Minimally Invasive Cardiac Surgery

Minimally Invasive
Cardiac Surgery
SECOND EDITION
Edited by
Daniel J. Goldstein, MD
Mehmet C. Oz, MD
i
MINIMALLY INVASIVE CARDIAC SURGERY
ii
CONTEMPORARY CARDIOLOGY
CHRISTOPHER P. CANNON, MD
SERIES EDITOR
Coronary Disease in Women: Evidence- Essentials of Bedside Cardiology: With a

Based Diagnosis and Treatment, Complete Course in Heart Sounds and
edited by Leslee J. Shaw, PhD and Murmurs on CD, Second Edition,
Rita F. Redberg, MD, FACC, 2004 by Jules Constant, MD, 2003
Complementary and Alternative Medicine Primary Angioplasty in Acute Myocardial Infarc-
in Cardiovascular Disease, tion, edited by James E. Tcheng, MD, 2002
edited by Richard A. Stein, MD and Cardiogenic Shock: Diagnosis and Treatment,
Mehmet C. Oz, MD, 2004 edited by David Hasdai, MD, Peter B.
Nuclear Cardiology, The Basics: How to Set Berger, MD, Alexander Battler, MD, and
Up and Maintain a Laboratory, by David R. Holmes, Jr., MD, 2002
Frans J. Th. Wackers, MD, PhD, Wendy Management of Cardiac Arrhythmias, edited
Bruni, CNMT, and Barry L. Zaret, MD, by Leonard I. Ganz, MD, 2002
P hD, 2004 Diabetes and Cardiovascular Disease, edited by
Minimally Invasive Cardiac Surgery, Michael T. Johnstone, MD and Aristidis
Second Edition, edited by Daniel J. Veves, MD, DSC, 2001
Goldstein, MD, and Mehmet C. Oz, MD, Blood Pressure Monitoring in Cardiovascular
2003 Medicine and Therapeutics, edited by
Cardiovascular Health Care Economics, William B. White, MD, 2001
edited by William S. Weintraub, MD, 2003 Vascular Disease and Injury: Preclinical Research,
Platelet Glycoprotein IIb/IIIa Inhibitors in edited by Daniel I. Simon, MD, and Campbell
Cardiovascular Disease, Second Edition, Rogers, MD, 2001
edited by A. Michael Lincoff, MD, 2003 Preventive Cardiology: Strategies for the
Heart Failure: A Clinician’s Guide to Ambulatory Prevention and Treatment of Coronary
Diagnosis and Treatment, edited by Mariell Artery Disease, edited by JoAnne Micale
L. Jessup, MD, and Evan Loh, MD, 2003 Foody, MD, 2001
Management of Acute Coronary Syndromes, Nitric Oxide and the Cardiovascular System,
Second Edition, edited by Christopher edited by Joseph Loscalzo, MD, PhD, and
P. Cannon, MD, 2003 Joseph A. Vita, MD, 2000
Aging, Heart Disease, and Its Management: Annotated Atlas of Electrocardiography:
Facts and Controversies, edited by Niloo A Guide to Confident Interpretation,
M. Edwards, MD, Mathew S. Maurer, MD, by Thomas M. Blake, MD, 1999
and Rachel B. Wellner, MPH, 2003 Platelet Glycoprotein IIb/IIIa Inhibitors in
Peripheral Arterial Disease: Diagnosis and Cardiovascular Disease,
Treatment, edited by Jay D. Coffman, edited by A. Michael Lincoff, MD, and
MD, and Robert T. Eberhardt, MD, 2003 Eric J. Topol, MD, 1999
Cardiac Repolarization: Bridging Basic and Minimally Invasive Cardiac Surgery, edited by
Clinical Science, edited by Ihor Gussak, Mehmet C. Oz, MD, and Daniel J.
MD , PhD, Charles Antzelevitch, Ph D, Goldstein, MD, 1999
Stephen C. Hammill, MD, Win K. Shen, Management of Acute Coronary Syndromes,
MD, and Preben Bjerregaard, MD, DMSc, edited by Christopher P. Cannon, MD,
2003 1999
iii
MINIMALLY INVASIVE
CARDIAC SURGERY
Second Edition
Edited by
DANIEL J. GOLDSTEIN, MD
Newark Beth Israel Medical Center, Newark, NJ
and
MEHMET C. OZ, MD
New York Presbyterian Hospital, New York, NY
HUMANA PRESS
TOTOWA, NEW JERSEY
iv
© 2004 Humana Press Inc.
999 Riverview Drive, Suite 208
Totowa, New Jersey 07512
www.humanapress.com
For additional copies, pricing for bulk purchases, and/or information about other Humana titles,
contact Humana at the above address or at any of the following numbers: Tel.: 973-256-1699;
Fax: 973-256-8341, E-mail: [email protected]; or visit our Website: www.humanapress.com
All rights reserved.
No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means,
electronic, mechanical, photocopying, microfilming, recording, or otherwise without written permission from the
Publisher.
All articles, comments, opinions, conclusions, or recommendations are those of the author(s), and do not necessarily reflect
the views of the publisher.
Due diligence has been taken by the publishers, editors, and authors of this book to assure the accuracy of the information
published and to describe generally accepted practices. The contributors herein have carefully checked to ensure that the
drug selections and dosages set forth in this text are accurate and in accord with the standards accepted at the time of
publication. Notwithstanding, as new research, changes in government regulations, and knowledge from clinical experi-
ence relating to drug therapy and drug reactions constantly occurs, the reader is advised to check the product information
provided by the manufacturer of each drug for any change in dosages or for additional warnings and contraindications.
This is of utmost importance when the recommended drug herein is a new or infrequently used drug. It is the responsibility
of the treating physician to determine dosages and treatment strategies for individual patients. Further it is the responsi-
bility of the health care provider to ascertain the Food and Drug Administration status of each drug or device used in their
clinical practice. The publisher, editors, and authors are not responsible for errors or omissions or for any consequences
from the application of the information presented in this book and make no warranty, express or implied, with respect to
the contents in this publication.
Cover illustrations: EndoWrist® Robotic Surgical Instruments © 2003 Intuitive Surgical, Inc. CORx System® ©2003
CardioVention, Inc. Figure 14 from “Minimally Invasive Conduit Harvesting” by K. D. Accola, et al. (Chapter 14).
Figure 2 from “Thoracoscopic Pericardial Surgery” by P. M. McFadden (Chapter 23).
Cover design by Patricia F. Cleary.
This publication is printed on acid-free paper. ∞

ANSI Z39.48-1984 (American National Standards Institute) Permanence of Paper for Printed Library Materials.
Photocopy Authorization Policy:

Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted
by Humana Press Inc., provided that the base fee of US $20 is paid directly to the Copyright Clearance Center at 222
Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license from the
CCC, a separate system of payment has been arranged and is acceptable to Humana Press Inc. The fee code for users
of the Transactional Reporting Service is: [1-58829-170-7/03 $25].
Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1
1-59259-416-6 (e-ISBN)
Library of Congress Cataloging-in-Publication Data

Minimally invasive cardiac surgery / edited by David J. Goldstein and Mehmet C. Oz.--2nd ed.
p.;cm. -- (Contemporary cardiology)
Includes biblioographical references and index.
ISBN 1-58829-170-7 (alk. paper)
1. Heart–Surgery–Miscellanea. 2. Coronary artery bypass–Miscellanea. 3. Operations, Surgical–Miscella
nea. I. Goldstein, David J., MD II. Oz, Mehmet, 1960- III. Contemporary cardiology (Totowa, N.J.: Unnum
bered)
[DNLM: 1. Cardiac Surgical Procedures. 2. Surgical Procedures, Minimally Invasive. WG 169 M665 2003]
RD598.M525 2003
617.4'12–dc21 2003040673
PREFACE v
In 1962, Thomas Kuhne coined the term “paradigm shift” while arguing that human
knowledge advances by quantum leaps with interspersed smaller steps. Preparation for
the major advance is generally not a concerted effort by thought leaders. Rather, a few
(or one) visionaries gain insights into a process and are able to definitely demonstrate the
accuracy of their worldview. Often, the epiphany does not occur during the intellectual
lifetime of the discoverers. Medicine has had numerous such “paradigm shifts” including
the compelling reworking of Galen’s concepts of the body. Of note, the scientific world
of the time explained the new views by arguing that the human body must have changed
between the time of ancient Greece and modern Europe.
The inauguration of cardiac surgery itself required profound shifts in medicine’s view
of physiology. Yet, over the ensuing 40 years, the field was fine tuned so we could
provide greater than 95% success rates in elective surgery with low cost and short hos-
pital stays. In some parts of the world, the procedures were viewed as commodities and
prices dropped as providers were unable to differentiate the quality of their work. As
patients and their physicians became more demanding, the desire to make the procedures
“minimally invasive” grew. In effect, what we were really searching for was a life saving
procedure that also preserved quality of life. In short, “minimally invasive” has really
been a code phrase for procedures that disrupt our quality of life the least. We will use
this definition as we explain the constant trade-offs between incision size and location,
cardiopulmonary bypass time, and expensive new technologies. The constant bottom
line is our need to bring value to the health care system and our patients, recognizing that
the early forays may not reflect longer term potential.
Our first edition argued that this evolution was really a revolution, as stunning reports
from countries outside North America, the birthplace of cardiac surgery, revealed that
heart surgery could be successfully performed without arresting the heart. In fact, these
investigators were able to export these radical concepts more readily to heart surgeons
around the globe because many in the field perceived the stagnation that arises when a
field has been mastered. Heart surgeons, generally a restless breed, often choose our
demanding field because they enjoy the challenge and the opportunity to engage these
new procedures was alluring. Many took the bait, not realizing the costs incurred, and
in some areas the pendulum of growth has swung back towards the status quo. If for no
other reason than to provide medium term updates, a second edition was warranted.
In addition, the overall rapid growth in the range and volume of minimally invasive
procedures since the first edition mandated an update. In this evolutionary cycle, some
procedures have become extinct even as others have become mainstays. New approaches,
unthinkable even three years ago, are being presented at our most prestigious national
meetings. In particular, the invasion of robotic technologies into the traditionally con-
servative cardiac operating theatre paralleled the stunning growth of laparoscopic chole-
cystectomy once it was discovered in the rural southeastern United States. Once the
possibility of this radically new approach was comprehended, new generations of pro-
v
vi Preface
cedures were inaugurated. Surgeons do not just perform the old operation with new tools.
We should create new procedures that are adapted to the new tools.
Finally, the fictitious border between different specialties focusing on atherosclerosis,
especially cardiac surgery, cardiology, and invasive radiology, has blurred. Our different
lineages are less important than our current procedural needs and the evolution of mini-
mally invasive heart surgery will demand a reworking of training programs and mindsets
—a healthy process that will mature over the next decade.
Daniel Goldstein, MD
Mehmet Oz, MD
CONTENTS vii
Preface ............................................................................................................................ v
Contributors ................................................................................................................... xi
Part I Physiology of Coronary Bypass Grafting With and

Without Cardiopulmonary Bypass
1 Pathophysiology of Cardiopulmonary Bypass ................................. 3
Ron G. H. Speekenbrink, Wim van Oeveren,
Charles R. H. Wildevuur, and León Eijsman
2 Endothelial Injury During Minimally Invasive
Bypass Grafting .......................................................................... 27
Robert J. Dabal, Craig R. Hampton, and Edward D. Verrier
Part II Minimally Invasive Coronary Bypass Grafting

3 Minimally Invasive Bypass Grafting:
A Historical Perspective ............................................................. 45
Enio Buffolo and Luís Roberto Gerola
4 Tailoring Minimal Invasive Coronary Bypass to the Patient ........ 53
Michael A. Borger and M. Anno Diegeler
5 OPCAB: A Primer on Technique ................................................... 67
John D. Puskas
6 Mechanical Stabilization: The Medtronic Octopus System ........... 75
Michael Gibson, Robert B. Beauford,
and Daniel J. Goldstein
7 Mechanical Stabilization Systems:
The Genzyme-OPCAB Elite System ........................................... 89
William E. Cohn and Marc Ruel
8 Mechanical Stabilization Systems:
The Guidant OPCAB System .................................................... 105
Marc W. Connolly and Valavanur A. Subramanian
9 The MIDCAB Operation .............................................................. 113
Mercedes K. C. Dullum and Albert J. Pfister
10 Minimally Invasive Coronary Artery Bypass Grafting
on the Beating Heart: The European Experience .................... 129
Antonio M. Calafiore, Michele Di Mauro, Alessandro Pardini,
Antonio Bivona, and Stefano D’Alessandro
vii
viii
viii Contents
11 Minimally Invasive Coronary Artery Bypass Grafting:

The South American Experience .............................................. 147
Federico J. Benetti and Maximo Guida
12 Reoperative Off-Pump CABG ...................................................... 157
Valavanur A. Subramanian, James D. Fonger,
and Nilesh U. Patel
13 Perioperative Evaluation of Graft Patency in OPCAB ................ 179
Vinod H. Thourani and John D. Puskas
14 Minimally Invasive Conduit Harvesting ...................................... 193
Kevin D. Accola, Mike Butkus, and Brenda Dickey
15 Hybrid Revascularization ............................................................. 207
Uwe Klima and Axel Haverich
16 Neurocognitive Issues in Off-Pump CABG ................................. 217
Ronald M. Lazar and Daniel F. Heitjan
17 Multivessel Off-Pump Revascularization
in High-Risk Patients: Severe Left Ventricular
Dysfunction ............................................................................... 229
Daniel J. Goldstein, Robert B. Beauford, Patricia Garland,
and Craig R. Saunders
18 Multivessel Off-Pump Revascularization
in High-Risk Populations: Octogenarians ............................... 241
Frederic Sardari, Robert B. Beauford,
and Daniel J. Goldstein
Part III Minimally Invasive Valvular Surgery

19 Minimally Invasive Mitral Valve Surgery ...................................... 255
Victor F. Chu, L. Wiley Nifong,
and W. Randolph Chitwood, Jr.
20 Port-Access Mitral Valve Surgery................................................ 273
Ashish S. Shah and Donald D. Glower
21 Minimally Invasive Aortic Valve Surgery ................................... 293
Jerome Sepic and Lawrence H. Cohn
Part IV Minimally Invasive Congenital, Pericardial,

and Arrhythmia Surgery
22 Strategies for Reducing Trauma in Congenital
Heart Surgery ............................................................................ 311
Redmond P. Burke and Robert L. Hannan
Contents ix
23 Thoracoscopic Pericardial Surgery .............................................. 327

P. Michael McFadden
24 Less Invasive Surgical Treatment of Atrial Fibrillation .............. 343
Mathew R. Williams and Michael Argenziano
Part V Miscellaneous Aspects of Minimally Invasive

Cardiac Surgery
25 The Economic Impact of Minimally Invasive
Cardiac Surgery ........................................................................ 361
Aftab R. Kherani, Elizabeth H. Burton, and Mehmet C. Oz
26 Minimally Invasive Cardiac Surgery: Quality-of-Life Issues ...... 369
Giulio Pompilio, Francesco Alamanni, and Paolo Biglioli
27 Experimental Percutaneous Mitral Valve Repair ......................... 383
Juan P. Umaña and Peter Fitzgerald
28 Alternative Anastomotic Techniques ........................................... 393
David A. D’Alessandro and Mehmet C. Oz
29 Making Cardiopulmonary Bypass Less Invasive ......................... 411
James R. Beck, Linda B. Mongero, and A. Kenneth Litzie
Part VI Robotic Surgery

30 Robotics and Telemanipulation: The Zeus™ System ................... 429
Hersh S. Maniar, Sunil M. Prasad,
and Ralph J. Damiano, Jr.
31 Robotics and Telemanipulation: The da Vinci™ System ............. 445
Volkmar Falk, Thomas Walther, and Friedrich W. Mohr
32 Totally Endoscopic Atrial Septal Defect Repair
with Robotic Assistance ........................................................... 461
Michael Argenziano, Mehmet C. Oz,
and Craig R. Smith, Jr.
Index ........................................................................................................................... 471

CONTRIBUTORS xi
KEVIN D. ACCOLA, MD, Florida Heart Institute, Florida Hospital, Orlando, FL

FRANCESCO ALAMANNI, MD, Department of Cardiovascular Surgery, University of Milan,
Centro Cardiologico Monzino IRCCS, Milan, Italy
MICHAEL ARGENZIANO, MD, Assistant Professor of Surgery, Director, Cardiac Robotics and
Arrhythmia Surgery, Division of Cardiothoracic Surgery, Columbia Presbyterian Medical
Center, New York, NY
ROBERT B. BEAUFORD, MD, Surgical Resident, Department of General Surgery, Brookdale
University Hospital Medical Center, Brooklyn, NY
JAMES R. BECK, BS, CCP, Co-Director, Perfusion Services, Columbia Presbyterian Medical
Center, New York, NY
FEDERICO J. BENETTI, MD, PhD, Fundacion Benetti, Buenos Aires, Argentina
PAOLO BIGLIOLI, MD, Department of Cardiovascular Surgery, University of Milan, Centro
Cardiologico Monzino IRCCS, Milan, Italy
ANTONIO BIVONA, MD, Department of Cardiology and Cardiac Surgery, “G. D’Annunzio”
University, S. Camillo de’ Lellis Hospital, Chieti, Italy
MICHAEL A. BORGER, MD, P hD, Division of Cardiovascular Surgery, Toronto General
Hospital, Toronto, Ontario, Canada
ENIO BUFFOLO, MD, PhD, Escola Paulista Medicina, São Paulo, Brazil
REDMOND P. BURKE, MD, Chief, Division of Cardiovascular Surgery, Miami Children’s
Hospital, Miami, FL
ELIZABETH H. BURTON, BA, Research Assistant, Division of Cardiothoracic Surgery,
Columbia University, College of Physicians and Surgeons, New York, NY
MIKE BUTKUS, P.A.-C., Florida Heart Institute, Florida Hospital, Orlando, FL
ANTONIO M. CALAFIORE, MD, Professor and Chairman, Department of Cardiology and
Cardiac Surgery, “G. D’Annunzio” University, S. Camillo de’ Lellis Hospital,
Chieti, Italy
W. RANDOLPH CHITWOOD, JR., MD, Professor and Chairman, Department of Surgery,
The Brody School of Medicine, East Carolina University, Greenville, NC
VICTOR F. CHU, MD, Fellow in Minimally Invasive and Robotic Surgery, Department of
Surgery, The Brody School of Medicine, East Carolina University, Greenville, NC
LAWRENCE H. COHN, MD, Virginia and James Hubbard Professor of Surgery, Harvard
Medical School, and Chief, Division of Cardiac Surgery, Brigham and Women’s
Hospital, Boston, MA
WILLIAM E. COHN, MD, Department of Cardiothoracic Surgery, Beth Israel Deaconess
Medical Center, Boston, MA
MARC W. CONNOLLY, MD, Department of Cardiothoracic Surgery, St. Michael’s Medical
Center, Newark, NJ
ROBERT J. DABAL, MD, Cardiothoracic Resident, Division of Cardiothoracic Surgery,
University of Washington, Seattle, WA
DAVID A. D’ALESSANDRO, MD, Division of Cardiothoracic Surgery, Columbia Presbyterian
Medical Center, New York, NY
STEFANO D’ALESSANDRO, MD, Department of Cardiology and Cardiac Surgery, “G.
D’Annunzio” University, S. Camillo de’ Lellis Hospital, Chieti, Italy
RALPH J. DAMIANO, JR., MD, Chief, Division of Cardiothoracic Surgery, Washington
University School of Medicine, St. Louis, MO
xi
xii Contributors
BRENDA DICKEY, RN, Florida Heart Institute, Florida Hospital, Orlando, FL

M. ANNO DIEGELER, MD, PhD, Department of Cardiothoracic Surgery, University of
Leipzig, Bad Neustadt, Germany
MICHELE DI MAURO, MD, Department of Cardiology and Cardiac Surgery, “G. D’Annunzio”
University, S. Camillo de’ Lellis Hospital, Chieti, Italy
MERCEDES K. C. DULLUM, MD, Washington Regional Cardiac Surgery P.C., Washington
Hospital Center, Washington, DC
L EÓN E IJSMAN , MD , P hD , Department of Thoracic Surgery, Onze Lieve Vrouwe
Gasthius, Amsterdam, The Netherlands
VOLKMAR FALK, MD, PhD, Department of Cardiac Surgery, Heartcenter, University of
Leipzig, Leipzig, Germany
PETER FITZGERALD, MD, Associate Professor of Medicine, Department of Cardiothoracic
Surgery, Stanford University School of Medicine, Palo Alto, CA
JAMES D. FONGER, MD, Division of Cardiovascular and Thoracic Surgery, Lenox Hill
Hospital, New York, NY
PATRICIA GARLAND, RN, Heart Hospital of New Jersey, Newark Beth Israel Medical Center,
Newark, NJ
LUÍS ROBERTO GEROLA, MD, PhD, Escola Paulista Medicina, São Paulo, Brazil
MICHAEL GIBSON, MD, Cardiothoracic Surgery Resident, Department of Cardiothoracic
Surgery, Newark Beth Israel Medical Center, Newark, NJ
DONALD D. GLOWER, MD, Duke University Medical Center, Durham, NC
DANIEL J. GOLDSTEIN, MD, Surgical Director of Cardiac Transplantation and Mechanical
Assistance, Department of Cardiothoracic Surgery, Heart Hospital of New Jersey,
Newark Beth Israel Medical Center, Newark, NJ
MAXIMO GUIDA, MD, Department of Cardiovascular Surgery, Hospital Mendez de
Valencia, Venezuela
CRAIG R. HAMPTON, MD, Cardiothoracic Research Fellow, Division of Cardiothoracic
Surgery, University of Washington, Seattle, WA
ROBERT L. HANNAN, MD, Attending Physician, Division of Cardiovascular Surgery,
Miami Children’s Hospital, Miami, FL
AXEL HAVERICH, MD, Professor of Surgery, Division of Cardiothoracic and Vascular
Surgery, Hannover Medical School, Hannover, Germany
DANIEL F. HEITJAN, PhD, Professor of Biostatistics, Joseph P. Mailman School of Public
Health, Columbia University, New York, NY
AFTAB R. KHERANI, MD, Resident in General Surgery, Duke University Medical Center,
Durham, NC
UWE KLIMA, MD. Associate Professor of Surgery, Division of Cardiothoracic and Vascular
Surgery, Hannover Medical School, Hannover, Germany
RONALD M. LAZAR, PhD, Associate Professor of Clinical Neuropsychology, Columbia
University College of Physicians & Surgeons, and Director, Cerebral Localization
Laboratory, Neurovascular Service, Neurological Institute of New York, Columbia-
Presbyterian Medical Center, New York, NY
A. KENNETH LITZIE, Director, Clinical Development, Cardiovention, Santa Clara, CA
HERSH S. MANIAR, MD, Division of Cardiothoracic Surgery, Washington University
School of Medicne, St. Louis, MO
Contributors xiii
P. MICHAEL MCFADDEN, MD, Surgical Director Lung Transplantation, Ochsner Clinic

Foundation, Clinical Professor of Surgery, Tulane University School of Medicine,
New Orleans, LA
FRIEDRICH W. MOHR, MD, PhD, Department of Cardiac Surgery, Heartcenter, University
of Leipzig, Leipzig, Germany
LINDA B. MONGERO, BS, CCP, Co-Director, Perfusion Services, Columbia Presbyterian
Medical Center, New York, NY
L. WILEY NIFONG, MD, Assistant Professor of Surgery, Department of Surgery, and
Director, Minimally Invasive and Robotic Surgery, The Brody School of Medicine,
East Carolina University, Greenville, NC
MEHMET C. OZ, MD, Associate Professor of Surgery, Director of the Cardiovascular
Institute, Columbia Presbyterian Medical Center, New York, NY
ALESSANDRO PARDINI, MD, Department of Cardiology and Cardiac Surgery,
“G. D’Annunzio” University, S. Camillo de’ Lellis Hospital, Chieti, Italy
NILESH U. PATEL, MD, Division of Cardiovascular and Thoracic Surgery, Lenox Hill
Hospital, New York, NY
ALBERT J. PFISTER, MD, Washington Regional Cardiac Surgery P.C., Washington Hospital
Center, Washington, DC
GIULIO POMPILIO, MD, PhD, Department of Cardiovascular Surgery, University of Milan,
Centro Cardiologico Monzino IRCCS, Milan, Italy
SUNIL M. PRASAD, MD, Division of Cardiothoracic Surgery, Washington University
School of Medicne, St. Louis, MO
JOHN D. PUSKAS, MD, MSc, Carlyle Fraser Heart Center, Division of Cardiothoracic Surgery,
Department of Surgery, Emory University School of Medicine, Atlanta, GA
MARC RUEL, MD, MPH, Department of Cardiothoracic Surgery, Beth Israel Deaconess
Medical Center, Boston, MA
FREDERIC SARDARI, MD, Department of Cardiothoracic Surgery, Saint Barnabas Hospital,
Heart Hospital of New Jersey, Newark, NJ
CRAIG R. SAUNDERS, MD, Chairman, Department of Cardiothoracic Surgery, Heart
Hospital of New Jersey, Newark, NJ
JEROME SEPIC, MD, Cardiac Surgery Research Fellow, Harvard Medical School, Brigham
and Women’s Hospital, Boston, MA
ASHISH S. SHAH, MD, Duke University Medical Center, Durham, NC
CRAIG R. SMITH, JR., MD, Calvin Barber Professor of Surgery, Chief, Division of
Cardiothoracic Surgery, Columbia Presbyterian Medical Center, New York, NY
RON G. H. SPEEKENBRINK, MD, PhD, Department of Thoracic Surgery, Onze Lieve
Vrouwe Gasthius, Amsterdam, The Netherlands
VALAVANUR A. SUBRAMANIAN, MD, Chairman, Department of Surgery, Lenox Hill Hospital,
New York, NY
VINOD H. THOURANI, MD, Carlyle Fraser Heart Center, Division of Cardiothoracic Sur-
gery, Department of Surgery, Emory University School of Medicine, Atlanta, GA
JUAN P. UMAÑA, MD, Chairman, Division of Cardiovascular Surgery, Fundacion
Cardioinfantil, Instituto de Cardiologia, Bogota, Columbia
xiv Contributors
WIN VAN OEVEREN, PhD, The Center for Blood Interaction Research, Department of
Cardiothoracic Surgery, University Hospital Groningen, Groningen, The Netherlands
EDWARD D. VERRIER, MD, William K. Edmark Professor of Cardiovascular Surgery, Vice
Chairman, Department of Surgery, Chief, Division of Cardiothoracic Surgery, University
of Washington, Seattle, WA
THOMAS WALTHER, MD, PhD, Department of Cardiac Surgery, Heartcenter, University of
Leipzig, Leipzig, Germany
CHARLES R. H. WILDEVUUR, MD, Department of Thoracic Surgery, Onze Lieve Vrouwe
Gasthius, Amsterdam, The Netherlands
MATHEW R. WILLIAMS, MD, Surgical Arrhythmia Program, Division of Cardiothoracic
Surgery, Columbia Presbyterian Medical Center, New York, NY
Chapter 1 / Pathophysiology of Cardiopulmonary Bypass 1
I PHYSIOLOGY OF CORONARY BYPASS

GRAFTING WITH AND WITHOUT
CARDIOPULMONARY BYPASS
1 Pathophysiology
of Cardiopulmonary Bypass
Ron G. H. Speekenbrink, MD, PHD,

Wim van Oeveren, PHD,
Charles R. H. Wildevuur, MD, PHD,
and León Eijsman, MD, PHD
CONTENTS
INTRODUCTION
MATERIAL-DEPENDENT ACTIVATION
METHODS TO ATTENUATE MATERIAL-DEPENDENT ACTIVATION
MATERIAL-INDEPENDENT ACTIVATION
FUTURE DEVELOPMENTS
REFERENCES
INTRODUCTION
From the earliest clinical experiences with cardiopulmonary bypass (CPB) for cardiac
operations, it was apparent that significant morbidity and even mortality were associated
with the CPB procedure itself (1). Often, only the contact of blood to the foreign material
of the extracorporeal circuit was held responsible. However, cardiopulmonary bypass
implies more than just connecting the circulation of the patient to an extracorporeal
circuit, resulting in the material-dependent activation of blood. With cardiopulmonary
bypass, a number of other nonphysiological events are introduced, including hemodilu-
tion, hypothermia, nonpulsatile blood flow, retransfusion of shed blood, and exclusion
of the metabolic function of the lung resulting in material-independent activation.
Together, these events cause the massive and systemic activation of the patient’s defense
mechanisms, with repercussions on nearly every end-organ system. Signs of this “whole-
body inflammatory reaction” can be observed in every postoperative patient. In a number
of patients, especially neonates, the elderly, and those undergoing large procedures or
From: Contemporary Cardiology: Minimally Invasive Cardiac Surgery, Second Edition
Edited by: D. J. Goldstein and M. C. Oz © Humana Press Inc., Totowa, NJ
3
4 Part I / Physiology of Bypass Grafting
those with severe comorbidities, this phenomenon can escalate into the so-called
postperfusion syndrome, which is characterized by elevated cardiac output with
decreased vascular resistance, capillary leak, and renal function impairment, a constel-
lation of factors that is associated with increased mortality (2).
In this chapter, we discuss the various biochemical mechanisms that underlie the
deleterious effects of cardiopulmonary bypass, methods to mitigate these effects, and
guidelines for future developments.
MATERIAL-DEPENDENT ACTIVATION
In the past, activation of the contact system by the nonbiocompatible surface of the
extracorporeal circuit was considered to be the initiating factor in blood activation in CPB
and to be responsible for most of the detrimental effects of CPB. Because the contact
system is linked to the other humoral defense systems, its activation would result in
subsequent activation of the kallikrein, fibrinolytic, and coagulation systems. The active
products of these systems would cause, directly or indirectly via activation of leukocytes,
platelets, and endothelium, a part of the detrimental effects of CPB (3). This concept is
supported by studies in simulated bypass models showing activation of the contact sys-
tem by the extracorporeal circuit (4). However, clinical observations of patients with a
deficiency in contact system proteins have undermined this concept. Indeed, patients
with a severe deficiency of the primary factor of contact activation, factor XII, were
shown to have similar patterns of thrombin generation as healthy patients (5,6). More-
over, data from recent clinical investigations have shown no change in the marker for
contact-activation kallikrein–C1 esterase inhibitor complex between prebypass and
bypass levels, and no increase of factor XIIa levels during cardiopulmonary bypass (7,8).
Furthermore, the levels of a second marker for contact activation factor XIIa–C1 esterase
inhibitor complex during cardiopulmonary bypass remained below the detection limit in
the majority of patients (7). Coagulation studies have shown that factor X activation and
thrombin generation precede factor IX activation during cardiopulmonary bypass (Fig. 1),
which indicates activation through the extrinsic (tissue factor) pathway and not through
the intrinsic (contact-phase) pathway (8,9). From these data it appears justified to con-
clude that the role of the contact system in cardiopulmonary bypass needs to be redefined.
It is undisputed that the contact of blood with foreign materials results in the activation
of the complement system. Complement is activated through one of two pathways, the
“classical” or the “alternative” pathway (Fig. 2). The latter is predominantly involved in
complement activation by biomaterials because it can be activated in the absence of
specific antibodies (10). Both pathways form complexes named “C3 convertases,” which
cleave the third component of complement, C3, generating the anaphylatoxin C3a and a
major cleavage fragment C3b. Accumulation of C3b molecules onto the surface in the
vicinity of C3 convertases changes the specificity of the C3-cleaving enzyme into a C5
convertase, resulting in the cleavage of C5, generation of the leukocyte-activating C5a,
and recruitment of the terminal complement complex (TCC) C5–C9 (11). During CPB,
C3a appears in the circulation after 10–20 min, followed by C5a and TCC (12).
Although the composition of the artificial surfaces plays a predominant role in comple-
ment activation during cardiopulmonary bypass, nonbiomaterial-dependent triggers can
also activate complement. Contact of air with blood as it occurs in bubble oxygenators
and in the cardiotomy suction line activates complement (13). This advantage of the
membrane oxygenator, having no direct blood–air contact, may be nullified by its larger
Fig. 1. Relation between the intrinsic (contact) pathway and the extrinsic pathway to the common
pathway of coagulation. Josso indicates the Josso loop, which links the extrinsic to the intrinsic
pathway. TF = tissue factor. Roman numerals represent coagulation factors.
surface (14). Dextran 70 and, to lesser extent, polygeline induce complement activation
(15). Classical pathway activation is also initiated on interaction of C1 with antigen–
antibody complexes and, in some instances, with other activators, including some bac-
terial and viral surfaces and bacterial endo- or exotoxins (16,17). Furthermore, activation
of the classical pathway occurs after administration of protamine through formation of
heparin–protamine complexes (18,19).
The effects of complement activation are mediated by the products C3a, C5a, and
TCC. TCC, also called the membrane attack complex, deposits on erythrocytes and
leukocytes to augment cell lysis and cell activation (20). We recently found that TCC
concentrations formed during pediatric CPB correlated with postoperative fever and gain
in body weight (21). C3a and C5a are also called the anaphylatoxins. The anaphylatoxins
have chemotactic activity for neutrophils and monocytes (22) and induce cytokine re-
lease. Binding of C5a and C5a desArg to specific receptors on neutrophils induces:
1. Aggregation of the cells and their adherence to endothelial cells (23,24)
2. Release of reactive oxygen species, which may damage the endothelium to which acti-
vated neutrophils have bound (25,26)
3. Release of lysosomal enzymes (27)
4. Neosynthesis and release of leukotrienes (28,29)
Fig. 2..Schematic of the complement system. Immune complexes or activating surfaces activate
the first step of the classic and alternative pathway respectively. This results in the formation of
pathway-specific C3 and C5 convertases and finally formation of the terminal complement com-
plex, which damages the cell membranes.
Special attention has been paid to the lysosomal enzyme elastase, which affects the
endothelial junctions (30) and enhances vascular permeability for blood proteins (31).
The appearance of proteins, including elastase, in the alveolar space may be an important
cause for surfactant dysfunction and decreased lung compliance, resulting in pulmonary
dysfunction and possibly in adult respiratory distress syndrome (32).
Cytokines (Table 1) are released by monocytes, macrophages, endothelium, and other
cells on stimulation with the anaphylatoxins or the lipopolysaccharide endotoxin, a con-
stituent of the cell wall of Gram-negative bacteria. A number of acute and chronic adverse
consequences, such as hypotension and an increased body temperature after extracorpo-
real circulation, may be attributed to the cytokine interleukin-1 (IL-1) (33). IL-1 is an
essential component of the inflammatory reaction and the immune response by its ability
to stimulate neutrophil degranulation (34) and to activate T- and B-cells (35,36). The
cytokine tumor necrosis factor (TNF), a mediator of the host response in sepsis, is derived
from monocytes and macrophages (37). It has some common properties with IL-1, such
as induction of fever (38). TNF induces procoagulant activity (39) and IL-1 release
(38,40) from endothelial cells by interaction with specific receptors. TNF generation has
been measured in CPB patients and was assumed to be associated with endotoxin release
Chapter 1 / Pathophysiology of Cardiopulmonary Bypass
Table 1
Properties of the Most Well Known Cytokines Released During and After Cardiac Surgery
Main Source Main Target Main Effects
TNF-α Monocytes, macrophages TNF receptors on Hypotension, fever,

neutrophils, monocytes, cytokine + APP synthesis/
endothelial cells, hepatocytes, release ↑, plasma
chondrocytes, astrocytes, albumin ↓, neutrophil/
fibroblasts, osteoclasts endothelial interaction ↑
Il-1 Monocytes, macrophages Il-1 receptors Hypotension, fever
Il-6 Fibroblasts, monocytes, Il-6 receptors on Hypotension, fever,
7
macrophages, T-lymphocytes hepatocytes APP synthesis/release ↑,

tissue iNOS expression↑,
Il-1-RA ↑,TNF-RA ↑
Il-8 Monocytes, macrophages Chemokine receptors on Chemoattraction and
T-cells, fibroblasts, activation of neutrophils
monocytes, and neutrophils
Il-10 Monocytes, macrophages, Il-10 receptor on macrophages Soluble TNF receptor ↑,
T- and B-lymphocytes and monocytes in the liver TNF/Il-1/Il-6 ↓, Th1-
lymphocyte response ↓
TNF-α = tumor necrosis factor α , Il = interleukin, APP = acute-phase response, iNOS = inducible nitric oxide synthase, RA = receptor antagonist, Th1
= T-helper type 1. (Adapted from P. Bruins. Changes in the inflammatory response during and after cardiac surgery. Thesis, University of Amsterdam, 2000.)
7
(41,42). Interleukin-6 (IL-6) is produced by activated monocytes, endothelial cells, fibro-

