simplified stop-flow method, or B QEH 94.8 Leic 99.2
the stop dialysate flow method. The method used should Bangor 100.0 Liv Ain 0.0
remain consistent within renal units and should be reported Basldn 95.5 Liv RI 0.0
to the Registry. Belfast 95.8 M RI 51.1
Bradfd 96.8 Middlbr 97.2
The RA clinical practice guidelines for HD dose apply specifi- Brightn 0.0 Newc 1.7
cally to patients undergoing thrice weekly HD. In these patients Bristol 100.0 Newry 87.1
it is recommended that blood for biochemical measurement Camb 96.8 Norwch 96.3
(including pre-dialysis urea for URR) should be taken before the Cardff 94.0 Nottm 93.7
mid-week dialysis session [9]. Carlis 100.0 Oxford 75.3
Carsh 86.8 Plymth 97.4
Chelms 98.1 Ports 95.8
Clwyd 94.3 Prestn 82.9
Colchr 89.6 Redng 3.2
Results Covnt 98.1 Salford 59.1
D & Gall 100.0 Sheff 94.4
Derby 94.3 Shrew 96.7
Data completeness Donc 95.4 Stevng 98.6
Data providing HD dose (URR) were available from Dorset 93.7 Sthend 97.8
63 of the 71 renal centres which submitted data to Dudley 91.6 Stoke 99.6
the UKRR (table 9.1). Data were available for Dundee 99.4 Sund 1.8
Dunfn 98.6 Swanse 55.1
75.2% (n = 15,286) of the total prevalent population Edinb 99.6 Truro 69.5
(n = 20,332) treated with HD who met the inclusion Exeter 98.8 Ulster 97.8
criteria for these analyses. Glasgw 98.7 West NI 95.0
Completeness in the 63 centres reporting URR data Glouc 100.0 Wirral 0.0
Hull 97.7 Wolve 88.2
was generally good, with 49 centres reporting data on
Inverns 98.6 Wrexm 96.2
more than 90% of patients. Three centres reported Ipswi 100.0 York 99.2
URR data on less than 50% of prevalent patients (Read- Kent 92.8 England 71.6
ing, Newcastle and Sunderland). URR data were not Klmarnk 100.0 N Ireland 95.4
received from eight centres (Brighton, London Barts, L Barts 0.0 Scotland 99.3
L Guys 73.3 Wales 82.5
London Kings, London Royal Free, London St Georges, L Kings 0.0 UK 75.2
Liverpool Aintree, Liverpool Royal Infirmary and
Wirral).
Several centres had a reduction in the completeness of
URR data submitted to the UKRR in 2012 compared with
2010 (data not shown). These changes may represent no data on this variable. All centres in Northern Ireland
changes in data extraction, or a move by centres to utilis- returned over 88% data.
ing Kt/V rather than URR as the preferred measure of For those centres that did return data, three dialysis
dialysis dose. sessions a week was most prevalent, although several
Of the total incident patient population (n = 4,387) centres reported .10% of the HD population under-
who started HD during 2011 and meeting the inclusion going HD for more or less than three sessions. For
criteria for URR analyses, 47.0% (n = 2,062) had URR example, Salford reported 22.9% of their prevalent
data available during the first quarter of treatment. haemodialysis population having more than three
Percentage completeness of data returns on the number sessions a week whereas Southend reported that 13.3%
of HD sessions varied across centres (table 9.2). Ten and Bradford 28.6% of their population in 2012 had
centres in England and two centres in Wales returned fewer than three sessions per week respectively.
181
Table 9.2. Percentage completeness for the number of dialysis sessions for prevalent patients on HD by centre, on 30/9/2012
Percentage
Percentage
Centre completeness ,3 sessions 3 sessions .3 sessions
England
B Heart 89.5 5.7 93.4 0.9
B QEH 0.0
Basldn 97.9 2.9 93.5 3.6
Bradfd 3.7 28.6 71.4 0.0
Brightn 99.3 0.0 99.7 0.3
Bristol 100.0 3.3 96.3 0.5
Camb 99.4 12.0 85.8 2.2
Carlis 86.7 9.6 90.4 0.0
Carsh 0.0
Chelms 100.0 8.7 90.4 0.9
Colchr 99.1 0.0 100.0 0.0
Covnt 2.2 0.0 100.0 0.0
Derby 89.8 0.6 99.4 0.0
Donc 99.4 1.3 98.7 0.0
Dorset 98.8 2.9 96.7 0.4
Dudley 97.3 3.5 96.5 0.0
Exeter 99.7 2.0 96.0 2.0
Glouc 0.0
Hull 2.9 11.1 88.9 0.0
Ipswi 86.8 6.1 93.9 0.0
Kent 98.2 6.0 92.9 1.2
L Barts 0.0
L Guys 0.0
L Kings 0.0
L Rfree 0.0
L St.G 67.4 0.6 99.4 0.0
L West 45.5 0.7 98.4 1.0
Leeds 15.7 2.9 95.7 1.4
Leic 98.7 0.4 99.6 0.0
Liv Ain 100.0 1.9 96.8 1.3
Liv RI 97.9 0.9 90.3 8.8
M RI 51.6 1.4 97.2 1.4
Middlbr 15.5 0.0 100.0 0.0
Newc 99.2 1.3 98.7 0.0
Norwch 98.2 2.5 96.0 1.4
Nottm 99.4 0.6 99.4 0.0
Oxford 0.0
Plymth 0.0
Ports 99.2 5.0 93.3 1.7
Prestn 0.0
Redng 100.0 0.4 99.6 0.0
Salford 99.7 0.6 76.5 22.9
Sheff 99.2 3.1 96.9 0.0
Shrew 100.0 5.0 93.8 1.3
Stevng 98.4 5.2 92.6 2.2
Sthend 99.1 13.3 86.7 0.0
Stoke 99.6 0.4 97.8 1.8
Sund 98.9 0.0 91.1 8.9
Truro 92.4 12.4 83.5 4.1
Wirral 92.4 2.5 87.4 10.1
Wolve 10.7 0.0 100.0 0.0
York 38.7 0.0 97.8 2.2
182
Percentage
Percentage
Centre completeness ,3 sessions 3 sessions .3 sessions
N Ireland
Antrim 99.2 0.8 99.2 0.0
Belfast 88.5 0.6 98.2 1.2
Newry 97.8 7.8 92.2 0.0
Ulster 96.8 1.1 97.8 1.1
West NI 98.4 0.0 95.2 4.8
Wales
Bangor 77.6 7.7 92.3 0.0
Cardff 0.0
Clwyd 94.4 3.0 97.0 0.0
Swanse 0.0
Wrexm 100.0 1.2 97.5 1.2
England 54.2 2.9 94.9 2.2

N Ireland 95.2 1.7 96.8 1.5
Wales 21.7 3.5 96.0 0.5
E, W & NI 53.9 2.9 95.0 2.1
Blank cells denote no data returned by that centre
Wide between centre variation in completeness of data the median URR for the UK from 2000 to 2012 is
on dialysis session time was also evident (table 9.3). In shown in figure 9.3. The proportion of patients attaining
centres that reported data the most frequently reported the RA guideline increased from 68.8% to 88.3% whilst
dialysis session length was 3–5 hours. the median URR has risen from 69.0% to 75.0%
during the same time period. There has been no substan-
Achieved URR tial change in the median URR between 2009 and 2012 in
For prevalent patients, the median URR (75.0% for the UK.
UK, centre range 70.5–81.0%) and percentage of patients
attaining the RA guideline of a URR .65% (88.4% for the
UK; centre range 69.7–100%) are shown in figures 9.1a Variation of achieved URR with time on dialysis
and figure 9.2 respectively. The median URR in women The proportion of patients who attained the RA
was 78.0% (95% CI 73.0–82.0%) compared with a UK guideline for HD was greater in those who had been
median in men of 74.0% (95% CI 69.0–78.0%) on RRT for the longest time (figure 9.4). In 2012, of
(figures 9.1b, 9.1c). those dialysed for less than 6 months, 74% had a URR
There continued to be variation between renal centres .65%, whilst 91% of patients who had survived and
in the percentage of prevalent patients with a URR of continued on RRT for more than two years attained
.65%, with 21 centres attaining the RA clinical practice the guideline target. In all strata of time on dialysis,
guideline in .90% of patients, 38 centres attaining the there has been an improvement in the proportion of
guideline in 70–90% of patients and one centre in less patients receiving the target dose of HD over the last
than 70% of patients (figure 9.2). There has been an 13 years.
improvement compared with 2010, when five centres The median URR during the first quarter of starting
reported fewer than 70% of their patients with a URR HD treatment of the incident HD population in the
of .65%. UK in 2011 was 67.5% (centre range 58.0–76.0%)
(figure 9.5a). At the end of one year for this incident
Changes in URR over time cohort, the median URR was higher and more uniform
The change in the percentage attainment of the cur- across renal centres (median URR 74.0%, centre range
rent RA clinical practice guidelines (URR .65%) and 69.0–80.0%) (figure 9.5b).
183
Table 9.3. Percentage completeness for time per dialysis session for prevalent patients on HD by centre, on 30/9/2012
Percentage per dialysis session

Percentage
Centre completeness ,3.5 hours 3.5–5 hours 5+ hours
England
B Heart 83.1 4.3 92.0 3.7
B QEH 0.0
Basldn 97.9 13.0 86.3 0.7
Bradfd 98.4 8.2 91.9 0.0
Brightn 97.7 3.4 96.6 0.0
Bristol 100.0 5.9 94.2 0.0
Camb 0.0
Carlis 86.7 7.7 92.3 0.0
Carsh 0.0
Chelms 100.0 9.6 90.4 0.0
Colchr 99.1 1.0 99.1 0.0
Covnt 7.8 44.0 56.0 0.0
Derby 89.8 1.3 98.7 0.0
Donc 99.4 12.4 87.6 0.0
Dorset 98.8 9.0 91.0 0.0
Dudley 97.3 6.3 93.7 0.0
Exeter 99.7 20.3 79.4 0.3
Glouc 0.0
Hull 2.9 11.1 88.9 0.0
Ipswi 86.8 3.0 97.0 0.0
Kent 98.2 14.6 85.4 0.0
L Barts 0.0
L Guys 18.5 0.0 100.0 0.0
L Kings 0.0
L Rfree 0.0
L St.G 61.8 0.0 100.0 0.0
L West 45.9 3.1 94.8 2.1
Leeds 100.0 8.3 91.7 0.0
Leic 91.9 2.6 97.0 0.4
Liv Ain 100.0 15.6 84.4 0.0
Liv RI 100.0 11.0 88.7 0.3
M RI 50.1 1.9 97.6 0.5
Middlbr 100.0 28.2 71.8 0.0
Newc 99.2 11.4 87.3 1.3
Norwch 98.2 22.3 77.7 0.0
Nottm 16.4 7.3 92.7 0.0
Oxford 0.0
Plymth 0.0
Ports 0.0
Prestn 0.9 0.0 100.0 0.0
Redng 91.3 1.3 98.3 0.4
Salford 97.2 11.1 88.9 0.0
Sheff 82.1 56.3 43.3 0.5
Shrew 99.4 25.8 74.2 0.0
Stevng 99.5 60.8 38.9 0.3
Sthend 99.1 23.8 76.2 0.0
Stoke 100.0 6.2 93.8 0.0
Sund 87.8 8.8 91.2 0.0
Truro 97.7 28.1 71.9 0.0
Wirral 94.2 16.7 81.5 1.9
Wolve 9.6 7.7 92.3 0.0
York 98.3 6.8 93.2 0.0
184
Percentage per dialysis session

Percentage
Centre completeness ,3.5 hours 3.5–5 hours 5+ hours
N Ireland
Antrim 98.4 0.8 99.2 0.0
Belfast 89.1 11.1 88.9 0.0
Newry 97.8 8.9 91.1 0.0
Ulster 96.8 3.3 96.7 0.0
West NI 98.4 11.3 88.7 0.0
Wales
Bangor 77.6 11.5 88.5 0.0
Cardff 0.0
Clwyd 94.4 29.9 70.2 0.0
Swanse 0.0
Wrexm 100.0 3.7 96.3 0.0
England 52.4 13.8 85.8 0.5

N Ireland 95.2 7.5 92.5 0.0
Wales 21.7 14.5 85.5 0.0
E, W & NI 52.3 13.4 86.2 0.4
Blank cells denote no data returned by that centre
Discussion a median URR of 75.0%. This increment will not only

reflect improvements in practice and delivery of dialysis,
The dose of delivered HD is recognised as having but also enhanced coverage and quality of the data
an important influence on outcome in established renal collected by the UKRR and renal centres over the years.
failure (ERF) patients treated with low flux HD. Survival Post hoc analyses of the HEMO study and observa-
has been shown to depend on achieving a minimum urea tional studies have suggested that women may benefit
clearance target [1–3]. It is therefore reassuring that from a higher dialysis dose than men [12, 13]. Current
the proportion of UK patients achieving the RA guideline RA guidelines do not differentiate on the basis of gen-
for URR has increased in the last decade, with 88.4% of the der [9]. It is an interesting observation that the UK
HD population achieving the URR guideline in 2012, with median URR achieved in women was higher than in
90
N = 15,286 Upper quartile
Median
85
Lower quartile
Urea reduction ratio (%)
80
75
70
65
60
4 Wrexm
5 B QEH
10 Colchr
5 L West
4 Ports
4 Belfast
13 Carsh
5 Donc
0 Glouc
2 Hull
13 Newry
5 West NI
1 Antrim
3 Bradfd
3 Camb
2 Covnt
6 Dorset
49 M RI
1 York
1 B Heart
0 Bangor
6 Cardff
0 D&Gall
1 Glasgw
1 Inverns
0 Klmarnk
0 Leeds
1 Leic
6 Sheff
2 Ulster
12 Wolve
0 Abrdn
5 Basldn
6 Derby
1 Dundee
7 Kent
4 Norwch
6 Nottm
17 Prestn
0 Stoke
45 Swanse
0 Carlis
2 Chelms
1 Exeter
27 L Guys
3 Middlbr
41 Salford
1 Stevng
2 Sthend
31 Truro
0 Airdrie
1 Dunfn
0 Edinb
3 Shrew
0 Bristol
6 Clwyd
8 Dudley
0 Ipswi
25 Oxford
3 Plymth
28 England
5 N Ireland
1 Scotland
17 Wales
25 UK
Centre
Fig. 9.1a. Median URR achieved in prevalent patients on HD by centre, 30/9/2012
185
186
Percentage of patients Urea reduction ratio (%) Urea reduction ratio (%)
60
65
70
75
80
85
90
60
65
70
75
80
85
90
50
60
70
80
90
100
4 Wrexm 4 Wrexm 4 Wrexm
10 Colchr 5 B QEH 5 B QEH
4 Ports 10 Colchr 10 Colchr
13 Newry 5 L West 5 L West
0 Klmarnk 4 Ports 4 Ports
5 Donc 5 Donc 4 Belfast
5 B QEH 13 Newry 13 Carsh
0 Bangor 4 Belfast 5 Donc
5 L West 13 Carsh 0 Glouc
Upper 95% Cl
Lower 95% Cl
0 Glouc 2 Covnt 1 Antrim
6 Dorset
% with URR >65%

6 Dorset 2 Hull
0 Leeds 0 Glouc 0 Klmarnk
3 Camb 2 Hull 12 Wolve
49 M RI 5 West NI 1 York
5 West NI 1 Antrim 41 Salford
2 Ulster 0 Bangor 5 West NI
2 Hull 3 Camb 3 Camb
6 Cardff
N = 15,286
4 Belfast 6 Cardff
6 Sheff 1 Inverns 6 Dorset
1 Glasgw 0 Leeds 1 Glasgw
1 Antrim 1 Leic 0 Leeds
13 Carsh 6 Sheff 49 M RI
1 Inverns 45 Swanse 13 Newry
0 Stoke 12 Wolve 0 D&Gall
1 York 1 York 1 Leic
1 Dunfn 5 Basldn 1 B Heart
6 Derby 2 Ulster 0 Bangor
2 Covnt 1 B Heart 3 Bradfd
0 Carlis 3 Bradfd 2 Covnt
45 Swanse 0 Carlis 6 Derby
5 Basldn 2 Chelms 1 Dundee
12 Wolve 6 Derby
17 Prestn
17 Prestn 1 Glasgw
6 Sheff
Centre
Centre
7 Kent 7 Kent
Centre
0 Klmarnk 45 Swanse
27 L Guys 2 Ulster
1 B Heart 49 M RI
4 Norwch 0 Abrdn
6 Nottm 1 Inverns
0 Edinb 6 Nottm
17 Prestn 7 Kent
4 Norwch 27 L Guys
0 Abrdn 0 D&Gall
0 Abrdn 4 Norwch
6 Cardff 6 Nottm
1 Leic 1 Dundee
1 Stevng
Fig. 9.2. Percentage of prevalent patients with URR >65% on HD by centre, 30/9/2012
1 Exeter 0 Edinb
Fig. 9.1c. Median URR achieved in male prevalent patients on HD by centre, 30/9/2012
1 Exeter 0 Stoke
3 Bradfd
Fig. 9.1b. Median URR achieved in female prevalent patients on HD by centre, 30/9/2012
27 L Guys 2 Chelms
2 Sthend
31 Truro 2 Sthend 8 Dudley
3 Shrew 0 Stoke 31 Truro
1 Dundee 31 Truro 5 Basldn
3 Middlbr 0 Airdrie 0 Edinb
0 D&Gall 1 Dunfn 1 Exeter
0 Airdrie 3 Middlbr 3 Middlbr
8 Dudley 3 Shrew 2 Sthend
0 Ipswi 1 Stevng 0 Airdrie
0 Bristol 8 Dudley 0 Bristol
N = 9,299
N = 5,987
25 Oxford 0 Ipswi 6 Clwyd

1 Stevng 25 Oxford 1 Dunfn
41 Salford 41 Salford 0 Ipswi
2 Chelms 0 Bristol 25 Oxford
3 Plymth 6 Clwyd 3 Plymth
6 Clwyd 3 Plymth 3 Shrew
28 England 28 England
Median
Median
28 England
5 N Ireland 5 N Ireland 5 N Ireland
17 Wales 17 Wales
Lower quartile
17 Wales
Lower quartile
Upper quartile
Upper quartile
25 UK 25 UK 25 UK
95
90
85
80
75
70
65
60 % of patients with Median URR

URR >65% and 95% CI and quartiles
55
50 Fig. 9.3. Change in the percentage of

prevalent patients on HD with URR >65%
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
and the median URR between 2000 and
Year 2012 in the UK
95
90
85
80
75
70
65
60
55 >2 yrs
1–2 yrs
50 6 m–1 yr
45 <6 months
40 Fig. 9.4. Percentage of prevalent patients
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
on HD achieving URR >65% by time on
Year
RRT between 1999 and 2012
85
80
75
70
65
60
55
50 Upper quartile
Median N = 2,062
45 Lower quartile
40
39 B QEH
40 Belfast
50 Stoke
23 Bradfd
46 Ports
36 Carsh
36 Glouc
38 Wolve
23 Hull
42 Liv RI
6 Ulster
9 Edinb
10 Leeds
10 York
22 Sheff
17 Dundee
42 Norwch
45 Nottm
17 B Heart
33 Basldn
40 Camb
16 Covnt
11 Middlbr
7 Prestn
27 Dunfn
45 Airdrie
35 Bristol
34 Stevng
21 Plymth
12 Leic
26 Abrdn
53 England
36 N Ireland
44 Scotland
72 Wales
53 UK
Centre
Fig. 9.5a. Median URR in the first quarter of starting RRT in incident patients who started haemodialysis in 2011
187
85
80
75
70
Upper quartile
65
Median
Lower quartile N = 1,462
60
B QEH
Carsh
L West
Ports
Belfast
Bradfd
Stoke
Cardff
Dundee
Glasgw
Hull
Leic
Nottm
B Heart
Camb
Prestn
Sheff
Abrdn
Exeter
Leeds
Middlbr
Covnt
Edinb
Kent
Ulster
Stevng
Bristol
England
N Ireland
Scotland
Wales
UK
Centre
Fig. 9.5b. Median URR one year after starting RRT for patients who started haemodialysis in 2011
men in this analysis. This may simply reflect differences be incorrectly categorised and introduce bias into the
in dietary intake and lower pre-dialysis serum urea median estimate of URR and the percentage achieving the
values in women, and as such does not necessarily URR RA standard. Although RA guidelines recommend
imply improved urea clearances for women [14, 15]. standardised methods for urea sampling, inconsistency in
In the prevalent haemodialysis population there was a sampling methodology for the post-dialysis urea sample
wide range (69.7–100%) of achievement of the RA may also play a part in the variations seen [9].
guideline for URR between different centres which is Debate continues as to the toxicity of urea, and how
likely to reflect genuine differences in HD dose with representative urea clearance is of other azotaemic
both individual and centre level contributors. Under- toxin clearances. In addition, the dialysis prescription
standing more fully individual renal centre practice should also be designed to achieve volume, sodium and
would be informative. In the incident population, the divalent cation balance and correct metabolic acidosis.
variation in the between centre median URR within the As such basing HD dosing simply on urea clearance is
first quarter for incident patients may represent variation criticised by some [13] arguing that patient outcomes
in dialysis prescription practice for patients starting RRT. are improved by longer treatment times independent of
Some renal centres may use dialysis initially as a ‘top-up’ urea removal [5, 17–22] and that clearance of ‘middle
in individuals with residual renal function, whilst other molecules’ has an important impact [23, 24]. However,
centres use a more standardised ‘full-dose’ approach to no consensus has yet emerged on alternative markers of
dialysis prescription, irrespective of residual function. HD adequacy. The UKRR has historically reported
Although evidence supports that preservation of residual URR, predominantly for logistical reasons with the
renal function is associated with improved survival [16], URR being the easiest measure to calculate, and the
how much individualisation of dialysis prescription measure of dialysis adequacy that is most complete
based on residual renal function is practiced across UK when returned to the UKRR. However, the limitations
renal centres and how this correlates with outcomes is of the URR are recognised. Although URR correlates
not currently known. Similarly, it is not known whether well with single pool Kt/V (spKt/V) in population
the decline in residual renal function is affected by studies, significant variability in correlation in individual
differences in centre practice approach to initiating patients occurs because URR fails to include both
dialysis. Varied completeness of data returns across other the contraction in extracellular volume (ECV) and the
important factors such as dialysis session also limits the urea generation during routine HD [11]. Neither URR
interpretation of the data, and increases the risk of nor spKt/V take into account post-dialysis urea rebound,
misclassification of patients in the presented analyses. For potentially resulting in an over-estimate of the amount of
example, some patients who were receiving more or less dialysis actually delivered. A possible move to reporting
frequent dialysis sessions than three times per week may eKt/V to the UKRR in addition to high quality data on
188
residual renal function, weights and dialysis prescription allow for evaluation of different approaches to the initia-
practice including duration and frequency of sessions tion of dialysis and the effect of residual renal function.
would enhance the quality of analyses the UKRR could
provide for the renal community, and would potentially Conflicts of interest: none
References
1 Gotch FA, Sargent JA: A mechanistic analysis of the National Coopera- 13 Port FK, Wolfe RA, Hulbert-Shearon TE, McCullough KP, Ashby VB,
tive Dialysis Study (NCDS). Kidney Int 1985;28:526–534 Held PJ. High dialysis dose is associated with lower mortality among
2 Owen WF, Lew NL, Liu Y, Lowrie EG, Lazarus JM: The Urea Reduction women but not among men. Am J Kidney Dis. 2004;43(6):1014
Ratio and Serum Albumin Concentration as Predictors of Mortality in 14 Lowrie EG: The Kinetic Behaviors of Urea and Other Marker Molecules
Patients Undergoing Hemodialysis. N Engl J Med 1993;329:1001–1006 During Hemodialysis. Am J Kidney Dis 2007;50:181–183
3 Held PJ, Port FK, Wolfe RA, Stannard DC, Carroll CE, Daugirdas JT, 15 Spalding EM, Chandna SM, Davenport A, Farrington K. Kt/V under-
Bloembergen WE, Greer JW, Hakim RM: The dose of hemodialysis estimates the haemodialysis dose in women and small men. Kidney
and patient mortality. Kidney Int 1996;50:550–556 Int 2008;74:348–355
4 Tentori F, Hunt WC, Rohrscheib M, Zhu M, Stidley CA, Servilla K, 16 Hanson JA, Hulbert-Shearon TE, Ojo AO, et al.: Prescription of twice-
Miskulin D, Meyer KB, Bedrick EJ, Johnson HK, Zager PG: Which weekly hemodialysis in the USA. Am J Nephrol 1999;19:625–633
Targets in Clinical Practice Guidelines Are Associated with Improved 17 Vanholder R, Eloot S, Van Biesen W: Do we need new indicators of
Survival in a Large Dialysis Organization? J Am Soc Nephrol 2007;18: dialysis adequacy based on middle-molecule removal? Nature Clinical
2377–2384 Practice Nephrology 2008;4:174–175
5 Locatelli F, Buoncristiani U, Canaud B, Kohler H, Petitclerc T, Zucchelli 18 Tattersall J, Martin-Malo A, Pedrini L, Basci A, Canaud B, Fouque D,
P: Dialysis dose and frequency. Nephrol Dial Transplant 2005;20: Haage P, Konner K, Kooman J, Pizzarelli F, Tordoir J, Vennegoor M,
285–296 Wanner C, ter Wee P, Vanholder R: EBPG guideline on dialysis
6 Depner TA: Assessing adequacy of hemodialysis: urea modeling. Kidney strategies. Nephrol Dial Transplant 2007;22:ii5–21
Int 1994;45:1522–1535 19 Saran R, Bragg-Gresham JL, Levin NW, Twardowski ZJ, Wizemann V,
7 Movilli E: Simplified approaches to calculate Kt/V. It’s time for agree- Saito A, Kimata N, Gillespie BW, Combe C, Bommer J, Akiba T,
ment. Nephrol Dial Transplant 1996;11:24–27 Mapes DL, Young EW, Port FK: Longer treatment time and slower
8 Vanbelleghem H, Vanholder R, Levin NW, Becker G, Craig JC, Ito S, Lau ultrafiltration in hemodialysis: Associations with reduced mortality in
J, Locatelli F, Zoccali C, Solez K, Hales M, Lameire N, Eknoyan G: The the DOPPS. Kidney Int 2006;69:1222–1228
Kidney Disease: Improving Global Outcomes website: Comparison of 20 Marshall MR, Byrne BG, Kerr PG, McDonald SP: Associations of
guidelines as a tool for harmonization. Kidney Int 2007;71:1054–1061 hemodialysis dose and session length with mortality risk in Australian
9 UK Renal Association Clinical Practice Guidelines Committee. Haemo- and New Zealand patients. Kidney Int 2006;69:1229–1236
dialysis, 2009 http://www.renal.org/Clinical/GuidelinesSection/Haemo- 21 Eloot S, Van Biesen W, Dhondt A, Van de Wynkele H, Glorieux G,
dialysis.aspx Verdonck P, Vanholder R: Impact of hemodialysis duration on the
10 European Best Practice Guidelines Expert Group on Haemodialysis. removal of uremic retention solutes. Kidney Int 2007;73:765–770
Nephrol Dial Transplant 2002:17(suppl 7):S16–S31 22 Basile C, Lomonte C: Dialysis time is the crucial factor in the adequacy of
11 NKF-KDOQI clinical practice guidelines; update 2006. Am J Kidney Dis hemodialysis. Kidney Int 2008;74:965–966
2006:48(suppl 1):S2–S90 23 Eloot S, Torremans A, De Smet R, Marescau B, De Deyn PP, Verdonck
12 Depner T, Daugirdas J, Greene T, Allon M, Beck G, Chumlea C, Delmez P, Vanholder R: Complex Compartmental Behavior of Small Water-
J, Gotch F, Kusek J, Levin N, Macon E, Milford E, Owen W, Star R, Toto Soluble Uremic Retention Solutes: Evaluation by Direct Measurements
R, Eknoyan G, Hemodialysis Study Group Dialysis dose and the effect of in Plasma and Erythrocytes. Am J Kidney Dis 2007;50:279–288
gender and body size on outcome in the HEMO Study. Kidney Int. 24 Davenport A. How best to improve survival in haemodialysis patients:
2004;65(4):1386 solute clearance or volume control? Kidney Int. 2011;80(10):1018–20
189
Chapter 10 Haemoglobin, Ferritin and
Erythropoietin amongst UK Adult Dialysis
Patients in 2012: National and
Anirudh Raoa, Julie Gilga, Andrew Williamsb

a
UK Renal Registry, Bristol, UK; bMorriston Hospital, Swansea, UK
Key Words
. The median Hb of prevalent patients on HD was
Anaemia . Chronic kidney disease . Dialysis . End stage renal 112 g/L with an IQR of 103–121 g/L.
disease . Epidemiology . Erythropoietin . Erythropoietin
. The median Hb of prevalent patients on PD was
stimulating agent . European Best Practice Guidelines . 114 g/L with an IQR of 105–123 g/L.
Ferritin . Haemodialysis . Haemoglobin . NICE . Peritoneal . 82% of HD and 85% of PD patients had Hb
dialysis . Renal Association 5100 g/L.
. 57% of HD patients and 55% of PD patients had Hb
5100 and 4120 g/L.
Summary . The median ferritin in HD patients was 431 mg/L
(IQR 285–623) and 95% of HD patients had a
In the UK in 2012: ferritin 5100 mg/L.
. The median Hb of patients at the time of starting In England, Wales and Northern Ireland in 2012:
dialysis was 100 g/L with 51% of patients having a
Hb 5100 g/L. . The median ferritin in PD patients was 285 mg/L
. The median Hb in patients starting haemodialysis (IQR 164–466) with 88% of PD patients having a
(HD) was 97 g/L (IQR 89–106) and in patients ferritin 5100 mg/L.
starting peritoneal dialysis (PD) was 109 g/L (IQR . The median erythropoietin stimulating agent (ESA)
99–118). dose was higher for HD than PD patients (7,248 vs.
. At start of dialysis, 54% of patients presenting early 4,250 IU/week).
had Hb 5100 g/L whilst 34% of patients presenting
late had Hb 5100 g/L.
191
Introduction recently updated KDOQI guidelines suggest ESAs should

not be used to maintain Hb concentration routinely
This chapter describes the UK Renal Registry (UKRR) above 115 g/L with careful consideration in patients
data relating to the management of anaemia in dialysis who require individualization of therapy for improve-
patients during 2012. The chapter reports on the ments in quality of life at Hb concentration above
analyses of submitted variables in the context of the UK 115 g/L [7]. The target of Hb 100–120 g/L has been
Renal Association – Anaemia in CKD guidelines and used for both HD and PD patients in keeping with the
recommendations. above recommendations. There are also some analyses
In this report, haemoglobin levels are given in g/L as showing attainment of the minimum standard of Hb
the majority of UK laboratories have now switched to 5100 g/L.
reporting using these units. In patients on peritoneal dialysis (PD), the timing of
Anaemia in adults with CKD is diagnosed when the the blood sample draw is not critical because plasma
Hb concentration is ,130 g/L in males and ,120 g/L volume in these patients remains relatively constant. In
in females [1]. The degree of renal impairment affects haemodialysis (HD) patients, interdialytic weight gain
the likelihood of any patient developing anaemia. contributes to a decrease in Hb level, whereas intradialy-
Although current treatment with ESAs is not rec- tic ultrafiltration leads to an increase. Thus, a predialysis
ommended unless Hb falls consistently below 110 g/L, sample underestimates the euvolaemic Hb level, whereas
other causes of anaemia should be excluded in patients a postdialysis sample overestimates the euvolaemic Hb.
with Hb below the normal range. Given the relationship between Hb level and the dialysis
The renal National Service Framework (NSF) part one related weight change, midweek pre-dialysis sampling is
[2] and the RA minimum standards document 3rd edi- recommended for regular Hb monitoring [8].
tion [3] state that individuals with chronic kidney disease The 2010 Renal Association (RA) Clinical Practice
(CKD) should achieve a haemoglobin (Hb) of at Guidelines document, revised European Best Practice
least 100 g/L within six months of being seen by a Guidelines (EBPGII), Dialysis Outcomes Quality Initiative
nephrologist, unless there is a specific reason why it is (DOQI) guidelines and UK NICE anaemia guidelines all
unachievable. The UKRR does not collect Hb measure- recommend a target serum ferritin greater than 100 mg/L
ments from patients with CKD six months after meeting and percentage transferrin saturation (TSAT) of more
a nephrologist. However, an indication of the attainment than 20% in patients with CKD. RA guidelines and EBP-
of this standard is given by the Hb of the incident patient GII recommend hypochromic red cells (HRC) less than
population at the start of dialysis. Achievement of these 10%. In addition, EBPGII recommends target reticulocyte
standards is mainly through the use of iron therapy Hb content (CHr) of greater than 29 pg/cell. KDOQI
(oral and intravenous) and erythropoietin stimulating recommends a serum ferritin .200 mg/L for HD patients.
agents (ESAs). The NICE guidelines suggest that a hypochromic red cell
The European Best Practice Guidelines (EBPG) value .6% indicates ongoing iron deficiency.
published in 2009 recommend that Hb values of 110– To achieve adequate iron status across a patient popu-
120 g/L should be generally sought in the CKD population, RA guidelines [6] advocate population target
lation without intentionally exceeding 130 g/L [4]. The medians for ferritin of 200–500 mg/L in HD patients
5th edition of the UK Renal Association’s Anaemia in and 100–500 mg/L for PD patients, for TSAT of 30–
CKD guideline was published at the end of 2010 and 40%, for hypochromic red cells of ,2.5% and CHr of
attempted to unify targets with those published in the 35 pg/cell. EBPGII comments that a serum ferritin target
2011 update NICE guideline on anaemia management for the treatment population of 200–500 mg/L ensures
in CKD and other guidelines [5, 6]. The target outcome that 85–90% of patients attain a serum ferritin of
Hb for RRT patients on ESA treatment in these guidelines 100 mg/L. All guidelines advise that serum ferritin levels
is between 100 and 120 g/L. The rationale behind choos- should not exceed 800 mg/L since the potential risk of
ing a wide target Hb range (100–120 g/L) is that when the toxicity increases without conferring additional benefit.
target Hb level is narrow (e.g. 100 g/L), variability in The KDOQI and NICE guidelines advise against intra-
achieved Hb levels around the target is high, the pro- venous iron administration to patients with a ferritin
portion of prevalent patients with achieved Hb levels .500 mg/L.
within the target range is low and ESA dose titration is Serum ferritin has some disadvantages as an index of
required frequently during maintenance therapy. The iron status. It measures storage iron rather than available
192
Chapter 10 Anaemia management in UK dialysis patients
iron, behaves as an acute phase reactant and is therefore renal centres in England, Wales and Northern Ireland; data
increased in inflammatory states, malignancy and liver from Scotland were provided by the Scottish Renal Registry.
For the analyses of Hb for incident patients, those patients
disease and may not accurately reflect iron stores if commencing RRT on PD or HD were included whilst those receiv-
measured within a week of the administration of intra- ing a pre-emptive transplant were excluded. Hb measurements
venous iron. Serum ferritin level is less reliable in the from after starting dialysis but still within the same quarter of
evaluation of iron stores in HD patients, because ferritin the year were used. Therefore, depending on when in the quarter
level is affected by other factors in addition to iron a patient started RRT the Hb could be from 0 to 90 days later. The
haemoglobin values the UKRR receives should be the closest
storage status. In relatively healthy HD patients, before available measurement to the end of the quarter. Patients who
widespread use of IV iron therapy, the finding of a ferritin died within the first 90 days on treatment were excluded. Results
level less than 50 ng/ml was not uncommon and was are also shown with the cohort subdivided into early and late
associated with absent bone marrow iron in approxi- presenters (date first seen by a nephrologist, 90 or more days
mately 80% of patients. However, in HD patients with and less than 90 days before starting dialysis respectively).
For the analyses of prevalent patients, those patients receiving
several comorbidities, absent iron stores may still be dialysis on 31st December 2012 were included if they had been on
found at ferritin levels approaching or even exceeding the same modality of dialysis in the same centre for at least three
200 ng/ml [9]. months. In order to improve completeness the last available
Of the alternative measures of iron status available, measurement for each patient from the last two quarters for Hb
HRC and CHr are generally considered superior to and from the last three quarters for ferritin was used. Scotland
was excluded from the analysis for ferritin for PD patients as
TSAT. Both however require specialised analysers to this data was not available.
which not all UK renal centres have easy access. Since The completeness of data items were analysed at both centre
TSAT is measured infrequently in many centres and and country level. As in previous years, all patients were included
most UK centres continue to use serum ferritin for in analyses but centres with less than 50% completeness were
routine iron management, ferritin remains the chosen excluded from the caterpillar and funnel plots showing centre
performance. Centres providing relevant data from less than 10
index of iron status for this report. patients were also excluded from the plots. The number preceding
Anaemia treatment in CKD patients has changed the centre name in the caterpillar plots indicates the percentage of
dramatically since the implementation of erythropoietin data missing for that centre.
stimulating agents (ESAs) into clinical practice in 1987. The data were analysed to calculate summary statistics
This has reduced the need for blood transfusions and including maximum, minimum and average (mean and median)
values. Standard deviations and inter-quartile ranges (IQR) were
improved quality of life for patients [10]. These agents also calculated. These are shown using caterpillar plots giving
are relatively expensive and thus approaches to achieving median values and the inter-quartile ranges.
optimal haemoglobin levels with the lowest possible The percentages achieving RA and other standards were calcu-
doses are desirable. The health economics of anaemia lated for Hb and ferritin. These are displayed using caterpillar
therapy using ESAs has been subject to a NICE systematic plots with the percentages meeting the targets and 95% confidence
intervals (CIs) shown. Funnel plots show the distribution of the
review [5] which concluded that treating to a target Hb percentages meeting the various targets and also whether any of
110–120 g/L is cost effective in HD patients. the centres are significantly different from the average.
The risks associated with low (,100 g/L) and high Longitudinal analysis was performed to show overall changes
(.130 g/L) Hb are not necessarily equivalent. Two in achievement of standards from 1998 to 2012.
important studies of patients not yet on dialysis, Erythropoietin data from the last quarter of 2012 were used to
define which patients were receiving ESAs. Scotland was excluded
CHOIR [11] and CREATE [12] showed an increased from this analysis as data regarding ESA was not included in its
risk of cardiovascular events amongst the patients return. Each individual was defined as being on ESA if a drug
assigned to the higher Hb targets. In the TREAT study type and/or a dose was present in the data. Centres reporting
[13] although there was no difference between the two fewer than 60% of HD patients or fewer than 45% of PD patients
arms in the primary outcome of death, cardiovascular being treated with ESAs were considered to have incomplete data
and were excluded from further analysis. It is recognised that these
event or end stage renal disease, there was an increase exclusion criteria are relatively arbitrary but they are in part based
in fatal or non-fatal stroke in the treatment arm. upon the frequency distribution graph of centres’ ESA use as it
appears in the data. The percentage of patients on ESAs is calcu-
lated from these data and incomplete data returns risk seriously
impacting on any conclusions drawn.
Methods For analyses of ESA dose, values are presented as weekly
erythropoietin dose. Doses of less than 150 IU/week (likely to be
The incident and prevalent RRT cohorts for 2012 were darbepoietin) were harmonised with erythropoietin data by multi-
analysed. The UKRR extracted quarterly data electronically from plying by 200. No adjustments were made with respect to route of
193
administration. Patients who were not receiving ESAs were not Median Hb of patients at the time of starting HD was
included in analyses of dose (rather than being included with 97 g/L (IQR 89–106 g/L) and in those starting PD was
dose = 0).
Until last year, reports have only used the dose from the final
109 g/L (IQR 99–118 g/L). When starting dialysis, 44%
quarter of the year. Now, as last year, starting with the cohort of of HD patients had a Hb 5100 g/L, compared with
patients receiving ESAs in the final quarter and having a dose 75% of PD patients.
value present for that quarter, any further dose values available Incident dialysis patients from 2011 were followed for
from the earlier three quarters of the year were used (provided one year and the median haemoglobin (and percentage
the patient was on the same treatment and receiving the same
drug in those quarters). The average (mean) of the available values
with a Hb 5100 g/L) of survivors on the same treatment
was then used in analyses rather than the dose in the final quarter. at the same centre after a year was calculated for each
The ESA data were collected electronically from renal IT quarter. Only patients who had Hb data for each of the
systems but in contrast to laboratory linked variables the ESA four time points were included in this analysis. This
data required manual data entry. The reliability depended upon was sub-analysed by modality and length of pre-RRT
the data source, whether the entry was linked to the prescription
or whether the prescriptions were provided by the primary care
care (figures 10.3 and 10.4). Hb was higher in the second
physician. In the latter case, doses may not be as reliably updated quarter on dialysis than during the quarter at start of
as the link between data entry and prescription is indirect. dialysis reflecting the benefits of treatment administered.
Over 76% of incident patients surviving to a year had Hb
5100 g/L regardless of the modality or the length of
pre-RRT care.
Results The annual distribution of Hb in incident dialysis
patients is shown in figure 10.5. Since 2006, the pro-
Anaemia management in incident dialysis patients portion of incident patients with Hb 5120 g/L has fallen
Haemoglobin in incident dialysis patients from 17% to 10% and the proportion of patients with Hb
The Hb at the time of starting RRT gives the only ,100 g/L continues to gradually increase over the years
indication of concordance with current anaemia manage- from 40% to 49%. In the 2012 cohort, 66% of patients
ment recommendations in the pre-dialysis (CKD 5 not in the late presentation group had Hb ,100 g/L com-
yet on dialysis) group. pared with 46% in the early presentation group.
The percentage of data returned and outcome Hb are
listed in table 10.1. Results are not shown for two centres ESA by time on dialysis in early vs. late presenters
(Kent and Inverness) because data completeness was less Incident dialysis patients from 2011 were followed for
than 50%. one year and the percentages receiving an ESA were
The median Hb of patients at the time of starting calculated for each quarter for survivors on the same
dialysis in the UK was 100 g/L. The median starting Hb treatment at the same centre after a year. This was
by centre is shown in figure 10.1. The percentage of sub-analysed by modality and length of pre-RRT care
patients having a Hb 5100 g/L has fallen over the last (figure 10.6). For HD patients at the start of treatment
couple of years to 51% from 55% in the 2009 cohort. there was a relatively small difference between early
The percentage starting with a Hb 5100 g/L by centre and late presenters in the percentage of patients receiving
is given in figure 10.2. an ESA. This difference had disappeared within one year
The variation in the proportion of patients starting of starting dialysis. For PD patients there was a more
renal replacement therapy with Hb 5100 g/L between marked difference between the early and late group
centres remained high (32–87%). Using only centres which was highest in the second quarter at more than
with time of presentation data, the median Hb in the 10%. The difference was lowest 1 year after starting
late presenters was 94 g/L with only 34% of patients dialysis. Caution is advised in interpreting this figure as
having a Hb 5100 g/L compared with a median Hb of the number of patients in the PD late group is relatively
101 g/L and 54% of the patients having a Hb 5100 g/L small (22).
in the early presenters group. In the late presenters group
there was a large variation between centres in percentage Anaemia management in prevalent dialysis patients
of patients having a Hb 5100 g/L (9%–64%). The lower Compliance with data returns for haemoglobin and
median Hb in late presenters may reflect inadequate serum ferritin and percentages on ESA are shown for
pre-dialysis care with limited anaemia management, the 71 renal centres in the UK in table 10.2 for both
anaemia of multisystem disease or inter-current illness. HD and PD patients. Completeness of data returns was
194
Table 10.1. Haemoglobin data for incident patients starting haemodialysis or peritoneal dialysis during 2012, both overall and by
presentation time
Early presenters only Late presenters only

All incident patients (590 days) (,90 days)
Centre % data N with Median % Hb Median % Hb Median % Hb

return data Hb g/L 5100 g/L Hb g/L 5100 g/L Hb g/L 5100 g/L
England
B Heart 100 96 96 39 94 37
B QEH 94 179 95 36 98 44 88 17
Basldn 100 45 94 40 97 44
Bradfd 97 56 103 59 104 64
Brightn 97 118 103 63 107 67 93 40
Bristol 100 128 97 45 99 48 85 26
Camb 94 81 100 51 102 58 94 36
Carlis 100 15 114 87 116 92
Carsh 99 204 103 60 103 63 99 48
Chelms 97 37 101 59 104 65
Colchr 52 14 97 36 97 42
Covnt 96 90 101 56 101 58 94 44
Derby 97 68 100 53 102 57 93 27
Donc 100 41 96 41 96 45
Dorset 97 63 106 57 106 59
Dudley 96 47 100 51 100 53
Exeter 100 125 102 57 103 61 97 44
Glouc 100 68 101 53 104 57 96 40
Hull 88 74 106 64 109 68
Ipswi 100 38 97 45 96 40 108 58
Kent 46 44
L Barts 100 241 99 49
L Guys 56 63 98 44
L Kings 99 114 96 42 96 43 96 39
L Rfree 68 140 103 55 105 60 98 44
L St.G 89 64 95 39
L West 79 176 105 69
Leeds 98 111 95 36 96 40 90 14
Leic 98 186 95 38 97 43 90 20
Liv Ain 98 57 102 58 103 60
Liv RI 95 70 102 51 104 55 95 41
M RI 97 116 98 47 97 46 104 64
Middlbr 98 93 93 32 97 38 83 16
Newc 98 82 102 57 101 56 109 64
Norwch 95 61 105 64
Nottm 99 72 98 49 100 51
Oxford 99 131 96 44 97 45 90 30
Plymth∗ 100 41 100 51
Ports 100 134 102 60 104 63 99 40
Prestn 100 116 99 45 99 45 99 43
Redng 100 67 103 61 108 71 94 31
Salford 90 110 99 47
Sheff 100 133 100 50 101 52 95 38
Shrew 100 49 106 57 106 56
Stevng 99 73 98 48 98 48 98 50
Sthend 100 25 99 48 100 53
Stoke 99 66 102 55 104 60 95 39
Sund 96 54 101 52 101 53
Truro 100 42 102 62 106 80 91 9
Wirral 98 44 104 70
Wolve 99 72 102 54 111 65 92 22
York 100 46 95 33 98 40 87 9
195
Early presenters only Late presenters only

All incident patients (590 days) (,90 days)
Centre % data N with Median % Hb Median % Hb Median % Hb

return data Hb g/L 5100 g/L Hb g/L 5100 g/L Hb g/L 5100 g/L
N Ireland
Antrim 100 26 102 54 104 58
Belfast 95 57 101 56 101 58 98 42
Newry 100 18 104 61 104 67
Ulster 100 21 109 71 109 76
West NI 89 16 98 38 98 36
Scotland
Abrdn 100 54 98 46
Airdrie 68 40 95 40
D & Gall 65 11 99 45
Dundee 89 33 98 42
Dunfn 77 20 107 60
Edinb 83 53 101 57
Glasgw 64 103 98 47
Inverns 46 6
Klmarnk 78 29 94 45
Wales
Bangor 95 18 102 67 101 64
Cardff 100 137 103 61 104 65 94 29
Clwyd 100 19 103 63 103 67
Swanse 99 97 99 46 103 58 89 16
Wrexm 97 30 108 67 109 71
England 93 4,480 100 51 101 53 94 34
N Ireland 97 138 103 57 104 59 95 38
Scotland 75 349 99 48
Wales 99 301 102 57 104 64 92 26
UK 92 5,268 100 51 101 54 94 34
Blank cells denote centres excluded from analyses due to poor data completeness or low patient numbers or because presentation time data not
available
∗
Plymouth, approximately 33% of incident patients were missing from the data extract
130
Median Hb
120
Lower quartile
Haemoglobin g/L
110
100
90
80
70
0 Carlis
0 Ulster
3 Wrexm
23 Dunfn
3 Dorset
12 Hull
0 Shrew
21 L West
5 Norwch
0 Newry
2 Wirral
3 Bradfd
32 L Rfree
3 Brightn
0 Redng
0 Clwyd
0 Cardff
1 Carsh
0 Truro
0 Exeter
0 Ports
1 Wolve
5 Bangor
2 Newc
2 Liv Ain
0 Antrim
5 Liv RI
1 Stoke
5 Belfast
4 Sund
3 Chelms
17 Edinb
0 Glouc
4 Covnt
3 Derby
6 Camb
0 Sheff
0 Plymth
4 Dudley
0 Sthend
0 Prestn
0 L Barts
1 Swanse
35 D&Gall
10 Salford
11 West NI
1 Stevng
1 Nottm
44 L Guys
3 M RI
36 Glasgw
0 Abrdn
11 Dundee
0 Bristol
0 Ipswi
48 Colchr
1 L Kings
0 B Heart
1 Oxford
0 Donc
0 York
2 Leic
11 L St.G
2 Leeds
6 B QEH
32 Airdrie
0 Basldn
22 Klmarnk
2 Middlbr
7 England
4 N Ireland
25 Scotland
1 Wales
8 UK
Centre
Fig. 10.1. Median haemoglobin for incident dialysis patients at start of dialysis treatment in 2012
196
100
N = 5,268 Upper 95% Cl
90
% with Hb >100 g/L
80 Lower 95% Cl
70
60
50
40
30
20
10
0
0 Carlis
0 Ulster
2 Wirral
21 L West
3 Wrexm
5 Bangor
5 Norwch
12 Hull
0 Clwyd
3 Brightn
0 Truro
0 Redng
0 Newry
0 Cardff
23 Dunfn
1 Carsh
0 Ports
3 Chelms
3 Bradfd
2 Liv Ain
2 Newc
3 Dorset
0 Shrew
0 Exeter
17 Edinb
5 Belfast
4 Covnt
32 L Rfree
1 Stoke
1 Wolve
0 Antrim
3 Derby
0 Glouc
4 Sund
5 Liv RI
0 Plymth
4 Dudley
6 Camb
0 Sheff
0 L Barts
1 Nottm
0 Sthend
1 Stevng
3 M RI
10 Salford
36 Glasgw
1 Swanse
0 Abrdn
35 D&Gall
0 Bristol
0 Prestn
22 Klmarnk
0 Ipswi
44 L Guys
1 Oxford
11 Dundee
1 L Kings
0 Donc
0 Basldn
32 Airdrie
11 L St.G
0 B Heart
2 Leic
11 West NI
2 Leeds
6 B QEH
48 Colchr
0 York
2 Middlbr
7 England
4 N Ireland
25 Scotland
1 Wales
8 UK
Centre
Fig. 10.2. Percentage of incident dialysis patients with Hb 5100 g/L at start of dialysis treatment in 2012
125 100
120 90
80
115
Haemoglobin g/L
70
110
60
105
50
PD – early PD – early
100
PD – late 40 PD – late
HD – early HD – early
95 30
HD – late HD – late
90 20
Start 3 months 6 months 12 months Start 3 months 6 months 12 months
Time since commencing dialysis Time since commencing dialysis
Fig. 10.3. Median haemoglobin, by time on dialysis and length of Fig. 10.4. Percentage of incident dialysis patients in 2011 with Hb
pre-RRT care, for incident dialysis patients in 2011 5100 g/L, by time on dialysis and by length of pre-RRT care
100
90
100
80
Percentage of incident patients
70 90
60
Percentage on ESA
80
50
40 70
30
60 PD – early
20 <90 90–99.9 PD – late
100–109.9 110–119.9 HD – early
10 50
>120 (Hb, g/L)
HD – late
0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 40
Start 3 months 6 months 12 months
Year of start Time since commencing dialysis
Fig. 10.5. Distribution of haemoglobin in incident dialysis patients Fig. 10.6. Percentage of incident dialysis patients in 2011 on ESA,
by year of start by time on dialysis and by length of pre-RRT care
197
Table 10.2. Percentage completeness of data returns for haemoglobin and serum ferritin and percentages on ESA for prevalent HD and
PD patients in 2012
HD PD
Centre N Hb Ferritin % on ESA N Hb Ferritin % on ESA
England
B Heart 401 100 100 77 42 100 98 48
B QEH 864 97 96 84 149 99 97 62
Basldn 150 98 97 91 28 100 100 61
Bradfd 189 98 98 96 24 100 100 83
Brightn 338 96 86 0 69 94 83 0
Bristol 461 100 100 92 56 100 100 66
Camb 324 95 76 43 32 100 97 59
Carlis 57 100 70 68 21 100 95 67
Carsh 698 95 92 0 97 98 99 0
Chelms 121 100 99 97 25 100 100 76
Colchr 108 93 95 29
Covnt 335 100 99 91 84 96 89 68
Derby 209 100 99 0 84 100 99 0
Donc 158 100 100 91 23 100 100 70
Dorset 244 100 98 97 38 95 87 68
Dudley 153 100 99 3 53 100 89 4
Exeter 351 100 100 93 69 100 100 72
Glouc 193 100 98 91 31 100 77 55
Hull 310 100 99 0 79 97 95 0
Ipswi 124 100 99 65 30 100 90 70
Kent 361 100 99 91 55 100 96 67
L Barts 846 100 99 0 167 99 95 0
L Guys 592 91 81 19 27 96 96 7
L Kings 460 100 97 0 76 100 99 0
L Rfree 668 86 81 0 102 99 86 0
L St.G 271 97 92 0 48 98 96 0
L West 1,342 98 99 0 47 98 98 0
Leeds 454 100 100 94 77 100 100 78
Leic 801 100 100 98 143 98 98 80
Liv Ain 166 99 98 0 17 100 100 0
Liv RI 345 99 99 0 55 98 96 0
M RI 474 93 92 0 76 100 100 0
Middlbr 312 98 98 78 8 88 88 75
Newc 262 100 100 69 37 86 92 0
Norwch 303 100 98 91 48 100 98 71
Nottm 355 100 100 90 72 100 100 69
Oxford 389 100 100 93 69 100 99 81
Plymth 119 100 98 0 31 97 77 0
Ports 510 100 99 10 78 100 100 12
Prestn 496 100 99 88 59 100 100 75
Redng 251 100 100 90 63 100 98 2
Salford 345 88 0 68 90 93 0 77
Sheff 562 100 100 86 67 100 100 60
Shrew 184 100 99 88 33 97 94 61
Stevng 380 99 99 0 27 100 89 0
Sthend 107 100 100 97 14 100 100 57
Stoke 294 86 99 1 69 100 99 0
Sund 184 99 93 95 17 100 94 65
Truro 134 99 99 0 19 100 89 0
Wirral 177 98 97 0 29 79 62 0
198
HD PD
Centre N Hb Ferritin % on ESA N Hb Ferritin % on ESA
Wolve 270 100 99 85 83 100 100 63