blasts, and T- and B-cells. It is therefore a general marker of blood and tissue damage,
observed in a variety of surgical interventions and diseases. Il-6 levels after off-pump
coronary bypass grafting (OPCAB) are similar to levels after conventional on-pump
CABG (43,44). In CPB, IL-6 is of interest by its correlation with the ischemic time during
crossclamping (45): 4 h after CPB, peak levels of IL-6 are observed (46). Next to its
function as a marker for inflammation, IL-6 generated in the ischemic myocardium
appears to induce intercellular adhesion molecule-1 (ICAM-1) on the myocyte surface,
which could be held responsible for granulocyte adhesion in the myocardial tissue after
reperfusion of the ischemic heart (47). IL-8, similar to IL-6, is produced in several cell
types, including alveolar macrophages, fibroblasts, lymphocytes, and endothelial cells.
IL-8 plays an important role in leukocyte activation and contributes to myocyte
reperfusion injury. Its responses have been reported to be at maximum during or 2 h after
CPB (45,48), which is faster than other proinflammatory cytokines. IL-8 release is thought
to be dependent on C5a generation (49). IL-10 was recently reported as an anti-inflam-
matory cytokine that deactivates monocytes and macrophages and thus likely reduces
release of IL-6, IL-8, and TNF. Its rapid release within 1 h after CPB might offer impor-
tant negative feedback to further TNF production (45).
Adhesion molecules (Table 2) play a major role in the recruitment of neutrophils to the
site of inflammation. Multiple steps are involved in this process. Each step is character-
ized by the participation of a different family of adhesion molecules. The rolling phase
of neutrophils over the endothelial layer is mediated by the selectin family, the E-, L-, and
P- selectins and their ligands. The next step, the activation and adhesion of the neutrophils
to the endothelium, is regulated by the integrin family and their ligands. The final step
of transendothelial migration is mediated by these two families of adhesion molecules,
the selectins and integrins. It has been demonstrated that soluble isoforms of these adhe-
sion molecules can be found in circulation. In addition, these soluble isoforms appear to
be useful markers of disease activity, and they have physiological effects (50).
E-selectin is a specific marker for endothelial activation. The soluble form is biologi-
cally active in its capacity to bind to neutrophils. High concentrations of (recombinant)
E-selectin can inhibit neutrophil adhesion. L-selectin is produced by leukocytes after
stimulation with chemotactic peptides, IL-8, or endotoxin. It enhances the binding of
leukocytes to (inflammed) endothelium. P-selectin is produced by platelets and endothe-
lium and mediates the interaction with neutrophils and some lymphocyte subsets. It has
been suggested that P-selectin has anti-inflammatory effects, shown by reduced oxygen
radical production by neutrophils and inhibition of integrin-mediated adhesion of neu-
trophils (50). However, blocking of P-selectin during experimental CPB with specific
monoclonal antibodies resulted in reduced levels of IL-6, IL-8, and metabolites of nitric
oxide. The most extensively studied integrin in the field of cardiopulmonary bypass
surgery is CD11b/CD18, previously called Mac-1 or CR3 (51,52). The increased adhe-
siveness of neutrophils following incubation with C5a and C5a desArg is dependent on
the enhancement of membrane expression of the adhesion-promoting glycoprotein Mac1
on the cells (53). The integrins bind to the ECAM-1, ICAM-1, and VCAM-1, which are
constitutionally expressed ligands on endothelium (54).
The expressions of CD11b/CD18 and L-selectin have been used as markers for neu-
trophil activation during cardiopulmonary bypass (55). CD18 expression increased
immediately at the start of CPB, while L-selectin was shed from the neutrophil surface,
Table 2
Specific Ligands and Cell Membrane Adhesion Molecules
Group/Type Expressed by Ligand
Ig-like
ICAM-1 Endothelial cells, leukocytes CD11a/CD18
CD11b/CD18
VCAM Endothelial cells VLA-4
PECAM Platelets
Selectins
E-selectin (ELAM-1) Endothelial cells, leukocytes L-selectin, ESL-1
L-selectin Leukocytes P-selectin
P-selectin (GMP-140) Endothelial cells, platelets L-selectin
β2-Integrins
CD11a/CD18 (LFA-1) Leukocytes ICAM-1, -2, and -3
CD11b/CD18 (MAC-1) Leukocytes ICAM-1, ICAM-2
CD11c/CD18 Monocytes
β1-Integrin
CD49d/CD29 (VAL-4) Endothelial cells VCAM/fibronectin
Ig = immunoglobulin; ICAM = intercellular adhesion molecule; PECAM = platelet-endothelial cell
adhesion molecule; VCAM = vascular cell adhesion molecule; VLA = very late antigen; LFA =
leukocyte function associated antigen; ESL = endothelial sialyl Lewis antigen. (Reproduced with
permission from P. Bruins. Changes in the inflammatory response during and after cardiac surgery.
Thesis, University of Amsterdam, 2000.)
shown by a gradual loss during 60 min of CPB. Similarly, instant activation of platelets,
reflected by a decreased expression of the adhesive receptor Gp1b after initiation of
cardiopulmonary bypass has been reported (56). Markers for activation of the comple-
ment, coagulation, or fibrinolytic systems do not show such rapid increases at the onset
of cardiopulmonary bypass, but rather increase slowly during the procedure. Therefore,
the stimulus for this rapid activation of platelets and neutrophils is unclear. It is possible
that direct activation by the foreign surface is involved. Another explanation could be the
production of trace amounts of activators, not distinguishable systemically, on the sur-
face of the extracorporeal circuit (57). Based on the observations of fast cellular
responses to extracorporeal circulation the generally used activation scheme, contact
activation → activation of humoral defense systems → activation of cells, probably needs
to be revised. Most likely, the activated cell membranes and the cell constituents form
major triggers for activation of the hemostatic and inflammatory reactions to CPB.
The endothelium plays an important role in the regulation of vascular tone. Two
systems are involved in this regulation: the nitric oxide (NO)/endothelin and the
prostacyclin/thromboxane system. Both systems are affected by the inflammatory
response after CPB. NO was formerly known as endothelium-derived relaxing factor. It
is formed by two synthases, a constitutive form, which produces NO in picomolar quan-
tities, and an inducible form, which produces nanomolar amounts of NO (58). The activ-
ity of the inducible form is increased by endotoxin, TNF, and IL-1. Increased NO
concentrations are found during and after cardiopulmonary bypass, and might have a role
in the vasoplegic syndrome affecting some patients after cardiac procedures (58). The
cardiodepressive effect of IL-6 and TNF was shown to result from increased NO produc-
tion through inducible synthase activation (59). Moreover, NO in high concentrations has
been implicated in vascular, lung, and bowel injury (60). The counterpart of NO is
endothelin, a small peptide with potent vasoconstrictive capacities that is produced by
endothelium, macrophages, and the hypothalamus (61,62). Institution of CPB results in
a rapid increase of endothelin, which is likely the result of a neurohumoral response to
decreased blood pressure. During CPB, a slow increase of endothelin concentrations can
be observed that correlates with endotoxin concentrations (63). Inappropriate endothelin
concentrations can cause pulmonary hypertension and myocardial ischemia, are associ-
ated with increased inotrope requirements and ICU stay, and might have a role in
perioperative gut ischemia (63–66).
Prostacyclin and thromboxane are byproducts of the cyclooxygenase pathway located
in platelets and endothelium. Prostacyclin (PGI2) is produced by endothelial cyclooxygenase.
It is a potent vasodilator and inhibits platelet adhesion to endothelium in synergy with
NO. Thromboxane B2 (TXB2) is produced by activated platelets and results in irrevers-
ible platelet aggregation and vasoconstriction. The balance between prostacyclin and
TXB2 can be modified with acetylsalicylic acid, which irreversibly inhibits platelet
cyclooxygenase. This is a common therapy in patients with coronary heart disease. During
cardiopulmonary bypass prostacyclin levels are increased due to the presence of heparin.
Following platelet activation, TXB2 levels are increased during and after cardiopulmo-
nary bypass (67). This results in a disturbed balance between prostacyclin and TXB2 in
the immediate postoperative period, which might have an influence on graft patency. It
has been advocated to continue acetylsalicylic acid therapy in the perioperative period,
but fear for increased postoperative bleeding discouraged many centers to adopt this
policy. Because treatment with aprotinin effectively inhibits the increased bleeding in
acetylsalicylic acid-treated patients, but maintains the effect of acetylsalicylic acid on
platelets, this policy is no longer warranted (68).
METHODS TO ATTENUATE MATERIAL-DEPENDENT ACTIVATION

Surface Coating of Extracorporeal Circuits
A large improvement in the biocompatibility of the extracorporeal circuit was expected
with the development of extracorporeal circuits that were lined with a more biocompatible
coating. The first and most extensively studied coating is heparin coating. Promising
recent developments include coatings of poly-2-methoxyethyl acrylate (69) synthetic
protein (70) and phosphorylcholine (71). The concept behind heparin coating is to mimic
the endothelial surface that contains heparan sulfate. Currently two types of heparin
coating are commercially available for extracorporeal circuits. In the first coating heparin
is ionically bound to the polymeric surface of the extracorporeal circuit (Duraflo II,
Baxter, Muskegon, MI). The second type of coating uses covalently bound heparin
(Carmeda, Medtronic, Minneapolis, MN; Bioline, Jostra, Hirrlingen, Germany). Both
coatings have been studied extensively during the past decade. The most striking effect
of heparin-coated circuits is the reduction of complement activation, which has been
estimated at 45% (72). Another study comparing Duraflo II with Carmeda coatings
demonstrated a 25% reduction with both coatings (73). The reductions are most promi-
nent in C5a and TCC levels, probably owing to their slower clearance. Secondary to the
reduced complement activation, the inflammatory responses of leukocytes, platelets, and

endothelium are attenuated, resulting in reduced lactoferin and myeloperoxidase levels
(74,75), IL-6, IL-8, and E-selectin (76,77), oxygen free-radical production (78), integrin
and selectin response of platelets (79), and platelet β-thromboglobulin release (80).
Although heparin coating is effective in reducing complement activation through the
alternative pathway, classical pathway activation after protamine infusion was also shown
to be reduced (81). This might indicate that a key component of the complement system
is inactivated or bound by the coating (82).
The levels of kallikrein–C1 esterase inhibitor complex, a marker for contact activa-
tion, were shown to be reduced during cardiopulmonary bypass with heparin-coated
circuits, but remained unchanged when uncoated circuits were used (7). Binding of FXII
to the coating could be responsible for this observation (83).
Contrary to initial expectations, thrombin generation and the activity of the fibrin-
olytic system are not reduced with heparin coating (84,85). Improved hemostasis,
reflected by a decrease in perioperative blood loss and transfusions with the use of coated
circuits has only been reported anecdotally. These results were obtained when heparin
coating was combined with a lower dose of systemic heparin (86,87). However, the use
of a decreased dose of heparin in conjunction with heparin-coated circuits is not consid-
ered to be safe and is indicated only in special circumstances (88).
Although substantial reductions in blood activation can be obtained with heparin-
coated circuits, it has proven difficult to translate these into improved clinical outcomes.
In one study, a decreased intrapulmonary shunt with improved respiratory index was
found after cardiopulmonary bypass with heparin-coated circuits. However, intubation
time and ICU stay were not affected (89). A composite score, consisting of intubation
time, the central–peripheral temperature difference, and postoperative fluid balance, was
significantly reduced with heparin-coated circuits (90). Similarly, a composite score of
adverse events after coronary surgery was improved by using heparin-coated circuits
(91,92). In a multicenter European trial, a significantly better postoperative recovery was
found in females and in patients with crossclamp times exceeding 60 min when heparin-
coated circuits were used (93).
Aprotinin
Aprotinin is a polypeptide processed from bovine lung that acts as an inhibitor of serine
proteases such as plasmin, trypsin, and kallikrein (94). It was first introduced into cardiac
surgery in an attempt to reduce complement activation during cardiopulmonary bypass.
Although this attempt failed, “bone-dry” operative fields with significantly reduced
postoperative blood loss and perioperative homologous blood product use were noted
(95). Since these initial reports, use of aprotinin has become widespread as an adjunct in
blood-saving programs in cardiac surgery. Two dosage regimens with aprotinin have
evolved, the low Groningen and the high Hammersmith dose, which have similar effec-
tiveness in reducing blood loss and transfusion requirements (96–98). More recently,
encouraging results were obtained with aprotinin administered topically to the pericar-
dial sac (99).
Owing to its nonspecific mode of action, aprotinin has a number of effects. Most
notably, aprotinin, in either dose, preserves platelet function by preventing the acute loss
of the Gp1b adhesive receptor expression on platelets that occurs at the onset of cardiop-
ulmonary bypass (56). The mechanism underlying this protective effect has not been
elucidated. Possibly, aprotinin inhibits the activity or production of trace amounts of

agonists on the surface of the extracorporeal circuit, or interferes with the interaction
between the platelet and the foreign surface. Inhibition of hyperfibrinolysis is a second
mechanism by which aprotinin reduces bleeding in cardiopulmonary bypass, and prob-
ably explains the efficacy of topically administered aprotinin and other antifibrinolytic
agents (100–102). Aprotinin has also been shown to protect platelets against the inhibi-
tory effect of heparin, a phenomenon present in 30% of patients that is not clarified (103).
A number of anti-inflammatory effects of aprotinin have been demonstrated. In simu-
lated bypass, aprotinin inhibited contact, complement, and neutrophil activation (104).
In vitro aprotinin inhibits ICAM-1 and VCAM expression on endothelial cells (105) and
inhibits shedding of L-selectin of leukocytes (106) and neutrophil transmigration (107).
In a murine bronchial epithelial cell line, aprotinin inhibited expression of the cytokine-
inducible form of nitric oxide synthase (108), which could account for the decreased
airway NO levels found during CPB in aprotinin-treated patients (109). During CPB,
aprotinin inhibits CD11b expression (110–112), and after CPB, reduced levels of TNF
(110), IL-8 (113), and IL-6 and increased IL-1 receptor antagonist (114) were found. In
a comparative study, the effects of aprotinin on CD11b expression and TNF levels were
comparable to those of methylprednisolone (110). Bronchoalveolar lavage fluid obtained
after CPB contained less neutrophils and less IL-8 in aprotinin-treated patients (115), and
lung tissue contained reduced levels of malondialdehyde, a marker of oxygen free radical
damage, higher glutathione peroxidase levels, and reduced leukocyte sequestration (116).
Although use of aprotinin in cardiac surgery is considered efficacious and safe (117),
many clinicians caution against liberal use of aprotinin, arguing that the inhibition of
fibrinolysis might increase thromboembolic complications, particularly the occlusion of
thrombogenic de-endothelialized vein grafts (118–120). Others have implicated inhibi-
tion of activated protein C, a pivotal factor in the regulation of coagulation, by aprotinin
as a cause for thromboembolic complications (121). Although thromboembolic compli-
cations after aprotinin treatment have been reported anecdotally, it appears to be safer to
use the lowest effective dose of aprotinin during CPB. Currently, the low-dosage regimen
and topical aprotinin meet this criterion. As for the inhibition of activated protein C, we
demonstrated that aprotinin treatment, either low or high dose, does not change the
pattern of activation of the protein C system (122).
Corticosteroids
Administration of high-dose corticosteroids has been shown to attenuate the inflam-
matory response induced by cardiopulmonary bypass and to improve the postoperative
course (41). Although inhibition of the alternative pathway by methylprednisolone has
been demonstrated (123), C3a and elastase levels during cardiopulmonary bypass were
not influenced by high-dose corticosteroids (124,125). The cytokine response during
cardiopulmonary bypass is markedly modulated by high-dose corticosteroids. Increase
of the inflammatory cytokines Il-6, Il-8, and TNF is prevented, whereas concentrations
of the anti-inflammatory Il-10 increase 10-fold (126–130). Reduced cytokine-mediated
activation of neutrophils results in reduced CD11b expression and leukotriene B4 release
(125,131). Moreover, leukocyte adhesion to endothelium is reduced because glucocor-
ticoids inhibit the expression of ICAM-1 to which CD11b adheres (132). The attenuated
inflammatory response with high-dose corticosteroids has been associated with enhanced
myocardial recovery, reduced pulmonary damage, and an overall better clinical recovery
after cardiopulmonary bypass (123,133,134).
MATERIAL-INDEPENDENT ACTIVATION
Cardiotomy Suction
Recently, it became clear that blood collected in the pericardium is highly activated
by tissue factor and t-PA and is rich in the highly procoagulant microparticles derived
from damaged platelets and erythrocytes (135–137). Tissue factor is not expressed by
pericardium, but enters the pericardium from the surgical wounds (138). Being a
mesothelial surface, pericardium is rich in t-PA (139). In vitro experiments have indicated
that the activation of pericardial blood is triggered by the extrinsic (tissue factor) coagu-
lation system and that the activation of fibrinolysis is secondary (135). Retransfusion of
the activated blood introduces fibrin(ogen) degradation products into the circulation,
which interfere with platelet receptors, fibrinogen binding to platelets, and clot formation
(140,141). Moreover, activators are retransfused, which can result in further systemic
activation and impaired hemostasis (142). Similarly, retransfusion of mediastinally shed
blood after operation was shown to result in a dose-dependent inflammatory response,
impaired hemostasis, and increased bleeding (143). Nevertheless, with a limited amount
of retransfusion, reductions in blood use could be achieved (144).
There are several ways to reduce the activation of pericardial blood. Use of a controlled
suction device, which incorporates a level sensor that is activated when blood accumu-
lates in the pericardium, minimizes superfluous suctioning and air entering the suction
line, and thus the formation of activating air–blood interfaces (145). Reduction of the
contact time between blood and pericardium might have additional effects.
Topical administration of aprotinin into the surgical wound and the pericardium can
inhibit the hyperfibrinolysis that occurs in the pericardial blood and improve hemostasis
(99). Because heparin levels in pericardial blood were shown to be lower than systemic
levels, topical administration of heparin might also reduce the activation of pericardial
blood, by reducing thrombin activity (135).
Ischemia and Reperfusion

Cooling and cardioplegia have been shown to attenuate the negative effects of
ischemia on the heart after crossclamping of the aorta by reducing the metabolic demand
of the myocardium. Nevertheless, ischemia will occur or is already present owing to the
disease process that is being treated. This ischemia will reduce high-energy phosphate
content of cells and may cause a degree of reversible and irreversible myocardial damage.
Proposed mediators of reperfusion injury following ischemia involve the generation of
oxygen free radicals, which are produced via the xanthine oxidase reaction (146) and by
activated neutrophils (147). Exposure of the ischemic endothelium to oxygen free radi-
cals induces a rapid upregulation of P-selectin and integrin expression (148). At
reperfusion, this will result in the accumulation of more activated neutrophils, which shed
their cytotoxic enzymes, cytokines, and oxygen free radicals on the endothelium, leading
to tissue injury. Damage to receptors involved in the activation of constitutive NO syn-
thase will reduce nitric oxide production and as a consequence coronary spasm and the
no-reflow phenomenon can occur (149). Possible ways to reduce reperfusion damage
include oxygen-radical scavengers (150), maintenance of physiological oxygen concen-

trations during CPB (151), inhibition of xanthine oxidase by allopurinol (152), or preven-
tion of ischemia by using continuous blood cardioplegia techniques (153).
Respiratory dysfunction is a well-recognized side effect of cardiac operations. One-
quarter of uncomplicated CPB patients still have a significant respiratory impairment 1
wk after operation (154). A proportion of these impairments can be attributed to deterio-
rated breathing mechanics as a result of surgical factors (e.g., wound pain, drains, effu-
sions). The effects of cardiopulmonary bypass primarily involve the gas flow and gas
exchange owing to parenchymal damage (155).
The bronchial circulation might be expected to prevent ischemia of the lungs during
CPB and crossclamping. However, in a piglet study a 10-fold decrease in bronchial artery
blood flow during CPB was demonstrated (156). Thus, low or absent flow in the pulmo-
nary circulation during aortic crossclamping is likely to occur and will result in similar
reperfusion phenomena as in the myocardium. This is supported by the salutary role of
inhaled NO, intravenous NO donors, and endothelin-1 antagonists in the treatment of
postperfusion pulmonary dysfunction (157–159). Apart from reperfusion injury, other
factors unique to the lung render it more susceptible to damage by cardiopulmonary
bypass. First, the lung is the filter of the venous circulation, meaning that all active or
activating substances and cells generated during CPB transit the pulmonary circulation.
Second, the lung capillaries are smaller in diameter than the average systemic capillaries,
resulting in preferential trapping of aggregates in the lung. Third, a considerable pool of
neutrophils is present in the lungs (160). The importance of neutrophils in inducing lung
damage is illustrated by the correlation of postoperative shunt fraction and respiratory
index with elastase levels (161). Animal experiments demonstrated that leukocyte deple-
tion by filtration reduced heart and lung reperfusion injury (162). Clinically, the use of
leukocyte filters transiently improved the pulmonary shunt fraction and the mean arterial
pressure (163). In another study, post-bypass filtration of 2 L heart–lung machine blood
significantly improved the postoperative lung function (164). Maintenance of some
pulmonary flow during aortic crossclamping, as is achieved with the use of two-stage
venous canulas, was shown to prevent the increase in extravascular lung water content
and preserve endothelin clearance (165,166). This concept is expanded by using the
Drew perfusion technique, in which no oxygenator is used but the patient’s lungs provide
oxygenation through separate perfusion of the systemic and pulmonary circulations (167).
Use of this technique resulted in reduced pulmonary leukocyte sequestration and comple-
ment activation (168).
Hemodilution
At the initiation of CPB, mixing of the patient’s blood with the relatively large
asanguineous pump prime results in a sudden hemodilution. Although moderate hemodi-
lution is considered beneficial in the setting of CPB, unwanted side effects do occur, and
dilution to hematocrits below 0.23 during CPB are associated with increased mortality
(169). Next to the apparent effect on oxygen delivery, hemodilution can reduce the
plasma colloid oncotic pressure to borderline values, resulting in a transcapillary oncotic
imbalance. Consequently, important fluid shifts toward the interstitial tissue take place,
contributing to edema formation, hypovolemia, and impaired oxygen delivery to vital
organs such as the digestive tract (170,171). Reduction of the priming volume of the
extracorporeal circuit was demonstrated to attenuate the hyperdynamic response to CPB,
as measured by fluid load, arterial pressure, cardiac index, vascular resistance, and oxy-
gen delivery (172). Furthermore, endotoxin levels, likely derived from ischemic intes-
tines, were reduced in these patients. Increasing the colloid oncotic pressure of the priming
solution by replacement of crystalloids with colloids similarly improved the postopera-
tive course and resulted in reduced hospital stay (173).
Reduction of priming volumes also results in important savings in the use of donor
blood by two mechanisms (172). First, the reduction of hemodilution will allow for
predonation of relatively large volumes of blood in the majority of patients while main-
taining a sufficiently high hematocrit during perfusion. Retransfusion of predonated
blood after perfusion has been shown to improve hemostasis (174). Second, the attenu-
ated hyperdynamic response will reduce the need for fluid administration and thus further
hemodilution during and after perfusion (173). Other methods to prevent or treat exces-
sive hemodilution during extracorporeal circulation, such as the use of blood cardiople-
gia or perioperative hemofiltration, were shown to further reduce the need for blood
transfusions in coronary surgery (175).
Hypothermia
Hypothermic perfusion has for a long time been a standard procedure during CPB. In
addition to its effect on cell metabolism, a reduction of the inflammatory reaction can be
anticipated. Indeed, IL-1, IL-6, and TNF production appeared higher at 37°C during in
vitro experiments and in clinical CPB (176). Similarly, elastase levels were higher after
normothermic perfusion (177). Also, the production of E-selectin was reduced, and
CD11b upregulation was delayed at low temperature, which could account for reduced
leukocyte adherence to the vasculature after reperfusion (178,179). Other researchers
have found a similar release of cytokines at normothermia (>36.5°C) compared to hypo-
thermia (180).
Although normothermic cardiopulmonary bypass appears to induce a more severe
inflammatory reaction on a biochemical level, this is not reflected in clinical parameters.
No difference in adverse events such as perioperative infarctions, use of an intra-aortic
balloon pump, length of ICU stay, or mortality was observed (181). However, vasopres-
sors are used more frequently in normothermic perfusion, probably as a result of reduced
endothelial endothelin-1 release, or an increased production of nitric oxide (181–183).
Other studies reported improved pulmonary function with shorter intubation times and
a reduced incidence of postoperative atrial fibrillation after normothermic CPB (184–186).
When considering normothermic CPB, neuroprotection is of special interest. Cogni-
tive impairment can be identified in up to 45% patients who undergo CPB and focal
deficits in 1–3% (187,188). The cerebral damage is caused mainly by emboli. Focal
defects are the result of large emboli that originate from surgical manipulation of the heart
and aorta. Microemboli are thought to be responsible for the more subtle neurological
defects detected in neuropsychological testing (189). With the use of a particle counter,
it was found that a bubble oxygenator produced more 15–80 µm microemboli than a
membrane oxygenator, but that 80% of all particles were produced in the blood circula-
tion owing to cardiotomy suction (190). The small capillary and arteriolar dilatations
(SCADs) that are caused by diffuse depositions of acellular fatty material found postmor-
tem in the brains of patients who have recently undergone CPB might be the result of these
microemboli (191). In a dog study, use of a cell saver instead of cardiotomy suction
reduced the number of SCADs by more than 50% (192).
Several studies have addressed the issue of perfusion temperature and cerebral protec-
tion, often with the use of an impressive battery of neuropsychological tests (193–200).
The results of these studies vary greatly. Some found a protective effect of hypothermia,
others of normothermia or of “tepid” perfusion. A meta-analysis by the Cochrane Col-
laboration could only demonstrate a trend toward fewer strokes with hypothermic per-
fusion, but this was offset by a trend toward increased incidence of non-stroke-related
mortality and morbidity (201). The current knowledge is far from conclusive and in many
centers an intermediate course with temperatures between 32 and 35ºC is followed.
Recently, specific markers for cerebral damage such as the S-100 protein and neuron-
specific enolase have become available (202,203). A correlation of CPB time with levels
of S-100 could be explained by intensified suction in these patients (204). Similarly,
intracardiac operations appeared to be associated with higher S-100 levels than CABG
operations and the use of an arterial filter with reduced S-100 levels in CABG operations
(205,206). In OPCAB, 10-fold reductions in S-100 levels compared to on-pump CABG
were observed (207). These assays will be of aid in the development of improved
neuroprotective strategies in cardiopulmonary bypass.
Heparin
Because blood will clot in an extracorporeal circuit, strict anticoagulation is manda-
tory. Traditionally, heparin is used for this purpose because of its easy dosage, control of
effectivity and the availability of an antidote. However, heparin does have side effects.
Despite adequate heparin levels, thrombin generation can be detected during cardiopul-
monary bypass. Moreover, heparin administration results in a rapid release of t-PA from
its body sources, which may induce fibrinolysis (208,209). Recently, in vitro inhibition
of platelet function by heparin was reported (210). This inhibition, which was present in
more than 30% of the study population, was associated with an increased postoperative
blood loss. Heparin also has proactivating properties on granulocytes and platelets
(211,212). With the neutralization of heparin with protamine, complexes are formed that
activate the complement system through the classical pathway. This classical pathway
activation correlates with postoperative pulmonary shunt fraction (213). The recombi-
nant form of platelet factor 4, a polypeptide present in platelets that binds and inhibits
heparin, could become an attractive alternative to protamine (214,215).
The disadvantages related to heparin and protamine have prompted a search for better
anticoagulants. Hirudin, a selective thrombin inhibitor derived from leeches, is frequently
mentioned as an attractive alternative. Animal studies comparing recombinant hirudin
with heparin demonstrated good clinical results without increased bleeding tendency
(216). Unlike heparin, however, hirudin only inhibits thrombin; it does not prevent its
formation. This could result in the escape of small amounts of thrombin, as was demon-
strated in vitro (217). The absence of an inhibitory effect of hirudin on components higher
in the coagulation cascade will not prevent an ongoing activation at this level and might
result in depletion of these factors (218). Finally, an antidote to hirudin is not available.
Based on these considerations, replacement of heparin with hirudin is not advised. Per-
haps there is a role for hirudin as an adjunct to heparin.
FUTURE DEVELOPMENTS
From the previous sections it is clear that many factors are involved in the detrimental
effects of cardiopulmonary bypass. Therefore, substantial improvements in the proce-
dure of cardiopulmonary bypass can only be obtained when a multifactorial approach is

followed, directed at both material-dependent and -independent factors. Thus,
biocompatibility of material surfaces has to be improved, and material-independent
sources of blood activation should be controlled by adaptation of perfusion techniques
and, when necessary, pharmacological intervention.
Based on current insights and available technologies, we propose a novel system for
CPB aimed at minimal disturbance of the patient’s homeostasis. Primary in this system
is a newly designed low-prime, closed-volume, and hemocompatible extracorporeal
circuit. The basic principle of this circuit is to abandon the use of gravity drainage and
use a venoarterial blood pump instead (219). An alternative is the use of vacuum-assisted
drainage (220). Both techniques allow for the use of smaller cannulae and tubings
and placement of the extracorporeal circuit close to the patient. Together with a low-
prime oxygenator with an integrated arterial filter, a drastic reduction of the prime volume
will be achieved. Hemodilution can be further minimized by using the technique of
retrograde autologous priming, in which a substantial portion of the priming volume can
be extracted from the extracorporeal circuit by controlled exsanguination of the patient
into the circuit while simultaneously draining priming solution (221). Handling of air in
such a system is more complicated and will require an advanced air-trapping mechanism.
The cardiotomy reservoir will be connected to a controlled suction or a cell-saving
device. All components of the circuit will be coated with heparin and primed with
aprotinin.
The described system is expected to provide a more physiological perfusion. The
reduced prime volumes will avoid hemodilution, hypooncotic pressures, and fluid shifts,
ensuring improved preservation of the patient’s autoregulatory mechanisms and better
hemodynamic stability and organ perfusion. The use of controlled suction will minimize
the contact time between blood and nonendothelialized tissues, thus avoiding activation
of the coagulation and fibrinolytic systems. Addition of heparin, hirudin, or aprotinin
to the pericardial sac can be of further aid to achieve this goal. An alternative to controlled
suction might be the use of a cell-saving device, which separates red cells from the fluid in
the pericardial sac. In the proposed system, heparin coating is used. However, it should
be emphasized that this coating results in only a 25–45% reduction in complement acti-
vation and, to a lesser extent, inhibits contact activation. Although contact activation can
be inhibited by the addition of aprotinin to the pump prime, the problem of complement
activation will persist. Because the mechanism of heparin coating is probably the result
of absorption of an essential factor of the complement system, it does not seem to be
pragmatic to improve the efficacy of the heparin coating. Instead, research should
be focused at the development of biologically active coatings that actually prevent the
activation of the humoral and cellular components of blood.
The use of improved perfusion equipment and techniques should be accompanied by
improved methods for monitoring the quality of the perfusion. Traditionally, perfusion
is guided by the values of flow, pressures, diuresis, and a limited number of laboratory
results. These do not adequately monitor the primary goal of perfusion, i.e., maintenance
of adequate microcirculation, but mainly reflect the macrocirculation. Moreover, the
target values for flow, perfusion pressure, and their interactions with temperature man-
agement and hematocrits resulted more from assumptions than from (patho)physiological
studies.
Direct monitoring of the microcirculation is possible with the technique of orthogonal

polarization spectral imaging (222). With this technique the microcirculation on mucous
membranes can be visualized up to the level of individual traveling erythrocytes. Cur-
rently this technique is in an experimental phase. Difficulties in the interpretation and
processing of data preclude its routine use. Furthermore, probes for measurement of the
microcirculation in the intestine, which is affected earliest during CPB, are not yet avail-
able. Alternatively, the adequacy of the microcirculation can be monitored by measure-
ment of biochemical markers for ischemic organ damage. These markers should be
sensitive and specific, be measurable by simple assays that can be performed on a routine
basis with results available during or shortly after the perfusion. Obviously, if beating
heart revascularization proves to be as safe and effective as conventional bypass grafting
in the short and long term, off-pump approaches will circumvent many of the pathophysi-
ological derangements associated with extracorporeal circulation.
REFERENCES
1. Kirklin JW. Open-heart surgery at the Mayo Clinic: the 25th anniversary. Mayo Clin Proc 1980;55:339–341.
2. Westaby S. Organ dysfunction after cardiopulmonary bypass. A systemic inflammatory reaction ini-
tiated by the extracorporeal circuit. Intensive Care Med 1987;13:89–95.
3. Edmunds LH Jr. Blood-surface interactions during cardiopulmonary bypass. J Card Surg 1993;8:
404–410.
4. Wachtfogel YT, Harpel PC, Edmunds LH Jr, Colman RW. Formation of C1s-C1-inhibitor, kallikrein-
C1-inhibitor and plasmin-alpha 2-plasmin-inhibitor complexes during cardiopulmonary bypass. Blood
1989;73:468–471.
5. Burman JF, Chung HI, Lane DA, Philippou H, Adami A, Lincoln JC. Role of factor XII in thrombin
generation and fibrinolysis during cardiopulmonary bypass. Lancet 1994;344:1192–1193.
6. Moorman RM, Reynolds DS, Communale ME. Management of cardiopulmonary bypass in a patient
with congenital factor XII deficiency. J Cardiothorac Vasc Anesth 1993;7:452–454.
7. te Velthuis H, Baufreton C, Jansen PG, et al. Heparin coating of extracorporeal circuits inhibits contact
activation during cardiac operations. J Thorac Cardiovasc Surg 1997;114:117–122.
8. Boisclair MD, Lane DA, Philippou H, et al. Mechanisms of thrombin generation during cardiopulmo-
nary bypass. Blood 1993;82:3350–3357.
9. Philippou H, Adami A, Boisclair MD, Lane DA. An ELISA for factor X activation peptide: application
to the investigation of thrombogenesis in cardiopulmonary bypass. Br J Haematol 1995;90:432–437.
10. Kazatchkine MD, Nydegger UE. The human alternative pathway. Biology and immunopathology of
activation and regulation. Prog Allergy 1982;30:193–234.
11. Müller-Eberhard HJ. Complement: Chemistry and pathways. In Gallin JI, Goldstein IM, Snyderman
R. Inflammation: Basic Principles and Clinical Correlates. New York: Raven Press, 1988:21–54.
12. Chenoweth DE, Cooper SW, Hugli TE, Stewart RW, Blackstone EH, Kirklin JW. Complement acti-
vation during cardiopulmonary bypass. N Engl J Med 1981;304:497–503.
13. Parker DJ, Cantrell JW, Karp RB, Stroud RM, Digerness SB. Changes in serum complement and
immunoglobins following cardiopulmonary bypass. Surgery 1972;71:824–827.
14. Videm V, Fosse E, Mollnes TE, Garred P, Svennevig JL. Complement activation with bubble and
membrane oxygenators in aortocoronary bypass grafting. Ann Thorac Surg 1990;50:387–391.
15. Videm V, Mollnes TE. Human complement activation by polygeline and dextran 70. Scand J Immunol
1994;39:314–320.
16. Cooper NR. The classical complement pathway: activation and regulation of the first complement
component. Adv Immunol 1985;37:151–216.
17. Loos M, Wellek B, Thesen R, Opferkuch W. Antibody-independent interaction of the first component
of complement with gram-negative bacteria. Infect Immun 1978;22:5–9.
18. Fehr J, Rohr H. In vivo complement activation by polyanion-polycation complexes: Evidence that C5a
is generated intravascularly during heparin-protamine interaction. Clin Immunol 1983;29:7–14.
19. Kirklin JK, Chenoweth DE, Naftel DC, et al. Effects of protamine administration after cardiopulmo-
nary bypass on complement, blood elements, and the hemodynamic state. Ann Thorac Surg 1986;41:
193–199.
20. Salama A, Hugo F, Heinrich D, et al. Deposition of terminal C5b-9 complement complexes on eryth-
rocytes and leukocytes during cardiopulmonary bypass. N Engl J Med 1988;318:408–414.
21. Schreurs HH, Wijers MJ, Gu YJ, et al. Heparin coated bypass circuits: effects on inflammatory response
in paediatric cardiac surgery. Ann Thorac Surg 1998;66:166–171.
22. Hugli TE, Müller-Eberhard HJ. Anaphylatoxins C3a and C5a: Adv Immunol 1978;26:1–53.
23. Charo IF, Yuen C, Perez HD, Goldstein IM. Chemotactic peptides modulate adherence of human
polymorphonuclear leukocytes to monolayers to monolayers of cultured endothelial cells. J Immunol
1986;136:3412–3419.
24. Tonnesen MG, Smedly LA, Henson PM. Neutrophil-endothelial cell interactions. J Clin Invest
1984;74:1581–1592.
25. Bender JG, van Epps DE. Stimulus interactions in release of superoxide anion (O2–) from human
neutrophils. Inflammation 1985;9:67–86.
26. Bender JG, Mc Phail LC, van Epps DE. Exposure of human neutrophils to chemotactic factors poten-
tiates activation of the respiratory burst enzyme. J Immunol 1983;130:2316–2323.
27. Henson PM, Zanolari B, Schwartzman NA, Hong SR. Intracellular control of human neutrophil secre-
tion. I. C5a-induced stimulus-specific desensitisation and the effects of cytochalasin B. J Immunol
1978;121:851–855.
28. Clancy RM, Dahinden CA, Hugli TE. Arachidonate metabolism by human polymorphonuclear leuko-
cytes stimulated by N-formyl-Met-Leu-Phe or complement component C5a is independent of phospho-
lipase activation. Proc Natl Acad Sci USA 1983;80:7200–7204.
29. Palmer RMJ, Salmon JA. Release of leukotriene B4 from human neutrophils and its relationship to
degranulation induced by n-formyl-methionyl-leucyl-phenylalanine, serum-treated zymosan and the
ionophore A23187. Immunology 1983;50:65–73.
30. Cochrane CG, Spragg RG, Revak SD. Studies on the pathogenesis of the adult respiratory distress
syndrome: evidence of oxidants in the broncheoalveolar lavage fluid. J Clin Invest 1983;71:754–761.
31. Royston D, Minty BD, Higenbottam TW, Wallwork J, Jones GJ. The effect of surgery with car-
diopulmonary bypass on alveolar-capillary barrier function in human beings. Ann Thorac Surg
1985;40:139–143.
32. Rinaldo JE, Rogers RM. Adult respiratory distress syndrome. Changing concepts of lung injury and
repair. N Engl J Med 1982;306:900–909.
33. Dinarello CA. Interleukin-1. Rev Infect Dis 1984;6:51–95.
34. Smith RJ, Speziale SC, Bowman BJ. Properties of interleukin-1 as a complete secretagogue for human
neutrophils. Biochem Biophys Res Commun 1982;130:1233–1240.
35. Mizel SB. Interleukin 1 and T cell activation. Immunol Rev 1982;63:51–72.
36. Falkoff RJM, Muraguchi A, Hong JX, Buttler JL, Dinarello CA, Fanci AS. The effects of interleukin
1 on human B cell activation and proliferation. J Immunol 1983;131:801–805.
37. Old LJ. Tumor necrosis factor (TNF). Science 1985;230:630–632.
38. Dinarello CA, Cannon JG, Wolff SM, et al. Tumor necrosis factor (cachectin) is an endogenous
pyrogen and induces production of interleukin 1. J Exp Med 1986;163:1433–1450.
39. Nawroth PP, Stern D. Modulation of endothelial cell hemostatic properties by tumor necrosis factor.
J Exp Med 1986;164:740–745.
40. Nawroth PP, Bank I, Handley D, Cassimeris J, Chess L, Stern D. Tumor necrosis factor/cachectin
interacts with endothelial cell receptors to induce release of interleukin 1. J Exp Med 1986;163:
1363–1375.
41. Jansen NJ, van Oeveren W, van de Broek L, et al. Inhibition by dexamethasone of the reperfusion
phenomena in cardiopulmonary bypass. J Thorac Cardiovasc Surg 1991;102:515–525.
42. Jansen NJ, van Oeveren W, Gu YJ, van Vliet MH, Eijsman L, Wildevuur CR. Endotoxin release and
tumor necrosis factor formation during cardiopulmonary bypass. Ann Thorac Surg 1992;54:744–748.
43. Fransen E, Maessen J, Dentener M, Senden N, Geskes G, Buurman W. Systemic inflammation present
in patients undergoing CABG without extracorporeal circulation. Chest 1998;113:1290–1295.
44. Schulze C, Conrad N, Schutz A, et al. Reduced expression of systemic proinflammatory cytokines after
off-pump versus conventional coronary artery bypass grafting. Thorac Cardiovasc Surg 2000;48:
364–369.
45. Wan S, Marchant A, DeSmet JM, et al. Human cytokine responses to cardiac transplantation and
coronary artery bypass grafting. J Thorac Cardiovasc Surg 1996;111:469–477.
46. Butler J, Chong GL, Baigrie RJ, Pillai R, Westaby S, Rocker GM. Cytokine responses to cardiopulmo-
nary bypass with membrane and bubble oxygenation. Ann Thorac Surg 1992;53:833–838.
47. Kukielka GL, Smith CW, Manning AM, Youker KA, Michael LH, Entman ML. Induction of
interleukin-6 synthesis in the myocardium. Potential role in postreperfusion inflammatory injury.
Circulation 1995;92:1866–1875.
48. Jorens PG, de Jongh R, de Backer W, et al. Interleukin-8 production in patients undergoing cardiop-
ulmonary bypass. The influence of pre-treatment with methylprednisolone. Am Rev Respir Dis
1993;148:890–895.
49. Ivey CL, Williams FW, Collins PD, Jose PJ, Williams TJ. Neutrophil chemoattractants generated in
two phases during reperfusion of ischemic myocardium in the rabbit. Evidence for a role for C5a and
interleukin-8. J Clin Invest 1995;95:2720–2728.
50. Gearing AJH, Newman W. Circulating adhesion molecules in disease. Immunol Today 1993;14:
506–512.
51. Gu YJ, van Oeveren W, Boonstra PW, de Haan J, Wildevuur CR. Leukocyte activation with increased
membrane expression of CR3 receptors induced by cardiopulmonary bypass. Ann Thorac Surg
1992;53:839–844.
52. Gillinov AM, Bator JM, Zehr KJ, et al. Neutrophil adhesion molecule expression during cardiopulmo-
nary bypass with bubble and membrane oxygenators. Ann Thorac Surg 1993;56:847–853.
53. Arnaout MA, Hakim RM, Todd RF III, Dana N, Colten HR. Increased expression of an adhesion-
promoting surface glycoprotein in the granulocytopenia of hemodialysis. N Engl J Med 1985;312:457–462.
54. Etzioni A. Adhesion molecules—their role in health and disease. Pediatr Res 1996;39:191–198.
55. Dreyer WJ, Michael LH, Millman EE, Berens KL. Neutrophil activation and adhesion molecule
expression in a canine model of open heart surgery with cardiopulmonary bypass. Cardiovasc Res
1995;29:775–781.
56. van Oeveren W, Eijsman L, Roozendaal KJ, Wildevuur CR. Platelet preservation by aprotinin during
cardiopulmonary bypass. Lancet 1988;19:644.
57. Moncada S, Higgs A. The L-arginine-nitric oxide pathway. N Engl J Med 1993;329:2002–2012.
58. Speziale G, Ruvolo G, Marino B. A role for nitric oxide in the vasoplegic syndrome. J Card Surg
(Torino) 1996;37:301–303.
59. Finkel MS, Oddis CV, Jacob TD, et al. Negative inotropic effects of cytokines on the heart mediated
by nitric oxide. Science 1992;257:387–389.
60. Alican I, Kubes P. A critical role for nitric oxide in intestinal barrier function and dysfunction. Am J
Physiol 1996;270:G225–G237.
61. Yanagisawa M, Kurihara H, Kimura S, et al. A novel potent vasoconstrictor peptide produced by
vascular endothelial cells. Nature 1988;332:411–415.
62. Yoshizawa T, Osamu S, Giaid A, et al. Endothelin, a novel peptide in the posterior pituitary system.
Science 1989;247:462–464.
63. te Velthuis H, Jansen PG, Oudemans-van Straaten HM, et al. Circulating endothelin in cardiac opera-
tions: influence of blood pressure and endotoxin. Ann Thorac Surg 1996;61:904–908.
64. Dorman BH, Bond BR, Clair MJ, et al. Temporal synthesis and release of endothelin within the
systemic and myocardial circulation during and after cardiopulmonary bypass: relation to postopera-
tive recovery. J Cardiothorac Vasc Anesth 2000;14:540–545.
65. Kirshbom PM, Tsui SS, Di Bernardo LR, et al. Blockade of endothelin-converting enzyme reduces
pulmonary hypertension after cardiopulmonary bypass and circulatory arrest. Surgery 1995;118:
440–444.
66. Matheis G, Haak T, Beyersdorf F, Baretti R, Polywka C, Winkelmann BR. Circulating endothelin in
patients undergoing coronary artery bypass grafting. Eur J Cardiothorac Surg 1995;9:269–274.
67. Nakamura H, Kim DK, Philbin DM, et al. Heparin-enhanced plasma phospholipase A2 activity and
prostacyclin synthesis in patients undergoing cardiac surgery. J Clin Invest 1995;95:1062–1070.
68. Tabuchi N, Gallandat Huet RC, Sturk A, Eijsman L, Wildevuur CR. Aprotinin effects on aspirin treated
platelets and hemostasis during cardiopulmonary bypass. Ann Thorac Surg 1994;58:1036–1039.
69. Suhara H, Sawa Y, Nishimura M, Oshiyama H, Yokoyama K, Saito N, Matsuda H. Efficacy of a new
coating material, PMEA, for cardiopulmonary bypass circuits in a porcine model. Ann Thorac Surg
2001;71:1603–1608.
70. Wimmer-Greinecker G, Matheis G, Martens S, Oremek G, Abdel-Rahman U, Moritz A. Synthetic
protein treated versus heparin coated cardiopulmonary bypass surfaces: similar clinical results and
minor biochemical differences. Eur J Cardiothorac Surg 1999;16:211–217.
71. De Somer F, Francois K, van Oeveren W, et al. Phosphorylcholine coating of extracorporeal circuits
provides natural protection against blood activation by the material surface. Eur J Cardiothorac Surg
200;18:602–606.
72. Videm V, Svennevig JL, Fosse E, Semb G, Osterud A, Mollnes TE. Reduced complement activation
with heparin-coated oxygenator and tubings in coronary bypass operations. J Thorac Cardiovasc Surg
1992;103:806–813.
73. Ovrum E, Mollnes TE, Fosse E, et al. Complement and granulocyte activation in two different types
of heparinized extracorporeal circuits. J Thorac Cardiovasc Surg 1995;110:1623–1632.
74. Lundblad R, Moen O, Fosse E. Endothelin-1 and neutrophil activation during heparin-coated cardiop-
ulmonary bypass. Ann Thorac Surg 1997;63:1361–1367.
75. Moen O, Fosse E, Brockmeier V, et al. Disparity in blood activation by two different heparin-coated
cardiopulmonary bypass systems. Ann Thorac Surg 1995;60:1317–1323.
76. Steinberg BM, Grossi EA, Schwartz DS, et al. Heparin bonding of bypass circuits reduces cytokine
release during cardiopulmonary bypass. Ann Thorac Surg 1995;60:525–529.
77. Weerwind PW, Maessen JG, van Tits LJ, et al. Influence of Duraflo II heparin-treated extracorporeal
circuits on the systemic inflammatory response in patients having coronary bypass. J Thorac Cardiovas
Surg 1995;110:1633–1641.
78. Bozdayi M, Borowiec J, Nilsson L, Venge P, Thelin S, Hansson HE. Effects of heparin-coating of
cardiopulmonary bypass circuits on in vitro oxygen free radical production during coronary bypass
surgery. Artif Organs 1996;20:1008–1016.
79. Moen O, Hogasen K, Fosse E, et al. Attenuation of changes in leukocyte surface markers and comple-
ment activation with heparin-coated cardiopulmonary bypass. Ann Thorac Surg 1997;63:105–111.
80. Fukutomi M, Kobayashi S, Niwaya K, Hamada Y, Kitamura S. Changes in platelet, granulocyte and
complement activation during cardiopulmonary bypass using heparin-coated equipment. Artif Organs
1996;20:767–776.
81. Gu YJ, van Oeveren W, Akkerman C, Boonstra PW, Huyzen RJ, Wildevuur CR. Heparin-coated circuits
reduce the inflammatory response to cardiopulmonary bypass. Ann Thorac Surg 1993;55:917–922.
82. te Velthuis H, Jansen PGM, Hack CE, Eijsman L, Wildevuur CR. Specific complement inhibition by
heparin-coated extracorporeal circuits. Ann Thorac Surg 1996;61:1153–1157.
83. van der Kamp KW, van Oeveren W. Contact, coagulation and platelet interaction with heparin treated
equipment during heart surgery. Int J Artif Organs 1993;16:836–842.
84. Gorman RC, Ziats N, Rao AK, et al. Surface-bound heparin fails to reduce thrombin formation during
clinical cardiopulmonary bypass. J Thorac Cardiovasc Surg 1996;111:1–12.
85. Ovrum E, Brosstad F, Am Holen E, Tangen G, Abdelnoor M. Effects on coagulation and fibrinolysis
with reduced versus full heparinization and heparin coated cardiopulmonary bypass. Circulation
1995;92:2579–2584.
86. von Segesser LK, Weiss BM, Garcia E, von Felten A, Turina MI. Reduction and elimination of systemic
heparinization during cardiopulmonary bypass. J Thorac Cardiovasc Surg 1992;103:790–799.
87. Ovrum E, Holen EA, Tangen G, et al. Completely heparinized cardiopulmonary bypass and reduced
systemic heparin: clinical and hemostatic effects. Ann Thorac Surg 1995;60:365–371.
88. Edmunds LH Jr. Surface-bound heparin; panacea or peril? Ann Thorac Surg 1994;85:285–286.
89. Ranucci M, Cirri S, Conti D, et al. Beneficial effects of Duraflo II heparin-coated circuits on post-
perfusion lung dysfunction. Ann Thorac Surg 1996 61:76–81.
90. Jansen PG, te Velthuis H, Huybrechts RA, et al. Reduced complement activation and improved post-
operative performance after cardiopulmonary bypass with heparin-coated circuits. J Thorac Cardiovasc
Surg 1995;110:829–834.
91. Jansen PG, Baufreton C, Le Besnerais P, Loisance DY, Wildevuur ChRH. Heparin-coated circuits and
aprotinin prime for coronary artery bypass grafting. Ann Thorac Surg 1996;61:1363–1366.
92. Baufreton C, Le Besnerais P, Jansen P, et al. Clinical outcome after coronary surgery with heparin-
coated extracorporeal circuits for cardiopulmonary bypass. Perfusion 1996;11:437–443.
93. Wildevuur CR, Jansen PG, Bezemer PD, et al. Clinical evaluation of Duraflo II treated extracorporeal
circuits (2nd version). The European Working Group on heparin coated extracorporeal circulation
circuits. Eur J Cardiothorac Surg 1997;11:616–623.
94. Verstraete M. Clinical application of inhibitors of fibrinolysis. Drugs 1985;29:236–261.
95. van Oeveren W, Jansen NJ, Bidstrup BP, et al. Effects of aprotinin on hemostatic mechanisms during
cardiopulmonary bypass. Ann Thorac Surg 1987;44:640–645.
96. Wildevuur CR, Eijsman L, Roozendaal KJ, Harder MP, Chang MP, van Oeveren W. Platelet preser-
vation during cardiopulmonary bypass with aprotinin. Eur J Cardiothor Surg 1989;3:533–538.
97. van Oeveren W, Harder MP, Roozendaal KJ, Eijsman L, Wildevuur CR. Aprotinin protects platelets
against the initial effect of cardiopulmonary bypass. J Thorac Cardiovasc Surg 1990;99:788–797.
98. Speekenbrink RG, Wildevuur CR, Sturk A, Eijsman L. Low-dose and high-dose aprotinin improve
hemostasis in coronary surgery. J Thorac Cardiovasc Surg 1996;112:523–530.
99. Tatar H, Cicek S, Demirkilic U, et al. Topical use of aprotinin in open heart operations. Ann Thorac
Surg 1993;55:659–661.
100. Speekenbrink RG, Vonk AB, Wildevuur CR, Eijsman L. Hemostatic efficacy of dipyridamole,
tranexamic acid and aprotinin in coronary bypass grafting. Ann Thorac Surg 1995;59:438–442.
101. Maquelin KN, Nieuwland R, Lentjes EG, et al. Aprotinin administration in the pericardial cavity does
not prevent platelet activation. J Thorac Cardiovasc Surg 2000:120:552–557.
102. Horrow JC, Hlavacek J, Strong MD, et al. Prophylactic tranexamic acid decreases bleeding after
cardiac operations. J Thorac Cardiovasc Surg 1990;99:70–74.
103. John LC, Rees GM, Kovacs IB. Reduction of heparin binding to and inhibition of platelets by aprotinin.
Ann Thorac Surg 1993;55:1175–1179.
104. Wachtfogel YT, Kucich U, Hack CE, et al. Aprotinin inhibits the contact, neutrophil, and platelet
activation systems during simulated extracorporeal perfusion. J Thorac Cardiovasc Surg 1993;106:1–10.
105. Asimakopoulos G, Lidington EA, Mason J, Haskard DO, Taylor KM, Landis RC. Effect of aprotinin
on endothelial cell activation. J Thorac Cardiovasc Surg 2001;122:123–128.
106. Asimakopoulos G, Taylor KM, Haskard DO, Landis RC. Inhibition of neutrophil L-selectin shedding:
a potential anti-inflammatory effect of aprotinin. Perfusion 2000;15:495–499.
107. Asimakopoulos G, Thompson R, Nourshargh S, et al. An anti-inflammatory property of aprotinin
detected at the level of leukocyte extravasation. J Thorac Cardiovasc Surg 2000;120:361–369.
108. Hill GE, Taylor JA, Robbins RA. Differing effects of aprotinin and e-aminocaproic acid on cytokine-
induced inducible nitric oxide synthase expression. Ann Thorac Surg 1997;63:74–77.
109. Hill GE, Springal DR, Robbins RA. Aprotinin is associated with a decrease in nitric oxide production
during cardiopulmonary bypass. Surgery 1997;121:449–455.
110. Hill GE, Alonso A, Spurzem JR, Stammers AH, Robbins RA. Aprotinin and methylprednisolone
equally blunt cardiopulmonary bypass-induced inflammation in humans. J Thorac Cardiovasc Surg
1995;110:1658–1662.
111. Asimakopoulos G, Kohn A, Stefanou DC, Haskard DO, Landis RC, Taylor KM. Leukocyte integrin
expression in patients undergoing cardiopulmonary bypass. Ann Thorac Surg 2000;69:1192–1197.
112. Alonso A, Whitten CW, Hill GE. Pump prime only aprotinin inhibits cardiopulmonary bypass-induced
neutrophil CD11b up-regulation. Ann Thorac Surg 1999;67:392–395.
113. Isbir CS, Dogan R, Demircin M, Yaylim I, Pasaoglu I. Aprotinin reduces the IL-8 after coronary artery
bypass grafting. Cardiovasc Surg 2001;9:403–406.
114. Tassani P, Augustin N, Barankay A, Braun SL, Zaccaria F, Richter JA. High-dose aprotinin modulates
the balance between proinflammatory and anti-inflammatory responses during coronary artery bypass
graft surgery. J Cardiothorac Vasc Anesth 2000;14:682–686.
115. Hill GE, Pohorecki R, Alonso A, Rennard SI, Robbins RA. Aprotinin reduces interleukin-8 production
and neutrophil accumulation after cardiopulmonary bypass. Anesth Analg 1996;83:696–700.
116. Rahman A, Ustunda B, Burma O, Ozercan IH, Cekirdekci A, Bayar MK. Does aprotinin reduce lung
reperfusion damage after cardiopulmonary bypass? Eur J Cardiothorac Surg 2000;18:583–588.
117. Rich JB. The efficacy and safety of aprotinin use in cardiac surgery. Ann Thorac Surg 199;66:S6–S11.
118. van Oeveren W, van Oeveren B, Wildevuur CR. Anticoagulation policy during use of aprotinin in
cardiopulmonary bypass. J Thorac Cardiovasc Surg 1992;104:210–211.
119. Feindt P, Seyfert U, Volkmar I, Huwer H, Kalweit G, Gams E. Is there a phase of hypercoagulability
when aprotinin is used in cardiac surgery? Eur J Cardiothor Surg 1994;8:308–314.
120. Bidstrup BP, Underwood SR, Sapsford RN. Effect of aprotinin (Trasylol) on aorta-coronary bypass
graft patency. J Thorac Cardiovasc Surg 1993;105:147–153.
121. Westaby S. Aprotinin in perspective. Ann Thorac Surg 1993;55:1033–1041.
122. Speekenbrink RG, Bertina RM, España F, Wildevuur CR, Eijsman L. Activation of the protein C
anticoagulant system during cardiopulmonary bypass and the influence of aprotinin. Ann Thorac Surg
1998;66:1998–2002.
123. Weiler JM, Packard B. Methylprednisolone inhibits the alternative and amplification pathways of
complement. Infect Immun 1982;38:122–126.
124. Boscoe MJ, Yewdall VM, Thompson MA, Cameron JS. Complement activation during cardiopulmo-
nary bypass: quantitative study of effects of methylprednisolone and pulsatile flow. Br Med J (Clin Res
Ed) 1983;287:1747–1750.
125. Jansen NJ, van Oeveren W, van Vliet M, Stoutenbeek CP, Eijsman L, Wildevuur CR. The role of
different types of corticosteroids on the inflammatory mediators in cardiopulmonary bypass. Eur J
Cardiothorac Surg 1991;5:211–217.
126. Hill GE, Snider S, Galbraith TA, Forst S, Robbins RA. Glucocorticoid reduction of bronchial epithelial
inflammation during cardiopulmonary bypass. Am J Respir Crit Care Med 1995;152:1791–1795.
127. Kawamura T, Inada K, Okada H, Okada K, Wakusawa R. Methylprednisolone inhibits increase of
interleukin 8 and 6 during open heart surgery. Can J Anaesth 1995;42:399–403.
128. Teoh KH, Bradley CA, Gauldie J, Burrows H. Steroid inhibition of cytokine-mediated vasodilation
after warm heart surgery. Circulation 1995;92:347–353.
129. Tabardel Y, Duchateau J, Schmartz D, et al. Corticosteroids increase blood interleukin-10 levels during
cardiopulmonary bypass in men. Surgery 1996;119:76–80.
130. Tassani P, Richter JA, Barankay A, et al. Does high-dose methylprednisolone in aprotinin-treated
patients attenuate the systemic inflammatory response during coronary artery bypass grafting proce-
dures? J Cardiothorac Vasc Anesth 1999;13:165–172.
131. Hill GE, Alonso A, Thiele GM, Robbins RA. Glucocorticoids blunt neutrophil CD11b surface glyco-
protein upregulation during cardiopulmonary bypass in humans. Anesth Analg 1994;79:23–27.
132. Cronstein BN, Kimmel SC, Levin RI, Martiniuk F, Weissman G. A mechanism for the anti-inflamma-
tory effects of corticosteroids: the glucocorticoid receptor regulates leukocyte adhesion to endothelial
cells and expression of endothelial-leukocyte adhesion molecule 1 and intercellular adhesion molecule
1. Proc Natl Acad Sci USA 1992; 89:9991–9995.
133. Busuttil RW, George WJ, Hewitt RL. Protective effect of methylprednisolone on the heart during
ischemic arrest. J Thorac Cardiovasc Surg 1975;70:955–965.
134. Hill DG, Aguilar MJ, Kosek JC, Hill JD. Corticosteroids and prevention of pulmonary damage follow-
ing cardiopulmonary bypass in puppies. Ann Thorac Surg 1976;22:36–40.
135. Tabuchi N, de Haan J, Boonstra PW, van Oeveren W. Activation of fibrinolysis in the pericardial cavity
during cardiopulmonary bypass. J Thorac Cardiovasc Surg 1993;106:828–833.
136. Nieuwland R, Berckmans RJ, Rotteveel-Eijkman RC, et al. Cell-derived microparticles generated in
patients during cardiopulmonary bypass are highly procoagulant. Circulation 1997;96:3534–3541.
137. Philippou H, Adami A, Davidson SJ, Pepper JR, Burman JF, Lane DA. Tissue factor is rapidly elevated
in plasma collected from the pericardial cavity during cardiopulmonary bypass. Thromb Haemost
2000;84:124–128.
138. Chung JH, Gikakis N, Rao AK, Drake TA, Colman RW, Edmunds LH Jr. Pericardial blood activates
the extrinsic coagulation pathway during clinical cardiopulmonary bypass. Circulation 1996;93:
2014–2018.
139. van Hinsbergh VW, Kooistra T, Scheffer MA, van Bockel JH, van Muijen GN. Characterization and
fibrinolytic properties of human omental tissue mesothelial cells. Comparison with endothelial cells.
Blood 1990;75:1490–1497.
140. Adelman B, Michelson AD, Loscalzo J, Greenberg J, Handin RI. Plasmin effect on platelet glycopro-
tein Ib-von Willebrand’s factor interaction. Blood 1985;65:32–40.
141. Coller BS. Platelet and thrombolytic therapy. N Engl J Med 1990;99:518–527.
142. de Haan J, Boonstra PW, Monnink SH, Ebels T, van Oeveren W. Retransfusion of suctioned blood
during cardiopulmonary bypass impairs hemostasis. Ann Thorac Surg 1995;59:901–907.
143. Schönberger JP, van Oeveren W, Bredee JJ, Everts PA, de Haan J, Wildevuur CR. Systemic blood
activation during and after autotransfusion Ann Thorac Surg 1994;57:1256–1262.
144. Schönberger JP, Bredee JJ, Speekenbrink RG, Everts PA, Wildevuur CR. Autotransfusion of shed
blood contributes additionally to blood saving in patients receiving aprotinin (2 million KIU). Eur J
145. Boonstra PW, van Imhoff GW, Eijsman L, et al. Reduced platelet activation and improved hemostasis
after controlled cardiotomy suction during clinical membrane oxygenator perfusions. J Thorac
Cardiovasc Surg 1985;89:900–906.
146. Menasché P, Piwnica A. Free radicals and myocardial protection: a surgical viewpoint. Ann Thorac
Surg 1989;47:939–945.
147. Royston D, Fleming JS, Desar JB, Westaby S, Taylor KM. Increased production of peroxidation
products associated with cardiac operations. J Thorac Cardiovasc Surg 1986;91:759–766.
148. Lefer AM. Role of selectins in myocardial ischemia-reperfusion injury. Ann Thorac Surg 1995;60:773–777.
149. Seccombe JF, Schaff HV. Coronary artery endothelial function after myocardial ischemia and
reperfusion. Ann Thorac Surg 1995;60:778–788.
150. Menasché P, Grousset C, Gauduel Y, Piwnica A. A comparative study of free radical scavengers in
cardioplegic solutions. Improved protection with peroxidase. J Thorac Cardiovasc Surg 1986;92:
264–271.
151. Kaneda T, Ku K, Inoue T, Onoe M, Oku H. Postischemic reperfusion injury can be attenuated by
oxygen tension control. Jpn Circ J 2001;65:213–218.
152. Bochenek A, Religa Z, Spyt TJ, Mistarz K, Bochenek Ad, Zembala M, Grzybek H. Protective influence
of pretreatment with allopurinol on myocardial function in patients undergoing coronary artery surgery.
Eur J Cardiothorac Surg 1990;4:538–542.
153. Lichtenstein SV, Kassam AA, El Dalati H, Cusimano RJ, Panos A, Slutsky AS. Warm heart surgery.
J Thorac Cardiovasc Surg 1991;101:269–274.
154. Taggart DP, El-Fiky MM, Carter R, Bowman A, Wheatley DJ. Respiratory dysfunction after uncom-
plicated cardiopulmonary bypass. Ann Thorac Surg 1993;56:1123–1128.
155. Ratcliff NB, Young WG Jr, Hackel DB, et al. Pulmonary injury secondary to extracorporeal circulation:
an ultrastructural study. J Thorac Cardiovasc Surg 1973;65:425–432.
156. Schlensak C, Doenst T, Preusser S, Wunderlich M, Kleinschmidt M, Beyersdorf F. Bronchial artery
perfusion during cardiopulmonary bypass does not prevent ischemia of the lung in piglets: assessment
of bronchial artery blood flow with fluorescent microspheres. Eur J Cardiothorac Surg 2001;19:
326–331.
157. Hillman ND, Cheifetz IM, Craig DM, Smith PK, Ungerleider RM, Meliones JN. Inhaled nitric oxide,
right ventricular efficiency, and pulmonary vascular mechanics: selective vasodilation of small pulmo-
nary vessels during hypoxic pulmonary vasoconstriction. J Thorac Cardiovasc Surg 1997;113:
1006–1013.
158. King RC, Binns OA, Kanithanon RC, et al. Low-dose sodium nitroprusside reduces pulmonary
reperfusion injury. Ann Thorac Surg 1997;63:1398–1404.
159. Pearl JM, Wellmann SA, McNamara JL, et al. Bosentan prevents hypoxia-reoxygenation-induced
pulmonary hypertension and improves pulmonary function. Ann Thorac Surg 1999;68:1714–1721.
160. MacNee W, Selby C. Neutrophil kinetics in the lungs. Clin Sci 1990;79:97–107.
161. Tönz M, Mihaljevic T, von Segesser LK, Fehr J, Schmid ER, Turina MI. Acute lung injury during
cardiopulmonary bypass: are the neutrophils responsible? Chest 1995;108:1551–1556.
162. Bando K, Pillai R, Cameron DE, et al. Leukocyte depletion ameliorates free radical-mediated lung
injury after cardiopulmonary bypass. J Thorac Cardiovasc Surg 1990;99:873–877.
163. Johnson D, Thomson D, Mycyk T, Burbridge B, Mayers I. Depletion of leucocytes transiently improves
postoperative cardiorespiratory status. Chest 1995;107:1253–1259.
164. Gu YJ, Vries AJ de, Boonstra PW, van Oeveren W. Leukocyte depletion results in improved lung
function and reduced inflammatory response after cardiac surgery. J Thorac Cardiovasc Surg
1996;112;494–500.
165. Boldt J, Zickmann B, Dapper F, Hempelmann G. Does the technique of cardiopulmonary bypass affect
lung water content? Eur J Cardiothorac Surg 1991;5:22–26.
166. Matheis G, Haak T, Beyersdorf F, Baretti R, Polywka C, Winkelmann BR. Circulating endothelin in
patients undergoing coronary artery bypass grafting. Eur J Cardiothorac Surg 1995;9:269–274.
167. Dobell AR, Bailey JS. Charles Drew and the origins of deep hypothermic circulatory arrest. Ann Thorac
Surg 1997;63:1193–1199.
168. Bochenek A, Religa Z, Kokot F, et al. Biocompatibility of extracorporeal circulation with auto-
oxygenation. Eur J Cardiothorac Surg 1992;6:397–402.
169. DeFoe GR, Ross CS, Olmstead EM, et al. Lowest hematocrit on bypass and adverse outcomes asso-
ciated with coronary artery bypass grafting. Ann Thorac Surg 2001;71:769–776.
170. Beattie HW, Evans G, Garnett ES, Webber CE. Sustained hypovolemia and extracellular fluid volume
expansion following cardiopulmonary bypass. Surgery 1972;71:891–897.
171. Utley JR, Wachtel C, Cain RB, Spaw AE, Collins JC, Stephens DB. Effects of hypothermia, hemodi-
lution, and pump oxygenation on organ water content, blood flow, and oxygen delivery, and renal
function. Ann Thorac Surg 1981;31:121–133.
172. Jansen PG, te Velthuis H, Bulder ER, et al. Reduction in prime volume attenuates the hyperdynamic
response after cardiopulmonary bypass. Ann Thorac Surg 1995;60:544–550.
173. Jansen PG, te Velthuis H, Wildevuur WR, et al. Cardiopulmonary bypass with modified fluid gelatin
and heparin-coated circuits. Br J Anaesth 1996;6:13–19.
174. Schönberger JP, Bredee JJ, Tjian D, Everts PA, Wildevuur CR. Intraoperative predonation contributes
to blood saving. Ann Thorac Surg 1993;56:893–898.
175. Schönberger JP, Woolley S, Tavilla G, et al. Efficacy and safety of a blood conservation program
including low-dose aprotinin in routine myocardial revascularization. J Cardiovasc Surg (Torino)
1996;37:35–44.
176. Menasché P, Haydar S, Peynet J, et al. A potential mechanism of vasodilation after warm heart surgery.
177. Menasché P, Peynet J, Lariviere J, et al. Does normothermia during cardiopulmonary bypass increase
neutrophil-endothelium interactions? Circulation 1994;90:II275–II279.
178. Haddix TL, Pohlman TH, Noel RF, Sato TT, Boyle EM Jr, Verrier ED. Hypothermia inhibits human
E-selectin transcription. J Surg Res 1996;64:176–183.
179. Menasché P, Peynet J, Haeffner-Cavaillon N, et al. Influence of temperature on neutrophil trafficking
during clinical cardiopulmonary bypass. Circulation 1995;92(suppl):II334–II340.
180. Frering B, Philip I, Dehoux M, Rolland C, Langlois JM, Desmonts JM. Circulating cytokines in patients
undergoing normothermic cardiopulmonary bypass. J Thorac Cardiovasc Surg 1994;108:636–641.
181. Birdi I, Regragui I, Izzat MB, Bryan AJ, Angelini GD. Influence of normothermic systemic perfusion
during coronary artery bypass operations: a randomized prospective study. J Thorac Cardiovasc Surg
1997;114:475–481.
182. Tonz M, Mihaljevic T, von Segesser LK, Shaw S, Luscher TF, Turina M. Postoperative hemodynamics
depend on cardiopulmonary bypass temperature: the potential role of endothelin-1. Eur J Cardiothorac
Surg 1997;11:157–161.
183. Ohata T, Sawa Y, Kadoba K, Kagisaki K, Suzuki K, Matsuda H. Role of nitric oxide in a temperature
dependent regulation of systemic vascular resistance in cardiopulmonary bypass. Eur J Cardiothorac
Surg 2000;18:342–347.
184. Ranucci M, Soro G, Frigiola A, et al. Normothermic perfusion and lung function after cardiopulmonary
bypass: effects in pulmonary risk patients. Perfusion 1997;12:309–315.
185. Ohata T, Sawa Y, Kadoba K, Masai T, Ichikawa H, Matsuda H. Effect of cardiopulmonary bypass under
tepid temperature on inflammatory reactions. Ann Thorac Surg 1997;64:124–128.
186. Adams DC, Heyer EJ, Simon AE, et al. Incidence of atrial fibrillation after mild or moderate hypoth-
ermic cardiopulmonary bypass. Crit Care Med 2000;28:574–545.
187. Vingerhoets G, Van Nooten G, Vermassen F, De Soete G, Jannes C. Short-term and long-term neurop-
sychological consequences of cardiac surgery with extracorporeal circulation. Eur J Cardiothorac Surg
1997;11:424–431.
188. Sotaniemi KA. Long-term neurologic outcome after cardiac operation. Ann Thorac Surg
1995;59:1336–1339.
189. Blauth CI. Macroemboli and microemboli during cardiopulmonary bypass. Ann Thorac Surg
1995;59:1300–1303.
190. Liu JF, Su ZK, Ding WX. Quantitation of particulate microemboli during cardiopulmonary bypass:
experimental and clinical studies. Ann Thorac Surg 1992;54:1196–1202.
191. Moody DM, Brown WR, Challa VR, Stump DA, Reboussin DM, Legault C. Brain microemboli
associated with cardiopulmonary bypass: a histologic and magnetic resonance imaging study. Ann
Thorac Surg 1995;59:1304–1307.
192. Kincaid EH, Jones TJ, Stump DA, et al. Processing scavenged blood with a cell saver reduces cerebral
lipid microembolization. Ann Thorac Surg 2000;70:1296–1300.
193. Plourde G, Leduc AS, Morin JE, et al. Temperature during cardiopulmonary bypass for coronary artery
operations does not influence postoperative cognitive function: a prospective, randomized trial.
194. Engelman RM, Pleet AB, Rousou JA, et al. What is the best perfusion temperature for coronary
revascularization? J Thorac Cardiovasc Surg 1996;112:1622–1632.
195. Regragui I, Birdi I, Izzat MB, et al. The effects of cardiopulmonary bypass temperature on
neuropsychologic outcome after coronary artery operations: a prospective randomized trial. J Thorac
196. Mora CT, Henson MB, Weintraub WS, et al. The effect of temperature management during cardiop-
ulmonary bypass on neurologic and neuropsychologic outcomes in patients undergoing coronary
revascularization. J Thorac Cardiovasc Surg 1996;112:514–522.
197. McLean RF, Wong BI, Naylor CD, et al. Cardiopulmonary bypass, temperature, and central nervous
system dysfunction. Circulation 1994;90:II250–II255.
198. Martin TD, Craver JM, Gott JP, et al. Prospective, randomized trial of retrograde warm blood cardiople-
gia: myocardial benefit and neurological threat. Ann Thorac Surg 1994;57:298–304.
199. Engelman RM, Pleet AB, Rousou JA, et al. Influence of cardiopulmonary bypass perfusion temperature
on neurologic and hematologic function after coronary artery bypass grafting. Ann Thorac Surg
1999;67:1547–1555.
200. Grimm M, Czerny M, Baumer H, et al. Normothermic cardiopulmonary bypass is beneficial for cog-
nitive brain function after coronary artery bypass grafting—a prospective randomized trial. Eur J
201. Rees K, Beranek-Stanley M, Burke M, Ebrahim S. Hypothermia to reduce neurological damage
following coronary artery bypass surgery (Cochrane Review). Cochrane Database Syst Rev
2001;1:CD002138.
202. Johnsson P, Lundqvist C, Lindgren A, Ferencz I, Alling C, Stahl E. Cerebral complications after cardiac
surgery assessed by S-100 and NSE levels in blood. J Cardiothorac Vasc Anesth 1995;9:694–699.
203. Ali MS, Harmer M, Vaughan R. Serum S100 protein as a marker of cerebral damage during cardiac
surgery. Br J Anaesth 2000;85:287–298.
204. Westaby S, Johnsson P, Parry A, et al. Serum S100 protein: a potential marker for cerebral events during
205. Taggart DP, Mazel JW, Bhattacharya K, et al. Comparison of serum S-100ß levels during CABG and
intracardiac operations. Ann Thorac Surg 1997;63:492–496.
206. Taggart DP, Bhattacharya K, Meston N, et al. Serum S-100 protein concentration after cardiac surgery:
a randomized trial of arterial line filtration. Eur J Cardiothorac Surg, 1997;11:645–649.
207. Anderson RE, Hansson LO, Vaage J. Release of S100B during coronary artery bypass grafting is
reduced by off-pump surgery. Ann Thorac Surg 1999;67:1721–1725.
208. Khuri SF, Valeri CR, Loscalzo J. Heparin causes platelet dysfunction and induces fibrinolysis before
209. Upchurch GR, Valeri CR, Khuri SF, et al. Effect of heparin on fibrinolytic activity and platelet function
in vivo. Am J Physiol 1996;271:528–534.
210. John LCH, Rees GM, Kovacs IB. Inhibition of platelet function by heparin. J Thorac Cardiovasc Surg
1993;105:816–822.
211. Wahba A, Black G, Koksch M, et al. Cardiopulmonary bypass leads to a preferential loss of activated
platelets. A flow cytometric assay of platelet surface antigens. Eur J Cardiothorac Surg 1996;10:768–773.
212. Videm V. Heparin in clinical doses “primes” granulocytes to subsequent activation as measured by
myeloperoxidase release. Scand J Immunol 1996;43:385–390.
213. Shastri KA, Logue GL, Stern MP, Rehman S, Raza S. Complement activation by heparin-protamine
complexes during cardiopulmonary bypass: effect of C4a null allele. J Thorac Cardiovasc Surg
1997;114:482–488.
214. Levy JH, Cormack JG, Morales A. Heparin neutralization by recombinant platelet factor 4 and prota-
mine. Anesth Analg 1995;81:35–37.
215. Dehmer GJ, Fisher M, Tate DA, Teo S, Bonnem EM. Reversal of heparin anticoagulation by recom-
binant platelet factor 4 in humans. Circulation 1995;91:2188–2194.
216. Riess FC, Potsch B, Behr I, et al. Recombinant hirudin as an anticoagulant during cardiac operations:
experiments in a pig model. Eur J Cardiothorac Surg 1997;11:739–745.
217. Bernabei A, Rao AK, Niewiarowski S, Colman RW, Sun L, Edmunds LH Jr. Recombinant
desulphatohirudin as a substitute for heparin during cardiopulmonary bypass. J Thorac Cardiovasc
Surg 1994;108:381–382.
218. Edmunds LH Jr. HIT, HITT and desulphatohirudin: look before you leap. J Thorac Cardiovasc Surg
1995;110:1–3.
219. Jegger D, Tevaearai HT, Horisberger J, et al. Augmented venous return for minimally invasive open
heart surgery with selective caval cannulation. Eur J Cardiothorac Surg 1999;16:312–316.
220. Nakanishi K, Shichijo T, Shinkawa Y, et al. Usefulness of vacuum-assisted cardiopulmonary bypass
circuit for pediatric open-heart surgery in reducing homologous blood transfusion. Eur J Cardiothorac
Surg 2001;20:233–238.
221. Rosengart TK, DeBois W, O’Hara M, et al.Retrograde autologous priming for cardiopulmonary bypass:
a safe and effective means of decreasing hemodilution and transfusion requirements. J Thorac
222. Groner W, Winkelman JW, Harris AG, et al. Orthogonal polarization spectral imaging: a new method
for study of the microcirculation. Nat Med 1999:10;1209–1213.
Chapter 2 / Endothelial Injury During Bypass Grafting 27
2 Endothelial Injury During Minimally