York 122 100 100 93 27 100 96 70
N Ireland
Antrim 126 100 100 92 10 100 100 80
Belfast 208 99 97 90 25 100 96 80
Newry 85 99 28 95 14 100 100 86
Ulster 101 100 100 93 6 100 100 100
West NI 129 98 59 92 15 100 100 67
Scotland
Abrdn 214 100 93 20 100
Airdrie 176 100 97 10 100
D & Gall 48 100 98 14 93
Dundee 171 99 88 19 95
Dunfn 140 100 89 20 95
Edinb 250 100 93 35 100
Glasgw 579 99 72 40 100
Inverns 73 100 64 15 93
Klmarnk 141 100 91 40 100
Wales
Bangor 82 100 100 79 14 100 100 50
Cardff 448 100 99 61 71 100 73 27
Clwyd 76 100 100 0 15 100 93 0
Swanse 308 100 100 92 54 100 89 78
Wrexm 86 100 73 91 20 100 45 55
England 18,324 98 95 88 2,864 98 92 69
N Ireland 649 99 82 92 70 100 99 80
Scotland 1,792 100 85 213 98
Wales 1,000 100 97 76 174 100 79 68
UK 21,765 98 93 87∗ 3,321 99 94∗ 69∗
∗
The overall averages given are for E, W & NI (not UK)
Blank cells denote centres with no PD patients or because data was not available
Percentages on ESA are shown, but it is believed that there were data problems for those centres with apparently less than 60% of HD patients or
45% of PD patients on ESA
The country level averages for the % on ESA are based only on those centres whose % was above the limits mentioned above
generally good for Hb and ferritin. The percentages on Haemoglobin in prevalent haemodialysis patients
ESA are shown as they appear in the data received by The median Hb of patients on HD in the UK was
the registry. For some centres, the ESA data was com- 112 g/L with an IQR of 103–121 g/L and 82% of HD
pletely missing and for others it appears to be partially patients had a Hb 5100 g/L (table 10.3). The median
complete with, for example, only 10 or 20% of patients Hb by centre is shown in figure 10.7. Compliance with
appearing to be on ESAs. It is believed that there were the target range of Hb 5100 and 4120 g/L continues
problems with data entry and/or data transfer in those to increase year on year, 52.7% in 2010, 56.1% in 2011
centres with apparently less than 60% of HD patients and 57% in 2012 (figure 10.8). The percentages of HD
or 45% of PD patients on ESA. These centres have been patients with Hb below 100 g/L and above 120 g/L, as
excluded from further analyses of ESA use. well as the percentages meeting the target, are shown
Summary statistics for haemoglobin, serum ferritin by centre in figure 10.9.
and ESA are shown for the 71 renal centres in the UK Funnel plots are shown for the minimum (Hb
in tables 10.3 for HD and 10.4 for PD patients 5100 g/L) and target range (Hb 5100 and 4120 g/L)
respectively. in figures 10.10 and 10.11 respectively. Many centres
199
Table 10.3. Summary statistics for haemoglobin, serum ferritin and ESA for prevalent HD patients in 2012
% Hb Median % % ferritin Median % with Hb

N with Median % Hb 100– ferritin ferritin .200 and % on ESA dose 5100 g/L and
Centre Hb data Hb g/L 5100 g/L 120 g/L mg/L 5100 mg/L 4500 mg/L ESA (IU/week) not on ESA
England
B Heart 401 108 70 52 333 94 57 77 6,667 21
B QEH 838 111 82 59 354 95 77 84 7,000 14
Basldn 147 108 67 47 339 93 72 91 6,000 6
Bradfd 186 112 78 52 497 95 39 96 6,500 4
Brightn 323 110 81 65 510 99 45
Bristol 461 113 85 57 564 96 31 92 7,500 8
Camb 309 113 85 59 306 88 56
Carlis 57 115 84 42 439 93 50 68 4,750 32
Carsh 660 111 84 70 375 95 63
Chelms 121 118 93 50 631 100 22 97 10,000 3
Colchr 100 117 89 50 500 99 48
Covnt 335 110 78 61 336 95 67 91 11,000 8
Derby 208 113 84 61 428 97 47
Donc 158 111 77 53 401 99 59 91 6,500 9
Dorset 244 115 85 52 453 97 51 97 9,250 3
Dudley 153 111 76 50 333 95 70
Exeter 351 112 83 62 265 90 62 93 7,500 6
Glouc 193 111 83 63 330 89 49 91 8
Hull 309 116 88 51 393 99 64
Ipswi 124 111 80 55 611 98 28 65 7,500 29
Kent 361 113 86 59 445 93 38 91 8,250 7
L Barts 844 109 76 61 432 95 53
L Guys 537 107 71 55 693 97 26
L Kings 460 107 73 61 579 98 35
L Rfree 576 112 84 58 425 91 41
L St.G 263 111 80 59 458 97 47
L West 1,314 117 91 54 477 99 50
Leeds 454 110 78 57 499 95 39 94 4,000 5
Leic 799 113 83 54 337 95 63 98 6,190 1
Liv Ain 164 110 78 59 703 98 22
Liv RI 343 118 83 41 475 92 35
M RI 439 114 82 53 396 94 56
Middlbr 307 112 79 56 676 94 22 78 5,000 18
Newc 262 116 84 50 424 95 43 69 11,025 28
Norwch 302 115 87 59 444 93 35 91 8,000 9
Nottm 354 113 84 62 582 99 24 90 7,500 10
Oxford 389 112 81 55 308 94 57 93 8,000 6
Plymth 119 112 83 60 752 97 22
Ports 509 117 89 49 357 97 67
Prestn 494 113 83 58 577 94 30 88 11
Redng 251 116 84 56 536 98 38 90 8
Salford 303 108 73 58 68 6,000 14
Sheff 562 112 79 54 488 96 45 86 7,500 11
Shrew 184 115 89 55 391 98 57 88 11
Stevng 376 114 86 60 521 97 37
Sthend 107 111 82 66 313 98 72 97 9,000 3
Stoke 254 115 84 54 405 97 49
Sund 183 111 81 56 615 95 26 95 5
Truro 133 111 83 66 460 97 52
Wirral 173 112 82 62 537 98 35
Wolve 269 115 86 53 473 96 44 85 6,750 14
York 122 110 75 57 414 97 69 93 4,000 6
200

N Ireland
Antrim 126 115 88 60 469 98 52 92 6,000 7
Belfast 205 111 78 57 434 95 41 90 8,000 7
Newry 84 112 86 62 95 4,300 5
Ulster 101 113 86 61 677 99 20 93 5,875 6
West NI 126 111 79 61 640 93 17 92 8,000 8
Scotland
Abrdn 213 108 69 50 634 99 32
Airdrie 176 113 86 62 669 99 30
D & Gall 48 108 85 67 648 96 23
Dundee 170 113 82 64 289 84 47
Dunfn 140 118 92 50 622 90 21
Edinb 249 119 91 47 372 94 47
Glasgw 573 115 85 53 437 96 44
Inverns 73 116 97 59 426 98 57
Klmarnk 141 113 82 52 332 91 54
Wales
Bangor 82 116 89 59 432 96 54 79 9,000 17
Cardff 447 112 83 58 301 94 64 61 33
Clwyd 76 113 89 61 358 100 68
Swanse 308 112 85 66 386 94 45 91 7,500 8
Wrexm 86 113 87 58 485 97 43 92 5,000 8
England 17,885 112 82 57 432 96 48 88 7,333 10
N Ireland 642 112 82 60 535 96 35 92 6,500 7
Scotland 1,783 114 85 54 448 94 40
Wales 999 113 85 60 348 95 56 76 7,500 21
UK 21,309 112 82 57 431 95 48 87 7,248 11
Blank cells denote centres excluded from analyses due to poor data completeness or low patient numbers or because the data item was not available
ESA data only shown for those centres for which the % on ESA was 60% or more
For ESA, the overall averages given are for E, W & NI not UK
135
130 Median Hb
Lower quartile
125
Haemoglobin g/L
120
115
110
105
100
95
0 Edinb
1 Liv RI
0 Chelms
0 Dunfn
0 Ports
2 L West
7 Colchr
0 Hull
0 Redng
0 Bangor
0 Newc
0 Inverns
0 Antrim
0 Dorset
14 Stoke
0 Wolve
0 Carlis
0 Shrew
0 Norwch
1 Glasgw
1 Stevng
7 M RI
0 Ulster
0 Nottm
0 Bristol
0 Leic
5 Camb
0 Clwyd
0 Wrexm
0 Prestn
0 Kent
0 Klmarnk
0 Airdrie
1 Dundee
0 Derby
1 Newry
2 Bradfd
14 L Rfree
2 Wirral
2 Middlbr
0 Exeter
0 Sheff
0 Plymth
0 Oxford
0 Swanse
0 Cardff
1 Belfast
2 West NI
0 Sthend
5 Carsh
1 Truro
0 Ipswi
0 Glouc
3 L St.G
1 Sund
0 Dudley
0 Donc
3 B QEH
4 Brightn
0 Covnt
0 Leeds
1 Liv Ain
0 York
0 L Barts
2 Basldn
0 B Heart
12 Salford
0 Abrdn
0 D&Gall
9 L Guys
0 L Kings
2 England
1 N Ireland
0 Scotland
0 Wales
2 UK
Centre
Fig. 10.7. Median haemoglobin in patients treated with HD by centre in 2012
201
80 Upper 95% CI
N = 21,309 % with Hb 100–120 g/L
70 Lower 95% CI
60
50
40
30
20
5 Carsh
0 D&Gall
0 Sthend
1 Truro
0 Swanse
4 Brightn
1 Dundee
0 Glouc
0 Nottm
0 Exeter
0 Airdrie
1 Newry
2 Wirral
0 L Barts
0 Ulster
0 Covnt
2 West NI
0 Derby
0 L Kings
0 Clwyd
1 Stevng
0 Plymth
0 Antrim
0 Kent
3 B QEH
1 Liv Ain
3 L St.G
0 Inverns
5 Camb
0 Norwch
0 Bangor
0 Wrexm
12 Salford
0 Cardff
0 Prestn
14 L Rfree
0 York
0 Leeds
0 Bristol
1 Belfast
1 Sund
0 Redng
2 Middlbr
0 Shrew
0 Ipswi
0 Oxford
9 L Guys
2 L West
14 Stoke
0 Leic
0 Sheff
0 Donc
7 M RI
0 Wolve
1 Glasgw
0 B Heart
2 Bradfd
0 Klmarnk
0 Dorset
0 Hull
0 Chelms
0 Abrdn
7 Colchr
0 Dunfn
0 Dudley
0 Newc
0 Ports
0 Edinb
2 Basldn
0 Carlis
1 Liv RI
2 England
1 N Ireland
0 Scotland
0 Wales
2 UK
Centre
Fig. 10.8. Percentage of HD patients with Hb 5100 and 4120 g/L by centre in 2012
complied well with respect to both the minimum and compliance with Hb 5100 and 4120 g/L is shown in
target range Hb standards. Some centres complied well figure 10.13. In 2012, 55% of prevalent PD patients had
with the percentage with Hb 5100 g/L (figure 10.10) a Hb within the target range. The distribution of Hb in
but had a poor compliance with percentage of patients PD patients by centre is shown in figure 10.14. The funnel
with Hb 5100 and 4120 g/L (figure 10.11). This plots for percentage with Hb 5100 g/L and for the
demonstrates that compliance with one standard can be percentage of patients with Hb 5100 and 4120 g/L
achieved without compliance with another standard. are shown in figures 10.15 and 10.16 respectively.
Table 10.3 can be used in conjunction with figures 10.10 Table 10.4 can be used in conjunction with figures 10.15
and 10.11 to identify centres. and 10.16 to identify centres in the funnel plot.
Haemoglobin in prevalent peritoneal dialysis patients Relationship between Hb in incident and prevalent dialysis
Overall, 85% of patients on PD had a Hb 5100 g/L patients in 2012
(table 10.4). The median Hb of patients on PD in the The relationship between the percentage of incident
UK in 2012 was 114 g/L with an IQR of 105–123 g/L. and prevalent dialysis (HD and PD) patients with a Hb
The median Hb by centre is shown in figure 10.12. The 5100 g/L is shown in figure 10.17. As expected, all
100
Hb >120 g/L
90 Hb 100–120 g/L
Hb <100 g/L
80
70
60
50
40
30
20
10
0
Belfast
Carsh
D&Gall
Sthend
Truro
Swanse
Brightn
Dundee
Glouc
Nottm
Exeter
Airdrie
Newry
Wirral
L Barts
Ulster
Covnt
West NI
Derby
L Kings
Clwyd
Stevng
Plymth
Antrim
Kent
B QEH
Liv Ain
L St.G
Camb
Inverns
Norwch
Bangor
Wrexm
Salford
Prestn
L Rfree
York
Leeds
Bristol
Sund
Redng
Middlbr
Shrew
Ipswi
L Guys
L West
Stoke
Leic
Sheff
Donc
M RI
Wolve
Glasgw
B Heart
Bradfd
Klmarnk
Dorset
Hull
Chelms
Abrdn
Colchr
Dunfn
Dudley
Newc
Ports
Edinb
Basldn
Carlis
Liv RI
England
N Ireland
Scotland
Wales
UK
Cardff
Oxford
Centre
Fig. 10.9. Distribution of haemoglobin in patients treated with HD by centre in 2012
202
100
percentages with serum ferritin 5100 mg/L, .200 mg/L
Solid lines show 95% limits and 4500 mg/L, and 5800 mg/L are shown in figures
95 10.20, 10.21 and 10.22 respectively. Most centres
achieved greater than 90% compliance with a serum
90 ferritin 5100 mg/L for HD patients. The HD population
had a median ferritin value of 431 mg/L, IQR 285–623.

85
Seventeen of the 69 centres who had returns for ferritin
had greater than 20% (21–47%) of their patients with
80
ferritin 5800 mg/L (figure 10.22). The serum ferritin
correlated poorly with median Hb achieved and ESA
75
dose (table 10.3).
70
Ferritin in prevalent peritoneal dialysis patients
65
The median and IQR for serum ferritin for patients
0 200 400 600 800 1,000 1,200 treated with PD are shown in figure 10.23. The percen-
Number of patients with data in centre tages with serum ferritin 5100 mg/L, .100 mg/L and
Fig. 10.10. Funnel plot of percentage of HD patients with Hb 4500 mg/L, and 5800 mg/L are shown in figures 10.24,
5100 g/L by centre in 2012 10.25 and 10.26 respectively. The PD population had a
lower median ferritin value (285 mg/L, IQR 164–466)
centres had a higher percentage of prevalent patients than the HD population. In 2012, 31 centres reported
achieving a Hb 5100 g/L than that for incident patients. less than 90% of PD patients compliant with serum
Overall in the UK, 83% of prevalent patients, compared ferritin 5100 mg/L, although this had little bearing on
with 51% of incident patients, had a Hb 5100 g/L in their achieved median Hb or median ESA dose when
2012. Compliance with ‘current’ minimum standards compared with other centres (table 10.4).
by year (1998–2012) for incident and prevalent patients
(all dialysis patients) is shown in figure 10.18. The decline Erythropoietin stimulating agents in prevalent haemodialysis
in achieving this standard for incident and prevalent patients
patients continues. As shown in previous reports there was substantial
variation in the average dose of ESA prescription used.
Ferritin in prevalent haemodialysis patients The median dose for prevalent HD patients in England,
The median and IQR for serum ferritin for patients Wales and Northern Ireland was 7,248 IU/week. The
treated with HD are shown in figure 10.19. The median dose varied from 4,000 IU/week (Leeds, York)
to 11,025 IU/week (Newcastle) with a median Hb for
80 these centres of 110 g/L (Leeds, York) and 116 g/L
Solid lines show 95% limits (Newcastle) (table 10.3). Over the last three years there
has been a fall in the median ESA dose, 8000 IU in
70
2010, 7,450 IU in 2011 and 7,248 IU in 2012.
60 Erythropoietin stimulating agents in prevalent peritoneal

dialysis patients
In 2012, the median dose was substantially lower in
50
prevalent PD patients at 4,250 (range 2,231–9,500)
IU/week (table 10.4) compared with HD patients.
40
ESA prescription and association with achieved haemoglobin
For HD patients, centre level median Hb is plotted
30
0 200 400 600 800 1,000 1,200 against median ESA dose in figure 10.27 and compliance
Number of patients with data in centre with the RA standards for Hb 5100 g/L and 4120 g/L is
Fig. 10.11. Funnel plot of percentage of HD patients with Hb plotted against median ESA dose in figure 10.28. For
5100 and 4120 g/L by centre in 2012 these figures, Hb data was only used for those patients
203
Table 10.4. Summary statistics for haemoglobin, serum ferritin and ESA for prevalent PD patients in 2012

England
B Heart 42 114 86 57 182 85 73 48 6,000 50
B QEH 147 114 81 49 308 85 66 62 5,000 37
Basldn 28 112 71 46 189 82 68 61 3,750 39
Bradfd 24 111 83 58 302 88 54 83 4,000 17
Brightn 65 113 88 52 314 95 74
Bristol 56 112 73 50 383 95 66 66 4,885 32
Camb 32 114 91 63 334 90 65 59 3,600 41
Carlis 21 116 95 62 346 95 65 67 4,125 33
Carsh 95 112 81 56 173 79 72
Chelms 25 119 96 48 200 76 60 76 4,000 24
Colchr n/a
Covnt 81 114 89 62 257 84 72 68 8,000 30
Derby 84 114 81 52 341 94 63
Donc 23 113 78 52 266 96 65 70 4,000 30
Dorset 36 120 92 44 347 94 61 68 2,900 31
Dudley 53 112 85 53 150 68 66
Exeter 69 114 96 67 212 83 74 72 4,000 28
Glouc 31 114 84 58 173 75 71 55 35
Hull 77 114 84 48 295 99 75
Ipswi 30 116 87 50 390 85 44 70 3,000 30
Kent 55 113 85 55 259 83 68 67 4,000 31
L Barts 165 113 78 44 307 89 63
L Guys 26 112 81 58 207 81 73
L Kings 76 110 84 58 219 83 77
L Rfree 101 110 76 53 430 95 49
L St.G 47 114 87 55 317 93 87
L West 46 114 83 46 251 89 76
Leeds 77 114 88 62 328 92 74 78 3,333 22
Leic 140 115 86 56 344 95 74 80 3,900 14
Liv Ain 17 112 76 53 434 100 59
Liv RI 54 115 83 52 325 85 49
M RI 76 116 84 54 174 83 70
Middlbr 7
Newc 32 114 88 50 426 97 50
Norwch 48 117 96 58 131 68 53 71 3,725 29
Nottm 72 113 83 60 339 93 71 69 3,333 29
Oxford 69 113 80 55 179 87 76 81 6,000 16
Plymth 30 119 90 53 345 92 58
Ports 78 119 95 53 310 96 72
Prestn 59 115 85 58 339 83 54 75 25
Redng 63 116 87 54 378 92 65
Salford 84 112 86 56 77 9,500 19
Sheff 67 113 85 60 538 97 42 60 5,292 39
Shrew 32 116 84 44 214 74 61 61 4,000 41
Stevng 27 109 78 59 196 75 63
Sthend 14 117 93 57 241 100 100 57 7,500 43
Stoke 69 115 86 59 447 94 50
Sund 17 117 82 41 570 94 25 65 2,231 29
Truro 19 114 89 63 268 100 82
Wirral 23 113 87 57 497 94 44
Wolve 83 116 88 51 244 76 54 63 4,000 36
York 27 109 81 59 170 88 73 70 4,000 30
204

N Ireland
Antrim 10 115 100 70 239 80 60 80 3,833 20
Belfast 25 114 88 56 221 96 75 80 3,000 20
Newry 14 108 86 71 192 64 57 86 2,458 14
Ulster 6
West NI 15 122 93 40 277 100 73 67 2,500 33
Scotland
Abrdn 20 115 85 55
Airdrie 10 113 90 70
D & Gall 13 115 92 69
Dundee 18 109 78 72
Dunfn 19 118 84 42
Edinb 35 113 86 60
Glasgw 40 113 90 60
Inverns 14 116 100 79
Klmarnk 40 111 73 45
Wales
Bangor 14 117 86 43 179 57 50 50 4,000 50
Cardff 71 110 87 65 151 67 63
Clwyd 15 108 73 53 238 86 64
Swanse 54 111 87 69 328 85 63 78 4,500 22
Wrexm 20 121 85 35 55 8,000 40
England 2,819 114 85 54 288 88 66 69 4,500 29
N Ireland 70 115 91 56 239 88 67 80 3,000 20
Scotland 209 114 85 58
Wales 174 112 86 60 198 76 64 68 6,000 31
UK 3,272 114 85 55 285 88 65 69 4,250 29
Blank cells denote centres excluded from analyses due to poor data completeness or low patient numbers or because the data item was not available
n/a – no PD patients
ESA data only shown for those centres for which the % on ESA was 45% or more
For ferritin and for ESA the overall averages given are for E, W & NI not UK

130 Median Hb
Lower quartile
Haemoglobin g/L
120
110
100
90
0 West NI
0 Wrexm
5 Dorset
3 Plymth
0 Chelms
0 Ports
5 Dunfn
0 Sthend
0 Sund
0 Bangor
0 Norwch
0 Ipswi
0 Redng
0 Wolve
0 Carlis
3 Shrew
0 M RI
7 Inverns
0 Antrim
2 Leic
0 Stoke
0 Prestn
7 D&Gall
2 Liv RI
0 Abrdn
0 Belfast
0 Truro
3 Hull
0 Derby
0 Camb
0 Glouc
0 Exeter
2 L St.G
4 Covnt
0 B Heart
0 Leeds
1 B QEH
14 Newc
2 L West
21 Wirral
0 Nottm
6 Brightn
0 Sheff
1 L Barts
0 Oxford
0 Donc
0 Kent
0 Glasgw
0 Edinb
0 Airdrie
2 Carsh
0 Bristol
4 L Guys
7 Salford
0 Dudley
0 Liv Ain
0 Basldn
0 Swanse
0 Klmarnk
0 Bradfd
1 L Rfree
0 L Kings
0 Cardff
0 Stevng
0 York
5 Dundee
0 Clwyd
0 Newry
2 England
0 N Ireland
2 Scotland
0 Wales
1 UK
Centre
Fig. 10.12. Median haemoglobin in patients treated with PD by centre in 2012
205
206
60
65
70
75
80
85
90
95
100
0
Percentage of patients Percentage of patients
0
10
20
30
40
50
60
70
80
90
100
10
20
30
40
50
60
70
80
90
Inverns 7 Inverns
100
Dundee 5 Dundee
25
Newry 0 Newry
Antrim 0 Antrim
Airdrie 0 Airdrie
D&Gall 7 D&Gall
Swanse 0 Swanse
50
Exeter 0 Exeter
Cardff 0 Cardff
5100 g/L by centre in 2012

Truro 0 Truro
Camb 0 Camb
Leeds 0 Leeds
Carlis 0 Carlis
75
Covnt 4 Covnt
Glasgw 0 Glasgw
Edinb 0 Edinb
Nottm 0 Nottm
Sheff 0 Sheff
Stoke 0 Stoke
100
Stevng 0 Stevng
York 0 York
Bradfd 0 Bradfd
Norwch 0 Norwch
Glouc 0 Glouc
Number of patients with data in centre

L Kings 0 L Kings
125
L Guys 4 L Guys
Prestn 0 Prestn
Sthend 0 Sthend
B Heart 0 B Heart

Wirral 21 Wirral
150
Leic 2 Leic

Belfast 0 Belfast
Salford 7 Salford
Fig. 10.15. Funnel plot of percentage of PD patients with Hb

Carsh 2 Carsh
L St.G 2 L St.G
Oxford 0 Oxford
Percentage of patients Abrdn 0 Abrdn
Kent 0 Kent
Centre
Centre
Redng 0 Redng
20
30
40
50
60
70
80
90
M RI 0 M RI
0
L Rfree 1 L Rfree
Plymth 3 Plymth
Clwyd 0 Clwyd
Liv Ain 0 Liv Ain
Dudley 0 Dudley
Ports 0 Ports
25
Derby 0 Derby
Fig. 10.14. Distribution of haemoglobin in patients treated with PD by centre in 2012

Fig. 10.13. Percentage of PD patients with Hb 5100 and 4120 g/L by centre in 2012
Brightn 6 Brightn
Donc 0 Donc
Liv RI 2 Liv RI
Wolve 0 Wolve
Bristol 0 Bristol
50
Ipswi 0 Ipswi
N = 3,272
Newc 14 Newc
B QEH 1 B QEH
Hull 3 Hull
Chelms 0 Chelms
75
Basldn 0 Basldn
L West 2 L West
Klmarnk 0 Klmarnk
Dorset 5 Dorset
Shrew 3 Shrew
L Barts 1 L Barts
5100 g/L and 4120 g/L by centre in 2012

Lower 95% Cl
Upper 95% Cl
0 Bangor
100
Bangor
Dunfn 5 Dunfn
Sund 0 Sund
West NI 0 West NI
Wrexm 0 Wrexm
Number of patients with data in centre

England 2 England
125
N Ireland 0 N Ireland
Hb <100 g/L
Hb >120 g/L
Scotland 2 Scotland
Wales 0 Wales
Hb 100–120 g/L
% with Hb >100 and <120 g/L
UK 1 UK
150
Fig. 10.16. Funnel plot of percentage of PD patients with Hb

30
40
50
60
70
80
90
100
Ferritin µg/L Percentage of patients
0
10
20
30
40
50
60
70
80
90
100
0
200
400
600
800
1,000
1,200
1998
Chapter 10
2 Plymth 1999 Carlis

2 Liv Ain 2000 Ulster
19 L Guys Wirral
0 Ulster 2001 L West
2 Middlbr Wrexm
3 Airdrie 2002 Bangor
2 D&Gall 2003 Norwch
41 West NI Hull
7 Abrdn 2004 Clwyd
1 Chelms 2005 Brightn
11 Dunfn Truro
7 Sund 2006 Redng
1 Ipswi 2007 Newry
0 Nottm Cardff
Incident patients
3 L Kings 2008 Dunfn
1 Prestn Carsh
0 Bristol 2009 Ports
3 Wirral 2010 Chelms
0 Redng Bradfd
1 Stevng 2011 Liv Ain
14 Brightn 2012 Newc
5 Colchr Dorset
0 Leeds Shrew
Year
2 Bradfd 1998 Exeter
0 Sheff Edinb
Lower 95% Cl
Upper 95% Cl
27 Wrexm 1999 Belfast
1 L West Covnt
1 Liv RI 2000 L Rfree
% with Hb >100 g/L

1 Wolve 2001 Stoke
0 Antrim Wolve
1 Truro 2002 Antrim
8 L St.G 2003 Derby
2 Dorset Glouc
1 Kent 2004 Sund
2 Norwch 2005 Liv RI
30 Carlis Plymth
28 Glasgw 2006 Dudley
3 Belfast Camb
Centre
Centre
1 L Barts 2007 Sheff
Prevalent patients
Fig. 10.19. Median ferritin in patients treated with HD by centre in 2012

0 Bangor 2008 L Barts
1 Derby Nottm
36 Inverns 2009 Sthend
19 L Rfree 2010 Stevng
0 Newc M RI
0 York 2011 Salford
1 Stoke 2012 Glasgw
0 Donc Swanse
8 M RI Abrdn
1 Hull D&Gall
1 Shrew Bristol
0 Swanse Prestn
8 Carsh Klmarnk
7 Edinb Ipswi
0 Clwyd L Guys
1 Ports Oxford
4 B QEH Dundee
3 Basldn L Kings
0 Leic Donc
Fig. 10.17. Percentage of incident and prevalent dialysis patients with Hb 5100 g/L by centre in 2012
1 Covnt Basldn
0 B Heart Airdrie
1 Dudley L St.G
9 Klmarnk B Heart
N = 20,332
with Hb 5100 g/L
2 Glouc Leic
0 Sthend West NI
0 Oxford Leeds
24 Camb B QEH
1 Cardff Colchr
12 Dundee York
0 Exeter Middlbr
6 England England
18 N Ireland N Ireland
15 Scotland Scotland
Incident dialysis patients
3 Wales Wales
Lower quartile
Upper quartile
Median ferritin
Prevalent dialysis patients
7 UK UK
prevalent dialysis patients (1998–2012)
Fig. 10.18. Percentage of incident and
207
Anaemia management in UK dialysis patients
208
Percentage of patients Percentage of patients Percentage of patients
0
10
20
30
40
50
60
0
10
20
30
40
50
60
70
80
90
75
80
85
90
95
0 York 4 B QEH 0 Clwyd 100
24 Camb 0 Sthend 1 Chelms
1 Dudley 3 Basldn 3 Airdrie
3 Basldn 1 Dudley 1 L West
36 Inverns 0 York 5 Colchr
0 Exeter 0 Clwyd 0 Ulster
0 Clwyd 1 Covnt 7 Abrdn
0 Sthend 1 Ports 0 Nottm
4 B QEH 1 Hull 0 Donc
1 Cardff 1 Cardff 1 Hull
Lower 95% Cl
Upper 95% Cl
Upper 95% Cl
Lower 95% Cl
0 Donc 0 Leic 14 Brightn
1 Covnt 8 Carsh 2 Liv Ain
8 Carsh 0 Exeter 0 Sthend
0 B Heart 0 Donc 36 Inverns
0 Leic 36 Inverns 1 Shrew
0 Oxford 0 Oxford 3 L Kings
% with ferritin >800 µg/L

1 Ports 1 Shrew 3 Wirral
1 Truro 0 B Heart 0 Antrim
1 Hull 8 M RI 0 Redng
12 Dundee 24 Camb 1 Ipswi
9 Klmarnk 9 Klmarnk 2 Plymth
8 M RI 0 Bangor 1 Truro
2 Glouc 1 L Barts 1 Stoke
1 Shrew 0 Antrim 19 L Guys
N = 20,332
N = 20,332
1 Wolve 1 Truro 1 Ports
1 L Barts 2 Dorset 1 Stevng
0 Bangor 1 L West 27 Wrexm
1 Derby 30 Carlis 8 L St.G
1 L West 1 Stoke 0 York
0 Sheff 2 Glouc 2 Dorset
1 Stoke 5 Colchr 1 Derby
5 Colchr 7 Edinb 0 Bangor
0 Antrim 1 Derby 1 Wolve
7 Edinb 8 L St.G 0 Sheff
0 Nottm 12 Dundee 28 Glasgw
0 Swanse 0 Swanse 0 Bristol
30 Carlis 0 Sheff 2 D&Gall
2 Dorset 14 Brightn 1 L Barts
Centre
Centre
Centre
28 Glasgw 28 Glasgw 4 B QEH

0 Newc 1 Wolve 0 Leeds
0 Leeds 0 Newc 0 Newc
2 Bradfd 27 Wrexm 0 Leic
27 Wrexm 3 Belfast 1 Covnt
3 Wirral 19 L Rfree 7 Sund
Fig. 10.22. Percentage of HD patients with ferritin 5800 mg/L by centre in 2012
Fig. 10.20. Percentage of HD patients with ferritin 5100 mg/L by centre in 2012
14 Brightn 2 Bradfd 1 Dudley

0 Bristol 0 Leeds 2 Bradfd
8 L St.G 1 Kent 8 Carsh
19 L Rfree 0 Redng 3 Belfast
2 Norwch 1 Stevng 2 Middlbr
0 Redng 2 Norwch 7 Edinb
1 Stevng 3 Wirral 0 Oxford
N = 20,332
1 Kent 1 Liv RI 8 M RI
3 Belfast 3 L Kings 0 B Heart
1 Ipswi 7 Abrdn 1 Cardff
1 Chelms 0 Bristol 1 Prestn
Fig. 10.21. Percentage of HD patients with ferritin .200 mg/L and 4500 mg/L by centre in 2012
1 Liv RI 3 Airdrie 0 Swanse
3 L Kings 1 Prestn 41 West NI
0 Ulster 1 Ipswi 3 Basldn
2 D&Gall 7 Sund 1 Kent
1 Prestn 19 L Guys 2 Norwch
41 West NI 0 Nottm 30 Carlis
7 Abrdn 2 D&Gall 1 Liv RI
Lower 95% CI
Upper 95% CI
7 Sund 2 Middlbr 19 L Rfree

2 Liv Ain 2 Plymth 9 Klmarnk
3 Airdrie 1 Chelms 11 Dunfn
19 L Guys 2 Liv Ain 0 Exeter
11 Dunfn 11 Dunfn 2 Glouc
2 Middlbr 0 Ulster 24 Camb
2 Plymth 41 West NI 12 Dundee
6 England 6 England 6 England
% with ferritin >200 and <500 µg/L
7 UK 7 UK 7 UK
Percentage of patients Percentage of patients Ferritin µg/L
0
10
20
30
40
50
60
70
80
90
100
30
40
50
60
70
80
90
100
0
200
400
600
800
Chapter 10
0 Sthend 0 West NI 6 Sund

4 L St.G 0 Sthend 0 Sheff
11 Truro 11 Truro 38 Wirral
1 L Kings 0 Liv Ain 1 Stoke
1 Oxford 5 Hull 0 Liv Ain
2 L West 8 Newc 14 L Rfree
4 Belfast 0 Sheff 8 Newc
5 Hull 0 Ports 10 Ipswi
Lower 95% Cl
Upper 95% Cl
Upper 95% Cl
Lower 95% Cl
2 Leic 4 Belfast 0 Bristol
0 Leeds 0 Donc 2 Redng
0 Exeter 14 L Rfree 13 Dorset
17 Brightn 2 Leic 5 Carlis
0 West NI 5 Carlis 23 Plymth

2 B Heart 17 Brightn 2 Leic
4 York 0 Bristol 1 Derby
4 L Guys 38 Wirral 0 Nottm
11 Covnt 1 Stoke 0 Prestn
% with ferritin >100 and <500 µg/L

1 Carsh 1 Derby 3 Camb
0 Ports 13 Dorset 0 Leeds
N = 2,855
0 Nottm 6 Sund 11 Swanse
23 Glouc 4 L St.G 4 Liv RI
0 M RI 0 Nottm 4 L St.G
4 Kent 0 Leeds 17 Brightn
N = 2,855
0 Basldn 2 Redng 0 Ports
3 B QEH 23 Plymth 3 B QEH
0 Bristol 3 Camb 5 L Barts
11 Dudley 2 L West 0 Bradfd
0 Donc 5 L Barts 5 Hull
5 Carlis 4 York 0 West NI
3 Camb 0 Bradfd 11 Truro
2 Redng 1 Oxford 0 Donc
Centre
7 Clwyd
Centre
Centre
7 Clwyd 4 Kent
Fig. 10.23. Median ferritin in patients treated with PD by centre in 2012
27 Cardff 11 Swanse 11 Covnt

5 L Barts 2 B Heart 2 L West
1 Derby 10 Ipswi 0 Wolve
11 Stevng 4 Liv RI 0 Sthend
Fig. 10.24. Percentage of PD patients with ferritin 5100 mg/L by centre in 2012
11 Swanse 3 B QEH 0 Antrim
6 Shrew 11 Covnt 7 Clwyd
13 Dorset 0 Prestn 4 Belfast
0 Antrim 4 Kent 1 L Kings
0 Chelms 0 M RI 6 Shrew
0 Liv Ain 1 L Kings 0 Exeter
23 Plymth 0 Exeter 4 L Guys
0 Newry 0 Basldn 0 Chelms
0 Prestn 4 L Guys 11 Stevng
Fig. 10.25. Percentage of PD patients with ferritin .100 mg/L and 4500 mg/L by centre in 2012
0 Wolve 0 Antrim 0 Newry
0 Bradfd 1 Carsh 0 Basldn
2 Norwch 0 Chelms 2 B Heart
1 Stoke 0 Wolve 1 Oxford
0 Bangor 11 Stevng 0 Bangor
8 Newc 23 Glouc 0 M RI
N = 2,855
4 Liv RI 6 Shrew 1 Carsh

14 L Rfree 2 Norwch 23 Glouc
38 Wirral 11 Dudley 4 York
10 Ipswi 27 Cardff 27 Cardff
0 Sheff 0 Newry 11 Dudley
6 Sund 0 Bangor 2 Norwch
Lower quartile
Upper quartile
Median ferritin
8 E, W & NI 8 E, W & NI 8 E, W & NI
209
Anaemia management in UK dialysis patients
60
Upper 95% Cl N = 2,855
50
Lower 95% Cl
40
30
20
10
0
0 Newry
0 Sthend
4 York
11 Truro
4 L St.G
7 Clwyd
11 Dudley
0 Bangor
1 Carsh
0 M RI
1 L Kings
0 Exeter
2 Norwch
2 L West
17 Brightn
0 Basldn
4 L Guys
27 Cardff
11 Covnt
3 B QEH
4 Belfast
0 Bradfd
11 Stevng
23 Glouc
1 Oxford
2 B Heart
0 Wolve
13 Dorset
0 Ports
2 Leic
2 Redng
6 Shrew
5 Hull
0 Nottm
5 L Barts
0 Leeds
0 Chelms
11 Swanse
0 Donc
4 Kent
3 Camb
0 Bristol
0 Prestn
1 Derby
23 Plymth
4 Liv RI
1 Stoke
38 Wirral
0 Liv Ain
10 Ipswi
0 Antrim
0 West NI
5 Carlis
0 Sheff
8 Newc
14 L Rfree
6 Sund
8 England
1 N Ireland
21 Wales
8 E, W & NI
Centre
Fig. 10.26. Percentage of PD patients with ferritin 5800 mg/L by centre in 2012
who were receiving an ESA and had dose data available. above. These figures show that 25% of HD patients had
There was no strong relationship in either figure. a Hb .120 g/L. Most of these patients (79%) were on
It is known that not all patients treated with dialysis ESAs. Whereas for PD, 30% of patients had a Hb
who have a Hb above 120 g/L are receiving ESA. It has .120 g/L, but only about 51% of these were on ESAs.
been suggested that it may be inappropriate to include
those patients not receiving ESA within the group not ESA prescription: age and modality associations
meeting this RA target. There are two reasons: firstly, The proportion of patients on an ESA was higher for
the high Hb remains outside the control of the clinician, HD (87%) than PD (69%) and this difference was present
and secondly, the recent trials suggesting that it may be and similar across all age groups (figure 10.31). The
detrimental to achieve a high Hb in renal patients were proportion of patients who maintained a Hb 5100 g/L
based only upon patients treated with ESAs [13, 14]. without requiring ESA (by age group and modality) is
Figures 10.29 and 10.30 show the percentages of HD shown in figure 10.32. This was highest in the 45–54
and PD patients in each centre whose Hb lies above, age group both for HD at 13.6% (95% CI: 12–15.5%)
within or below the RA guidelines of 100–120 g/L. and PD at 33.8% (95% CI: 28–40%).
These charts also show the proportion of patients with
a Hb above the upper limit who were receiving, or were ESAs and time on renal replacement therapy
not receiving an ESA. These analyses are restricted to The percentage of patients on ESA by time on RRT
the centres with acceptable ESA returns as stipulated and dialysis modality is shown in figure 10.33. This is a
120 70
Compliance with Hb 100–120 g/L
115
Median Hb g/L
60
110
50
105
100 40
3,000 5,000 7,000 9,000 11,000 3,000 5,000 7,000 9,000 11,000
Median ESA dose (IU/week) Median ESA dose (IU/week)
Fig. 10.27. Median Hb versus median ESA dose in HD patients Fig. 10.28. Compliance with Hb 100–120 g/L versus median ESA
on ESA, by centre in 2012 dose in HD patients on ESA, by centre in 2012
210
100
90
80
70
60
Hb >120 g/L – not on ESA
50
Hb >120 g/L – on ESA
40 Hb 100–120 g/L
Hb <100 g/L
30
20
10
0
Belfast
Antrim
Newry
West NI
Ulster
Bradfd
Sthend
Dorset
York
Middlbr
Nottm
Basldn
Bristol
Leic
Ipswi
Glouc
Exeter
Redng
Sheff
Covnt
Wrexm
Wolve
B Heart
Carlis
Sund
Shrew
Norwch
Salford
Prestn
Oxford
Donc
Leeds
Chelms
Bangor
Swanse
Cardff
B QEH
Newc
Kent
England
N Ireland
Wales
E, W & NI
Centre
Fig. 10.29. Distribution of haemoglobin in patients treated with HD and the proportion of patients with Hb .120 g/L receiving ESA by
centre in 2012
cross-sectional analysis at the final quarter of 2012. Resistance to ESA therapy

Patients who had previously changed RRT modality Figure 10.34 shows the frequency distribution of
were included in this analysis. The proportion of PD weekly ESA dose by treatment modality adjusted for
patients requiring ESA rises with duration of RRT from weight. Data regarding prevalence of ESA resistance in
69% after 3–12 months, to 81% after 10 or more years. the literature in the ERF population is very sparse. RA
This almost certainly reflects loss of residual renal guidelines define resistance to ESA therapy as ‘failure
function. For at least the first 10 years on RRT, a greater to reach the target Hb level despite SC epoetin dose
percentage of HD patients are receiving ESA treatment >300 IU/kg/week (450 IU/kg/week IV epoetin) or
than patients on PD for any given duration on RRT. darbepoetin dose >1.5 mcg/kg/week’. For the purposes
100
90
80
70
60
Hb >120 g/L – not on ESA
50
Hb >120 g/L – on ESA
40 Hb 100–120 g/L
Hb <100 g/L
30
20
10
0
Belfast
Antrim
Newry
West NI
Bradfd
Sthend
Dorset
York
Nottm
Basldn
Bristol
Leic
Ipswi
Camb
Glouc
Exeter
Sheff
Covnt
Wrexm
Wolve
B Heart
Carlis
Sund
Shrew
Norwch
Salford
Prestn
Oxford
Donc
Leeds
Chelms
Bangor
Swanse
B QEH
Kent
England
N Ireland
Wales
E, W & NI
Centre
Fig. 10.30. Distribution of haemoglobin in patients treated with PD and the proportion of patients with Hb .120 g/L receiving ESA by
centre in 2012
211
100 100
HD HD
PD PD
Percentage of patients on ESA (95% CIs)

Percentage of patients (95% CIs)
90 90
80 80
70 70
60 60
50 50
18–34 35–44 45–54 55–64 65–74 75+ 3 months 1–2 years 2–3 years 3–5 years 5–10 years >10 years
Age group (years) to <1 year
Time on RRT
Fig. 10.31. Percentage of dialysis patients on ESA, by age group
Fig. 10.33. Percentage of patients on ESA by time on RRT (2012)
and treatment modality (2012)
100–120 g/L. The percentage of patients with ESA resist-

of this analysis the centres were restricted to those with ance, defined by those failing to reach target Hb .100 g/L
good completeness for weight (over 75%) and ESA dose were 0.5% for HD and 0.6% for PD. Caution needs to be
data (33 centres for HD and 22 centres for PD). As per applied when interpreting these results as the numbers
the above definition and assuming that HD patients for the above calculations are small.
largely receive ESA intravenously and PD patients receive
ESA subcutaneously, the prevalence of high doses of ESA Success with guideline compliance
was 1.0% (N = 72) and 2.2% (N = 12) for HD and PD Compliance with current minimum standards by year
patients respectively. For these patients the dose range (1998 to 2012) is shown in figure 10.35 for prevalent
for HD was 453–772 IU/kg/week and for PD 312–535 patients (by treatment modality).
IU/kg/week. For patients on HD with high ESA doses, The Renal Association guidelines state that centres
45% (N = 32) had Hb ,100 g/L and 28% were within should audit the ‘Proportion of patients on renal
100–120 g/L. For patients on PD with high ESA doses, replacement therapy with Hb level <100 g/L who are
25% (N = 3) had a Hb ,100 g/L and 67% were within
HD
50 PD
45
HD
40 PD 40
35
30
30
25 20
20
10
15
10
0
1–<50
50–<100
100–<150
150–<300
300–<450
450 or more
0
18–34 35–44 45–54 55–64 65–74 75+
Age group (years) ESA dose (IU/kg/week)
Fig. 10.32. Percentage of whole cohort (2012) who are not on Fig. 10.34. Frequency distribution of mean weekly ESA dose
ESA and have Hb 5100 g/L, by age group and treatment modality corrected for weight in 2012
212
100
90
80
70
Upper 95% Cl
60 % with Hb >100 g/L
Lower 95% Cl
50
40
30
Fig. 10.35. Percentage of prevalent HD
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
and PD patients (1998–2012) with Hb
Year 5100 g/L
not prescribed an ESA’. Figure 10.36 shows the Hb >120 g/L’. Table 10.5 shows that the percentage of
percentage of anaemic patients (Hb ,100 g/L) receiving all patients treated with an ESA and having Hb
an ESA. A minority of patients had a Hb ,100 g/L and .120 g/L ranged between 7–39% for HD and between
were not receiving ESA therapy. Across the age groups 0–33% for PD. For HD, there was a small percentage of
this was between 7–10% for HD patients and 2–13% patients having ferritin levels ,100 mg/L and being on
for PD patients. There are several potential explanations an ESA (0–7%). The percentages were somewhat higher
for this. Treatment with ESA may have been stopped in for PD (0–21%).
patients who were unresponsive or avoided in those Renal Association guidelines state that ‘Each renal
with malignancy. Others may have been on ESA treat- unit should audit the type, route and frequency of
ment but not had it recorded. administration and weekly dose of ESA prescribed’.
The Renal Association guideline states that centres Table 10.6 shows the percentage completeness for type,
should audit the ‘Proportion of patients with serum route and frequency of administration for centres
ferritin levels <100 mg/L treated with an ESA’ & (N = 40) reporting ESA data. The completeness was
‘The proportion of patients treated with an ESA with generally good for drug type and dose but patchy for
frequency and route of administration.
100
Discussion
90
Anaemia is one of the major problems that contribute
80
to high comorbidity and poor outcome in dialysis
patients. Since the introduction of human recombinant
erythropoietin for treating CKD-related anaemia over
70 two decades ago, attention has shifted from treating
severe anaemia in dialysis patients to preventing anaemia
60 pre-dialysis and to correcting anaemia within defined
HD target limits. Renal centres strive to meet the Renal
PD Association standards in order to prevent adverse out-
50
18–34 35–44 45–54 55–64 65–74 75+ comes associated with low Hb such as impaired quality
Age group (years) of life, increased hospitalisation, increased cardiovascular
Fig. 10.36. Percentage of patients with Hb ,100 g/L who were on events and increased cardiovascular and all-cause
ESA, by age group and treatment modality (2012) mortality.
213
Table 10.5. Percentage of patients with serum ferritin levels ,100 mg/L and on ESA and percentage of patients with Hb .120 g/L and
on ESA by modality
HD PD
% with Hb .120 g/L % with ferr ,100 mg/L % with Hb .120 g/L % with ferr ,100 mg/L
Centre and on ESA and on ESA and on ESA and on ESA
England
B Heart 7 1 5 6
B QEH 15 1 11 5
Basldn 16 6 0 4
Bradfd 24 4 13 5
Bristol 23 3 7 0
Camb 6 4
Carlis 21 4 19 6
Chelms 39 0 32 13
Covnt 13 3 16 6
Donc 20 0 13 0
Dorset 31 2 33 0
Exeter 18 7 13 7
Glouc 15 7 13 15
Ipswi 15 1 17 12
Kent 22 6 15 8
Leeds 19 2 16 4
Leic 29 5 24 2
Middlbr 17 3
Newc 22 2
Norwch 22 5 17 20
Nottm 15 0 11 1
Oxford 22 5 17 5
Prestn 19 2 14 14
Redng 23 2
Salford 13 21
Sheff 20 1 7 0
Shrew 29 2 22 10
Sthend 16 2 14 0
Sund 22 5 18 0
Wolve 23 1 18 14
York 15 1 15 4
N Ireland
Antrim 25 1 20 10
Belfast 18 4 20 0
Newry 21 7 21
Ulster 21 1
West NI 13 4 33 0
Wales
Bangor 21 1 7 7
Cardff 15 3
Swanse 14 2 6 6
Wrexm 24 2 20
England 20 3 16 6
N Ireland 19 3 24 7
Wales 16 3 9 5
E, W & NI 20 3 16 6
Blank cells denote centres excluded from analyses due to poor completeness or small numbers with data
214
Table 10.6. Percentage completeness for type, dose, route and frequency of administration of ESA
HD PD
% with % % % with % with % % % with

N on drug with with administration N on drug with with administration
Centre ESA type dose frequency route ESA type dose frequency route
England
B Heart 307 100 100 0 0 20 100 100 0 0
B QEH 726 100 100 100 0 92 100 100 100 0
Basldn 137 100 99 100 100 17 100 100 100 100
Bradfd 181 100 91 0 0 20 100 100 0 0
Bristol 422 100 100 0 0 37 100 100 0 0
Camb 19 100 100 0 0
Carlis 39 100 100 0 0 14 100 100 0 0
Chelms 117 100 100 100 100 19 100 100 100 100
Covnt 305 100 99 0 0 57 100 98 0 0
Donc 144 100 100 100 100 16 100 100 100 94
Dorset 236 100 100 97 100 26 100 100 100 100
Exeter 327 100 100 0 0 50 100 100 0 0
Glouc 175 100 0 0 0 17 100 0 0 0
Ipswi 81 100 100 0 0 21 100 100 0 0
Kent 330 100 100 100 100 37 100 100 100 100
Leeds 427 100 90 0 0 60 100 100 0 0
Leic 782 100 98 0 0 115 100 93 0 0
Middlbr 242 100 100 0 0 6 100 100 0 0
Newc 180 100 100 0 0
Norwch 275 100 100 100 100 34 100 100 97 100
Nottm 318 100 99 0 0 50 100 80 0 0
Oxford 360 100 100 0 0 56 100 100 0 0
Prestn 435 100 8 0 0 44 100 0 0 0
Redng 227 100 0 0 0
Salford 236 100 97 100 0 69 100 96 100 0
Sheff 486 100 100 0 0 40 100 100 0 0
Shrew 162 100 99 87 94 20 100 100 100 100
Sthend 104 100 95 0 0 8 100 75 0 0
Sund 174 100 28 0 0 11 100 100 0 0
Wolve 230 100 100 0 0 52 100 100 0 0
York 113 100 100 0 0 19 100 100 0 0
N Ireland
Antrim 116 100 100 100 100 8 100 100 100 100
Belfast 187 100 100 99 100 20 100 100 100 100
Newry 81 100 100 93 100 12 100 100 100 92
Ulster 94 100 100 100 100 6 100 100 100 100
West NI 119 100 100 98 100 10 100 100 100 100
Wales
Bangor 65 100 96 0 0 7 100 100 0 0
Cardff 273 100 0 0 0
Swanse 282 100 100 100 99 42 100 98 100 98
Wrexm 78 100 99 99 100 11 100 92 83 100
England 8,278 100 88 28 17 1,046 100 92 31 16
N Ireland 597 100 100 98 100 56 100 100 100 98
Wales 698 100 61 51 51 60 100 97 85 87
E, W & NI 9,573 100 86 34 24 1,162 100 93 37 24
Blank cells denote centres excluded from analyses due to poor completeness or small numbers with data
215
Haemoglobin outcomes for patients on HD and PD in with ferritin levels ,100 mg/L and receiving an ESA.
the UK were largely compliant with the RA minimum There was substantial variation between centres in the
standard of Hb 5100 g/L (82% and 85% respectively). average dose of ESA prescribed. Attainment of Hb
As would be anticipated, a greater proportion of targets correlates poorly with median ferritin and ESA
prevalent patients (83%) than incident patients (51%) usage.
had a Hb 5100 g/L in 2012. In the UK, the median Hb Resistance to ESA has consistently been shown to be
of patients on HD was 112 g/L with an IQR of 103– associated with an increased risk of death and cardio-
121 g/L, and the median Hb of patients on PD was vascular events in CKD patients [14–17]. There is for
114 g/L with an IQR of 105–123 g/L. the first time an attempt to describe the prevalence of
Compliance with advice regarding iron stores as ESA resistance in the UK and this was 0.5% and 0.6%
reflected by ferritin remained stable in the UK with for HD and PD patients respectively. Bearing in mind
95% of HD patients and 88% of PD patients achieving the limitations of relatively small numbers involved
a serum ferritin greater than 100 mg/L. in the calculations, one possible reason that could explain
The analysis of ESA usage is limited by incomplete the low prevalence is that this group of patients have poor
data returns. From the available data, 87% of HD patients survival. This again emphasises the need for better data
and 69% of PD patients were on ESA treatment in returns and with improved completeness future analysis
England, Wales and Northern Ireland. The percentage of could look into whether this translates to poor patient
patients treated with an ESA and having Hb .120 g/L outcomes for the UK dialysis population.
ranged between centres from 7–39% for HD and from
0–33% for PD. There was a small percentage of patients Conflicts of interest: none
References
1 KDIGO clinical practice guideline for anemia in chronic kidney disease. 10 Eschbach JW, et al. Correction of the Anemia of End-Stage Renal
Summary of recommendations statements. Kidney International Disease with Recombinant Human Erythropoietin. New England
Supplements 2012;2:283–287 Journal of Medicine 1987:316(2);73–78
2 Department of Health Renal Team National Service Framework for 11 Singh AK, et al. Correction of Anemia with Epoetin Alfa in Chronic
Renal Services: Part One – Dialysis and transplantation, 2004, Depart- Kidney Disease. New England Journal of Medicine 2006:355(20);2085–
ment of Health: London 2098
3 Renal Association. Treatment of adults and children with renal failure: 12 Drüeke TB, et al. Normalization of Hemoglobin Level in Patients with
standards and audit measures. 3rd Edition., 2002, Royal College of Chronic Kidney Disease and Anemia. New England Journal of Medicine
Physicians of London and the Renal Association: London 2006:355(20):2071–2084
4 Locatelli F, et al. Anaemia management in patients with chronic kidney 13 Pfeffer MA, et al. A Trial of Darbepoetin Alfa in Type 2 Diabetes and
disease: a position statement by the Anaemia Working Group of Euro- Chronic Kidney Disease. New England Journal of Medicine 2009:
pean Renal Best Practice (ERBP). Nephrology Dialysis Transplantation 361(21);2019–2032
2009;24(2):348–354 14 Gomez-Alamillo C, et al. Erythropoietin Resistance as Surrogate
5 National Institute for Health and Clinical Excellence (NICE). Anaemia Marker of Graft and Patient Survival in Renal Transplantation: 3-Year
management in people with chronic kidney disease (CG114), 2011 Prospective Multicenter Study. Transplantation Proceedings 2010;
6 Renal Association Clinical Practice Guidelines Committee: Haemodialy- 42(8):2935–2937
sis, 5th Edition. 2010. http://www.renal.org/clinical/guidelinessection/ 15 Chung S, et al. Relationship between erythropoietin resistance index and
haemodialysis.aspx left ventricular mass and function and cardiovascular events in patients
7 Kidney Disease: Improving Global Outcomes (KDIGO) Anemia Work on chronic hemodialysis. Hemodialysis International 2012; 16(2):181–
Group. KDIGO Clinical Practice Guideline for Anemia in Chronic 187
Kidney Disease. Kidney inter 2012:2(suppl 1);279–335 16 Panichi V, et al. Anaemia and resistance to erythropoiesis-stimulating
8 Movilli E, et al. Predialysis versus postdialysis hematocrit evaluation agents as prognostic factors in haemodialysis patients: results from the
during erythropoietin therapy. American journal of kidney diseases: RISCAVID study. Nephrology Dialysis Transplantation, 2011;26(8):
the official journal of the National Kidney Foundation 2002:39(4);850– 2641–2648
853 17 Solomon SD, et al. Erythropoietic Response and Outcomes in Kidney
9 Kalantar-Zadeh K, et al. Diagnosis of iron deficiency anemia in renal Disease and Type 2 Diabetes. New England Journal of Medicine
failure patients during the post-erythropoietin era. American Journal 2010;363(12):1146–1155
of Kidney Diseases 1995;26(2):292–299
216
Chapter 11 Blood Pressure Profile of
Prevalent Patients receiving Renal
Replacement Therapy in 2012:
Anirudh Raoa, David Pitchera, Ken Farringtonb

a
UK Renal Registry, Bristol, UK; bLister Hospital, Stevenage, UK
Key Words respectively. The median SBP of patients on PD