Invasive Bypass Grafting
Robert J. Dabal, MD, Craig R. Hampton, MD,

and Edward D. Verrier, MD
CONTENTS
INTRODUCTION
OVERVIEW OF ENDOTHELIAL FUNCTION
ACUTE ENDOTHELIAL INJURY
CHRONIC ENDOTHELIAL INJURY
CONCLUSION
REFERENCES
INTRODUCTION
Over the last few decades, the vascular endothelium has emerged as a central mediator
of the biochemical events that underlie the preoperative, operative, and postoperative
course of nearly all patients who undergo cardiovascular interventional procedures and
cardiovascular surgery. While initially felt to be a passive bystander of the whole-body
response to cardiac surgery, it is now clear that the endothelium is a dynamic organ
which is a central regulator of vascular tone, vasomotor function, coagulation, and cel-
lular interactions. Although in the physiologically unstressed state the role of the endo-
thelium is to maintain intravascular homeostasis, it also serves a critical role in the
response to injury. Specifically, in terms of cardiovascular surgery, the endothelium
modulates the systemic inflammatory response, changing vasomotor tone, impacting
coagulation responses, initiating the intimal hyperplastic response, and mediating the
chronic changes leading to atherosclerosis.
Enhanced understanding of these complex biochemical cascades has suggested a causal
role in the morbidity and mortality of cardiovascular operations with cardiopulmonary
bypass (CPB). Accordingly, in order to blunt the systemic and myocardial consequences
of these biochemical alterations and improve clinical outcome, surgical techniques and

27
management strategies have been altered. Recognizing that the utilization of CPB has
been considered a primary stimulator of endothelial cell activation in the myocardial,
pulmonary, and systemic vasculature, there has been a major shift toward the use of
minimally invasive techniques and the elimination of CPB. However, off-pump mini-
mally invasive surgery employs novel techniques, such as vessel snaring, intraluminal
shunts, rotation of the heart, and high-frequency gas insufflation, which can disrupt the
endothelium acutely and lead to short- and long-term consequences (1,2). With the
increasing use of off-pump techniques, recent investigations have attempted to characterize
the cellular and biochemical consequences of eliminating CPB. More important, increased
understanding of these cellular events, and their causal relationship to clinical outcomes,
will allow a more precise characterization of the relative advantages and disadvantages of
off-pump techniques in cardiovascular surgery. This chapter highlights the current under-
standing of the role of the vascular endothelium in cardiovascular surgery, with emphasis
on the effects of eliminating CPB. Moreover, the endothelial effects of some unique aspects
of off-pump surgical techniques (e.g., target vessel snaring) will be discussed.
OVERVIEW OF ENDOTHELIAL FUNCTION

While it was initially felt to be a passive barrier between the intravascular compart-
ment and the body’s tissues, it is now clear that the endothelium is a critical regulator of
inflammation, hemostasis, thrombosis, and fibrinolysis. It is important to note that the
endothelium regulates membrane permeability, lipid transport, vasomotor tone, coagu-
lation, fibrinolysis, and inflammation, and maintains vascular wall structure (Fig. 1). The
endothelium performs these functions by numerous mechanisms, including (1) the
expression of biologically active agents such as surface proteins and soluble factors; (2)
homeostatic regulation of the coagulation and fibrinolytic cascades with a dynamic phe-
notype that can be thromboresistant or prothrombotic; (3) the regulation of cellular
trafficking at the blood–endothelium interface, including the recruitment of cellular
elements to the interstitial space; and (4) the alteration of the arterial wall, including the
intima and media, through the release of growth factors and extracellular matrix (3).
In the inactivated state, the endothelium serves to promote blood flow. However, the
quiescent endothelium may become activated upon exposure to myriad stimuli, including
oxidative stress from ischemia, proinflammatory cytokines (interleukins, tumor necrosis
factor [TNF]), infectious stimuli (lipopolysaccharide [LPS]), increased shear forces, and
physical injury. This activation alters the endothelial cell phenotype. It disrupts the
barrier function of the endothelium, causes enhanced vasoconstriction, alters coagula-
tion, and stimulates leukocyte adhesion and smooth muscle cell proliferation. The net
effect of this phenotypic shift, if excessive and unbalanced, includes severe tissue dam-
age, impaired organ function, and an abnormal fibroproliferative response (3). Moreover,
in addition to the local consequences, endothelial activation may also affect distant tissue
beds through activation of systemic cascades. Thus, the vascular endothelium is a central
regulator of a wide variety of humoral and cellular processes, which are important in
nonpathological and pathological states. Accordingly, the importance of understanding
the molecular details and consequences of activated endothelium in minimally invasive
off-pump cardiovascular surgery is clear.
Vasomotor Function
Vascular tone is controlled largely by the endothelium, which produces a variety of
vasoactive substances that can cause both relaxation and contraction of the affected blood
Fig. 1. Endothelial cell responses to injury.
vessels. While nitric oxide (NO) is perhaps the best-known relaxant factor produced by
the endothelium, prostacyclin (PGI2), bradykinin, endothelium-derived hyperpolarizing
factor, and adenosine are also produced, resulting in vessel dilation and prevention of
thrombosis.
Derived from L-arginine by a family of nitric oxide synthases, NO, formerly called
endothelium-derived relaxing factor, is a vasodilator and potent inhibitor of platelet and
monocyte activation. In the resting state, there is a continuous basal release of NO from
the endothelium. Activation of the endothelium by either platelets or the coagulation
cascade results, however, in increased NO production, which prevents vasoconstriction,
platelet aggregation, and thrombus formation (4).
These vasorelaxing factors are balanced by a group of vasconstrictive substances
including arachidonic acid metabolites including prostanoids (e.g., thromboxane A2,
prostaglandin H2) and the leukotrienes, oxygen free radicals, endothelin, angiotensin II,
and cytokines, including tumor necrosis factor-alpha (TNF-α) and interleukin-1. In par-
ticular, endothelin (ET), working through membrane ET receptors and G-protein-linked
cell-signaling pathways, causes vasoconstriction at both low and high doses. While
intramyocardial vessels are particularly sensitive to endothelin, the epicardial coronary
vessels also respond to endothelin by regulating blood flow.
In patients undergoing cardiovascular surgery with CPB, the finely controlled balance
between vasoconstriction and vasorelaxation is frequently tipped toward a state of
impaired vasomotor control. A variety of conditions commonly seen in the cardiac sur-
gical patient population may contribute to altered endothelial vasomotor function,
including advanced age, diabetes mellitus, hypertension, chronic nicotine use, hypercho-
lesterolemia, and atherosclerosis (5,6)—all of which may predispose the patient to throm-
bosis, hypertension, and atherogenesis. This predisposition reflects both a lack of baseline
vasorelaxation with decreases in NO production and an overproduction of potent vaso-
constrictors such as thromboxane A2, endothelin, and angiotensin II. These alterations in
vasomotor function are particularly deleterious in the patient undergoing cardiovascular
surgery because they can lead to coronary spasm, spasm of the internal mammary con-
duit, and microcirculatory no-reflow leading to acute myocardial ischemia, arrhythmias,
and death (4,7,8).
Off-pump coronary artery bypass grafting (OPCAB) may provide some advantages to
postoperative endothelial vasomotor function by eliminating CPB and reducing myocar-
dial and pulmonary ischemia. Although it is difficult to measure endothelial vasomotor
function directly in humans, a number of investigators have measured coronary (9) and
pulmonary (10) endothelial vasomotor function indirectly in patients undergoing OPCAB
and conventional CAB with CPB.
To assess coronary endothelial-dependent and -independent vasomotor responses,
Lockowandt and Franco-Cereceda gave a bolus injection of either acetylcholine (Ach)
or adenosine (ADO) into a vein graft anastomosed to a coronary target vessel in patients
undergoing OPCAB and compared them to a case-matched cohort undergoing CAB with
CPB (9). Then, flow in the vein graft was measured, as an indirect indicator of coronary
target vessel diameter. They demonstrated that elimination of CPB significantly reduced
dysfunction of coronary endothelium as measured by endothelium-dependent vasorelaxation.
Since there was no true negative control group, the extent to which OPCAB, via brief
periods of target vessel occlusion, impairs endothelium-dependent vasorelaxation was
not determined. In a separate study, these same investigators assessed endothelium-
dependent vasorelaxation in an animal model following 10 min of vessel occlusion, and
demonstrated impairment of endothelium-dependent vasorelaxation after this brief
period of ischemia (11). Taken together, these results suggest that occlusion of the target
vessel during OPCAB may result in some impairment of endothelium-dependent
vasorelaxation, although it seems to be significantly reduced when CPB is eliminated.
Recognizing that CPB results in little or no flow in the pulmonary circulation with
subsequent impaired endothelium-dependent vasorelaxation (12), Angdin et al. assessed
the effect of OPCAB on pulmonary endothelial vasomotor function following CAB. In
a similar study design comparing OPCAB to a case-matched cohort undergoing CAB
with CPB, they demonstrated significantly improved endothelial function when CPB was
eliminated. Compared to preoperative response to Ach (control), there was attenuated
endothelial-dependent vasorelaxation in both groups undergoing CAB. These data are
consistent with the aforementioned coronary endothelium data—that OPCAB signifi-
cantly attenuates, but does not eliminate, endothelial vasomotor dysfunction of the coro-
nary and pulmonary vascular beds following CAB. More important, the long-term
consequences of these observations are not known. It is tempting to postulate that a
reduction in vasomotor dysfunction following CAB may reduce the risk of thrombosis
and vasospasm following coronary revascularization, thus improving both short- and
long-term patency, but this remains unproven.
Coagulation
The vascular endothelium is a master regulator of hemostasis and thrombosis. To this
end, it is strategically positioned at the interface of the blood and the tissues and is the
source, either directly or indirectly, of the majority of regulatory molecules of the hemo-
static and thrombotic cascades. As a result, quiescent endothelium maintains the fluidity
of blood and promotes flow, while activated endothelium potently causes coagulation,
through activation of platelets and the coagulation cascades. In this regard, the endothe-
lium finely orchestrates the pro- and anti-thrombotic effector molecules in the vascular
space, which has been reviewed in detail elsewhere (13,14).
Prostacyclin (PGI2) and NO are two of the best-characterized endothelial-derived
inhibitors of platelet aggregation and thrombosis (13). Both potently inhibit platelet
aggregation and simultaneously cause vasodilation, thus serving to limit the extent of
intravascular thrombus formation at sites of endothelial injury or stress. Moreover, PGI2
and NO may act synergistically to inhibit platelet aggregation (15). Taken together,
normal functioning of PGI2 and NO are critical for normal hemostasis, and abnormal
function of either may predispose to excessive platelet aggregation and thrombosis.
An additional endothelial-derived molecule important in suppressing platelet activa-
tion is ecto-adenosine diphosphatase (ADPase) (13,14). In contrast to NO and PGI2,
which are secreted to act in a paracrine fashion, ADPase is expressed on the cell surface,
where it enzymatically degrades ADP, a potent potentiator of platelet recruitment and
activation, thereby limiting the extent of platelet activation (16).
In addition to the endothelial-derived antiplatelet molecules listed above, the endot-
helium also antagonizes the coagulation cascade through synthesis of multiple mol-
ecules, including (1) surface-expressed thrombomodulin, (2) heparin-like molecules, (3)
protein S, and (4) tissue-factor pathway inhibitor (TFPI) (17).
Thrombomodulin is a cell surface protein that is present in all endothelial cells, and,
by competing with catalytic activity of thrombin, inhibits the procoagulant effects of
thrombin. Furthermore, thrombomodulin binds factor Xa and inhibits subsequent activa-
tion of prothrombin. Taken together, thrombomodulin inhibits coagulation at multiple
regulatory steps in the coagulation cascade.
The expression of heparin-like proteoglycans on the surface of the endothelium pro-
vides another endothelial-generated anticoagulant mechanism. Heparin-like molecules
on endothelium serves as a cofactor for the anticoagulant antithrombin III (ATIII), thereby
increasing its activity and inhibition of the coagulation factors thrombin, factor Xa, IXa,
and XIIa.
Protein S is a vitamin K-dependent glycoprotein that is a cofactor for activated protein
C, thereby enhancing inactivation of factor Va and inhibiting coagulation. Protein S is
synthesized by the vascular endothelium, in addition to the liver and megakaryocytes.
An additional mechanism whereby the endothelium inhibits coagulation is through
tissue factor pathway inhibitor (TFPI), which antagonizes the factor VIIa/tissue factor
complex, the main initiator of the extrinsic coagulation cascade. It is important to note
that, unlike other extravascular cells (e.g., mononuclear phagocytes), endothelial cells do
not normally express tissue factor, the primary trigger of the coagulation cascade. Upon
exposure to various stresses, tissue factor is expressed on the surface of endothelial cells,
thereby promoting coagulation (18–22). TFPI is constitutively expressed by endothelial
cells and is present in low concentrations in the circulation. As mentioned, TFPI inhibits
factor VIIa/tissue factor, and does so much more efficiently when TFPI is complexed
with factor Xa (23).
Indeed, in addition to the regulation of platelet activity and the coagulation system, the
vascular endothelium also modulates the fibrinolytic cascade. The vascular endothelium
is the primary source of tissue-type plasminogen activator (tPA), required for initiation
of fibrinolysis. When bound to fibrin, tPA efficiently generates plasmin from its precur-
sor, plasminogen, effectively localizing plasmin formation to sites of clot. In addition to
constitutive expression and release, tPA is synthesized and released in response to throm-
bin, shear stress, vasopressin, acidosis, and hypoxia (13,14,24). The endothelium also
synthesizes the major inhibitor of tPA, plasminogen activator inhibitor-1 (PAI-1), which
is usually present in significant excess relative to tPA. Like other procoagulant mol-
ecules, PAI-1 synthesis is increased in response to inflammation, hormones, and
cytokines, while tPA levels change very little. In clinical situations where cytokine
release may be significant (e.g., sepsis, SIRS following coronary artery bypass [CAB +
CPB]), it has been suggested that this disparate activation of PAI-1, compared to tPA,
may provide a biochemical basis toward an early procoagulant state (25,26).
In summary, the vascular endothelium plays an active role in resisting adherence and
activation of platelets, and inhibiting activation of the coagulation cascade, resulting in
a thromboresistant barrier between the blood and tissues. To these ends, there is a com-
plex interplay between a variety of endothelial-derived antiplatelet factors, anticoagulant
mediators, and modulators of fibrinolysis. However, the quiescent thromboresistant
phenotype of the endothelium can rapidly change upon exposure to cellular stresses
including hypoxia, mechanical trauma, and proinflammatory cytokines, among others.
Many of these cellular stresses are integral to cardiac surgery with CPB, leading to
profound endothelial cell activation.
There is a vast literature, including recent reviews, describing the effects of coronary
revascularization with cardiopulmonary bypass (CPB) on the coagulation and fibrin-
olytic cascades, as well as on platelet function (27–29). Abnormalities of these systems
may contribute to significant morbidity and mortality following cardiovascular opera-
tions with CPB, including postoperative hemorrhage and tendency for thromboembolic
complications. The biochemical alterations are complex and likely result from simulta-
neous activation of coagulation, fibrinolysis, and platelets. Following CPB, thrombin is
generated in significant amounts, as measured by prothrombin fragments 1+2 (F1 + 2)
and by thrombin–antithrombin complex (TAT) levels (30). Although initially thought to
be produced primarily by activation of the intrinsic coagulation cascade, when factor XII
becomes activated after contacting the CPB circuit, recent evidence suggests that coagu-
lation is initiated via the tissue factor pathway of the extrinsic coagulation cascade (31,32).
Activated vascular endothelium is likely a significant modulator of these alterations,
given its upregulation of tissue factor and TFPI in response to proinflammatory mediators
(e.g., C5a) known to be increased after CAB with CPB (33). Platelet activation also
contributes to the hemostatic disturbances following CAB with CPB, resulting, at least
in part, from endothelial cell activation.
There are both qualitative and quantitative platelet defects following cardiac surgery
with CPB, with various causes, including endothelial-derived substances (27). The vas-
cular endothelium responds to humoral mediators elaborated during CAB with CPB,
such as thrombin, by releasing von Willebrand factor (vWF), P-selectin, and other
proinflammatory mediators, such as IL-8 (34). The net effect of these endothelial-derived
mediators is to activate and aggregate platelets at sites of tissue injury. If this process is
widespread in multiple vascular beds, such as following CAB with CPB, then thromb-
ocytopenia may result and the remaining platelets are qualitatively abnormal, as demon-
strated by prolonged bleeding time in patients after CPB (35). The observation that von
Willebrand factor, as a marker of endothelial cell activation, increases after cardiac
surgery with CPB provides evidence of in vivo endothelial cell activation and injury (36).
Finally, activation of the fibrinolytic cascade, further disturbing the coagulation
abnormalitites, has been documented in numerous studies following CPB (37,38). More-
over, it has been demonstrated that following cold cardioplegia administration for CPB,
the coronary circulation releases tPA (39), thus supporting the notion of endothelial cell
activation and initiation of fibrinolysis.
Taken together, these data provide compelling evidence for a critical role of the vas-
cular endothelium in modulating platelets, coagulation, and fibrinolysis in cardiovascu-
lar surgery. Although these abnormalities have been well documented with the use of
CPB, much less is known about these systems when CPB is eliminated in OPCAB.
Casati and colleagues assessed changes in biochemical markers of coagulation, fibrin-
olysis, and platelets in patients undergoing OPCAB and compared them to a case-matched
cohort undergoing CAB with CPB (40). To account for the reduced hemodilution in
OPCAB patients, they adjusted the biochemical results for hemodilution. In both groups,
there was a comparable consumption of prothrombin and fibrinogen, consistent with
activation of the coagulation cascade via the tissue factor/VII pathway, independent of
the CPB circuit and contact activation of factor XII. When CPB was used, there was a
transient reduction in platelet counts, with plasminogen activation, and increased
D-dimer formation (reflecting fibrinolysis)—an effect that was not seen in the OPCAB
group. By 24 h postoperatively, the hemostatic profiles were similar between groups.
Although there is a paucity of data examining the coagulation disturbances following
OPCAB, multiple studies have demonstrated reduced bleeding and transfusion require-
ments when CPB is eliminated (41,42). Additional studies are warranted to better char-
acterize the benefits of OPCAB with respect to disturbances of coagulation.
Cell–Cell Interactions
Under normal physiological conditions, inflammatory cells in the bloodstream do not
adhere to the endothelium. Endothelial cell activation, however, can result in the expres-
sion of a variety of inflammatory mediators and cell adhesion molecules. These adhesion
molecules, including the selectins, the integrins, and the immunoglobulin supergene
family (e.g., ICAM-1), lead to interactions of neutrophils, monocytes, platelets, and
lymphocytes with the endothelium. Following adherence, these cells are able to invade
the internal elastic lamina and release a variety of substances that act on the cellular milieu
and promote chemotaxis, thereby defining the vessel’s response to injury.
Monocytes and platelets have classically been the two cell types most intimately
linked with the regulation of the inflammatory response. The monocyte is able to trans-
migrate across vessel walls and transform into tissue macrophages or foam cells. In
addition to the elaboration of inflammatory mediators, these cells are able to ingest lipids
and scavenge oxidized low-density lipoproteins, producing free radicals in the process.
Platelets, via elaboration of platelet-derived growth factor, stimulate vascular smooth
muscle cell proliferation after adhering to collagen in the basement membrane.
While the various cell types involved in inflammation are important, an understanding
of the mediators produced by these cells is also critical. A complex network of cytokines
and growth factors regulates the inflammatory state via autocrine and paracrine signal-
ing. Classically, these mediators are divided into growth factors, chemokines, and pro-
or anti-inflammatory factors. However, most factors have multiple roles that are influ-
enced by the setting in which they are expressed. The redundancy and complexity of this
system of inflammatory mediation makes it challenging to target any single component
this system to improve clinical endpoints.
A final mediator of the inflammatory response that is critical in the setting of endot-
helial damage is nitric oxide (NO). With endothelial damage, there are decreases in NO,
prostacyclin, and adenosine—all of which have been shown to be potent antiadhesive

mediators. It has therefore been suggested that NO or NO-donor therapy may be benefi-
cial in protecting the myocardium from endothelial injury by inhibiting neutrophil adher-
ence and accumulation (43).
ACUTE ENDOTHELIAL INJURY

While the whole-body inflammatory response to CPB is largely avoided with off-pump
techniques (44), several mechanisms by which endothelial cell damage can occur remain.
Furthermore, despite the attenuation of systemic inflammation, OPCAB can lead to regional
myocardial injury with localized effects, such as endothelial dysfunction or myocardial
stunning (9,45). The endothelial stress response during OPCAB is likely regional, rather
than widespread as seen in after CAB with CPB (29), and results from the novel techniques
employed during OPCAB including target vessel snaring, intraluminal shunts, and high-
frequency gas insufflation (1,46,47). Loosely categorized, these etiologies may be divided
into mechanical trauma of the vasculature and ischemia reperfusion injury.
Mechanical Trauma
The most obvious source of damage in minimally invasive cardiac surgery comes from
the temporary disruption of coronary blood flow during completion of the distal anasto-
mosis. A variety of clamps and stabilization devices have been employed with the end
result of stabilization of the target coronary artery in a hemostatic field. The full effects
of these types of occlusion have been shown both in animal models and, more recently,
in human studies (47–52). However, the evidence regarding the acute impact of operative
control of the coronary arteries on endothelial function is somewhat conflicting, as other
reports have shown no endothelial dysfunction with ischemic periods up to 1 h in a swine
model (51). Finally, it has been shown that the type of technique used (i.e., bulldog
occluders, prolene sutures, vascular snares) has an impact on the degree of endothelial
disruption present (50,52). Local microthrombus (50) or acute vasospasm (53) may then
result from this endothelial disruption.
High-flow gas insufflation is another technique used to maintain coronary arteriotomy
visualization during minimally invasive techniques for coronary revascularization. The
effects of this insufflation on the coronary vascular endothelium have not been fully
elucidated. However, it has been suggested that gas insufflation results in almost com-
plete denudation of the exposed endothelium (46,54). Humidification of the insufflated
gas may attenuate but not alleviate this type of endothelial injury (46).
Ischemia/Reperfusion
Minimally invasive techniques, which avoid CPB and cardiac arrest, may not cause
global myocardial ischemia, but local myocardial ischemia does still occur. Moreover,
patients with coronary artery disease often suffer from myocardial ischemia at baseline,
and the main goal in correction of this problem is the restoration of oxygenated blood to
the ischemic myocardium. Interventional procedures in the cardiac catheterization lab as
well as surgical procedures, by necessity, incur a period of superimposed complete
ischemia on a background of chronic low-grade ischemia. Reperfusion of this ischemic
myocardium results in further but necessary myocardial injury that can be either revers-
ible or irreversible. The key determinant of irreversible damage is the length of the
ischemic time, with 20 min accepted as the maximal length for reversible defects. Beyond
20 min, myocardial necrosis is inevitable, with concomitant endothelial cell damage.

Elevation of lactate levels indicating anaerobic metabolism occur within 10 min of coro-
nary occlusion in both animals and humans (55). The role of apoptosis, or programmed
cell death, in the setting of acute myocardial ischemia is less clear (56).
Although the exact mechanism of ischemia/reperfusion injury is not known, a simpli-
fied paradigm suggests that oxidative stress initiates numerous cell-signaling cascades
that result in transcriptional activation of multiple stress-responsive gene products, which,
in total, result in tissue injury (57,58). In addition, with a shift to anaerobic metabolism,
endothelial hypoxic stress induces transcriptional activation of multiple stress-response
gene products (59). Although this stress response likely involves multiple transcriptional
regulators, NF-κB has received the most attention and is the best characterized (60,61).
NF-κB, a redox-sensitive transcription factor, exists in a latent form in the cytoplasm
and upon activation translocates to the nucleus with binding of the protein to DNA in
target genes, thus promoting transcription and translation (62). This molecule has been
shown—through deletional and mutational analyses—to be necessary for activation of
many of the genes involved in ischemia/reperfusion injury including IL-1, IL-8,
E-selectin, VCAM-1, ICAM-1, MCP-1, inducible nitric oxide synthase, and tissue factor
(61). This transcriptional activation in turn leads to activation of endothelial and other
cells, including platelets and leukocytes. Taken together, injury is potentiated upon
reperfusion with oxygenated blood at the completion of coronary revascularization (63).
In particular, the phenomenon of neutrophil adherence to the endothelium following
ischemia/reperfusion is mediated via a variety of substances including the cytokine
IL-1 (64) and the cell adhesion markers E-selectin and P-selectin (65,66).
While localized ischemia would seem to be less deleterious than global ischemia from
CPB, there are studies that show that the endothelial response is similar. Indeed, in
patients undergoing OPCAB there is a comparable increase in serum vascular endothelial
growth factor (VEGF) compared to those undergoing CAB with CPB (67). Recognizing
that vascular endothelium responds to hypoxic stress with transcriptional activation of
myriad genes, including VEGF (58), these data imply that OPCAB may cause a profound
response of the vascular endothelium despite elimination of cardioplegia and CPB.
The increased use of minimally invasive techniques of cardiac surgery has led to
interest in techniques to avoid or reduce the amount of ischemia/reperfusion injury.
Emphasis has been placed on ischemic preconditioning (IP) of the myocardium as a
mechanism of inducing cardioprotection against ischemia. Ischemic preconditioning of
the myocardium is a phenomenon whereby brief periods of sublethal ischemia protect the
heart against subsequent prolonged ischemia (68). There are two phases, or windows, of
protection against ischemia—a classic, or early, window of protection that begins within
minutes of the preconditioning stimulus and lasts up to 4 h, which is followed by a late,
or second, window of protection that ensues 24 h after preconditioning, and lasts up to
4 d (69). In contrast to the early phase of protection, which is effective against both
necrotic and apoptotic cell death, the second window of protection is also effective
against myocardial stunning (70). As a result of its longer duration of action and its
efficacy against myocardial stunning, the second window of protection may ultimately
have greater translational potential in cardiac surgery. While there is some evidence that
IP may reduce postoperative myocardial dysfunction in patients undergoing OPCAB, the
clinical utility of either phase of IP is not known at this time (71). Other techniques that
may be useful in protecting against myocardial ischemia/reperfusion injury include
hypothermia (72,73) and the use of aprotinin (74).
CHRONIC ENDOTHELIAL INJURY

While the acute damage to the endothelium is somewhat different with minimally
invasive techniques, the chronic endothelial response to injury is quite similar and may
limit the long-term patency of bypass grafts. Both the development of intimal hyperplasia
at the sites of vessel injury and the progression of underlying atherosclerotic disease lead
to reductions in coronary blood flow over time, which may lead to the need for further
procedures (47,75). In support, Hangler et al. examined CAB grafts in patients undergo-
ing cardiac transplantation who had previously been revascularized with OPCAB (50).
Examination of coronary target vessels occluded with snares revealed focal endothelial
denudation, local microthrombosis, atherosclerotic plaque rupture, and injury to target-
vessel side branches. These data support a chronic endothelial response following OPCAB
that may affect long-term graft patency and clinical course. To assess this postulate,
Gundry et al. compared long-term (7-yr) intervention-free outcome in patients subjected
to OPCAB vs a case-matched cohort subjected to CAB with CPB (76). In this study, twice
as many patients revascularized with OPCAB (30% vs 16%) required repeat catheteriza-
tion during the observation period, and nearly three times as many patients in the OPCAB
group required a second intervention (20% vs 7%). While it has been suggested that use
of newer stabilizing devices and occlusion techniques may attenuate endothelial damage
with improved long-term patency rates following OPCAB, there are limited data to
make this determination (47,77). Moreover, recognizing that OPCAB results in compa-
rable levels of endothelial-derived vascular mitogens such as VEGF (67) that are thought
to be important mediators in the vascular response to injury, additional investigation is
warranted into the endothelial effects of OPCAB surgical techniques, with an emphasis
on long-term graft patency and clinical outcome.
Intimal Hyperplasia
During the course of cardiovascular revascularization, there is damage to both the
conduit being harvested and the native vessel, which results in fibroproliferation and the
formation of a neointima (78). The intimal response to arterial injury can be divided into
three phases. Beginning within 24 h of injury, there is vascular smooth muscle prolifera-
tion. With denudation of the endothelium, platelets become adherent to the vessel wall
and become activated. These activated platelets are responsible for the release of a variety
of substances that stimulate smooth muscle division and migration into the intima. Smooth
muscle proliferation is also stimulated in the media by substances such as fibroblast
growth factor. Over the period from 1 to 2 wk after injury, there is a gradual relocation
of smooth muscle cells from the media to the intima with formation of a neointima. This
neointima then expands via further smooth muscle cell proliferation, which can eventu-
ally result in luminal obstruction.
Clearly, the vascular smooth muscle cell is the central cell type involved in neointimal
hyperplasia. These cells are critical in modulating vascular tone with their contractile
properties and are also the principal manufacturers of the extracellular matrix, including
proteoglycans and collagen. Regulation of smooth muscle cell proliferation is quite
variable, however, and is subject to fluctuations related to interactions with inflammatory
mediators such as tumor necrosis factor and interferon (78).
One interesting factor that influences the degree of neointimal hyperplasia is the
choice of bypass conduit, which has important implications for postoperative vascular
tone. The patency of mammary grafts is clearly superior to that of vein grafts. This
difference is likely secondary to differences in endothelial function. Mammary grafts are

sensitive to relaxation from acetylcholine, bradykinin, thrombin, and adenosine diphos-
phate. Vein grafts are not sensitive to thrombin or adenosine diphosphate and only weakly
sensitive to acetylcholine. In fact, vein grafts exhibit a converse effect from thrombin or
adenosine diphosphate from platelets, including contraction and release of growth fac-
tors, which stimulate smooth muscle cell proliferation. Also, mammary grafts produce
more NO and prostacyclin than vein grafts (4).
Atherosclerosis
Since the early 1970s it has been recognized that atherosclerosis results from an
endothelial response to chronic injury characterized by neutrophil, lymphocyte, platelet,
and macrophage adhesion and migration into the subendothelium (79). Multiple risk
factors exist in cardiac surgical patients that predispose them to this chronic type of
endothelial injury, including nicotine abuse, hypertension, diabetes, and hypercholesterolemia.
The underlying pathophysiology of chronic endothelial injury and atherosclerosis is
complex and not completely defined. Chronic injury leads to disruption of the normal
endothelial monolayer via a loss of endothelial cell ability to replicate. This, in turn,
exposes the subendothelium to a variety of activated inflammatory cells including acti-
vated monocytes, platelets, and T-lymphocytes, which stimulate smooth muscle cell
proliferation and coverage of the denuded area (80,81) via the elaboration of multiple
growth factors. These growth factors in combination with locally expressed cytokines
stimulate the smooth muscle cells to allow for deposition of extracellular matrix with the
establishment of a fibrous plaque.
One of the key substances implicated in the progression of atherosclerosis is NO.
Dysregulation of NO has been shown to result in an impairment of endothelium-depen-
dent relaxation that causes vascular spasm with increased shear stress, increased smooth
muscle cell proliferation from the loss of NO antimitogenic effects, increased platelet
adherence, and even thrombosis (82).
Endothelin has also been demonstrated to play a key role in atherosclerotic vascular
disease. Low-density lipoproteins stimulate increased endothelin production from the
endothelium via increased gene transcription. Also, there is a down-regulation of endothelin
receptor expression and increased production of endothelin from vascular smooth muscle
cells. As a result, endothelin levels correlate with the extent of atherosclerosis present (83).
While the underlying pathophysiology of atherosclerosis is less clearly defined, the
natural history and progression of atherosclerosis are quite well established. Beginning
as lipid deposits in the intima, there is further accumulation of intracellular lipids in foam
cells. Following this, there is the gradual buildup of extracellular pools of lipid that
eventually becomes clinically evident and leads to distortion of blood vessel architecture.
Further smooth muscle cell proliferation and collagen deposition leads to greater surface
defects from the fibroatheroma, which can result in vessel occlusion from either hemor-
rhage or thrombosis (82). The end clinical result of this plaque rupture can be acute
myocardial infarction and sudden death (84).
CONCLUSION
The endothelium plays a central role in cardiovascular function. It helps to regulate
vasomotor function, resists intravascular thrombosis by maintaining a state of anticoagu-
lation, and works to prevent the activation of a variety of intravascular inflammatory
cells. Cardiovascular surgery, however, results in damage of the endothelium that