Diastolic blood pressure . Epidemiology . Established renal was 137 mmHg. Transplant recipients had a median
failure . Haemodialysis . Peritoneal dialysis . Pulse pressure SBP of 134 mmHg.
. Systolic blood pressure . Transplant . In 2012, median diastolic blood pressures (DBPs)
were 71 mmHg (pre-HD), 67 mmHg (post-HD),
78 mmHg (PD) and 79 mmHg (transplant).
Summary . In England, Wales and Northern Ireland only
26% of PD patients achieved the Renal Association
. Data completeness was better for haemodialysis guideline of SBP ,130 mmHg and DBP
(HD) patients (75% for pre-HD measurements) ,80 mmHg.
than for peritoneal dialysis (PD) patients (51%) or . In England, Wales and Northern Ireland only
transplant recipients (41%). 27% of transplant patients achieved the Renal
. In 2012, median pre- and post-HD systolic blood Association guideline of SBP ,130 mmHg and
pressures (SBPs) were 140 mmHg and 128 mmHg DBP ,80 mmHg.
217
Introduction pressure measurements should be made to facilitate

good management of the HD session. (2D)
The aetiology of hypertension in established renal
failure is multifactorial and interpreting blood pressure Guideline 5.4 – CVD: Hypertension in dialysis
(BP) values in this cohort of patients is challenging. In patients
dialysis patients there is a complex interplay between It would be sensible to avoid sustained BP extremes
volume overload with salt (and water) which may be and, in order to try to provide some guidance we suggest
appropriately addressed by dialysis, and vasoconstriction that systolic blood pressure during the inter-dialytic
caused by neurohumoral mechanisms which may require period on HD and for PD patients should not regularly
additional treatment with antihypertensive drugs. These exceed >160 mmHg. (2C)
mechanisms lead to cardiovascular dysfunction and
may be important in the observation of the ‘U-shaped’ Guideline 5.5 – CVD: Hypotension/Hypertension in
mortality curve seen in relation to BP in dialysis patients dialysis patients
[1, 2]. Original descriptions at the individual patient level We suggest that systolic blood pressure should
were confounded by unmeasured case-mix, with comor- not routinely be treated with pharmacological agents
bidity associated with both lower BP and lower survival. with antihypertensive properties if SBP is regularly
Similar patterns have now been reported at centre level <120 mmHg pre dialysis.
[3]. It is possible that the association can be overcome
by longer or more frequent sessions of dialysis and careful Guideline 5.7 – CVD: Hypertension in dialysis
attention to dry-weight [4, 5]. Iatrogenic factors such as patients
erythropoiesis stimulating agents (ESA) [6] in dialysis We suggest that hypertension on dialysis should be
patients and ciclosporin [7] in transplant patients may managed by ultrafiltration in the first instance. (2D)
also contribute to high BP. Further, BP in dialysis patients
Blood pressure in peritoneal dialysis patients should
varies as much within individuals as it does between
be <130/80 mmHg (Good Practice).
individuals [8]. The extent of this variability appears to
be as important as the absolute value in predicting cardio- The target blood pressure for renal transplant
vascular mortality in haemodialysis patients [9]. The patients is <130/80 mmHg (Good practice).
optimal measure of BP therefore remains the subject
of considerable controversy, with ambulatory BP predict- These guidelines are consistent with international
ing mortality better than pre- or post-dialysis BP [10]. guidelines [20, 21].
There is some evidence to suggest that pre-dialysis This chapter reports UK Renal Registry (UKRR) data
systolic blood pressures (SBPs) .150–160 [11–13] are completeness for BP for adult renal centres in England,
associated with excess mortality in haemodialysis patients Northern Ireland and Wales and presents centre level
and other data suggesting that very high SBP (.200) average BP attainment for patients on haemodialysis
pre-dialysis seems to confer an adverse prognosis [14]. (HD), peritoneal dialysis (PD) and with a functioning
Conversely, lowering BP too aggressively may lead to kidney transplant at the end of December 2012.
intradialytic hypotension [15], which is an independent
predictor of mortality [16, 17]. Data from a number of
studies suggest excess mortality associated with pre-
dialysis SBP ,120 mmHg [14, 18]. Methods
The Renal Association guidelines updated in August
2010 and in operation during the period in which the All adult patients in England, Wales and Northern Ireland
receiving renal replacement therapy (RRT) (HD, PD and trans-
audit data in this chapter were collected [19] stated: plant recipients) on 31st December 2012 were considered for
inclusion in the analyses.
The method of data extraction employed is described in chapter
Guideline 5.2 – CVD: Hypertension in dialysis 15 of the 11th UKRR Annual Report [22]. The UKRR extracts
patients quarterly laboratory, clinical and demographic data for all patients
receiving RRT in the 62 renal centres in England, Northern Ireland
We suggest that pre- and post-dialysis blood pressure and Wales. Data on some variables from the nine Scottish renal
(measured after completion of dialysis, including centres are sent annually to the Scottish Renal Registry. However,
washback) should be recorded and intra-dialytic blood BP measurements are only collected from the Scottish Registry
218
Chapter 11 Blood pressure in UK RRT patients
for HD patients and therefore PD and transplant patients from SBP of patients on PD was 137 mmHg. Transplant
Scottish renal centres are excluded from all BP analyses. recipients had a median SBP of 134 mmHg. Median
Patients who had been on the same modality and at the same
renal centre for three months and with a valid BP reading in either
DBP was 71 mmHg (pre-HD), 67 mmHg (post-HD),
the fourth or the third quarter of 2012 were included. This 78 mmHg (PD) and 79 mmHg (transplant).
included incident patients starting RRT during 2012 who were
still alive on 31st December 2012. Analyses used the last recorded Relationship between the centre mean and the
BP from quarter four, however, if this was missing, the last proportion above a threshold BP in that centre
recorded BP from quarter three was used instead. BP data from
quarter two were used for patients at renal centres in Scotland
As the distribution of BP in each centre approximates
because BP data from quarters three and four were unavailable. a normal distribution (data not shown), the population
Analyses were performed for each RRT modality (HD, PD mean of each BP variable should predict the number of
and transplant). Most UK renal centres manage HD, PD and individuals above (or below) a predefined threshold or
transplant patients. However, Colchester had no PD patients standard (Rose and Day 1990). As these assumptions
and four centres (Bangor, Colchester, Liverpool Aintree, Wirral)
had no transplant patients under their care.
were confirmed in the 13th UKRR Annual Report [23]
All patients meeting the criteria above were included in the only median BP data by centre are presented below.
overall national analyses, but renal centres with less than 50%
data completeness for any modality, or fewer than 20 patients Centre-specific analyses of BP in haemodialysis
with results, were excluded from the centre level analysis for patients
that modality. The number preceding the centre name in each
figure corresponds to the percentage of missing data for that
Figures 11.2 and 11.3 illustrate the median and IQR
centre. pre-dialysis SBP and DBP in each centre supplying data
Patients on HD were analysed both by pre-dialysis and post- on .50% of patients. The median HD pre-dialysis SBP
dialysis BP. The BP components analysed included systolic and pre-dialysis DBP for the UK were 140 mmHg and
blood pressure, diastolic blood pressure (DBP) and pulse pressure 71 mmHg respectively. Figures 11.4 and 11.5 illustrate
(PP). The data were analysed to produce summary statistics
(mean, median, maximum, minimum). Standard deviation and the equivalent analyses for post-dialysis BP.
quartile ranges were also calculated. Median BP and inter-quartile There remains marked centre variation. The difference
ranges (IQRs) are presented for each analysis as caterpillar plots. between the centres with the lowest and highest median
In addition, the percentage of HD patients with pre-dialysis systo- SBP was .20 mmHg. Comparison with previous UKRR
lic BP ,120 mmHg, between 120–160 mmHg, .160 mmHg; PD reports showed that in general, the same centres can be
and transplant patients attaining Renal Association standards for
BP (<130/80 mmHg) in individual renal centres and each nation found at roughly the same place in the distribution
were calculated and are presented with 95% confidence intervals from year to year.
in caterpillar plots.
Chi-squared tests were used in the analyses of the 2012 BP data Adherence to guidelines
to test for statistically significant differences between renal centres Figures 11.6, 11.7 and 11.8 illustrate the percentages
and between nations. All statistical analyses were performed using
SAS version 9.3. (with 95% confidence intervals (CIs)) of HD patients
achieving SBP in the range 120–160 mmHg,
,120 mmHg and .160 mmHg respectively. There was
marked variation (45–80%) between centres achieving
Results their pre-dialysis SBP readings in the range 120–
160 mmHg. The vast majority of centres had greater
Data completeness than 50% of their patients falling in the range 120–
Data extracts were received from all 62 centres in 160 mmHg. Thirty-five of the centres had greater than
England, Wales and Northern Ireland. Data complete- 20% of their patients with a pre-dialysis SBP
ness is summarised in table 11.1. Overall, completeness ,120 mmHg and there were also 35 centres who had
was very similar to that previously reported. greater than 20% of their patients with a pre-dialysis
SBP .160 mmHg.
BP on each modality
Figure 11.1 gives the median and IQR for SBP, DBP Centre-specific analyses of BP in peritoneal dialysis
and PP in prevalent HD patients (pre- and post-dialysis), patients
PD and transplant patients. Figures 11.9 and 11.10 illustrate the median and IQR
In 2012, the median pre- and post-HD SBPs were SBP and DBP in each centre supplying data on .50%
140 mmHg and 128 mmHg respectively. The median of eligible patients. Figure 11.11 gives the percentage of
219
Table 11.1. Percentage of patients by renal centre for whom BP readings were received by the UKRR, by modality
% completed data % completed data
Centre Pre-HD Post-HD PD Transplant Centre Pre-HD Post-HD PD Transplant

England Prestn 20 0 0 0
B Heart 98 98 2 3 Redng 95 100 0 0
B QEH 94 93 84 93 Salford 97 97 0 0
Basldn 98 93 96 2 Sheff 99 96 99 97
Bradfd 3 2 96 69 Shrew 99 99 0 1
Brightn 54 68 0 0 Stevng 94 91 63 23
Bristol 99 99 91 72 Sthend 99 99 0 61
Camb 100 100 97 97 Stoke 95 95 1 0
Carlis 100 100 5 0 Sund 99 99 0 0
Carsh 92 92 1 0 Truro 83 82 68 19
Chelms 100 98 96 94 Wirral 94 93 14 n/a
Colchr 99 99 n/a n/a Wolve 99 99 98 95
Covnt 100 100 95 81 York 100 98 96 53
Derby 99 95 99 83 N Ireland
Donc 100 95 91 100 Antrim 98 92 100 65
Dorset 100 96 58 81 Belfast 94 87 16 45
Dudley 95 93 47 16 Newry 99 98 71 86
Exeter 100 99 94 92 Ulster 99 94 100 90
Glouc 100 100 90 89 West NI 98 92 100 93
Hull 97 97 89 25 Scotland
Ipswi 100 100 0 0 Abrdn 99 99 n/a n/a
Kent 98 98 98 85 Airdrie 94 94 n/a n/a
L Barts 0 0 0 0 D & Gall 96 96 n/a n/a
L Guys 0 0 0 0 Dundee 99 96 n/a n/a
L Kings 0 0 0 0 Dunfn 96 95 n/a n/a
L Rfree 93 91 99 77 Edinb 94 93 n/a n/a
L St.G 59 60 0 0 Glasgw 95 88 n/a n/a
L West 0 0 0 0 Inverns 96 95 n/a n/a
Leeds 100 97 99 96 Klmarnk 99 99 n/a n/a
Leic 97 96 81 48 Wales
Liv Ain 98 98 12 n/a Bangor 98 98 100 n/a
Liv RI 97 95 2 2 Cardff 4 29 51 98
M RI 0 0 0 0 Clwyd 100 92 0 0
Middlbr 97 96 88 46 Swanse 100 100 96 100
Newc 100 100 0 0 Wrexm 100 99 25 0
Norwch 95 90 4 41 England 73 71 50 37
Nottm 100 100 97 87 N Ireland 97 91 64 60
Oxford 96 95 43 16 Scotland 96 93 n/a n/a
Plymth 59 10 65 85 Wales 57 67 61 84
Ports 100 100 85 19 UK 75 74 51∗ 41∗
∗
UK % completeness for PD and transplant excludes Scotland
160
Upper quartile
Median
140
Lower quartile
120
BP mmHg
100 Systolic Diastolic Pulse pressure
80
60
40
pre-HD post-HD PD Tx pre-HD post-HD PD Tx pre-HD post-HD PD Tx
Modality Fig. 11.1. Summary of BP achievements
220
BP mmHg BP mmHg BP mmHg
40
50
60
70
80
90
100
90
100
110
120
130
140
150
160
170
180
100
110
120
130
140
150
160
170
180
190
Chapter 11
4 Dundee 1 Klmarnk 1 Klmarnk

32 Brightn 1 Colchr 2 Basldn
1 Colchr 1 Dundee 1 Colchr
3 Hull 6 B QEH 0 Leeds
5 Stoke 2 B Heart 0 York
5 Inverns 0 Camb 0 Ipswi
1 Klmarnk 8 Carsh 2 Antrim
7 Edinb 1 Newry 6 B QEH
12 Glasgw 1 Wolve 6 Edinb
Lower quartile
Lower quartile
Upper quartile
Upper quartile
Lower quartile
2 Basldn
Upper quartile
4 Leic 3 Leic
0 Redng 46 Brightn 0 Camb
8 West NI 0 Dorset 0 Exeter
0 Camb 0 Leeds 5 Stoke
1 Exeter 2 Liv Ain 17 Truro
0 Swanse 4 D&Gall 2 West NI
Median pre HD systolic BP
Median pre HD diastolic BP
Median post HD systolic BP

Fig. 11.2. Median systolic BP: pre-HD
1 Abrdn 0 Donc 8 Carsh
Fig. 11.4. Median systolic BP: post-HD

Fig. 11.3. Median diastolic BP: pre-HD
7 B QEH 5 Dudley 1 Derby
7 Basldn 5 Glasgw 1 Dundee
7 Dudley 3 Leic 5 Glasgw
1 Sund 3 Liv RI 3 Liv RI
2 York 5 Norwch 0 Swanse
0 Carlis 0 Swanse 0 Covnt
N = 16,233
0 Wrexm 4 Dunfn
N = 16,233
9 L Rfree
N = 16,003
3 Leeds 0 York 0 Glouc
5 Liv RI 0 Nottm 1 Sthend
1 Bristol 5 Redng 1 Sund
5 Dunfn 2 Antrim 6 Wirral
3 Salford 0 Covnt 1 Abrdn
1 Wolve 0 Exeter 6 Airdrie
5 Derby 0 Ipswi 1 Bristol
8 Antrim 41 L St.G 0 Dorset
40 L St.G 3 Middlbr 3 Hull
5 Oxford 0 Ports 2 Liv Ain
2 Liv Ain 17 Truro 3 Middlbr
4 Dunfn 3 Salford
Centre
Centre
0 Ports
Centre
7 L Rfree 46 Brightn
2 Bangor 5 Stoke 0 Carlis
8 Carsh 6 Wirral 4 D&Gall
0 Covnt 1 Abrdn 2 Kent
4 Dorset 1 Bristol 7 L Rfree
4 Sheff 1 Shrew 41 L St.G
1 Wrexm 1 Derby 5 Redng
18 Truro 6 Edinb 0 Chelms
5 Donc 3 Hull 0 Newc
2 Kent 2 Kent 5 Norwch
1 Sthend 0 Newc 0 Ports
4 Middlbr 1 Sheff 0 Wrexm
0 Newc 1 Sthend 0 Donc
10 Norwch 1 Sund 1 Shrew
9 Stevng 2 West NI 5 Dudley
7 Wirral 6 Belfast 41 Plymth
0 Glouc 0 Carlis 1 Sheff
2 B Heart 6 Stevng 2 Bangor
0 Ipswi 4 Oxford 2 B Heart
2 Newry 1 Ulster 1 Newry
1 Shrew 2 Bangor 0 Nottm
2 Chelms 0 Chelms 1 Wolve
4 D&Gall 0 Clwyd 6 Belfast
6 Airdrie 0 Glouc 4 Inverns
0 Nottm 3 Salford 0 Clwyd
13 Belfast 6 Airdrie 4 Oxford
8 Clwyd 41 Plymth 1 Ulster
6 Ulster 4 Inverns 6 Stevng
26 UK 25 UK 25 UK
221
Blood pressure in UK RRT patients
222
Percentage of patients Percentage of patients BP mmHg
0
10
20
30
40
50
30
40
50
60
70
80
90
40
50
60
70
80
1 Klmarnk 1 Newry 4 Dundee 90
0 Leeds 5 Norwch 32 Brightn
0 Ipswi 2 West NI 1 Colchr
3 Hull 0 Chelms 1 Klmarnk
2 Basldn 0 Dorset 0 Camb
0 York 0 Donc 0 Redng
3 Liv RI 1 Colchr 2 B Heart
2 Liv Ain 4 D&Gall 7 B QEH
2 Antrim 5 Dudley 12 Glasgw

Lower quartile
0 Wrexm 41 L St.G
Upper quartile
1 Wrexm
0 Exeter 2 Basldn 8 Carsh
6 B QEH 2 Kent 4 Dorset
3 Leic 6 Belfast 7 Dudley
3 Middlbr 6 Edinb 5 Stoke
5 Redng 1 Ulster 0 Swanse
5 Glasgw 2 Bangor 5 Donc
Median post HD diastolic BP
0 Camb 1 Sheff
Fig. 11.5. Median diastolic BP: post-HD

10 Norwch
1 Sund 0 Glouc 1 Wolve
6 Wirral 17 Truro 1 Abrdn
1 Colchr 5 Stoke 4 Leic
1 Wolve 0 Exeter 5 Liv RI
4 Dunfn 2 Antrim 2 Newry
1 Bristol 0 Wrexm 40 L St.G
N = 16,003
46 Brightn 3 Salford 1 Exeter

8 Carsh 1 Abrdn 9 L Rfree
1 Derby 6 Airdrie 3 Leeds
0 Covnt 1 Shrew 2 Liv Ain
1 Dundee 3 Leic 4 Middlbr
17 Truro 0 Covnt
6 Edinb 0 Swanse 0 Ports
0 Swanse 6 B QEH 9 Stevng
0 Carlis 1 Derby 18 Truro
7 L Rfree 0 Ports 7 Basldn
Fig 11.7. Percentage of patients with pre-dialysis SBP ,120 mmHg

5 Stoke 0 Newc 1 Bristol
1 Abrdn 1 Dundee 4 D&Gall
Centre
Centre
Centre
1 Sthend 0 Clwyd 5 Dunfn

0 Clwyd 0 York 3 Hull
0 Newc 8 Carsh 5 Inverns
3 Salford 1 Bristol 0 Nottm
4 Oxford 0 Camb 2 York
0 Ports 6 Wirral 5 Oxford
41 Plymth 2 Liv Ain 1 Sthend
6 Airdrie 6 Stevng 0 Carlis
0 Nottm 46 Brightn 0 Covnt
5 Dudley 2 B Heart 4 Sheff
0 Glouc 3 Liv RI 1 Shrew
N = 16,233
1 Shrew 7 L Rfree 8 West NI
4 Inverns 0 Ipswi 8 Antrim
2 West NI 4 Dunfn 5 Derby
0 Dorset 1 Sthend 7 Edinb
Fig 11.6. Percentage of patients achieving pre-dialysis SBP readings in the range 120–160 mmHg
4 D&Gall 0 Nottm 2 Kent
2 Kent 5 Glasgw 0 Newc
1 Sheff 41 Plymth 1 Sund
41 L St.G 1 Sund 7 Wirral
6 Stevng 0 Leeds 13 Belfast
N = 16,233
Lower 95% Cl
Upper 95% Cl
2 B Heart 3 Middlbr 3 Salford
6 Belfast 5 Redng 6 Ulster
2 Bangor 1 Klmarnk 2 Bangor
0 Donc 4 Oxford 0 Ipswi
0 Chelms 0 Carlis 0 Glouc
5 Norwch 3 Hull 2 Chelms
1 Ulster 4 Inverns 6 Airdrie
1 Newry 1 Wolve 8 Clwyd
Lower 95% Cl
Upper 95% Cl

% with BP in range
% with pre HD systolic BP <120 mmHg

25 UK 25 UK 26 UK
BP mmHg BP mmHg Percentage of patients
0
10
20
30
40
50
50
60
70
80
90
100
110
100
110
120
130
140
150
160
170
Chapter 11
4 Inverns
10 Glouc 10 Glouc 4 Oxford
1 Wolve
0 Carlis
4 York 1 Leeds 2 B Heart
41 Plymth
4 Basldn 4 York 6 Stevng
0 Nottm
49 Cardff 6 Exeter 1 Ulster
1 Sthend
Lower quartile
Lower quartile
Upper quartile
Upper quartile
1 Sund
3 Camb 4 Chelms 5 Redng
Median systolic BP
Median diastolic BP
7 L Rfree
Fig. 11.9. Median systolic BP: PD

19 Leic 4 Swanse 3 Middlbr
Fig. 11.10. Median diastolic BP: PD

3 Hull
4 Swanse 9 Bristol 0 Ports
1 Shrew
2 Bangor
1 L Rfree 9 Donc 0 Clwyd
N = 1,583
N = 1,583
0 Newc
1 Sheff 19 Leic 4 Dunfn
6 Airdrie
16 B QEH 15 Ports 6 Belfast
3 Salford
46 Brightn
1 Derby 49 Cardff 1 Sheff
1 Dundee
9 Donc 35 Plymth 5 Glasgw
1 Bristol
5 Covnt 16 B QEH 1 Abrdn
8 Carsh
15 Ports 5 Covnt 0 Donc
0 Swanse
Fig 11.8. Percentage of patients with pre-dialysis SBP .160 mmHg
2 Kent
4 Chelms 11 Hull 6 Wirral
Centre
Centre
Centre
1 Derby
42 Dorset 2 Wolve 41 L St.G
0 Covnt
6 Exeter 4 Bradfd 0 Glouc
0 Camb
3 Liv RI
11 Hull 2 Kent 4 D&Gall
2 Liv Ain
2 Kent 1 L Rfree 5 Stoke
1 Newry
0 Chelms
N = 16,233
1 Leeds 4 Basldn
6 B QEH
0 York
9 Bristol 3 Camb 3 Leic
17 Truro
3 Nottm 1 Derby 5 Dudley
0 Ipswi
2 Wolve 42 Dorset 0 Dorset
6 Edinb
5 Norwch
Lower 95% Cl
Upper 95% Cl
35 Plymth 3 Nottm 0 Leeds

0 Wrexm
4 Bradfd 1 Sheff 2 Antrim
1 Klmarnk
50 England 50 England 0 Exeter
2 West NI
1 Colchr
36 N Ireland 36 N Ireland 2 Basldn
27 England
39 Wales 39 Wales 3 N Ireland
4 Scotland
43 Wales
% with pre HD systolic BP >160 mmHg
49 E, W & NI 49 E, W & NI
25 UK
223
Blood pressure in UK RRT patients
80
N = 1,583 Upper 95% CI
70 % with BP in range
Lower 95% CI
60
50
40
30
20
10
0
49 Cardff
10 Glouc
4 York
19 Leic
1 Leeds
3 Camb
1 L Rfree
4 Chelms
11 Hull
2 Kent
15 Ports
35 Plymth
4 Swanse
6 Exeter
1 Derby
16 B QEH
4 Bradfd
9 Bristol
3 Nottm
2 Wolve
9 Donc
5 Covnt
42 Dorset
1 Sheff
4 Basldn
50 England
36 N Ireland
39 Wales
49 E, W & NI
Centre
Fig. 11.11. Percentage of patients with BP ,130 mmHg systolic and ,80 mmHg diastolic: PD
patients meeting the audit standard of BP ,130/ Discussion

80 mmHg.
The possibility of information bias in these analyses Blood pressure control amongst HD patients in the
cannot be excluded since BP data are extracted from UK remained poor in 2012. Nearly half of centres had
the routine clinical record. greater than 20% of their patients with pre-dialysis systo-
lic BP ,120 mmHg. There were also nearly half who had
Centre-specific analysis of BP in transplant patients greater than 20% of their patients with pre-dialysis
Figures 11.12 and 11.13 illustrate the median and IQR systolic BP .160 mmHg. There continues to be marked
SBP and DBP in each centre supplying data on .50% of variation between centres in attainment of nationally
eligible patients and figure 11.14 illustrates the percentage agreed BP standards for those on PD and those with
of patients meeting the audit standard of BP ,130/ functioning kidney transplants.
80 mmHg. High BP is common in HD patients and contributes to
As with PD patients, the possibility of information bias the observed excess of cardiovascular morbidity and
in these analyses cannot be excluded. mortality in these patients [24]. However, there is still
170
Median systolic BP
160
Lower quartile
150
BP mmHg
140
130
120
110
28 Bristol
8 Exeter
7 West NI
11 Glouc
14 Newry
39 Sthend
0 Swanse
31 Bradfd
23 L Rfree
47 York
19 Dorset
15 Kent
4 Leeds
35 Antrim
0 Donc
19 Covnt
13 Nottm
2 Cardff
5 Wolve
7 B QEH
6 Chelms
10 Ulster
3 Camb
15 Plymth
3 Sheff
17 Derby
63 England
40 N Ireland
16 Wales
59 E, W & NI
Centre
Fig. 11.12. Median systolic BP: transplant
224
Chapter 11 Blood pressure in UK RRT patients
100
Upper quartile
Median diastolic BP N = 9,938
Lower quartile
90
BP mmHg
80
70
60
28 Bristol
11 Glouc
47 York
23 L Rfree
19 Dorset
10 Ulster
3 Camb
4 Leeds
14 Newry
6 Chelms
8 Exeter
7 West NI
13 Nottm
0 Swanse
3 Sheff
35 Antrim
7 B QEH
31 Bradfd
2 Cardff
19 Covnt
0 Donc
15 Kent
5 Wolve
17 Derby
39 Sthend
15 Plymth
63 England
40 N Ireland
16 Wales
59 E, W & NI
Centre
Fig. 11.13. Median diastolic BP: transplant
60
N = 9,938 Upper 95% CI
% with BP in range
50
Lower 95% CI
40
30
20
10
0
28 Bristol
7 West NI
14 Newry
47 York
11 Glouc
8 Exeter
23 L Rfree
0 Swanse
19 Dorset
4 Leeds
2 Cardff
10 Ulster
13 Nottm
5 Wolve
15 Kent
31 Bradfd
19 Covnt
0 Donc
3 Camb
7 B QEH
6 Chelms
15 Plymth
35 Antrim
17 Derby
39 Sthend
3 Sheff
63 England
40 N Ireland
16 Wales
Centre 59 E, W & NI
Fig. 11.14. Percentage of patients with BP ,130 mmHg systolic and ,80 mmHg diastolic: transplant
no clarity about how and when to measure BP, or about The utility of UKRR data could be enhanced by
BP targets in the haemodialysis population. collection of data on intra-dialytic weight gain, the use
Reliance upon immediate pre-dialysis and/or post- of BP lowering drugs and the frequency of intra-dialytic
dialysis BP measurements alone to detect hypertension hypotension. Future registry analyses should include
in patients undergoing haemodialysis may be misleading systolic BP as an independent risk factor in models for
[25]. Pre-dialysis BP may substantially overestimate predictors of death and variation in survival on dialysis.
mean ambulatory inter-dialytic BP [26]. For pre-dialysis
SBP the overestimate may range from 6–18 mmHg Conflicts of interest: none
depending on the timing of the measurement and for
DBP from 3–9 mmHg. In contrast, post-dialysis mea-
surements underestimate mean systolic BP by approxi- References
mately 4–14 mmHg for SBP and 1 mmHg for DBP.
There are suggestions that post-dialysis BP may be 1 Kalantar-Zadeh K KR, Kopple JD. Reverse epidemiology of blood
more reflective of mean inter-dialytic BP [25, 26]. pressure in dialysis patients. Kidney Int. 2005;67(5):2067
225
2 Levin NW, Kotanko P, Eckardt KU, Kasiske BL, Chazot C, Cheung AK, 14 Li Z, Lacson E, Lowrie EG, Ofsthun NJ, Kuhlmann MK, Lazarus JM,
Redon J, Wheeler DC, Zoccali C, London GM: Blood pressure in chronic et al. The epidemiology of systolic blood pressure and death risk in
kidney disease stage 5D-report from a Kidney Disease: Improving Global hemodialysis patients. Am. J. Kidney Dis. 2006 Oct;48(4):606–615
Outcomes controversies conference. Kidney Int. 2010;77:273–28 15 Davenport A, Cox C, Thuraisingham R. Achieving blood pressure
3 Robinson BM TL, Zhang J, Wolfe RA, Goodkin DA, Greenwood RN, targets during dialysis improves control but increases intradialytic
Kerr PG, Morgenstern H, Li Y, Pisoni RL, Saran R, Tentori F, Akizawa hypotension. Kidney Int. 2008 Mar;73(6):759–764
T, Fukuhara S, Port FK. Blood pressure levels and mortality risk among 16 Shoji T, Tsubakihara Y, Fujii M, Imai E. Hemodialysis-associated
hemodialysis patients in the Dialysis Outcomes and Practice Patterns hypotension as an independent risk factor for two-year mortality in
Study. Kidney Int 2012;82(5):570–580 hemodialysis patients. Kidney Int. 2004 Sep;66(3):1212–1220
4 Chazot C, Vo-Van C, Deleaval P, Lorriaux C, Hurot JM, Mayor B, 17 Burton JO, Jefferies HJ, Selby NM, McIntyre CW. Hemodialysis-induced
Jean G: Predialysis systolic blood pressure evolution in incident cardiac injury: determinants and associated outcomes. Clin J Am Soc
hemodialysis patients: effects of the dry weight method and prognostic Nephrol. 2009 May;4(5):914–920
value. Blood Purif. 2012;33:275–283 18 Zager PG, Nikolic J, Brown RH, Campbell MA, Hunt WC, Peterson D,
5 The FHN Trial Group. In-Center Hemodialysis Six Times per Week et al. ‘‘U’’ curve association of blood pressure and mortality in hemo-
versus Three Times per Week. New England Journal of Medicine. 2010; dialysis patients. Medical Directors of Dialysis Clinic, Inc. Kidney Int.
363:2287–2300 1998 Aug;54(2):561–569
6 Krapf R, Hulter HN. Arterial Hypertension Induced by Erythropoietin 19 Holt S, Goldsmith D: UK Renal Association Clinical Practice Guidelines
and Erythropoiesis-Stimulating Agents (ESA). Clinical Journal of the Committee: 2010 RA Guidelines – Cardiovascular disease in CKD,
American Society of Nephrology 2009;4(2):470–480 5th Edition. 2010. http://www.renal.org/Clinical/GuidelinesSection/
7 Oberbauer R, Kreis H, Johnson RW, Mota A, Claesson K, Ruiz JC, CardiovascularDiseaseInCKD.aspx
Wilczek H, Jamieson N, Henriques AC, Paczek L, Chapman J, Burke 20 Levin NW, Kotanko P, Eckardt KU, et al. Blood pressure in chronic
JT: Rapamune Maintenance Regimen Study Group: Long-term improve- kidney disease stage 5D-report from a Kidney Disease: Improving Global
ment in renal function with sirolimus after early cyclosporine with- Outcomes controversies conference. Kidney Int 2010;77(4):273–284
drawal in renal transplant recipients: 2–year results of the Rapamune 21 KDIGO clinical practice guideline for the care of kidney transplant
Maintenance Regimen Study. Transplantation 2003;76:364–370 recipients. Am J Transplant 2009;9(suppl 3):S1–155
8 Rohrscheib MR, Myers OB, Servilla KS, Adams CD, Miskulin D, 22 Ansell D, Tomson CR. UK Renal Registry 11th Annual Report
Bedrick EJ, Hunt WC, Lindsey DE, Gabaldon D, Zager PG: Age-related (December 2008): Chapter 15 The UK Renal Registry, UKRR database,
blood pressure patterns and blood pressure variability among hemo- validation and methodology. Nephron Clin Pract 2009;111(suppl 1):
dialysis patients. Clin J Am Soc Nephrol 2008;3:1407–1414 c277–285
9 Di Iorio B DML, Torraca S, Sirico ML, Guastaferro P, Chiuchiolo L, 23 Webb L, Tomson CRV, Casula A, Farrington K: Registry 13th Annual
Nigro F, De Blasio A, Romano P, Pota A, Rubino R, Morrone L, Report (December 2010): Chapter 11 Blood Pressure Profile of Prevalent
Lopez T, Casino FG. Variability of blood pressure in dialysis patients: Patients receiving Renal Replacement Therapy in England, Wales and
a new marker of cardiovascular risk. J Nephrol 2013;26(1):173–182 Northern Ireland in 2009: National and Centre-Specific Analyses.
10 Agarwal R. Blood pressure and mortality among hemodialysis patients. Nephron Clin Pract 2011;119(suppl 2):c215–c224
Hypertension 2010;55(3):762–768 24 Foley RN, Parfrey PS, Harnett JD, Kent GM, Murray DC, Barre PE:
11 Zoccali C. Arterial pressure components and cardiovascular risk in end- Impact of hypertension on cardiomyopathy, morbidity and mortality
stage renal disease. Nephrol. Dial. Transplant. 2003 Feb;18(2):249–252 in end-stage renal disease. Kidney Int 1996;49:1379–1385
12 Mazzuchi N, Carbonell E, Fernández-Cean J. Importance of blood 25 Agarwal R, Peixoto, AJ, Santos SFF, Zoccali C. Pre- and post-dialysis
pressure control in hemodialysis patient survival. Kidney Int. 2000 blood pressures are imprecise estimates of interdialytic ambulatory
Nov;58(5):2147–2154 blood pressure. Clin J Am Soc Nephrol 2006;1:389–398
13 Tomita J, Kimura G, Inoue T, Inenaga T, Sanai T, Kawano Y, et al. Role 26 Mitra S, Chandna SM, Farrington K. What is hypertension in chronic
of systolic blood pressure in determining prognosis of hemodialyzed haemodialysis? The role of interdialytic blood pressure monitoring.
patients. Am. J. Kidney Dis. 1995 Mar;25(3):405–412 Nephrol Dial Transplant. 1999;14(12):2915–21
226
Chapter 12 Biochemical Variables
amongst UK Adult Dialysis patients in
2012: National and Centre-specific
Analyses
Johann Nicholasa, Catriona Shawb, David Pitcherb, Anne Dawnayc

a
Royal Wolverhampton NHS Trust, UK; bUK Renal Registry, Bristol, UK; cUniversity College London Hospitals,
London, UK
Key Words
. 77% of HD and 78% of PD patients had adjusted
Bicarbonate . Biochemical variables . Calcium . Cholesterol . calcium between 2.2–2.5 mmol/L.
Dialysis . Haemodialysis . Parathyroid hormone . Peritoneal
. 58% of HD and 65% of PD patients had a serum
dialysis . Phosphate . Quality improvement PTH between 16–72 pmol/L.
. 16% of HD and 10% of PD patients had a serum
PTH .72 pmol/L.
Summary . Simultaneous control of all three parameters within
current audit standards was achieved by 51% of HD
. 56% of HD patients and 61% of PD patients and PD patients.
achieved the audit measure for phosphate. . 59% of HD and 80% of PD patients achieved the
. 32% of HD and 32% of PD patients had a serum audit measure for bicarbonate.
phosphate above the audit standard range.
227
Introduction modality or renal centre in the last 90 days. Haemodialysis (HD)

and peritoneal dialysis (PD) cohorts were analysed separately. A
The UK Renal Registry (UKRR) collects routine bio- full definition of this cohort including inclusion and exclusion
criteria is included in appendix B www.renalreg.com.
chemical data from clinical information systems in The biochemical variables analysed were phosphate, calcium,
renal centres in England, Wales and Northern Ireland parathyroid hormone, bicarbonate and cholesterol. The method
and receives data from Scotland via the Scottish Renal of data collection and validation by the UKRR has been described
Registry. Annual cross sectional analyses are undertaken elsewhere [3]. For each quarter of 2012 the UKRR extracted bio-
on some of these variables to determine centre level per- chemical data electronically from clinical information systems in
UK dialysis centres. The UKRR does not collect data regarding
formance against national (Renal Association) clinical different assay methods mainly because a single dialysis centre
performance measures [1]. This enables UK renal centres may process samples in several different laboratories. Scottish
to compare their own performance against each other centres have only been included in analyses relating to phosphate
and to the UK average performance [2]. Currently the control, with data for their prevalent dialysis cohort being supplied
directly by the Scottish Renal Registry. The audit measure used for
5th edition of the UK Renal Association clinical practice
serum phosphate was 1.1–1.7 mmol/L in both the HD and PD
guidelines is in practice [1]. This edition commenced in a cohorts [1, 3]. For centres providing adjusted calcium values,
graded manner in 2009 and includes an expanded num- these data were analysed directly as it is these values on which
ber of guideline modules compared to previous editions. clinical decisions within centres are based. For centres providing
Audit measures for kidney disease increasingly include unadjusted calcium values, a formula in widespread use was
used to calculate adjusted calcium [4]. The audit measure for
tighter specification limits in conjunction with a growing
adjusted calcium depends on a local reference range [1, 3]. The
evidence base. Out of range observations (e.g. hyperpho- UKRR has used the RA guideline standard of adjusted calcium
sphataemia and hypophosphataemia) need to be inter- between 2.2–2.5 mmol/L as the audit measure for these analyses.
preted cautiously as they may relate to different clinical There are also a variety of methods and reference ranges in use
problems or population characteristics. These will there- to measure parathyroid hormone (PTH). To enable some form
of comparative audit the UKRR has used 2–9 times the median
fore require different strategies to improve centre per- upper limit of the reference range (8 pmol/L) as the audit measure
formance of clinical audit measures. To supplement in line with the 5th edition of the Renal Association clinical
these performance analyses, summary statistical data practice guidelines that were current during 2012 and KDIGO
have been provided to enhance understanding of the 2009 guidance [3, 5]. This equates to a PTH of 16–72 pmol/L.
population characteristics of each centre and longitudinal The audit measure used for serum bicarbonate in the HD cohort
was 18–24 mmol/L as per the updated haemodialysis guidelines
analyses demonstrate changes over time. and in the PD cohort was 22–30 mmol/L [1]. A summary of the
current Renal Association audit measures and conversion factors
to SI units are given in table 12.1.
Quarterly values were extracted from the database for the last
two quarters for calcium, phosphate and bicarbonate; the last
Methods three quarters for PTH and the entire year for cholesterol.
Patients who did not have these data were excluded from the
These analyses relate to biochemical variables in the prevalent analyses. Data completeness was analysed at centre and country
dialysis cohort in England, Wales and Northern Ireland in 2012. level. All patients were included in analyses but centres with less
Scotland is also included in analyses pertaining to phosphate than 50% completeness were excluded from plots showing centre
control. The cohort studied were patients prevalent on dialysis level performance. Data were also excluded from plots when
treatment on 31st December 2012, excluding patients receiving there were less than 20 patients with data both at centre or
dialysis for less than 90 days and those who had changed country level. These data were analysed to calculate summary
Table 12.1. Summary of clinical audit measures and conversion factors from SI units
Biochemical variable Clinical audit measure Conversion factor from SI units
Phosphate HD Patients: 1.1–1.7 mmol/L mg/dl = mmol/L × 3.1

PD Patients: 1.1–1.7 mmol/L
Calcium (adjusted) Normal range (ideally <2.5 mmol/L) mg/dl = mmol/L × 4
Parathyroid hormone 2–9 times upper limit of normal ng/L = pmol × 9.5
Bicarbonate HD Patients: 18–24 mmol/L mg/dl = mmol/L × 6.1
PD Patients: 22–30 mmol/L
Cholesterol No audit measure mg/dl = mmol/L × 38.6
228
Chapter 12 Management of biochemical variables
descriptive statistics (maximum, minimum, mean and median The data completeness for serum phosphate across the
values in addition to standard deviation and quartile ranges). UK was 96% for HD patients and 98% for PD patients
Where applicable, the percentage achieving the Renal Associ-
ation standard or other surrogate clinical performance measure
although there was considerable variation between
was also calculated. centres (tables 12.2 and 12.4). The individual centre
The simultaneous control of all three components of bone and means and standard deviations are shown in tables 12.2
mineral disorder (BMD) parameters were analysed in combi- and 12.4. Fifty-six percent (95% CI 55–57%) of HD
nation. Thus, the control of none, one, two or three parameters, patients and 61% (95% CI 59–63%) of PD patients
as well as an analysis of combinations of calcium-PTH, calcium-
phosphate and phosphate-PTH were collated, with an emphasis
achieved a phosphate level within the target range speci-
on evaluating the effective management and prevention of severe fied by the RA clinical audit measure (tables 12.3, 12.5).
hyperparathyroidism (maintaining PTH 472 pmol/L). For the The proportion of HD patients with hyperphosphataemia
purpose of this analysis, the corrected calcium standard of was 32% and the proportion with hypophosphataemia
between 2.2–2.5 mmol/L, a phosphate level being maintained at was 12% (table 12.3, figures 12.1, 12.2). The proportion
or below 1.7 mmol/L and a PTH level being at or below
72 pmol/L, were evaluated in combination.
of PD patients with hyperphosphataemia was 32% and
The analyses presented in this chapter are descriptive. As data the proportion with hypophosphataemia was 7%
are provided unadjusted for confounding factors and due to con- (table 12.5, figures 12.3, 12.4). Longitudinal analysis
cerns regarding measurement error in many of the biochemical showed a trend towards improved phosphate control
parameters, hypothesis testing was not utilised. for England, Northern Ireland and Wales combined
Centres report several biochemical variables with different
levels of accuracy, leading to problems in comparative evaluation.
between 2002 and 2012 that has plateaued in recent
For example, in the case of serum bicarbonate, data can be sub- years (figure 12.5).
mitted as integer values but some centres submit data to one dec- There was significant between centre variation in the
imal place. All data has been rounded up in an attempt to make all proportion of patients below, within and above the phos-
centres more comparable. phate range specified by the clinical performance
The number preceding the centre name in each figure indicates
the percentage of missing data for that centre. Funnel plot analysis measure (figures 12.1–12.4). Of note, the percentage of
was used to identify ‘outlying centres’ [6]. The percentage achiev- PD patients achieving the target decreased substantially
ing each standard was plotted against centre size along with the from 2011 for Birmingham Heartlands (from 66% to
upper and lower 95% and 99.9% limits. Centres can be identified 43%) and for Cambridge (from 72% to 47%). The same
on these plots by looking up the number of patients treated in each fall was not seen for HD patients at these centres. If the
centre provided in the relevant table and finding this value on the
x-axis. Longitudinal analyses were performed for some data to phosphate analyses for both HD and PD patients were
calculate overall changes in achievement of a performance conducted in the same laboratories for each centre, it
measure annually from 2002 to 2012 and were recalculated for suggests that this was not due to any change in laboratory
each previous year using the rounding procedure. All data were methods.
unadjusted for case-mix.
Adjusted calcium
In 2012, the following Renal Association clinical practice
guideline regarding calcium management was applicable:
Results and discussions
Guideline 2.2 CKD-MBD: Serum calcium in dialysis
Mineral and bone variables patients (stage 5D)
Phosphate
In 2012 the following Renal Association clinical ‘We suggest that serum calcium, adjusted for albumin
practice guideline regarding phosphate management concentration, should be maintained within the normal
was applicable: reference range for the laboratory used, measured before
a ‘‘short-gap’’ dialysis session in haemodialysis patients.
Guideline 3.2 CKD-MBD: Serum phosphate in Ideally, adjusted serum calcium should be maintained
dialysis patients between 2.2 and 2.5 mmol/L, with avoidance of hyper-
calcaemic episodes (2D)’ [3]
‘We suggest that serum phosphate in dialysis patients,
measured before a ‘‘short-gap’’ dialysis session in haemo- The current guidelines are based upon adjusted serum
dialysis patients, should be maintained between 1.1 and calcium. A variety of formulae have been proposed to
1.7 mmol/L (2C)’ [3] permit calculation of the ‘adjusted’ total calcium (i.e. an
229
Table 12.2. Summary statistics for phosphate in haemodialysis patients in 2012

% Patients with data Lower Upper
Centre completeness N Mean SD Median quartile quartile
England
B Heart 100.0 401 1.7 0.6 1.6 1.3 2.0
B QEH 96.2 831 1.6 0.5 1.5 1.3 1.8
Basldn 97.3 146 1.4 0.4 1.4 1.2 1.7
Bradfd 98.4 186 1.4 0.6 1.3 1.0 1.7
Brightn 95.6 323 1.6 0.5 1.6 1.2 1.9
Bristol 100.0 461 1.6 0.5 1.5 1.3 1.9
Camb 95.7 310 1.5 0.4 1.5 1.2 1.8
Carlis 100.0 57 1.8 0.7 1.6 1.4 1.9
Carsh 93.4 652 1.6 0.5 1.5 1.3 1.9
Chelms 100.0 121 1.5 0.4 1.5 1.2 1.8
Colchr 92.6 100 1.6 0.4 1.6 1.3 1.8
Covnt 99.7 334 1.7 0.5 1.6 1.3 2.0
Derby 99.5 208 1.6 0.5 1.5 1.3 1.9
Donc 100.0 158 1.5 0.5 1.4 1.2 1.7
Dorset 99.6 243 1.6 0.5 1.5 1.2 1.9
Dudley 100.0 153 1.7 0.5 1.6 1.3 2.0
Exeter 100.0 351 1.5 0.5 1.5 1.2 1.8
Glouc 100.0 193 1.5 0.5 1.4 1.2 1.8
Hull 100.0 310 1.5 0.5 1.5 1.2 1.8
Ipswi 100.0 124 1.5 0.5 1.4 1.2 1.7
Kent 98.3 355 1.7 0.5 1.6 1.3 1.9
L Barts 99.8 844 1.6 0.5 1.6 1.3 1.9
L Guys 89.0 527 1.5 0.5 1.5 1.2 1.8
L Kings 99.8 459 1.5 0.4 1.4 1.2 1.7
L Rfree 84.4 564 1.5 0.5 1.5 1.2 1.8
L St.G 96.3 261 1.5 0.5 1.5 1.2 1.8
L West 98.6 1,323 1.5 0.5 1.5 1.2 1.8
Leeds 100.0 454 1.6 0.5 1.5 1.2 1.9
Leic 99.8 799 1.7 0.5 1.6 1.3 2.0
Liv Ain 98.2 163 1.5 0.5 1.4 1.1 1.8
Liv RI 99.4 343 1.5 0.5 1.4 1.1 1.8
M RI 92.2 437 1.6 0.5 1.5 1.2 1.9
Middlbr 99.4 310 1.6 0.5 1.5 1.3 1.9
Newc 100.0 262 1.6 0.5 1.5 1.3 1.9
Norwch 100.0 303 1.6 0.5 1.6 1.3 1.9
Nottm 99.7 354 1.5 0.5 1.5 1.2 1.8
Oxford 100.0 389 1.6 0.5 1.6 1.3 1.9
Plymth 100.0 119 1.5 0.5 1.4 1.2 1.8
Ports 99.8 509 1.7 0.5 1.6 1.4 2.0
Prestn 99.6 494 1.7 0.5 1.6 1.3 1.9
Redng 100.0 251 1.5 0.4 1.5 1.2 1.8
Salford 88.1 304 1.5 0.6 1.5 1.1 1.8
Sheff 99.8 561 1.6 0.5 1.6 1.3 1.8
Shrew 99.5 183 1.6 0.6 1.5 1.2 1.8
Stevng 99.2 377 1.7 0.5 1.6 1.3 1.9
Sthend 100.0 107 1.6 0.5 1.6 1.3 1.9
Stoke 86.1 253 1.5 0.5 1.4 1.2 1.8
Sund 0.0 0
Truro 99.3 133 1.5 0.5 1.4 1.2 1.8
Wirral 97.7 173 1.5 0.5 1.5 1.2 1.8
Wolve 98.9 267 1.5 0.6 1.4 1.1 1.8
York 100.0 122 1.5 0.5 1.4 1.1 1.7
N Ireland
Antrim 100.0 126 1.4 0.5 1.3 1.1 1.7
Belfast 99.0 206 1.5 0.5 1.5 1.2 1.8
Newry 100.0 85 1.7 0.5 1.6 1.3 2.0
Ulster 100.0 101 1.6 0.4 1.5 1.3 1.7
West NI 100.0 129 1.7 0.5 1.6 1.3 1.9
230

Scotland
Abrdn 94.4 202 1.5 0.5 1.4 1.1 1.8
Airdrie 93.8 165 1.5 0.5 1.4 1.1 1.9
D & Gall 95.8 46 1.5 0.5 1.5 1.2 1.8
Dundee 98.8 169 1.6 0.5 1.6 1.4 1.9
Dunfn 95.0 133 1.7 0.5 1.7 1.4 2.0
Edinb 94.0 235 1.7 0.5 1.6 1.3 1.9
Glasgw 86.0 498 1.6 0.5 1.6 1.3 1.9
Inverns 74.0 54 1.8 0.6 1.8 1.4 2.1
Klmarnk 88.7 125 1.4 0.5 1.4 1.0 1.7
Wales
Bangor 100.0 82 1.6 0.4 1.5 1.3 1.9
Cardff 99.3 445 1.6 0.5 1.5 1.2 1.8
Clwyd 100.0 76 1.6 0.6 1.5 1.3 1.9
Swanse 100.0 308 1.5 0.5 1.4 1.2 1.8
Wrexm 100.0 86 1.3 0.4 1.3 1.1 1.6
England 96.4 17,662 1.6 0.5 1.5 1.2 1.8
N Ireland 99.7 647 1.6 0.5 1.5 1.2 1.8
Scotland 90.8 1,627 1.6 0.5 1.5 1.2 1.9
Wales 99.7 997 1.5 0.5 1.5 1.2 1.8
UK 96.2 20,933 1.6 0.5 1.5 1.2 1.8
Blank cells denote no data returned
Table 12.3. Percentage of haemodialysis patients within, below and above the range specified in the RA audit measure for phosphate
(1.1–1.7 mmol/L) in 2012
Change in %
% phos Lower Upper % phos % phos within range 95% LCL 95% UCL
Centre N 1.1–1.7 mmol/L 95% CI 95% CI ,1.1 mmol/L .1.7 mmol/L from 2011 change change
England
B Heart 401 52.4 47.5 57.2 11.0 36.7 −3.7 −10.5 3.2
B QEH 831 58.1 54.7 61.4 11.0 30.9 −1.4 −6.1 3.4
Basldn 146 62.3 54.2 69.8 17.1 20.6 7.9 −3.7 19.4
Bradfd 186 50.5 43.4 57.7 25.8 23.7 1.4 −8.9 11.7
Brightn 323 55.1 49.6 60.5 9.9 35.0 0.9 −6.9 8.7
Bristol 461 53.8 49.2 58.3 10.0 36.2 −2.0 −8.5 4.4
Camb 310 65.2 59.7 70.3 9.7 25.2 0.3 −7.1 7.7
Carlis 57 52.6 39.8 65.1 8.8 38.6 −0.8 −19.1 17.4
Carsh 652 58.7 54.9 62.5 8.9 32.4 −3.2 −8.5 2.1
Chelms 121 65.3 56.4 73.2 8.3 26.5 7.9 −4.5 20.3
Colchr 100 71.0 61.4 79.0 4.0 25.0 9.0 −4.0 22.0
Covnt 334 56.6 51.2 61.8 6.0 37.4 −4.7 −12.2 2.8
Derby 208 55.8 49.0 62.4 9.6 34.6 −2.2 −12.0 7.6
Donc 158 64.6 56.8 71.6 12.7 22.8 2.5 −8.2 13.2
Dorset 243 54.7 48.4 60.9 14.0 31.3 −11.6 −20.5 −2.8
Dudley 153 52.9 45.0 60.7 7.2 39.9 1.5 −10.1 13.0
Exeter 351 58.1 52.9 63.2 12.5 29.3 −3.4 −10.8 4.0
Glouc 193 59.1 52.0 65.8 14.5 26.4 −9.8 −19.4 −0.1
Hull 310 60.0 54.4 65.3 13.6 26.5 −1.7 −9.4 6.0
Ipswi 124 59.7 50.8 67.9 16.1 24.2 6.0 −6.4 18.3
Kent 355 53.5 48.3 58.7 9.0 37.5 −3.1 −10.4 4.3
L Barts 844 51.4 48.1 54.8 12.0 36.6 0.6 −4.2 5.4
L Guys 527 59.0 54.8 63.1 14.0 26.9 4.0 −2.2 10.1
L Kings 459 64.1 59.6 68.3 13.5 22.4 2.4 −3.9 8.8
L Rfree 564 56.9 52.8 61.0 14.5 28.6 0.8 −5.0 6.6
231
Change in %
L St.G 261 55.2 49.1 61.1 18.8 26.1 2.8 −5.7 11.2
L West 1,323 57.9 55.2 60.5 15.0 27.1 3.4 −0.4 7.2
Leeds 454 51.1 46.5 55.7 15.6 33.3 −4.6 −11.0 1.9
Leic 799 52.4 49.0 55.9 8.5 39.1 −8.8 −13.7 −3.9
Liv Ain 163 56.4 48.7 63.9 18.4 25.2 −5.7 −16.5 5.2
Liv RI 343 53.6 48.4 58.9 19.0 27.4 −0.6 −8.0 6.7
M RI 437 51.0 46.4 55.7 15.1 33.9 −2.8 −9.6 4.0
Middlbr 310 55.5 49.9 60.9 11.3 33.2 −2.7 −10.7 5.3
Newc 262 56.5 50.4 62.4 10.3 33.2 2.3 −6.4 11.0
Norwch 303 59.4 53.8 64.8 6.6 34.0 −1.8 −9.7 6.0
Nottm 354 57.9 52.7 63.0 15.3 26.8 −5.6 −12.7 1.4
Oxford 389 54.5 49.5 59.4 10.0 35.5 2.1 −5.0 9.2
Plymth 119 60.5 51.5 68.9 14.3 25.2 0.3 −12.0 12.7
Ports 509 52.1 47.7 56.4 9.0 38.9 5.2 −1.1 11.4
Prestn 494 51.6 47.2 56.0 9.9 38.5 −2.9 −9.1 3.4
Redng 251 58.2 52.0 64.1 14.3 27.5 −5.7 −14.2 2.9
Salford 304 53.0 47.3 58.5 21.1 26.0 2.5 −5.6 10.5
Sheff 561 59.5 55.4 63.5 8.6 31.9 −0.8 −6.6 4.9
Shrew 183 54.1 46.8 61.2 16.4 29.5 1.5 −8.9 11.9
Stevng 377 56.5 51.4 61.4 6.9 36.6 4.8 −2.3 11.9
Sthend 107 46.7 37.5 56.2 13.1 40.2 5.5 −7.6 18.6
Stoke 253 57.7 51.5 63.7 13.0 29.3 −3.9 −12.2 4.3
Truro 133 57.1 48.6 65.3 16.5 26.3 −3.0 −14.7 8.7
Wirral 173 56.7 49.2 63.8 14.5 28.9 −1.4 −12.0 9.2
Wolve 267 54.3 48.3 60.2 18.7 27.0 2.9 −5.4 11.2
York 122 59.0 50.1 67.4 16.4 24.6 4.2 −8.4 16.8
N Ireland
Antrim 126 57.1 48.4 65.5 22.2 20.6 4.7 −7.7 17.1
Belfast 206 51.5 44.7 58.2 17.0 31.6 −2.5 −12.1 7.2
Newry 85 50.6 40.1 61.0 10.6 38.8 −1.4 −15.9 13.0
Ulster 101 68.3 58.6 76.6 7.9 23.8 3.0 −10.0 15.9
West NI 129 53.5 44.9 61.9 7.8 38.8 −3.2 −15.2 8.8
Scotland
Abrdn 202 56.9 50.0 63.6 17.3 25.7 2.5 −7.3 12.3
Airdrie 165 49.1 41.5 56.7 23.6 27.3 −4.7 −15.6 6.2
D & Gall 46 56.5 42.1 70.0 13.0 30.4 −1.0 −22.0 20.0
Dundee 169 52.1 44.6 59.5 10.7 37.3 −4.5 −15.1 6.1
Dunfn 133 56.4 47.9 64.6 2.3 41.4 −2.7 −14.5 9.1
Edinb 235 54.5 48.1 60.7 7.7 37.9 −2.9 −11.9 6.1
Glasgw 498 51.6 47.2 56.0 11.9 36.6 −4.8 −11.0 1.4
Inverns 54 38.9 26.9 52.4 9.3 51.9 −5.6 −22.9 11.8
Klmarnk 125 52.8 44.1 61.4 25.6 21.6 −4.2 −16.3 7.9
Wales
Bangor 82 64.6 53.8 74.2 4.9 30.5 −0.1 −14.6 14.4
Cardff 445 58.7 54.0 63.1 10.6 30.8 3.2 −3.3 9.7
Clwyd 76 54.0 42.7 64.8 11.8 34.2 −2.0 −18.9 14.9
Swanse 308 62.3 56.8 67.6 10.7 27.0 −0.3 −7.9 7.2
Wrexm 86 59.3 48.7 69.1 23.3 17.4 17.8 3.0 32.7
England 17,662 56.2 55.5 56.9 12.4 31.4 −0.7 −1.7 0.3
N Ireland 647 55.5 51.6 59.3 13.9 30.6 −0.2 −5.6 5.2
Scotland 1,627 52.7 50.2 55.1 13.2 34.1 −3.2 −6.6 0.2
Wales 997 60.0 56.9 63.0 11.3 28.7 2.5 −1.8 6.8
UK 20,933 56.1 55.4 56.8 12.4 31.5 −0.7 −1.7 0.2
232
Table 12.4. Summary statistics for phosphate in peritoneal dialysis patients in 2012
England
B Heart 100.0 42 1.7 0.4 1.75 1.5 2
B QEH 98.0 146 1.6 0.5 1.5 1.2 1.8
Basldn 96.4 27 1.5 0.4 1.4 1.2 1.8
Bradfd 95.8 23 1.7 0.4 1.7 1.4 2
Brightn 94.2 65 1.6 0.4 1.6 1.2 1.9
Bristol 100.0 56 1.8 0.5 1.7 1.5 2
Camb 100.0 32 1.5 0.4 1.45 1.05 1.8
Carlis 100.0 21 1.6 0.4 1.4 1.3 1.7
Carsh 97.9 95 1.6 0.4 1.5 1.3 1.8
Chelms 100.0 25 1.5 0.4 1.5 1.2 1.8
Colchr
Covnt 91.7 77 1.4 0.3 1.4 1.2 1.6
Derby 100.0 84 1.5 0.4 1.4 1.2 1.7
Donc 100.0 23 1.7 0.4 1.6 1.3 1.9
Dorset 92.1 35 1.5 0.3 1.5 1.2 1.7
Dudley 100.0 53 1.8 0.4 1.8 1.5 2.2
Exeter 98.6 68 1.6 0.4 1.6 1.3 1.8
Glouc 96.8 30 1.6 0.4 1.6 1.3 1.8
Hull 96.2 76 1.7 0.4 1.6 1.4 1.9
Ipswi 100.0 30 1.6 0.4 1.55 1.3 1.8
Kent 98.2 54 1.7 0.5 1.6 1.4 1.9
L Barts 98.8 165 1.6 0.5 1.5 1.2 1.8
L Guys 96.3 26 1.5 0.4 1.5 1.3 1.7
L Kings 100.0 76 1.6 0.4 1.5 1.3 1.8
L Rfree 99.0 101 1.6 0.4 1.5 1.4 1.8
L St.G 97.9 47 1.5 0.3 1.5 1.3 1.7
L West 100.0 47 1.5 0.4 1.4 1.2 1.7
Leeds 100.0 77 1.7 0.4 1.7 1.4 1.9
Leic 97.9 140 1.6 0.4 1.6 1.3 1.8
Liv Ain 100.0 17
Liv RI 98.2 54 1.5 0.4 1.45 1.2 1.7
M RI 100.0 76 1.7 0.4 1.65 1.4 1.9
Middlbr 87.5 7
Newc 86.5 32 1.7 0.3 1.7 1.45 2
Norwch 100.0 48 1.5 0.4 1.5 1.3 1.75
Nottm 100.0 72 1.6 0.5 1.6 1.3 1.85
Oxford 100.0 69 1.7 0.5 1.6 1.3 1.9
Plymth 93.6 29 1.6 0.4 1.5 1.2 1.8
Ports 100.0 78 1.6 0.4 1.5 1.4 1.8
Prestn 98.3 58 1.7 0.4 1.7 1.4 2
Redng 100.0 63 1.5 0.4 1.5 1.3 1.7
Salford 93.3 84 1.6 0.5 1.6 1.3 1.9
Sheff 100.0 67 1.7 0.4 1.6 1.4 1.9
Shrew 97.0 32 1.8 0.5 1.6 1.45 1.95
Stevng 100.0 27 1.5 0.3 1.5 1.4 1.7
Sthend 100.0 14
Stoke 100.0 69 1.6 0.5 1.7 1.3 1.9
Sund 100.0 17
Truro 100.0 19
Wirral 72.4 21 1.6 0.3 1.5 1.4 1.6
Wolve 97.6 81 1.6 0.5 1.5 1.3 1.9
York 100.0 27 1.7 0.4 1.6 1.4 2
N Ireland
Antrim 100.0 10
Belfast 100.0 25 1.5 0.4 1.6 1.2 1.8
Newry 100.0 14
Ulster 100.0 6
West NI 100.0 15
233