impairs normal function and alters homeostasis, with the balance tipped toward tissue
injury. A recent trend toward minimally invasive techniques of cardiovascular
revascularization has led to the avoidance of endothelial damage from extracorporeal
circulation. However, elimination of CPB has not abrogated the problem of endothelial
damage. Indeed, minimally invasive techniques cause acute endothelial injury by
employing new techniques such as vessel snaring, intraluminal shunts, and trauma from
high-flow gas insufflation. Certainly, these new techniques have not mitigated the chronic
response to endothelial injury, intimal hyperplasia, or the progression of atheroscle-
rosis.
As techniques of minimally invasive surgery for cardiac revascularization evolve, it
will be increasingly important to further delineate the endothelial response to injury and
to employ techniques that minimize this trauma and maximize endothelial preservation.
A thorough understanding of endothelial biology will enable the cardiovascular surgeon
to greatly improve the long-term success of cardiac revascularization.
REFERENCES
1. Perrault LP, Menasche P, Wassef M, et al. Endothelial effects of hemostatic devices for continuous
cardioplegia or minimally invasive operations. Ann Thorac Surg 1996;62(4):1158–1163.
2. Chavanon O, Perrault LP, Menasche P, Carrier M, Vanhoutte PM. As originally published in 1996:
Endothelial effects of hemostatic devices for continuous cardioplegia or minimally invasive operations.
Updated in 1999. Ann Thorac Surg 1999;68(3):1118–1120.
3. Verrier ED, Boyle EMJ. Endothelial cell injury in cardiovascular surgery: an overview. Ann Thorac
Surg 1997;64(4):S2–S8.
4. Ruschitzka FT, Noll G, Luscher TF. The endothelium in coronary artery disease. Cardiology
1997;88(suppl 3):3–19.
5. Vita JA, Treasure CB, Nabel EG, et al. Coronary vasomotor response to acetylcholine relates to risk
factors for coronary artery disease. Circulation 1990;81(2):491–497.
6. Ludmer PL, Jelwyn AP, Shook TL, et al. Paradoxical vasoconstriction induced by acetylcholine in
atherosclerotic coronary arteries. N Engl J Med 1986;315(17):1046–1051.
7. Pearson PJ, Evora PR, Schaff HV. Bioassay of EDRF from internal mammary arteries: implications for
early and late bypass graft patency. Ann Thorac Surg 1992;54(6):1078–1084.
8. Nonami Y. The role of nitric oxide in cardiac surgery. Surg Today 1997;27(7):583–592.
9. Lockowandt U, Franco-Cereceda A. Off-pump coronary bypass surgery causes less immediate postop-
erative coronary endothelial dysfunction compared to on-pump coronary bypass surgery. Eur J
Cardiothorac Surg 2001;20(6):1147–1151.
10. Angdin M, Settergren G, Vaage J. Better preserved pulmonary endothelium-dependent vasodilation
with off- pump coronary surgery. Scand Cardiovasc J 2001;35(4):264–269.
11. Lockowandt U, Liska J, Franco-Cereceda A. Short ischemia causes endothelial dysfunction in porcine
coronary vessels in an in vivo model. Ann Thorac Surg 2001;71(1):265–269.
12. Angdin M, Settergren G. Acetylcholine reactivity in the pulmonary artery during cardiac surgery in
patients with ischemic or valvular heart disease. J Cardiothorac Vasc Anesth 1997;11(4):458–462.
13. Pearson JD. Endothelial cell function and thrombosis. Baillieres Best Pract Res Clin Haematol
1999;12(3):329–341.
14. Wu KK, Thiagarajan P. Role of endothelium in thrombosis and hemostasis. Annu Rev Med
1996;47:315–331.
15. Radomski MW, Palmer RM, Moncada S. Comparative pharmacology of endothelium-derived relaxing
factor, nitric oxide and prostacyclin in platelets. Br J Pharmacol 1987;92(1):181–187.
16. Marcus AJ, Safier LB. Thromboregulation: multicellular modulation of platelet reactivity in hemostasis
and thrombosis. FASEB J 1993;7(6):516–522.
17. Wu C. Heat shock transcription factors: structure and regulation. Annu Rev Cell Dev Biol 1995;11:441–469.
18. Ogawa S, Gerlach H, Esposito C, Pasagian-Macaulay A, Brett J, Stern D. Hypoxia modulates the barrier
and coagulant function of cultured bovine endothelium. Increased monolayer permeability and induc-
tion of procoagulant properties. J Clin Invest 1990;85(4):1090–1098.
19. Camera M, Giesen PL, Fallon J, et al. Cooperation between VEGF and TNF-alpha is necessary for
exposure of active tissue factor on the surface of human endothelial cells. Arterioscler Thromb Vasc Biol
1999;19(3):531–537.
20. Lyberg T, Galdal KS, Evensen SA, Prydz H, et al. Cellular cooperation in endothelial cell thromboplas-
tin synthesis. Br J Haematol 1983;53(1):85–95.
21. Bevilacqua MP, Pober JS, Majeau GR, Cotran RS, Gimbrone MA. Interleukin 1 (IL-1) induces biosyn-
thesis and cell surface expression of procoagulant activity in human vascular endothelial cells. J Exp
Med 1984;160(2):618–623.
22. Brox JH, Osterud B, Bjorklid E, Fenton JW 2nd. Production and availability of thromboplastin in
endothelial cells: the effects of thrombin, endotoxin and platelets. Br J Haematol 1984;57(2):239–246.
23. Bajaj MS, Birktoft JJ, Steer SA, Bajaj SP. Structure and biology of tissue factor pathway inhibitor.
Thromb Haemost 2001;86(4):959–972.
24. Emeis JJ, vanden Eijnden-Schrauwen Y, vanden Hoogen CM, de Priester W, Westmuckett A, Lupu F.
An endothelial storage granule for tissue-type plasminogen activator. J Cell Biol 1997;139(1):245–256.
25. Colucci M, Paramo JA, Collen D. Generation in plasma of a fast-acting inhibitor of plasminogen
activator in response to endotoxin stimulation. J Clin Invest 1985;75(3):818–824.
26. van der Poll T, Levi M, Buller HR, et al. Fibrinolytic response to tumor necrosis factor in healthy
subjects. J Exp Med 1991;174(3):729–732.
27. Weerasinghe A, Taylor KM. The platelet in cardiopulmonary bypass. Ann Thorac Surg
1998;66(6):2145–2152.
28. Raymond PD, Marsh NA. Alterations to haemostasis following cardiopulmonary bypass and the rela-
tionship of these changes to neurocognitive morbidity. Blood Coagul Fibrinolysis 2001;12(8):601–618.
29. Boyle EM Jr, Verrier ED, Spiess BD. Endothelial cell injury in cardiovascular surgery: the procoagulant
response. Ann Thorac Surg 1996;62(5):1549–1557.
30. Boisclair MD, Lane DA, Philippou H, Sheikh S, Hunt B. Thrombin production, inactivation and expres-
sion during open heart surgery measured by assays for activation fragments including a new ELISA for
prothrombin fragment F1 + 2. Thromb Haemost 1993;70(2):253–258.
31. Boisclair MD, Lane DA, Philippou H, et al. Mechanisms of thrombin generation during surgery and
cardiopulmonary bypass. Blood 1993;82(11):3350–3357.
32. Philippou H, Adami A, Boiselair MD, Lane DA. An ELISA for factor X activation peptide: application
to the investigation of thrombogenesis in cardiopulmonary bypass. Br J Haematol 1995;90(2):432–437.
33. Ikeda K, Nagasawa K, Horiuchi T, Tsuru T, Nishizaka H, Niho Y. C5a induces tissue factor activity on
endothelial cells. Thromb Haemost 1997;77(2):394–398.
34. van Mourik JA, Romani De Wit T, Voorberg J. Biogenesis and exocytosis of Weibel-Palade bodies.
Histochem Cell Biol 2002;117(2):113–122.
35. Harker LA, Malpass TW, Branson HE, Hessel EA 2nd, Slichter SJ. Mechanism of abnormal bleeding
in patients undergoing cardiopulmonary bypass: acquired transient platelet dysfunction associated with
selective alpha-granule release. Blood 1980;56(5):824–834.
36. Holdright DR, Hunt BJ, Parratt R, et al. The effects of cardiopulmonary bypass on systemic and coronary
levels of von Willebrand factor. Eur J Cardiothorac Surg 1995;9(1):18–21.
37. Hunt BJ, Parratt RN, Segal HC, Sheikh S, Kallis P, Yacoub M. Activation of coagulation and fibrinolysis
during cardiothoracic operations. Ann Thorac Surg 1998;65(3):712–718.
38. Tabuchi N, de Haan J, Boonstra PW, van Oeveren W. Activation of fibrinolysis in the pericardial cavity
during cardiopulmonary bypass. J Thorac Cardiovasc Surg 1993;106(5):828–833.
39. Valen G, Eriksson E, Risberg B, Vaage J. Fibrinolysis during cardiac surgery. Release of tissue plasmi-
nogen activator in arterial and coronary sinus blood. Eur J Cardiothorac Surg 1994;8(6):324–330.
40. Casati V, Gerli C, Franco A, et al. Activation of coagulation and fibrinolysis during coronary surgery:
on-pump versus off-pump techniques. Anesthesiology 2001;95(5):1103–1109.
41. Nader ND, Khadra WZ, Reich NJ, Bacon DR, Salerno TA, Panos AL. Blood product use in cardiac
revascularization: comparison of on- and off-pump techniques. Ann Thorac Surg 1999;68(5):1640–1643.
42. Ascione R, Williams S, Lloyd CT, Sundaramoorthi T, Pitsis AA, Angelini GD. Reduced postoperative
blood loss and transfusion requirement after beating-heart coronary operations: a prospective random-
ized study. J Thorac Cardiovasc Surg 2001;121(4):689–696.
43. Vinten-Johansen J, Sato H, Zhao ZQ. The role of nitric oxide and NO-donor agents in myocardial
protection from surgical ischemic-reperfusion injury. Int J Cardiol 1995;50(3):273–281.
44. Ascione R, Lloyd CT, Underwood MJ, Lotto AA, Pitsis AA, Angelini GD. Inflammatory response
after coronary revascularization with or without cardiopulmonary bypass. Ann Thorac Surg
2000;69(4):1198–1204.
45. Grubitzsch H, Ansorge K, Wollert GH, Eckel L. Stunned myocardium after off-pump coronary artery
bypass grafting. Ann Thorac Surg 2001;71(1):352–355.
46. Okazaki Y, Takarabe K, Murayama J, et al. Coronary endothelial damage during off-pump CABG
related to coronary-clamping and gas insufflation. Eur J Cardiothorac Surg 2001;19(6):834–839.
47. Perrault LP, Nickner C, Desjardins N, Carrier M. Effects on coronary endothelial function of the Cohn
stabilizer for beating heart bypass operations. Ann Thorac Surg 2000;70(3):1111–1114.
48. Walia AS, Kole SD. Clamp for coronary artery operations. Ann Thorac Surg 1998;65(5):1475–1476.
49. Bandyopadhyay A, Kapoor L, Gan M. Bulldog with spikes: clamp for coronary artery operations. Ann
Thorac Surg 1999;67(2):594–595.
50. Hangler HB, Pfaller K, Antretter H, Dapunt OE, Bonatti JO. Coronary endothelial injury after local
occlusion on the human beating heart. Ann Thorac Surg 2001;71(1):122–127.
51. Perrault LP, Menasche P, Biolouard JP, et al. Snaring of the target vessel in less invasive bypass
operations does not cause endothelial dysfunction. Ann Thorac Surg 1997;63(3):751–755.
52. Sokullu O, Karabulut H, Gercekoglu H, et al. Coronary artery stabilization causes endothelial damage:
an electron microscopic study on dogs. Cardiovasc Surg 2001;9(4):407–410.
53. Fonger JD, Yang XM, Cohen RA, Haudenschild CC, Shemin RJ. Human mammary artery endothelial
sparing with fibrous jaw clamping. Ann Thorac Surg 1995;60(3):551–555.
54. Burfeind WR Jr, Duhaylongsod FG, Annex BH, Samuelson D. High-flow gas insufflation to facilitate
MIDCABG: effects on coronary endothelium. Ann Thorac Surg 1998;66(4):1246–1249.
55. Hall RI, O’Regan N, Gardner M. Detection of intraoperative myocardial ischaemia—a comparison
among electrocardiographic, myocardial metabolic, and haemodynamic measurements in patients with
reduced ventricular function. Can J Anaesth 1995;42(6):487–494.
56. Elsasser A, Suzuki K, Lorenz-Meyer S, Bode C, Schaper J. The role of apoptosis in myocardial ischemia:
a critical appraisal. Basic Res Cardiol 2001;96(3):219–226.
57. Boyle EM Jr, Canty TG Jr, Morgan EN, Yun W, Pohlman TH, Varrier ED. Treating myocardial
ischemia-reperfusion injury by targeting endothelial cell transcription. Ann Thorac Surg
1999;68(5):1949–1953.
58. Faller DV. Endothelial cell responses to hypoxic stress. Clin Exp Pharmacol Physiol 1999;26(1):74–84.
59. Pohlman TH, Harlan JM. Adaptive responses of the endothelium to stress. J Surg Res 2000;89(1):85–119.
60. Parry GCN, Mackman N. NF-kB mediated transcription in human monocytic cells and endothelial cells.
Trends Cardiovasc Med 1998;8:138–142.
61. De Martin R, Hoeth M, Hofer-Warbinek R, Schmid JA. The transcription factor NF-kappa B and the
regulation of vascular cell function. Arterioscler Thromb Vasc Biol 2000;20(11):E83–E88.
62. Collins T, Read MA, Neish AS, Whitley MZ, Thanos D, Maniatis T. Transcriptional regulation of
endothelial cell adhesion molecules: NF-kappa B and cytokine-inducible enhancers. FASEB J
1995;9(10):899–909.
63. Johnson M, Pohlman TH, Verrier ED. Neutrophil antiadhesion therapy for myocardial ischemia: clinical
potential. Clin Immunother 1995;3:8–14.
64. Shreeniwas R, Koga S, Karakurum M, et al. Hypoxia-mediated induction of endothelial cell interleukin-
1 alpha. An autocrine mechanism promoting expression of leukocyte adhesion molecules on the vessel
surface. J Clin Invest 1992;90(6):2333–2339.
65. Shen I, Verrier ED. Expression of E-selectin on coronary endothelium after myocardial ischemia and
reperfusion. J Card Surg 1994;9(3 suppl):437–441.
66. Winn RK, Liggitt P, Vedder NB, Paulson JC, Harlan JM. Anti-P-selectin monoclonal antibody attenu-
ates reperfusion injury to the rabbit ear. J Clin Invest 1993;92(4):2042–2047.
67. Burton PB, Owen VJ, Hafizi S, et al. Vascular endothelial growth factor release following coronary
artery bypass surgery: extracorporeal circulation versus “beating heart” surgery. Eur Heart J
2000;21(20):1708–1713.
68. Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in
ischemic myocardium. Circulation 1986;74(5):1124–1136.
69. Baxter GF, Goma FM, Yellon DM. Characterisation of the infarct-limiting effect of delayed precon-
ditioning: timecourse and dose-dependency studies in rabbit myocardium. Basic Res Cardiol
1997;92(3):159–167.
70. Sun JZ, Tang XL, Knowlton AA, Park SW, Qiu Y, Bolli R. Late preconditioning against myocardial
stunning. An endogenous protective mechanism that confers resistance to postischemic dysfunction
24 h after brief ischemia in conscious pigs. J Clin Invest 1995;95(1):388–403.
71. Laurikka J, Wu ZK, Iusalo P, et al. Regional ischemic preconditioning enhances myocardial perfor-
mance in off-pump coronary artery bypass grafting. Chest 2002;121(4):1183–1189.
72. Hale SL, Kloner RA. Myocardial hypothermia: a potential therapeutic technique for acute regional
myocardial ischemia. J Cardiovasc Electrophysiol 1999;10(3):405–413.
73. Dave RH, Hale SL, Kloner RA. Hypothermic, closed circuit pericardioperfusion: a potential
cardioprotective technique in acute regional ischemia. J Am Coll Cardiol 1998;31(7):1667–1671.
74. Hendrikx M, Rega F, Jamaer L, Valkenborgh T, Gutermann H, Mecs U. Na(+)/H(+)-exchange inhibition
and aprotinin administration: promising tools for myocardial protection during minimally invasive
CABG. Eur J Cardiothorac Surg 2001;19(5):633–639.
75. Allaire E, Clowes AW. Endothelial cell injury in cardiovascular surgery: the intimal hyperplastic
response. Ann Thorac Surg 1997;63(2):582–591.
76. Gundry SR, Romano MA, Shattuck OH, Razzouk AJ, Bailey LL. Seven-year follow-up of coronary
artery bypasses performed with and without cardiopulmonary bypass. J Thorac Cardiovasc Surg
1998;115(6):1273–1277; discussion 1277–1278.
77. Jansen EW, Borst C, Lahpor JR, et al. Coronary artery bypass grafting without cardiopulmonary bypass
using the octopus method: results in the first one hundred patients. J Thorac Cardiovasc Surg
1998;116(1):60–67.
78. Schwartz SM, deBlois D, O’Brien ER. The intima. Soil for atherosclerosis and restenosis. Circ Res
1995;77(3):445–465.
79. Ross R, Glomset JA. Atherosclerosis and the arterial smooth muscle cell: proliferation of smooth muscle
is a key event in the genesis of the lesions of atherosclerosis. Science 1973;180(93):1332–1339.
80. Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature 1993;362(6423):801–809.
81. Ross R. Cell biology of atherosclerosis. Annu Rev Physiol 1995;57:791–804.
82. Billiar TR. Nitric oxide. Novel biology with clinical relevance. Ann Surg 1995;221(4):339–349.
83. Lerman A, Edwards BS, Hallett JW, Heublein DM, Sandberg SM, Burnett JC Jr. Circulating and tissue
endothelin immunoreactivity in advanced atherosclerosis. N Engl J Med 1991;325(14):997–1001.
84. Selzman CH, Miller SA, Harken AH. Therapeutic implications of inflammation in atherosclerotic
cardiovascular disease. Ann Thorac Surg 2001;71(6):2066–2074.
Chapter 3 / Minimally Invasive Bypass Grafting 43
II MINIMALLY INVASIVE CORONARY

BYPASS GRAFTING
3 Minimally Invasive Bypass Grafting

A Historical Perspective
Enio Buffolo, MD, PHD

and Luís Roberto Gerola, MD, PHD
After some isolated cases of direct myocardial revascularization performed by Goetz

(1) in 1961 and years later by Kolessov (2), myocardial revascularization was standard-
ized by Favaloro (3) and others using cardiopulmonary bypass and cardioplegic arrest.
In 1975, Trapp and Bisarya (4) published for the first time a consecutive series of 63
patients who were operated on without cardiopulmonary bypass. In this series, the right
coronary artery and left anterior descending (LAD) artery were grafted with saphenous
vein graft or internal mammary artery. The authors reported no deaths related to the
cardiac operation and only one death due to stroke. In this landmark paper, the authors
presented the bases of beating-heart revascularization technique including a system to
maintain myocardial perfusion with a perfusion catheter and some maneuvers to stabilize
the heart to perform the distal anastomosis (Fig. 1). At that time, it was considered
dangerous and/or impossible to completely occlude the coronary artery without
perioperative myocardial infarction.
A similar study was done by Ankeney (5) in the same year with satisfactory results,
but these isolated studies were soon abandoned due to technical difficulties, the ease of
performing a delicate anastomosis in a quiet arrested heart, and the belief that distal
coronary perfusion was imperative to avoid myocardial infarction, even for short periods
of coronary occlusion.
In 1981, our group (6) and Benetti (7) in Argentina, working independently, began the
clinical application of beating-heart revascularization. These were the first reports
describing coronary revascularization without cardiopulmonary bypass on consecutive

45
46 Part II / Minimally Invasive Bypass Grafting
Fig. 1. Original method of distal coronary perfusion and maneuvers to stabilize the heart. The
distal perfusing catheter is passed through the side of the internal mammary artery (A) prior to
being led into the distal left anterior descending artery (B). With this technique, perfusion of the
distal coronary artery is carried out to support the heart up until the moment the three interrupted
sutures are placed in the side of the internal mammary artery (C,D).
patients after the techniques of myocardial revascularization with extracorporeal circu-

lation and cardiac arrest had been established.
In our series of 80 patients operated on without cardiopulmonary bypass, and paral-
leling the experience of Trapp et al., only atherosclerotic disease of the right coronary
artery and left anterior descending artery were treated. The technique was premised on
the use of pharmacological manipulation with intravenous calcium-channel antagonist
(verapamil) before coronary occlusion. This created a situation of controlled hypotension
and reduction of myocardial metabolic need, permitting occlusion of the coronary arter-
ies without myocardial injuries. To our knowledge, this represented the first stabilization
maneuver used to facilitate coronary suturing.
Encouraged by this early experience, we initiated a program of myocardial
revascularization without extracorporeal circulation with the goal of employing it in all
applicable patients and, indeed, it became our routine procedure for patients with one- or
two-vessel disease that did not include the circumflex artery circulation.
Naturally, many questions remained unanswered at that time. The first issue we
addressed was whether the snaring of the coronary arteries caused damage to the coronary
arterial wall. In an experimental study in human cadavers (8), we evaluated the potential
for local trauma to the native coronary artery caused by the snaring tourniquets. We
applied both 5-0 polypropylene and 2-0 polyester snares to the proximal and distal right
coronary and LAD arteries in 25 isolated fresh human cadaver hearts. A total of 100
points of snare application to the native coronary vessel were induced and then investi-
gated histologically.
The results suggested a direct relationship between the severity of the arterial lesion,
induced by the snares, and the degree of local atherosclerotic disease in the native coro-
nary artery. Compression and bucking of elastic lamellae with medial fracture were seen
when snares were applied to a region with marked atherosclerotic disease (9) (Fig. 2).
Based on these findings, we developed a surgical technique employing very soft snares
to avoid damage in the coronary wall. Nowadays, we avoid the use of tourniquets alto-
gether in the distal portion of the target coronary artery, as they are unnecessary in
most instances.
It is important to note that issues surrounding the quality of the anastomosis, its short-
and long-term patency, neurological effects, impact on blood use, and costs needed to be
investigated before widespread use of beating-heart revascularization could be supported.
Patency of the coronary anastomosis was analyzed in an unpublished study from our
group. Sixty patients with single-vessel disease (LAD lesion) were randomized to receive
a left internal thoracic artery-to-LAD graft by conventional cardiopulmonary bypass and
cardioplegic arrest or using our off-pump techniques. All patients were submitted to
coronary arteriograms before hospital discharge. Early patency of left internal thoracic
artery to LAD artery was 96% for each group without any statistical difference. This
experience convinced us that two-vessel bypass grafting (LAD and right coronary arter-
ies) could be performed effectively with low morbidity and mortality.
With this conviction, we began to study whether more difficult patients could safely
undergo off-pump myocardial revascularization. Our first objective was to evaluate
patients with acute myocardial infarction who successfully received intravenous strep-
tokinase in the early era of thrombolytic therapy. Twenty-five such patients were sub-
mitted to off-pump myocardial revascularization. The time from acute myocardial
infarction to operation ranged from 1 to 21 d (mean 8 d). There was no in-hospital
mortality, and no patients had hemorraghic complications (10). These results showed that
in the burgeoning era of thrombolytic therapy, it was possible to operate in the subset of
patients with evolving myocardial infarction after thrombolysis.
Another focus of interest was the application of off-pump revascularization to the
elderly, a subject that is covered in more contemporary fashion in Chapter 18. In 1994,
we presented our results in 265 patients older than 70 years of age who underwent
coronary revascularization. Of these, 204 were operated on with conventional bypass
grafting and 61 patients were operated on with beating-heart techniques. The hospital
mortality was 7.8% in patients operated with cardiopulmonary bypass and 1.6% in
patients operated off-pump. Moreover, length of stay and the need for blood transfusion
were lower in patients operated without extracorporeal circulation (11).
A particularly attractive concept of beating-heart techniques was the potential benefit
to postoperative neurological function. To this effect, we undertook a study in which
Fig. 2. Histological aspect of a coronary artery after snaring demonstrating lesion of the media
layer.
cerebral microembolic signals were analyzed by transcranial Doppler in a randomized

clinical trial comparing off-pump vs on-pump revascularization (12). The number of
microembolic signals was significant higher in patients operated on with cardiopulmo-
nary support, although there was no significant clinical difference. Another clinical
comparative study was performed to evaluate neurological complications. We prospec-
tively examined 81 patients, before and up to 7 d after surgery, to compare the neurologi-
cal morbidity between on-pump and off-pump procedures. Neurological abnormalities
were found in 16 patients operated on-pump and 12 patients in the off-pump cohort.
Prolonged coma occurred in one patient who underwent conventional CABG. Permanent
stroke occurred in two patients in the on-pump group and one patient in the off-pump
cohort. No clinical difference between the two groups was found (13).
To address the inflammatory response associated with bypass grafting, we studied the
release of cytokines—tumor necrosis factor-alpha and interleukin-6—as markers of the
inflammatory response in patients undergoing conventional coronary revascularization.
Thirty patients were studied; 15 patients received adjunctive methylprednisolone and 15

did not. In this study we demonstrated that the systemic adverse effects caused by inflam-
matory response after cardiopulmonary bypass were minimized by the use of a corticos-
teroid (14). We then extrapolated this concept by investigating the effect of avoiding
extracorporeal circulation on the release of inflammatory cytokines. Not surprisingly,
patients who underwent off-pump CABG demonstrated significantly reduced levels of
cytokine release (15).
An important contribution to the developing field of off-pump coronary artery bypass
grafting was the introduction by Rivetti et al. (16) of a temporary intraluminal shunt that
facilitated anastomosis and preserved distal coronary perfusion. We commonly employ
this shunt when grafting the right coronary artery, and we believe that it is very important
when no backflow can be seen after coronary occlusion.
For many years to follow, skepticism regarding the quality of the anastomosis com-
bined with the natural resistance of surgeons to change routine procedures, rendered us
one of a couple of groups in the world persisting with the clinical application of off-pump
revascularization.
A landmark event in the field occurred when Benetti proposed a small left thoracotomy
to graft left internal thoracic artery to LAD artery (17), a technique that received the often-
confusing eponym MIDCAB, for minimally invasive direct coronary artery bypass. This
procedure, widely applied and popularized by Calafiore (18), could be considered a
natural evolution of the off-pump technique, as it highlighted the minimal-access poten-
tial of coronary revascularization.
The initial results with MIDCAB were mixed, and our group published less than
optimal early patency rates for left internal thoracic artery to LAD (19) anastomoses.
Compromising with the minimal access concept, we moved to performing a full sterno-
tomy under a small skin incision. Perhaps the most important legacy of the MIDCAB or
LAST (left anterior small thoracotomy) operation, as it is also referred to in the literature,
was that it brought attention to cardiac surgeons worldwide that myocardial revascularization
could be performed without extracorporeal circulation, much in the same way that
Kolessov had done in the past.
Notably, widespread criticism for this procedure soon followed, mainly because of its
restricted applicability to patients with single-vessel disease. But this procedure proved
that coronary artery bypass performed off-pump through a small incision promoted rapid
recovery, avoided blood transfusion, and resulted in fewer postoperative complications,
shortened hospital stay, and reduced costs.
Unlike minimally invasive surgery of the abdomen or other body cavities, minimal
invasive cardiac surgery has two dimensions: one is avoidance of cardiopulmonary
bypass, and the other is limited access. While most agree that the former is of greater
physiological importance, the latter seems to be the more attractive to patients.
While full sternotomy allowed the surgeon to reach all coronary branches, heart
movement continued to challenge the creation of reliable anastomoses and revascularization
of the posterior and lateral walls of the heart. The introduction of mechanical stabilizers
as an effective and reproducible means to promote local reduction of heart motion rep-
resents perhaps the most important factor in the expansion of beating-heart surgery.
The importance of stabilizers was well documented by Calafiore (20), who demon-
strated improved early patency of the left internal thoracic artery and declared a “new era”
of cardiac surgery following clinical introduction of the stabilizer. A discussion of the
leading stabilizing systems appears later in this book (Chapters 6–8).
Many groups, including ours, adopted the full sternotomy-limited skin incision–
mechanical stabilizer approach to off-pump coronary artery bypass with the conviction
that two-vessel grafting was as safe and reliable as conventional bypass grafting.
The issue of circumflex artery revascularization, however, remained a major obstacle
to complete off-pump revascularization. Exposure of the lateral and posterior walls
demanded greater heart mobilization, with more possibilities for hemodynamic instability.
Lima proposed maneuvers for presentation of the marginal branches of the circumflex
artery with little hemodynamic instability (21). With several sutures positioned strategi-
cally between the inferior vena cava and the left inferior pulmonary vein, the heart can
be “verticalized” outside of the chest, creating a true “ectopia cordis.”
Some modification in the positioning of the operation table, associated with other
maneuvers, allows good exposure of the circumflex branches. The successful application
of the “LIMA stitches” and its subsequent modifications allowed complete revascularization
and markedly increased the proportion of patients undergoing off-pump bypass grafting.
With this evolution and the introduction by a burgeoning industry of sophisticated
enabling stabilizing systems, several groups worldwide have moved toward application
of off-pump revascularization to all patients.
We believe that off-pump coronary artery bypass grafting or OPCAB is not a surgical
procedure for all patients, and its ubiquitous application should not be seen as a laudable
accomplishment. Rather, the technique of beating-heart revascularization should be tai-
lored to the patient’s anatomy and needs. In our experience, patients with hypertrophic
left ventricles, diffuse atheromatous coronary arteries, and intramyocardial LAD artery
are not good candidates for off-pump revascularization.
In almost 20 yr performing coronary artery bypass without extracorporeal circulation in
over 3235 patients up to December 2002, we have been rewarded with a 2.8% in-hospital
mortality and low morbidity. Following introduction of mechanical stabilizers and tech-
niques for exposure of the circumflex artery branches, our proportion of patients undergo-
ing off-pump revascularization has doubled from 21% to 49%. We expect that in the next
few years, as better stabilizers, improved surgical maneuvers, and other technologies such
as facilitated anastomotic devices are introduced, our indications will expand.
We believe that myocardial revascularization without cardiopulmonary bypass is a
superior treatment for coronary insufficiency and, although, some comparative studies
(22,23) with low-risk patients do not demonstrate improved clinical outcomes, the real
superiority will become more obvious as OPCAB is applied to more patients with severe
preoperative comorbidities (24–26). It follows, then, that if OPCAB is a better technique
for high-risk patients, it should be a better technique for all patients. While this is our
belief, clinical confirmation is needed. The decision to use cardiopulmonary bypass must
be individualized, but we do believe that today’s coronary surgeon must be able to
perform OPCAB and conventional CABG with equal expertise.
REFERENCES
1. Goetz RH, Rohman M, Haller JD, Dee R, Rosenak SS. Internal mammary-coronary anastomosis. A non-
suture method employing tantalum rings. J Thorac Cardiovasc Surg 1961;41:378–386.
2. Kolessov VL. Mammary artery coronary anastomosis as method of treatment for angina pectoris.
3. Favaloro RG. Saphenous vein autograft replacement of severe segmental coronary artery occlusion
technique. Ann Thorac Surg 1968;5:334–339.
4. Trapp WG, Bisarya R. Placement of coronary bypass graft without pump oxygenator. Ann Thorac Surg
1975;19:1–9.
5. Ankeney JL. To use or not to use pump oxygenator in coronary bypass operation. Ann Thorac Surg
1975;19:108–109.
6. Buffolo E, Andrade JCS, Succi JE, et al. Revascularização direta do miocárdio sem circulação
extracorpórea. Descrição da técnica e resultados iniciais. Arq Bras Cardiol 1982;38:365–373.
7. Benetti FJ. Direct coronary surgery with saphenous vein bypass without cardiopulmonary bypass or
cardiac arrest. J Cardiovasc Surg. 1985;26:217–222.
8. Gerola LR, Moura LAR, Buffolo E, Leão LEV, Soares HC, Gallucci C. Garroteamento da artéria
coronária na revascularização do miocárdio. Relação entre o grau de aterosclerose e a lesão vascular:
estudo experimental. Rev Brasil Cir Cardiov 1987;2:64–69.
9. Gerola LR, Moura LAR, Leão LEV, Soares HC, Branco JNR, Buffolo E. Arterial wall damage caused
by snaring of the coronary arteries during off-pump revascularization. Heart Surgery Forum
2000;3(2):103–107.
10. Vega H, Leão LEV, Silva LA, et al. Surgical myocardial revascularization without using extracorporeal
circulation in patients with acute myocardial infarct treated previously with intravenous streptokinase.
Rev Hosp S Paulo 1990;2:75–78.
11. Buffolo E, Summo H, Aguiar LF, Teles CA, Branco JNR. Myocardial revascularization in patients 70
years of age and older without the use of extracorporeal circulation. Am J Geriatr Cardiol 1997;6:6–9.
12. Malheiros SMF, Massao AR, Gabbai AA, et al. Is the number of microembolic signals related to
neurologic outcome in coronary bypass surgery? Arc Neuropsychiatr 2001;59:1–5.
13. Malheiros SMF, Brucki SMD, Gabbai AA, et al. Neurological outcome in coronary artery surgery with
and without cardiopulmonary bypass. Acta Neurol Scand 1995;92:256–260.
14. Brasil LA, Gomes WJ, Salomão R, Fonseca JHP, Branco JNR, Buffolo E. Uso de corticóide como
inhibidor da resposta inflamatoria sistêmica induzida pela circulação extracorpórea. Rev Bras Cir
Cardivasc 1999;14(3):254–268.
15. Brasil LA, Gomes WJ, Salomão R, Buffalo E. Inflammatory response after myocardial revascularization
with or without cardiopulmonary bypass. Ann Thorac Surg 1998;66:56–59.
16. Rivetti LA, Gandra SMA. Initial experience using an intraluminal shunt during revascularization of the
beating heart. Ann Thorac Surg 1997;63:1742–1747.
17. Benetti FJ. Video assisted coronary bypass surgery. J Cardiac Surg 1995;10:620–625.
18. Calafiore AM, Gianmarco GD, Teodori G, et al. Left anterior descending coronary artery grafting via
left anterior small thoracotomy without cardiopulmonary bypass. Ann Thorac Surg 1996;61:1658–1665.
19. Buffolo E, Teles C, Aguiar LF, et al. Left anterior small thoracotomy and left anterior descending
anastomoses: immediate postoperative analyses. J Am Clin Cardiol 1997;29:716.
20. Calafiore AM, Vitolla G, Mazzei V, Teodori G, Di Gianmarco G, Iovino T, Iaco A. The LAST operation:
technique and results before and after stabilization era. Ann Thorac Surg 1998;66:998–1001.
21. Lima RC. Padronização técnica de revascularização do miocárdio da artéria circunflexa e seus ramos
sem circulação extracorpórea. Tese Doutorado. São Paulo: Universidade Federal de São Paulo/Escola
Paulista de Medicina.
22. Vural KM, Tasdemi O, Karagaz H, Emir M, Tarcon O, Bayazit K. Comparison of the early results of
coronary artery bypass grafting with and without extracorporeal circulation. Thorac Cardiovasc Surg
1995;43(6):620–625.
23. Kshettry VR, Flavin TF, Emery RW, Nicoloff DM, Arom K, Peterson RJ. Does multivessel off-pump
coronary artery bypass reduce postoperative morbidity? Ann Thorac Surg 2000;69:1725–1730.
24. Yokoyama T, Baumgartner FJ, Gheissari A, Capouya ER, Panagiotides GP, Declusion RJ. Off-pump
versus on-pump coronary bypass in high-risk subgroups. Ann Thorac Surg 2000;70:1546–1550.
25. Güller M, Kirali K, Toker ME, et al. Different CABG methods in patients with chronic obstructive
pulmonary disease. Ann Thorac Surg 2001;71:152–157.
26. Ricci M, Karamanoukian HL, Abraham R, et al. Stroke in octogenarians undergoing coronary artery
surgery with and without cardiopulmonary bypass. Ann Thorac Surg 2000;69:1471–1475.
Chapter 4 / Tailoring to the Patient 53
4 Tailoring Minimal Invasive Coronary

Bypass to the Patient
Michael A. Borger, MD, PHD

and M. Anno Diegeler, MD, PHD
CONTENTS
INTRODUCTION
OPCAB
MIDCAB
PARTIAL STERNOTOMY CABG
PORT-ACCESS CABG
DRESDEN TECHNIQUE
TECAB
SUMMARY
REFERENCES
INTRODUCTION
Application of minimally invasive cardiac surgery techniques, in particular coronary
bypass, is becoming increasingly widespread. A review of surgical practice in 72 North
American hospitals, encompassing 17,000 coronary bypass operations performed in
1999, revealed that 16% of procedures were performed off-pump (1).
While a large number of minimally invasive coronary bypass techniques have been
developed, some controversy exists over what actually constitutes “minimally invasive.”
For the purposes of this chapter we will define minimal invasive coronary bypass as any
procedure performed without the aid of cardiopulmonary bypass (CPB) and/or through
a minimal access approach (that is, nonmedian sternotomy).
Furthermore, the substantial number of newly developed operations has led to a con-
fusing array of names and acronyms, including OPCAB, MIDCAB, TECAB, and others.
Our definition of each of these acronyms will be addressed in their respective subsequent
sections. Each operative technique offers certain advantages and disadvantages over the