Scotland
Abrdn 100.0 20 1.7 0.3 1.75 1.4 1.9
Airdrie 100.0 10
D & Gall 92.9 13
Dundee 94.7 18
Dunfn 95.0 19
Edinb 100.0 35 1.7 0.5 1.7 1.2 2.1
Glasgw 100.0 40 1.7 0.4 1.7 1.4 1.9
Inverns 86.7 13
Klmarnk 100.0 40 1.6 0.3 1.6 1.4 1.85
Wales
Bangor 100.0 14
Cardff 98.6 70 1.5 0.4 1.45 1.2 1.7
Clwyd 100.0 15
Swanse 98.2 53 1.5 0.4 1.5 1.3 1.7
Wrexm 95.0 19
England 97.8 2,802 1.6 0.4 1.6 1.3 1.8
N Ireland 100.0 70 1.6 0.4 1.6 1.2 1.8
Scotland 97.7 208 1.7 0.4 1.6 1.4 1.9
Wales 98.3 171 1.5 0.4 1.5 1.3 1.8
UK 97.9 3,251 1.6 0.4 1.6 1.3 1.8
Blank cells denote centres excluded from analyses due to low patient numbers or poor data completeness
Table 12.5. Percentage of peritoneal dialysis patients within, below and above the range specified in the RA audit measure for
phosphate (1.1–1.7 mmol/L) in 2012
Change in %
England
B Heart 42 42.9 28.9 58.0 7.1 50.0 −22.9 −44.2 −1.7
B QEH 146 67.8 59.8 74.9 6.9 25.3 4.1 −6.8 15.0
Basldn 27 59.3 40.3 75.8 11.1 29.6 0.9 −26.1 28.0
Bradfd 23 56.5 36.3 74.8 0.0 43.5 1.0 −26.6 28.6
Brightn 65 49.2 37.4 61.2 10.8 40.0 −16.4 −33.2 0.4
Bristol 56 53.6 40.6 66.1 1.8 44.6 −2.4 −20.6 15.8
Camb 32 46.9 30.6 63.9 25.0 28.1 −25.0 −48.3 −1.7
Carlis 21 76.2 54.0 89.7 0.0 23.8 n/a n/a n/a
Carsh 95 64.2 54.1 73.2 5.3 30.5 0.1 −13.7 13.8
Chelms 25 52.0 33.1 70.4 16.0 32.0 2.0 −26.6 30.6
Covnt 77 76.6 65.9 84.8 9.1 14.3 9.1 −5.0 23.2
Derby 84 63.1 52.3 72.7 16.7 20.2 −2.5 −16.6 11.5
Donc 23 56.5 36.3 74.8 8.7 34.8 −19.7 −46.9 7.6
Dorset 35 65.7 48.8 79.4 11.4 22.9 −9.3 −29.6 11.0
Dudley 53 43.4 30.8 56.9 1.9 54.7 −3.5 −22.9 15.8
Exeter 68 63.2 51.2 73.8 10.3 26.5 −5.6 −21.9 10.7
Glouc 30 63.3 45.1 78.4 6.7 30.0 7.1 −17.3 31.4
Hull 76 61.8 50.5 72.0 2.6 35.5 0.5 −15.0 16.0
Ipswi 30 63.3 45.1 78.4 6.7 30.0 3.3 −21.3 27.9
Kent 54 53.7 40.5 66.5 11.1 35.2 −16.8 −34.3 0.8
L Barts 165 58.2 50.5 65.5 11.5 30.3 −13.6 −24.1 −3.2
L Guys 26 73.1 53.3 86.6 7.7 19.2 23.1 −2.1 48.3
L Kings 76 63.2 51.8 73.2 9.2 27.6 −1.1 −16.7 14.5
L Rfree 101 62.4 52.6 71.3 7.9 29.7 −7.6 −21.4 6.2
234

Change in %
L St.G 47 74.5 60.2 84.9 2.1 23.4 11.7 −6.5 29.9

L West 47 70.2 55.8 81.5 8.5 21.3 4.6 −16.4 25.6
Leeds 77 58.4 47.2 68.9 3.9 37.7 −5.8 −20.9 9.4
Leic 140 63.6 55.3 71.1 7.9 28.6 3.4 −8.0 14.8
Liv RI 54 72.2 58.9 82.5 5.6 22.2 0.3 −16.4 17.0
M RI 76 52.6 41.5 63.5 6.6 40.8 −3.7 −19.8 12.4
Newc 32 56.3 39.0 72.1 0.0 43.8 −2.3 −25.2 20.6
Norwch 48 60.4 46.1 73.1 14.6 25.0 −12.5 −31.2 6.2
Nottm 72 52.8 41.3 64.0 9.7 37.5 0.1 −16.1 16.3
Oxford 69 55.1 43.3 66.3 5.8 39.1 3.9 −12.1 19.8
Plymth 29 62.1 43.6 77.6 6.9 31.0 1.0 −22.8 24.7
Ports 78 65.4 54.2 75.1 2.6 32.1 12.8 −2.5 28.1
Prestn 58 58.6 45.7 70.5 1.7 39.7 −4.3 −22.4 13.7
Redng 63 68.3 55.9 78.5 7.9 23.8 −1.2 −16.9 14.5
Salford 84 54.8 44.1 65.0 7.1 38.1 −0.6 −15.2 14.1
Sheff 67 58.2 46.2 69.4 3.0 38.8 −17.7 −34.1 −1.3
Shrew 32 62.5 44.9 77.3 0.0 37.5 −6.7 −31.1 17.7
Stevng 27 81.5 62.5 92.1 7.4 11.1 23.8 −0.2 47.8
Stoke 69 55.1 43.3 66.3 7.3 37.7 −14.5 −30.5 1.5
Wirral 21 76.2 54.0 89.7 4.8 19.1 20.2 −6.5 46.8
Wolve 81 59.3 48.3 69.4 7.4 33.3 −0.4 −16.7 15.8
York 27 55.6 36.9 72.8 3.7 40.7 −4.4 −32.9 24.1
N Ireland
Belfast 25 48.0 29.6 66.9 16.0 36.0 −23.4 −49.2 2.3
Scotland
Abrdn 20 50.0 29.4 70.6 0.0 50.0 −4.6 −34.8 25.7
Edinb 35 45.7 30.2 62.1 8.6 45.7 −2.9 −26.2 20.5
Glasgw 40 52.5 37.3 67.3 5.0 42.5 −2.7 −26.5 21.1
Klmarnk 40 65.0 49.2 78.1 5.0 30.0 0.5 −21.9 22.9
Wales
Cardff 70 64.3 52.5 74.6 11.4 24.3 1.3 −13.7 16.2
Swanse 53 69.8 56.3 80.6 7.6 22.6 2.5 −15.6 20.5
England 2,802 60.9 59.0 62.6 7.5 31.7 −2.3 −4.8 0.3
N Ireland 70 62.9 51.0 73.3 5.7 31.4 −7.9 −23.7 7.9
Scotland 208 54.3 47.5 61.0 4.8 40.9 1.2 −8.9 11.2
Wales 171 66.7 59.3 73.3 8.2 25.2 0.9 −8.9 10.8
UK 3,251 60.8 59.1 62.5 7.3 31.9 −2.1 −4.5 0.3
estimation of the expected total calcium were the serum adjustment equations normalised to a mean calcium of
albumin normal) from the total calcium and albumin 2.4 mmol/L. Until this process is complete, differences
concentration, but there are no data to support the use between laboratories in the reported adjusted calcium
of mathematical corrections of serum calcium amongst are likely to continue.
patients with ERF. This topic was discussed in con- Meanwhile, centres must work with their laboratories
siderable detail in the 2009 report and most of the short- to ensure that the calcium results are adjusted correctly
comings remain. However the ongoing restructuring of for the methods in use. These problems must be borne
pathology into a smaller number of services together in mind when trying to interpret the following figures
with harmonisation should increase measurement that compare serum adjusted calcium achieved in differ-
uniformity across laboratories and hence renal centres. ent renal centres. These issues raise the question as to
UK laboratories are still in the process of adopting the whether these comparisons between centres of achieve-
guidelines to harmonise albumin-adjusted calcium refer- ment of the calcium guidelines are of value, and also
ence ranges to 2.2–2.6 mmol/L using method-specific raises questions about the guidelines themselves.
235
80
N = 20,983 Upper 95% Cl
% with phos 1.1–1.7 mmol/L
70
Lower 95% Cl
60
50
40
30
20
7 Colchr
0 Ulster
0 Chelms
4 Camb
0 Bangor
0 Donc
0 L Kings
0 Swanse
3 Basldn
0 Plymth
0 Hull
0 Ipswi
0 Sheff
0 Norwch
0 Wrexm
0 Glouc
0 York
11 L Guys
7 Carsh
1 Cardff
0 Redng
4 B QEH
0 Exeter
0 Nottm
1 L West
14 Stoke
0 Antrim
1 Truro
6 Abrdn
16 L Rfree
2 Wirral
0 Covnt
4 D&Gall
1 Stevng
0 Newc
2 Liv Ain
5 Dunfn
0 Derby
1 Middlbr
4 L St.G
4 Brightn
0 Dorset
0 Oxford
6 Edinb
1 Wolve
1 Shrew
0 Clwyd
0 Bristol
1 Liv RI
2 Kent
0 West NI
12 Salford
0 Dudley
11 Klmarnk
0 Carlis
0 Leic
0 B Heart
1 Dundee
0 Ports
0 Prestn
14 Glasgw
1 Belfast
0 L Barts
0 Leeds
8 M RI
0 Newry
2 Bradfd
6 Airdrie
0 Sthend
26 Inverns
4 England
0 N Ireland
9 Scotland
0 Wales
4 UK
Centre
Fig. 12.1. Percentage of haemodialysis patients with phosphate within the range specified by the RA clinical audit measure
(1.1–1.7 mmol/L) by centre in 2012
80 To try and better understand the varation in current

75 Solid lines show 95% limits
laboratory assays utilised and practice in adjustment
70 formulae applied it is proposed to undertake a short
65 survey of all renal centres in 2013.

60
The audit measure for calcium in the current Renal
Association clinical practice guidelines does not specify
55
a lower limit for calcium and advises that adjusted
50
calcium should ideally be within the normal range as
45
per earlier guidance. Previously the UKRR used 2.2–
40 2.5 mmol/L as the audit measure for adjusted calcium
35 and in the absence of any change in guidance has
0 100 200 300 400 500 600 700 800 900 1,000 1,100 1,200 1,300 1,400
Number of patients with data in centre maintained this range in this report to allow consistency.
Fig. 12.2. Funnel plot of percentage of haemodialysis patients
The data for adjusted calcium was 97% complete for HD
with phosphate within the range specified by the RA clinical patients and 98% complete for PD patients overall,
audit measure (1.1–1.7 mmol/L) by centre in 2012 although there was between centre variation (tables 12.6,
100
N = 20,933 Upper 95% Cl
90 % with phos 1.1–1.7 mmol/L
Lower 95% Cl
80
70
60
50
40
30
20
0 Stevng
8 Covnt
28 Wirral
0 Carlis
2 L St.G
4 L Guys
2 Liv RI
0 L West
2 Swanse
0 Redng
2 B QEH
8 Dorset
0 Ports
0 Klmarnk
1 Cardff
2 Carsh
2 Leic
0 Ipswi
3 Glouc
1 Exeter
0 L Kings
0 Derby
3 Shrew
1 L Rfree
6 Plymth
4 Hull
0 Norwch
4 Basldn
2 Wolve
2 Prestn
0 Leeds
0 Sheff
1 L Barts
4 Bradfd
0 Donc
14 Newc
0 York
0 Stoke
0 Oxford
7 Salford
2 Kent
0 Bristol
0 Nottm
0 M RI
0 Glasgw
0 Chelms
0 Abrdn
6 Brightn
0 Belfast
0 Camb
0 Edinb
0 Dudley
0 B Heart
2 England
0 N Ireland
2 Scotland
2 Wales
2 UK
Centre
Fig. 12.3. Percentage of peritoneal dialysis patients with phosphate within the range specified by the RA clinical audit measure
(1.1–1.7 mmol/L) by centre in 2012
236
90 are shown in figure 12.10. The percentage of patients

Solid lines show 95% limits achieving the audit standard for calcium appears to
80
have plateaued for both HD and PD patients in recent
70 years. However, centres should be aware that achieve-

ment of the audit standard can mask population shifts
60 in concentration. This can be illustrated by data from
50
the Royal Free for HD patients: in 2011 30% had an
adjusted calcium ,2.2 mmol/L, 65% were within range,
40 and 5% were .2.5 mmol/L; in 2012 4% had an adjusted
calcium ,2.2 mmol/L, 77% were within range and 19%
30
15 35 55 75 95 115 135 155 175 were .2.5 mmol/L (date not shown). A similar pattern
Number of patients with data in centre was observed in PD patients. However, the figures for
Fig. 12.4. Funnel plot of percentage of peritoneal dialysis patients unadjusted calcium remained stable. This shift can be
with phosphate within the range specified by the RA clinical audit attributed to a change in the equation used to adjust
measure (1.1–1.7 mmol/L) by centre in 2012 calcium that was introduced on July 6th 2012 before
the UKRR collection of data in the last two quarters.
The new equation increased adjusted calcium values by
approximately 0.2 mmol/L. It has since been recognised
12.8). Seventy-seven percent (95% CI 76–77%) of HD that the new equation was over-adjusting calcium results
patients and 78% (95% CI 77–80%) of PD patients and a revised equation has been introduced from 17th
achieved adjusted calcium between 2.2–2.5 mmol/L October 2013.
(tables12.7, 12.9), not significantly different from 2011. Similar to that seen in earlier phosphate analyses, there
The proportion of HD patients with hypercalcaemia was significant between centre variation in unadjusted
was 12% and the proportion with hypocalcaemia was analyses for the proportion of patients below, within
11%. For peritoneal dialysis patients the proportion of and above the range specified by the clinical performance
patients with hypercalcaemia was 16% and the pro- measure (figures 12.6–12.10). There was greater variation
portion with hypocalcaemia was 6% (tables 12.7, 12.9, in the proportion of patients within range for adjusted
figures 12.6–12.9). The changes in the percentages calcium than phosphate, most notably for HD patients.
above, below and within range for the period 2002 to The funnel plot shows a greater number of centres out-
2012 for England, Northern Ireland and Wales combined lying the 3SD limit indicating over dispersion in the
70
60
50 % with phos >1.7

% with phos 1.1–1.7

40 % with phos <1.1mmol/L
30
20
10

0 Fig. 12.5. Longitudinal change in
percentage of patients with phosphate
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
below, within and above the 2010 RA

Year
standard by dialysis modality 2000–2012
237
Table 12.6. Summary statistics for adjusted calcium in haemodialysis patients in 2012
England
B Hearta 100.0 401 2.5 0.2 2.5 2.4 2.6
B QEH 96.8 836 2.2 0.2 2.2 2.1 2.3
Basldn 98.0 147 2.4 0.2 2.4 2.3 2.5
Bradfd 98.4 186 2.4 0.2 2.4 2.3 2.5
Brightn 67.2 227 2.3 0.2 2.3 2.2 2.4
Bristol 100.0 461 2.4 0.2 2.4 2.3 2.5
Camb 95.1 308 2.3 0.1 2.3 2.2 2.4
Carlis 100.0 57 2.3 0.2 2.3 2.2 2.4
Carsh 93.3 651 2.4 0.2 2.3 2.2 2.5
Chelms 100.0 121 2.3 0.1 2.3 2.2 2.4
Colchr 92.6 100 2.4 0.1 2.4 2.3 2.5
Covnt 100.0 335 2.3 0.2 2.3 2.2 2.4
Derby 99.5 208 2.5 0.2 2.4 2.4 2.5
Donc 100.0 158 2.4 0.1 2.3 2.3 2.5
Dorset 99.6 243 2.4 0.2 2.3 2.2 2.4
Dudley 100.0 153 2.3 0.2 2.3 2.2 2.4
Exeter 100.0 351 2.3 0.2 2.3 2.2 2.4
Glouc 100.0 193 2.4 0.1 2.4 2.3 2.5
Hull 100.0 310 2.3 0.2 2.4 2.2 2.4
Ipswi 100.0 124 2.4 0.2 2.4 2.3 2.5
Kent 98.9 357 2.4 0.2 2.4 2.3 2.5
L Barts 99.8 844 2.3 0.2 2.3 2.1 2.4
L Guys 89.0 527 2.3 0.2 2.3 2.2 2.4
L Kings 99.8 459 2.3 0.1 2.3 2.2 2.4
L Rfreea 84.4 564 2.4 0.2 2.4 2.3 2.5
L St.G 96.7 262 2.3 0.2 2.3 2.2 2.4
L Westb 91.6 1,229 2.4 0.2 2.4 2.3 2.5
Leeds 100.0 454 2.4 0.2 2.4 2.3 2.5
Leic 99.8 799 2.4 0.2 2.4 2.3 2.5
Liv Ain 98.2 163 2.4 0.2 2.3 2.3 2.5
Liv RI 99.4 343 2.4 0.2 2.3 2.3 2.5
M RI 92.2 437 2.4 0.2 2.4 2.3 2.5
Middlbr 99.4 310 2.3 0.2 2.3 2.2 2.4
Newc 100.0 262 2.3 0.2 2.3 2.2 2.4
Norwch 100.0 303 2.4 0.2 2.4 2.3 2.6
Nottm 99.7 354 2.4 0.2 2.4 2.3 2.5
Oxford 100.0 389 2.4 0.2 2.4 2.3 2.5
Plymth 100.0 119 2.4 0.2 2.4 2.3 2.5
Ports 99.8 509 2.4 0.2 2.4 2.3 2.5
Prestn 99.6 494 2.3 0.2 2.3 2.2 2.4
Redng 100.0 251 2.3 0.2 2.3 2.2 2.4
Salford 88.4 305 2.4 0.2 2.4 2.3 2.5
Sheff 99.8 561 2.3 0.2 2.3 2.2 2.4
Shrew 100.0 184 2.3 0.2 2.3 2.2 2.4
Stevng 99,0 376 2.4 0.2 2.4 2.3 2.5
Sthend 100.0 107 2.4 0.2 2.4 2.3 2.5
Stoke 85.0 250 2.4 0.2 2.4 2.3 2.5
Sund 99.5 183 2.4 0.2 2.4 2.2 2.5
Truro 99.3 133 2.4 0.2 2.3 2.2 2.4
Wirral 97.7 173 2.4 0.2 2.4 2.3 2.5
Wolve 99.6 269 2.4 0.2 2.4 2.3 2.5
York 100.0 122 2.4 0.1 2.4 2.3 2.5
N Ireland
Antrim 99.2 125 2.4 0.1 2.4 2.3 2.5
Belfast 99.0 206 2.3 0.2 2.3 2.2 2.4
Newry 100.0 85 2.4 0.2 2.4 2.3 2.4
Ulster 100.0 101 2.4 0.2 2.4 2.3 2.5
West NI 100.0 129 2.3 0.2 2.3 2.2 2.4
238

Wales
Bangor 100.0 82 2.4 0.2 2.4 2.3 2.5
Cardffb 99.3 445 2.4 0.2 2.4 2.3 2.5
Clwyd 100.0 76 2.3 0.2 2.3 2.2 2.4
Swanse 100.0 308 2.3 0.2 2.3 2.2 2.4
Wrexm 100.0 86 2.4 0.1 2.4 2.3 2.5
England 96.4 17,662 2.4 0.2 2.4 2.2 2.5
N Ireland 99.5 646 2.4 0.2 2.4 2.3 2.4
Wales 99.7 997 2.3 0.2 2.3 2.2 2.5
E, W & NI 96.7 19,305 2.4 0.2 2.4 2.2 2.5
a
London Royal Free and Birmingham Heartlands had changes in their calcium assay/albumin adjustment calculations in 2012
b
These centres supplied uncorrected calcium and were corrected using the formula: adjusted calcium = unadjusted calcium + [(40-
albumin) × 0.02]
Table 12.7. Percentage of haemodialysis patients within, below and above the range for adjusted calcium (2.2–2.5 mmol/L) in 2012
% Lower Upper % % Change in % 95% 95%

adjusted Ca 95% 95% adjusted Ca adjusted Ca within range LCL UCL
Centre N 2.2–2.5 mmol/L CI CI ,2.2 mmol/L .2.5 mmol/L from 2011 change change
England
B Heart 401 57.9 53.0 62.6 4.0 38.2 −15.8 −22.2 −9.4
B QEH 836 70.7 67.5 73.7 26.8 2.5 −2.3 −6.6 2.1
Basldn 147 82.3 75.3 87.7 6.1 11.6 3.2 −6.0 12.5
Bradfd 186 73.1 66.3 79.0 3.8 23.1 −4.3 −13.1 4.6
Brightn 227 78.4 72.6 83.3 15.4 6.2 3.9 −4.1 11.8
Bristol 461 76.6 72.5 80.2 4.8 18.7 1.9 −3.7 7.5
Camb 308 87.3 83.1 90.6 6.5 6.2 6.7 1.0 12.4
Carlis 57 79.0 66.5 87.6 15.8 5.3 −2.1 −16.7 12.5
Carsh 651 81.6 78.4 84.4 9.4 9.1 2.6 −1.7 7.0
Chelms 121 84.3 76.7 89.8 9.9 5.8 −4.4 −13.1 4.3
Colchr 100 87.0 78.9 92.3 0.0 13.0 11.0 0.3 21.7
Covnt 335 77.6 72.8 81.8 10.8 11.6 9.0 2.3 15.7
Derby 208 77.4 71.2 82.6 2.4 20.2 1.3 −7.0 9.7
Donc 158 86.7 80.5 91.2 5.1 8.2 −0.2 −7.7 7.3
Dorset 243 84.8 79.7 88.8 7.4 7.8 4.1 −2.8 10.9
Dudley 153 78.4 71.2 84.2 12.4 9.2 6.8 −3.2 16.8
Exeter 351 76.1 71.3 80.2 14.0 10.0 −6.1 −12.1 0.0
Glouc 193 86.5 81.0 90.7 5.2 8.3 0.7 −6.2 7.7
Hull 310 76.8 71.8 81.1 13.6 9.7 −4.4 −10.9 2.0
Ipswi 124 79.8 71.9 86.0 7.3 12.9 3.0 −7.3 13.3
Kent 357 70.3 65.4 74.8 5.3 24.4 −4.0 −10.6 2.6
L Barts 844 66.7 63.5 69.8 26.0 7.4 −2.1 −6.6 2.4
L Guys 527 73.8 69.9 77.4 14.8 11.4 2.5 −3.0 8.0
L Kings 459 81.9 78.1 85.2 14.8 3.3 −3.4 −8.3 1.5
L Rfree 564 77.0 73.3 80.2 3.9 19.2 11.7 6.4 16.9
L St.G 262 80.5 75.3 84.9 11.1 8.4 −2.4 −8.9 4.2
L West∗ 1,229 71.4 68.8 73.8 9.1 19.5 −3.9 −7.4 −0.4
Leeds 454 79.5 75.6 83.0 5.7 14.8 2.8 −2.6 8.1
Leic 799 79.0 76.0 81.7 9.1 11.9 −2.0 −6.0 1.9
Liv Ain 163 79.8 72.9 85.2 6.8 13.5 −2.3 −11.0 6.5
Liv RI 343 80.8 76.2 84.6 7.6 11.7 5.8 −0.3 11.9
M RI 437 74.8 70.6 78.7 5.5 19.7 0.0 −6.0 5.9
Middlbr 310 76.1 71.1 80.6 17.7 6.1 4.7 −2.4 11.8
239

Nottm 354 83.1 78.8 86.6 3.1 13.8 2.1 −3.5 7.6
Oxford 389 78.9 74.6 82.7 8.2 12.9 −1.6 −7.3 4.1
Plymth 119 87.4 80.1 92.3 4.2 8.4 12.6 2.9 22.3
Ports 509 80.0 76.3 83.2 5.5 14.5 −0.2 −5.2 4.8
Prestn 494 75.1 71.1 78.7 18.6 6.3 −2.7 −8.0 2.7
Redng 251 80.5 75.1 84.9 15.5 4.0 −2.5 −9.2 4.3
Salford 305 71.5 66.2 76.3 6.2 22.3 −3.8 −10.8 3.3
Sheff 561 77.7 74.1 81.0 13.6 8.7 1.1 −3.8 6.0
Shrew 184 71.7 64.8 77.8 22.8 5.4 −1.3 −10.5 8.0
Stevng 376 80.1 75.7 83.8 5.1 14.9 1.0 −4.8 6.7
Sthend 107 76.6 67.7 83.7 9.4 14.0 3.0 −8.4 14.3
Stoke 250 78.0 72.4 82.7 8.4 13.6 −0.1 −7.1 7.0
Sund 183 77.1 70.4 82.6 10.4 12.6 4.6 −4.7 13.8
Truro 133 73.7 65.6 80.5 12.8 13.5 −4.6 −14.7 5.6
Wirral 173 81.5 75.0 86.6 12.1 6.4 −1.2 −9.4 7.0
Wolve 269 76.6 71.2 81.3 7.4 16.0 −0.9 −7.9 6.1
York 122 91.0 84.5 94.9 2.5 6.6 7.5 −1.5 16.4
N Ireland
Antrim 125 84.0 76.5 89.4 1.6 14.4 2.0 −7.3 11.4
Belfast 206 82.0 76.2 86.7 13.1 4.9 −0.3 −7.7 7.1
Newry 85 84.7 75.4 90.9 9.4 5.9 6.7 −4.4 17.9
Ulster 101 81.2 72.4 87.7 6.9 11.9 3.0 −8.1 14.1
West NI 129 83.7 76.3 89.1 7.8 8.5 1.6 −7.5 10.7
Wales
Bangor 82 82.9 73.2 89.6 1.2 15.9 −7.7 −17.9 2.6
Cardff ∗ 445 73.3 69.0 77.2 11.0 15.7 −5.4 −11.0 0.2
Clwyd 76 73.7 62.7 82.4 21.1 5.3 7.6 −8.0 23.2
Swanse 308 75.3 70.2 79.8 18.2 6.5 0.4 −6.3 7.1
Wrexm 86 88.4 79.7 93.6 5.8 5.8 10.6 −0.7 21.9
England 17,662 76.5 75.9 77.2 10.9 12.6 0.0 −0.9 0.9
N Ireland 646 83.0 79.9 85.7 8.4 8.7 2.0 −2.1 6.2
Wales 997 76.0 73.3 78.6 12.7 11.2 −1.6 −5.3 2.1
E, W & NI 19,305 76.7 76.1 77.3 10.9 12.4 0.0 −0.8 0.9
∗
albumin) × 0.02]
Table 12.8. Summary statistics for adjusted calcium in peritoneal dialysis patients in 2012
England
B Heart 100.0 42 2.5 0.2 2.4 2.3 2.6
B QEH 98.7 147 2.3 0.1 2.3 2.2 2.4
Basldn 96.4 27 2.4 0.2 2.5 2.3 2.6
Bradfd 100.0 24 2.4 0.2 2.4 2.3 2.4
Brightn 94.2 65 2.4 0.8 2.3 2.2 2.4
Bristol 100.0 56 2.5 0.1 2.4 2.4 2.5
Camb 100.0 32 2.4 0.1 2.3 2.3 2.4
Carlis 100.0 21 2.3 0.1 2.3 2.2 2.3
Carsh 97.9 95 2.4 0.2 2.3 2.3 2.5
Chelms 100.0 25 2.4 0.1 2.4 2.3 2.5
Colchra
Covnt 95.2 80 2.3 0.2 2.3 2.2 2.4
Derby 100.0 84 2.5 0.2 2.5 2.4 2.6
Donc 100.0 23 2.4 0.2 2.4 2.2 2.5
240

Dorset 73.7 28 2.4 0.1 2.4 2.3 2.5

Dudley 100.0 53 2.4 0.2 2.4 2.3 2.5
Exeter 98.6 68 2.3 0.1 2.3 2.2 2.4
Glouc 96.8 30 2.4 0.2 2.4 2.3 2.4
Hull 96.2 76 2.5 0.1 2.4 2.4 2.5
Ipswi 100.0 30 2.4 0.1 2.4 2.3 2.5
Kent 98.2 54 2.5 0.2 2.5 2.4 2.6
L Barts 98.8 165 2.3 0.2 2.3 2.2 2.4
L Guys 96.3 26 2.4 0.1 2.35 2.3 2.5
L Kings 100.0 76 2.3 0.1 2.2 2.2 2.3
L Rfree 99.0 101 2.5 0.2 2.4 2.3 2.5
L St.G 97.9 47 2.4 0.1 2.4 2.4 2.5
L Westb 100.0 47 2.5 0.1 2.5 2.4 2.6
Leeds 100.0 77 2.4 0.2 2.4 2.3 2.5
Leic 97.9 140 2.4 0.2 2.4 2.3 2.5
Liv Ain 100.0 17
Liv RI 98.2 54 2.4 0.2 2.3 2.2 2.4
M RI 100.0 76 2.5 0.2 2.5 2.35 2.6
Middlbr 87.5 7
Newc 86.5 32 2.3 0.1 2.3 2.3 2.4
Norwch 100.0 48 2.5 0.1 2.5 2.4 2.6
Nottm 100.0 72 2.4 0.2 2.4 2.3 2.5
Oxford 100.0 69 2.4 0.2 2.4 2.3 2.5
Plymth 96.8 30 2.4 0.1 2.4 2.4 2.5
Ports 100.0 78 2.4 0.2 2.4 2.3 2.5
Prestn 98.3 58 2.4 0.2 2.4 2.3 2.5
Redng 100.0 63 2.3 0.1 2.3 2.3 2.4
Salford 93.3 84 2.5 0.2 2.45 2.4 2.6
Sheff 100.0 67 2.4 0.1 2.3 2.3 2.4
Shrew 97.0 32 2.3 0.2 2.3 2.2 2.4
Stevng 100.0 27 2.4 0.2 2.4 2.3 2.5
Sthend 100.0 14
Stoke 87.0 60 2.4 0.2 2.5 2.3 2.5
Sund 100.0 17
Truro 100.0 19
Wirral 72.4 21 2.4 0.2 2.3 2.3 2.4
Wolve 98.8 82 2.4 0.2 2.4 2.3 2.5
York 100.0 27 2.4 0.1 2.4 2.4 2.5
N Ireland
Antrim 100.0 10
Belfast 100.0 25 2.3 0.2 2.3 2.2 2.4
Newry 100.0 14
Ulster 100.0 6
West NI 100.0 15
Wales
Bangor 100.0 14
Cardffb 98.6 70 2.4 0.2 2.4 2.3 2.5
Clwyd 100.0 15
Swanse 98.2 53 2.3 0.2 2.3 2.2 2.4
Wrexm 95.0 19
England 97.5 2,793 2.4 0.2 2.4 2.3 2.5
N Ireland 100.0 70 2.4 0.2 2.4 2.3 2.5
Wales 98.3 171 2.4 0.2 2.4 2.3 2.5
E, W & NI 97.6 3,034 2.4 0.2 2.4 2.3 2.5
Blank cells denote centres excluded from the analysis due to low patient numbers
a
No PD patients
b
albumin) × 0.02]
241
Table 12.9. Percentage of peritoneal dialysis patients within, below and above the range for adjusted calcium (2.2–2.5 mmol/L) in 2012
England
B Heart 42 71.4 56.1 83.0 0.0 28.6 −7.5 −26.4 11.3
B QEH 147 79.6 72.3 85.4 15.0 5.4 1.5 −7.8 10.9
Basldn 27 66.7 47.3 81.7 3.7 29.6 0.0 −25.9 25.9
Bradfd 24 79.2 58.7 91.1 8.3 12.5 8.8 −14.9 32.5
Brightn 65 81.5 70.2 89.2 6.2 12.3 −4.4 −17.1 8.3
Bristol 56 76.8 64.0 86.0 1.8 21.4 12.4 −4.1 28.9
Camb 32 90.6 74.7 96.9 0.0 9.4 9.4 −7.5 26.3
Carlis 21 81.0 58.9 92.7 19.1 0.0 n/a n/a n/a
Carsh 95 80.0 70.8 86.9 10.5 9.5 −7.0 −17.5 3.6
Chelms 25 88.0 68.7 96.1 4.0 8.0 −7.5 −22.9 8.0
Covnt 80 81.3 71.2 88.4 10.0 8.8 7.2 −5.8 20.2
Derby 84 65.5 54.7 74.8 1.2 33.3 −13.7 −26.7 −0.7
Donc 23 82.6 61.8 93.3 4.4 13.0 −3.1 −24.6 18.4
Dorset 28 89.3 71.6 96.5 0.0 10.7 23.4 5.3 41.5
Dudley 53 81.1 68.4 89.5 1.9 17.0 −4.6 −19.0 9.8
Exeter 68 82.4 71.4 89.7 14.7 2.9 5.3 −8.6 19.2
Glouc 30 86.7 69.4 94.9 6.7 6.7 5.4 −12.8 23.6
Hull 76 76.3 65.5 84.5 0.0 23.7 1.6 −12.1 15.4
Ipswi 30 76.7 58.5 88.5 6.7 16.7 −3.3 −24.2 17.5
Kent 54 55.6 42.2 68.1 1.9 42.6 −13.3 −30.9 4.3
L Barts 165 75.8 68.6 81.7 14.6 9.7 1.9 −7.7 11.6
L Guys 26 88.5 69.7 96.2 7.7 3.9 13.5 −6.7 33.7
L Kings 76 76.3 65.5 84.5 21.1 2.6 −9.4 −22.0 3.2
L Rfree 101 73.3 63.8 81.0 2.0 24.8 −3.0 −15.7 9.7
L St.G 47 87.2 74.4 94.2 0.0 12.8 8.8 −6.0 23.6
L West∗ 47 63.8 49.3 76.2 0.0 36.2 −1.8 −23.2 19.6
Leeds 77 85.7 76.0 91.9 2.6 11.7 5.5 −6.2 17.1
Leic 140 77.9 70.2 84.0 5.0 17.1 −5.5 −14.7 3.8
Liv RI 54 79.6 66.8 88.4 7.4 13.0 0.7 −14.4 15.8
M RI 76 65.8 54.5 75.5 2.6 31.6 −4.6 −19.7 10.4
Newc 32 84.4 67.5 93.3 9.4 6.3 8.8 −9.4 27.0
Norwch 48 64.6 50.2 76.7 0.0 35.4 −16.3 −33.9 1.3
Nottm 72 81.9 71.3 89.2 2.8 15.3 13.0 −0.8 26.8
Oxford 69 78.3 67.0 86.5 4.4 17.4 −1.0 −14.1 12.1
Plymth 30 83.3 65.7 92.9 0.0 16.7 2.3 −16.1 20.6
Ports 78 78.2 67.7 86.0 1.3 20.5 −4.3 −16.7 8.1
Prestn 58 82.8 70.8 90.5 6.9 10.3 −4.3 −17.5 8.9
Redng 63 93.7 84.3 97.6 4.8 1.6 7.5 −2.5 17.5
Salford 84 64.3 53.5 73.8 0.0 35.7 −7.0 −20.7 6.7
Sheff 67 92.5 83.3 96.9 3.0 4.5 14.8 2.0 27.5
Shrew 32 71.9 54.2 84.7 18.8 9.4 10.3 −14.0 34.7
Stevng 27 85.2 66.5 94.3 0.0 14.8 −11.0 −26.3 4.3
Stoke 60 73.3 60.8 83.0 6.7 20.0 10.3 −6.0 26.5
Wirral 21 76.2 54.0 89.7 4.8 19.1 −4.6 −28.3 19.1
Wolve 82 82.9 73.2 89.6 6.1 11.0 2.3 −10.5 15.0
York 27 85.2 66.5 94.3 3.7 11.1 n/a n/a n/a
N Ireland
Belfast 25 64.0 44.0 80.1 20.0 16.0 −7.4 −32.6 17.8
Wales
Cardff ∗ 70 80.0 69.0 87.8 8.6 11.4 6.1 −6.9 19.1
Swanse 53 81.1 68.4 89.5 11.3 7.6 −0.5 −15.6 14.6
England 2,793 78.0 76.4 79.5 6.1 15.9 0.3 −1.8 2.5
N Ireland 70 71.4 59.8 80.8 8.6 20.0 −7.0 −21.6 7.5
Wales 171 81.3 74.7 86.5 8.8 9.9 4.7 −3.8 13.1
E, W & NI 3,034 78.1 76.5 79.5 6.3 15.7 0.4 −1.7 2.5
∗
albumin) × 0.02]
242
50
60
70
80
90
100
0
100
50
60
70
80
90
100
40
50
60
70
80
90
100
Chapter 12
0 Redng 0 York
0 Sheff 0 Wrexm
0 Plymth
0 Camb 5 Camb
200 300
26 Dorset 7 Colchr
4 L Guys 0 Donc
0 Glouc
400
0 Chelms
0 Dorset
2 L St.G 0 Newry
Upper 95% Cl
Lower 95% Cl
500
3 Glouc 0 Chelms
0 Leeds 1 Antrim
% with calc 2.2–2.5

0 West NI
600
0 Stevng
0 Nottm
0 York 0 Bangor
700
14 Newc 2 Basldn
3 Plymth 1 Belfast
1 Wolve 0 L Kings
7 Carsh
N = 3,034
2 Prestn 2 Wirral
Number of patients in centre

0 Donc 0 Ulster
1 Exeter 1 Liv RI
0 Nottm 3 L St.G
0 Redng
6 Brightn 1 Stevng
5 Covnt 0 Ports
0 Dudley 0 Ipswi

2 Swanse 2 Liv Ain
0 Leeds

0 Carlis 0 Leic
2 Carsh 0 Carlis
800 900 1,000 1,100 1,200 1,300

1 Cardff 0 Oxford
2 Liv RI 0 Dudley
33 Brightn
1 B QEH 15 Stoke
Centre
Centre
2012
0 Bradfd 0 Sheff
0 Oxford 0 Covnt
0 Ports 0 Derby
1 Sund
2 Leic 16 L Rfree
0 Bristol 0 Hull
0 Ipswi 0 Sthend
4 Hull 0 Wolve
0 L Kings 0 Bristol
1 Middlbr
28 Wirral 0 Exeter
1 L Barts 0 Swanse
13 Stoke 0 Newc
1 L Rfree 0 Prestn
8 M RI
3 Shrew 11 L Guys
0 B Heart 1 Truro
4 Basldn 0 Clwyd
0 M RI 1 Cardff
N = 19,305
2 Bradfd
0 Derby 0 Shrew
0 Norwch 12 Salford
7 Salford 8 L West
0 Belfast 3 B QEH
1 Kent
0 L West 0 Norwch
2 Kent 0 L Barts
Fig. 12.6. Percentage of haemodialysis patients with adjusted calcium within range (2.2–2.5 mmol/L) by centre in 2012
2 England 0 B Heart
Lower 95% Cl
4 England
Upper 95% Cl
0 N Ireland
Fig. 12.8. Percentage of peritoneal dialysis patients with adjusted calcium within range (2.2–2.5 mmol/L) by centre in 2012
0 N Ireland
2 Wales 0 Wales
% with calc 2.2–2.5
2 E, W & NI 3 E, W & NI
with adjusted calcium within range (2.2–2.5 mmol/L) by centre in
243
Management of biochemical variables
100 The data for parathyroid hormone were 83% complete

Solid lines show 95% limits for both HD and PD patients overall, although there was
90 between centre variation (tables 12.10, 12.12). Fifty-eight
percent (95% CI 57–58%) of HD patients and 65% (95%
80 CI 63–67%) of PD patients achieved a parathyroid

hormone between 16–72 pmol/L (tables 12.11, 12.13). In
70 2010, when the PTH standard target was 16–32 pmol/L,
28% (95% CI 27–29%) of HD patients and 31% (95% CI
60 29–32%) of PD patients achieved the RA standard.
In 2012, the proportion of HD patients with a
50 parathyroid hormone above the upper limit of the
range (.72 pmol/L) was 16% and the proportion with
40
10 30 50 70 90 110 130 150 170
parathyroid hormone below the lower limit of the
Number of patients in centre range was 27%. The proportion of PD patients with
Fig. 12.9. Funnel plot of percentage of peritoneal dialysis patients
parathyroid hormone above the upper limit of the
with adjusted calcium within range (2.2–2.5 mmol/L) by centre range was 10% and the proportion below the lower
in 2012 limit of the range was 25% (tables 12.11, 12.13,
figures 12.11–12.14). Again there was significant between
data, possibly due to differences in calcium adjustment centre variation in unadjusted analyses for the proportion
factors between centres. of patients below, within and above the range specified by
the clinical performance measure.
Parathyroid hormone A significant contributor to centre variation will be the
At the beginning of 2012 the following RA guideline assay used to measure PTH. This has been demonstrated
for PTH applied: by a study undertaken by the Scottish Clinical Bio-
Guideline 4.2.1 CKD-MBD: Target range of serum chemistry Managed Diagnostic Network in association
PTH in patients on dialysis with the Scottish Renal Registry. Analysis of samples
from 106 haemodialysis patients by six different PTH
‘We suggest that the target range for parathyroid immunoassays in common use showed a 1.2- to 2.7-
hormone measured using an intact PTH assay should fold variation in results in spite of similar reference
be between 2 and 9 times the upper limit of normal ranges for each method [7]. Since current guidelines
for the assay used (2C)’ [3] refer to multiples of the upper reference limit, 53% of
100
90
80
of patients
70
60
% with corrected Ca >2.5 mmol/L
50 % with corrected Ca 2.2–2.5 mmol/L
Percentage
% with corrected Ca <2.2 mmol/L

40
30
20
10
Fig. 12.10. Longitudinal change in
0
percentage of patients with adjusted
calcium ,2.2 mmol/L, 2.2–2.5 mmol/L
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
and .2.5 mmol/L by dialysis modality

Year 2002–2012
244
Table 12.10. Summary statistics for PTH in haemodialysis patients in 2012

England
B Heart 96.0 385 49.4 43.7 37 20 61
B QEH 0.2 2
Basldn 97.3 146 43.0 44.9 33 16 55
Bradfd 97.9 185 31.3 35.8 17 8 43
Brightn 80.2 271 42.0 41.5 31 12 58
Bristol 98.1 452 41.4 49.6 27 14 48
Camb 68.5 222 45.5 72.3 29 17 46
Carlis 100.0 57 26.3 29.8 18 11 30
Carsh 0.4 3
Chelms 99.2 120 41.5 32.1 33 19 54
Colchr 95.4 103 33.3 35.5 23 12 37
Covnt 98.8 331 42.2 42.6 30 16 54
Derby 98.6 206 33.1 27.9 26 16 43
Donc 99.4 157 43.6 35.4 35 21 55
Dorset 99.2 242 27.7 33.6 19 9 34
Dudley 96.1 147 50.4 53.8 31 14 62
Exeter 99.2 348 22.0 22.9 14 7 29
Glouc 99.5 192 35.9 33.1 28 15 48
Hull 98.4 305 45.3 47.0 31 16 59
Ipswi 100.0 124 36.6 31.8 30 14 46
Kent 98.3 355 44.2 37.3 38 19 57
L Barts 98.6 834 51.1 47.9 37 20 66
L Guys 77.0 456 48.3 48.0 34 16 62
L Kings 96.5 444 46.3 44.4 32.5 15 66
L Rfree 80.5 538 37.8 39.2 28 14 50
L St.G 92.3 250 57.6 53.7 42 22 74
L West 75.5 1,013 62.8 62.4 43 20 87
Leeds 98.7 448 39.9 38.5 28 14 54
Leic 98.5 789 42.4 45.4 28 11 57
Liv Ain 95.2 158 28.4 33.2 19 7 37
Liv RI 97.1 335 37.8 36.4 28 11 50
M RI 90.1 427 47.6 45.5 34 14 66
Middlbr 93.3 291 52.6 48.4 38 22 67
Newc 99.6 261 37.7 36.4 28 14 50
Norwch 94.4 286 36.9 33.4 29 15 46
Nottm 99.4 353 45.9 49.9 31 17 56
Oxford 98.5 383 51.5 42.2 41 19 70
Plymth 96.6 115 28.5 29.9 19 9 39
Ports 94.7 483 42.2 52.1 25 10 51
Prestn 1.4 7
Redng 100.0 251 37.6 37.6 30 16 47
Salford 84.9 293 34.7 32.1 25 12 46
Sheff 96.8 544 42.2 42.0 31 17 54
Shrew 99.5 183 37.1 40.0 19 10 48
Stevng 98.2 373 45.4 42.5 38 19 57
Sthend 90.7 97 55.6 59.2 37 20 57
Stoke 87.8 258 51.1 43.2 39.5 23 64
Sund 97.8 180 46.8 50.9 30 14 60
Truro 97.8 131 25.7 37.1 16 6 31
Wirral 96.6 171 38.3 35.5 31 15 48
Wolve 97.0 262 32.5 40.0 21 10 40
York 96.7 118 26.2 29.2 18.5 7 36
N Ireland
Antrim 100.0 126 33.2 31.5 23 15 42
Belfast 97.1 202 37.1 43.2 23.5 13 48
Newry 100.0 85 25.3 27.3 16 9 30
Ulster 100.0 101 22.5 23.3 16 9 28
West NI 100.0 129 36.0 29.4 29 15 47
245

Wales
Bangor 98.8 81 25.2 24.5 19 10 31
Cardff 96.7 433 37.3 30.6 29 19 47
Clwyd 100.0 76 32.4 31.6 24.5 14 41
Swanse 72.7 224 40.1 37.0 30.5 16 52
Wrexm 96.5 83 18.4 15.6 19 4 29
England 82.3 15,085 43.4 45.4 30 15 56
N Ireland 99.1 643 32.3 34.1 22 13 41
Wales 89.7 897 34.7 31.6 27 15 44
E, W & NI 83.2 16,625 42.5 44.4 29 14 54
Table 12.11. Percentage of haemodialysis patients within, below and above the range for PTH (16–72 pmol/L) in 2012
PTH 95% 95% PTH PTH within range LCL UCL
Centre N 16–72 pmol/L CI CI ,16 pmol/L .72 pmol/L from 2011 change change
England
B Heart 385 63.9 59.0 68.5 16.6 19.5 10.4 3.5 17.3
Basldn 146 62.3 54.2 69.8 23.3 14.4 −0.4 −11.7 11.0
Bradfd 185 43.8 36.8 51.0 46.0 10.3 −6.5 −16.8 3.8
Brightn 271 53.5 47.6 59.4 29.9 16.6 6.4 −1.8 14.7
Bristol 452 57.5 52.9 62.0 28.1 14.4 1.3 −5.2 7.8
Camb 222 66.7 60.2 72.6 23.9 9.5 1.5 −7.3 10.3
Carlis 57 57.9 44.8 69.9 36.8 5.3 14.8 −3.3 32.9
Chelms 120 68.3 59.5 76.0 18.3 13.3 12.2 −0.1 24.5
Colchr 103 50.5 40.9 60.0 37.9 11.7 −7.7 −21.4 6.1
Covnt 331 61.6 56.3 66.7 24.2 14.2 5.5 −2.0 13.0
Derby 206 68.9 62.3 74.9 24.8 6.3 1.2 −8.0 10.4
Donc 157 71.3 63.8 77.9 15.3 13.4 6.9 −3.5 17.3
Dorset 242 52.1 45.8 58.3 41.3 6.6 2.3 −6.8 11.4
Dudley 147 51.0 43.0 59.0 27.2 21.8 11.3 −0.4 23.1
Exeter 348 41.7 36.6 46.9 53.5 4.9 −1.8 −9.3 5.7
Glouc 192 64.1 57.0 70.5 25.5 10.4 6.7 −3.2 16.5
Hull 305 57.7 52.1 63.1 24.3 18.0 5.2 −2.8 13.1
Ipswi 124 59.7 50.8 67.9 29.0 11.3 −3.4 −15.6 8.7
Kent 355 67.3 62.3 72.0 15.5 17.2 −2.6 −9.4 4.3
L Barts 834 59.4 56.0 62.6 19.2 21.5 −3.7 −8.4 1.1
L Guys 456 55.3 50.7 59.8 24.6 20.2 5.4 −1.1 11.9
L Kings 444 53.6 49.0 58.2 25.5 21.0 3.8 −2.8 10.5
L Rfree 538 59.7 55.5 63.7 28.6 11.7 0.3 −5.6 6.3
L St.G 250 56.0 49.8 62.0 18.4 25.6 0.1 −8.6 8.7
L West 1,013 50.5 47.5 53.6 19.2 30.3 0.5 −3.9 4.9
Leeds 448 55.6 50.9 60.1 28.4 16.1 −0.8 −7.3 5.6
Leic 789 50.2 46.7 53.7 31.7 18.1 −1.3 −6.2 3.7
Liv Ain 158 50.0 42.3 57.7 43.7 6.3 −2.8 −15.0 9.4
Liv RI 335 55.5 50.2 60.8 32.2 12.2 2.9 −4.5 10.3
M RI 427 51.8 47.0 56.5 26.2 22.0 −7.1 −14.0 −0.3
Middlbr 291 62.2 56.5 67.6 16.5 21.3 2.8 −5.3 10.9
Newc 261 60.5 54.5 66.3 27.2 12.3 0.5 −8.0 9.1
Norwch 286 61.5 55.8 67.0 26.2 12.2 −0.3 −8.4 7.8
Nottm 353 60.1 54.9 65.0 23.0 17.0 6.0 −1.1 13.2
Oxford 383 58.2 53.2 63.1 18.0 23.8 −2.9 −9.9 4.1
Plymth 115 52.2 43.1 61.1 40.0 7.8 6.8 −6.0 19.6
246

Ports 483 47.0 42.6 51.5 36.2 16.8 3.1 −3.4 9.5
Redng 251 65.7 59.7 71.4 24.7 9.6 −3.1 −11.4 5.2
Salford 293 55.3 49.6 60.9 33.5 11.3 8.9 0.7 17.1
Sheff 544 63.6 59.5 67.5 22.1 14.3 3.6 −2.2 9.3
Shrew 183 49.7 42.5 56.9 35.0 15.3 −8.6 −18.8 1.7
Stevng 373 66.0 61.0 70.6 15.6 18.5 −0.6 −7.4 6.1
Sthend 97 66.0 56.0 74.7 15.5 18.6 5.0 −8.2 18.3
Stoke 258 66.7 60.7 72.2 14.3 19.0 3.8 −4.2 11.8
Sund 180 54.4 47.1 61.6 27.8 17.8 −1.8 −12.4 8.8
Truro 131 45.0 36.7 53.6 48.1 6.9 −1.0 −12.9 11.0
Wirral 171 64.9 57.5 71.7 25.2 9.9 −3.5 −14.5 7.6
Wolve 262 52.7 46.6 58.7 38.9 8.4 9.3 0.9 17.6
York 118 48.3 39.4 57.3 45.8 5.9 4.1 −8.8 16.9
N Ireland
Antrim 126 67.5 58.8 75.1 25.4 7.1 1.1 −10.6 12.8
Belfast 202 57.4 50.5 64.1 32.2 10.4 −4.6 −14.2 4.9
Newry 85 45.9 35.6 56.5 48.2 5.9 −8.1 −22.5 6.3
Ulster 101 47.5 38.0 57.2 48.5 4.0 8.5 −5.1 22.2
West NI 129 66.7 58.1 74.3 25.6 7.8 −10.2 −21.0 0.6
Wales
Bangor 81 56.8 45.9 67.1 40.7 2.5 −1.5 −16.6 13.5
Cardff 433 71.1 66.7 75.2 18.2 10.6 5.7 −0.5 11.9
Clwyd 76 61.8 50.5 72.0 29.0 9.2 17.0 0.2 33.8
Swanse 224 63.0 56.4 69.0 24.1 13.0 2.5 −6.4 11.4
Wrexm 83 51.8 41.1 62.3 48.2 0.0 3.1 −12.3 18.4
England 15,085 57.0 56.2 57.8 26.5 16.5 1.9 0.7 3.0
N Ireland 643 58.2 54.3 61.9 34.2 7.6 −3.0 −8.3 2.3
Wales 897 65.2 62.0 68.3 25.4 9.4 4.6 0.1 9.1
E, W & NI 16,625 57.5 56.7 58.2 26.8 15.8 1.8 0.8 2.9
Table 12.12. Summary statistics for PTH in peritoneal dialysis patients in 2012

England
B Heart 76.2 32 52.6 31.5 44.5 33.0 65.5
B QEH 0.0 0
Basldn 96.4 27 35.4 25.3 29.0 19.0 49.0
Bradfd 91.7 22 45.6 51.1 27.0 14.0 57.0
Brightn 88.4 61 32.2 30.9 23.0 14.0 37.0
Bristol 94.6 53 34.9 33.3 25.0 14.0 44.0
Camb 100.0 32 32.2 27.6 29.5 14.5 38.5
Carlis 95.2 20 30.6 24.0 26.5 13.5 38.0
Carsh 0.0 0
Chelms 100.0 25 36.2 16.3 37.0 24.0 51.0
Colchr∗
Covnt 92.9 78 28.3 28.3 19.5 12.0 34.0
Derby 98.8 83 28.2 23.6 25.0 15.0 33.0
Donc 100.0 23 42.3 36.4 32.0 19.0 65.0
Dorset 73.7 28 28.2 20.0 26.0 16.5 38.0
Dudley 86.8 46 20.9 16.9 17.5 9.0 28.0
247

Exeter 98.6 68 23.8 23.9 17.0 10.0 29.0

Glouc 80.7 25 23.1 18.5 19.0 8.0 34.0
Hull 88.6 70 25.7 27.5 18.0 10.0 32.0
Ipswi 96.7 29 53.5 46.4 37.0 23.0 78.0
Kent 90.9 50 35.7 27.3 29.0 19.0 48.0
L Barts 88.6 148 34.5 26.1 27.0 14.0 46.0
L Guys 96.3 26 37.5 19.2 39.5 25.0 49.0
L Kings 98.7 75 45.4 37.5 37.0 17.0 70.0
L Rfree 69.6 71 42.1 47.0 33.0 18.0 46.0
L St.G 89.6 43 37.4 29.7 30.0 22.0 46.0
L West 97.9 46 37.0 31.1 29.0 17.0 44.0
Leeds 100.0 77 46.5 34.8 37.0 26.0 57.0
Leic 95.8 137 37.3 35.6 26.0 12.0 52.0
Liv Ain 94.1 16
Liv RI 96.4 53 28.0 22.2 20.0 13.0 39.0
M RI 97.4 74 43.5 33.8 35.5 21.0 62.0
Middlbr 75.0 6
Newc 86.5 32 33.9 22.8 27.5 22.0 45.0
Norwch 89.6 43 28.5 24.3 21.0 13.0 40.0
Nottm 98.6 71 50.9 45.5 38.0 21.0 66.0
Oxford 95.7 66 51.0 36.5 40.0 25.0 66.0
Plymth 90.3 28 28.7 40.1 12.5 9.5 29.5
Ports 98.7 77 41.1 35.3 31.0 18.0 53.0
Prestn 15.3 9
Redng 96.8 61 33.9 42.7 24.0 15.0 37.0
Salford 93.3 84 37.0 35.3 26.5 15.5 45.0
Sheff 79.1 53 33.2 22.1 31.0 19.0 42.0
Shrew 93.9 31 37.2 37.4 29.0 19.0 48.0
Stevng 92.6 25 31.9 29.7 29.0 10.0 38.0
Sthend 92.9 13
Stoke 94.2 65 58.7 52.2 40.0 26.0 68.0
Sund 100.0 17
Truro 94.7 18
Wirral 62.1 18
Wolve 94.0 78 28.6 18.4 27.0 15.0 40.0
York 92.6 25 29.9 23.2 22.0 13.0 49.0
N Ireland
Antrim 100.0 10
Belfast 92.0 23 27.4 18.3 21.0 15.0 38.0
Newry 100.0 14
Ulster 100.0 6
West NI 100.0 15
Wales
Bangor 100.0 14
Cardff 98.6 70 43.8 31.9 37.0 20 61
Clwyd 73.3 11
Swanse 92.6 50 29.3 20.9 26.0 16 36
Wrexm 95.0 19
England 82.3 2,358 36.3 33.2 27.0 15 47
N Ireland 97.1 68 26.0 17.9 21.5 12.5 37.5
Wales 94.3 164 34.5 26.2 29.5 17 46
E, W & NI 83.3 2,590 35.9 32.5 27.0 15 47
Blank cells denote centres excluded from analyses due to small numbers or poor data completeness
∗
No PD patients
248
Table 12.13. Percentage of peritoneal dialysis patients within, below and above the range for PTH (16–72 pmol/L) in 2012
England
B Heart 32 75.0 57.4 87.0 3.1 21.9 −0.7 −21.1 19.7
Basldn 27 81.5 62.5 92.1 14.8 3.7 27.3 2.6 52.1
Bradfd 22 54.6 34.1 73.5 27.3 18.2 10.1 −17.9 38.1
Brightn 61 63.9 51.3 74.9 27.9 8.2 2.0 −15.0 19.0
Bristol 53 60.4 46.8 72.5 30.2 9.4 4.2 −14.2 22.7
Camb 32 65.6 47.9 79.8 25.0 9.4 −6.3 −28.9 16.4
Carlis 20 60.0 38.0 78.6 35.0 5.0 n/a n/a n/a
Chelms 25 88.0 68.7 96.1 12.0 0.0 28.0 3.0 53.0
Covnt 78 60.3 49.1 70.5 32.1 7.7 14.3 −1.4 30.0
Derby 83 73.5 63.0 81.9 25.3 1.2 −1.3 −14.2 11.7
Donc 23 65.2 44.3 81.6 21.7 13.0 −4.8 −32.7 23.2
Dorset 28 67.9 48.9 82.4 25.0 7.1 4.0 −19.4 27.3
Dudley 46 52.2 38.0 66.1 45.7 2.2 −6.9 −27.4 13.6
Exeter 68 50.0 38.3 61.7 47.1 2.9 0.9 −16.6 18.3
Glouc 25 56.0 36.6 73.7 40.0 4.0 7.6 −18.6 33.8
Hull 70 54.3 42.6 65.5 38.6 7.1 3.6 −13.0 20.2
Ipswi 29 58.6 40.4 74.8 13.8 27.6 −8.1 −32.7 16.6
Kent 50 64.0 50.0 76.0 24.0 12.0 −13.1 −30.0 3.9
L Barts 148 63.5 55.5 70.9 26.4 10.1 −4.5 −15.3 6.3
L Guys 26 80.8 61.3 91.8 15.4 3.9 14.1 −9.3 37.5
L Kings 75 58.7 47.3 69.2 21.3 20.0 3.5 −12.9 19.8
L Rfree 71 67.6 55.9 77.4 22.5 9.9 10.1 −5.2 25.5
L St.G 43 69.8 54.6 81.6 18.6 11.6 5.9 −13.5 25.4
L West 46 69.6 55.0 81.1 19.6 10.9 0.8 −20.0 21.7
Leeds 77 66.2 55.0 75.9 14.3 19.5 −7.8 −22.1 6.4
Leic 137 59.9 51.4 67.7 27.0 13.1 4.2 −7.6 16.0
Liv RI 53 64.2 50.5 75.8 32.1 3.8 −9.5 −26.8 7.7
M RI 74 73.0 61.8 81.9 13.5 13.5 3.4 −11.4 18.2
Newc 32 75.0 57.4 87.0 18.8 6.3 8.3 −12.7 29.4
Norwch 43 51.2 36.6 65.6 41.9 7.0 5.0 −16.6 26.6
Nottm 71 63.4 51.6 73.7 14.1 22.5 2.3 −13.6 18.2
Oxford 66 65.2 53.0 75.6 10.6 24.2 4.4 −11.4 20.1
Plymth 28 39.3 23.3 58.0 53.6 7.1 −16.6 −41.2 8.0
Ports 77 67.5 56.4 77.0 22.1 10.4 8.4 −7.1 23.9
Redng 61 68.9 56.3 79.2 26.2 4.9 1.2 −14.7 17.2
Salford 84 61.9 51.1 71.6 25.0 13.1 10.8 −3.9 25.4
Sheff 53 73.6 60.2 83.7 18.9 7.6 2.5 −15.3 20.3
Shrew 31 67.7 49.7 81.7 19.4 12.9 −5.0 −29.8 19.9
Stevng 25 52.0 33.1 70.4 36.0 12.0 −29.8 −55.2 −4.5
Stoke 65 67.7 55.5 77.9 10.8 21.5 9.9 −6.7 26.5
Wolve 78 71.8 60.9 80.7 25.6 2.6 2.4 −12.8 17.7
York 25 64.0 44.0 80.1 28.0 8.0 −6.0 −33.5 21.5
N Ireland
Belfast 23 65.2 44.3 81.6 30.4 4.4 15.2 −11.6 42.1
Wales
Cardff 70 65.7 53.9 75.8 17.1 17.1 3.9 −11.1 18.9
Swanse 50 72.0 58.1 82.7 24.0 4.0 16.7 −2.2 35.6
England 2,358 64.4 62.5 66.3 25.1 10.5 2.3 −0.4 5.0
N Ireland 68 64.7 52.7 75.1 33.8 1.5 4.7 −11.7 21.2
Wales 164 68.3 60.8 75.0 23.2 8.5 8.0 −2.2 18.1
E, W & NI 2,590 64.7 62.8 66.5 25.2 10.1 2.7 0.1 5.3
249
250
30
40
50
60
70
80
0
30
40
50
60
70
80
20
30
40
50
60
70
80
90
100
100
0 Chelms 1 Donc
4 Basldn 3 Cardff
1 Derby
200
4 L Guys 1 Chelms
24 B Heart 0 Antrim
14 Newc 2 Kent
0 West NI
21 Sheff 31 Camb
Upper 95% Cl
Lower 95% Cl
1 Derby 12 Stoke
3 M RI 9 Sthend
% with iPTH 16–72

7 Swanse 2 Stevng
0 Redng
6 Wolve 3 Wirral
10 L St.G 1 Glouc
2 L West 4 B Heart
Solid
3 Sheff
3 Redng 27 Swanse
Solid
26 Dorset 3 Basldn
lines
N = 16,625
Number of patients in centre

6 Shrew 7 Middlbr
lines
0 Clwyd
300 400 500 600 700 800

6 Stoke
show
1 Covnt
30 L Rfree 6 Norwch
show
1 Ports 0 Newc
0 Leeds 1 Nottm
95%95%
0 Ipswi
limits
1 Cardff 19 L Rfree
0 Camb 1 L Barts
limits
8 Belfast 2 Oxford
900 1,000 1,100

0 Donc 0 Carlis
2 Hull
4 Oxford 2 Bristol
4 Liv RI 3 Belfast
Centre
Centre
7 York 1 Bangor
8 L St.G
9 Kent 1 Leeds
12 Brightn 3 Liv RI
11 L Barts 15 Salford
1 Nottm 23 L Guys
2 Sund
7 Salford 3 L Kings
5 Bristol 20 Brightn
7 Covnt 3 Wolve
3 Plymth
5 Carlis 1 Dorset
4 Leic 3 Wrexm
1 L Kings 10 M RI
3 Ipswi 4 Dudley
25 L West
19 Glouc 5 Colchr
8 Bradfd 2 Leic
5 Liv Ain
N = 2,590
11 Hull
1 Shrew
Fig. 12.11. Percentage of haemodialysis patients with PTH within range (16–72 pmol/L) by centre in 2012
13 Dudley 3 York
7 Stevng 0 Ulster
Fig. 12.13. Percentage of peritoneal dialysis patients with PTH within range (16–72 pmol/L) by centre in 2012
10 Norwch 5 Ports
1 Exeter 0 Newry
2 Truro
10 Plymth 2 Bradfd
18 England 1 Exeter
18 England
Lower 95% Cl
Upper 95% Cl
3 N Ireland
1 N Ireland
6 Wales 10 Wales
% with iPTH 16–72

with PTH within range (16–72 pmol/L) by centre in 2012
17 E, W & NI 17 E, W & NI
100 consider two principal factors: adoption of a common

Dotted
Dotted lines
lines show
show 99.9%
99.9%limits
limits
Solid reference preparation for standardisation, such as the
90 Solid lines
lines show
show 95%
95%limits
limits
WHO international standard 95/646, and selection of
pairs of antibodies that do not detect PTH fragments
80
such as 7–84 that accumulate in CKD. Meanwhile
70
Almond et al. [7] and a recent editorial review [10]
urge adoption of assay-specific action limits for PTH in
60 CKD patients. However this approach raises a number
of difficult governance issues. There is already evidence
50 that the manufacturers of the major diagnostic platforms
used throughout the world have started to respond. The
40 Roche assay used by Almond et al. [7] was PTH (intact)
that was not standardised and cross-reacted with PTH 7–
30
84. Roche have recently launched the more expensive
20
PTH (1–84) that is standardised against the WHO inter-
10 30 50 70 90 110 130 150 170 national standard 95/646 and has 40.1% cross-reactivity
Number of patients in centre with both PTH (1–34) and PTH (7–84) (information
Fig. 12.14. Funnel plot of percentage of peritoneal dialysis supplied by Roche Diagnostics).
patients with PTH within range (16–72 pmol/L) by centre in 2012
Simultaneous control of corrected calcium, phosphate and
patients were classified differently by different methods PTH in preventing severe hyperparathyroidism
with implications for treatment e.g. with Cinacalcet. In Data points to perform the bone mineral disease (BMD)
an excellent accompanying editorial, Garrett and Gold- combination analyses were available from 58 HD and 45
smith [8] also highlighted the high biological variability PD centres, covering 16,300 HD and 2,377 PD patients,
of PTH and its poor ability to predict skeletal or patient from England, Wales and Northern Ireland. The ranges
outcomes. Whether more accurate and specific assays used for this audit were adjusted calcium 2.2–2.5 mmol/
would improve this or whether PTH will be supplanted L, phosphate 41.7 mmol/L, and PTH 472 pmol/L.
by other markers such as bone specific alkaline phospha- Tables 12.14 and 12.15 identify each centre and detail
tase that also have greater pre-analytical stability remains the numbers of patients who had received HD and PD
to be determined [9]. and the results of the BMD combination analyses.
Improvement of PTH assays to achieve consensus Figures 12.15 and 12.16, demonstrate the caterpillar
results within CKD patients requires manufacturers to plots of all centres and the percentage achievement of
Table 12.14. Percentage of haemodialysis patients within the ranges specified for the simultaneous combinations of control of bone and
mineral disorder parameters in preventing severe hyperparathyroidism in 2012
Number of parameters
Centre N None One Two Three
England
B Heart 385 5.2 23.1 36.1 35.6
Basldn 145 0.7 7.6 35.9 55.9
Bradfd 185 0.5 15.7 28.1 55.7
Brightn 206 2.4 15.5 33.5 48.5
Bristol 452 1.5 16.6 36.9 44.9
Camb 214 0.9 8.4 29.0 61.7
Carlis 57 1.8 10.5 38.6 49.1
Chelms 120 0.8 12.5 27.5 59.2
Colchr 98 0.0 7.1 36.7 56.1
Covnt 331 2.1 15.7 36.9 45.3
Derby 206 1.0 12.1 36.9 50.0
Donc 157 1.3 5.7 34.4 58.6
251
Dorset 241 0.4 10.4 30.3 58.9

Dudley 147 3.4 17.0 40.1 39.5
Exeter 348 1.7 6.9 38.8 52.6
Glouc 192 0.5 7.3 34.4 57.8
Hull 305 3.6 11.1 35.4 49.8
Ipswi 124 0.0 8.9 37.9 53.2
Kent 351 2.6 18.8 39.9 38.7
L Barts 834 4.1 19.3 41.1 35.5
L Guys 452 3.5 13.1 36.3 47.1
L Kings 444 1.8 11.5 33.1 53.6
L Rfree 531 2.4 10.9 34.5 52.2
L St.G 249 2.0 13.3 38.2 46.6
L West 1,007 3.1 17.5 44.1 35.4
Leeds 448 2.2 15.0 33.5 49.3
Leic 789 2.3 16.9 37.3 43.6
Liv Ain 158 1.9 7.0 31.0 60.1
Liv RI 335 1.8 8.7 36.1 53.4
M RI 426 2.6 17.1 39.0 41.3
Middlbr 291 2.1 17.9 38.5 41.6
Newc 261 2.3 12.6 37.9 47.1
Norwch 286 2.8 13.3 41.6 42.3
Nottm 353 2.0 13.0 28.9 56.1
Oxford 383 3.1 17.0 36.0 43.9
Plymth 115 1.7 7.0 27.8 63.5
Ports 483 2.5 14.1 40.0 43.5
Redng 251 2.4 7.2 35.1 55.4
Salford 292 2.1 12.3 34.6 51.0
Sheff 544 1.3 14.0 36.6 48.2
Shrew 182 2.2 15.4 35.7 46.7
Stevng 370 1.9 16.2 37.6 44.3
Sthend 97 3.1 18.6 38.1 40.2
Stoke 236 1.3 12.7 41.1 44.9
Truro 131 0.0 11.5 35.1 53.4
Wirral 171 0.6 13.5 29.2 56.7
Wolve 261 1.1 8.8 37.2 52.9
York 118 0.0 7.6 24.6 67.8
N Ireland
Antrim 125 0.8 7.2 27.2 64.8
Belfast 202 1.5 12.4 31.2 55.0
Newry 85 2.4 7.1 38.8 51.8
Ulster 101 0.0 6.9 32.7 60.4
West NI 129 1.6 12.4 33.3 52.7
Wales
Bangor 81 0.0 8.6 33.3 58.0
Cardff 432 1.9 12.7 37.0 48.4
Clwyd 76 1.3 11.8 42.1 44.7
Swanse 224 1.8 12.1 35.3 50.9
Wrexm 83 0.0 1.2 25.3 73.5
England 14,762 2.2 14.0 36.7 47.1
N Ireland 642 1.2 9.8 32.1 56.9
Wales 896 1.5 11.0 35.6 51.9
E, W & NI 16,300 2.2 13.7 36.4 47.8
Target range: adjusted calcium 2.2–2.5 mmol/L; phosphate 4 1.7 mmol/L; PTH 4 72 pmol/L
252
Table 12.15. Percentage of peritoneal dialysis patients within the ranges specified for the simultaneous combinations of control of bone
and mineral disorder parameters in preventing severe hyperparathyroidism in 2012
England
B Heart 32 3.1 25.0 40.6 31.3
Basldn 27 0.0 14.8 37.0 48.1
Bradfd 21 4.8 14.3 38.1 42.9
Brightn 61 0.0 9.8 44.3 45.9
Bristol 53 0.0 13.2 49.1 37.7
Camb 32 0.0 3.1 40.6 56.3
Carlis 20 0.0 20.0 10.0 70.0
Chelms 25 0.0 8.0 28.0 64.0
Covnt 76 1.3 2.6 28.9 67.1
Derby 83 1.2 7.2 37.3 54.2
Donc 23 0.0 21.7 21.7 56.5
Dorset 22 0.0 9.1 13.6 77.3
Dudley 46 0.0 6.5 60.9 32.6
Exeter 68 0.0 4.4 38.2 57.4
Glouc 25 0.0 8.0 32.0 60.0
Hull 69 1.4 10.1 42.0 46.4
Ipswi 29 6.9 10.3 37.9 44.8
Kent 48 2.1 18.8 47.9 31.3
L Barts 147 0.7 13.6 34.7 51.0
L Guys 25 0.0 4.0 28.0 68.0
L Kings 75 1.3 17.3 33.3 48.0
L Rfree 71 1.4 15.5 35.2 47.9
L St.G 43 0.0 4.7 37.2 58.1
L West 46 0.0 13.0 43.5 43.5
Leeds 77 2.6 14.3 35.1 48.1
Leic 136 2.9 11.0 32.4 53.7
Liv RI 53 0.0 7.5 32.1 60.4
M RI 74 4.1 20.3 35.1 40.5
Newc 32 0.0 12.5 40.6 46.9
Norwch 43 2.3 9.3 44.2 44.2
Nottm 71 2.8 15.5 39.4 42.3
Oxford 66 4.5 15.2 42.4 37.9
Plymth 27 0.0 7.4 44.4 48.1
Ports 77 1.3 13.0 35.1 50.6
Redng 61 0.0 6.6 23.0 70.5
Salford 83 1.2 24.1 34.9 39.8
Sheff 53 0.0 9.4 37.7 52.8
Shrew 31 3.2 9.7 45.2 41.9
Stevng 25 0.0 8.0 24.0 68.0
Stoke 57 5.3 22.8 28.1 43.9
Wolve 77 0.0 5.2 41.6 53.2
York 25 0.0 16.0 36.0 48.0
N Ireland
Belfast 23 0.0 17.4 43.5 39.1
Wales
Cardff 69 2.9 8.7 33.3 55.1
Swanse 50 0.0 6.0 34.0 60.0
England 2,235 1.4 12.1 36.6 49.9
N Ireland 23 0.0 17.4 43.5 39.1
Wales 119 1.7 7.6 33.6 57.1
E, W & NI 2,377 1.4 11.9 36.5 50.1
Target range: adjusted calcium 2.2–2.5 mmol/L; phosphate 4 1.7 mmol/L; PTH 4 72 pmol/L
253
100
N = 15,564 Upper 95% Cl
90 % with phos calc and PTH all in range
80 Lower 95% Cl
70
60
50
40
30
20
10
0
3 Wrexm
3 York
1 Antrim
3 Plymth
34 Camb
0 Ulster
5 Liv Ain
1 Chelms
1 Dorset
1 Donc
1 Bangor
1 Glouc
3 Wirral
9 Colchr
1 Nottm
3 Basldn
2 Bradfd
0 Redng
3 Belfast
3 L Kings
2 Truro
3 Liv RI
0 Ipswi
3 Wolve
0 West NI
1 Exeter
21 L Rfree
0 Newry
15 Salford
27 Swanse
1 Derby
2 Hull
1 Leeds
0 Carlis
39 Brightn
4 Cardff
3 Sheff
0 Newc
24 L Guys
1 Shrew
8 L St.G
1 Covnt
20 Stoke
2 Bristol
0 Clwyd
3 Stevng
2 Oxford
2 Leic
5 Ports
6 Norwch
7 Middlbr
10 M RI
9 Sthend
4 Dudley
3 Kent
4 B Heart
1 L Barts
25 L West
19 England
1 N Ireland
10 Wales
18 E, W & NI
Centre
Fig. 12.15. Percentage of HD patients achieving simultaneous control of all three BMD parameters in preventing severe hyper-
parathyroidism by centre in 2012
simultaneous control of all three BMD parameters for Figures 12.17 and 12.18 are funnel plots of all centres
HD and PD patients. who contributed data to these analyses based on the size
Control of none of the parameters of BMD was found of the centre and the percentage of patients achieving the
in 2.2% of HD patients and 1.4% of PD patients; of one control of all three BMD parameters. In HD patients,
parameter in 13.7% of HD and 11.9% of PD patients; there was a negative trend observed between centre size
of two parameters in 36.4% of HD and 36.5% of PD and the simultaneous control of all three BMD param-
patients; and of all three parameters in 47.8% of HD eters as identified in this analysis. No such trend was
and 50.1% of PD patients (tables 12.14, 12.15). observed in PD patients, perhaps because PD centres
The details of single parameters alone and combi- are all of a small size.
nations of adjusted calcium, phosphate and PTH are
detailed in table 12.16 (aggregate information has been Mineral and bone variables
presented as a percentage measure for all centres with There are convincing observational data that hyper-
valid data). phosphataemia is associated with increased mortality in
100
N = 2,297 Upper 95% Cl
90 % with phos calc and PTH all in range
80 Lower 95% Cl
70
60
50
40
30
20
10
0
42 Dorset
3 Redng
5 Carlis
7 Stevng
7 L Guys
10 Covnt
0 Chelms
4 Liv RI
19 Glouc
7 Swanse
10 L St.G
1 Exeter
0 Donc
0 Camb
3 Cardff
1 Derby
5 Leic
7 Wolve
21 Sheff
12 L Barts
1 Ports
4 Basldn
13 Plymth
0 Leeds
7 York
1 L Kings
30 L Rfree
14 Newc
13 Hull
12 Brightn
3 Ipswi
10 Norwch
17 Stoke
2 L West
13 Bradfd
1 Nottm
6 Shrew
3 M RI
8 Salford
8 Belfast
4 Oxford
5 Bristol
13 Dudley
24 B Heart
13 Kent
19 England
3 N Ireland
6 Wales
18 E, W & NI
Centre
Fig. 12.16. Percentage of PD patients achieving simultaneous control of all three BMD parameters in preventing severe hyper-
parathyroidism by centre in 2012
254
Table 12.16. Average control of BMD parameters in preventing severe hyperparathyroidism across renal centres in 2012
HD PD
BMD combination of parameters Avg (%) Min (%) Max (%) Avg (%) Min (%) Max (%)
None 1.8 0.0 5.2 1.3 0.0 6.9

One 12.2 1.2 23.1 11.9 2.6 25.0
Two 35.2 24.6 44.1 36.1 10.0 60.9
Three 50.8 35.4 73.5 50.8 31.3 77.3
Adj.Ca alone 4.2 0.0 9.1 4.0 0.0 13.0
Phosphate alone 2.1 0.0 7.3 1.8 0.0 6.5
PTH alone 5.8 1.0 13.0 6.2 0.0 15.0
Adj.Ca and phosphate 5.1 0.0 17.3 3.3 0.0 10.3
Adj.Ca and PTH 18.4 10.3 29.4 20.4 5.0 47.8
Phosphate and PTH 11.8 6.1 17.4 12.4 0.0 30.4
Adj.Ca = adjusted calcium
Avg = average
dialysis patients but the data linking calcium and bone and mineral diseases may become available to aid
parathyroid hormone to patient survival are less clear in better analyses of these parameters.
[11–15]. A recent cohort study has demonstrated that Finally, it is important to consider data quality and the
simultaneous achievement of all three audit measures potential for measurement bias particularly in light of the
does appear to be associated with better outcomes [16]. variability in assay methods for parathyroid hormone as
The UKRR has consistently demonstrated between discussed above. However, detecting these centre level
centre variation in achievement of audit measures for differences is an important step in understanding the
bone and mineral parameters but little is understood factors associated with exceptional performance.
about the causes of this ‘centre effect’. The complexity
of the clinical processes required to manage mineral Bicarbonate
and bone disorders is probably further confounded by In 2012 the following Renal Association clinical
case-mix. In the future, with centres moving to newer practice guidelines regarding bicarbonate management
IT systems, medications used in the management of were applicable:
80 80
Dotted lines show 99.9% limits Dotted lines show 99.9% limits
Solid lines show 95% limits Solid lines show 95% limits
70 70
60 60
50 50
40 40
30 30
20 20
0 200 400 600 800 1,000 1,200 0 20 40 60 80 100 120 140 160
Number of HD patients in centre Number of PD patients in centre
Fig. 12.17. Funnel plot for percentage of HD patients achieving Fig. 12.18. Funnel plot for percentage of PD patients achieving
simultaneous control of all three BMD parameters in preventing simultaneous control of all three BMD parameters in preventing
severe hyperparathyroidism by centre in 2012 severe hyperparathyroidism by centre in 2012
255
Haemodialysis Guideline 6.3: Pre-dialysis serum Citing evidence for reduced risk of adverse events, the
bicarbonate concentrations haemodialysis module of the 5th edition of the Renal
Association clinical practice guidelines published in
‘We suggest that pre-dialysis serum bicarbonate con- December 2009 [1, 17–18] recommended a target range
centrations, measured with minimum delay after vene- for serum bicarbonate of 18–24 mmol/L, a reduction
puncture, should be between 18 and 24 mmol/l. (2C)’ from the previous guideline range of 20–26 mmol/L.
[17] Bicarbonate data were 91% complete for both HD and
PD patients (tables 12.17, 12.19). A lower bicarbonate
Peritoneal Dialysis Guideline 6.2 – PD: Metabolic RA target range in haemodialysis patients was introduced
factors in 2010. The proportion of patients achieving the
audit measure was 59% in 2012 (95% CI 58–60%)
‘We recommend that plasma bicarbonate should be (table 12.18); the mean bicarbonate was 24 mmol/L
maintained within the normal range’ [18] (table 12.17). The proportion achieving the standard in
Table 12.17. Summary statistics for serum bicarbonate in haemodialysis patients by centre in 2012

England
B Heart 93.8 376 21.8 3.0 22 20 24
B QEH 96.2 831 23.6 2.6 24 22 25
Basldn 96.7 145 23.2 3.2 24 22 25
Bradfd 98.4 186 24.0 3.2 24 22 26
Brightn 90.8 307 23.2 3.1 23 21 25
Bristol 100.0 461 22.7 2.4 23 21 24
Camb 94.8 307 23.7 2.2 24 22 25
Carlis 100.0 57 22.0 3.0 22 20 25
Carsh 92.8 648 23.4 3.9 24 21 26
Chelms 100.0 121 21.9 2.0 22 21 23
Colchr 92.6 100 24.6 1.7 25 23 26
Covnt 98.5 330 24.5 3.3 24 22 27
Derby 99.5 208 22.1 2.7 22 20 24
Donc 100.0 158 23.1 3.1 23 21 25
Dorset 99.6 243 22.8 2.8 23 21 24
Dudley 100.0 153 23.7 2.7 24 22 25
Exeter 100.0 351 21.0 2.6 21 19 23
Glouc 100.0 193 24.0 2.6 24 22 26
Hull 100.0 310 22.0 2.3 22 21 23
Ipswi 100.0 124 23.0 2.8 23 21 25
Kent 99.5 359 21.8 2.7 22 20 23
L Barts 66.4 562 22.4 3.1 22 21 24
L Guys 71.6 424 22.3 3.0 22 20 24
L Kings 99.8 459 26.5 2.2 26 25 28
L Rfree 82.9 554 23.1 2.9 23 21 25
L St.G 97.4 264 26.6 2.8 26.5 25 28.5
L West 65.6 880 19.2 2.7 19 17 21
Leeds 100.0 454 22.3 3.6 22 20 25
Leic 99.5 797 24.8 3.7 24 22 27
Liv Ain 98.2 163 24.0 2.9 24 22 26
Liv RI 99.4 343 27.3 2.9 28 26 29
M RI 92.2 437 24.1 3.1 24 22 26
Middlbr 99.4 310 27.2 3.2 27 25 29
Newc 100.0 262 25.0 3.2 26 24 27
Norwch 99.3 301 24.0 2.9 24 22 26
Nottm 94.4 335 25.1 3.0 25 23 27
Oxford 100.0 389 23.5 3.2 24 22 25
Plymth 100.0 119 25.6 2.6 26 24 27
256

Ports 99.8 509 22.9 2.8 23 21 25

Prestn 98.8 490 23.3 3.0 23 21 25
Redng 100.0 251 24.5 2.4 24 23 26
Salford 9.0 31
Sheff 99.8 561 24.8 3.2 25 23 27
Shrew 100.0 184 24.5 2.9 24 22 26
Stevng 97.9 372 23.3 3.1 23 21 25
Sthend 100.0 107 25.1 3.6 25 23 27
Stoke 33.3 98
Sund 99.5 183 26.5 3.1 27 25 29
Truro 99.3 133 21.5 2.4 21 20 23
Wirral 97.7 173 24.3 2.8 25 22 26
Wolve 98.9 267 21.9 2.6 22 20 24
York 100.0 122 23.6 2.7 23 22 25
N Ireland
Antrim 97.6 123 23.6 2.8 23 22 25
Belfast 99.0 206 23.8 2.4 24 22 25
Newry 100.0 85 22.4 2.8 22 20 24
Ulster 100.0 101 23.3 2.9 23 22 25
West NI 100.0 129 24.1 2.7 24 22 26
Wales
Bangor 100.0 82 24.9 3.4 25 23 27
Cardff 96.0 430 22.8 4.0 22 20 25
Clwyd 100.0 76 22.6 2.7 23 21 24
Swanse 100.0 308 24.5 2.9 24 23 26
Wrexm 100.0 86 21.4 2.1 22 20 23
England 90.1 16,502 23.5 3.5 23 21 26
N Ireland 99.2 644 23.6 2.7 23 22 25
Wales 98.2 982 23.4 3.5 23 21 26
E, W & NI 90.8 18,128 23.5 3.5 23 21 26
Blank cells denote centres excluded from analyses due to poor data completeness
Table 12.18. Percentage of haemodialysis patients within, below and above the range for bicarbonate (18–24 mmol/L) by centre in 2012

bicarb 95% 95% bicarb bicarb within range LCL UCL
Centre N 18–24 mmol/L CI CI ,18 mmol/L .24 mmol/L from 2011 change change
England
B Heart 376 78.5 74.0 82.3 5.9 15.7 52.7 46.5 59.0
B QEH 831 62.1 58.7 65.3 0.8 37.1 8.8 4.0 13.5
Basldn 145 69.0 61.0 76.0 3.5 27.6 −7.2 −17.6 3.3
Bradfd 186 51.1 43.9 58.2 2.7 46.2 −4.9 −15.1 5.4
Brightn 307 64.2 58.7 69.3 3.9 31.9 1.8 −5.9 9.6
Bristol 461 75.5 71.4 79.2 3.0 21.5 25.0 19.0 31.1
Camb 307 64.8 59.3 70.0 0.7 34.5 0.5 −7.0 8.0
Carlis 57 63.2 50.0 74.6 8.8 28.1 −2.4 −19.9 15.1
Carsh 648 52.9 49.1 56.8 6.5 40.6 −8.3 −13.7 −2.9
Chelms 121 86.8 79.5 91.7 2.5 10.7 25.0 14.3 35.8
Colchr 100 42.0 32.7 51.9 0.0 58.0 27.0 15.1 38.9
Covnt 330 53.3 47.9 58.7 1.8 44.9 18.8 11.3 26.2
Derby 208 76.9 70.7 82.2 2.9 20.2 23.2 14.1 32.3
Donc 158 67.7 60.1 74.6 1.3 31.0 1.1 −9.4 11.5
257

Dorset 243 74.1 68.2 79.2 1.2 24.7 11.0 2.6 19.4
Dudley 153 58.8 50.9 66.3 2.0 39.2 3.2 −8.3 14.7
Exeter 351 82.1 77.7 85.7 9.1 8.8 12.7 6.3 19.1
Glouc 193 58.0 51.0 64.8 0.5 41.5 3.4 −6.6 13.4
Hull 310 86.1 81.8 89.6 2.3 11.6 3.7 −2.1 9.4
Ipswi 124 70.2 61.6 77.6 0.8 29.0 35.7 24.1 47.4
Kent 359 81.9 77.6 85.5 3.9 14.2 6.4 0.4 12.4
L Barts 562 71.7 67.8 75.3 5.7 22.6 25.2 20.1 30.3
L Guys 424 75.0 70.7 78.9 4.3 20.8 13.0 6.4 19.6
L Kings 459 17.0 13.8 20.7 0.0 83.0 −13.3 −18.9 −7.8
L Rfree 554 69.7 65.7 73.4 2.5 27.8 0.1 −5.4 5.5
L St.G 264 20.1 15.7 25.3 0.0 79.9 −9.3 −16.6 −2.0
L West 880 71.7 68.6 74.6 25.3 3.0 −2.9 −7.1 1.2
Leeds 454 63.7 59.1 68.0 8.8 27.5 −10.5 −16.4 −4.5
Leic 797 47.8 44.4 51.3 2.6 49.6 −3.1 −8.0 1.8
Liv Ain 163 57.1 49.4 64.4 0.0 42.9 −6.3 −17.1 4.4
Liv RI 343 13.4 10.2 17.4 0.6 86.0 −45.5 −51.8 −39.3
M RI 437 54.9 50.2 59.5 2.1 43.0 −7.1 −13.8 −0.4
Middlbr 310 20.3 16.2 25.2 0.0 79.7 −2.7 −9.4 4.0
Newc 262 32.1 26.7 38.0 3.1 64.9 12.9 5.4 20.4
Norwch 301 57.8 52.2 63.3 0.7 41.5 1.9 −6.1 9.9
Nottm 335 39.1 34.0 44.4 1.5 59.4 10.4 3.4 17.3
Oxford 389 61.4 56.5 66.2 2.8 35.7 14.7 7.7 21.6
Plymth 119 28.6 21.2 37.3 0.0 71.4 −50.5 −61.3 −39.6
Ports 509 70.9 66.8 74.7 3.1 25.9 −0.5 −6.2 5.2
Prestn 490 60.8 56.4 65.0 3.1 36.1 0.9 −5.2 7.1
Redng 251 49.8 43.7 56.0 0.8 49.4 −3.9 −12.6 4.9
Sheff 561 46.5 42.4 50.7 0.7 52.8 −2.2 −8.1 3.6
Shrew 184 50.5 43.4 57.7 0.0 49.5 −9.1 −19.4 1.1
Stevng 372 68.8 63.9 73.3 1.6 29.6 10.5 3.7 17.4
Sthend 107 36.5 27.9 46.0 1.9 61.7 −11.8 −24.7 1.1
Sund 183 20.2 15.0 26.7 1.6 78.1 17.1 10.7 23.5
Truro 133 82.7 75.3 88.2 4.5 12.8 37.8 27.3 48.3
Wirral 173 47.4 40.1 54.8 0.6 52.0 1.2 −9.6 11.9
Wolve 267 78.3 72.9 82.8 3.8 18.0 3.1 −3.9 10.1
York 122 62.3 53.4 70.4 1.6 36.1 20.6 8.1 33.0
N Ireland
Antrim 123 63.4 54.6 71.4 1.6 35.0 41.6 30.3 52.9
Belfast 206 63.6 56.8 69.9 0.5 35.9 10.4 0.9 19.9
Newry 85 72.9 62.6 81.3 3.5 23.5 32.1 18.6 45.7
Ulster 101 71.3 61.7 79.3 2.0 26.7 5.0 −7.8 17.7
West NI 129 53.5 44.9 61.9 1.6 45.0 −14.4 −26.1 −2.7
Wales
Bangor 82 47.6 37.0 58.3 1.2 51.2 14.6 −0.1 29.3
Cardff 430 64.0 59.3 68.4 5.8 30.2 −0.2 −6.6 6.1
Clwyd 76 77.6 66.9 85.6 2.6 19.7 −0.3 −14.5 13.8
Swanse 308 51.0 45.4 56.5 1.6 47.4 16.1 8.5 23.7
Wrexm 86 89.5 81.1 94.5 3.5 7.0 0.4 −8.9 9.7
England 16,502 58.5 57.8 59.3 3.9 37.6 4.1 3.0 5.1
N Ireland 644 64.0 60.2 67.6 1.6 34.5 13.3 8.0 18.6
Wales 982 61.8 58.7 64.8 3.7 34.5 7.0 2.7 11.3
E, W & NI 18,128 58.9 58.2 59.6 3.8 37.3 4.5 3.5 5.6
258
Table 12.19. Summary statistics for serum bicarbonate in peritoneal dialysis patients by centre in 2012

England
B Heart 97.6 41 21.2 2.6 21 20 23
B QEH 92.0 137 24.2 3.5 25 22 26
Basldn 96.4 27 27.6 3.9 27 24 30
Bradfd 95.8 23 26.3 2.5 27 24 28
Brightn 79.7 55 25.3 3.5 26 22 28
Bristol 100.0 56 22.5 2.7 23 21 24
Camb 93.8 30 27.8 3.4 29 25 30
Carlis 100.0 21 22.3 3.8 23 21 24
Carsh 90.7 88 26.7 3.5 27 25 29
Chelms 100.0 25 24.7 3.4 26 23 27
Colchr∗
Covnt 89.3 75 26.2 2.8 26 24 28
Derby 100.0 84 24.3 3.1 25 22 26
Donc 100.0 23 25.5 3.4 26 22 29
Dorset 68.4 26 24.5 3.5 25 23 28
Dudley 98.1 52 27.3 3.1 28 26 29
Exeter 100.0 69 21.7 3.1 22 20 24
Glouc 93.6 29 25.2 2.9 26 23 26
Hull 94.9 75 26.1 3.0 27 25 28
Ipswi 100.0 30 28.2 3.1 29 26 31
Kent 98.2 54 23.9 2.8 23 22 26
L Barts 97.6 163 23.9 2.7 24 22 26
L Guys 96.3 26 23.8 2.7 24 22 26
L Kings 98.7 75 27.1 2.9 27 25 29
L Rfree 81.4 83 26.6 3.0 27 25 29
L St.G 97.9 47 27.5 2.6 28 26 29
L West 100.0 47 21.5 2.7 22 20 24
Leeds 100.0 77 26.0 3.1 26 24 29
Leic 96.5 138 26.8 3.7 27 25 29
Liv Ain 100.0 17
Liv RI 98.2 54 25.8 3.4 26 24 28
M RI 98.7 75 26.0 3.2 26 24 28
Middlbr 87.5 7
Newc 86.5 32 24.1 2.5 24 23 26
Norwch 100.0 48 22.9 3.2 23 21 24
Nottm 55.6 40 28.6 3.2 28 26 31
Oxford 76.8 53 25.1 3.6 25 24 27
Plymth 93.6 29 25.2 2.3 26 24 27
Ports 98.7 77 25.6 2.6 26 24 27
Prestn 98.3 58 24.8 3.7 26 22 28
Redng 100.0 63 27.8 2.6 28 26 29
Salford 11.1 10
Sheff 100.0 67 26.0 3.3 26 24 28
Shrew 97.0 32 26.5 3.3 27 25 29
Stevng 96.3 26 26.0 2.5 26 25 27
Sthend 100.0 14
Stoke 98.6 68 26.9 3.5 27 25 29
Sund 100.0 17
Truro 94.7 18
Wirral 75.9 22 25.1 2.2 25 24 27
Wolve 97.6 81 24.6 2.7 25 23 26
York 100.0 27 26.3 2.5 27 24 29
259

N Ireland
Antrim 30.0 3
Belfast 96.0 24 25.4 3.3 25 23 28
Newry 57.1 8
Ulster 100.0 6
West NI 86.7 13
Wales
Bangor 100.0 14
Cardff 93.0 66 26.5 3.9 26 23 30
Clwyd 100.0 15
Swanse 100.0 54 25.6 3.3 27 24 28
Wrexm 95.0 19
England 91.2 2,611 25.4 3.5 26 23 28
N Ireland 77.1 54 24.6 2.9 24 23 27
Wales 96.6 168 25.9 3.4 26 24 28
E, W & NI 91.2 2,833 25.4 3.5 26 23 28
Blank cells denote low patient numbers or poor data completeness
∗
No PD patients
PD patients was 80% (CI 78–81%) (table 12.20). Collec- (tables 12.18, 12.20). The UKRR has previously con-
tively there was significant inter-centre variation for ducted a limited survey into the possible underlying
both HD and PD (tables 12.18, 12.20, figures 12.19, causes of this variation. The study predominantly looked
12.20). There was even greater between centre variation at measures of sample processing and of dialysis treat-
in the proportion of patients with bicarbonate values ment. It did not adjust for case-mix and was unable to
above and below the specified range for the audit measure detect any significant differences between centres.
Table 12.20. Percentage of peritoneal dialysis patients within, below and above the range for bicarbonate (22–30 mmol/L) by centre in 2012
England
B Heart 41 41.5 27.6 56.9 58.5 0.0 −47.7 −65.8 −29.6
B QEH 137 72.3 64.2 79.1 24.1 3.7 −8.4 −18.4 1.5
Basldn 27 74.1 54.7 87.1 3.7 22.2 −21.8 −40.1 −3.4
Bradfd 23 100.0 0.0 100.0 0.0 0.0 22.2 6.5 37.9
Brightn 55 80.0 67.4 88.6 14.6 5.5 1.3 −13.4 16.1
Bristol 56 62.5 49.3 74.1 37.5 0.0 −27.3 −42.2 −12.5
Camb 30 76.7 58.5 88.5 6.7 16.7 −4.6 −24.9 15.7
Carlis 21 57.1 36.0 76.0 42.9 0.0 24.0 24.0 24.0
Carsh 88 80.7 71.1 87.6 8.0 11.4 4.5 −7.6 16.7
Chelms 25 76.0 55.8 88.8 24.0 0.0 −14.9 −35.5 5.7
Covnt 75 86.7 77.0 92.7 8.0 5.3 1.1 −9.9 12.2
Derby 84 82.1 72.5 88.9 15.5 2.4 −3.3 −14.1 7.5
Donc 23 78.3 57.2 90.7 17.4 4.4 −17.0 −36.1 2.2
Dorset 26 73.1 53.3 86.6 23.1 3.9 −1.3 −23.2 20.6
Dudley 52 84.6 72.1 92.1 3.9 11.5 5.5 −9.7 20.6
Exeter 69 55.1 43.3 66.3 44.9 0.0 −30.2 −44.9 −15.5
Glouc 29 86.2 68.5 94.7 6.9 6.9 −7.5 −22.6 7.6
Hull 75 90.7 81.7 95.5 8.0 1.3 2.8 −7.1 12.8
Ipswi 30 73.3 55.0 86.1 0.0 26.7 −16.7 −35.8 2.5
Kent 54 79.6 66.8 88.4 20.4 0.0 10.8 −5.0 26.6
L Barts 163 79.1 72.2 84.7 20.3 0.6 −8.0 −16.3 0.2
L Guys 26 76.9 57.2 89.3 23.1 0.0 −5.2 −26.8 16.3
260

L Kings 75 81.3 70.9 88.6 1.3 17.3 −15.8 −25.5 −6.2

L Rfree 83 83.1 73.5 89.8 9.6 7.2 −0.6 −12.0 10.8
L St.G 47 87.2 74.4 94.2 4.3 8.5 18.6 2.7 34.5
L West 47 53.2 39.1 66.8 46.8 0.0 −18.7 −39.8 2.4
Leeds 77 81.8 71.6 88.9 11.7 6.5 −7.1 −18.1 3.9
Leic 138 79.0 71.4 85.0 5.8 15.2 3.1 −6.9 13.0
Liv RI 54 83.3 71.0 91.1 7.4 9.3 −0.9 −14.6 12.8
M RI 75 86.7 77.0 92.7 6.7 6.7 −0.6 −11.6 10.3
Newc 32 81.3 64.1 91.3 18.8 0.0 −0.3 −18.6 18.0
Norwch 48 68.8 54.4 80.2 29.2 2.1 2.8 −16.1 21.6
Nottm 40 75.0 59.5 86.0 0.0 25.0 0.0 0.0 0.0
Oxford 53 86.8 74.8 93.6 9.4 3.8 22.0 6.3 37.6
Plymth 29 89.7 72.4 96.6 10.3 0.0 0.5 −14.5 15.4
Ports 77 93.5 85.3 97.3 3.9 2.6 5.7 −3.6 14.9
Prestn 58 81.0 68.9 89.2 17.2 1.7 1.4 −13.3 16.1
Redng 63 82.5 71.2 90.1 0.0 17.5 7.5 −6.2 21.2
Sheff 67 80.6 69.4 88.4 9.0 10.5 2.8 −11.8 17.4
Shrew 32 81.3 64.1 91.3 9.4 9.4 −2.8 −22.5 17.0
Stevng 26 92.3 73.9 98.1 3.9 3.9 0.3 −14.5 15.1
Stoke 68 75.0 63.4 83.9 8.8 16.2 19.8 4.0 35.5
Wirral 22 95.5 73.9 99.4 4.6 0.0 28.8 8.0 49.6
Wolve 81 86.4 77.1 92.3 13.6 0.0 −0.7 −11.9 10.5
York 27 96.3 77.9 99.5 3.7 0.0 1.3 −10.6 13.2
N Ireland
Belfast 24 83.3 63.1 93.6 12.5 4.2 15.5 −7.4 38.3
Wales
Cardff 66 74.2 62.4 83.4 9.1 16.7 −6.9 −20.2 6.4
Swanse 54 83.3 71.0 91.1 13.0 3.7 7.8 −7.8 23.4
England 2,611 79.2 77.6 80.7 14.2 6.6 −2.2 −4.4 0.0
N Ireland 54 87.0 75.2 93.7 11.1 1.9 8.1 −5.8 22.0
Wales 168 81.0 74.3 86.2 10.1 8.9 −1.4 −9.5 6.8
E, W & NI 2,833 79.5 77.9 80.9 13.9 6.6 −1.9 −4.0 0.1
100
100
90 Solid lines show 95% limits 90
80
70 80
60
70
50
40
60
30
20
50 Dotted lines show 99.9% limits
10 Solid lines show 95% limits
0 40
0 100 200 300 400 500 600 700 800 900 0 20 40 60 80 100 120 140
Number of patients in centre Number of patients in centre
Fig. 12.19. Funnel plot for percentage of haemodialysis patients Fig. 12.20. Funnel plot for percentage of peritoneal dialysis
within the range for bicarbonate (18–24 mmol/L) by centre in patients within the range for bicarbonate (22–30 mmol/L) by
2012 centre in 2012
261
However, it is possible that there may be unmeasured We recommend that a total cholesterol of <4 mmol/L
processes including dialysis and oral bicarbonate pre- or a 25% reduction from baseline, or a fasting low den-
scription that might account for the variation observed sity lipoprotein (LDL)-cholesterol of <2 mmol/L or a
[19]. 30% reduction from baseline, should be achieved,
whichever is the greatest reduction in all patients.
Total cholesterol
There is no audit standard for total cholesterol in the Statins should not be withdrawn from patients in
Renal Association clinical practice guidelines. Current whom they were previously indicated and should con-
guidance on lipid management states: tinue to be prescribed when such patients start renal
replacement therapy (RRT) or change modality.’ [20]
‘We recommend that statins (or 3 hydroxy-3 methyl-
glutaryl-coenzyme A reductase inhibitors) should be Total cholesterol data were 82% complete for HD
considered for primary prevention in all CKD Stages patients and 78% complete for PD patients. As there
1–4 and transplant patients with a 10-year risk of are no specific audit measures for total cholesterol, sum-
cardiovascular disease, calculated as >20% according mary data are presented for each dialysis centre
to the Joint British Societies’ Guidelines – JBS2 (British (tables 12.21, 12.22, figures 12.21, 12.22). There are a
Hypertension Society British Cardiac Society 2005). number of case-mix factors (comorbidity, inflammation,
Table 12.21. Summary statistics for total cholesterol in haemodialysis patients by centre in 2012
England
B Heart 98.8 396 4.2 1.1 4.1 3.3 4.8
B QEH 93.8 810 4.0 1.1 3.8 3.2 4.6
Basldn 98.0 147 3.9 1.0 3.7 3.2 4.6
Bradfd 91.5 173 3.8 1.0 3.7 3.1 4.4
Brightn 32.8 111
Bristol 94.8 437 4.1 1.1 3.9 3.4 4.7
Camb 83.3 270 3.8 1.1 3.6 3.0 4.5
Carlis 100.0 57 4.2 1.2 4.1 3.3 5.0
Carsh 85.8 599 4.1 1.1 3.9 3.4 4.8
Chelms 88.4 107 3.6 0.9 3.5 2.9 4.1
Colchr 50.9 55 3.8 1.1 3.7 2.9 4.5
Covnt 0.3 1
Derby 95.7 200 4.2 1.1 4.0 3.4 4.9
Donc 97.5 154 3.8 0.9 3.7 3.1 4.4
Dorset 91.4 223 3.9 1.0 3.9 3.2 4.6
Dudley 89.5 137 3.7 1.0 3.6 3.0 4.3
Exeter 96.6 339 3.9 1.0 3.8 3.2 4.5
Glouc 95.9 185 4.0 1.1 4.0 3.3 4.6
Hull 23.6 73
Ipswi 90.3 112 3.8 1.0 3.6 3.1 4.3
Kent 92.0 332 4.0 1.1 3.9 3.2 4.7
L Barts 94.1 796 4.1 1.1 4.0 3.4 4.9
L Guys 39.5 234
L Kings 89.8 413 3.8 0.9 3.7 3.1 4.4
L Rfree 61.8 413 4.1 1.2 3.9 3.3 4.7
L St.G 90.4 245 4.0 1.1 3.9 3.3 4.6
L West 83.5 1,120 3.6 0.9 3.5 2.9 4.2
Leeds 98.5 447 3.8 0.9 3.7 3.2 4.4
Leic 95.3 763 3.8 1.0 3.8 3.1 4.4
Liv Ain 88.6 147 4.0 1.1 3.8 3.1 4.8
Liv RI 96.5 333 3.8 1.1 3.8 3.0 4.5
M RI 91.1 432 4.0 1.1 3.8 3.2 4.7
Middlbr 80.8 252 4.3 1.1 4.2 3.5 4.9
Newc 100.0 262 3.8 1.0 3.7 3.1 4.3
Norwch 95.7 290 4.0 1.0 3.9 3.2 4.6
262