53
other coronary bypass procedures. One of the most difficult tasks for cardiac surgeons is
to determine which operation is appropriate for which patients. The purpose of this
chapter is to provide a framework for identifying the optimal surgical technique for
specific patient populations—that is, tailoring the coronary bypass operation to the
patient’s needs.
A few cautionary words are necessary before proceeding. First, it should be stressed
that each of the minimal invasive techniques discussed below is associated with a learn-
ing curve of skill and knowledge acquisition. Therefore, surgeons must take into account
their own level of experience, as well as the experience of their respective institutions,
when determining the optimal surgical approach for specific patient populations. Second,
individual patient preferences should always be considered when determining the appro-
priate operation. Patient preferences may become more important over time because of
increasing access to medical information through the Internet and patient-to-patient
communications. However, we strongly recommend against surgeons performing opera-
tions that are clearly beyond their own level of comfort, experience, or knowledge simply
to meet patient expectations. Strong patient preferences may therefore require referral to
a surgeon or to a center that is experienced with performing the procedure in question.
Third, it should be remembered that conventional coronary bypass surgery is always a
viable alternative to each of the minimal invasive techniques listed below and is still a
kind of “gold standard” for coronary revascularization. Fourth, surgeons should not
hesitate to employ conventional coronary bypass in patients who do not meet minimal
invasive inclusion criteria. Because the focus of this book is minimal invasive cardiac
surgery, however, we will not discuss conventional on-pump coronary bypass in any
detail.
OPCAB
For the purpose of this chapter we will define off-pump coronary artery bypass
(OPCAB) as coronary revascularization on a beating heart, without the aid of CPB,
through a standard median sternotomy. Improved stabilizing devices (see Chapters 6–8)
have enabled surgeons to perform beating-heart surgery with greater facility and sense
of comfort, and have allowed operating times to become similar to or less than conven-
tional coronary bypass surgery (2).
The main advantage of OPCAB is the avoidance of CPB with all of its documented
deleterious effects. Avoiding CPB may lower the risk of neurological complications,
renal dysfunction, bleeding, and systemic inflammatory activation (all discussed in more
detail below). Another major advantage of OPCAB is the similarity of its operative
technique to conventional coronary bypass. The full sternotomy approach allows
revascularization of all major coronary territories, an option that is not possible for other
minimal invasive techniques using a limited access. Furthermore, rapid institution of
CPB is easily achieved, which increases the safety of the procedure, particularly during
the learning period.
The main disadvantage of OPCAB is that the complete sternotomy approach increases
surgical trauma, increases the risk of postoperative dehiscence and instability, and, most
important, increases the risk of wound infection. The approach is therefore not really
“minimally invasive,” and no benefit can be gained in postoperative patient comfort or
time to ambulation.
Another disadvantage is that beating-heart surgery is technically more demanding and

may compromise the quality of revascularization, particularly for surgeons with limited
experience. However, a number of studies from experienced centers have revealed that
OPCAB results in excellent early and intermediate graft patency rates, roughly equiva-
lent to conventional coronary bypass (3,4). Although long-term results are not yet avail-
able, there is no foreseeable reason to expect diminished patency rates over time compared
to conventional techniques.
The encouraging results for OPCAB have led investigators to try to determine which
patient populations are optimal for this technique. Although some surgeons state that
OPCAB should be employed in all patients, we do not endorse such a rigid recommen-
dation. Table 1 summarizes our current recommendations for the indications and
contraindications for OPCAB coronary revasularization.
Cardiac Indications and Contraindications for OPCAB

The types of cardiac pathology that are suitable and unsuitable for OPCAB surgery are
fairly well agreed upon in the literature. OPCAB is indicated for multivessel coronary
disease because all territories can be adequately exposed with pericardial stay sutures and
stabilizing devices. Hemodynamic compromise may occur, particularly during exposure
of the circumflex territory. Exposure of the posterior left ventricle impairs cardiac func-
tion by obstructing inflow to the right ventricle and by inducing diastolic dysfunction in
the left ventricle. Coronary flow may also become impaired, particularly in the territory
of the circumflex artery (5). Furthermore, mitral valve incompetence may increase during
cardiac displacement and atrial fibrillation may occur. The majority of these hemody-
namic problems are temporary, however, and most can be managed conservatively with
monitoring, physiological support, and patience. We have found that measuring left atrial
pressure can also be helpful and is easily performed. In addition, atrial pacing at a rate of
90/min often provides further hemodynamic stabilization.
Some types of coronary artery lesions preclude the use of beating-heart surgery. Dif-
fuse coronary artery calcification makes temporary vessel occlusion nearly impossible
and therefore OPCAB should be avoided in these patients. A deep intramyocardial left
anterior descending coronary artery (LAD) severely complicates the dissection and
exposure of this vessel on a beating heart, thereby necessitating conventional coronary
bypass under cardioplegic arrest to achieve a reproducible quality of revascularization.
Temporary occlusion of a dominant right coronary artery with a noncritical stenosis may
cause significant ischemia and arrhythmias (6). A shunt, or a more distally performed
occlusion at the posterior interventricular artery, may enable OPCAB to be performed in
such patients. Some authors state that severe left main stenosis carries increased risk for
OPCAB surgery, but this point is controversial (7).
Other types of cardiac pathology that require increased caution during OPCAB
revascularization are severe left ventricular (LV) dysfunction, markedly elevated LV
end-diastolic pressure, and moderate to severe mitral regurgitation. Although these
patients could benefit the most by avoiding CPB, their hemodynamic management
requires experienced cardiac anesthetic and surgical teams. If OPCAB is attempted in
these patients, emergency conversion to CPB (so-called crash CPB) should be avoided
whenever possible, because the subsequent results are particularly poor. In addition, we
recommend against OPCAB surgery in patients with uncontrolled ventricular
arrhythmias. Such patients should undergo revascularization using conventional coro-

nary bypass techniques.
Noncardiac Indications and Contraindications for OPCAB

OPCAB plays an important role for patients who are at high risk for complications of
CPB (see Table 1). Elderly patients may benefit from OPCAB, predominantly by low-
ering the risk of neurological injury. Elderly age is consistently identified as one of the
most important risk factors for stroke during coronary artery bypass grafting (CABG).
Ricci et al. compared outcomes in 97 octogenarians undergoing beating-heart surgery to
172 octogenarians having conventional coronary bypass (8). These investigators found
a 0% incidence of stroke in OPCAB patients, compared to 9% for conventional patients.
In addition, OPCAB may lower the incidence of other complications in elderly patients.
Demers and Cartier demonstrated lower rates of atrial fibrillation and transfusion
requirements in elderly patients undergoing OPCAB surgery when compared to conven-
tional surgery (9).
Patients with a calcified ascending aorta may benefit the most from OPCAB surgery.
Off-pump total arterial revascularization can be performed without aortic cannulation or
crossclamping, thereby eliminating the most important cause of embolic stroke during
coronary bypass surgery (10). Gaudino et al. compared two techniques for coronary
revascularization in 211 patients with diffuse atherosclerosis of the ascending aorta (11).
They found that OPCAB resulted in a lower incidence of neurological complications,
renal insufficiency, and intensive care unit (ICU) and hospital lengths of stay when
compared to revascularization during ventricular fibrillation. Although total arterial
revascularization eliminates the need for ascending aortic manipulation, saphenous vein
grafts can also be used in patients with a diseased ascending aorta. Newly developed
automated devices allow for proximal anastomoses to be constructed between the saphe-
nous vein and the ascending aorta, while avoiding clamping of the ascending aorta.
OPCAB may also be indicated in patients at high risk for postoperative neurological
impairment. OPCAB is associated with decreased cerebral microemboli, the predomi-
nant cause of postoperative neuropsychological impairment (12). We used transcranial
Doppler monitoring to demonstrate that patients undergoing off-pump surgery experi-
enced on average 11 cerebral microemboli during the operation compared to 394 in
conventional bypass patients (13). Lowering the number of cerebral microemboli during
coronary bypass was directly associated with better postoperative cognitive perfor-
mance (12,13).
Patients with severe carotid stenosis or a history of previous neurologic injury are par-
ticularly high-risk groups for postoperative neurologic sequelae and may benefit substan-
tially from avoiding the systemic inflammatory effects and micro- and macroemboli
associated with CPB. Patients with a prior history of stroke or TIA, or possibly degenerative
disorders such as Parkinson’s, may be optimal candidates for beating-heart surgery. In
addition, patients with recent stroke or long-standing hypertension may benefit from a
lower risk of cerebral hemorrhage, because lower levels of anticoagulation are required.
Another subpopulation that may benefit from beating-heart surgery is the patient with
renal dysfunction. Avoiding CPB appears to lower the risk of perioperative renal
ischemia, which is poorly tolerated in patients with preexisting renal impairment. Ascione
et al. performed a randomized trial of beating-heart vs conventional surgery in 50 CABG
patients (14). These investigators found that OPCAB resulted in better postoperative
glomerular filtration rate and renal tubular function than conventional surgery. Improved
renal function should decrease the risk of renal failure, a major cause of postoperative
morbidity and resource utilization, in patients with preoperative renal insufficiency. This
may also be important for insulin-dependent diabetics, even with normal creatinine lev-
els, since post-CPB renal failure is not uncommon in such patients.
OPCAB may play an important role for patients with chronic obstructive pulmonary
disease (COPD). Systemic inflammatory activation and capillary leakage during CPB
cause postoperative pulmonary edema and markedly impaired pulmonary function. The
impaired pulmonary function is temporary and well tolerated by most patients, but may
lead to respiratory failure, prolonged intubation, and/or adult respiratory distress syn-
drome in patients with preexisting COPD. Guler and colleagues prospectively assessed
58 CABG patients with COPD (15). Beating-heart surgery patients (OPCAB and
MIDCAB) had better postoperative forced expiratory volumes (FEV1) than conventional
patients. Improved pulmonary function may lead to decreased risk of postoperative res-
piratory failure, an important cause of prolonged ICU stay.
Finally, OPCAB may be important for patients with a history of bleeding diathesis.
Several investigators have demonstrated that OPCAB is associated with improved plate-
let counts and platelet function, decreased fibrinolysis, and decreased blood loss when
compared to conventional surgery. These findings may be important for patients with
documented bleeding diathesis or preoperative anemia, as well as in patients who refuse
blood transfusions (i.e., Jehovah’s Witness patients).
There are very few noncardiac contraindications for OPCAB surgery (Table 1). How-
ever, one special circumstance deserves note. Chavanon et al. found that iatrogenic aortic
dissection was 50 times more likely to occur during beating-heart surgery than during
conventional CABG (16). Iatrogenic dissection was caused by applying a partial occlud-
ing aortic crossclamp in patients with a dilated ascending aorta. Although the incidence
of aortic dissection was still low in OPCAB patients (1%), this devastating complication
should obviously be avoided whenever possible. Because of their experience, these
investigators now recommend against using a partial occluding clamp in patients with an
ascending aortic diameter greater than 4.0–4.5 cm (9).
MIDCAB
Minimal invasive direct coronary artery bypass (MIDCAB) is another common form
of minimal invasive coronary surgery, comprising approx 10% of all beating-heart
operations (2). This procedure is discussed in great detail by Dr. Dullum in Chapter 9. For
the purpose of this chapter we will define MIDCAB as coronary revascularization through
an anterolateral thoracotomy. The procedure entails dissection of the left internal mam-
mary artery (LIMA) with specialized rib retractors through a small (6–8 cm) anterolateral
thoracotomy. The LIMA is then anastomosed to the LAD through the thoracotomy
incision, using myocardial stabilizers similar to those used for OPCAB. Although
MIDCAB can be performed with the assistance of port-access CPB and cardioplegic
arrest (see below), beating-heart techniques are employed in the vast majority of cases.
MIDCAB is usually performed for isolated LAD lesions, but radial artery “T” grafts
can also be used to bypass favorable diagonal or intermediate branches. A right antero-
lateral thoracotomy (“right MIDCAB”) can be used for patients with isolated proximal
RCA lesions. The right internal mammary artery (RIMA) is harvested using the same
Table 1
Current Recommendations for OPCAB
Pathology Indications Contraindications
Cardiac Multivessel disease Intramyocardial LAD
Diffusely calcified coronaries
RCA stenosis < 80%
Severe LV dysfunction
↑↑ LVEDP
Moderate MR
Ventricular arrhythmias
Noncardiac Elderly age Dilated ascending aorta
Calcified ascending aorta
Neurologic history
Renal dysfunction
COPD
Bleeding diatheses
LAD = left anterior descending artery, RCA = right coronary artery, LV = left ventricle,
LVEDP = left ventricular end-diastolic pressure, MR = mitral regurgitation, COPD =
chronic obstructive pulmonary disease.
technique as for the LIMA and anastomosed to the mid-RCA. In the vast majority of
patients, however, the RCA is diseased down to the crux cordis segment. Because the
distal branches of the RCA are inaccessible through this limited access, the classical right
MIDCAB is rarely performed.
A special type of limited access procedure can be performed through a transabdominal
approach (also known as the subxiphoid procedure) (17). This approach is explored by
Dr. Benetti in great detail in Chapter 11. The transabdominal approach involves a chev-
ron-shaped incision in the subxiphoid region, followed by harvesting of the right gastro-
epiploic artery (GEA) and LIMA, if necessary. The GEA is anastomosed to the RCA, or
one of its branches, and the LIMA is anastomosed to the LAD. Although transabdominal
technique has the benefit of being feasible for multivessel coronary disease, the approach
is markedly different than most cardiac operations and therefore is not used by many
cardiac surgeons.
Another special type of MIDCAB surgery is the endoscopic coronary artery bypass
(ENDOCAB) procedure. ENDOCAB involves endoscopic harvesting of the LIMA (or
RIMA) through three small ports, followed by anastomosis of the LIMA to the LAD (or
RIMA to the RCA) through a very small (4–5 cm) anterolateral thoracotomy. As with
other MIDCAB procedures, ENDOCAB is usually performed on a beating heart, but can
also be done with port-access CPB and cardioplegic arrest.
The major advantages of MIDCAB surgery are avoidance of CPB and avoidance of
a complete sternotomy. The benefits of surgery without CPB were covered in detail
above. The benefits of surgery without a sternotomy are decreased risk of sternal wound
infection and decreased time to ambulation. Several investigators, including our group
(18), have demonstrated that MIDCAB is associated with lower postoperative pain scores
(except in the immediate postoperative period) and better quality of life than conven-
tional coronary bypass surgery. These benefits are particularly notable with recently
developed chest retractors that eliminate the need for rib removal or sternocostal
disarticulation.
The main concern about MIDCAB surgery, just as for OPCAB, was the quality of
coronary anastomoses. Several surgeons have questioned the ability to perform anasto-
moses on a beating heart through a small thoracotomy. However, increasing evidence has
revealed that excellent patency rates can be achieved with the MIDCAB procedure. In
one of the largest series to date, we examined graft patency rates in 618 MIDCAB patients
(19). Early postoperative angiography, performed in 585 patients, revealed an overall
patency rate of 97.4%. Six-month postoperative angiography was performed in 353
patients and revealed an overall patency of 95.8%, increasing to 97.0% for the last year
of the study. Significant graft stenosis was present in 4.5% of patients at 6 mo. Several
other investigators have confirmed the excellent results achievable with MIDCAB
revascularization (4).
One definite disadvantage of MIDCAB is the inability to institute CPB through an
anterolateral thoracotomy. If hemodynamic instability is to occur during the operation,
patients must be placed on CPB through the femoral vessels or through a separate median
sternotomy incision. Fortunately, with proper patient selection, this is a rare occurrence.-
Another disadvantage of MIDCAB surgery is the requirement for single-lung ventila-
tion. The lung should be deflated during IMA dissection, thereby necessitating double-
lumen endotracheal intubation. Single-lung ventilation may increase anesthesia time and
is contraindicated in patients with severe COPD or pulmonary hypertension. The final
disadvantage of MIDCAB is the possibility of postoperative lung herniation. The devel-
opment of new chest retractors has eliminated the need for rib resection or sternocostal
disarticulation, reducing the risk of lung herniation.
Cardiac Indications and Contraindications for MIDCAB

Table 2 summarizes our current recommendations for proper patient selection for
MIDCAB surgery. As mentioned previously, the predominant cardiac indication for
MIDCAB is isolated high-grade LAD disease. In a subset of patients from the above-
mentioned study (19), we randomized 200 patients with isolated high-grade LAD steno-
sis to receive MIDCAB or PTCA. Because of the surgical nature of the MIDCAB
procedure, these patients had a slightly higher prevalence of periprocedural adverse
events than PTCA patients. However, 6-mo follow-up revealed that MIDCAB patients
had significantly less angina, fewer stenoses, and fewer reinterventions than PTCA
patients, particularly those with a complex (type B or C) LAD lesion (19). Therefore,
MIDCAB may be the procedure of choice for patients with isolated, complex, high-grade
disease of the LAD.
MIDCAB can also be attempted in patients with less severe stenosis of the LAD,
particularly if a shunt is employed. However, there is an increased risk of temporary
ischemia and hemodynamic instability during vessel occlusion in these patients. Patients
with exceptionally favorable diagonal or intermediate vessels may also undergo bypass
of these vessels with a “T” graft, but this is a relatively uncommon and challenging
procedure. We would recommend this technique only in patients with a normal body
mass index. MIDCAB is also the option of choice for redo LAD revascularization,
provided the LIMA was not used during the previous operation.
Occasionally, a “right MIDCAB” can be used to revascularize isolated RCA disease.
However, this is a more technically demanding procedure than conventional MIDCAB,
and it results in significantly lower graft patency rates (4). The transabdominal procedure
can be used in patients with both LAD and RCA disease, but again this technique is not
commonly used. The subxiphoid approach may be the best alternative, however, for
patients with isolated RCA disease who have undergone a previous CABG operation,
particularly if a patent LIMA to LAD graft is present.
A special indication for MIDCAB surgery is very-high-risk patients with multivessel
disease who cannot tolerate CPB. In such patients PTCA and stenting of the RCA and/
or circumflex territory is perfromed, followed by MIDCAB revascularization of the
LAD. This so-called hybrid procedure has been performed in several centers with satis-
factory results (20) and is discussed in great detail in Chapter 15. Isolated MIDCAB may
also be indicated for patients with LAD disease and “no-option” RCA or circumflex
disease. The procedure results in an incomplete revascularization. However, the options
for treating the culprit lesion are limited in such patients, and often their angina can be
relieved with a MIDCAB alone.
The cardiac contraindications to MIDCAB are similar to those for OPCAB. We
strongly recommend against MIDCAB in patients with an intramyocardial or diffusely
calcified or diseased LAD. We also recommend against MIDCAB in patients with a small
(<1.5 mm) LAD or an occluded LAD that cannot be localized on the angiogram. MIDCAB
may also be difficult in patients with severe left ventricular hypertrophy, since the LAD
is often notably displaced to the left.
Noncardiac Indications and Contraindications for MIDCAB

Table 2 summarizes the major noncardiac variables to be considered when deciding
if a patient is suitable for MIDCAB surgery. In general, MIDCAB patients fall into two
groups. One group is comprised of young, low-risk patients with LAD disease that is
favorable for MIDCAB revascularization. Such patients would do well with MIDCAB
or conventional coronary bypass, but MIDCAB may be a better option because of quicker
rehabilitation and better postoperative quality of life. The second MIDCAB group is
comprised of very-high-risk patients with isolated LAD disease or multivessel disease
that is to be treated with the hybrid procedure. As discussed above, high-risk patients may
benefit greatly by avoiding the deleterious effects of CPB. The noncardiac indications for
MIDCAB in such patients are essentially the same as for OPCAB (see Table 1). There-
fore, elderly patients, patients with a calcified ascending aorta, and patients with preop-
erative renal insufficiency are likely to benefit from MIDCAB surgery if the coronary
anatomy is favorable.
The noncardiac contraindications to MIDCAB are severe COPD and severe obesity.
As mentioned previously, MIDCAB requires single-lung ventilation and therefore should
not be used in patients with an FEV1 of less than 1.0 L. Severe obesity (body mass index
> 35) is also a contraindication to MIDCAB because of the resultant difficulty with
surgical exposure. The MIDCAB approach may also be difficult in women with large
breasts. A previous laparotomy is a contraindication for patients in whom a transabdomi-
nal procedure is being considered
PARTIAL STERNOTOMY CABG

Another minimal invasive coronary bypass operation is the partial sternotomy CABG.
In this technique, the lower half of the sternum is divided in the midline up to the third
intercostal space. The rest of the procedure is performed using a specialized sternal
retractor (21). Alternatively, an “L”- or “T”-shaped sternotomy can be employed (22).
Table 2
Current Recommendations for MIDCAB
Cardiac Isolated high-grade (>80%) Intramyocardial LAD
LAD stenosis Diffusely calcified LAD
Redo LAD bypass (if LIMA LAD < 1.5 mm
not previously used) Severe LV dysfunction
RCA stenosis (“right Ventricular arrhythmias
MIDCAB” or subxiphoid)
Multivessel disease (hybrid
procedure)
Noncardiac Young, active patients Severe COPD (FEV1 < 1.0 L)
Same indications as OPCAB Severe obesity (BMI > 35)
in high-risk patients Previous laparotomy
(subxiphoid procedure only)
LAD = left anterior descending artery, LIMA = left internal mammary artery, RCA = right
coronary artery, LV = left ventricle, COPD = chronic obstructive pulmonary disease, BMI =
body mass index.
The LITA and RITA can be adequately harvested through this incision, or the GEA can
be harvested through a small subxiphoid extension. The partial sternotomy CABG allows
for beating-heart revascularization of the LAD and RCA territories. Saphenous vein
grafts are not used because the ascending aorta is not exposed.
As with the OPCAB and MIDCAB procedures, the main advantage of the partial
sternotomy CABG is the avoidance of CPB. The advantages of a partial sternotomy over
a complete sternotomy (OPCAB) are a smaller incision (10–12 cm), less traction on the
brachial plexus, and lower risk of postoperative sternal dehiscence or infection. The
advantages of the partial sternotomy procedure over MIDCAB are the ability to perform
double-vessel revascularization, the familiar surgical technique, and the ease of conver-
sion to full sternotomy and CPB.
The main disadvantage of partial sternotomy CABG is prolonged operative times,
particularly in the early stages of learning this procedure (22). The difficulties with
surgical exposure, as well as the nominal advantages over a complete sternotomy tech-
nique, have led most surgeons to choose conventional OPCAB over partial sternotomy CABG.
Table 3 displays our current recommendations for partial sternotomy CABG. The
main cardiac indication is double-vessel (LAD and RCA) disease. If the RCA is
bypassable in its midsection, then the RITA is employed. If a posterior branch of the RCA
needs to be bypassed, then the GEA is used. The cardiac contraindications to partial
sternotomy CABG are the same as for OPCAB surgery.
Because partial sternotomy CABG is performed without the use of CPB, the noncar-
diac indications are the same as for OPCAB. The only noncardiac contraindication, if the
GEA is to be used, is a previous laparotomy.
PORT-ACCESS CABG
Port-access CABG refers to coronary revascularization, with CPB support and
cardioplegic arrest, through a small anterior thoracotomy (23). The patient is placed on
Table 3
Current Recommendations for Partial Sternotomy CABG
Cardiac Double-vessel disease (LAD Same as for OPCAB
and RCA)
Noncardiac Same as for OPCAB Previous laparotomy (if GEA
is to be used)
LAD = left anterior descending artery, RCA = right coronary artery, GEA = gastroepiploic
artery.
CPB through the femoral vessels and an endoaortic balloon is used to separate the coro-
nary and systemic circulations.
The main advantage of port-access CABG is that distal anastomoses are performed on
a quiescent, bloodless field through a minimal access incision. In addition, patients with
double-vessel disease (LAD and circumflex territory) can be revascularized successfully
with this procedure.
There are several disadvantages to port-access CABG. The main disadvantage is the
risk of retrograde aortic dissection, particularly in patients with peripheral vascular dis-
ease. Port-access CABG is contraindicated in patients with peripheral vascular disease,
aortic aneurysmal dilation, or calcification or plaques in the ascending aorta. Aortic
regurgitation is another contraindication to this procedure, although retrograde cardiople-
gia can be employed through a transjugular coronary sinus catheter in such patients.
Another major disadvantage of port-access CABG is the risk of endoaortic balloon
migration with resultant cerebral ischemia. In addition, the procedure entails prolonged
operative and CPB times, and is more expensive than all alternative techniques at the
present time.
Multi-institution registries have demonstrated that port-access CABG can be per-
formed in low-risk patients with acceptable results (23). Because of the risks of major
complications, however, we do not currently recommend this procedure over MIDCAB
(for isolated LAD disease) or OPCAB (for multivessel disease).
DRESDEN TECHNIQUE
Another form of minimal invasive coronary bypass is the Dresden technique (24). For
this procedure, an anterolateral thoracotomy is performed in the third intercostal space,
with disarticulation of the adjacent sternocostal junctions. The aorta and right atrium are
directly cannulated through this incision, and the aorta is occluded with a conventional clamp.
The Dresden technique has the same main advantage as port-access CABG—coronary
surgery on an arrested heart—without the major disadvantages of endoaortic balloon
migration or retrograde aortic dissection. The presence of peripheral vascular disease is
not a limitation for this technique. In addition, the Dresden technique can be used for
patients with double-vessel disease, i.e., LAD and high obtuse marginal branches of the
circumflex. Endoscopic harvesting of both thoracic arteries can be performed with this
approach and may actually enhance the ease of the procedure, particularly if using the
recently available da Vinci Telemanipulator system (Intuitve Surgical; Mountain View, CA).
The main disadvantage of the Dresden technique is the limited and unfamiliar access
to the heart. The more posterior marginal branches as well as the right coronary artery are
difficult to expose. In addition, the third rib is disarticulated from its sternal connection
and may cause significant postoperative discomfort. Procedure times are prolonged and
technical demands are increased. Perhaps because of these limitations, the Dresden tech-
nique is currently performed in few surgical centers.
TECAB
Total endoscopic coronary artery bypass (TECAB) is an experimental procedure
currently being examined in a few cardiac surgery centers. LIMA harvesting and coro-
nary revascularization is performed through four very small (1–2 cm) incisions, using
robotic techniques (25,26). The procedure was originally performed with port-access
CPB support, but concerns about its complications has led to the use of beating-heart
techniques.
The main advantage of TECAB is decreased postoperative pain and time to ambulation.
The main disadvantages are markedly prolonged operative times, a high rate of conver-
sion to MIDCAB, the need for a specialized surgical team, and increased costs.
TECAB is still an experimental surgical approach and is currently performed in low-
risk patients with isolated, high-grade LAD stenoses. The RIMA can also be used for
diagonal or intermediate coronary bypass in exceptional patients (25). Contraindications
for TECAB are an intramyocardial, diffusely diseased, or noncritically stenosed LAD.
Any clinical pulmonary disease, chest deformations, or previous surgery of the left chest
are also contraindications to TECAB.
It should be stressed that TECAB should be reserved for centers that are experienced
in robotic techniques. Owing to the steep learning curve, this approach should only be
performed under a strict scientific protocol. Further developments in endoscopic stabi-
lizers and automated anastomotic devices may increase the feasibility of this procedure
in the future.
SUMMARY
The purpose of this chapter was to make current recommendations about which mini-
mal invasive coronary bypass procedure is suitable for which patient groups. The recom-
mendations are likely to change, however, as more data and new enabling technologies
become available. As stated earlier, every minimal invasive procedure has its own learn-
ing curve of skill and knowledge acquisition. When deciding which technique to use,
surgeons must be aware of their own limitations as well as the limitations of their insti-
tutions. Standard conventional coronary bypass should always be considered an option.
We currently recommend OPCAB surgery for patients with multivessel coronary
disease who are at high risk for complications during CPB. OPCAB should be strongly
considered in elderly patients and in patients with ascending aortic atherosclerosis, pre-
vious neurological injury, renal impairment, or COPD.
MIDCAB revascularization is the procedure of choice for patients with isolated, high-
grade LAD stenosis. Low-risk and high-risk patients may benefit significantly from
this procedure.
The partial sternotomy CABG and Dresden technique are other viable minimal inva-
sive operations, but are not commonly performed. The subxiphoid procedure may be
particularly helpful for redo CABG patients requiring isolated RCA revascularization.
We currently do not recommend port-access CABG because of the risk of major
complications. TECAB is currently an experimental procedure, but may become feasible
in the future.
REFERENCES
1. Brown PP, Mack MJ, Simon AW, et al. Comparing clinical outcomes in high-volume and low-volume
off-pump coronary bypass operation programs. Ann Thorac Surg 2001;72:S1009–S1015.
2. Mack MJ. Is there a future for minimally invasive cardiac surgery? Eur J Cardiothorac Surg
1999;16:S119–S125.
3. Puskas JD, Thourani VH, Marshall JJ, et al. Clinical outcomes, angiographic patency, and resource
utilization in 200 consecutive off-pump coronary bypass patients. Ann Thorac Surg 2001;71:477–484.
4. Stanbridge RDL, Hadjinikolaou LK. Technical adjuncts in beating heart surgery. Comparison of
MIDCAB to off-pump sternotomy: a meta-analysis. Eur J Cardiothorac Surg 1999;16:S24–S33.
5. Grundeman PF, Borst C, van Herwaarden JA, Verlaan CW, Jansen EW. Vertical displacement of the
beating heart by the octopus tissue stabilizer: influence on coronary flow. Ann Thorac Surg
1998;65:1348–1352.
6. Diegeler A, Matin M, Falk V, et al. Indication and patient selection in minimally invasive and off-pump
coronary artery bypass grafting. Eur J Cardiothorac Surg 1999;16:S79–S82.
7. Yeatman M, Caputo M, Ascione R, Ciulli F, Angelini GD. Off-pump coronary artery bypass surgery for
critical left main stem disease: safety, efficacy and outcome. Eur J Cardiothorac Surg 2001;19:239–244.
surgery with and without cardiopulmonary bypass. Ann Thorac Surg 2000;69:1471–1475.
9. Demers P, Cartier R. Multivessel off-pump coronary artery bypass surgery in the elderly. Eur J
10. Borger MA, Ivanov J, Weisel RD, Rao V, Peniston CM. Stroke during coronary bypass surgery: prin-
cipal role of cerebral macroemboli. Eur J Cardiothorac Surg 2001;19:627–632.
11. Gaudino M, Glieca F, Alessandrini F, et al. The unclampable ascending aorta in coronary artery bypass
patients: a surgical challenge of increasing frequency. Circulation 2000;102:1497–1452.
12. Borger MA, Peniston CM, Weisel RD, Vasiliou M, Green REA, Feindel CM. Neuropsychological
impairment after coronary bypass surgery: effect of gaseous emboli during perfusionist interventions.
13. Diegeler A, Hirsch R, Schneider F, et al. Neuromonitoring and neurocognitive outcome in off-pump
versus conventional coronary bypass operation. Ann Thorac Surg 200;69:1162–1166.
14. Ascione R, Lloyd CT, Underwood MJ, Gomes WJ, Angelini GD. On-pump versus off-pump coronary
revascularization: evaluation of renal function. Ann Thorac Surg 1999;68:493–498.
15. Guler M, Kirali K, Toker ME, et al. Different CABG methods in patients with chronic obstructive
pulmonary disease. Ann Thorac Surg 2001;71:152–157.
16. Chavanon O, Carrier M, Cartier R, et al. Increased incidence of acute ascending aortic dissection with
off-pump aortocoronary bypass surgery? Ann Thorac Surg 2001;71:117–121.
17. Subramanian VA, Patel NU. Transabdominal minimally invasive direct coronary artery bypass grafting
(MIDCAB). Eur J Cardiothorac Surg 2000;17:485–487.
18. Diegeler A, Walther T, Metz S, et al. Comparison of MIDCAB versus conventional CABG surgery
regarding pain and quality of life. Heart Surg Forum 1999;2:290–295.
19. Diegeler A, Spyrantis N, Matin M, et al. The revival of surgical treatment for isolated proximal high
grade LAD lesions by minimally invasive coronary artery bypass grafting. Eur J Cardiothorac Surg
2000;17:501–504.
20. Cohen HA, Zenati M, Smith AJC, et al. Feasibility of combined percutaneous transluminal angioplasty
and minimally invasive direct coronary artery bypass in patients with multivessel coronary artery dis-
ease. Circulation 1998;98:1048–1050.
21. Niinami H, Takeuchi Y, Ichikawa S, Suda Y. Partial median sternotomy as a minimal access for off-
pump coronary artery bypass grafting: feasibility of the lower-end sternal splitting approach. Ann
Thorac Surg 2001;72:S1041–S1045.
22. Lichtenberg A, Klima U, Harringer W, Kim PY, Haverich A. Mini-sternotomy for off-pump coronary
artery bypass grafting. Ann Thorac Surg 2000;69:1276–1277.
23. Grossi EA, Groh MA, Lefrak EA, et al. Results of a prospective multicenter study on port-access
coronary bypass grafting. Ann Thorac Surg 1999;68:1475–1477.
24. Gulielmos V, Brandt M, Dill HM, et al. Coronary artery bypass grafting via median sternotomy or lateral
thoracotomy. Eur J Cardiothorac Surg 1999;16:S48–S52.
25. Kappert U, Cichon R, Schneider J, et al. Technique of closed chest coronary artery surgery on the beating
heart. Eur J Cardiothorac Surg 2001;20:765–769.
26. Mohr FW, Falk V, Diegeler A, et al. Computer-enhanced “robotic” cardiac surgery: experience in 148
patients. J Thorac Cardiovasc Surg 2001;121:842–853.
Chapter 5 / OPCAB: A Primer on Technique 67
5 OPCAB
A Primer on Technique
John D. Puskas, MD, MSC

CONTENTS
INTRODUCTION
PREOPERATIVE EVALUATION
OPERATIVE PROCEDURE
SEQUENCE OF GRAFTING IN OPCAB
CARDIAC DISPLACEMENT AND PRESENTATION OF CORONARY
TARGETS
CORONARY STABILIZATION AND GRAFTING
PERFUSION-ASSISTED OPCAB
SUMMARY
REFERENCES
INTRODUCTION
Coronary artery bypass without the use of cardiopulmonary bypass off-pump coronary
artery bypass (OPCAB), is gaining popularity due to recent scientific reports showing its
safety. OPCAB is performed to avoid the diffuse inflammatory response, multiorgan
dysfunction, and neurological complications associated with cardiopulmonary bypass.
The authors and others have recently shown excellent graft patency, improved outcomes,
and lower costs with OPCAB (1).
Experience with OPCAB over 5 yr has allowed an evolution in surgical technique.
Early in the experience, it was common to require inotropic support for hemodynamic
instability during OPCAB. Since refining techniques for cardiac positioning and stabi-
lization, use of inotropes is infrequent. With patience and persistence, OPCAB can be
performed in the vast majority of coronary revascularization patients. Patients for whom
OPCAB may be inappropriate are those in cardiogenic shock, those suffering ischemic