Nottm 99.2 352 4.0 1.0 3.9 3.3 4.7

Oxford 54.2 211 3.8 1.0 3.6 2.9 4.5
Plymth 90.8 108 3.8 0.9 3.7 3.1 4.3
Ports 67.1 342 4.0 1.2 3.9 3.1 4.8
Prestn 75.2 373 3.9 1.0 3.8 3.1 4.5
Redng 96.4 242 3.8 1.0 3.7 3.1 4.3
Salford 49.3 170
Sheff 90.9 511 4.1 1.1 4.0 3.3 4.8
Shrew 96.7 178 4.0 1.0 3.9 3.3 4.5
Stevng 20.0 76
Sthend 94.4 101 3.8 1.1 3.8 3.1 4.5
Stoke 90.8 267 3.7 0.9 3.7 3.0 4.2
Sund 98.4 181 3.8 1.0 3.7 3.2 4.3
Truro 98.5 132 4.0 1.1 3.9 3.2 4.7
Wirral 91.5 162 3.8 1.1 3.6 3.0 4.4
Wolve 98.5 266 4.3 1.1 4.2 3.5 4.9
York 98.4 120 4.2 1.1 4.1 3.4 4.9
N Ireland
Antrim 98.4 124 3.7 1.1 3.5 3.0 4.3
Belfast 82.7 172 3.8 1.0 3.7 3.1 4.5
Newry 100.0 85 3.5 0.8 3.5 3.1 3.8
Ulster 100.0 101 3.8 1.1 3.6 3.1 4.6
West NI 100.0 129 3.7 0.8 3.7 3.1 4.2
Wales
Bangor 95.1 78 4.2 1.1 3.9 3.5 5.0
Cardff 95.3 427 4.1 1.1 4.0 3.3 4.8
Clwyd 97.4 74 3.9 1.0 3.8 3.3 4.4
Swanse 99.0 305 4.0 1.2 3.8 3.2 4.7
Wrexm 67.4 58 4.0 1.3 4.0 3.2 4.8
England 81.1 14,857 3.9 1.1 3.8 3.2 4.6
N Ireland 94.1 611 3.7 1.0 3.6 3.1 4.3
Wales 94.2 942 4.0 1.1 3.9 3.3 4.7
E, W & NI 82.2 16,410 3.9 1.1 3.8 3.2 4.6
Blank cells denote poor data completeness
Table 12.22. Summary statistics for total cholesterol in peritoneal dialysis patients by centre in 2012
England
B Heart 95.2 40 5.0 1.2 4.9 4.2 5.7
B QEH 96.0 143 4.6 1.2 4.5 3.8 5.3
Basldn 96.4 27 4.5 1.5 4.4 3.4 5.2
Bradfd 87.5 21 4.0 0.9 4.1 3.3 4.7
Brightn 18.8 13
Bristol 82.1 46 5.3 1.5 5.0 4.4 6.4
Camb 100.0 32 4.6 1.1 4.5 3.8 5.3
Carlis 95.2 20 4.4 0.8 4.4 3.8 4.9
Carsh 21.7 21
Chelms 88.0 22 4.6 1.2 4.4 3.6 5.3
Colchr∗
Covnt 0.0 0
Derby 86.9 73 4.9 1.5 4.6 3.6 5.9
Donc 52.2 12
Dorset 65.8 25 5.1 1.5 4.7 4.0 5.6
263

Dudley 60.4 32 4.3 1.1 4.4 3.4 5.0

Exeter 95.7 66 4.7 1.3 4.6 4.0 5.4
Glouc 83.9 26 4.6 1.0 4.4 4.0 5.0
Hull 25.3 20
Ipswi 96.7 29 4.5 1.1 4.5 3.7 5.6
Kent 87.3 48 4.6 1.4 4.7 3.5 5.4
L Barts 100.0 167 4.7 1.3 4.6 3.8 5.3
L Guys 55.6 15
L Kings 98.7 75 4.5 1.3 4.3 3.6 5.1
L Rfree 81.4 83 4.9 1.8 4.6 3.6 5.6
L St.G 97.9 47 4.4 1.2 4.4 3.4 5.4
L West 100.0 47 4.6 1.1 4.4 3.6 5.5
Leeds 90.9 70 4.3 0.9 4.1 3.5 4.9
Leic 96.5 138 4.5 1.2 4.4 3.6 5.1
Liv Ain 94.1 16
Liv RI 94.6 52 4.6 1.5 4.3 3.5 5.1
M RI 98.7 75 4.4 1.4 4.2 3.6 5.2
Middlbr 37.5 3
Newc 89.2 33 4.8 1.4 4.8 3.8 5.4
Norwch 100.0 48 4.9 1.4 4.7 3.9 5.6
Nottm 88.9 64 4.6 1.3 4.4 3.6 5.2
Oxford 58.0 40 4.7 1.5 4.4 3.7 5.4
Plymth 61.3 19
Ports 97.4 76 4.6 1.2 4.4 3.8 5.1
Prestn 86.4 51 4.4 1.2 4.0 3.6 4.7
Redng 71.4 45 4.3 1.5 4.0 3.4 4.9
Salford 85.6 77 4.6 1.3 4.5 3.9 5.4
Sheff 50.8 34 4.7 1.1 4.6 3.9 5.3
Shrew 81.8 27 4.7 1.3 4.5 3.7 5.6
Stevng 44.4 12
Sthend 78.6 11
Stoke 100.0 69 4.0 1.1 4.0 3.1 4.8
Sund 70.6 12
Truro 79.0 15
Wirral 58.6 17
Wolve 81.9 68 5.4 1.6 5.2 4.3 6.6
York 88.9 24 5.4 1.4 5.4 4.4 6.1
N Ireland
Antrim 100.0 10
Belfast 96.0 24 5.2 1.3 5.0 4.5 5.8
Newry 100.0 14
Ulster 100.0 6
West NI 100.0 15
Wales
Bangor 100.0 14
Cardff 57.8 41 4.8 1.2 4.8 3.9 5.6
Clwyd 86.7 13
Swanse 79.6 43 4.8 1.4 4.5 3.9 5.3
Wrexm 60.0 12
England 78.4 2,246 4.6 1.3 4.5 3.7 5.3
N Ireland 98.6 69 4.6 1.2 4.6 3.8 5.1
Wales 70.7 123 4.8 1.3 4.7 4.0 5.5
E, W & NI 78.4 2,438 4.7 1.3 4.5 3.7 5.3
Blank cells denote low patient numbers or poor data completeness
∗
No PD patients
264
100
N = 16,410 Upper 95% Cl
90 % with chol <4
Lower 95% Cl
80
70
60
50
40
30
20
0 Newry
12 Chelms
17 L West
0 West NI
2 Antrim
9 Plymth
10 Dudley
9 Stoke
10 Ipswi
8 Bradfd
49 Colchr
4 Redng
10 L Kings
0 Newc
17 Camb
17 Belfast
8 Wirral
46 Oxford
2 Basldn
5 Leic
0 Ulster
3 Liv RI
2 Sund
25 Prestn
2 Leeds
3 Donc
3 Exeter
11 Liv Ain
9 Dorset
9 M RI
6 Sthend
38 L Rfree
6 B QEH
1 Swanse
3 Shrew
33 Ports
3 Clwyd
10 L St.G
1 Nottm
4 Norwch
1 Truro
5 Bangor
5 Bristol
8 Kent
14 Carsh
5 Cardff
9 Sheff
6 L Barts
4 Glouc
4 Derby
33 Wrexm
1 B Heart
2 York
0 Carlis
19 Middlbr
1 Wolve
19 England
6 N Ireland
6 Wales
18 E, W & NI
Centre
Fig. 12.21. Median total cholesterol in haemodialysis patients by centre in 2012
70
60 % with chol <4
Lower quartile
50
40
30
20
10
0
0 Stoke
29 Redng
4 Basldn
12 Bradfd
14 Prestn
5 Liv RI
9 Leeds
13 Kent
2 L St.G
40 Dudley
1 L Kings
1 M RI
4 Leic
19 L Rfree
13 Derby
0 L Barts
3 Ports
11 Nottm
0 Camb
3 Ipswi
11 Newc
5 Carlis
0 L West
4 B QEH
42 Oxford
14 Salford
12 Chelms
0 Norwch
42 Cardff
49 Sheff
18 Shrew
20 Swanse
16 Glouc
4 Exeter
34 Dorset
18 Wolve
18 Bristol
5 B Heart
4 Belfast
11 York
22 England
1 N Ireland
29 Wales
22 E, W & NI
Centre
Fig. 12.22. Median total cholesterol in peritoneal dialysis patients by centre in 2012
malnutrition) which may account for any inter-centre

variation in addition to differences in prescription of
lipid lowering medication and other therapies known to
influence serum lipid concentration such as steroids or
sevelamer as examples.
References
1 Renal Association. Clinical Practice Guidelines. 5th Edition. http://www. 3 Steddon S, Sharpes E. Renal Association Clinical Practice Guideline.
renal.org/Clinical/GuidelinesSection/Guidelines.aspx CKD-Mineral and Bone Disorders 2010 http://www.renal.org/Clinical/
2 Ansell D, Feehally J, Feest T, Tomson C, Williams AJ, Warwick G: U.K. GuidelinesSection/CKD-MBD.aspx
Renal registry report 2007. Chapter 9. Management of biochemical vari- 4 Morton AR, Garland JS, Holden RM: Is the calcium correct? Measuring
ables. UK Renal Registry, Bristol, UK 2007 serum calcium in dialysis patients. Semin Dial;23:283–289
265
5 Kidney Disease: Improving Global Outcomes (KDIGO) CKD–MBD 13 Noordzij M, Korevaar JC, Boeschoten EW, Dekker FW, Bos WJ, Krediet
Work Group. KDIGO clinical practice guideline for the diagnosis, RT: The kidney disease outcomes quality initiative (k/doqi) guideline
evaluation, prevention, and treatment of chronic kidney disease–mineral for bone metabolism and disease in ckd: Association with mortality in
and bone disorder (CKD–MBD). Kidney International 2009; 76(suppl dialysis patients. Am J Kidney Dis 2005;46:925–932
113):S1–S130 14 Stevens LA, Djurdjev O, Cardew S, Cameron EC, Levin A: Calcium,
6 Spiegelhalter DJ: Funnel plots for comparing institutional performance. phosphate, and parathyroid hormone levels in combination and as a
Statistics in Medicine 2005;24:1185–1202 function of dialysis duration predict mortality: Evidence for the com-
7 Almond A, Ellis AR, Walker SW: Current parathyroid hormone plexity of the association between mineral metabolism and outcomes.
immunoassays do not adequately meet the needs of patients with J Am Soc Nephrol 2004;15:770–779
chronic kidney disease. Ann Clin Biochem 2011;49:63–67 15 Block GA, Klassen PS, Lazarus JM, Ofsthun N, Lowrie EG, Chertow GM:
8 Garrett G, Goldsmith DJA: Parathyroid hormone measurements, guide- Mineral metabolism, mortality, and morbidity in maintenance hemo-
lines statements and clinical treatments: a real world cautionary tale. dialysis. J Am Soc Nephrol 2004;15:2208–2218
Ann Clin Biochem 2011;49:4–6 16 Danese MD, Belozeroff V, Smirnakis K, Rothman KJ: Consistent control
9 Gardham C, Stevens PE, Delaney MP, LeRoux M, Coleman A, Lamb EJ: of mineral and bone disorder in incident hemodialysis patients. Clin J
Variability of parathyroid hormone and other markers of bone mineral Am Soc Nephrol 2008;2:2
metabolism in patients receiving haemodialysis. Clin J Am Soc Nephrol 17 Mactier R, Hoenich N, Breen C. Renal Association Clinical Practice Guide-
2010;5:1261–1267 line Haemodialysis 2009 http://www.renal.org/Clinical/GuidelinesSection/
10 Sturgeon CM, Sprague SM, Metcalfe W: Variation in parathyroid Haemodialysis.aspx
hormone immunoassay results – a critical governance issue in the man- 18 Woodrow G, Davies S. Renal Association Clinical Practice Guideline Per-
agement of chronic kidney disease. Nephrol Dial Transplant 2011;26: itoneal Dialysis 2010 http://www.renal.org/Clinical/GuidelinesSection/
3440–3445 PeritonealDialysis.aspx
11 Kalantar-Zadeh K, Kuwae N, Regidor DL, Kovesdy CP, Kilpatrick RD, 19 Ansell DF, T.G. (eds): Renal registry 7th annual report. Chapter 6:
Shinaberger CS, McAllister CJ, Budoff MJ, Salusky IB, Kopple JD: Adequacy of haemodialysis and serum bicarbonate, 2004, pp 59–86
Survival predictability of time-varying indicators of bone disease in 20 Holt S, Goldsmith D. Renal Association Clinical Practice Guidelines
maintenance hemodialysis patients. Kidney Int 2006;70:771–780 Cardiovascular Disease in CKD 2010 http://www.renal.org/Clinical/
12 Melamed ML, Eustace JA, Plantinga L, Jaar BG, Fink NE, Coresh J, GuidelinesSection/CardiovascularDiseaseInCKD.aspx
Klag MJ, Powe NR: Changes in serum calcium, phosphate, and pth
and the risk of death in incident dialysis patients: A longitudinal
study. Kidney Int 2006;70:351–357
266
Chapter 13 Clinical, Haematological and
Biochemical Parameters in Patients
Receiving Renal Replacement Therapy in
Paediatric Centres in the UK in 2012:
Rishi Pruthia, Heather Maxwellb, Anna Casulaa, Fiona Braddona, Malcolm Lewisc,
Catherine O’Briend, Jelena Stojanovice, Yincent Tsef, Carol Inwardg, Manish D Sinhae
a
UK Renal Registry, Bristol, UK; bRoyal Hospital for Sick Children (Yorkhill), Glasgow, UK; cManchester Children’s
Hospital, Manchester, UK; dBirmingham Children’s Hospital, Birmingham, UK; eEvelina Childrens Hospital, London, UK;
f
Royal Victoria Infirmary, Newcastle, UK; gBristol Royal Hospital for Children, Bristol, UK
Key Words
. Median height z-score for children on dialysis was
Biochemical variables . Children . Dialysis . ERF . −2.0 and for children with a functioning transplant
Haemoglobin . Height . Quality improvement . Transplant . −1.3.
Weight . 76% of transplant patients, 57% of haemodialysis
patients and 56% of peritoneal dialysis patients
had a systolic blood pressure within the 90th per-
centile standard.
Summary . 92% of transplant patients, 74% of HD patients and
83% of PD patients had a haemoglobin within or
. Median weight z-score for children on dialysis was above the age appropriate standard.
−1.1 whereas children with a functioning transplant . 50% of HD patients and 56% of PD patients
had a near normal weight (median z-score 0.1). achieved the audit standard for phosphate.
267
Introduction Data collection

The data presented in this report relate to the annual census
This report focuses on the following variables for the date of 31st December 2012.
Those paediatric centres with access to renal IT systems sub-
prevalent paediatric dialysis and transplantation cohort mitted encrypted electronic data directly to the UK Renal Registry
on 31st December 2012: (UKRR). Those centres without access, sent paper (Belfast and
Liverpool) or electronic returns (Filemaker systems in table 13.1)
1. The completeness of data returns to the renal in the original BAPN database format which were then entered
registry into the original BAPN database as in previous years. Complete
2. The anthropometric characteristics in children with transfer to the UKRR encrypted database is still awaited.
established renal failure (ERF) Governance, reporting and standardisation
3. Blood pressure control in children with ERF Information governance, reporting and standardisation were
4. Anaemia control in children with ERF all performed in an identical manner to previous analyses to
5. Key biochemical findings in this population. allow comparison [1]. Where the value of clinical parameters in
childhood varies with age and size, data are presented as z-scores.
Analyses of prevalent paediatric patients aged ,16
years receiving renal replacement therapy for the year Anthropometry
The reference range for height (Ht), weight (Wt) and body
2012 and for the period 2001 to 2012 inclusive are mass index (BMI) in childhood varies with gender and age. BMI
reported. A single dataset was collected for each patient was calculated using the formula BMI = Wt (kg)/ Ht (m)2. Height
per year during this time period. Due to low numbers of and weight were adjusted for age. To account for discrepancies in
patients in each cohort, no incident cohort analyses have linear growth secondary to renal disease, BMI was expressed
been undertaken. Centre specific data for each paediatric according to height-age, rather than chronological age. The
International Obesity Taskforce (IOTF) definition proposed by
nephrology centre in the UK has also been provided. Cole et al [2] was used to define overweight and obesity; z-scores
were calculated based on the British 1990 reference data for height
and weight [3].
Methods Blood pressure (BP)

The reference range for blood pressure varies with gender, age
and height. The data is therefore presented as z-scores based on
There were 13 centres providing care for children requiring data from the fourth report of the National High Blood Pressure
renal replacement therapy in the UK, ten of which also provided Education Programme (NHBPEP) working group in the United
surgical renal transplant services. All 13 centres provided out- States [4].
patient and inpatient follow up for children who had received
kidney transplants. Centres are listed in table 13.1 and appendix K. Laboratory values
Haemoglobin (Hb), ferritin (Ferr), calcium (Ca) and phosphate
(Phos) were analysed using age related laboratory reference
Table 13.1. Paediatric renal centres, their abbreviations and IT ranges as in table 13.2. Data analysis is presented for each centre
systems individually and at a national level for each variable.
Renal
Paediatric centre Abbreviation IT system
Belfast∗ Blfst_P Mediqal Statistical analyses

Birmingham Bham_P Proton
Bristol Brstl_P Proton Data were analysed to calculate summary statistics
Cardiff Cardf_P Proton
Glasgow Glasg_P Filemaker (maximum, minimum, mean and median values in
Leeds Leeds_P Proton addition to standard deviation and quartile ranges).
Liverpool Livpl_P None Where applicable, the percentage achieving the audit
London Evelina L Eve_P Filemaker standard was also calculated. If a patient had missing
London Great Ormond Street L GOSH_P Filemaker
Manchester Manch_P Filemaker
data, they were excluded from the relevant analyses.
Newcastle Newc_P Clinical Longitudinal analyses of attainment of standards over
Vision time were also performed. These were based on a single
Nottingham Nottm_P Proton data point per ERF patient per year collected as described
Southampton Soton_P Bespoke previously. Cautious interpretation of these analyses is
∗
New system installed, although paper submission received in 2012 required due to changing audit standards over time and
268
Chapter 13 Paediatric biochemistry
Table 13.2. Summary of relevant biochemical clinical audit measures
Age
Parameter ,1 year 1–,6 years 6–12 years .12 years

Haemoglobin (g/L), NICE guideline CG 114 Maintain 95–115 Maintain 100–120 100–120 100–120
for ,2 years for .2 years
Ferritin (mg/L) 200–500 200–500 200–500 200–500
Corrected calcium (mmol/L) 2.24–2.74 2.19–2.69 2.19–2.69 2.15–2.55
Phosphate (mmol/L) 1.10–1.95 1.05–1.75 1.05–1.75 1.05–1.75
2
eGFR ml/min/1.73 m (transplant patients) Estimated GFR (eGFR) as per Schwartz formula: (height × k)/ plasma creatinine
The value for k is that in use at the reporting centre
Parathyroid hormone (individual centre units) Within twice the normal range
Levels may be maintained within normal range if growing appropriately
variable data returns for previous years. All analyses were Calcium, phosphate and parathyroid hormone (PTH) levels
done using SAS 9.3. Phosphate and calcium should be kept within the normal range
[5]. For analyses of calcium and phosphate, the age related ranges
as described previously have been used [1]. PTH levels should be
kept less than twice the upper limit of normal.
Standards
Standards are from the treatment of adults and children with Results
renal failure, Renal Association 2002 guidelines [5] unless other-
wise stated.
Data completeness
Anthropometry Tables 13.3 and 13.4 show the completeness of data
‘Height and weight should be monitored at each clinic visit. returns for transplant and dialysis patients for 2012.
Measures of supine length or standing head circumference should In 2012, overall completeness was good, with vir-
be measured during each visit up to two years of age and 6 tually all data variables showing a significant rise in
monthly up to 5 years of age. All measurements should be plotted
on European reference growth charts for healthy children.’ completeness compared to 2011, maintaining the
improvement noted in data returns over recent years.
Blood pressure The only exception were data returns for cholesterol
‘Blood pressure varies throughout childhood and should be which continued to remain poor with four centres
maintained within 2 standard deviations of the mean for normal
reporting on data for ,50% patients, it is planned that
children of the same height and sex. Systolic blood pressure
during PD or post-HD should be maintained at <90th percentile analysis of this data will be included in next year’s
for age, gender and height.’ report.
The analyses of blood pressure in this report present the
achievement of blood pressures at or below the 90th percentile. Height, weight and BMI
Anaemia
Figures 13.1 and 13.4 show that children receiving
Guidance on the management of anaemia in adults and chil- renal replacement therapy were short for their age;
dren with chronic kidney disease was updated and published by those on dialysis were significantly shorter that those
the National Institute for Clinical Excellence (NICE) in February with renal transplants. The overall median z-score was
2011 (Clinical Guideline 114) [6]. The recommendation in this −1.3 in the transplanted group and −2.0 in the dialysis
guidance is that in children with chronic kidney disease, treatment
should maintain stable haemoglobin levels between 100 and
group, p , 0.0001.
120 g/L in children above 2 years of age and between 95 and Children with a functioning kidney transplant had a
115 g/L in children below 2 years of age. These NICE standards median weight z-score of 0.1, (figure 13.2), whilst those
have been adopted for this report. on dialysis had a significantly lower weight z-score than
269
Table 13.3. Percentage data completeness for transplant patients ,16 years old by centre for each variable and total number of patients
per centre in 2012
Transplant
patients Systolic IV
Centre N Height Weight BMI BP Hb Creat Ferr EPO iron Chol HCO3 PTH Ca Phos
Bham_P 59 100.0 100.0 100.0 100.0 100.0 100.0 50.9 8.5 8.5 78.0 100.0 88.1 100.0 100.0
Blfst_P∗ 21 95.2 100.0 95.2 100.0 100.0 100.0 19.1 100.0 76.2 61.9 100.0 9.5 100.0 100.0
Brstl_P 35 94.3 97.1 94.3 97.1 100.0 100.0 68.6 100.0 100.0 74.3 100.0 80.0 100.0 100.0
Cardf_P 17 100.0 100.0 100.0 100.0 100.0 100.0 94.1 100.0 100.0 47.1 100.0 100.0 100.0 100.0
Glasg_P 27 100.0 100.0 100.0 100.0 100.0 100.0 85.2 100.0 100.0 37.0 100.0 100.0 100.0 100.0
L Eve_P 62 98.4 100.0 98.4 100.0 100.0 100.0 100.0 100.0 100.0 75.8 100.0 96.8 100.0 100.0
L GOSH_P 113 93.8 96.5 93.8 93.8 100.0 94.7 99.1 94.7 97.4 8.9 98.2 96.5 100.0 100.0
Leeds_P 57 96.5 98.3 96.5 98.3 100.0 100.0 35.1 96.5 96.5 93.0 93.0 35.1 98.3 93.0
Livpl_P 21 95.2 95.2 95.2 95.2 95.2 95.2 81.0 90.5 90.5 85.7 95.2 85.7 95.2 95.2
Manch_P 30 96.7 100.0 96.7 100.0 100.0 100.0 80.0 100.0 100.0 73.3 100.0 100.0 100.0 100.0
Newc_P 24 100.0 100.0 100.0 100.0 100.0 100.0 87.5 100.0 100.0 83.3 100.0 91.7 100.0 100.0
Nottm_P 54 92.6 96.3 92.6 94.4 98.2 98.2 85.2 100.0 100.0 24.1 98.2 57.4 98.2 98.2
Soton_P 14 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
UK 534 96.4 98.3 96.4 97.6 99.6 98.5 77.3 88.0 87.6 56.2 98.5 80.5 99.4 98.9
∗
Belfast do not routinely measure PTH in transplant patients
that of healthy children with a median of −1.1 children are either overweight or obese, compared to
(figure 13.5), p , 0.0001. 25.7% of children on dialysis.
Body mass index in children, reported here based on An analysis was performed excluding patients with
‘height age’, with a functioning transplant in 2012 showed syndromes and those born prematurely whose growth
inter-centre variation with a median z-score of 1.0 might be compromised. Table 13.5 shows that 27.7% of
(figure 13.3) which was significantly higher than the patients with a functioning transplant had a height
median BMI z-score in those on dialysis which was ,2SD, whilst the proportion below the normal range
0.40 (figure 13.6), p = <0.0001. This is also highlighted was even greater amongst those on haemodialysis
in figure 13.7 which shows that 42.3% of transplanted (50.0%) and those on peritoneal dialysis (41.2%),
Table 13.4. Percentage data completeness for dialysis patients ,16 years old by centre for each variable and total number of patients
per centre in 2012
Dialysis
patients Systolic IV
Centre N Height Weight BMI BP Hb Ferr EPO iron Chol HCO3 PTH Ca Phos
Bham_P 21 100.0 100.0 100.0 100.0 100.0 90.5 4.8 4.8 90.5 100.0 100.0 100.0 100.0
Blfst_P 6 83.3 100.0 83.3 83.3 100.0 66.7 100.0 100.0 50.0 83.3 100.0 100.0 100.0
Brstl_P 5 100.0 100.0 100.0 100.0 100.0 100.0 100.0 80.0 80.0 100.0 100.0 100.0 100.0
Cardf_P 3 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Glasg_P 13 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 53.9 100.0 100.0 100.0 100.0
L Eve_P 12 91.7 91.7 91.7 91.7 100.0 91.7 100.0 100.0 25.0 100.0 100.0 100.0 100.0
L GOSH_P 25 100.0 100.0 100.0 100.0 100.0 88.0 96.0 100.0 72.0 100.0 100.0 100.0 100.0
Leeds_P 8 87.5 100.0 87.5 100.0 100.0 100.0 100.0 100.0 87.5 100.0 100.0 100.0 100.0
Livpl_P 4 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 75.0 100.0 100.0 100.0 100.0
Manch_P 22 95.5 100.0 95.5 95.5 100.0 95.5 100.0 100.0 13.6 95.5 100.0 100.0 100.0
Newc_P 3 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Nottm_P 15 93.3 100.0 93.3 100.0 100.0 93.3 100.0 100.0 66.7 100.0 100.0 100.0 100.0
Soton_P 8 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 62.5 100.0 87.5 100.0 100.0
UK 145 96.6 99.3 96.6 97.9 100.0 93.1 85.5 85.5 50.0 98.6 99.3 100.0 100.0
Blank cell denotes data items which could not be sent by centre due to technical reasons
270
2
Upper quartile N = 515
1 Median
Lower quartile
Height z-score
–1
–2
–3
–4
Bham_P
Livpl_P
Cardf_P
Nottm_P
Glasg_P
LGOSH_P
LEve_P
Soton_P
Blfst_P
Newc_P
Manch_P
Leeds_P
Brstl_P
UK
Fig. 13.1. Median height z-scores for
Centre transplant patients ,16 years in 2012
3
N = 525 Upper quartile
2 Median
Lower quartile
Weight z-score
–1
–2
–3
Blfst_P
Livpl_P
Cardf_P
Leeds_P
L GOSH_P
L Eve_P
Soton_P
Brstl_P
Nottm_P
Glasg_P
Bham_P
Newc_P
Manch_P
UK
Fig. 13.2. Median weight z-scores for

Centre
transplant patients ,16 years in 2012
2
BMI z-score
0
–1 Median
Lower quartile
–2
Manch_P
Glasg_P
Bham_P
Livpl_P
L GOSH_P
Newc_P
Nottm_P
Leeds_P
Soton_P
Blfst_P
L Eve_P
Brstl_P
Cardf_P
UK
Fig. 13.3. Median BMI z-scores for

Centre transplant patients ,16 years in 2012
p , 0.01. Analysis by age showed that amongst dialysis change during this time in the overall use of growth
and transplanted patients the greatest proportion of hormone with a significant proportion of children
children with a height ,2SD was in the 2–4.99 years under 16 years with a height under 2SD not receiving
age group. growth hormone. Only 29.2% of dialysis patients with
Figure 13.8 shows the use of growth hormone in all a height below the normal range and 11.9% with a
ERF children under 16 years with a height under 2SD functioning transplant who were short received growth
in the UK between 2001 and 2012. There has been little hormone treatment.
271
2
1
Median
0 Lower quartile
Height z-score
–1
–2
–3
–4
–5
–6
–7
L Eve_P
L GOSH_P
Glasg_P
Nottm_P
Livpl_P
Bham_P
Brstl_P
Leeds_P
Manch_P
Soton_P
Blfst_P
Cardf_P
Newc_P
UK
Fig. 13.4. Median height z-scores for
Centre dialysis patients ,16 years in 2012
2
1
0
Weight z-score
–1
–2
–3
Upper quartile N = 144
–4 Median
–5 Lower quartile
–6
L Eve_P
Brstl_P
Cardf_P
Glasg_P
Blfst_P
Bham_P
L GOSH_P
Livpl_P
Soton_P
Nottm_P
Manch_P
Leeds_P
Newc_P
UK
Fig. 13.5. Median weight z-scores for

6
Upper quartile
5
Median N = 140
4
Lower quartile
3
BMI z-score
2
1
0
–1
–2
–3
–4
Newc_P
Leeds_P
Livpl_P
L GOSH_P
Manch_P
Nottm_P
Bham_P
Blfst_P
Soton_P
Glasg_P
L Eve_P
Brstl_P
Cardf_P
UK
Fig. 13.6. Median BMI z-scores for

Blood pressure For children with a functioning kidney transplant,

Analyses of blood pressure levels have shown that blood 76.3% had a systolic BP ,90th percentile which was
pressure was higher in children receiving renal replace- slightly lower than last year when 81.1% of such children
ment therapy than in healthy children (figures 13.9, achieved the target (table 13.6). In comparison, 56.7% of
13.10). There was wide inter-centre variation in systolic children on haemodialysis had a systolic BP ,90th
blood pressure, particularly in dialysis patients. The UK percentile whilst 56.2% of children receiving peritoneal
median z-score was 1.0 for dialysis patients and 0.40 for dialysis achieved this (table 13.6). The results for haemo-
transplant patients. dialysis and peritoneal dialysis were slightly worse than
272
70 those achieved in the previous year (66.7% and 66.2%

Underweight
60 Normal respectively) although absolute numbers were small.
50 Overweight When analysing data by age, blood pressure control

Obese
40 was slightly worse in the 0–4.99 year age group for
30 dialysis patients with little difference noted amongst
20 transplanted age groups.
10
0 Haemoglobin
Transplant Dialysis The analyses in this report show that many children
Treatment modality
receiving dialysis were anaemic, with 25.7% of haemo-
Fig. 13.7. BMI categorisation in children ,16 years by modality dialysis and 17.3% of peritoneal dialysis patients having
in 2012 a haemoglobin level below the standard (table 13.7).
Table 13.5. Percentage of patients aged 2–16 years old with height under 2SDs in 2012∗
Transplant patients Haemodialysis patients Peritoneal dialysis patients
Centre Patients with data (N) % ,2SD Patients with data (N) % ,2SD Patients with data (N ) % ,2SD
Bham_P 57 26.3 10 80.0 10 40.0
Blfst_P 14 42.9 2 50.0 2 50.0
Brstl_P 24 41.7 2 50.0 2 50.0
Cardf_P 16 37.5 1 100.0 1 0.0
Glasg_P 25 8.0 3 33.3 6 33.3
L Eve_P 54 27.8 5 0.0 4 50.0
L GOSH_P 95 21.1 10 40.0 5 20.0
Leeds_P 40 27.5 2 50.0 3 33.3
Livpl_P 18 22.2 2 50.0 1 100.0
Manch_P 20 35.0 5 80.0 7 42.9
Newc_P 21 33.3 1 100.0 1 0.0
Nottm_P 39 30.8 4 25.0 6 50.0
Soton_P 10 50.0 3 33.3 3 66.7
UK 433 27.7 50 50.0 51 41.2
Age group
2–4.99 years 40 35.0 9 77.8 11 54.6
5–11.99 years 203 30.05 23 47.83 20 45.0
12–15.99 years 190 23.68 18 38.89 20 30.0
∗
Preterm children and patients with a syndromic diagnosis were excluded from analyses
45
Transplant patients
40 Dialysis patients
Percentage of patients on GH
35
30
25
20
15
10
0 Fig. 13.8. Use of growth hormone in

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 children ,16 years with a height under
Year 2SD in the UK between 2001 and 2012
273
3
Dotted line shows the 90th centile
2
1
SBP z-score
–1 Upper quartile
Median
–2
Lower quartile N = 514
–3
Brstl_P
Leeds_P
Bham_P
Glasg_P
Cardf_P
Livpl_P
Manch_P
Nottm_P
L GOSH_P
Newc_P
Soton_P
Blfst_P
L Eve_P
UK
Fig. 13.9. Median systolic blood pressure
z-scores for transplant patients ,16 years
Centre in 2012
1
SBP z-score
–1 Upper quartile
Median
–2
Lower quartile N = 140 Dotted line shows the 90th centile
–3
Bham_P
Cardf_P
Leeds_P
Brstl_P
Manch_P
Nottm_P
Blfst_P
Livpl_P
Glasg_P
L Eve_P
Newc_P
Soton_P
L GOSH_P
UK
Fig. 13.10. Median systolic blood

pressure z-scores for dialysis patients
Centre ,16 years in 2012
Table 13.6. Percentage of patients ,16 years achieving the standards for systolic blood pressure in 2012
Patients with Below 90th Patients with Below 90th Patients with Below 90th
Centre data (N) percentile data (N) percentile data (N ) percentile
Bham_P 59 67.8 11 54.6 10 20.0
Blfst_P 20 75.0 2 50.0 3 66.7
Brstl_P 33 57.6 3 33.3 2 0.0
Cardf_P 17 76.5 2 0.0 1 100.0
Glasg_P 27 74.1 4 50.0 9 77.8
L Eve_P 61 95.1 7 100.0 4 75.0
L GOSH_P 106 84.0 15 66.7 10 80.0
Leeds_P 55 49.1 3 33.3 4 0.0
Livpl_P 20 85.0 2 50.0 2 50.0
Manch_P 29 79.3 7 57.1 14 42.9
Newc_P 24 87.5 2 50.0 1 100.0
Nottm_P 49 75.5 4 0.0 10 70.0
Soton_P 14 92.9 5 80.0 3 100.0
UK 514 76.3 67 56.7 73 56.2
Age group
0–4.99 years 49 73.5 20 45.0 29 51.7
5–11.99 years 239 73.6 28 53.6 23 52.2
12–15.99 years 226 79.7 19 73.7 21 66.7
274
Table 13.7. Percentage of patients ,16 years old achieving the haemoglobin standard in 2012
Patients % % lower Patients % % lower Patients % % lower

with data achieving then with data achieving then with data achieving then
Centre (N) standard standard (N) standard standard (N) standard standard
Bham_P 59 91.5 8.5 11 72.7 27.3 10 80.0 20.0
Blfst_P 21 90.5 9.5 3 100.0 0.0 3 100.0 0.0
Brstl_P 35 94.3 5.7 3 33.3 66.7 2 50.0 50.0
Cardf_P 17 94.1 5.9 2 50.0 50.0 1 100.0 0.0
Glasg_P 27 96.3 3.7 4 100.0 0.0 9 88.9 11.1
L Eve_P 62 95.2 4.8 8 100.0 0.0 4 75.0 25.0
L GOSH_P 113 90.3 9.7 15 93.3 6.7 10 90.0 10.0
Leeds_P 57 87.7 12.3 3 33.3 66.7 5 40.0 60.0
Livpl_P 20 95.0 5.0 2 50.0 50.0 2 100.0 0.0
Manch_P 30 93.3 6.7 7 71.4 28.6 15 86.7 13.3
Newc_P 24 91.7 8.3 2 0.0 100.0 1 100.0 0.0
Nottm_P 53 88.7 11.3 5 80.0 20.0 10 80.0 0.0
Soton_P 14 92.9 7.1 5 40.0 60.0 3 100.0 0.0
UK 532 91.7 8.3 70 74.3 25.7 75 82.7 17.3
Age group
0–4.99 years 50 88.0 12.0 22 54.6 45.5 29 86.2 13.8
5–11.99 years 245 91.4 8.6 28 75.0 25.0 24 83.3 16.7
12–15.99 years 237 92.8 7.2 20 95.0 5.0 22 77.3 22.7
This compared to only 8.3% of patients with a function- because of a higher proportion of historical missing data.
ing transplant having haemoglobin below the standard. The attainment of the haemoglobin standard in trans-
Analysis by age showed that the proportion of children plant patients was assessed for different levels of graft
on haemodialysis with haemoglobin below the standard function (figure 13.12) and with the use of MMF as
was greatest for those under five years although this immunosuppressant therapy (figure 13.13). Figure 13.12
was not statistically significant. demonstrates that haemoglobin standard attainment
Figure 13.11 shows that the percentage of dialysis was worse for patients with transplant dysfunction with
patients achieving or exceeding the treatment standards only 79.5% of patients with an eGFR of ,45 achieving
for haemoglobin has increased over the last decade, or exceeding the standard for haemoglobin compared
with little change noted in transplanted patients. Attain- to 95.4% of patients with an eGFR of .60. As for the
ment of ferritin standards are more difficult to interpret impact of MMF, figure 13.12 shows that patients using
100
90
80
70
60
% Dialysis with Hb in range
50
% Transplant with Hb in range
40
30
20
10 Fig. 13.11. The percentage of patients
0 ,16 years achieving the treatment
2001–2003 2004–2006 2007–2009 2010–2012 standard for haemoglobin between
Year group 2001–2012, by treatment modality
275
100 2007–2012, although during this time period there was a

marked rise in missing data for MMF (48% missing data,
80
compared to 14% during earlier years) making it difficult

60
to draw any significant conclusions.
Regarding the use of erythropoietin (ESA) and IV iron,
40 figure 13.14 shows that there has been little change in the
use of these agents in transplanted patients over the last
20 decade; in dialysis patients the use of ESA appears to
0
have stabilised following the initial fall below 90% first
eGFR <45 eGFR 45–60 eGFR >60 observed in 2009. The use of IV iron in dialysis patients
showed a small increase over last year. Table 13.8
Fig. 13.12. The achievement of haemoglobin treatment standards
in paediatric transplant patients ,16 years, by the level of graft shows that the majority of patients on dialysis (with a
function haemoglobin above or below range) were on ESA with
This figures combines all data from 2001–2012. little change over time.
MMF as immunosuppressant therapy were more likely to Phosphate, calcium, PTH and bicarbonate
have haemoglobin concentrations below the standard, In 2012 in the UK as a whole, 50% of haemodialysis
which was statistically significant p , 0.001. Whilst this patients and 56% of peritoneal dialysis patients had a
was noted between 2001–2006, this was not seen between phosphate within the target range (table 13.9). The
100
90
80
70
60
Above standard
50
Below-standard
40
30
20
10
0
On MMF Not on MMF On MMF Not on MMF On MMF Not on MMF On MMF Not on MMF
(n = 145) (n = 851) (n = 190) (n = 676) (n = 324) (n = 423) (n = 296) (n = 349)
2001–2003 2004–2006 2007–2009 2010–2012
Year group
Fig. 13.13. The achievement of haemoglobin treatment standards in paediatric transplant patients ,16 years, by use of MMF between
2001–2012
100
90
% dialysis with data using IViron
80 % dialysis with data using EPO
70 % transplant with data using IViron

% transplant with data using EPO
60
50
40
30
20
10 Fig. 13.14. The use of erythropoietin
0 and IV iron in paediatric patients
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 ,16 years between 2001 and 2012 by
Year treatment modality
276
Table 13.8. Proportion of paediatric RRT patients on ESA, by between assays used at different centres which may
haemoglobin attainment, across time further complicate interpretation of results. No signifi-
Hb below standard Hb above standard cant age related differences were observed.
Time period % on ESA % on ESA For the first time this year, data are presented on the
bicarbonate levels achieved in children on dialysis and
Transplant patients those transplanted (table 13.12). It is important to high-
2001–2003 15.2 3.8
2004–2006 23.2 4.2 light that some centres reported having normal ranges
2007–2009 23.2 6.6 extending below 20 mmol/L. It was observed that more
2010–2012 21.3 6.4 children were acidotic (bicarbonate level ,20 mmol/L)
Dialysis patients on haemodialysis (18.8%) as compared to peritoneal
2001–2003 92.7 89.9 dialysis (2.7%), this perhaps reflects the timing of blood
2004–2006 98.9 93.0 testing performed. Transplanted patients had the highest
2007–2009 95.7 90.6 percentage (92.1%) of patients with a bicarbonate in
2010–2012 82.0 86.8
range (20–30 mmol/L) with 7.2% of patients having a
bicarbonate ,20 mmol/L. No significant age related
achievement of the standard for calcium was better with differences were observed.
80% of children on dialysis (haemodialysis and peritoneal
dialysis) having a calcium level within the target range
(table 13.10). As for PTH, only 43.5% of children on
HD and 30.7% on PD had a PTH within the target Discussion
range with wide inter-centre variation (table 13.11). In
comparison, 84.2% of patients with a functioning trans- This year 92% of data returns were submitted electro-
plant achieved a PTH within the target range. Caution nically with most centres now having electronic systems,
should be exercised in the interpretation of these analyses albeit currently without the facility for automatic data
as these analyses represent measurements performed extraction. As this is developed over the coming years,
once per year per patient. Further, there are differences it will allow downloads of data at multiple time points
Table 13.9. Achievement of the phosphate standard in dialysis patients ,16 years in 2012
Haemodialysis patients Peritoneal dialysis patients
Patients with % within % below % above Patients with % within % below % above
Centre data (N) standard standard standard data (N) standard standard standard
Bham_P 11 36.4 18.2 45.5 10 50.0 0.0 50.0
Blfst_P 3 33.3 33.3 33.3 3 0.0 33.3 66.7
Brstl_P 3 33.3 66.7 0.0 2 50.0 0.0 50.0
Cardf_P 2 0.0 50.0 50.0 1 100.0 0.0 0.0
Glasg_P 4 25.0 0.0 75.0 9 55.6 0.0 44.4
L Eve_P 8 25.0 12.5 62.5 4 100.0 0.0 0.0
L GOSH_P 15 66.7 13.3 20.0 10 50.0 20.0 30.0
Leeds_P 3 66.7 0.0 33.3 5 60.0 0.0 40.0
Livpl_P 2 100.0 0.0 0.0 2 50.0 0.0 50.0
Manch_P 7 57.1 14.3 28.6 15 66.7 0.0 33.3
Newc_P 2 100.0 0.0 0.0 1 100.0 0.0 0.0
Nottm_P 5 80.0 20.0 0.0 10 50.0 0.0 50.0
Soton_P 5 40.0 20.0 40.0 3 33.3 0.0 66.7
UK 70 50.0 17.1 32.9 75 56.0 4.0 40.0
Age group
0–4.99 years 22 54.6 22.7 22.7 29 55.2 0.0 44.8
5–11.99 years 28 50.0 17.9 32.1 24 50.0 8.3 41.7
12–15.99 years 20 45.0 10.0 45.0 22 63.6 4.6 31.8
277
Table 13.10. Achievement of the adjusted calcium standard in dialysis patients ,16 years in 2012
Haemodialysis patients Peritoneal dialysis patients
Patients with % within % below % above Patients with % within % below % above
Centre data (N) standard standard standard data (N) standard standard standard
Bham_P 11 54.6 0.0 45.5 10 80.0 0.0 20.0
Blfst_P 3 66.7 0.0 33.3 3 33.3 0.0 66.7
Brstl_P 3 66.7 0.0 33.3 2 100.0 0.0 0.0
Cardf_P 2 100.0 0.0 0.0 1 100.0 0.0 0.0
Glasg_P 4 50.0 25.0 25.0 9 66.7 0.0 33.3
L Eve_P 8 87.5 12.5 0.0 4 75.0 0.0 25.0
L GOSH_P 15 100.0 0.0 0.0 10 90.0 0.0 10.0
Leeds_P 3 66.7 33.3 0.0 5 100.0 0.0 0.0
Livpl_P 2 100.0 0.0 0.0 2 50.0 0.0 50.0
Manch_P 7 71.4 14.3 14.3 15 80.0 13.3 6.7
Newc_P 2 100.0 0.0 0.0 1 100.0 0.0 0.0
Nottm_P 5 100.0 0.0 0.0 10 80.0 0.0 20.0
Soton_P 5 80.0 20.0 0.0 3 100.0 0.0 0.0
UK 70 80.0 7.1 12.9 75 80.0 2.7 17.3
Age group
0–4.99 years 22 86.4 9.1 4.6 29 79.3 3.5 17.2
5–11.99 years 28 85.7 3.6 10.7 24 83.3 0.0 16.7
12–15.99 years 20 65.0 10.0 25.0 22 77.3 4.6 18.2
per year for each patient allowing more meaningful The data for each section are discussed below, but
analyses. The recently updated NEW paediatric dataset often the results throw up as many questions as they
is now being issued to system providers so that it can answer. There are several areas where more detailed
be incorporated in software upgrades. analysis may help to identify obstacles as to why there
Table 13.11. Percentage of patients ,16 years achieving the PTH standard in 2012
Patients % % Patients % % Patients % %

with data achieving above with data achieving above with data achieving above
Centre (N) standard standard (N) standard standard (N) standard standard
Bham_P 52 61.5 38.5 11 36.4 63.6 10 20.0 80.0
Blfst_P 3 33.3 66.7 3 100.0 0.0
Brstl_P 28 82.1 17.9 3 100.0 0.0 2 50.0 50.0
Cardf_P 17 82.4 17.7 2 50.0 50.0 1 0.0 100.0
Glasg_P 27 96.3 3.7 4 0.0 100.0 9 33.3 66.7
L Eve_P 60 90.0 10.0 8 50.0 50.0 4 25.0 75.0
L GOSH_P 109 84.4 15.6 15 53.3 46.7 10 50.0 50.0
Leeds_P 3 33.3 66.7 5 40.0 60.0
Livpl_P 18 100.0 0.0 2 50.0 50.0 2 100.0 0.0
Manch_P 30 93.3 6.7 7 42.9 57.1 15 6.7 93.3
Newc_P 22 100.0 0.0 2 0.0 100.0 1 0.0 100.0
Nottm_P 31 83.9 16.1 5 60.0 40.0 10 30.0 70.0
UK 430 84.2 15.8 69 43.5 56.5 75 30.7 69.3
Age group
0–4.99 years 45 91.1 8.9 22 22.7 77.3 29 37.9 62.1
5–11.99 years 192 83.9 16.2 27 63.0 37.0 24 33.3 66.7
12–15.99 years 193 82.9 17.1 20 40.0 60.0 22 18.2 81.8
∗
Blank cells denote modalities where data completeness was ,50%
278
Table 13.12. Centre analysis of bicarbonate levels (mmol/L) in patients under 16years old by treatment modality, in 2012
Patients Patients Patients

with data % % % with data % % % with data % % %
Centre (N) ,20 20–30 .30 (N) ,20 20–30 .30 (N) ,20 20–30 .30
Bham_P 59 8.5 89.8 1.7 11 18.2 81.8 0.0 10 0.0 80.0 20.0
Blfst_P 21 0.0 100.0 0.0 2 0.0 100.0 0.0 3 0.0 100.0 0.0
Brstl_P 35 2.9 97.1 0.0 3 33.3 66.7 0.0 2 0.0 100.0 0.0
Cardf_P 17 23.5 76.5 0.0 2 0.0 100.0 0.0 1 0.0 100.0 0.0
Glasg_P 27 37.0 63.0 0.0 4 100.0 0.0 0.0 9 22.2 66.7 11.1
L Eve_P 62 6.5 93.6 0.0 8 25.0 75.0 0.0 4 0.0 100.0 0.0
L GOSH_P 113 2.7 96.5 0.9 15 13.3 80.0 6.7 10 0.0 90.0 10.0
Leeds_P 53 0.0 100.0 0.0 3 0.0 100.0 0.0 5 0.0 80.0 20.0
Livpl_P 20 40.0 60.0 0.0 2 0.0 100.0 0.0 2 0.0 100.0 0.0
Manch_P 30 3.3 93.3 3.3 7 28.6 71.4 0.0 14 0.0 85.7 14.3
Newc_P 24 8.3 91.7 0.0 2 0.0 100.0 0.0 1 0.0 100.0 0.0
Nottm_P 53 0.0 98.1 1.9 5 0.0 80.0 20.0 10 0.0 70.0 30.0
Soton_P 14 0.0 100.0 0.0 5 0.0 100.0 0.0 3 0.0 100.0 0.0
UK 528 7.2 92.1 0.8 69 18.8 78.3 2.9 74 2.7 83.8 13.5
Age group
0–4.99 years 50 12.0 88.0 0.0 22 13.6 86.4 0.0 29 3.5 72.4 24.1
5–11.99 years 244 6.6 92.2 1.2 27 18.5 77.8 3.7 23 4.4 87.0 8.7
12–15.99 years 234 6.8 92.7 0.4 20 25.0 70.0 5.0 22 0.0 95.5 4.6
,20 mmol/L was defined as being acidotic, although it is worth noting some centres report having normal ranges extending below 20
has been little apparent change in the attainment of many with ERF for a significant part of their childhood are
standards over the last few years. more likely to have impaired growth than those who
have had better health for part of their childhood and
Anthropometry may be part of the explanation.
Children on renal replacement therapy are short for There have been initiatives to try and improve growth,
their age. Excluding children and young people with such as using rhGH, improved nutrition and avoiding the
syndromes and those born prematurely, who are more use of steroids post transplant. Just under a third of dialy-
likely to be short, just over a quarter of transplant sis patients, and 11.9% of transplant patients, who were
patients, 50% of HD patients and 41% of PD patients short for their age, were on growth hormone treatment.
had a height that was below the normal range. The figures The low uptake of rhGH within the UK ERF population
would be lower if all children on RRT were included. where overall 32.8% of patients have a height below the
Children aged less than five years who were on dialysis normal range, remains disappointing. However in the
seemed to be most affected. Growth in the pre-school transplant group it is important to remember that these
years is faster than in later years and so it is not surprising data are cross-sectional and although some children are
that dialysis at this age can have a deleterious effect on short, they may be growing at a rate above normal and
growth. It is a sobering thought that nearly half of therefore would not fall into the category for whom
children on dialysis have a height below the normal rhGH is appropriate.
range. Whilst transplantation improves the situation, a The use of steroids post-transplant can affect growth
quarter remain short. and varies from centre to centre. It would be interesting
The cross-sectional data presented here are little differ- to compare those centres which avoid steroids to those
ent from previous reports; indeed there appears to have where steroids are used as standard post-transplant.
been little change since 1999 which is disappointing [7]. Furthermore, it may be that many different factors not
There may be a number of reasons for this. Over the included here have an influence on growth and that
last few years, there has been an increase in the number further in depth studies will highlight these. There is
of infants and young children receiving RRT. Children therefore scope to increase the use of rhGH in these
279
patients. An analysis evaluating final adult height may Anaemia

add to our understanding. The proportion of short trans- A significant proportion of dialysis patients (25.7%
planted children varied by centre and it would be inter- HD, 17.3% PD) were anaemic; this is little changed
esting to see if this relates to the centres’ likelihood of from previous reports. The proportion of transplant
using steroids post transplant. patients with a haemoglobin within the recommended
In this report for the first time, BMI based on height- range however has improved and is due to the change
age as opposed to their chronological age is reported in standard used.
which is more appropriate given a cohort of children For transplant patients, the chances of a haemoglobin
who have growth restriction. Overweight and obesity level below the standard were greater with reduced GFR
are also defined as per IOTF definitions. These definitions and with the use of MMF. This highlights the importance
are different to those used in previous reports and likely of calculating GFR for transplant patients, rather than
to account for the small differences in reported data. using creatinine alone. A lower GFR should highlight
Overall, little change in weight SDs and BMI SDs since the need to check that the haemoglobin is within the
1999 in both transplanted children and those on dialysis recommended range. Since 2000, the proportion of
were observed. Recent reports from the ERA-EDTA patients with a haemoglobin within range who were on
Registry [8] highlight the high prevalence rates of excess MMF has increased and remained stable in this year’s
weight in UK children following renal transplantation. report.
Furthermore, a report from the BAPN analysing the Whilst there are indicators to help identify those trans-
longitudinal change in BMI following transplantation plant patients at risk of anaemia, it is more difficult to
highlights rates of excessive weight (overweight and highlight those at risk within the dialysis populations.
obese) significantly worse than the background UK As expected patients on HD seem more at risk and the
childhood population [9]. These data together highlight risk of anaemia may be higher for those aged less than
the need for urgent work to understand factors that five years. Of those with a haemoglobin below range,
lead to excess weight gain in this high risk cohort for over 90% of patients were on ESAs, although the pro-
adverse cardiovascular outcomes. portion on IV iron or with a low ferritin was less clear.
Of transplant patients with a low haemoglobin, 21%
Blood pressure were on ESAs compared with 15% between 2001–2003.
There is an increasing body of evidence supporting the It is important to highlight here that it is beyond the
role of optimal blood pressure control in the manage- scope of the registry to be able to report on dose adjust-
ment of CKD [10, 11]. There is also an increasing ments that would likely improve understanding of these
awareness of the importance of cardiovascular morbidity data. It would be helpful to study dialysis patients in
in paediatric patients with CKD and ERF. Despite this, more detail to see if there are any factors which help
there remains scope for improvement in BP control. As identify those children at highest risk of anaemia.
BP changes during childhood, it is important to manage Detailed data on ferritin and IV iron would be needed
blood pressure using percentiles in the clinic rather than for this subgroup of patients. The results of the national
using the absolute measurements alone. The authors audit on anaemia in the UK paediatric ERF population
hope that it may be possible at some point to include may help to shed some further light on this.
the degree of proteinuria for transplant patients in the
analysis. Biochemistry
There was a wide range of median systolic BP scores The numbers of paediatric patients on dialysis were
in different centres and it might be helpful to reflect on small but phosphate control appears to be worse in
the different strategies in each centre and their effect on patients on HD than in patients on PD. Results for cal-
outcomes. It is hoped that the clinical application of cium were little different between the dialysis groups,
recently developed guidelines by the BAPN for the whilst patients on PD had worse PTH concentrations
management of hypertension following transplantation than those on haemodialysis. Data were less complete
would help in improving blood pressure control [12]. for PTH in the transplant group which might imply
Once again the authors would highlight that these that the complications of reduced GFR might sometimes
data reflect single measurements per year often per- be overlooked in this group of patients. It would be useful
formed using BP instruments that employ different to include vitamin D (calcidiol) concentrations in the
techniques. parameters studied. Moving to multiple time point
280
reporting of data in future reports will allow better platforms have the additional potential to display percen-
interpretation of biochemistry results. A higher propor- tiles and SDs and it may be that these functionalities will
tion of subjects on HD were acidotic compared to those help make clinicians aware of patient’s results and
on PD, with the best results in transplanted patients. achievement of targeted clinical standards. Automatic
calculations of e.g. eGFR in transplant patients may
help to point out that some patients have lower GFRs
that make them susceptible to anaemia. The likelihood
Summary of complete electronic reporting in the near future with
plans for quarterly reporting in the format of the recently
In summary, continued efforts are being made to finalised NEW paediatric dataset will undoubtedly
move towards universal electronic reporting from UK improve quality of data and their reporting, allowing
paediatric centres. Whilst this is ongoing, most centres improvements in patient care.
are moving to using electronic systems which incorporate
an electronic patient record. These improved electronic Conflicts of interest: none
References
1 Hussain F, Castledine C, Schalkwyk DV, Sinha MD, Lewis MA, Inward 8 Bonthuis M, van Stralen KJ, Verrina E, Groothoff JW, Alonso Melgar A,
C. UK Renal Registry 12th Annual Report (December 2009): Chapter 12 Edefonti A, Fischbach M, Mendes P, Molchanova EA, Paripovic D,
Clinical, Haematological and Biochemical Parameters in Patients Peco-Antic A, Printza N, Rees L, Rubik J, Stefanidis CJ, Sinha MD,
receiving Renal Replacement Therapy in Paediatric Centres in the UK Zagozdzon I, Jager KJ, Schaefer F. Underweight, overweight and obesity
in 2008: national and centre-specific analyses. Nephron Clin Pract in paediatric dialysis and renal transplant patients. Nephrol Dial Trans-
2010;115(suppl 1):c289–c308 plant. 2013; doi: 10.1093/ndt/gft259. [Epub ahead of print]
2 TJ Cole, KM Flegal, D Nicholls, AA Jackson. Body Mass Index cut offs to 9 Plumb LA, Pitcher D, Tse Y, Shield JP, Inward C, Sinha MD on behalf of
define thinness in children and adolescents: international study. BMJ the British Association for Paediatric Nephrology. Longitudinal changes
2007;335(7612):194 in Body Mass Index following renal transplantation in UK children.
3 Freeman JV CT, Chinn S et al. Cross sectional stature and weight Nephrol Dial Transplant. 2013; doi: 10.1093/ndt/gft395. [Epub ahead
reference curves for the UK, 1990. Arch Dis Child 1995;73:17–24. of print]
4 National High Blood Pressure Education Program Working Group on 10 Strict blood-pressure control and progression of renal failure in children,
High Blood Pressure in Children and Adolescents.The Fourth Report ESCAPE Trial Group, N Engl J Med. 2009;361(17):1639–50
on the Diagnosis, Evaluation, and Treatment of High Blood Pressure 11 Sinha MD, Gilg JA, Kerecuk L, Reid CJD; on behalf of the British
in Children and Adolescents. Pediatrics 2004;114(2):555–76. Association for Paediatric Nephrology. Progression to hypertension in
5 BAPN clinical standards http://www.bapn.org/clinical_standards.html. non-hypertensive children following renal transplantation. Nephron
6 NICE clinical guideline 114. Anaemia management in people with Dial Transp 2012;27(7):2990–2996
chronic kidney disease. London: National Institute for Health and 12 BAPN Standards for Hypertension in Paediatric Renal Transplant
Clinical Excellence, 2011. Recipients. http://www.bapn.org/assets/clinical_standards/BAPN%
7 Pruthi R, Sinha MD, Casula A, Tse Y, Maxwell H, O’Brien C, Lewis M, 20standards%20for%20Hypertension%20in%20Renal%20Transplant%
Inward C. UK Renal Registry 14th Annual Report (December 2010): 20Recipients.pdf (last accessed 22nd September 2013)
Chapter 5 Demography of the UK Paediatric Renal Replacement
Therapy Population in 2010. Nephron Clin Prac 2012;120(suppl 1):
c93–c103
281
Chapter 14 2012 Multisite Dialysis Access
Audit in England, Northern Ireland and
Wales and 2011 PD One Year Follow-up:
Victoria Briggsa, David Pitcherb, Catriona Shawb, Richard Fluckc, Martin Wilkiea
a
Sheffield Teaching Hospitals NHS Foundation Trust, UK; bUK Renal Registry, Bristol, UK; cDerby Hospitals Foundation Trust, UK
. Initial surgical assessment was a key determinant of