67
arrhythmias, and those with physical conditions that profoundly limit rotation of the heart
(pectus excavatum or previous left pneumonectomy). Intramyocardial or unusually small
or calcified coronary arteries may be safely bypassed off-pump only with the benefit of
considerable experience.
PREOPERATIVE EVALUATION
Coronary revascularization candidates must undergo a complete history and physical
exam. If the Allen’s test is inconclusive, radial and ulnar artery duplex examinations are
performed before radial arteries are harvested. Pulmonary function tests are performed
only in patients with severe chronic obstructive pulmonary disease (COPD) or active
pulmonary disease. Criteria for preoperative carotid duplex examination include left
main disease, peripheral vascular disease, carotid bruits, history of stroke or transient
ischemic attack, heavy tobacco use, or age over 65 yr. If significant carotid disease is
suggested by duplex examination, then a carotid angiogram is obtained. When > 79%
stenosis is present, staged carotid endarterectomy followed by coronary revascularization
is performed.
OPERATIVE PROCEDURE
All patients undergoing OPCAB require invasive monitoring during operation. At a
minimum, an arterial line and central venous line are required. Significantly depressed
left ventricular function mandates a Swan–Ganz catheter. Those patients with severe
ventricular dysfunction may benefit from preoperative intra-aortic balloon pump (2).
Good communication between the surgeon and the anesthesiologist throughout the pro-
cedure is essential. Multiple displacements of the heart during OPCAB subject the patient
to repeated hemodynamic changes and require an attentive anesthesiologist throughout
the procedure.
The OPCAB procedure differs in many ways from on-pump CABG, but the skin entry
and sternotomy are identical. A #10 blade knife is used through the skin and fat, down
to fascia. Electrocautery is then utilized only selectively for bleeding and to divide the
muscles and fascia at the middle of the sternum. Standard sternotomy is performed with
a sternal saw. An upward-lifting Favoloro-type mammary retractor is used to harvest the
LIMA and/or the RIMA. This retractor aids in exposure of the arteries and minimizes
chest wall trauma. Heparin is given (1.5 mg/kg) with a target ACT > 300 s. Heparin (3000
units) is reinfused every 30 min to maintain an ACT > 300 s. This regimen is prompted
by the recognition that heparin metabolism is more rapid in warm OPCAB patients than
in those cooled on cardiopulmonary bypass.
After heparin is given, the mammary is divided and a mixture of papaverine and
lidocaine is injected into the lumen of the mammary artery and allowed to reside there
for 15–30 min. The Medtronic sternal retractor (Octobase™) is placed. This retractor is
designed specifically for the Medtronic Octopus III (Medtronic, Minneapolis, MN)
stabilizer, which can be secured to any aspect of the retractor.
A wide “T”-shaped pericardiotomy is performed, dividing the pericardium from the
diaphragm down towards—but not into—the left and right phrenic nerves. The left and
right pericardiophrenic artery and vein branches are carefully clipped and divided to
avoid postoperative bleeding. Both pleural spaces are opened widely. Care is taken
during the dissection to clip any large vessels encountered and to avoid the phrenic
nerves. It is important to divide the diaphragmatic muscle slips that insert on the right side
of the xiphoid, to allow elevation of the right hemisternal border, creating space for
rightward cardiac displacement. Similarly, excision of a large right-sided pericardial fat
pad will provide additional room. Placement of two rolled towels under the right limb of
the sternal retractor elevates the right sternal edge, allowing the heart to be repositioned
toward the right without compression against the sternum or retractor. One or two heavy
pericardial sutures are placed on the left pericardium above the phrenic nerve. It is
important to divide the left side of the pericardium off the diaphragm toward the phrenic
nerve, so that traction on this left pericardium may assist in rotating and displacing the
heart toward the right to visualize the left lateral wall.
The most important traction suture is a deep posterior pericardial suture placed
approximately two-thirds of the way between the inferior vena cava and the left pulmo-
nary vein at the point where the pericardium reflects over the left atrium. Care should be
taken with placement of this suture to avoid the aorta, esophagus, left lung, and pulmo-
nary veins. The suture is covered with a rubber catheter to prevent trauma to the epicar-
dium. When this suture is retracted toward the patient’s feet, it elevates the base of the
heart toward the ceiling and lifts the apex vertically with remarkably little change in
hemodynamics. When the deep pericardial traction suture is retracted toward the left
shoulder, it rotates the heart from left to right. A variety of cotton slings may be applied
at the base of this suture to aid in displacing the heart into the right pleural cavity. The
slings may be particularly helpful during OPCAB in the setting of cardiomegaly.
All patients undergo epiaortic ultrasound, which may add only 1–2 min to the proce-
dure. Epiaortic ultrasound guides the surgeon in individualized placement of aortic clamps
and proximal anastomoses to reduce the risk of embolism of atherosclerotic debris from
the ascending aorta. Grade IV or V atherosclerosis of the ascending aorta precludes aortic
crossclamping and mandates a change in the site of proximal anastomoses. Other possible
sites for proximal anastomoses include the innominate artery, LIMA, and RIMA.
The heart is allowed to roll with gravity into the left or right chest, facilitated by table
rotation and traction suture(s) and occasionally a cotton sling. The heart should never be
compressed against the sternum or pericardium. The right pericardial traction sutures
are released when exposing the left side of the heart and similarly the left traction sutures are
released when exposing the right coronary artery. Pericardial sutures on both the right and
left sides are never under tension simultaneously when displacing the heart to expose
coronary targets. Gentle application of these techniques maintains stable hemodynamics
while providing excellent exposure. On rare occasions, bradycardia and cardiomegaly
may coexist and the resulting ventricular distention may hinder effective cardiac dis-
placement. Temporary epicardial atrial pacing may significantly reduce cardiac size and
improve target vessel exposure during OPCAB in this difficult circumstance.
SEQUENCE OF GRAFTING IN OPCAB

In off-pump coronary surgery, the chosen sequence of grafting is important to main-
tain hemodynamic stability and avoid critical ischemia. As a general rule, the col-
lateralized vessel(s) are grafted first and then reperfused by performing the proximal
anastomosis(es) or unclamping the internal mammary flow or connecting the perfusion-
assisted direct coronary artery bypass (PADCAB) apparatus. The last coronary target
grafted is the collateralizing vessel(s). This strategy obviates interrupting vital flow from
the collateralizing vessel(s) to the collateralized vessels until after the collateralized
vessel(s) have been grafted.
Performing the proximal anastomoses first makes estimation of graft length difficult.
At times the proximal anastomoses may be performed early in the operative sequence to
aid in early reperfusion of a collateralized vessel. If the LIMA → LAD graft must be
performed first, it may be necessary to leave a long mammary pedicle to avoid tension
on the LIMA anastomosis during subsequent cardiac displacement to expose lateral wall
targets.
A preferred sequence of grafting is as follows.
1. Perform the anastomosis to the completely occluded, collaterized vessel(s) first. The
collateralizing vessel may then be safely grafted. This strategy will minimize myocardial
ischemia.
2. The LIMA → LAD anastomosis should be performed first if the LAD is collateralized
or in cases of tight left main stenosis. This anastomosis is performed last when the LAD
is the collateralizing vessel.
3. The proximal anastomosis can be performed first or early after the distal anastomosis if
the target is a critical, collateralized vessel. This allows simultaneous perfusion during
the occlusion of the collateralizing vessel and minimizes overall myocardial ischemia.
4. Beware of the large proximally stenosed right coronary artery. The right coronary artery,
particularly if large and dominant, can cause significant problems when occluded during
OPCAB. Acute occlusion of a moderately stenotic right coronary artery may lead to
severe hemodynamic compromise due to bradycardia. The surgeon must be prepared to
use an intracoronary shunt or epicardial pacing to correct bradyarrhythmias promptly.
5. Beware of mitral regurgitation in OPCAB. Prolonged cardiac displacement combined
with mitral regurgitation may contribute to a downward hemodynamic spiral. Progres-
sive elevation of pulmonary pressures and increasing regurgitation on TEE may signal
impending right heart falure. Ischemic mitral regurgitation should be addressed early in
the procedure. This is accomplished by grafting and perfusing the culprit vessel respon-
sible for papillary muscle dysfunction.
6. Finally, graft sequence should be individualized, depending on anatomic patterns of
coronary occlusion and collateralization, myocardial contractility, atherosclerosis of the
ascending aorta, conduit availability, and graft geometry.
CARDIAC DISPLACEMENT AND PRESENTATION

OF CORONARY TARGETS
It is important to understand that the cardiac displacement techniques for exposure
of the inferior and lateral vessels are different. The lateral vessels are approached by
allowing the base of the heart to descend and rolling the apex of the heart under the right
sternal border. As mentioned, the right pleural cavity is opened and the traction sutures
on the right pericardium are released. The left-sided traction sutures are pulled up taut on
the retractor and the table is rotated sharply to the right to aid in rolling the heart into the
right chest. The deep stitch is pulled toward the patient’s left shoulder and secured to the
drapes. The Octopus III stabilizer (Medtronic, Minneapolis, MN) (see Chapter 6) is
mounted on the right side of the retractor and its arm reaches across the heart, both aiding
in presentation and accomplishing stabilization of the obtuse marginal coronary arteries.
A cotton sling may facilitate exposure of the marginal vessel in cases of cardiomegaly.
For the RCA and inferior wall vessels such as the PDA and LV branch or PLOM, the
deep stitch is pulled toward the feet and clamped to the drapes. The Octopus III stabilizer
is attached to the left limb of the OPCAB sternal retractor (“Octobase,” Medtronic,
Minneapolis, MN). The patient is placed in Trendelenburg’s position with the bed tilted
to the right. The base of the heart is elevated. The apex is oriented vertically.
In contrast to targets on the inferior and lateral walls, the anterior vessels (LAD and
diagonals) are exposed with very little manipulation of the heart. The deep stitch is
secured to the left drapes and the Octopus III brought onto the heart from the caudal aspect
or left limb of the retractor. Care is taken to divide the pericardium to allow the mammary
pedicle to fall posteriorly in the apex of the left chest.
CORONARY STABILIZATION AND GRAFTING

The Octopus III is a suction device, not a compression device. It is therefore possible
to achieve good tissue capture while applying the device at the mechanical median of the
cardiac cycle, rather than vigorously compressing the cardiac chambers. Thus, stabiliza-
tion is optimized, while mechanical interference with ventricular function is minimized.
Once the device is applied, a few seconds may be needed for the heart to recover. If
hemodynamics are compromised, the degree of compression should be reduced and the
mechanical median of the cardiac cycle should be more clearly identified by releasing the
knob of the Octupus arm while maintaining suction. The suction is maintained to avoid
losing tissue capture. After the appropriate position for the limb is determined, the
Octopus arm is tightened once more. The malleable pods on the Octopus III allow one
to spread the epicardium adjacent to the coronary targets, significantly improving visu-
alization of the vessel. The malleable pods may be bent up or down, or in a curve. They
may be bent or rotated independent of each other to accommodate irregular epicardial
surfaces.
After optimal exposure is obtained, a soft silastic vessel loop (Quest Medical, Allen,
TX) is placed around the target vessel for occlusion. The loop is placed proximal to the
chosen anastomatic site—never distally. It is wise to avoid entrance into the ventricle and
trauma to the epicardial veins with the vessel loops. When this occurs, a superficial
epicardial suture will reliably stop bleeding. The vessel loop may be directed out of the
surgeon’s field of view with the aid of a loose pericardial suture serving as a “pulley.”
Once the distal anastomosis is underway, it is critical for the anesthesia team to com-
municate continuously with the surgical team. Any changes in hemodynamics should be
addressed quickly. Bradyarrhythmias may be promptly and easily treated with atrial and/
or ventricular pacing. Occlusion of the right coronary artery proximal to the AV nodal
artery may cause bradycardia, which responds reliably to epicardial pacing.
The target vessel is opened with a coronary knife and the arterotomy is extended with
coronary scissors. The field is kept free of blood by dispersing the retrograde bleeding
with a humidified CO2 blower (DLP, Medtronic, Minneapolis, MN). The mister-blower
utilizes warm, humidified, pH-balanced fluid and carbon dioxide to clear the target site
of blood and help expose the intima of the coronary artery. It is important to blow on the
target only when placing the needle through the tissue, to minimize intimal trauma.
(Endothelial injury is discussed at length in Chapter 2.) Good visualization is critical for
a precise anastomosis. The intima of both the conduit and coronary artery is visualized
with each stitch. Optical magnification (3.5× loups), headlight, and Castro needle drivers
are used on all anastomoses. Each distal anastomosis is constructed with 8-0 monofila-
ment suture, to optimize precision, unless severe calcification requires a larger, stron-
ger needle.
An intracoronary (or aortocoronary) shunt may be placed if significant hemodynamic

compromise occurs due to ischemia after target vessel occlusion. The shunts (Medtronic,
Minneapolis, MN) range in size from 1.5 to 3.0 mm in 0.25-mm increments. These are
easily placed and removed. They are used infrequently, but are kept available in the room
for all cases. Intracoronary shunts may be particularly helpful with large right coronary
arteries (where bradyarrhythmias may occur), intramyocardial vessels (where placement
of an occlusive vessel loop may be hazardous), and with critical anatomy (where occlu-
sion of a key collateralizing vessel may be poorly tolerated). The coronary shunt is
removed prior to tying the suture on the distal and flow is reestablished. Air is allowed
to expel from the anastomosis prior to tying the suture. The conduit should be occluded
with an atraumatic bulldog clamp until the proximal anastomosis is performed, in order
to prevent retrograde bleeding and loss of coronary perfusion pressure.
At least two corporations have recently introduced cardiac positioning devices
(Expose™, Guidant and Starfish™, Medtronic), which use suction to attach to the apex
of the heart and can elevate and displace the heart to provide exposure of coronary targets
with little hemodynamic compromise. These devices are aids in exposure/presentation
and are not designed for coronary stabilization. They are used in conjunction with coro-
nary stabilizers, such as the Octupus III, and may facilitate OPCAB exposure, particu-
larly in cases of cardiomegaly and depressed left ventricular function.
PERFUSION-ASSISTED OPCAB
With experience and gentle application of the principles described above, virtually all
coronary vessels can be safely exposed, stabilized, and grafted during OPCAB. However,
the cumulative effect of sequential coronary occlusions can lead occasionally to a down-
ward spiral of hemodynamic stability. At times, it may be helpful to provide accessory
perfusion to the myocardium while other vessels are occluded. Perfusion-assisted direct
coronary artery bypass (PADCAB) allows for direct perfusion of myocardium subtended
by a coronary bypass target artery, either during performance of the distal anastomosis,
by means of an olive-tipped intracoronary catheter, or after completion of the distal
anastomosis, by providing controlled flow down the conduit. Inflow to the circuit and
pump is provided by a catheter placed in the ascending aorta or femoral artery. The Quest
Medical MPS (Quest Medical, Allen, TX) allows for exact control of coronary perfusion
pressure (3,4). Pharmacological additives and temperature control may accentuate its
protective effects. The coronary perfusion pressure during PADCAB is independent of
systemic pressure. This technique is especially helpful in collateralized targets, as coro-
nary flow may be driven through collaterals to supply adjacent myocardium. It is also
possible to measure and document graft patency and flows with the circuit. Multiple
grafts may be perfused simultaneously by use of a multilimbed perfusion set. It is impor-
tant to continue flow through all the grafts simultaneously, when the proximal anasto-
moses are performed. Each should be disconnected from the multilimb perfusion set
separately to perform its proximal anastomosis. PADCAB is used selectively by the
authors to minimize regional ischemia and improve myocardial protection in cases of
critical coronary anastomosis and profound cardiac dysfunction. Some practitioners use
PADCAB routinely, to optimize myocardial perfusion and hemodynamic stability for all
OPCAB cases.
Proximal anastomoses to the aorta are performed with an aortic partial occlusion
clamp. The systolic pressure is brought down to approx 90–95 mmHg before application
of the clamp. Once the clamp is applied, the aortotomies are created with a 4.0-mm aortic
punch. The application of the aortic clamp is guided by the results of epiaortic ultrasound
scanning as discussed above. Vein graft anastomoses are constructed with 5-0 or 6-0
monofilament suture and arterial grafts with 7-0 monofilament suture. Any graft taken
as a “T” off the IMA is anastomosed with 8-0 monofilament. Air is expelled by tying the
final suture after removing the cross-clamp. The vein grafts are kept occluded until
punctured with a 25-gauge needle to expel air. Arterial grafts are not punctured but are
allowed to backbleed prior to cross-clamp removal.
After completion and reperfusion of all grafts, protamine is administered (0.75–1.0
mg/kg) to correct the ACT to approx 150 s. As hemostasis is being achieved, three chest
tubes are routinely placed, one in each pleural space and one in the mediastinum. Tem-
porary pacing wires are used only if the patient requires epicardial pacing immediately
prior to chest closure.
The chest is closed in standard fashion. Nine sternal wires are used routinely to facili-
tate a tight sternal closure. Interrupted fascial sutures below the xiphoid ensure a tight
fascial closure. The subdermis and skin are closed with running absorbable sutures.
SUMMARY
As with other new techniques, there is a learning curve associated with OPCAB. With
persistence and patient attention to detail, this technique can be mastered and performed
reproducibly and reliably. OPCAB may show improved outcomes compared with
conventional CABG, especially in patients for whom cardiopulmonary bypass has
elevated risks. It is therefore important for all cardiac surgeons to be comfortable with this
technique.
REFERENCES
utilization in 200 consecutive off-pump coronary bypass patients. Ann Thorac Surg 2001;71(5):1477–
1483; discussion 1483–1484.
2. Murphy DA, Craver JM, Jones EL, et al. Surgical management of acute myocardial ischemia following
percutaneous transluminal coronary angioplasty. Role of the intra-aortic balloon pump. J Thorac
Cardiovasc Surg 1984;87(3):332–339.
3. Guyton RA, Thourani VH, Puskas JD, et al. Perfusion-assisted direct coronary artery bypass: selective
graft perfusion in off-pump cases. Ann Thorac Surg 2000;69(1):171–175.
4. Puskas JD, Vinten-Johansen J, Muraki S, Guyton RA. Myocardial protection for off-pump coronary
artery bypass surgery. Semin Thorac Cardiovasc Surg 2001;13(1):82–88.
Chapter 6 / The Medtronic Octopus System 75
6 Mechanical Stabilization
The Medtronic Octopus System
Michael Gibson, MD, Robert B. Beauford, MD,

and Daniel J. Goldstein, MD
CONTENTS
INTRODUCTION
DEVELOPMENT OF THE OCTOPUS TISSUE STABILIZER
HEMODYNAMIC CHANGES ENCOUNTERED WITH THE OCTOPUS TISSUE
STABILIZER
HISTOLOGICAL CHANGES ASSOCIATED WITH THE OCTOPUS TISSUE
STABILIZER
OCTOPUS OFF-PUMP CLINICAL EXPERIENCE
CONCLUSION
REFERENCES
INTRODUCTION
Worldwide, coronary artery bypass grafting (CABG) using cardiopulmonary bypass
(CPB) is performed in over 750,000 patients a year. However, the morbidity associated
with extracorporeal circulation and oxygenation is indisputable (1). This fact, combined
with developments in enabling technologies and early data suggesting improved coro-
nary surgery outcomes by avoiding extracorporeal circulation (2,3), has led to the resur-
gence of beating-heart or off-pump coronary artery bypass (OPCAB). The initial
skepticism over OPCAB hinged largely on the lack of good mechanical stabilization
technology for operating on the beating heart. Currently, several methods of immobili-
zation have been introduced to accomplish this task. Of the proposed mechanical, elec-
trical, and pharmacological stabilization techniques, the mechanical approach has proven
to be the most popular. Early experience with OPCAB relied on the use of sutures to

75
Fig. 1. The Medtronic Octopus 1 tissue stabilizer system. (Courtesy of Medtronic Cardiac Sur-
gery Technologies.)
stabilize the coronary target. At present several suction and pressure systems developed
by industry have entered the marketplace. Table 1 details the different mechanical methods.
Borst and colleagues (4) at Utrecht University Hospital in The Netherlands pioneered
the development of an immobilization device based on the principle of suction. Now
known as the Medtronic-Utrecht Octopus® Tissue Stabilization System, it uses a table-
mounted mechanical stabilizer combined with suction for epicardial fixation. This device
has evolved significantly since its introduction in 1995. This chapter deals with the
development of the Octopus, analysis and comparison of the different Octopus proto-
types, hemodynamic changes encountered with its use, histological effects of suction,
and clinical experience associated with its use. Two of the other leading stabilization
systems, the Genzyme OPCAB elite system, and the Guidant OPCAB system are dis-
cussed in great detail in Chapters 7 and 8.
DEVELOPMENT OF THE OCTOPUS TISSUE STABILIZER

The evolution of local immobilization devices began in 1994 with the utilization of
epicardial stay sutures in the porcine heart. This progressed into an epicardial ring sutured
into place. Ultimately, these early attempts at local stabilization proved to be cumber-
some and ineffective. Finally, attachment to the epicardium was achieved by suction—
and thus the birth of the Medtronic Octopus system. The early system consisted of two
prototypes, an “Encircling Octopus,” which surrounds the anastomosis circumferentially
and the “endo-Octopus,” which is placed on either side of the coronary artery.
The first-generation Octopus (Fig. 1) involved an adjustable table-mounted clamp
with an articulating arm with a “ball-and-sleeve” design and a tissue stabilizer. The tissue
Chapter 6 / The Medtronic Octopus System
Table 1
Cardiac Muscle Stabilizationa
Epicardial Stay sutures Pressure/friction
stay sutures + vessel loops Suction method technique
Movement reduction +/– + +++ ++
Presentation +/– + ++ +/–
77
Efficacy at all sites + + ++ +/–

Stability versus slipping + ++ ++ +/–
Ease of application +++ +++ + +/–
Learning curve – – ++ +
Potential damage to coronaries – + – –
Risk of bleeding – + – –
aAdapted (with permission) from ref. 1, p. 61.
77
Fig. 2. A pictograph of the Medtronic Octopus 2+ attached to the sternal retractor overlying the
heart. (Courtesy of Medtronic Cardiac Surgery Technologies.)
stabilizer had numerous configurations and a malleable arm for positioning. Minimiza-
tion of motion of the beating heart was considerable. However, there was still some
movement in the long axis and difficulty with exposure of the circumflex system. Its
bulkiness led to difficulty with manipulation and positioning. These concerns led to the
development of the Octopus 2+ tissue stabilizer (Fig. 2), which allowed increased motion
control at the anastomosis. Additionally, improved multivessel access and mounting was
obtained with the four-cup pod stabilizer, a stronger articulating arm, and an improved
retractor mount (instead of a table mount). To allow improved ease with coronary anas-
tomosis, the Octopus 2+ LP (low-profile) tissue stabilizer (Fig. 3) was developed with
a 23% decrease in the pod height.
Additional refinements led to the Octopus 3 (Fig. 4), which consists of malleable
stabilizer pods to allow better adaptability to patient heart contours. This leads to easier
Fig. 3. The Medtronic Octopus 2+ LP tissue stabilizer system. (Courtesy of Medtronic Cardiac
Surgery Technologies.)
positioning for exposure of the posterior heart. The Octopus 3 design with its malleable
stabilizer pods is manipulated with suction for improved anastomotic site exposure.
Table 2 compares and contrasts the different Octopus tissue stabilizer models. As incom-
plete revascularization in off-pump patients became a growing issue, so did questions
about access to the lateral and inferior walls of the heart. This, combined with growing
concerns about hemodynamic compromise associated with deep pericardial sutures, led
to the develpment of the Octopus Starfish™ heart postitioner (Fig. 5) as an adjunct to the
Octopus 3 stabilizer. This multiappendage suction device was designed to be connected
to the apex of the heart and greatly enhances exposure.
Further developments came with introduction of Octopus System II, which consists of
the Octopus 4 stabilizer (Fig. 6) and Starfish 2 positioner. Both now have a turret design
with 360° movement. This dramatically increases the arm reach and positioning possi-
bilities. Additionally, the devices are less obstructive owing to a smaller “whaletail” for
locking the position and a single vacuum line (versus two) on the Octopus 4.
Fig. 4. The Octopus 3 system (Courtesy of Medtronic Cardiac Surgery Technologies.)
HEMODYNAMIC CHANGES ENCOUNTERED

WITH THE OCTOPUS TISSUE STABILIZER
As the popularity of off-pump coronary revascularization has grown, more investiga-
tions have been carried out to prove its safety. Concerns have arisen regarding difficulty
in grafting the circumflex coronary artery territory and posterior surface, maintenance of
adequate blood pressure and forward output with cardiac manipulation, and development
of myocardial ischemia with cardiac manipulation. These hemodynamic changes, which
could potentially lead to the inability to perform OPCAB, have been evaluated using the
Octopus Tissue Stabilizer. Figure 7 and Table 3 demonstrate changes in hemodynamic
parameters with various maneuvers. These maneuvers include fixation with the Octopus,
vertical displacement of the heart, Trendelenburg 20° positioning, and return to baseline.
Hemodynamic changes associated with positioning included decreased stroke volume
(SV), increased heart rate (HR), and decreased cardiac output (CO) with vertical dis-
placement. These are thought to be secondary to biventricular pump failure, which
improves with Trendelenburg positioning at the expense of augmented left and right
ventricular preloads and an increased heart rate (5). Complete inability to obtain hemo-
Table 2
Comparison of Octopus Designs
Octopus
Circumflex
System Movement/stabilization Ease of grafting Anastomotic ease
1 a. Increased z motiona Difficult Difficult
b. Less stable due to long
articulating arm
2+ a. Decreased z motion Improved compared Improved

to Octopus 1
b. Improved stability
2+ LP a. Decreased z motion Improved compared Decreased pod height

to Octopus 1 allows increased ease
b. Improved stability
3 a. Improved contact with Improved compared Potentially difficult

the posterior heart to Octopus 1 due to increased y motion
b. Increased y motion
aMotion in the z direction relative to the x-y plane.
Fig. 5. The Octopus Starfish heart positioner.

Fig. 6. The Octopus 4 system.
dynamic stability, infrequent as it is, would preclude proceeding with beating-heart

surgery with the aid of the Octopus Tissue Stabilizer. It is critical to emphasize the
importance of an experienced and vigilant anesthesiologist who can continuously moni-
tor hemodynamic changes and who can judiciously administer volume, inotropic therapy,
and/or anti-ischemic medications, as well as modify bed positioning to attenuate these
hemodynamic changes.
HISTOLOGICAL CHANGES ASSOCIATED

WITH THE OCTOPUS TISSUE STABILIZER
Application of the Octopus Tissue Stabilizer uses 400 mmHg of suction distributed
over two pods with four 5-mm suction domes each. This applied suction consistently
leads to visible well-circumscribed epicardial hematomas located at the suction pods of
the Octopus. To assess their significance, if any, these epicardial hematomas underwent
histological evaluation (Fig. 8) by Borst and colleagues at 1–4 h, 48 h, and 6 wk post-
operatively (4). Concern for underlying coronary arterial vessel injury was addressed.
Table 4 characterizes the progression of these lesions.
These studies demonstrated no significant pathological abnormality created by the
suction pods of the Octopus tissue stabilizer. No injury to the endothelium or the media
could be identified, even in coronary or lymph vessels contained within the initial visible
lesions. Additionally, no mural thrombi were identified, proving that use of the Octopus
tissue stabilizer system with OPCAB could be undertaken without concern for epicardial
injury.OPCAB revealed initial concerns with hemodynamic instability with heart and
Fig. 7. Relative changes in hemodynamic parameters during vertical displacement of the beating
porcine heart by the Utrecht Octopus and the effect of head-down tilt (BASE = pericardial control
position; FIX = fixation of the suction tentacles to the posterior cardiac wall; DIS (1) = displace-
ment of the heart by the Octopus; TREND = Trendelenburg maneuver (20° head-down tilt); DIS
(2) = retracted heart with table returned to horizontal position and FREE = pericardial position after
release of the Octopus. (Statistical comparison with control values: *p < 0.05, **p < 0.01, #p <
0.001. (Reprinted with permission from the Society of Thoracic Surgeons [Ann Thorac Surg
1997;63:90–91]) (5).
Fig. 8.
84
volume; HR=heart rate; CO=cardiac output; MAP=mean arterial pressure; RVEDP=right ventricular end-diastolic
pressure; LVEDP=left ventricular end-diastolic pressure; MRAP=mean right atrial pressure; MLAP=mean left atrial pressure;
Beating
Adapted (with permission) from “changes in hemodynamic parameters with various maneuvers,” Ann Thorac Surg 1997; 63,S91 (5).
Freely
Heart
987

3.7
68
57
45
5
7
4
7
Displacement
Vertical
2.7
44
63
34
8
7
5
6
894
Changes in Hemodynamic Parameters with Various Maneuvers
Trendelenburg
Head Down
20-degree
3.7
62
61
49
11
11
931
7
9
Displacement
Table 3
Vertical
2.6
43
62
34
921
8
7
5
8
Fixation
Octopus
3.5
with
65
55
44
918
5
7
4
7
SVR=systemic vascular resistance.

Values
Basal
3.8
5
7
4
7
75
52
46
931
SVR (dyn s cm–5)
RVEDP (mmHg)
LVEDP (mmHg)
MRAP (mm Hg)
MLAP (mm Hg)
MAP (mmHg)
aSV=stroke
CO (L/min)
HR ( bpm)
Variablea
SV (mL)
Fig. 8. (facing page) Histology of suction lesions (A,B) Acute suction lesion induced by the
encircling Octopus. (A) After 45 min of suction on top of the LAD and its concomitant vein. Note
the hemorrhage in the upper part of the thick epicardial layer (E) and the absence of mural thrombi
in the vessels. M, myocardium. Magnification: 18X. (B) Detail of A, showing the intact layers of the
arterial wall close to the superficial hemorrhage of the LAD. Magnification 165X. (C,D) Suction
lesion 48 h after immobilization by the encircling Octopus. (C) The arrows point at the edge of the
myocardial lesion induced by 22 min of suction at the OM area. Magnification: 18X. (D) Detail of
C showing intramural coronary artery branch embedded in suction lesion consisting of interstitial
hemorrhage and infiltrate. Note the uninjured layers of the arterial wall. The myocytes show a mixed
continued on page 86
85
Table 4
Progression of Suction Induced Trauma
Post-op Visible lesion Hemorrhage Infiltrate Dead myocytes Coronary injury
1–4 h + + – – –
48 h + – +a +b –
6 wk – – – – –
aDensely cellular fibroblastic tissue consisting of cells with prominent nucleoli and exhibiting mitotic
figures.
bFew.
OCTOPUS OFF-PUMP CLINICAL EXPERIENCE

To assess the safety and efficacy of this unique stabilization system, we surveyed the
largest clinical experiences from the English literature. These results are summarized
in Table 5. Van Dijk et al. (6) randomized 281 patients to OPCAB and conventional
CABG. Their conversion rate was 1.4% and there were no reported device-related com-
plications. They concluded that omitting cardiopulmonary bypass led to reduced cardiac
enzyme release, reduced use of blood products, and a slightly shorter hospital stay. They
reported that, in selected patients, off-pump CABG is safe and yields a short-term cardiac
outcome comparable to that of conventional CABG. Hart et al. (7) conducted a retrospec-
tive study of all Octopus-supported OPCAB patients by the seven training institutions for
the device. The authors studied 1582 patients, of whom 44 (2.8%) were converted to
conventional CABG. They point out that only three (0.2%) were converted urgently; the
other 41 were converted “electively.” The authors ultimately concluded that Octopus-
supported OPCAB was a safe procedure with widening applicability. Spooner et al. (8)
reviewed a 2-yr three-institution experience with the Octopus system involving 456
patients. Conversion to CABG occurred in 13 of 469 patients (2.8%), four of which were
secondary to hemodynamic instability. They concluded that the Octopus devices were
safely applied with low morbidity. Whitman et al. (10) studied the uniform safety of
OPCAB using the Octopus system at three institutions among 239 patients. Among the
three institutions, 10 patients (4.0%) were converted. They concluded that stabilization
with the Octopus system provided predictable, reproducible immobilization with mini-
mal morbidity and decreased costs. Hart (11), in a separate later study, surveyed 230
multivessel Octopus-supported OPCAB patients. Four of these patients (1.7%) were
converted to conventional CABG and were subsequently excluded from analysis. Of the
remaining 226, Hart pointed out that all regions of the coronary circulation were acces-
continued from page 85

appearance of injured and normal cells. Magnification: 270X. (E,F) Suction lesion 48 h after immo-
bilization by the encircling Octopus. (E) Longitudinal section through the OM branch shows a patent
artery without mural thrombus within the lesion induced by 45 min of suction. E, epicardium. M,
myocardium. Magnification: 50X. (F) Detail of E, showing interstitial hemorrhage and infiltrate in
the epicardium adjacent to the artery. Note the absence of any injury to the arterial wall. Magnifica-
tion: 165X. (G,H) Suction site 6 wk after immobilization by the encircling Octopus. (G) Typical
example of tissue section at suction site (OM area) identified by intraoperative suture wire (arrow).
In spite of 25 min of immobilization, no fibrous tissue was found in the myocardium. Magnification:
18X. (H) Detail of G, showing normal myocytes but a fuzzy rather than well-demarcated transition
between myocardium (M). (Reproduced with permission from ref. (1).
Chapter 6 / The Medtronic Octopus System
Table 5
Octopus Off-Pump Clinical Experience
Authors Van Dijk et al. (6) Hart et al (7) Spooner et al (8) Whitman (10) Hart (11)b
Number of patients 142 1582 456 239 226

Study design Multicenter randomized Multicenter retrospective Multicenter retrospective Multicenter retrospective retrospective
prospective trial review review review review
Demographics
Mean age (yr) 61 63.9 NA 62.3 NA
Mean LVEF 77% nL 23% moda 55.4 NA 56.6 51
% diabetics 9 26.4 NA 26 30.1
Data
Mean no. of grafts/patient 2.4 2.31 1.9 1.8 2.7
Operating time (h) 4.2NA NA NA NA
Conversion rate % 7.7 Elective 2.6 / urgent 0.2 Elective 2.6 / urgent 0.2 4.11 .7
Intraoperative IABP % 0.7 0.4 NA 0
87
Results (%)d
Mortality (2.87) 0 1 0.32 0 0
Post-op MI (1.09) 3.5c 1.30.8 0 0.4
Atrial fib (19.37) 20 14.9 13.3 11 NA
Stroke (1.65) 0.7 0.6 0.2 0 0.4
Renal Failure (3.14) 0 0.9 NA NA 0.4
Reop for bleeding (2.32) 1.2 1 0.8 0
Follow-up
Mean follow-up 1 mo NA NA NA NA
Recurrent angina NA NA NA NA NA
Angioplasty 1.4 NA NA NA NA
Reop CABG 0 NA 1.3 NA NA
aNormal left ventricular function reported in 77% patients (109) “moderate left ventricular function” reported in the remaining 23%.
blater study done by Hart which included a cohort of new patients.
crate of MI is reported as outcome 1-month after surgery, specific details about the immediate perioperative period was unavailable.
dresult variables are displayed as a percent with STS benchmarks for all CABG patients in parenthesis.
eLVEF=left ventricular ejection fraction; IABP=intraaortic balloon pump; MI=myocardial infarction; fib=fibrillation; CABG=coronary artery bypass graft.
87
sible. He further concluded that the Octopus system is safe, efficacious, and applicable
in a wide range of patients.
Of the represented studies, there were no reported device-related complications. Fur-
thermore, when looking at end points of mortality, atrial fibrillation, stroke, postopera-
tive renal failure, and reoperation for bleeding, OPCAB with the aide of the Octopus
compares favorably with published STS benchmarks for all CABG patients (9).
CONCLUSION
Myocardial presentation and anastomotic site immobilization remain the most critical
factors influencing the accuracy of the off-pump anastomosis. Early clinical use of patient
positioning. Most authors acknowledge these perturbations but have shown them to be
safely and easily overcome. This requires a dedicated and vigilant anesthesia team,
judicious use of inotropes, and continuous echocardiographic and Swan–Ganz monitor-
ing. Other concerns involving mechanical trauma resulting from application of the
Medtronic Octopus tissue stabilizer system have been addressed by histopathological
examination, which has demonstrated that, even in the presence of visible lesions, there
exists no microscopic evidence of trauma to underlying tissue structures or blood vessels.
Finally, results of recent publications relating to complications and mortality compare
favorably with regard to Society of Thoracic Surgery benchmarks for all CABG patients.
In short, OPCAB utilizing the Medtronic Octopus tissue stabilizer system is a safe and
efficacious alternative to conventional coronary artery bypass grafting with CPB.
REFERENCES
1. Jansen Erik WL. Towards Minimally Invasive Coronary Artery Bypass Grafting. Utrecht, 1998.
2. Mack MJ. Perspectives on minimally invasive coronary artery surgery. Current assessment and future
directions. Int J Cardiol 1997;62(suppl 1):S73–S79
3. Bowles BJ, Lee JD, Dang CR, et al. Coronary artery bypass performed without the use of cardiopulmo-
nary bypass is associated with reduced cerebral microemboli and improved clinical results. Chest
2001;119(1):25–30.
4. Borst C, Jansen EWL, Tulleken CAF, et al. Coronary artery bypass grafting without interruption of
native coronary flow using a novel anastomosis site restraining device (‘Octopus®’). J Am Coll Cardiol
1996;27:1356–1364.
5. Grundeman PF, Borst C, van Herwaarden JA, et al. Hemodynamic changes during displacement of the
beating heart by the utrecht Octopus® method. Ann Thorac Surg 1997;63:S88–S92.
6. van Dijk, Nierich AP, Jansen EWL, et al. Early outcome after off-pump versus on-pump coronary bypass
surgery. Circulation 2001;104:1761–1766.
7. Hart JC, Spooner TH, Pym J, et al. A review of 1582 consecutive Octopus® off-pump coronary bypass
patients. Ann Thorac Surg 2000;70:1017–1020.
8. Spooner TH, Hart JC, Pym J. A two-year, three institution experience with the Medtronic Octopus®:
systematic off-pump surgery. Ann Thorac Surg 1999;68:1478–1481.
9. Society of Thoracic Surgeons National Database Benchmarks 1997 Coronary artery bypass only patients.
10. Whitman GJ, Hart JC, Crestanello JA, Spooner TH. Uniform safety of beating heart surgery using the
Octopus® tissue stabilization system. J Card Surg 1999 Sep–Oct;14(5):323–329.
11. Hart JC. Multivessel off-pump coronary bypass with the Octopus® experience in 226 patients. J Card
Surg 2000 jul–Aug;15(4):266–272.
Chapter 7 / The Genzyme-OPCAB Elite System 89
7 Mechanical Stabilization Systems