Key Words
the likelihood of AVF formation; 70.4% of patients
Access . Chronic kidney disease . Diabetes . Dialysis . End
assessed by a surgeon at least three months before
stage renal disease . Established renal failure . Haemodialysis commencing dialysis started on an AVF. By con-
. Peritoneal dialysis . Prevalence . Primary Care Trust . Renal
trast, only 9.7% of patients not surgically assessed
replacement therapy . Transplantation . Treatment modality at least three months before commencing dialysis
used an AVF as first dialysis access.
. Length of time known to nephrology services and
Summary likelihood of commencing dialysis using either an
AVF or a PD catheter are strongly associated. For
. Data are presented from the first combined vascular patients presenting late, 84.6% started on a line
and peritoneal dialysis access audit. (TL/NTL). Amongst patients known to the centre
. In 2012, 51 centres in England, Wales and Northern for at least a year only 33.9% started via a line.
Ireland (representing 82% of all centres) returned . Data on PD catheter failure rates at one year were
data on first access from 3,720 incident haemodialy- poorly completed. Of 44 centres who reported
sis (HD) patients and 1,018 incident peritoneal data on PD patients in 2011, only 28 completed
dialysis (PD) patients. the one year follow up request, returning data on a
. Of the incident HD patients, 38.3% started therapy total of 649 patients.
on an arteriovenous fistula (AVF), 36.9% on a tun- . For centres returning data on one year peritoneal
nelled line (TL), 23.5% on non-tunnelled line (NTL) dialysis outcomes, the majority of centres maintained
and 1.2% by means of arteriovenous graft (AVG). .50% of patients on PD at one year, however only
. Referral time had an influence on PD catheter five centres maintained .80% on PD at one year.
insertion technique: of patients starting PD within . Further enhancement of data fields, improved data
90 days of initial referral, 50.6% underwent percuta- completeness and accuracy of returns will be essen-
neous PD catheter insertion. This contrasts with tial to improve the quality of future audits.
patients known to renal services in excess of 90 . Further work is required to define optimal dialysis
days, 32.4% of whom underwent percutaneous PD access care pathways that are comprehensive, high
catheter insertion. quality and responsive to patient needs.
283
Introduction operational effectiveness (surgical referral, conversion

rates between access types) and documentation of com-
This report represents the first combined vascular and plications continue to be the main endpoints of this
peritoneal dialysis access audit in England, Wales and joint access audit.
Northern Ireland. Previously, vascular and peritoneal There is substantial evidence to suggest that prompt
dialysis access audits have been published separately [1, permanent vascular access is clinically advantageous.
2, 3]. Indeed, current best practice indicates that vascular access
Dialysis access (regardless of modality) should be should be in place by a minimum of six months before
timely, minimise complications and maintain functional- starting treatment [6]. Observational data has repeatedly
ity for as long as it is required. Both haemodialysis (HD) demonstrated a strong association between the use of
and peritoneal dialysis (PD) require good functional central venous catheters and increased mortality and
access in order for the renal replacement technique to morbidity [2, 7]. Similarly, patients presenting late com-
be successful. mencing dialysis via a PD catheter rather than a tunnelled
The Department of Health National service framework line are also less likely to experience bacteraemia [8].
for Renal Services 2004 [4] states that by 2014: Whilst this, in part, may reflect late presentation and
co-morbidity, studies attempting to correct for this
‘All children, young people and adults approaching
have identified an independent effect of access on patient
established renal failure are to receive timely preparation
outcomes [7, 9]. Permanent vascular access delivers a
for renal replacement therapy so the complications and
higher, more effective dialysis dose, and those with
progression of their disease are minimised, and their
venous catheters may require an increase in frequency
choice of clinically appropriate treatment options is
and duration of dialysis to compensate. Permanent
maximised.’
vascular access will also remain functional for much longer
‘All children, young people and adults with estab-
than a venous catheter, requiring fewer hospital admis-
lished renal failure are to have timely and appropriate
sions with attendant health economic benefits [7, 9].
surgery for permanent vascular or peritoneal dialysis
The provision of high quality PD access is equally
access, which is monitored and maintained to achieve
important. The National Institute of Health and Clinical
its maximum longevity.’
Excellence (NICE) has recommended that PD should be
Previously reported vascular access and peritoneal offered as a first-line therapy for the majority of patients
access audits [1, 2, 3] have therefore been performed with established renal failure (ERF) on the basis of
with the intention of providing clinically useful infor- equivalent outcomes with haemodialysis [10]. Despite
mation relating to timely and appropriate access inter- this guidance, PD is only used for 20% of UK dialysis
ventions in order to achieve permanent access based on patients. Furthermore, the UK Renal Registry (UKRR)
these recommendations and quality requirements. The 2012 annual report documents a 10–fold national vari-
core principal of these audits has been to highlight the ation in PD utilisation between otherwise similar renal
performance variation of renal centres across England, clinical centres [11].
Wales and Northern Ireland and explore factors that The term established renal failure used within this
may contribute to the provision of excellent quality chapter is synonymous with the terms end stage renal
vascular and peritoneal access. failure and end stage renal disease, which are in more
High quality vascular access represents a key modifiable widespread international usage. Patients have disliked
risk factor for patients on dialysis and is an important the term ‘end stage’ which reflects the inevitable outcome
measure of clinical care [5]. Whilst it is possible to of this disease.
postulate plausible factors that influence access provision, The PD audit work was supported by funding from
such as variation in patient demographics and physician the Healthcare Quality Improvement Partnership (HQIP).
attitudes, the exact reasons for such variations are
unknown. Audit is essential to define relevant issues
relating to HD access formation and PD catheter inser-
tion, and to understand practice variation with the aim Methods
of standardising the provision of a high-quality service
to all patients who require it. Determination of the type All adult renal centres in England, Wales and Northern Ireland
of access first used for dialysis, investigation of were contacted regarding vascular and peritoneal access for all
284
Chapter 14 Multisite dialysis access audit
incident dialysis patients in 2012. Data were collected using Results

Microsoft Excel spreadsheets circulated by the UK Renal Registry.
Of 62 centres contacted, data were received from 51 centres. Data Data completeness
fields were refined from the audit performed in 2011 based on the
quality of the returned questionnaires and the feedback received Fifty-one centres returned data on first dialysis access
from centres. on 3,720 incident HD patients and 1,018 incident PD
Patients who were identified by the renal centres as having patients. The UKRR incident patient data for the same
acute kidney injury (AKI) in the free text fields or patients who year were 3,818 HD and 1,035 PD, thus there were access
were reported to have recovered renal function within three returns on 97% of HD and 98% of PD patients.
months were categorised as having AKI for the purposes of this
audit and excluded (n = 367/5,105). The remaining records Forty-one patients were excluded from all the analyses
received were validated against the UKRR database to confirm due to missing RRT start date or first access type.
that the population collected at each centre for the audit was the Figure 14.1 illustrates the data completeness for key
same as, or representative of, the incident population at that variables.
centre as collected via the usual UKRR methodology. Data
checks were made by cross-referencing with the UKRR database.
Any patients identified from the UKRR as not incident to Variations in first dialysis access
dialysis between 1st January 2012 and 31st December 2012 were Patient demographics
excluded. The cross-referencing also enabled ascertainment of The median patient age when starting RRT was 67
information on mortality within three months of commencing years in the HD cohort and 59 years for patients com-
dialysis.
mencing PD. Overall, 62.6% of the patients were male,
Centres who reported data on PD patients in the 2011 vascular
and peritoneal access audit were asked to complete a one year 37.4% female; the proportional distribution of the sexes
follow up of their PD patients. Additional information was was similar for both the HD and PD subgroups.
requested on the date of PD catheter failure, the reason for catheter A significant proportion of patients starting dialysis had
failure, the number of catheters used during the year, and the diabetes (43.0%), however diabetes associated nephropa-
modality in use at one year after starting PD.
Patients starting HD were grouped by type of first vascular
thy was the primary renal disease (PRD) in only 26.1%
access: arteriovenous fistula (AVF), arteriovenous graft (AVG), (table 14.1). There was however, a large volume of missing
tunnelled dialysis line (TL), non-tunnelled dialysis line (NTL). data relating to diabetes status (1,144 patients on HD
Patients starting PD were categorised by the insertion technique: (31.1%) and 204 patients on PD (20.1%)).
laparoscopic, peritoneoscopic, open surgery, percutaneous. Access Table 14.2 presents HD and PD patient subgroups
at three months was defined as the type of access in use at three
months after starting dialysis. If a patient was no longer receiving
stratified by age, gender, dichotomised body mass index
dialysis at three months then the reason was recorded instead, for (BMI) (,30 or 530), PRD, referral time (,90 or 590
example died or transplanted. Referral time was defined as the days) and surgical assessment status.
number of days between the date of first being seen by a renal There was an apparent association between the access
physician and the date of commencing dialysis. A patient was modality (HD vs. PD), referral time (,90 days vs. 590
classified as presenting late if they had a referral time of less
than 90 days. In the analyses involving whether or a not a patient
days) and surgical assessment status in excess of three
had received surgical assessment at least three months before months prior to dialysis start. The following observations
starting dialysis, patients were excluded if they were categorised can be made:
as a late presenter. For HD:
Access failure was defined as the access no longer being
usable for treatment. Data about the date and cause of access
failure were collected. Access failure was censored for death,
. AVF was the initial access for 38.3% of patients, with
transplantation, withdrawal from renal replacement therapy 1.2% on an AVG, 36.9% on a tunnelled line and
(RRT) and elective switching of access type. It was the intention 23.5% on a non-tunnelled line.
to only capture access failures relating to the first type of access. . Patients aged 60 or over were more likely to initiate
If the reason recorded for access failure was incompatible with RRT on an AVF (40.7%) when compared to patients
the first type of access recorded then the data was not included
in this analysis.
,60 years (33.9%). Similarly, older patients were
Separate or combined analyses have been performed for less likely to start on a tunnelled line (33.3% vs.
incident HD patients and incident PD patients as appropriate. 43.7%).
Due to the exploratory nature of the audit the analyses have . Patients with polycystic kidney disease (PKD) as
been limited to descriptive statistics of frequencies, percentages and primary renal diagnosis were most likely to start
unadjusted associations between variables. If a centre had more
than 50% missing returns for a particular data field, then all
on an AVF (65.5%).
patients from the centre were excluded from analyses involving . Patients who had been seen by a surgeon at least
that data field. The data were analysed using SAS 9.3. three months before starting dialysis were more
285
DOB Incident HD
Incident PD
Gender
Postcode
PRD
BMI
First seen date
Assessed by surgeon
Data field
RRT start date

First modality
First access
Access at 3 months
Date of 1st access
Diabetes Fig. 14.1. Data completeness for key
Insertion technique variables, stratified by first modality
HD = haemodialysis; PD = peritoneal
Peritonitis
dialysis; DOB = date of birth; PRD =
0 10 20 30 40 50 60 70 80 90 100 primary renal diagnosis; BMI = body mass
Percentage index
likely to start on an AVF than those not assessed techniques, 34.6% using percutaneous techniques
(67.7% vs. 5.6%). and only 3.0% inserted using a peritoneoscope.
. Of those referred at least 90 days prior to commen- . Patients who were assessed by a surgeon at least
cing dialysis, 50.1% started on an AVF compared to three months before starting dialysis were more
only 4.3% of those starting more acutely. likely to undergo laparoscopic placement (24.4%
vs. 5.9% for non-surgical assessment) and were
For PD:
less likely to have open surgical placement (36.8%
. PD catheters were placed in 44.4% of patients by using vs. 55.6%) or percutaneous catheter placement
open surgical techniques, 18.1% using laparoscopic (33.4% vs. 37.6%).
Table 14.1. Patient demographics
Total HD PD
N = 4,697 N = 3,682 N = 1,015
Med (IQR) Med (IQR) Med (IQR)

Age 65 (52, 75) 67 (54, 76) 59 (47, 71)
BMI 27 (24, 32) 27 (23, 32) 27 (24, 31)
N (%) N (%) N (%)
Gender Female 1,759 (37.4) 1,372 (37.3) 387 (38.1)
Male 2,938 (62.6) 2,310 (62.7) 628 (61.9)
Diabetes Missing 1,348 (28.7) 1,144 (31.1) 204 (20.1)
Yes 2,018 (43.0) 1,503 (40.8) 515 (50.7)
No 1,331 (28.3) 1,035 (28.1) 296 (29.2)
PRD Diabetes 1,227 (26.1) 980 (26.6) 247 (24.3)
Glomerulonephritis 610 (13.0) 446 (12.1) 164 (16.2)
Hypertension 374 (8.0) 289 (7.8) 85 (8.4)
Other 784 (16.7) 654 (17.8) 130 (12.8)
Polycystic kidney 257 (5.5) 171 (4.6) 86 (8.5)
Pyelonephritis 274 (5.8) 209 (5.7) 65 (6.4)
Renovascular disease 298 (6.3) 251 (6.8) 47 (4.6)
Uncertain aetiology 693 (14.8) 521 (14.1) 172 (16.9)
Missing 180 (3.8) 161 (4.4) 19 (1.9)
IQR = interquartile range; BMI = body mass index; PRD = primary renal diagnosis; HD = haemodialysis; PD = peritoneal dialysis
286
Table 14.2. Patient characteristics stratified by type of first dialysis access
% of HD patients % of PD patients
HD PD Open Laparo- Peritoneo- Percuta-

Variable N AVF AVG TL NTL N∗ surgery scopic scopic neous
Total patients 3,682 1,412 46 1,358 866 813 361 147 24 281
% 38.3 1.2 36.9 23.5 44.4 18.1 3.0 34.6
Age at first dialysis ,60 1,269 33.9 1.1 43.7 21.4 421 43.7 18.3 3.3 34.7
560 2,413 40.7 1.3 33.3 24.7 392 45.2 17.9 2.6 34.4
BMI (kg/m2) 430 1,056 42.8 1.3 32.9 23.0 263 58.9 11.8 7.2 22.1
.30 432 53.2 2.3 29.6 14.8 97 76.3 8.2 3.1 12.4
PRD Diabetes 980 41.4 1.7 39.3 17.6 202 44.1 18.8 1.0 36.1
GN 446 39.5 0.2 37.0 23.3 131 44.3 17.6 4.6 33.6
Hypertension 289 48.4 1.0 34.9 15.6 64 42.2 25.0 4.7 28.1
Other 654 21.4 1.1 42.2 35.3 111 46.8 18.0 3.6 31.5
PKD 171 65.5 1.2 25.7 7.6 72 45.8 9.7 4.2 40.3
Pyelo 209 40.2 3.3 35.9 20.6 45 42.2 11.1 2.2 44.4
RVD 251 37.8 0.0 33.9 28.3 41 46.3 24.4 4.9 24.4
Uncertain 521 43.6 1.2 33.2 22.1 133 45.9 14.3 2.3 37.6
Referral time (days) ,90 853 4.3 0.6 48.5 46.5 85 22.4 23.5 3.5 50.6
590 2,538 50.1 1.3 33.6 15.0 720 47.4 17.4 2.9 32.4
Assessed by surgeon No 1,435 5.6 0.3 53.5 40.6 306 55.6 5.9 1.0 37.6
Yes 1,690 67.7 2.0 21.4 8.9 386 36.8 24.4 5.4 33.4
∗
PD patients with missing insertion technique are excluded
Patients from centres with more than 50% missing data for a variable are excluded from the table for that variable
AVF = arteriovenous fistula; AVG = arteriovenous graft; TL = tunnelled line; NTL = non-tunnelled line; GN = glomerulonephritis; BMI = body
mass index; PRD = primary renal diagnosis; GN = glomerulonephritis; PKD = polycystic kidney disease; Pyelo = pyelonephritis; RVD = reno-
vascular disease
. Referral time had an influence on PD catheter inser- cohort of patients who present later and in whom a PRD
tion technique; 50.6% of patients referred less than cannot be ascertained.
90 days before starting dialysis underwent percuta- Patients with body mass index (BMI) .30kg/m2 were
neous insertion compared to 32.4% of patients more likely to undergo open surgical placement (76.3%)
known longer to the service. These data were than those with BMI 430kg/m2 (58.9%) (figure 14.3).
reversed for general surgical insertion: 22.4% of The percutaneous approach was nearly half as likely to
patients who presented late versus 47.4% of patients be used in patients in the higher BMI category (12.4%)
who did not present late. compared with those with a lower BMI (22.1%). Equally,
peritoneoscopic placement in the higher BMI category
The proportional distribution of HD access modality was 50% less likely than in the lower BMI group (3.1%
was similar for different primary renal disease diagnoses vs. 7.2%). It should be noted that the analysis was limited
(figure 14.2). Of note, patients with polycystic kidney due to a high proportion of missing data for BMI.
disease were more likely to start HD on an AVF. This Patients aged less than 60 at the point of commencing
likely results from the opportunity for timely access RRT were less likely than older patients to start dialysis
preparation as these patients are often known to renal using an AVF (33.9% vs. 40.7%) (figure 14.4). The reason
services for many years before dialysis is required and for this is unknown but may reflect patient engagement
indeed there is also evidence of a higher transplantation with renal services or varying progression of chronic
rate amongst this group [12]. Where no primary renal kidney disease in the older population [13, 14, 15]. Simi-
diagnosis was available (either missing or coded as larly, utility of non-tunnelled lines was lower in younger
uncertain aetiology), the numbers of patients starting dialysis patients (21.4% vs. 24.7%) in contrast to the use
dialysis with a tunnelled or non-tunnelled dialysis venous of tunnelled lines which were more common in those
catheter were higher, suggesting that this may represent a aged less than 60 (43.7% vs. 33.3%).
287
AVF
Other (654)
AVG
TL
Diabetes (980)
NTL
GN (446)
Primary renal diagnosis
Pyelonephritis (209)
Hypertension (289)
Fig. 14.2. Type of haemodialysis access
RVD (251) stratified by primary renal disease
Number of patients in each primary renal
No data (161) diagnosis group in brackets
Primary renal diagnosis groups sorted by
Uncertain (521) percentage of tunnelled lines
AVF = arteriovenous fistula; AVG =
PKD (171) arteriovenous graft; TL = tunnelled line;
NTL = non-tunnelled line; GN =
0 20 40 60 80 100 glomerulonephritis; RVD = reno-vascular
Percentage disease; PKD = polycystic kidney disease
First dialysis access and renal centre Use of a PD catheter as first access varied between
Large variations were apparent between centres when 44.4% (Wolverhampton) and 0% (Colchester) (figure 14.5).
considering patients commencing dialysis via an AVF Centres that had high usage of AVFs as starting access
(figure 14.5). At one end of the spectrum was Ulster were also more likely to start patients on a PD catheter.
who reported a total of 27 patients with 7.4% starting There was some evidence (p = 0.02) that the proportion
on an AVF, 0% on an AVG, 48.1% starting on a tunnelled of HD patients starting on an AVF increased as the
line, 33.3% using a non-tunnelled line and 11.1% PD proportion of dialysis patients starting on PD increased.
catheter. In contrast, Liverpool Aintree reported a total This may indicate variation in local processes for access
of 57 patients with 54.4% using an AVF, 3.5% on an planning and delivery.
AVG, 5.3% using a tunnelled line, 15.8% on a non-tun- The current audit question asked centres to report
nelled line and 21% on a PD catheter. which type of access was used for the first ever dialysis
session. The problem with this audit question is that
100
100
80 NTL
TL
AVG
80 AVF
60
Percentage
60
Percentage
40 Percutaneous
Peritoneoscopic
Laparoscopic 40
20 Open surgery
20
0
2 2
<30 kg/m >30 kg/m No data
(263 patients) (97 patients) (82 patients)
0
BMI group <60 years >60 years
(1,269 patients) (2,413 patients)
Fig. 14.3. Method of PD catheter insertion stratified by body Age at start of HD
mass index
BMI = body mass index Fig. 14.4. Type of haemodialysis access stratified by age group
All patients from centres with more than 50% missing data for BMI AVF = arteriovenous fistula; AVG = arteriovenous graft; TL =
were excluded tunnelled line; NTL = non-tunnelled line
288
Liv Ain (57) AVF

Newry (18) AVG
Nottm (86) TL
NTL
Swanse (111)
PD catheter
Donc (39)
Exeter (132)
Derby (77)
Redng (78)
B Heart (98)
M RI (71)
Brightn (131)
Wirral (33)
Basldn (54)
Wrexm (27)
L Kings (116)
Antrim (21)
York (51)
Leeds (131)
L St.G (54)
Belfast (76)
Truro (47)
West NI (14)
Ports (149)
Sheff (139)
Glouc (73)
Centre
L Rfree (198)
Oxford (148)
Norwch (67)
Leic (206)
Clwyd (22)
Plymth (33)
Liv RI (76)
Salford (117)
Bristol (162)
Sthend (24)
Bradfd (66)
Stoke (72)
Hull (94)
Prestn (121)
Sund (68)
Newc (96)
Wolve (81)
L Barts (280)
Chelms (59)
Middlbr (106)
Cardff (145)
Bangor (19)
Ulster (27)
B QEH (189)
Colchr (28)
L West (310)
Total (4,697)
0 20 40 60 80 100
Percentage
Fig. 14.5. Type of first dialysis access stratified by centre

Centres are ordered by the percentage of patients using a tunnelled line
AVF = arteriovenous fistula; AVG = arteriovenous graft; TL = tunnelled line; NTL = non-tunnelled line; PD = peritoneal dialysis
289
Antrim (16)
B Heart (79)
B QEH (160)
Bangor (13)
Basldn (40)
Belfast (67)
Bradfd (55)
Brightn (89)
Bristol (144)
Cardff (120)
Chelms (52)
Clwyd (18)
Colchr (28)
Derby (48)
Donc (32)
Exeter (114)
Glouc (56)
Hull (58)
L Barts (207)
L Kings (87)
L Rfree (141)
L St.G (46)
L West (288)
Leeds (107)
Leic (167)
Centre
Liv Ain (45)

Liv RI (52)
M RI (39)
Middlbr (99)
Newc (73)
Newry (12)
Norwch (47)
Nottm (48)
Oxford (112)
Plymth (25)
Ports (115)
Prestn (108)
Redng (57)
Salford (87)
Sheff (115)
Sthend (22)
Stoke (58)
Sund (61)
Swanse (76) AVF
Truro (40) AVG
Ulster (24) TL
West NI (10) NTL
Wirral (24)
Wolve (45)
Wrexm (19)
York (37)
Total (3682)
0 20 40 60 80 100
Percentage
Fig. 14.6. Type of first access for haemodialysis stratified by centre

AVF = arteriovenous fistula; AVG = arteriovenous graft; TL = tunnelled line; NTL = non-tunnelled line
290
Antrim (5) Open surgery

Brightn (42) Laparoscopic
L West (22) Peritoneoscopic
Percutaneous
Sthend (2)
Missing data
Wolve (36)
Salford (30)
L Kings (29)
Derby (29)
B Heart (19)
Liv RI (24)
Leic (39)
Plymth (8)
Redng (21)
Stoke (14)
Ports (34)
Glouc (17)
Bangor (6)
Liv Ain (12)
Belfast (9)
Basldn (14)
Donc (7)
Hull (36)
L St.G (8)
Leeds (24)
Newry (6)
Centre
Prestn (13)
Sheff (24)
Swanse (35)
York (14)
Nottm (38)
Chelms (7)
Oxford (36)
B QEH (29)
Truro (7)
Exeter (18)
Bradfd (11)
L Rfree (57)
Cardff (25)
Middlbr (7)
Sund (7)
Ulster (3)
West NI (4)
Wirral (9)
Wrexm (8)
Bristol (18)
Clwyd (4)
L Barts (73)
M RI (32)
Newc (23)
Norwch (20)
Colchr (0)
Total (1,015)
0 20 40 60 80 100
Percentage
Fig. 14.7. PD catheter insertion technique stratified by centre
291
many centres use a non-tunnelled line for a few days for ,90days). Amongst the 977 patients for whom data
while either a tunnelled line for HD or a PD catheter is were reported, 43.1% started dialysis on a tunnelled
placed, and therefore in retrospect the access used for line, 41.5% on a non-tunnelled line, 11.0% using a PD
the fourth dialysis session may provide a better catheter with only 4.0% having first access documented
description of the dialysis access selected for patients as an AVF. There was, however, wide variation amongst
presenting late. centres and clearly an understanding of practice patterns
Consideration of haemodialysis access separately from could lead to potential improvements in access service
the PD group revealed wide variation in the use of AVFs provision. There may also be reporting differences
for first HD (figure 14.6). This was demonstrated with the which need to be explored. Non-tunnelled haemodialysis
range being from 8.3% in Ulster to 70.8% in Derby lines are often used as a bridge to a more definitive form
(38.3% of HD patients at all centres). Central venous of access and it would be important to know what access
lines were clearly the main form of access where an was used at the end of the first week. As discussed above,
AVF was not available. The centres with highest tun- revision of the question used in the audit to investigate
nelled line use were London West (67.3%), Wolverhamp- the access used for the fourth rather than the first dialysis
ton (64.4%), Bangor (61.5%), and Colchester (60.7%). session in patients presenting late may provide more
Two centres reported non-tunnelled lines as the starting valuable information.
form of access in more than 50% of HD patients (Reading Only 13 centres reported that more than 15% of
54.4%, Exeter 58.9%). It will be important to understand patients presenting late had a peritoneal dialysis catheter
the variations in practice patterns that lie behind these inserted for use as first dialysis access. As the large part of
statistics which were not provided by current data. the remainder of patients presenting late start dialysis
Eighteen centres reported less than 10 patients using using a tunnelled vascular line, the centres that were
PD catheters for first dialysis in 2012 (figure 14.7). For able to make use of PD catheters for patients presenting
a total of 1,015 first PD catheters the insertion techniques late had a lower requirement for tunnelled or non-tun-
were 35.6% open surgical, 14.5% laparoscopic, 2.4% nelled lines. However, the number of patients presenting
peritoneoscopic and 27.7% percutaneous. Insertion tech- late reported in some centres was extremely small and it
nique was not reported for the remaining 19.9%. There is difficult to make firm observations about clinical path-
seems to be a strong tendency for many centres to rely ways for the development of dialysis access in this cohort.
on one single approach to PD catheter placement, it is Figure 14.9 combines PD and HD access data to
notable that 22 centres reported using a single technique demonstrate the association between referral time to
for all of their patients. This is important if evidence were renal services and the type of access used for the first
to suggest a benefit to offering an individualised tech- treatment. A strong relationship is seen between being
nique (e.g. percutaneous approach for low BMI patients known to the renal centre for more than a year and the
without previous surgery, or an open surgical approach likelihood of commencing dialysis using either an AVF
for more complex patients). Only 19 centres reported or a PD catheter. For patients presenting late, 84.6%
using the percutaneous technique at all and these start on some form of central venous line; however,
were Antrim, Birmingham Heartlands, Bangor, Belfast, amongst patients known to the centre for a year or
Brighton, Derby, Gloucester, Leicester, London Kings, more this percentage falls to 33.9%. Amongst HD
London West, Liverpool Aintree, Liverpool Royal Infirm- patients there was a strong relationship between being
ary, Plymouth, Portsmouth, Reading, Salford, Southend, known to the centre for more than a year and the use
Stoke and Wolverhampton. Amongst these centres were of AVF in preference to a venous line. Figure 14.9
some of those with the highest proportion of patients demonstrates that as the time known to renal services
using a PD catheter as first access (Wolverhampton 44%, increases, the proportion of patients starting dialysis on
Derby 34%, Brighton 32%, Liverpool RI 26%, Salford a line falls, whilst the proportion starting with an AVF
25%, Antrim 24%, London Kings 22%). Of the 20 centres or PD catheter increases. The number of patients starting
with the lowest PD usage as first access only three used dialysis with an AVG appears to remain the same regard-
the percutaneous approach. less of the referral time, but numbers are very small.
First dialysis access and referral time First dialysis access and surgical assessment
Figure 14.8 shows first access for centres providing Figure 14.10 shows the variation in centres according
data for patients presenting late (known to renal services to whether PD catheters were inserted at least two
292
West NI (2) AVF

Norwch (2) AVG
Derby (17) TL
NTL
L Kings (10)
PD catheter
Wolve (22)
Salford (25)
Liv Ain (6)
Newry (3)
L Rfree (54)
Nottm (11)
L Barts (88)
Liv RI (15)
B Heart (6)
York (14)
Brightn (36)
Wirral (15)
Donc (8)
M RI (8)
Belfast (16)
Hull (17)
Bradfd (10)
Sheff (30)
Ports (23)
Oxford (27)
Stoke (18)
Centre
Leeds (18)
B QEH (59)
Cardff (20)
Middlbr (28)
Redng (28)
Leic (47)
Bangor (4)
Sund (7)
Bristol (16)
Chelms (26)
Truro (18)
Swanse (26)
L West (73)
Antrim (1)
Exeter (32)
Basldn (6)
Ulster (8)
L St.G (6)
Sthend (6)
Plymth (4)
Clwyd (2)
Colchr (2)
Wrexm (2)
Glouc (13)
Prestn (21)
Newc (21)
Total (977)
0 20 40 60 80 100
Percentage
Fig. 14.8. Type of access for the first dialysis in patients presenting to a nephrologist ,90 days prior to dialysis start
AVF = arteriovenous fistula; AVG = arteriovenous graft; TL = tunnelled line; NTL = non-tunnelled line; PD = peritoneal dialysis
293
50 Total (844)
AVF
Cardff (25)
45 AVG
Redng (21)
TL
Norwch (20)
40 NTL
Glouc (17)
PD Catheter
Bradfd (11)
35 Sund (7)
Antrim (5)
30 West NI (4)
Percentage
Ulster (3)
25 Swanse (35)
Prestn (13)
20 Sheff (24)
Newc (23)
15 B QEH (29)
B Heart (19)
10 Oxford (36)
Exeter (18)
5 Liv RI (24)
Ports (33)
0
Basldn (14)
<90 days 90–365 days >365 days
Wirral (9)
Referral time
Centre
Bristol (12)
L St.G (8)
Fig. 14.9. Type of first dialysis access stratified by referral time
Chelms (7)
AVF = arteriovenous fistula; AVG = arteriovenous graft; TL = Donc (7)
tunnelled line; NTL = non-tunnelled line; PD = peritoneal dialysis Truro (7)
L Rfree (56)
York (13)
Brightn (42)
Hull (36)
weeks prior to commencing dialysis. Renal Association Salford (30)
Peritoneal Access Clinical Guidelines state that [16]: Bangor (6)
Nottm (38)
‘Whenever possible, catheter insertion should be M RI (32)
performed at least 2 weeks before starting peritoneal Stoke (14)
Plymth (8)
dialysis. Small dialysate volumes in the supine position L Kings (29)
can be used if dialysis is required earlier.’ Leeds (24)
Liv Ain (12)
This guideline was intended to reduce the risk of Newry (6)
Sthend (2)
dialysate leakage following catheter insertion, however Belfast (9)
it may actually have resulted in patients being less likely Wrexm (8)
Derby (29)
to use the PD catheter for early start PD and therefore Wolve (19)
possibly be exposed to the hazards of a central venous 0 20 40 60 80 100
line. It will be important to understand the association Percentage of PD patients
between early use and catheter outcomes. This has Fig. 14.10. Percentage of patients with PD catheter insertion .2
been explored in previous publications demonstrating a weeks before starting dialysis
modest increase in dialysate leakage can be mitigated
by careful preventative management [17]. It is quite
possible that this guideline has been a disincentive to three months prior to starting PD were less likely to
using PD for patients presenting late or for acute kidney have catheter placement via open surgical technique
injury and revision should be considered in the next than those who did not, possibly because such patients
iteration of the guideline. were more likely to have the laparoscopic approach.
From figure 14.11 it is clear that PD patients seen by a There does not appear from this data to be a relationship
surgeon at least three months prior to starting RRT were between surgical assessment and percutaneous catheter
more likely to have a laparoscopic insertion. Of those placement.
receiving surgical assessment at least three months This relationship was very different from that between
prior to commencing dialysis, 24.4% underwent laparo- surgical assessment and AVF formation (see the next
scopic insertion vs. 5.9% of those who did not. Indeed, section). It is quite possible that the time required to
patients who underwent surgical assessment at least plan PD catheter placement is less than that required
294
100 West NI (9)

Hull (43)
Sthend (16)
Glouc (41)
80
Redng (29)
L West (215)
B Heart (74)
60 L St.G (40)
Percentage
Newc (52)
Brightn (58)
Belfast (52)
40 Percutaneous Wolve (31)
Peritoneoscopic Liv Ain (37)
Laparoscopic Nottm (40)
20 Open surgery Oxford (87)
Wrexm (17)
Exeter (82)
0 Antrim (15)
Yes No No data Bangor (9)
(368 patients) (252 patients) (65 patients) Ports (91)
Colchr (22)
Assessed by surgeon at least three months before starting PD
Basldn (32)
Fig. 14.11. PD catheter insertion technique stratified by surgical Centre Ulster (16)
assessment Leeds (90)
Sund (53)
Liv RI (40)
Chelms (24)
L Kings (38)
Bradfd (46)
for AVF formation where vein mapping may be Salford (71)
necessary. L Rfree (102)
Figure 14.12 highlights the proportion of patients who Truro (22)
Wirral (11)
had been referred for surgical assessment at least three Prestn (87)
months prior to starting dialysis. Six renal centres were B QEH (104)
excluded because they returned data regarding surgical Stoke (41)
Sheff (88)
assessment or first seen date on fewer than half of their Derby (38)
patients (Clwyd, London Barts, Leicester, Manchester Donc (25)
Royal Infirmary, Norwich, Plymouth). There was con- Swanse (50)
Cardff (101)
siderable variation between the remaining renal centres. Newry (10)
Assessed
Not assessed
Overall, the proportion referred to a surgeon was highest Middlbr (72)
No data
York (25)
in York (92.0%) and Middlesbrough (91.7%). Out of Total (2,246)
2,246 patients with a referral time to nephrological 0 20 40 60 80 100
services of more than 90 days, 67% per cent had Percentage
been referred to a surgeon at least three months prior Fig. 14.12. Frequency of surgical assessment more than three
to dialysis start. months prior to starting dialysis
A detailed understanding of factors that prevent
patients from being assessed for access in a timely fashion
is required. These may reflect organisational factors or
clinical uncertainty around the need for dialysis. AVF whereas of those who were not seen by a surgeon
Figure 14.13 demonstrates a strong relationship only 9.7% did. Clearly, timely surgical assessment is a
between being assessed by a surgeon at least three months key component of the clinical pathway to fistula
before starting dialysis and the likelihood of starting on placement.
an AVF. This relationship was much stronger than that If data from figures 14.11 to 14.13 are considered
between surgical assessment and method of PD catheter together, the importance of timely referral for surgical
placement. This suggests that the role of surgical assess- assessment (if haemodialysis is the selected modality) is
ment is more important in relation to AVF placement. clear. Without such assessment, patients are more likely
Of those assessed by a surgeon at least three months to require temporary haemodialysis access such as a
prior to starting dialysis, 70.4% started dialysis on an tunnelled or non-tunnelled dialysis catheter.
295
100 No data/LTFU
NTL 3%
Tx/Died/Stop
TL
9%
AVG
80 AVF AVF
31%
PD Catheter
20%
60
Percentage
40
NTL
1% AVG
1%
20
0 TL
Yes No No data 35%
(1,549 patients) (714 patients) (34 patients)
Fig. 14.14. Type of dialysis access at three months
Assessed by a surgeon at least three months before starting dialysis
AVF = arteriovenous fistula; AVG = arteriovenous graft; TL =
Fig. 14.13. Type of haemodialysis access stratified by surgical tunnelled line; NTL = non-tunnelled line; PD = peritoneal dialysis;
assessment Tx = transplanted; LTFU = lost to follow up
AVF = arteriovenous fistula; AVG = arteriovenous graft; TL =
tunnelled line; NTL = non-tunnelled line
through a tunnelled line at three months. This may

suggest that obtaining definitive access for HD within
Dialysis access at three months after starting RRT three months of starting treatment remains a challenge.
The type of access used three months after starting Figures 14.14 and 14.15 demonstrate the differences in
dialysis gives an important insight into the responsive- access outcomes in aggregate and stratified by centre
ness of the access formation pathway. Table 14.3 respectively. By three months, 30.9% of patients were dia-
expresses the proportion of patients still dialysing using lysing using an AVF (range 7.5% London Barts to 59.6%
a particular form of access as a percentage of the access Liverpool Aintree); 0.9% were using an AVG (0% many
they originally started dialysis with. For example, 87.2% sites to 6.1% Exeter); 34.5% tunnelled lines (5.3% Liver-
of patients starting dialysis with an AVF were still using pool Aintree to 77.7% London West); 1.3% non-tun-
this at three months and 83.4% of patients starting on nelled lines; and 19.7% were using a PD catheter (0%
PD remained on this modality at three months. Of Plymouth to 48.1% Wolverhampton).
patients starting dialysis via a tunnelled line, the majority The majority (59.8%) of patients presenting late were
continued to use this form of access at three months being dialysed using tunnelled lines at three months
(72.6%) and of 864 patients who commenced dialysis after dialysis start (figure 14.16). The between centre
via a non-tunnelled line, 502 (58.1%) were dialysing range was from 0% in three centres (Clwyd, Newry,
Table 14.3. Type of dialysis access at 3 months stratified by first access type
Access in use at Access in use at three months

first dialysis (N)
AVF AVG TL NTL PD catheter Transplanted Died Stopped/LTFU No data
AVF (1,358) 87.2 0.3 3.7 0.0 0.9 0.9 4.6 0.1 2.4
AVG (46) 2.2 71.7 6.5 0.0 2.2 0.0 6.5 0.0 10.9
TL (1,328) 11.0 0.2 72.6 0.4 2.7 1.3 8.4 0.3 3.2
NTL (864) 8.4 0.1 58.1 6.4 5.2 0.0 16.4 0.3 5.0
PD (963) 0.4 0.1 5.7 0.0 83.4 2.4 2.5 0.2 5.3
AVF = arteriovenous fistula; AVG = arteriovenous graft; TL = tunnelled line; NTL = non-tunnelled line; PD = peritoneal dialysis;
LTFU = lost to follow up
296
Liv Ain (57) AVF

Bradfd (66) AVG
TL
York (51) NTL
L St.G (54) PD Catheter
B Heart (98) Tx/Died/Stop
No data/LTFU
Donc (39)
Antrim (21)
Cardff (145)
Salford (117)
Derby (77)
Sthend (24)
Plymth (33)
Leeds (131)
Sund (68)
Prestn (121)
Colchr (28)
Stoke (72)
Sheff (139)
Clwyd (22)
Exeter (132)
Liv RI (76)
B QEH (189)
Ports (149)
Swanse (111)
Centre
Newry (18)
Glouc (73)
Truro (47)
L Rfree (198)
Wirral (33)
Wrexm (27)
Bristol (162)
Oxford (148)
Leic (206)
L Kings (116)
Newc (96)
Bangor (19)
Basldn (54)
Brightn (131)
Belfast (76)
West NI (14)
Wolve (81)
Nottm (86)
Hull (94)
Middlbr (106)
Chelms (59)
Redng (78)
Ulster (27)
L West (310)
L Barts (280)
0 20 40 60 80 100
Percentage
Fig. 14.15. Type of dialysis access at three months stratified by centre

Tx = transplanted; LTFU = lost to follow up
297
No data/LTFU AVF Reported causes of access failures amongst peritoneal

4% 8%
Tx/Died/Stop AVG dialysis patients are not included here as the numbers
14% 0% reported were too low to make firm conclusions.
2011 PD access audit one-year follow-up

Centres who reported on PD patients in the 2011 vas-
PD Catheter cular and peritoneal access audits were asked to complete
13%
a one year follow up of their PD patients. The additional
information requested was the date of catheter failure, the
NTL reason for catheter failure, the number of catheters used
1% TL
60%
during the year, and the modality in use at one year
after starting PD. Of 44 centres who reported data on
PD patients in 2011, 28 completed the one year follow
Fig. 14.16. Type of dialysis access at three months in patients
referred to renal services less than 90 days before starting dialysis up request returning data on 649 (70.9%) patients.
AVF = arteriovenous fistula; AVG = arteriovenous graft; The reported numbers were too low to draw firm con-
TL = tunnelled line; NTL = non-tunnelled line; PD = peritoneal clusions. Unsurprisingly the principal causes of catheter
dialysis; Tx = transplanted; LTFU = lost to follow up failure were flow or infection related (figure 14.20).
Figure 14.21 is a funnel plot which graphically
displays the unadjusted percentage of PD patients
experiencing a catheter failure within one year of com-
Sunderland) to 93.1% at London West (figure 14.17). mencement of RRT across multiple renal centres
Amongst patients presenting late, only 8.0% were using according to Speigelhalter’s method [18]. PD catheter
an AVF at three months (individual centres ranged failure was censored for transplantation, elective trans-
from 0% in 16 centres to 75% in Plymouth). PD catheters fer to HD or death. The bold dotted line represents the
were used by 12.7% of patients (range 0% in 14 centres to mean one-year catheter failure (23.0%). The 95% (solid
85.7% in Sunderland). These percentages must be inter- lines) and 99.9% (dotted lines) binomial control limits
preted with caution as reported numbers of patients (essentially corresponding to 2 and 3 standard devi-
presenting late tended to be low in many centres. ations) were superimposed to indicate possible outlier
Figure 14.18 shows comparative access failure for the thresholds for ‘alert’ and ‘alarm’ [19]. The results have
different access types within three months. This was to be cautiously interpreted due to the extent of and
defined as a documented date of failure/discontinuation variation in missing data, small numbers of patients in
recorded within three months of starting dialysis unless some centres and non-adjustment for any patient
a centre comment indicated that it was a planned discon- related factors.
tinuation. Failure rates were generally less than 5%, apart Of the centres for which data were available (n = 28),
from AVGs where it was closer to 15%. There were no outlier centres were identified with failure rates above
deficiencies in the way that failure was recorded in this the upper 95% ‘alert’ or 99.9% ‘alarm’ limits for PD
audit, however it is interesting that for most forms of catheter failures. Such data is suggestive of the absence
access the failure rates are rather similar at three months. of outlier centres with abnormally poor one year catheter
Numbers of access failures reported were small, survival rates relative to the other centres. Contrastingly,
however it can be seen from figure 14.19 that there was four renal centres reported one-year catheter failure rates
relatively poor reporting of the reason for failures. This below the 95% control limit. Furthermore, of these, one
may reflect local documentation procedure. Infectious centre reported a one-year catheter failure rate of zero.
causes were reported as contributing to 26.1% of access This centre was thus considered as an ‘alarm’ outlier
failures of tunnelled lines and 12.1% of non-tunnelled raising questions over data integrity or accuracy.
lines, and stenosis was reported as contributing to Of note, although the overall mean one-year catheter
22.7% of AVF failures. Steal syndrome was also a com- failure rate was similar to that which was recommended
mon reason for failure in AVF and AVG (29.5% and in the guidelines issued by the ISPD/RA [16, 20] (23%
28.6% respectively). This data should be regarded as vs. 20%), reported failure rates of as low as 10% raise
provisional and would benefit from further detailed questions of whether such modest targets should be
exploration in future audit. revised to improve practice [21].
298
Plymth (4)
York (14)
Cardff (20)
Wirral (15)
Bangor (4)
Stoke (18)
Salford (25)
Derby (17)
Sthend (6)
L St.G (6)
Liv Ain (6)
Exeter (32)
B QEH (59)
Sund (7)
Donc (8)
Hull (17)
Bradfd (10)
Glouc (13)
Chelms (26)
Swanse (26)
Liv RI (15)
Truro (18)
Prestn (21)
Wolve (22)
Centre
Ports (23)
Leic (47)
Middlbr (28)
Sheff (30)
Brightn (36)
L West (73)
L Barts (88)
L Rfree (54)
Antrim (1)
B Heart (6)
Belfast (16)
Newc (21)
Oxford (27)
Basldn (6)
Ulster (8)
AVF
Leeds (18)
AVG
Redng (28) TL
L Kings (10) NTL
Colchr (2) PD Catheter
Tx/Died/Stop
West NI (2) No data/LTFU
Wrexm (2)
Nottm (11)
Newry (3)
Clwyd (2)
Total (951)
0 20 40 60 80 100
Percentage
Fig. 14.17. Type of dialysis access at three months in patients referred to renal services less than 90 days before starting dialysis,
stratified by centre
Tx = transplanted; LTFU = lost to follow up
299
16
14
12
10
Percentage
Fig. 14.18. Percentage of patients

2
experiencing failure of first access within
three months, by type of first access
0 AVF = arteriovenous fistula; AVG =
Percuta- Peritoneo- Open Missing Laparo- TL NTL AVF AVG arteriovenous graft; TL = tunnelled line;
neous scopic surgery data scopic
NTL = non-tunnelled line; PD = peritoneal
PD catheter insertion technique Type of haemodialysis vascular access dialysis
Discussion and recommendations data across a range of fields exist. Equally, there
remain ambiguities in the data fields which need to
. This multisite dialysis access audit from England, be refined to simplify collection and improve accu-
Wales and Northern Ireland has provided important racy. It may be preferable to collect dialysis access
information regarding the variation in access pro- at the fourth rather than the first dialysis session
vision and outcomes. Although this audit represents since non-tunnelled lines are often used for one or
an important advance for the UK, data collection is two dialysis sessions before more permanent access
still not optimal as significant amounts of missing is achieved (PD catheter or tunnelled line).
Infection
Sepsis
AVF
Failure to develop
(88 access failures)
Stenosis
Thrombosis
Steal syndrome
Other
AVG Unknown
First access type
(7 access failures)
NTL
(33 access failures)
Fig. 14.19. Reported causes of access

TL failure within three months
(23 access failures) (haemodialysis) stratified by first access
type
AVF = arteriovenous fistula;
0 20 40 60 80 100 AVG = arteriovenous graft; TL = tunnelled
Percentage line; NTL = non-tunnelled line
300
Flow related
Laparoscopic Surgical related
(10 failures) PD related
Infection related
Unknown
Open
(72 failures)
Insertion technique
Peritoneoscopic
(6 failures)
Percutaneous
(35 failures)
Missing
(25 failures)
Fig. 14.20. Causes of dialysis access
0 20 40 60 80 100 failure within one year of PD catheter
Percentage insertion
. It is clear from the data that many centres still utilise . The practice of PD catheter insertion in patients
high numbers of tunnelled and non-tunnelled dialy- presenting late was used by relatively few centres.
sis catheters especially in patients presenting late. Of Only 13 out of 50 centres with sufficient data on
concern is that tunnelled lines continue to be used in patients presenting late placed a peritoneal dialysis
approximately a third of patients three months post catheter in more than 15% of patients as first dialysis
dialysis start and this figure is higher for patients access. If the National Service Specification for
presenting late (60%). dialysis recommendation that PD catheters should
. Surgical assessment is of high importance in the devel- be placed within 72 hours of being required is to
opment of permanent vascular access (AVF/AVG). be complied with, a significant practice change is
Whereas, in those assessed by a surgeon at least needed [22]. This timeframe may be shortened in
three months prior to starting dialysis, 70.4% received the future. It is relevant here that 50% of centres
an AVF, only 9.7% of those not assessed did. This only reported using a single technique for PD
strong relationship was not seen between surgical catheter insertion.
assessment and PD catheter placement, apart from . Variation demonstrated in PD catheter functional-
the use of the laparoscopic insertion technique. ity suggests that further exploration of centre
specific practice around PD access would also be
60 of value.
. The guideline recommending that PD catheters
50 should be inserted at least two weeks prior to use
[16] should be reconsidered since it may be a
disincentive to using PD for patients presenting
40
late.
30
20
Acknowledgement
10
Thanks are expressed to the Healthcare Quality
Improvement Partnership (HQIP) who contributed to
0
0 20 40 60 80 the funding of the PD access audit and also the renal
Number of patients with data in centre centres for their assistance in providing the data.
Fig. 14.21. Funnel plot of the percentage of PD catheter failures
within one year of insertion Conflicts of interest: none
301
References
1 Briggs V, et al. UK Renal Registry 15th annual report: Chapter 8 UK 12 Lobbedez T. et al. Peritoneal dialysis in polycystic kidney disease
multisite peritoneal dialysis access catheter audit for first PD catheters patients. Report from the French peritoneal dialysis registry (RDPLF).
2011. Nephron Clin Pract 2013;123(suppl 1):165–181 Nephrol Dial Transplant 2011;26:7:2332–2339
2 NHS Information Centre. National Kidney Care Audit Vascular Access 13 O’Hare AM. et al. Mortality risk stratification in chronic kidney disease:
Report. 2011 [cited 2013 1st October]; Available from: http://www.hqip. one size for all ages? J Am Soc Nephrol 2006;17:3:846–853
org.uk/assets/NCAPOP-Library/VAReport2011Interactive03082011– 14 Roderick PJ. Chronic kidney disease in older people: a cause for concern?
FINAL.pdf Nephrol Dial Transplant 2011;26:10:3083–3086
3 Fluck R, Pitcher D, Steenkamp R. Vascular Access Audit Report 2012: 15 Eriksen BO, Ingebretsen OC. The progression of chronic kidney disease:
UK Renal Registry and NHS Kidney Care 2012 a 10-year population-based study of the effects of gender and age. Kid-
4 National Service Framework for Renal Services. Dialysis and Transplan- ney Int 2006;69:2:375–382
tation 2004, Department of Health 16 Wilkie M, Jenkins S, Shrestha B. Renal Association Peritoneal dialysis
5 Ravani P, et al. Associations between hemodialysis access type and clini- access clinical guidelines. 2009. 31st August 2013. http://www.renal.
cal outcomes: a systematic review. J Am Soc Nephrol 2013;24:3:465–473 org/Clinical/GuidelinesSection/PeritonealAccess.aspx
6 National Kidney Foundation Kidney Disease Outcomes Quality Initiat- 17 Povlsen JV, Ivarsen P. How to start the late referred ESRD patient
ive (NKF KDOQI). Clinical Practice Guidelines and Clinical Practice urgently on chronic APD. Nephrol Dial Transplant 2006;21(suppl 2):
Recommendations 2006 Updates – Vascular Access 2006 ii:56–59
7 Perl J. et al. Hemodialysis vascular access modifies the association 18 Spiegelhalter DJ. Funnel plots for comparing institutional performance.
between dialysis modality and survival. J Am Soc Nephrol 2011;22:6: Stat Med 2005; 24:8:1185–1202
1113–1121 19 Spiegelhalter D. Funnel plots for institutional comparison. Qual Saf
8 Koch M. et al. Comparable outcome of acute unplanned peritoneal Health Care 2002;11:4:390–391
dialysis and haemodialysis. Nephrol Dial Transplant 2012;27:1:375–380 20 Figueiredo A. et al. Clinical practice guidelines for peritoneal access.
9 Coentrao L. et al. Effects of starting hemodialysis with an arteriovenous Perit Dial Int 2010;30:4:424–429
fistula or central venous catheter compared with peritoneal dialysis: a 21 Singh, N. et al. Risk factors associated with peritoneal dialysis catheter
retrospective cohort study. BMC Nephrol 2012;13:88 survival: a 9-year single-center study in 315 patients. J Vasc Access
10 National Institute for Health and Clinical Excellence, NICE clinical 2010;11:4:316–322
guideline 125 – Peritoneal dialysis in the treatment of stage 5 chronic 22 NHS England 2013/14 NHS Standard contract for renal dialysis:
kidney disease 2011 Peritoneal (adult) 2013
11 Shaw C. et al. UK Renal Registry 15th annual report: Chapter 2 UK RRT
prevalence in 2011: national and centre-specific analyses. Nephron Clin
Pract 2013;123(suppl 1):29–54
302
Chapter 15 Epidemiology of Reported
Infections amongst Patients Receiving
Dialysis for Established Renal Failure in
England from May 2011 to April 2012: a
Joint Report from Public Health England
and the UK Renal Registry
Lisa Crowleya, David Pitcherb, Jennie Wilsonc, Rebecca Guyd, Richard Flucka
a
Royal Derby Hospital, Derby, UK; bUK Renal Registry, Bristol, UK; cInstitute for Practice, Interdisciplinary Research and
Enterprise, University of West London, London, UK; dCentre for Infectious Disease surveillance and Control, PHE,
London, UK
Key Words
. In the same period there were 138 Clostridium
Clostridium difficile . Eschericia coli . Dialysis . Established difficile infection episodes with a rate of 0.61 per
renal failure . Infection . Staphylococcus 100 prevalent dialysis patients per year.
. Methicillin sensitive Staphylococcus aureus (MSSA)
bacteraemia rates were 1.15 per 100 prevalent
Summary dialysis patients per year with 322 episodes of
blood stream infection reported.
. From May 1st 2011 to April 30th 2012 there were 49 . Eschericia coli data were available from June 2011
episodes of methicillin resistant Staphylococcus and showed a reported rate of 0.92 per 100 prevalent
aureus (MRSA) bacteraemia in end stage renal dialysis patients per year.
failure patients on dialysis. This represents a further . In each infection type the presence of a central
slight decline in MRSA bacteraemia rates which venous catheter appeared to correlate with increased
have been falling since data collection began in 2007. risk.
303
Introduction E. coli BSI, in patients with ERF who were receiving

dialysis in England. This is the first UKRR report which
Infection remains the second leading cause of death in will contain data on Clostridium difficile (C. difficile)
patients with established renal failure (ERF) who receive and E. coli infections reported by laboratories as being
renal replacement therapy (RRT). The high rates of associated with ERF patients receiving dialysis.
systemic infection reported in haemodialysis patients
are related to their impaired immune system, the high
number of invasive procedures they are exposed to and
the type of vascular access used [1]. Methods
Previous UK Renal Registry (UKRR) reports have
detailed the epidemiology of staphylococcal bacteraemias The report covers the period of 1st May 2011 to 30th April
in patients with ERF receiving dialysis [2]. These were 2012. In choosing this time frame it is important to note that
the data on MSSA reported here overlaps with the period included
joint reports from the UKRR and the Health Protection in the previous UKRR report where MSSA cases from January
Agency (HPA). As of 1st April 2013 the HPA has now 2011 to June 2011 were reported.
become part of Public Health England (PHE) within It should also be noted that even though reporting is man-
the Department of Health. In addition to staphylococcal datory for these data collections (MRSA, MSSA and E. coli BSI
bacteraemias, surveillance has been expanded to incor- and CDI) completion of renal failure and dialysis information is
currently conducted on a voluntary basis depending on the data
porate Escherichia coli (E. coli) bloodstream infections entry policy within the reporting NHS acute Trust. Therefore a
(BSIs). As well as the mandatory reporting of methicillin reported infection rate of zero for an individual centre may
resistant Staphylococcus aureus (MRSA) BSIs, methicillin represent a difference in reporting policy.
sensitive Staphylococcus aureus (MSSA) BSIs have been The methods used have been described in previous UKRR
mandatory to report since January 2011 and E. coli reports [4]. Briefly, four stages of data collection and validation
were undertaken:
BSIs since June 2011; Clostridium difficile infection
(CDI) reporting has been mandatory for all patients 1 Identification of bacteraemias (and CDI) potentially associ-
aged two and above since 2007. CDIs are reported ated with dialysis patients. This data was captured by the
microbiology laboratory using the clinical details provided
according to a national testing protocol although during and the setting in which the sample was obtained.
the timeframe of this report there may have been some 2 This record was ‘shared’ with the parent renal centre. The
inter-hospital variation in testing methods [2]. microbiology laboratory attributed the record to the renal
MRSA BSI surveillance is the only data collection of centre responsible for the dialysis of the patient which in
the four which displays a prompt for additional renal turn triggered an email alert to the identified contact within
the parent renal centre.
failure information although the unprompted feature is 3 The renal centre then completed the additional renal data on
available for the other collections; however, completion the case via the HCAI-DCS website.
of renal failure information is not a requirement for 4 An additional validation and data capture step has been
any of these data collections. The data is supplied by clini- introduced as not all records were shared or completed.
This involved emailing clinical or infection control leads at
cal staff and captured using a secure web-based system,
the parent centre to finish incomplete records and confirm
the Healthcare Associated Infection Data Capture System that records associated with their centre were related to
(HCAI-DCS). As in previous reports, a final round of patients in ERF requiring dialysis.
data validation was also undertaken which involved This data reporting mechanism applies only to centres in England
emailing the clinical or infection control leads at each and is not utilised in Wales, Scotland or Northern Ireland.
centre in order for them to check the details and accept
the record. The dataset included dialysis modality, type For each infection, the number of individual episodes is shown
of dialysis access and use of non-tunnelled venous cath- alongside centre-specific rates which were calculated using the
number of prevalent dialysis patients according to 2011 data [5].
eters within the preceding 28 days. The previous report The collection period for E. coli BSIs was eleven months compared
confirmed that while dialysis patients remain at increased to twelve months of collection for the other infections. The rates
risk from MRSA there has been a continued year on year presented for E. coli have been adjusted accordingly to show the
decline in the number of bacteraemias [3]. The report rate per 100 prevalent dialysis patients per year. Data on the
also provided the first experience of MSSA BSI reporting type of access in use at the time of infection was also provided.
In order to adjust for variation in precision of estimated rate,
for the first six months of mandatory surveillance. the rate of bacteraemia/CDI per 100 prevalent dialysis patients
This report covers one year of reporting for MRSA, per year has been plotted against the centre size in a funnel plot.
MSSA and CDI, and eleven months of reporting for This process has been repeated for each infection. In the case of
304
Chapter 15 Epidemiology of infection in dialysis patients
MRSA a comparative box plot to demonstrate the overall trend is venous catheter (46.7%). In total, 51.1% of cases occurred
also included. in patients with either a tunnelled or non-tunnelled cath-
eter in situ, 8.9% of cases occurred in patients with an
arteriovenous graft while 37.8% occurred in patients
with an arteriovenous fistula (table 15.2).
Results
Clostridium difficile
Methicillin resistant Staphylococcus aureus In total, 172 episodes of CDI were reported to PHE in
In total, 53 MRSA bacteraemias were reported to PHE the period covered by this report. Of these, only one
as being associated with a dialysis patient during the time episode was shared and completed in full. A further 26
frame of this report. Of these, four episodes were rejected episodes were shared of which two were rejected. Of
by the parent centre because they occurred in patients the remaining unshared episodes, a further 32 were either
with acute kidney injury (AKI) rather than the patient rejected by the main centre or a main centre could not be
being in ERF. This left a total of 49 episodes of MRSA identified. This left a total of 138 infections in dialysis
bacteraemia within the time period. These episodes patients giving a rate for England of 0.61 infections per
were split between 42 patients registering one episode, 100 prevalent dialysis patients per year. Fourteen centres
two patients registering two episodes and one patient did not report any CDI episodes and the highest reported
registering three infection episodes. rate was 4.44 per 100 prevalent dialysis patients
The overall infection rate for England was 0.22 per 100 (table 15.3, figure 15.4). A funnel plot was created to
prevalent dialysis patients per year. This rate represents a display the rate compared to centre size (figure 15.5).
further year on year fall in the MRSA bacteraemia rates in Amongst patients for whom the type of access was
England as illustrated by the box plot in figure 15.1. known, 49.5% of patients had a line at the time of the
Centre level data can be seen in table 15.1 and includes infection (47.4% tunnelled catheter, 2.1% non-tunnelled
the absolute number of episodes and rates based on catheter), 42.3% of patients had an arteriovenous fistula
using the number of 2011 dialysis patients as the denomi- and 2.1% of patients an arteriovenous graft (table 15.4).
nator. The majority of centres did not report any MRSA Six (6.2%) episodes occurred in peritoneal dialysis
bacteraemia episodes. Only two centres had an infection patients where access was via a Tenchkoff catheter.
rate in excess of one per 100 prevalent dialysis patients
per year (figure 15.2). In order to adjust for variation in Methicillin sensitive Staphylococcus aureus
precision of estimated rate, the rate has been plotted In total, 322 episodes of MSSA bacteraemia were
against centre size in a funnel plot (figure 15.3). reported to PHE. However, 61 of these episodes were
Amongst patients for whom the type of access at the excluded leaving a final total of 261 bacteraemia episodes
time of infection was known, the highest proportion of within the time frame. The main reasons for exclusion
infections occurred in patients with a tunnelled central were a) the patient was unknown to the allocated centre
and b) an inability to identify the centre responsible for
the dialysis care. The majority of episodes were reported
3.5
in haemodialysis patients, with just six reported episodes
MRSA rate per 100 dialysis patients
3.0 amongst peritoneal dialysis patients.

2.5
The overall MSSA bacteraemia rate for England was
1.15 per 100 prevalent dialysis patients per year. There
2.0 was considerable variation in both the bacteraemia rate
1.5 at each centre and also in the number of individual
infection episodes at an individual centre which ranged
1.0
from 0 to 25 (table 15.5). The highest rate reported was
0.5 3.83 per 100 prevalent dialysis patients per year
(figure 15.6). Figure 15.7 is a funnel plot displaying the
0.0
2007/08 2008/09 2009/10 2010/11 2011/12 centre rates plotted against the size of the centre. A
Year number of centres reported a zero infection rate. Centres
Fig. 15.1. Box and whisker plot of MRSA rates by renal centre per reporting no MSSA infections are Birmingham Heart-
100 prevalent dialysis patients per year by reporting year lands, Chelmsford, Nottingham, Plymouth, Ipswich,
305
Table 15.1. Centre-specific data for MRSA bacteraemia episodes by access type, 1/05/2011 to 30/04/2012
MRSA bacteraemia episodes

Prevalent patients on 31/12/2011 (1/5/2011–30/4/2012) Rate per 100
dialysis
Centre HD PD Dialysis Total AVF AVG NTC TC PD UK patients
B Heart 446 46 492 0 0 0 0 0 0 0 0.00
B QEH∗ 894 167 1,061 0 0 0 0 0 0 0 0.00
Basldn 155 26 181 2 2 0 0 0 0 0 1.10
Bradfd 196 32 228 1 1 0 0 0 0 0 0.44
Brightn 340 80 420 1 0 1 0 0 0 0 0.24
Bristol∗ 474 66 540 1 0 0 0 0 0 1 0.19
Camb∗ 371 41 412 0 0 0 0 0 0 0 0.00
Carlis 66 24 90 0 0 0 0 0 0 0 0.00
Carsh 753 103 856 3 0 0 0 3 0 0 0.35
Chelms 119 26 145 0 0 0 0 0 0 0 0.00
Colchr 120 0 120 0 0 0 0 0 0 0 0.00
Covnt∗ 362 90 452 0 0 0 0 0 0 0 0.00
Derby 207 112 319 0 0 0 0 0 0 0 0.00
Donc 162 26 188 0 0 0 0 0 0 0 0.00
Dorset 239 53 292 0 0 0 0 0 0 0 0.00
Dudley 146 53 199 0 0 0 0 0 0 0 0.00
Exeter 376 78 454 0 0 0 0 0 0 0 0.00
Glouc 194 39 233 0 0 0 0 0 0 0 0.00
Hull 323 89 412 0 0 0 0 0 0 0 0.00
Ipswi 125 31 156 1 0 0 0 0 0 1 0.64
Kent 376 68 444 2 1 0 0 1 0 0 0.45
L Barts∗ 899 171 1,070 0 0 0 0 0 0 0 0.00
L Guys∗ 607 33 640 1 0 0 0 0 0 1 0.16
L Kings 468 89 557 1 1 0 0 0 0 0 0.18
L Rfree∗ 711 94 805 1 0 1 0 0 0 0 0.12
L St.G∗ 294 55 349 1 0 0 0 1 0 0 0.29
L West∗ 1,412 35 1,447 4 1 0 0 3 0 0 0.28
Leeds∗ 513 92 605 2 0 1 0 1 0 0 0.33
Leic∗ 854 159 1,013 8 3 0 1 4 0 0 0.79
Liv Ain 179 15 194 0 0 0 0 0 0 0 0.00
Liv RI∗ 381 74 455 2 0 0 0 2 0 0 0.44
M RI∗ 481 91 572 4 2 0 0 2 0 0 0.70
Middlbr 315 18 333 0 0 0 0 0 0 0 0.00
Newc∗ 265 48 313 0 0 0 0 0 0 0 0.00
Norwch 309 59 368 0 0 0 0 0 0 0 0.00
Nottm∗ 402 92 494 0 0 0 0 0 0 0 0.00
Oxford∗ 419 92 511 1 0 0 1 0 0 0 0.20
Plymth∗ 132 47 179 0 0 0 0 0 0 0 0.00
Ports∗ 524 95 619 1 0 0 0 1 0 0 0.16
Prestn 520 65 585 4 2 0 0 1 0 1 0.68
Redng 272 88 360 0 0 0 0 0 0 0 0.00
Salford 363 113 476 0 0 0 0 0 0 0 0.00
Sheff ∗ 591 62 653 3 2 1 0 0 0 0 0.46
Shrew 187 35 222 0 0 0 0 0 0 0 0.00
Stevng 412 30 442 0 0 0 0 0 0 0 0.00
Sthend 122 18 140 1 0 0 0 1 0 0 0.71
Stoke 318 82 400 2 2 0 0 0 0 0 0.50
Sund 178 17 195 2 0 0 0 1 1 0 1.03
Truro 152 26 178 0 0 0 0 0 0 0 0.00
Wirral 196 42 238 0 0 0 0 0 0 0 0.00
Wolve 307 71 378 0 0 0 0 0 0 0 0.00
York 144 25 169 0 0 0 0 0 0 0 0.00
England 19,371 3,283 22,654 49 17 4 2 21 1 4 0.22
∗
Transplant centres
AVF = arteriovenous fistula; AVG = arteriovenous graft; NTC = non-tunnelled catheter; TC = tunnelled catheter; PD = peritoneal dialysis;
UK = unknown access type
306
England
Basldn
Sund
Leic *
Sthend
M RI *
Prestn
Ipswi
Stoke
Sheff *
Kent
Centre
Liv RI *
Bradfd
Carsh
Leeds *
L St.G *
L West *
Brightn
Oxford *
Bristol *
L Kings
Ports *
Fig. 15.2. MRSA bacteraemia rate per
L Guys *
100 prevalent dialysis patients per year by
L Rfree *
renal centre
0 1 2 3 4 5 6 Centres with no reported infection episodes are not
MRSA rate per 100 prevalent dialysis patients displayed
Portsmouth and Reading. At present it is not clear was in situ at the time of infection for 54.1% whilst 35.4%
whether this represents lack of reporting rather than no had a native arteriovenous fistula (table 15.6).
reportable episodes.
Amongst patients with an MSSA episode and for Escherichia coli
whom the type of access was known, a tunnelled catheter A total of 284 episodes of E. coli bacteraemia were
reported in dialysis patients. A total of 93 episodes were
excluded from the final total (the highest number of
3.0
Dotted lines show 99.9% limits Table 15.2. Type of renal access in patients with established
Solid lines show 95% limits renal failure where record shared and completed for the MRSA
2.5 bacteraemia episodes
MRSA rate per 100 dialysis patients
MRSA bacteraemia
2.0 (1/5/2011–30/4/2012)
1.5
Renal access type N % Access class
Unknown 4
1.0 Haemodialysis
Other 0
Arteriovenous fistula 17 37.8 46.7
0.5
Arteriovenous graft 4 8.9
Non-tunnelled catheter 2 4.4 51.1
0.0 Tunnelled catheter 21 46.7
0 200 400 600 800 1,000 1,200 1,400 1,600 Peritoneal dialysis 1 2.2 2.2
Number of dialysis patients Total 49
Fig. 15.3. Funnel plot of the MRSA bacteraemia rate per 100 Total known access 45
prevalent dialysis patients per year by renal centre
307
Table 15.3. Centre-specific data for Clostridium difficile episodes by access type, 1/05/2011 to 30/04/2012
Clostridium difficile episodes

dialysis
B Heart 446 46 492 0 0 0 0 0 0 0 0.00
B QEH∗ 894 167 1,061 0 0 0 0 0 0 0 0.00
Basldn 155 26 181 0 0 0 0 0 0 0 0.00
Bradfd 196 32 228 8 3 0 0 5 0 0 3.51
Brightn 340 80 420 2 1 0 0 0 0 1 0.48
Bristol∗ 474 66 540 1 1 0 0 0 0 0 0.19
Camb∗ 371 41 412 3 0 0 0 0 0 3 0.73
Carlis 66 24 90 4 0 0 0 0 2 2 4.44
Carsh 753 103 856 9 3 0 0 5 1 0 1.05
Chelms 119 26 145 0 0 0 0 0 0 0 0.00
Colchr 120 0 120 1 0 0 0 1 0 0 0.83
Covnt∗ 362 90 452 2 0 0 1 1 0 0 0.44
Derby 207 112 319 0 0 0 0 0 0 0 0.00
Donc 162 26 188 2 0 1 0 1 0 0 1.06
Dorset 239 53 292 8 0 0 0 7 0 1 2.74
Dudley 146 53 199 3 0 0 0 0 0 3 1.51
Exeter 376 78 454 0 0 0 0 0 0 0 0.00
Glouc 194 39 233 2 0 0 0 0 0 2 0.86
Hull 323 89 412 3 1 0 0 0 2 0 0.73
Ipswi 125 31 156 0 0 0 0 0 0 0 0.00
Kent 376 68 444 7 1 0 0 6 0 0 1.58
L Barts∗ 899 171 1,070 0 0 0 0 0 0 0 0.00
L Guys∗ 607 33 640 5 3 0 0 2 0 0 0.78
L Kings 468 89 557 1 1 0 0 0 0 0 0.18
L Rfree∗ 711 94 805 5 3 0 0 2 0 0 0.62
L St.G∗ 294 55 349 1 0 0 0 1 0 0 0.29
L West∗ 1,412 35 1,447 9 0 0 0 0 0 9 0.62
Leeds∗ 513 92 605 0 0 0 0 0 0 0 0.00
Leic∗ 854 159 1,013 3 0 0 0 0 0 3 0.30
Liv Ain 179 15 194 0 0 0 0 0 0 0 0.00
Liv RI∗ 381 74 455 7 0 0 0 0 0 7 1.54
M RI∗ 481 91 572 7 1 0 0 6 0 0 1.22
Middlbr 315 18 333 5 0 0 0 0 0 5 1.50
Newc∗ 265 48 313 0 0 0 0 0 0 0 0.00
Norwch 309 59 368 0 0 0 0 0 0 0 0.00
Nottm∗ 402 92 494 0 0 0 0 0 0 0 0.00
Oxford∗ 419 92 511 2 1 0 1 0 0 0 0.39
Plymth∗ 132 47 179 1 0 0 0 0 0 1 0.56
Ports∗ 524 95 619 1 0 0 0 0 0 1 0.16
Prestn 520 65 585 1 1 0 0 0 0 0 0.17
Redng 272 88 360 0 0 0 0 0 0 0 0.00
Salford 363 113 476 1 1 0 0 0 0 0 0.21
Sheff∗ 591 62 653 9 4 0 0 4 0 1 1.38
Shrew 187 35 222 8 5 1 0 0 0 2 3.60
Stevng 412 30 442 3 0 0 0 3 0 0 0.68
Sthend 122 18 140 0 0 0 0 0 0 0 0.00
Stoke 318 82 400 5 3 0 0 1 1 0 1.25
Sund 178 17 195 0 0 0 0 0 0 0 0.00
Truro 152 26 178 4 4 0 0 0 0 0 2.25
Wirral 196 42 238 0 0 0 0 0 0 0 0.00
Wolve 307 71 378 2 2 0 0 0 0 0 0.53
York 144 25 169 3 2 0 0 1 0 0 1.78
England 19,371 3,283 22,654 138 41 2 2 46 6 41 0.61
∗
Transplant centres
308
England
Carlis
Shrew
Bradfd
Dorset
Truro
York
Kent
Liv RI *
Dudley
Middlbr
Sheff *
Stoke
M RI *
Donc
Carsh
Glouc
Centre
Colchr
L Guys *
Camb *
Hull
Stevng
L West *
L Rfree *
Plymth *
Wolve
Brightn
Covnt *
Oxford *
Leic *
L St.G *
Salford
Bristol *
L Kings
Prestn Fig. 15.4. CDI rate per 100 prevalent
Ports * dialysis patients per year by renal centre
0 1 2 3 4 5 6 7 8 Centres with no reported infection episodes are not
CDI rate per 100 prevalent dialysis patients displayed
5
Dotted lines show 99.9% limits Table 15.4. Type of renal access in patients with established renal
Solid lines show 95% limits failure where record shared and completed for Clostridium difficile
episodes
4
CDI rate per 100 dialysis patients
Clostridium difficile episodes

(1/5/2011–30/4/2012)
3
2 Unknown 41
Haemodialysis
Other 0
1 Arteriovenous fistula 41 42.3 44.3
0 Tunnelled catheter 46 47.4
0 200 400 600 800 1,000 1,200 1,400 1,600 Peritoneal dialysis 6 6.2 6.2
Fig. 15.5. Funnel plot of the CDI rate per 100 prevalent dialysis Total known access 97
patients per year by renal centre
309
Table 15.5. Centre-specific data for MSSA bacteraemia episodes by access type, 1/05/2011 to 30/04/2012
MSSA bacteraemia episodes

dialysis
B Heart 446 46 492 0 0 0 0 0 0 0 0.00
B QEH∗ 894 167 1,061 10 4 0 0 6 0 0 0.94
Basldn 155 26 181 1 0 0 0 1 0 0 0.55
Bradfd 196 32 228 3 0 0 0 3 0 0 1.32
Brightn 340 80 420 7 4 2 0 1 0 0 1.67
Bristol∗ 474 66 540 6 3 0 0 3 0 0 1.11
Camb∗ 371 41 412 7 0 0 0 0 0 7 1.70
Carlis 66 24 90 3 0 0 0 3 0 0 3.33
Carsh 753 103 856 9 4 2 0 2 1 0 1.05
Chelms 119 26 145 0 0 0 0 0 0 0 0.00
Colchr 120 0 120 4 0 0 0 4 0 0 3.33
Covnt∗ 362 90 452 2 1 0 1 0 0 0 0.44
Derby 207 112 319 6 4 0 0 2 0 0 1.88
Donc 162 26 188 1 0 1 0 0 0 0 0.53
Dorset 239 53 292 3 1 0 0 2 0 0 1.03
Dudley 146 53 199 1 0 0 0 0 0 1 0.50
Exeter 376 78 454 6 3 1 0 1 0 1 1.32
Glouc 194 39 233 1 0 0 0 0 0 1 0.43
Hull 323 89 412 3 1 1 0 1 0 0 0.73
Ipswi 125 31 156 0 0 0 0 0 0 0 0.00
Kent 376 68 444 3 0 0 0 3 0 0 0.68
L Barts∗ 899 171 1,070 1 0 0 0 0 0 1 0.09
L Guys∗ 607 33 640 10 2 0 1 7 0 0 1.56
L Kings 468 89 557 3 2 0 0 1 0 0 0.54
L Rfree∗ 711 94 805 18 6 1 0 10 0 1 2.24
L St.G∗ 294 55 349 4 3 0 0 1 0 0 1.15
L West∗ 1,412 35 1,447 20 0 0 1 19 0 0 1.38
Leeds∗ 513 92 605 2 2 0 0 0 0 0 0.33
Leic∗ 854 159 1,013 11 0 0 0 0 0 11 1.09
Liv Ain 179 15 194 2 0 0 0 0 0 2 1.03
Liv RI∗ 381 74 455 13 4 0 0 9 0 0 2.86
M RI∗ 481 91 572 4 1 0 0 3 0 0 0.70
Middlbr 315 18 333 4 2 0 0 2 0 0 1.20
Newc∗ 265 48 313 0 0 0 0 0 0 0 0.00
Norwch 309 59 368 7 0 0 0 0 0 7 1.90
Nottm∗ 402 92 494 0 0 0 0 0 0 0 0.00
Oxford∗ 419 92 511 6 0 0 0 0 0 6 1.17
Plymth∗ 132 47 179 0 0 0 0 0 0 0 0.00
Ports∗ 524 95 619 0 0 0 0 0 0 0 0.00
Prestn 520 65 585 2 1 0 0 1 0 0 0.34
Redng 272 88 360 0 0 0 0 0 0 0 0.00
Salford 363 113 476 6 0 0 0 0 0 6 1.26
Sheff∗ 591 62 653 25 11 0 0 11 0 3 3.83
Shrew 187 35 222 5 2 2 0 0 0 1 2.25
Stevng 412 30 442 7 1 1 1 3 1 0 1.58
Sthend 122 18 140 4 1 0 0 3 0 0 2.86
Stoke 318 82 400 6 2 0 0 1 3 0 1.50
Sund 178 17 195 5 3 0 0 1 0 1 2.56
Truro 152 26 178 4 2 0 0 2 0 0 2.25
Wirral 196 42 238 3 0 0 0 0 0 3 1.26
Wolve 307 71 378 8 4 0 0 3 1 0 2.12
York 144 25 169 5 0 0 1 4 0 0 2.96
England 19,371 3,283 22,654 261 74 11 5 113 6 52 1.15
∗
Transplant centres
310
England
Sheff *
Carlis
Colchr
York
Sthend
Liv RI *
Sund
Shrew
Truro
L Rfree *
Wolve
Norwch
Derby
Camb *
Brightn
Stevng
L Guys *
Stoke
L West *
Exeter
Bradfd
Centre
Wirral
Salford
Middlbr
Oxford *
L St.G *
Bristol *
Leic *
Carsh
Liv Ain
Dorset
B QEH *
Hull
M RI *
Kent
Basldn
L Kings
Donc
Dudley
Covnt *
Glouc
Prestn
Fig. 15.6. MSSA bacteraemia rate per 100
Leeds * prevalent dialysis patients per year by renal
L Barts * centre
0 1 2 3 4 5 6 7 Centres with no reported infection episodes are not
MSSA rate per 100 prevalent dialysis patients displayed
exclusions amongst the infections surveyed) with the centre whilst a further 96 were shared but not completed
commonest reason for exclusion being the patient was (12 of these episodes were rejected).
unknown to the parent centre. The number of bacterae- The overall infection rate for England was 0.92 per 100
mia episodes included totalled 191. Only eight of the prevalent dialysis patients per year (range 0 to 4.85)
records were both shared and completed by the parent (table 15.7). As with MSSA there was considerable
311
5 Table 15.6. Type of renal access in patients with established renal

Dotted lines show 99.9% limits failure where record shared and completed for MSSA bacteraemia
Solid lines show 95% limits episodes
4
MSSA rate per 100 dialysis patients
MSSA bacteraemia
(1/5/2011–30/4/2012)
3
2
Unknown 52
Haemodialysis
Other 0
Arteriovenous fistula 74 35.4 40.7
1
Tunnelled catheter 113 54.1
0 Perioneal dialysis 6 2.9 2.9
0 200 400 600 800 1,000 1,200 1,400 1,600
Fig. 15.7. Funnel plot of the MSSA bacteraemia rate per 100 Total known access 209
prevalent dialysis patients per year by renal centre
variation in the bacteraemia rates between centres and reduction in the number of central venous catheters
(figure 15.8). However, when centre size was taken into are likely to be amongst the contributing factors [6].
account, all the centres fell within the expected range This report also presents the first full year of reporting
(figure 15.9). of MSSA bacteraemias. There is higher incidence of
Amongst patients where the type of access was known, MSSA bacteraemia episodes in England (compared to
a slim majority (52.6%) had an arteriovenous fistula as MRSA) with an overall infection rate of 1.15 per 100
their mode of access whilst a tunnelled central venous prevalent dialysis patients per year. Again the presence
catheter was the next most common access type of a central venous catheter confers an increased risk of
(35.3%) (table 15.8). MSSA bacteraemia on the patient. There was a very
considerable between centre variation in terms of bacter-
aemia rates. These variations may be due to reporting
bias because of the voluntary nature of MSSA dialysis
Discussion information reporting and the fact that this is the first
full year. The difference in rates between MRSA and
The data presented are from one year of infections in MSSA are notable. The higher rate suggests that MSSA
ERF patients receiving dialysis that have been reported to bacteraemia continues to be a significant issue in the
PHE. This represents the fifth full year of reporting of dialysis population. Analyzing the discrepancy between
MRSA BSIs in dialysis patients. These data demonstrate the two rates is beyond the scope of this report but it
a further slight fall in the infection rate for MRSA in does raise the possibility that while screening and
England in comparison to the report in 2011 (0.25 per decolonization programmes for MRSA have been suc-
100 dialysis patients/year in 2009 vs. 0.22 per 100 dialysis cessful, the reduction of MRSA strains has left patients
patients/year in 2011). Just over half of these infections still vulnerable to MSSA.
occurred in patients with a tunnelled or non-tunnelled The first 12 months of Clostridium difficile reporting
venous catheter in comparison to patients with an show an overall infection rate of 0.61 per 100 prevalent
arteriovenous fistula. Assuming a catheter rate of 25%, dialysis patients per year and once again demonstrates
this would suggest that there remains an increased risk of a degree of variation between centres. It is again worth
infection in patients with central venous access as opposed noting that the presence of a central venous catheter
to an arteriovenous fistula. The reasons for the decline in appears to correlate with an increased risk of infection
infection rate are likely to be multifactorial. Enhanced with nearly half of dialysis patients who recorded an
screening programmes, attention to the care of access episode of CDI being dialysed via a tunnelled or
312
Table 15.7. Centre-specific data for Escherichia coli bacteraemia episodes by access type, 1/06/2011 to 30/04/2012
E. coli bacteraemia episodes

dialysis
Centre HD PD Dialysis Total AVF AVG NTC TC PD UK patientsa
B Heart 446 46 492 2 2 0 0 0 0 0 0.44
B QEHb 894 167 1,061 9 3 0 0 5 0 1 0.93
Basldn 155 26 181 1 0 0 0 1 0 0 0.60
Bradfd 196 32 228 0 0 0 0 0 0 0 0.00
Brightn 340 80 420 6 3 1 0 0 0 2 1.56
Bristolb 474 66 540 3 2 1 0 0 0 0 0.61
Cambb 371 41 412 8 0 0 0 0 0 8 2.12
Carlis 66 24 90 4 1 0 0 0 1 2 4.85
Carsh 753 103 856 14 4 0 1 7 2 0 1.78
Chelms 119 26 145 0 0 0 0 0 0 0 0.00
Colchr 120 0 120 4 1 0 0 3 0 0 3.64
Covntb 362 90 452 2 1 0 0 1 0 0 0.48
Derby 207 112 319 2 2 0 0 0 0 0 0.68
Donc 162 26 188 0 0 0 0 0 0 0 0.00
Dorset 239 53 292 0 0 0 0 0 0 0 0.00
Dudley 146 53 199 2 0 0 0 0 0 2 1.10
Exeter 376 78 454 3 2 0 0 1 0 0 0.72
Glouc 194 39 233 5 0 0 0 0 0 5 2.34
Hull 323 89 412 4 4 0 0 0 0 0 1.06
Ipswi 125 31 156 0 0 0 0 0 0 0 0.00
Kent 376 68 444 5 3 0 0 2 0 0 1.23
L Bartsb 899 171 1,070 1 0 0 0 0 0 1 0.10
L Guysb 607 33 640 6 4 0 0 2 0 0 1.02
L Kings 468 89 557 3 2 0 0 1 0 0 0.59
L Rfreeb 711 94 805 6 3 0 0 3 0 0 0.81
L St.Gb 294 55 349 0 0 0 0 0 0 0 0.00
L Westb 1,412 35 1,447 5 0 0 0 0 0 5 0.38
Leedsb 513 92 605 11 7 1 0 3 0 0 1.98
Leicb 854 159 1,013 13 0 0 0 0 0 13 1.40
Liv Ain 179 15 194 4 0 0 0 0 0 4 2.25
Liv RIb 381 74 455 5 3 0 2 0 0 0 1.20
M RIb 481 91 572 4 1 0 0 3 0 0 0.76
Middlbr 315 18 333 5 0 0 0 0 0 5 1.64
Newcb 265 48 313 6 0 0 0 0 0 6 2.09
Norwch 309 59 368 1 0 0 0 0 0 1 0.30
Nottmb 402 92 494 0 0 0 0 0 0 0 0.00
Oxfordb 419 92 511 5 2 3 0 0 0 0 1.07
Plymthb 132 47 179 0 0 0 0 0 0 0 0.00
Portsb 524 95 619 1 0 0 0 1 0 0 0.18
Prestn 520 65 585 9 5 0 0 2 1 1 1.68
Redng 272 88 360 0 0 0 0 0 0 0 0.00
Salford 363 113 476 1 0 0 0 1 0 0 0.23
Sheffb 591 62 653 8 3 0 1 3 1 0 1.34
Shrew 187 35 222 3 2 0 0 0 0 1 1.47
Stevng 412 30 442 6 2 0 0 4 0 0 1.48
Sthend 122 18 140 4 3 0 0 1 0 0 3.12
Stoke 318 82 400 0 0 0 0 0 0 0 0.00
Sund 178 17 195 2 2 0 0 0 0 0 1.12
Truro 152 26 178 1 0 0 0 1 0 0 0.61
Wirral 196 42 238 1 0 0 0 0 0 1 0.46
Wolve 307 71 378 4 3 0 0 0 1 0 1.15
York 144 25 169 2 0 0 0 2 0 0 1.29
England 19,371 3,283 22,654 191 70 6 4 47 6 58 0.92
a
Rate per year calculated from the eleven month collection period; bTransplant centres
313
England
Carlis
Colchr
Sthend
Glouc
Liv Ain
Camb *
Newc *
Leeds *
Carsh
Prestn
Middlbr
Brightn
Stevng
Shrew
Leic *
Sheff *
York
Kent
Liv RI *
Wolve
Centre
Sund
Dudley
Oxford *
Hull
L Guys *
B QEH *
L Rfree *
M RI *
Exeter
Derby
Truro
Bristol *
Basldn
L Kings
Covnt *
Wirral
B Heart
L West *
Norwch
Salford
Fig. 15.8. Escherichia coli bacteraemia
Ports *
rate per 100 prevalent dialysis patients per
L Barts *
year by renal centre
0 1 2 3 4 5 6 7 8 Centres with no reported infection episodes are not
E. coli rate per 100 prevalent dialysis patients displayed
non-tunnelled catheter at the time. This may underline about recent marked increases in the number of cases
the vulnerability to infection in this group of patients [7]. However, reporting of E. coli bacteraemia in patients
and the increased likelihood that they are exposed to in ERF is relatively new and as a result there was incon-
courses of antibiotics. sistency in reporting by microbiology laboratories and a
Lastly the report also considers the first eleven months high proportion of records were excluded due to the
of Escherichia coli reporting (beginning in June 2011). A patient not being in or known to the allocated main
national system for capturing data on E. coli bacteraemia centre. There were again noticable variations in infection
has been established in England in response to concern rate between centres, although this variation should be
314
6 these infections. Increased awareness of infection report-

Solid lines show 95% limits ing amongst both renal centres and microbiology units
5 would also help to improve the robustness of this data set.
E. coli rate per 100 dialysis patients
Summary
3
The data presented on bacteraemias occurring in ERF

2
are as reported to Public Health England. These data
demonstrate a further fall in the number and rate of
1 MRSA bloodstream infections in England continuing
the downward trend observed over the previous five
0 years. They also show a substantial incidence of MSSA
0 200 400 600 800 1,000 1,200 1,400 1,600
Number of dialysis patients
BSI in the first 12 months of reporting. Data are also
included for CDI and E. coli BSI. In each infection the
Fig. 15.9. Funnel plot of the Escherichia coli bacteraemia rate per presence of a central venous catheter appears to confer
100 prevalent dialysis patients per year by renal centre
a greater risk. Considerable regional variation is noted
treated with caution because of the inconsistency in that may be at least partially explained by differences in
reporting. It is also worth noting that nationally, the reporting policies. Further work is needed to establish
reported rates of E. coli bacteraemia are more than the overall trend in MSSA, CDI and E. coli. Finally,
three times that of MSSA so it is possible that there there is a need for consistency of reporting which
were a similar number of infections reported would enable trends to be more clearly defined.
inaccurately.
It is again noticeable that a high proportion of E. coli Conflicts of interest: none
infections occur in patients with a tunnelled catheter.
E. coli is traditionally associated with urinary tract and
other infections more than catheter related sepsis.
Again this may highlight the increased vulnerability of Acknowledgements
patients reliant on lines for their dialysis access. Further
work is needed over the next cycle to identify trends in The authors wish to acknowledge the help of our
colleagues at renal centres across the country for their
assistance in compiling this report.
Table 15.8. Type of renal access in patients with established renal
failure where record shared and completed for Escherichia coli BSI
episodes
Escherichia coli BSI References

(1/6/2011–30/4/2012)
1 Bray BD, Boyd J, Daly C, Donaldson K, Doyle A, Fox JG, et al. Vascular
Renal access type N % Access class access type and risk of mortality in a national prospective cohort of
haemodialysis patients. Qjm-an International Journal of Medicine.
Unknown 50 2012;105(11):1097–1103
Haemodialysis 2 https://www.gov.uk/government/uploads/system/uploads/attachment_
data/file/215135/dh_133016.pdf
Other 8
3 Fluck R, Wilson J, Tomson CRV. UK Renal Registry 12th Annual Report
Arteriovenous fistula 70 52.6 57.1 (December 2009): Chapter 12 Epidemiology of methicillin resistant
Arteriovenous graft 6 4.5 Staphylococcus aureus Bacteraemia Amongst Patients Receiving Dialysis
Non-tunnelled catheter 4 3.0 38.3 for Established Renal Failure in England in 2008: a joint report from
Tunnelled catheter 47 35.3 the UK Renal Registry and the Health Protection Agency. Nephron
Peritoneal dialysis 6 4.5 4.5 Clinical Practice. 2010;115:c261–c270
4 Crowley L, Wilson J, Guy R, Pitcher D, Fluck R. UK Renal Registry 14th
Total 191 Annual Report: Chapter 12 Epidemiology of Staphylococcus aureus
Total known access 133 Bacteraemia Amongst Patients Receiving Dialysis for Established Renal
Failure in England in 2009 to 2011: A Joint Report from the Health
315
Protection Agency and the UK Renal Registry. Nephron Clinical Practice. bacteraemia: data from the national mandatory surveillance of MRSA
2012;120:c233–c245 bacteraemia in England, 2006–2009 J Hosp Infect. 2011 Nov;79(3):
5 Shaw C, Pruthi R, Pitcher D, Fogarty D. UK Renal Registry 15th Annual 211–217
Report: Chapter 2 UK RRT Prevalence in 2011: National and Centre- 7 Wilson J, Elgohari S, Livermore DM, Cookson B, Johnson A, Lamagni T,
Specific Analyses. Nephron Clinical Practice. 2013;123:29–54 Chronias A, Sheridan E. Trends among pathogens reported as causing
6 Wilson J, Guy R, Elgohari S, Sheridan E, Davies J, Lamagni T, Pearson A. bacteraemia in England, 2004–2008 Clin Microbiol Infect. 2011
Trends in sources of methicillin-resistant Staphylococcus aureus (MRSA) Mar;17(3):451–458
316
Appendix A: The UK Renal Registry Statement of
Purpose
This appendix is available on the web only and can be found at http://www.karger.com/Journal/Home/228539 or
www.renalreg.com
Appendix B: Definitions and Analysis Criteria

www.renalreg.com
Appendix C: Renal Services Described for

Non-physicians
www.renalreg.com
Appendix D: Methodology used for Analyses of

PCT/HB Incidence and Prevalence
Rates and of Standardised Ratios
www.renalreg.com
Appendix E: Methodology for Estimating

Catchment Populations of Renal
Centres in England for Dialysis
Patients
www.renalreg.com
317
Appendix F: Additional data tables for 2011

incident and prevalent patients
www.renalreg.com
Appendix G: UK Renal Registry Dataset

Specification
This appendix is available on the web only and can be found at www.renalreg.com
Appendix H: Coding: Ethnicity, EDTA Primary

Renal Diagnoses, EDTA Causes of
Death
www.renalreg.com
318
Appendix I Acronyms and Abbreviations
used in the Report
ACE (inhibitor) Angiotensin converting enzyme (inhibitor)

AKI Acute kidney injury
ANZDATA Australia and New Zealand Dialysis and Transplant Registry
APD Automated peritoneal dialysis
ADPKD Autosomal dominant polycystic kidney disease
APKD Adult polycystic kidney disease
ATTOM Access to transplant and transplant outcome measures
AV Arteriovenous
AVF Arteriovenous fistula
AVG Arteriovenous graft
BAPN British Association of Paediatric Nephrology
BCG Bromocresol green
BCP Bromocresol purple
BMD Bone mineral disease
BMI Body mass index
BP Blood pressure
BSI Blood stream infection
BTS British Transplant Society
Ca Calcium
CAB Clinical Affairs Board (Renal Association)
CABG Coronary artery bypass grafting
CAPD Continuous ambulatory peritoneal dialysis
CCL Clinical Computing Limited
CCPD Cycling peritoneal dialysis
CDI Clostridium difficile infection
Chol Cholesterol
CHr Target reticulocyte Hb content
CI Confidence interval
CK Creatine kinase
CKD Chronic kidney disease
CKD-EPI Chronic kidney disease epidemiology collaboration
CK-MB Creatine kinase isoenzyme MB
COPD Chronic obstructive pulmonary disease
CRF Chronic renal failure
cRF Calculated HLA antibody reaction frequency
CRP C-reactive protein
CVVH Continuous veno-venous haemofiltration
CXR Chest x-ray
DBP Diastolic blood pressure
DCCT Diabetes Control and Complications Trial
DH Department of Health
319
DM Diabetes mellitus
DOB Date of birth
DOPPS Dialysis Outcomes and Practice Patterns Study
E&W England and Wales
E, W & NI England, Wales and Northern Ireland
EBPG European Best Practice Guidelines
ECG Electrocardiogram
EDTA European Dialysis and Transplant Association
EF Error factor
eGFR Estimated glomerular filtration rate
Ei Expected cases in area i
ECD Extended Criteria Donor
EDTA European Dialysis and Transplant Association
eKt/V Equilibrated Kt/V
EPO Erythropoietin
ERA European Renal Association
ERA–EDTA European Renal Association–European Dialysis and Transplant Association
ERF Established renal failure
ESA Erythropoiesis stimulating agent
ESRD End stage renal disease
ESRF End stage renal failure
EWNI England, Wales and Northern Ireland
Ferr Ferritin
FEV1 Forced expiratory volume in 1 second
FVC Forced vital capacity
GFR Glomerular filtration rate
GH Growth hormone
GN Glomerulonephritis
HA Health Authority
HB Health board
Hb Haemoglobin
HbA1c Glycated Haemoglobin
HBeAg Hepatitis B e antigen
HCAI-DCS Healthcare-associated infection data collection system
HD Haemodialysis
HDF Haemodialysis filtration
HDL High-density lipoprotein
HLA Human leucocyte antigen
HPA Health Protection Agency
HQIP Health Quality Improvement Partnership
HR Hazard ratio
HRC Hypochromic red blood cells
Ht Height
ICU Intensive care unit
IDMS Isotope dilution mass spectrometry
IDOPPS International Dialysis Outcomes and Practice Patterns Study
IFCC International Federation of Clinical Chemistry & Laboratory Medicine
IHD Ischaemic heart disease
IMD Index of Multiple Deprivation
IOTF International Obesity Taskforce
IPD Intermittent peritoneal dialysis
IQR Inter-quartile range
ISPD International Society for Peritoneal Dialysis
IT Information technology
IU International units
IV Intra venous
KDIGO Kidney Disease: Improving Global Outcomes
KDOQI Kidney Disease Outcomes Quality Initiative
KM Kaplan Meier
Kt/V Ratio between the product of urea clearance (K, in ml/min) and dialysis session duration (t, in minutes) divided
by the volume of distribution of urea in the body (V, in ml)
320
Appendix I Acronyms and abbreviations
LA Local Authority
LCL Lower confidence limit
LDL Low-density lipoprotein
LTFU Lost to follow-up
M:F Male:Female
MAP Mean arterial blood pressure
MDRD Modification of diet in renal disease
MI Myocardial infarction
MMF Mycophenolate mofetil
MRSA Methicillin resistant Staphylococcal aureus
MSSA Methicillin sensitive Staphylococcal aureus
N Number
N Ireland Northern Ireland
NE North East
NEQAS UK National External Quality Assessment Scheme
NHBPEP National high blood pressure education programme
NHS National Health Service
NHS BT National Health Service Blood and Transplant
NI Northern Ireland
NICE National Institute for Health and Clinical Excellence
NISRA Northern Ireland Statistic and Research Agency
NMO Non-mixed origin
NRS National Records of Scotland
NSF National service framework
NTC Non-tunnelled dialysis catheter
NTL Non-tunnelled line
NW North West
O/E Observed/expected
ODT Organ Donation and Transplantation (a Directorate of NHS Blood and transplant)
Oi Observed cases in area i
ONS Office for National Statistics
OR Odds ratio
PAS Patient Administration System
PCT Primary Care Trust
PD Peritoneal dialysis
PHE Public Health England
Phos Phosphate
PIAG Patient Information Advisory Group
PKD Polycystic kidney disease
PMARP Per million age related population
PMCP Per million child population
PMP Per million population
PP Pulse pressure
PRD Primary renal disease
PTH Parathyroid hormone
PUV Posterior urethral valves
PVD Peripheral vascular disease
QOF Quality and Outcomes Framework
QUEST Quality European Studies
RA Renal Association
rhGH Recombinant human growth hormone
RI Royal Infirmary
RNSF Renal National Service Framework (or NSF)
RR Relative risk
RRDSS Renal Registry data set specification
RRT Renal replacement therapy
RVD Renovascular disease
SAR Standardised acceptance ratio (=O/E)
SAS Statistical Analysis System
SBP Systolic blood pressure
SD Standard deviation
321
SES Socio-economic status

SHA Strategic health authority
SHARP Study of Heart and Renal Protection
SI System International (units)
SMR Standardised mortality ratios
spKt/V Single pool Kt/V
SPR Standardised prevalence ratio (=O/E)
SR Standardised ratio (used to cover either SAR or SPR)
SUS Secondary uses service
SW South West
TC Tunnelled dialysis catheter
TL Tunnelled line
TSAT Transferrin saturation
TWL Transplant waiting list
Tx Transplant
UCL Upper confidence limit
UK United Kingdom
UKRR UK Renal Registry
UKT UK Transplant (now ODT)
URR Urea reduction ratio
US United States
USA United States of America
USRDS United States Renal Data System
WHO World health organization
Wt Weight
322
Appendix J Laboratory Conversion
Factors
Conversion factors from SI units
Albumin g/dl = g/L × 0.1

Aluminium mg/L = mmol/L × 27.3
Bicarbonate mg/dl = mmol/L × 6.1
Calcium mg/dl = mmol/L × 4
Calcium × phosphate mg2/dl2 = mmol2/L2 × 12.4
Cholesterol mg/dl = mmol/L × 38.6
Creatinine mg/dl = mmol/L × 0.011
Glucose mg/d = mmol/L × 18
Phosphate mg/dl = mmol/L × 3.1
PTH ng/L = pmol/L × 9.5
Urea mg/dl = mmol/L × 6.0
Urea nitrogen mg/dl = mmol/L × 2.8
323
Appendix K Renal Centre Names and
Abbreviations used in the Figures and
Data Tables
Adult Centres
City Hospital Abbreviation
England
Basildon Basildon Hospital Basldn
Birmingham Heartlands Hospital B Heart
Birmingham Queen Elizabeth Hospital B QEH
Bradford St Luke’s Hospital Bradfd
Brighton Royal Sussex County Hospital Brightn
Bristol Southmead Hospital Bristol
Cambridge Addenbrooke’s Hospital Camb
Carlisle Cumberland Infirmary Carlis
Carshalton St Helier Hospital Carsh
Chelmsford Broomfield Hospital Chelms
Colchester Colchester General Hospital Colchr
Coventry University Hospital Coventry Covnt
Derby Royal Derby Hospital Derby
Doncaster Doncaster Royal Infirmary Donc
Dorset Dorset County Hospital Dorset
Dudley Russells Hall Hospital Dudley
Exeter Royal Devon and Exeter Hospital Exeter
Gloucester Gloucestershire Royal Hospital Glouc
Hull Hull Royal Infirmary Hull
Ipswich Ipswich Hospital Ipswi
Kent Kent and Canterbury Hospital Kent
Leeds St James’s University Hospital and Leeds General Infirmary Leeds
Leicester Leicester General Hospital Leic
Liverpool Aintree University Hospital Liv Ain
Liverpool Royal Liverpool University Hospital Liv RI
London St. Bartholomew’s Hospital and The Royal London Hospital L Barts
London St George’s Hospital and Queen Mary’s Hospital L St. G
London Guy’s Hospital and St Thomas’ Hospital L Guys
London Hammersmith, Charing Cross, St Mary’s L West
London King’s College Hospital L Kings
London Royal Free, Middlesex and UCL Hospitals L Rfree
Manchester Manchester Royal Infirmary M RI
Middlesbrough The James Cook University Hospital Middlbr
Newcastle Freeman Hospital and Royal Victoria Infirmary Newc
Norwich Norfolk and Norwich University Hospital Norwch
325
City Hospital Abbreviation
Nottingham Nottingham City Hospital Nottm

Oxford John Radcliffe Hospital and Churchill Hospital Oxford
Plymouth Derriford Hospital Plymth
Portsmouth Queen Alexandra Hospital Ports
Preston Royal Preston Hospital Prestn
Reading Royal Berkshire Hospital Redng
Salford Salford Royal Hospital Salford
Sheffield Northern General Hospital Sheff
Shrewsbury Royal Shrewsbury Hospital Shrew
Southend Southend Hospital Sthend
Stevenage Lister Hospital Stevng
Stoke University Hospital of North Staffordshire Stoke
Sunderland Sunderland Royal Hospital Sund
Truro Royal Cornwall Hospital Truro
Wirral Arrowe Park Hospital Wirral
Wolverhampton New Cross Hospital Wolve
York The York Hospital York
Wales
Bangor Ysbyty Gwynedd Bangor
Cardiff University Hospital of Wales Cardff
Clwyd Glan Clwyd Hospital Clwyd
Swansea Morriston Hospital Swanse
Wrexham Wrexham Maelor Hospital Wrexm
Scotland
Aberdeen Aberdeen Royal Infirmary Abrdn
Airdrie Monklands Hospital Airdrie
Dumfries Dumfries & Galloway Royal Infirmary D & Gall
Dundee Ninewells Hospital Dundee
Dunfermline Queen Margaret Hospital Dunfn
Edinburgh Royal Infirmary of Edinburgh Edinb
Glasgow Western Infirmary, Glasgow Royal Infirmary and Stobhill Hospitals Glasgw
Inverness Raigmore Hospital Inverns
Kilmarnock University Hospital Crosshouse Klmarnk
Northern Ireland
Antrim Antrim Area Hospital Antrim
Belfast Belfast City Hospital Belfast
Londonderry & Omagh Altnagelvin Area and Tyrone County Hospitals West NI
Newry Daisy Hill Hospital Newry
Ulster Ulster Hospital Ulster
Paediatric Centres
City Hospital Abbreviation Country
Belfast Royal Belfast Hospital for Sick Children Blfst_P N Ireland

Birmingham Birmingham Children’s Hospital Bham_P England
Bristol Bristol Royal Hospital for Children Brstl_P England
Cardiff KRUF Children’s Kidney Centre Cardf_P Wales
Glasgow Royal Hospital for Sick Children Glasg_P Scotland
Leeds Leeds Children’s Hospital Leeds_P England
Liverpool Alder Hey Children’s Hospital Livpl_P England
London Guy’s Hospital (Evelina) – Paediatric L Eve_P England
London Great Ormond Street Hospital for Children LGOSH_P England
Manchester Royal Manchester Children’s Hospital Manch_P England
Newcastle Great North Children’s Hospital Newc_P England
Nottingham Nottingham Children’s Hospital Nottm_P England
Southampton Southampton General Hospital – Paediatric Soton_P England
326

Report 2013

Uploaded by

Copyright:

Available Formats

Report 2013

Uploaded by

Document Information

Original Description:

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

Report 2013

Uploaded by

Copyright:

Available Formats

The Renal Association

UK Renal Registry 16th Annual Report: Introduction

UK Renal Registry 16th Annual Report: Introduction 1

Chapter 2 UK RRT Prevalence in 2012: National and Centre-specific Analyses 37

Chapter 9 Adequacy of Haemodialysis in UK Adult Patients in 2012: National and Centre-specific

Chapter 13 Clinical, Haematological and Biochemical Parameters in Patients Receiving Renal

Appendix B Definitions and Analysis Criteria 317

Appendix C Renal services described for non-physicians 317

Appendix G UK Renal Registry dataset specification 318

Appendix I Acronyms and Abbreviations used in the Report 319

Appendix J Laboratory Conversion Factors 323

Damian Fogarty, Ron Cullen

. In 2012, 88% of prevalent HD patients achieved a

Primary Date Cause Average

Ulster 100.0 100.0 100.0 100.0 100.0 100.0 N Ireland

Primary Date Cause Average

Liv Ain 100.0 100.0 100.0 0.0 0.0 60.0 England

Abrdn 100.0 65.7 Scotland

Julie Gilga, Anirudh Raoa, Damian Fogartyab

Introduction social and demographic factors such as underlying

England N Ireland Scotland Wales UK

Number starting RRT 5,826 186 519 360 6,891

centres, ethnicity coding is performed by clinical staff and

Table 1.2. Continued

Table 1.2. Continued

Table 1.2. Continued

2.8 ethnic minority populations in general compared with

Table 1.3. Number of patients starting RRT by renal centre 2007–2012

Year Catchment 2012

Table 1.3. Continued

Year Catchment 2012

Year Age group (years)

Fig. 1.6. Median age of incident RRT patients by centre in 2012

80 diagnoses (partly due to the reasons mentioned above).

Percentage in each ethnic group

Table 1.6. Continued

Percentage in each ethnic group

Table 1.7. Continued

Diagnosis England N Ireland Scotland Wales UK

Diabetes 25.5 22.5 27.9 31.9 25.9

Status at 90 days of only those

Table 1.11. Continued

Status at 90 days of only those

Modality change over time

Age ,65 (%) Age 565 (%) All patients (%)

Table 1.12. Continued

Age ,65 (%) Age 565 (%) All patients (%)

First treatment N Later modality 90 days 1 year 3 years 5 years

starting RRT allocated and thus, the eGFR used for

PD sis, cryoglobulinemic glomerulonephritis, myelomatosis/

estimated and to allow these to be shown for subgroups of patients. Results

England Norwch 87 74 93.1 64.9

Fig. 1.11. Percentage presenting late (2011/2012)

Percentage presenting late

Centre N with data Overall (95% CI) Non-acute∗ Non-diab PRD

Table 1.15. Continued

Percentage presenting late

Centre N with data Overall (95% CI) Non-acute∗ Non-diab PRD