The Genzyme-OPCAB Elite System
William E. Cohn, MD and Marc Ruel, MD, MPH

CONTENTS
INTRODUCTION
THE GENZYME-OPCAB ELITE SYMSTEM AND IMMOBILIZER PLATFORM
EXPOSURE OF THE INFERIOR AND LATERAL WALLS
MECHANISM OF ADVERSE HEMODYNAMIC EFFECTS
EXPOSURE TECHNIQUES
CONCLUSION
REFERENCES
INTRODUCTION
Off-pump coronary artery bypass grafting (OPCAB) has been performed since the
early years of coronary surgery, with case series documenting its feasibility dating back
to the 1970s (1–4). Nevertheless, OPCAB only recently gained widespread acceptance
and entered the mainstream of clinical practice, propelled by a greater awareness of the
potential morbidity of cardiopulmonary bypass and aortic manipulation.
The introduction of self-retaining coronary stabilizers and the development of tech-
niques for their use are key factors that have led to the recent popularity of OPCAB. When
used properly, stabilization systems provide a relatively bloodless and motionless field
that allows for the construction of a technically accurate anastomosis. One such system
is the Genzyme-OPCAB Elite System (Genzyme Surgical, Fall River, MA), which we
have been involved with since its inception.
Although most surgeons can intuitively use stabilizers to accomplish off-pump
revascularization of the left anterior descending and diagonal coronary arteries, a more
systematic approach is required to successfully graft vessels on the inferior, posterior,

89
Fig. 1. Five of the 11 early prototypes of the Genzyme stabilizer are shown. Each was fashioned
from a soup spoon or ladle. The iteration on the far right most closely resembles the first commer-
cial product.
and lateral walls of the left ventricle. This chapter describes the mechanism of function
and detailed instructions for use of the Genzyme system, as well as a number of expo-
sure and stabilization techniques that have contributed to improving the feasibility, safety,
and popularity of multivessel OPCAB worldwide.
THE GENZYME-OPCAB ELITE SYSTEM

AND IMMOBILIZER PLATFORM
At the heart of the Genzyme Elite OPCAB system is the Immobilizer coronary stabi-
lizer. The Immobilizer utilizes a proprietary mechanism to effect coronary stabilization
and local coronary occlusion. This mechanism, called coronary capture, was developed
to limit residual motion at the anastomotic site and to minimize physical forces and
geometric distortion of the coronary required to achieve hemostasis.
The Immobilizer evolved from fairly humble beginnings. The first 11 prototypes were
cut from the bodies of spoons and spatulas purchased at a grocery store not far from
Boston’s Beth Israel and Deaconess hospitals (Fig. 1). Refinements in design resulted in
progressive improvement in function. After a number of iterations, the stabilizer was
taken from the animal lab to the operating room, where it performed well. After several
additional refinements, a licensing agreement was reached between the Beth Israel Hos-
pital and Genzyme Surgical to produce and market the stabilizer. Initially marketed under
the name “Cohn Cardiac Stabilizer,” it found immediate support in the cardiac surgical
community due to the extremely stable operative field it provided. Since its release in
January 1998, it has undergone further refinement, evolving into the current Immobilizer
platform (Fig. 2).
Fig. 2. Three iterations of the commercial stabilizer platform are shown. The Cohn Cardiac
Stabilizer on the left provided excellent stability, but the small window was felt to crowd the
anastomosis. It was soon followed by the “Large Mouth” Cohn, which provided more room. The
current platform, the Immobilizer, eliminates the need to pass the tapes through the center window,
allowing for faster setup.
The Immobilizer Platform and Coronary Capture

The Immobilizer is an ovoid plate in the center of which lies a rectangular foramen or
window. At the proximal and distal ends of the window are plastic buttresses that project
slightly below the plane of the plate. Each buttress is flanked by a pair of cleats that allow
a silicone elastic tape to be affixed under tension. At one end of the plate is a detachable
segment that facilitates removal of the Immobilizer after graft completion. At the oppo-
site end is a pivot that connects the Immobilizer to the articulating arm (Fig. 3).
The coronary capture mechanism integrated into the design of the Immobilizer plat-
form results in compression of the coronary between a single taut segment of silicone
elastic tape and a flat plastic surface. As a result, intimate contact is maintained between
the epicardial surface and the stabilizer resulting in improved stability. More important,
the flattening of the coronary results in gentle coronary occlusion. This method of coro-
nary occlusion is advantageous from several perspectives. As the coronary is compressed
between two flat structures, coronary distortion is minimized. Whereas many techniques
require circumferential snaring and coronary strangulation to effect occlusion, the
biplanar compression effected by the Immobilizer is more analogous to that seen with
atraumatic vascular clamps (Fig. 4). This is especially important if the coronary segment
being snared is diseased, as circumferential snaring of an atherosclerotic vessel is more
likely to result in plaque rupture and vascular injury (Fig. 5). Biplanar compression may
result in less distortion of posterior or hemicircumferential plaque. Furthermore, because
the silicone elastic tapes are pulled to the sides, only a small fraction of the tape tension,
that portion that is perpendicular to the stabilizer, is transmitted to the coronary, decreas-
ing the risk of coronary injury (Fig. 6).
Many find that the capture technology utilized by the Genzyme Immobilizer, when
used correctly, provides improved anastomotic site stability. This is due in part to the fact
Fig. 3. The five key features of the Immobilizer Platform are (A) the anastomotic window, (B) the
proximal and distal buttresses, (C) the paired flanking tape cleats, (D) the detachable segment, and
(E) the metal pivot.
Fig. 4. When using the Immobilizer, the coronary is occluded by biplanar compression between
the flat undersurface of the plastic buttress and a straight segment of silicone elastic tape.
that the Immobilizer stabilizes the coronary segment directly, rather than indirectly by
stabilizing the adjacent epicardium. It is well recognized by OPCAB surgeons that
excessive downward pressure on the heart surface paradoxically compromises rather
than improves stability. The upward force created by the silicone elastic tapes ensures
intimate contact between the textured bottom surface of the Immobilizer and the epicar-
dium, despite the use of nominal downward pressure. This allows stability to be opti-
mized and prevents stabilizer slippage during grafting.
Fig. 5. Circumferential strangulation has the potential to injure delicate coronary arteries, espe-
cially if applied to a diseased vessel segment.
Fig. 6. A simple vector analysis illustrates that only a fraction of the tape tension A is transmitted
to the coronary artery through the normal vector B. The maximum tape tension A is limited by the
tape cleat, which will hold only 2 N.
Immobilizer Use
Prior to positioning the Immobilizer, silicone elastic tapes are placed deep to the target
coronary artery approx 2.0 cm apart and as perpendicular as possible with respect to the
long axis of the vessel. Accurate spacing is essential to achieve good hemostasis (Fig. 7).
Although the needle on the tapes is intentionally blunt, care is taken to avoid injury to
septal perforators and coronary veins and to ensure that the tapes pass deep to the artery.
Fig. 7. Placing the tapes symmetrically with respect to the long axis of the coronary is essential
so that the target site remains centered in the anastomotic window. Spacing the tapes 2 cm apart
facilitates occlusion of the coronary when the tapes are tightened by ensuring that the tapes and
proximal and distal buttresses are aligned.
Creating a defect in the epicardium prior to passing the needle allows passage with less
force and greater tactile sense, minimizing the risk of vascular injury. If resistance is
encountered during this maneuver, the needle is withdrawn and redirected until a resis-
tance-free path is identified. A careful review of the preoperative cineangiogram allows
one to identify the location of large septal perforators so that they are not included in the
segment isolated between proximal and distal tapes, which can result in troublesome
backbleeding.
The Immobilizer is then positioned over the target coronary artery such that the
intended anastomotic site is centered in the rectangular window, and the articulating
green arm is locked into position. The base of the articulating arm locks securely to a
metal rail that runs along the entire length of both sternal blades, as well as the rack of
the Elite OPCAB retractor. Like other systems, the stabilizer arm and retractor have been
designed to accommodate an infinite number of positions and configurations (Fig. 8).
The Elite retractor is heavier and more robust than most other commercially available
retractors, which many feel contributes to the stability of the system. A single locking
mechanism secures both the flexible arm configuration and the Immobilizer pivot.
After the Immobilizer is locked in position, the tapes are placed under tension and
wedged in opposing cleats. Once this is accomplished, the artery is compressed against
the posterior aspect of the proximal and distal buttresses, effecting temporary coronary
occlusion and isolation, as described above.
The decision of whether to use a single proximal tape or proximal and distal flanking
tapes is complex and is in part determined by the surgeon’s preference. Generally, com-
pletely occluded vessels that fill by distal collaterals are more likely to have brisk
backbleeding if only a proximal tape is used. Similarly, the rich septal collaterals between
the LAD and PDA vessels generally make flanking tapes a good idea. Furthermore, it
should be noted that often stability is improved with flanking tapes cleated under tension.
As some surgeons prefer not to manipulate the coronary distal to the anastomotic site
unless necessary, however, a good rule of thumb is to place flanking tapes, and cleat one
or both only as needed for hemostasis.
Fig. 8. The articulating flexible arm and the Immobilizer pivot are locked securely in position by
tightening a single mechanism. The flexible arm can be attached to either blade of the retractor A,
B as well as to the rack C.
The Immobilizer Platform and Intracoronary Shunts

In the early evolution of OPCAB, placement of intracoronary shunts was felt to be a
necessary step to avoid ischemia and hemodynamic instability while the distal anastomo-
sis was being constructed. With increased experience, there has been a trend away from
shunting, as it has been shown to be superfluous in the overwhelming majority of grafts.
Nevertheless, there remain occasional situations where shunting may be prudent. These
include the large dominant RCA with a moderate stenosis that must be grafted proximal
to the origin of the PDA branch, the LAD with moderate stenosis that must be grafted in
the proximal half of the vessel, and any situation where a large coronary with moderate
stenosis must be grafted prior to revascularization of an adjacent collateral dependent
territory. Although intracoronary shunts can be readily employed with the Genzyme
Immobilizer, in this setting we prefer the use of aortocoronary shunts. Aortocoronary
shunts consist of an arterial source catheter, generally a no. 5 French cannula inserted in
the ascending aorta, hooked by large-bore iv tubing to a silicone olive-tipped coronary
catheter. The Immobilizer is positioned as described earlier, but only a proximal tape is
used. The appropriately sized olive-tipped catheter is inserted into the distal coronary
lumen providing flow. Although there is a moderate amount of line resistance in this
arrangement, the blood delivered to the coronary distal to the anastomotic site is much
greater than that seen with traditional intracoronary shunts, which convey the already
compromised proximal flow through a resistive tube. As aortocoronary shunts are linear
rather than “T”-shaped, and extend only down the distal vessel, they are easier to remove
after completion of the anastomosis (Fig. 9).
EXPOSURE OF THE INFERIOR AND LATERAL WALLS

FOR PERFORMING MULTIVESSEL OPCAB
Clearly, one of the most important developments in the evolution of multivessel
OPCAB has been the techniques for exposing branches of the circumflex and right
coronary arteries on a beating heart. Understanding the mechanisms of hemodynamic
compromise during manipulation of the heart is essential in learning to perform OPCAB.
Fig. 9. A short segment of plastic tubing A conveys arterial blood from a no. 5 French cannula
placed in the ascending aorta through a Nylon olive-tipped cannula B to the distal LAD.
These mechanisms are stabilizer-independent, and can be seen with pressure-type and
epicardial suction-type stabilizers, as well as with the Genzyme Immobilizer.
MECHANISMS OF ADVERSE HEMODYNAMIC EFFECTS

Left ventricular compression, right ventricular compression, and coronary insuffi-
ciency are responsible for the majority of adverse hemodynamic effects resulting from
exposure and stabilization of the heart during OPCAB (5–7). These mechanisms are
reviewed below.
Left Ventricle
The application of pressure with the coronary stabilizer during exposure of the circum-
flex system constrains the lateral wall of the left ventricle, leading to diastolic dysfunction
and decreased left ventricular end-diastolic volume (LVEPV), stroke volume, and cardiac
output (7). Although volume loading and Trendelenburg positioning can attenuate these
effects by elevating left ventricular filling pressures, these measures may also exacerbate
intraoperative third-space fluid sequestration. The use of short-acting α-adrenergic agents
can help in maintaining perfusion pressure in this setting, but may be associated with an
increased risk of perioperative mesenteric ischemia (8). Furthermore, mitral regurgita-
tion can also occur during exposure of the lateral wall, possibly related to deformation
of subvalvular structures in combination with coronary insufficiency. Exposure tech-

niques that minimize or eliminate left ventricular deformation during OPCAB are there-
fore desirable.
Right Ventricle
Exposure of the lateral wall of the left ventricle displaces the heart toward the right,
resulting in right ventricular compression against the pericardium and right hemisternum.
The relatively low pressure in the right ventricle makes it particularly vulnerable to
deformation and to reduction in right ventricular end-diastolic volume (RVEDV) (6,9),
with decreased right ventricular stroke volume leading to poor left ventricular filling and
a drop in cardiac output. Volume loading and Trendelenburg positioning can only par-
tially compensate for this effect, and the recourse to intraoperative right ventricular assist
devices has been proposed as palliation for right ventricular compromise (10,11); pref-
erable, however, are exposure techniques that minimize right ventricular deformation
and render the use of these costly devices unnecessary.
Coronary Insufficiency
Decompensated coronary disease, coronary snaring, and reduced cardiac output dur-
ing heart manipulation can also lead to hemodynamic instability, which is best prevented
by planning a revascularization strategy that minimizes myocardial ischemia and by
using exposure techniques that maintain coronary perfusion pressure at all times. Useful
axioms include the revascularization of occluded or severely stenosed coronary arteries
first, the construction of proximal anastomoses after each distal anastomosis that
involves a free graft, the availability of temporary pacing during snaring of a moderately
stenosed right coronary artery, the constant optimization of afterload and coronary per-
fusion pressure by the anesthesiologist, and the selective use of intracoronary shunts,
aortocoronary shunts, or assisted coronary perfusion devices such as the PADCAB™
system (12). The use of an intra-aortic balloon pump can also prove useful in select cases (13).
EXPOSURE TECHNIQUES
Depending on the size and shape of the heart and the dimensions of the chest cavity,
visualization of the proximal obtuse marginal and posterolateral branches of the circum-
flex system may be obstructed by the left hemisternum. Although intuitive attempts at
improving exposure may involve displacing the coronary stabilizer’s arm toward the
right (thereby increasing its application pressure), this maneuver usually leads to hemo-
dynamic problems as well as paradoxically poor stabilization. Maximized access and
stability at the anastomotic site with minimal compromise of cardiac performance is most
readily achieved when the stabilizer is applied with little or no pressure; thus, coronary
stabilizers should be used only to stabilize the anastomotic field rather than assist at
retracting the heart. Exposure is best obtained with maneuvers such as the use of deep
pericardial sutures, right pleurotomy and pericardiotomy, right hemisternal elevation,
and apical suction devices, each of which is described below.
Deep Pericardial Sutures

One of the most important advances in exposure techniques for OPCAB has been the
use of deep pericardial sutures. These sutures, when placed under tension, create a ridge
Fig. 10. Placement of deep pericardial sutures results in rotation of the heart to an “apex up”
orientation. Often the heart protrudes above the plane of the chest wall in this position.
of pericardium that supports the base of the lateral left ventricle adjacent to the atrioven-
tricular groove and allows the heart to be rotated rightward to assume an “apex-up”
position (Fig. 10). In this subluxed position, the apex of the heart points toward the ceiling
and protrudes through the sternotomy incision, often in a plane above that of the sternal
retractor. This generally allows for adequate exposure of the lateral and inferior aspects
of the left ventricle before applying the coronary stabilizer.
Placement of deep pericardial sutures may vary between surgeons and according to the
patient’s anatomy. Generally, one or two 2-0 silk sutures are placed in the pericardium
posterior to the left phrenic nerve immediately anterior to one or both left pulmonary
veins. Two additional sutures are placed deep in the oblique sinus behind the left atrium,
and to the left and posterior to the inferior vena cava (Fig. 11). Care must be taken when
placing these sutures to avoid injury to underlying structures such as the esophagus and
lung (14). The suture in the oblique sinus is particularly important to obtain good expo-
sure of the lateral wall near the base of the heart. One commonly used technique, called
the “stockinet,” consists of placing a single deep pericardial suture in this location to
secure the midpoint of a 50-cm gauze strip deep in the oblique sinus. Subsequent traction
on the two ends of the strip can then be adjusted to optimize exposure of different surfaces
of the heart.
Regardless of the technique used, deep pericardial traction sutures generally allow for
presentation of the lateral, inferior, and even posterior wall of the heart with little change
in left ventricular geometry or LVEDV, providing adequate access for multivessel graft-
ing while maintaining hemodynamic stability. In some patients, however, additional
maneuvers such as those outlined below are necessary to accomplish this aim effectively.
Right Pleurotomy and Pericardiotomy

Right pleurotomy and right vertical pericardiotomy constitute helpful technical
adjuncts in patients with difficult lateral wall exposure. Human pericardium is quite
Fig. 11. Deep pericardial sutures are placed posterior to the left phrenic nerve adjacent to the left
upper and lower pulmonary veins A, behind the left atrium B, and posterior and to the left of the
inferior vena cava C. The operative photo shows the heart being lifted and rotated to the right to
allow suture placement.
flexible but very inelastic. This inelasticity accounts for the profound hemodynamic
effects caused by acute tamponade from a relatively small volume of intrapericardial
fluid. As such, the posterior pericardium, right lateral pericardium, and diaphragmatic
pericardium therefore constitute a fixed-volume cusp or pocket. It is into this pocket that
the right ventricle is compressed during extreme rightward rotation of the heart. By
opening the right pleura widely and incising the pericardium, the pocket is vented, allow-
ing the heart to herniate into the right pleural space while maintaining RVEDV.
Right lateral pericardiotomy can be performed with a right vertical incision extending
from the cut edge of the initial anterior pericardiotomy, 2 cm cephalad and parallel to the
diaphragm, down to the level of the inferior vena cava with care taken at avoiding injury
to the right phrenic nerve (Fig. 12). The 2-cm rim of pericardium on the diaphragm
facilitates closure of the pericardiotomy once grafting is complete. In some circum-
stances, one may also benefit from removing the right pericardiophrenic fat pad, which
can be quite large, and decreasing tidal volume to provide additional room for the easily
deformed right ventricle. Caution should be taken when measuring right-sided grafts, as
closure of the lateral pericardial incision may affect their lie.
Many surgeons skilled in the OPCAB procedure avoid right pleurotomy and lateral
pericardiotomy on the contention that incising the lateral pericardium and entering an
additional body cavity is inconsistent with the objective of decreased invasiveness. While
many cases can indeed be performed successfully without it, this maneuver facilitates the
reproducible performance of a precise anastomosis during multivessel OPCAB. Further-
more, right pleurotomy does not leave a scar, nor is it associated with significant morbid-
ity or additional length of stay after coronary surgery.
Right Hemi-Sternal Elevation

Asymmetric right hemisternal elevation can further improve lateral wall exposure
when used in conjunction with the techniques described above. Once the right lateral
Fig. 12. The pericardial incision is made along the dotted lines shown. The drawing on the right
shows the resulting vertical incision in the right lateral pericardium and the widely opened right
pleural space.
Fig. 13. After creating a right vertical pericardiotomy, opening the right pleural space, and elevat-
ing the right sternum, the heart can be readily rotated. The operative photo on the right shows the
lateral wall exposure that results. The tip of the forceps is on the proximal aspect of the obtuse
marginal branch, where it emerges from the fat in the atrioventricular groove. The left atrial
appendage A is in the center of the operative field, whereas the apex of the heart B is under the right
hemisternum.
pericardium and right pleura are incised, it is often the right half of the sternum and right
blade of the retractor that limits rightward displacement of the subluxed apex of the heart.
By elevating these structures, the apex can clear the posterior aspect of the chest wall.
This allows the entire heart to rotate into the right pleural space with little change in left
and right ventricular geometry or end-diastolic volume (Fig. 13). As a result of these
maneuvers, the proximal obtuse marginal and posterolateral branches are brought toward
the center of the operative field and into the surgeon’s view.
Fig. 14. The HMD (Heart Manipulation Device) when used in conjunction with other maneuvers
described, provides quick and reliable access to every aspect of the heart with little compromise
in hemodynamic stability.
Right hemisternal elevation is best accomplished if carried out as the sternotomy

incision is being performed. By applying anterior traction on the right chest wall when
the retractor is first spread, costal microfractures that inevitably form during sternal
opening will occur predominantly in the right-sided ribs. The resulting increase in flex-
ibility of the right chest wall will subsequently prevent the need for exaggerated tilting
of the sternal retractor, which is often required to overcome the left chest wall’s tendency
to rise after left internal mammary artery harvest. The diaphragmatic insertion on the
inferior aspect of the right hemisternum can also be released to facilitate right hemisternal
elevation and create additional room for displacement of the apex of the heart into the
right pleural space.
Apical Suction Devices

Self-retaining apical suction retractors represent a welcome addition to the operative
armamentarium of OPCAB surgeons. The Genzyme HMD (Heart Manipulation Device)
was recently introduced to work in concert with the Genzyme-OPCAB Elite System to
position the heart for exposure of its lateral and posterior aspects (Fig. 14). Although the
use of these devices involves supplemental expenses, the easy affixation of the suc-
tion cup to the cardiac apex or adjacent inferior or lateral wall and the rapid locking
into a chosen position make them an attractive adjunct for challenging patients. Apical
suction can be used alone or in conjunction with the other exposure techniques described
above. The HMD suction head is smaller and lower profile than other commercially
available devices, and can be placed in close proximity to the Immobilizer platform. It
is often most effective when placed on the heart wall being grafted halfway between the
apex of the heart and the anastomotic site. For smaller hearts, application to the apex may
suffice. Like the Immobilizer arm, the HMD can attach to the rack or either blade of the
Elite retractor, and can assume a wide range of configurations. When used in conjunction
with right hemisternal elevation, wide right pleurotomy, and right vertical pericardiotomy,
the HMD provides excellent access to every aspect of the beating heart with little hemo-
dynamic compromise (Fig. 14).
CONCLUSION
Introduction of the Immobilizer, recent changes in the stabilizer arm, and most recently,
introduction of the Genzyme Heart Manipulation Device, have resulted in continued
improvement in the Genzyme-OPCAB Elite System from both performance and ease-
of-use perspectives. It is likely that continued improvements in tools and exposure tech-
niques over the next several years will make multivessel OPCAB an increasingly
attractive approach for the next generation of cardiothoracic surgeons.
REFERENCES
1. Ankeney JL. To use or not to use the pump oxygenator in coronary bypass operations. Ann Thorac Surg
1975;19:108–109.
2. Trapp WJ, Bisarya R. Placement of coronary artery bypass graft without pump oxygenator. Ann Thorac
Surg 1975;19:1–9.
3. Benetti FG, Naselli G, Wood M, Geffner L. Direct myocardial revascularization without extracorporeal
circulation: experience in 700 patients. Chest 1991;100:312–316.
4. Buffolo E, Andrade JCS, Branco JNR, Aguiar LF, Teles CA, Gomes WJ. Coronary artery bypass
grafting without cardiopulmonary bypass. Ann Thorac Surg 1992;61:63–66.
5. Nierich AP, Diephuis J, Jansen EW, Borst C, Knape JT. Heart displacement during off-pump CABG:
how well is it tolerated? Ann Thorac Surg 2000;70(2):466–472.
6. Mathison M, Edgerton JR, Horswell JL, Akin JJ, Mack MJ. Analysis of hemodynamic changes during
beating heart surgical procedures. Ann Thorac Surg 2000;70(4):1355–1360; discussion 1360–1361.
7. Biswas S, Clements F, Diodato L, Hughes GC, Landolfo K. Changes in systolic and diastolic function
during multivessel off-pump coronary bypass grafting. Eur J Cardiothorac Surg 2001;20(5):913–917.
8. Reilly PM, Wilkins KB, Fuh KC, Haglund U, Bulkley GB. The mesenteric hemodynamic response to
circulatory shock: an overview. Shock 2001;15(5):329–343.
9. Grundeman PF, Borst C, Verlaan CW, Meijburg H, Moues CM, Jansen EW. Exposure of circumflex
branches in the tilted, beating porcine heart: echocardiographic evidence of right ventricular deforma-
tion and the effect of right or left heart bypass. J Thorac Cardiovasc Surg 1999;118(2):316–323.
10. Lima LE, Jatene F, Buffolo E, et al. A multicenter initial clinical experience with right heart support and
beating heart coronary surgery. Heart Surg Forum 2001;4(1):60–64.
11. Sharony R, Autschbach R, Porat E, et al. Right heart support during off-pump coronary artery bypass
surgery—a multi-center study. Heart Surg Forum 2002;5(1):13–16.
12. Guyton RA, Thourani VH, Puskas JD, et al. Perfusion-assisted direct coronary artery bypass: selective
graft perfusion in off-pump cases. Ann Thorac Surg 2000;69(1):171–175.
13. Kim KB, Lim C, Ahn H, Yang JK. Intraaortic balloon pump therapy facilitates posterior vessel off-pump
coronary artery bypass grafting in high-risk patients. Ann Thorac Surg 2001;71(6):1964–1968.
14. Zamvar V, Deglurkar I, Abdullah F, Khan NU. Bleeding from the lung surface: a unique complication
of off-pump CABG operation. Heart Surg Forum 2001;4(2):172–173.
Chapter 8 / The Guidant OPCAB System 105
8 Mechanical Stabilization Systems

The Guidant OPCAB System
Marc W. Connolly, MD
and Valavanur A. Subramanian, MD
CONTENTS
INTRODUCTION
GOAL OF OPCAB SURGERY
TECHNICAL CONSIDERATIONS AND STRATEGIES
INSTITUTIONAL OPCAB EXPERIENCE
CONCLUSION
REFERENCES
INTRODUCTION
In January 1997, the Guidant Corporation Cardiac Surgery Group introduced the first
widely used beating-heart stabilizer system for MIDCAB procedures. From the experi-
ence gained since these initial MIDCAB procedures (1,2), the Access MV OPCAB
System™ was introduced in 1998 for multivessel, sternotomy off-pump CABG, leading
to greater surgeon acceptance of the OPCAB procedure and to the development of further
enabling technologies and devices. Introduced in 2000, the latest Guidant OPCAB Sys-
tem (Axius™ Stabilizer and Xpose™ Apical Positioning Device) has greatly assisted the
surgeon in providing improved hemodynamic stability and access to lateral wall circum-
flex vessels—increasing the overall comfort level of the surgeon.
Presently, at Lenox Hill Hospital, 98% of all CABG procedures are performed off-
pump, with the 1–2% on-pump cases performed for deep intramyocardial left anterior
descending vessels and severe diffuse distal disease requiring extensive endarterecto-
mies. This chapter outlines our OPCAB operative technique with the use of Guidant
Axius Stabilizer with the Xpose Apical Positioning Device System, and briefly, discuss
our OPCAB hospital outcomes.

105
GOAL OF OPCAB SURGERY

The primary goal of OPCAB is to obtain complete revascularization using standard
anastomotic techniques to all critically stenosed coronary arteries without the use of
cardiopulmonary bypass (CPB). To avoid the use of CPB and its potential adverse
perioperative complications, hemodynamic stability needs to be maintained and coro-
nary ischemia avoided or minimized throughout the entire procedure. A consistent, well-
planned operative strategy is implemented to achieve this goal.
TECHNICAL CONSIDERATIONS AND STRATEGIES

Cardiac Anesthesia
All patients are monitored intraoperatively with arterial line pressure, Swan–Ganz
catheter with mixed venous oximetry, central body temperature, foley urimetry, and
transesophageal echocardiography.
Intravenous neosynephrine and nitroglycerin are most commonly used for pharmaco-
logical support. Occasionally, Primacor (Sanofi–Synthelabo) is used in patients with
poor left ventricular function, elevated pulmonary pressures, and/or hemodynamically
compromising acute ischemia. Rarely, levophed and IABP are used to support hemody-
namics.
Coronary Artery Anatomy

Coronary artery anatomy strategies have been discussed in detail previously (3). To
avoid intraoperative ischemia, the sequence of grafting, coronary shunting, and perfusion
of completed bypass grafts by performing proximals first or with the use of perfusion
cannulae should be well thought out before skin incision. An individual patient’s coro-
nary anatomy and lesions, left ventricular function, and regional wall motion abnormali-
ties should be taken into preoperative consideration when planning a patient’s
intraoperative strategy to avoid ischemia and hemodynamic compromise, leading to
conversion to cardiopulmonary bypass support.
Sternotomy and Pericardiotomy

The skin incision is made from two to three finger breadths below the sternal notch to
the inferior tip of the xiphoid. Sternotomy is performed in usual fashion. The Guidant
OPCAB retractor is placed and the sternum is opened, manually elevating the right side
to increase room for placement of the right ventricle during positioning for lateral wall
exposure.
The pericardium is incised with electrocautery completely along the diaphragm, then
superiorly with the superior mediastinal fat pad to the innominate vein. The right peri-
cardium is further loosened by incising its diaphragmatic attachments toward the vena
cava, carefully avoiding the phrenic nerve. Also, the pericardial attachment along the
right ascending aorta near the innominate artery is incised to increase laxity of the right
pericardium. Having a loose, movable right pericardium is critical to avoid right ventricu-
lar compression and hemodynamic instability when performing lateral and posterior wall
coronary artery anastomoses. Infrequently, the right pleura is opened further to increase
the space under the right sternum to displace the right heart during positioning.
Pericardial sutures are placed only along the anterior edge of the left side, never on the
right. Pulling up on these left pericardial sutures can usually expose the left anterior
descending without having to place deep pericardial sutures or use the Xpose.
Fig. 1. Anterior positioning with Xpose apical device and LAD stabilization with Axius suction
foot. (Used by permission of Guidant Corporation, Santa Clara, CA.)
The Guidant OPCAB retractor provides excellent mediastinal exposure. The retractor’s
curved sternal blades pull the heart anteriorly toward the surgeon as the sternum is
opened. The numerous suture guides along both sides provide for placement of silastic
tapes, coronary artery traction sutures, and pericardial sutures, thus avoiding unnecessary
clamps. A deep large blade retractor is also available for obese patients.
Heart Displacement
The Xpose Apical Positioning Device has greatly improved exposure to all vessels
while maintaining hemodynamic stability. The Xpose Apical suction device provides
excellent access to the anterior (Fig. 1), anterolateral (Fig. 2), lateral (Fig. 3), posterior
(Fig. 4), and inferior vessels (Figs. 5 and 6). In large hearts, the suction cup can even be
placed on the lateral apical wall to gain access to circumflex vessels near the AV groove,
or on the right ventricle to expose the left anterior descending or distal right coronary artery.
The most important aspect of heart positioning and displacement is to prevent com-
pression and decreased filling of the right ventricle. Elongating the heart by pulling out
and upward with the Xpose improves right ventricular and left ventricular preload and
function during displacement. As discussed previously, a loose right-sided pericardium
or opening into the right pleural space helps avoid right ventricular compression and
hemodynamic deterioration.
In over 600 cases using Xpose, we have not experienced any avulsion of the epicar-
dium, but bruising or an epicardial “hickey” is common without adverse sequelae. Dullum
Fig. 2. Anterolateral positioning to access intermediate ramus and anterior obtuse marginal vessels
with antegrade placement of mechanical Axius stabilizer. (Used by permission of Guidant Cor-
poration, Santa Clara, CA.)
Fig. 3. Lateral exposure of circumflex vessels using Xpose apical device with antegrade placement
of Axius suction stabilizer. (Used by permission of Guidant Corporation, Santa Clara, CA.)
Fig. 4. Posterior exposure to posterior lateral ascending or posterior obtuse marginal vessels using
Xpose apical position device and Axius suction stabilizer. Notice apex of heart pulled anteriorly
and cephalad. (Used by permission of Guidant Corporation, Santa Clara, CA.)
Fig. 5. Inferior exposure of posterior descending artery with Xpose apical device and retrograde
placement of Axius suction stabilizer. Apex of heart pulled anterior and cephalad. (Used by
permission of Guidant Corporation, Santa Clara, CA.)
Fig. 6. Inferior exposure with Xpose apical device and antegrade placement of Axius mechani-
cal stabilizing foot. (Used by permission of Guidant Corporation, Santa Clara, CA.)
and Resano (4) describe their initial experience with the Xpose in five case reports,
successfully gaining access to lateral and inferior wall vessels, while maintaining hemo-
dynamic stability. We avoid using the Xpose on the apex in very small, female hearts with
large, wrap-around left anterior descending coronary arteries. In these instances,
compression and occlusion of the coronary artery can occur, producing ischemia and
hemodynamic compromise. In two patients where this was experienced, the Xpose devise
was moved laterally away from the left anterior descending and the procedures were
completed uneventfully.
Vessel Stabilization
The Guidant Axius Off-Pump System has various stabilizing technologies available
to suit the particular needs and comfort level of the individual surgeon. The Axius sta-
bilizing arm has link technology for accurate placement of the stabilizing foot without a
protruding stabilizing arm extending above the sternal plane. The Axius stabilizer can be
placed anywhere on the OPCAB retractor to gain access to the desired coronary artery
and is available in two types: suction (Fig. 1) or mechanical (Fig. 2). The foot can be
placed retrograde (Fig. 5) or antegrade (Fig. 6) to the vessel, depending on the surgeon’s
preference. An earlier-generation Access Ultima™ Stabilizing System with an offset
mechanical foot and curved fluted shaft is also available, and is still preferred by some
surgeons.
We encourage surgeons to try all available Guidant stabilizing systems with the Xpose
Apical Positioner for the various vessel exposures to find what system provides him or
her the best access and stabilization with consistent hemodynamic stability. The melding
of an individual surgeon with specific available technology produces a comfort level for
consistent, reproducible results.
Proximal Anastomosis
Proximal anastomoses in a deep chest with a hyperdynamic heart that produces move-
ment of the aorta can be more difficult than distal anastomoses. To prevent loss of aortic
vascular control with the partial occlusion clamp and potential aortic complications, great
care should be taken when performing the proximal anastomoses.
The adventitial layer between the aorta and main pulmonary artery is widely separated
to allow for proper placement of the partial occlusion clamp. The anesthesiologist is
asked to maintain the systolic arterial pressure between 90 and 110 mmHg before place-
ment of the partial occlusion clamp. If the aorta and partial occlusion clamp are moving
too much for proper suture placement, a supporting towel, lap pad, or the hand of the
assistant can stabilize the clamp while suturing. The aortic clamp is gently removed after
completion of proximal anastomoses.
Proximal anastomoses can be performed before distals to reperfuse coronary beds as
coronary anastomoses are completed to prevent ischemia. Proximal anastomotic devices
may also be used in diseased ascending aortas, thereby avoiding partial clamping.
LENOX HILL OPCAB EXPERIENCE

Beginning in May 1999, Lenox Hill Hospital began a dedicated OPCAB program. All
eligible multivessel CABG patients underwent unselected, intention-to-treat OPCAB
procedures. Our experience gained from the use of the Access Ultima OPCAB System
(Guidant, Santa Clara, CA), made available in May 1999, provided a comfort level and
confidence in approaching our CABG patients in this intention-to-treat model. The
development of the Axius and Xpose systems has assisted in decreasing our conversion
rate to cardiopulmonary bypass.
Betweem April 1998 and January 2002, over 1300 patients underwent multivessel full
sternotomy OPCAB procedures. The mean number of grafts was 3.22. Conversion to
CPB occurred in 3.1% of patients over the entire series, improved to 1.4% in the last
6 mo. The most common etiology of conversion to CPB was hemodynamic ischemic
deterioration and presence of deep intramyocardial left anterior descending arteries.
A retrospectively comparison to conventional CABG performed in previous years at
our institution demonstrated that OPCAB patients had a significantly less overall
perioperative complication rate (10% vs 18%, p < 0.0001), deep sternal infection (1.4%
vs 3.2%, p < 0.02), reoperation for bleeding (2.2% vs 6.3%, p < 0.0001), hospital mor-
tality (1.6% vs 3.4%), and stroke rate (1.1% vs 3.1%, p < 0.004). Length of stay was also
significantly decreased (6.1 d vs 7.2 d, p < 0.01). The perioperative myocardial infarction
rate was similar in both groups (0.8% OPCAB vs 1.1% conventional).
CONCLUSION
With over 1300 OPCAB and 500 MIDCAB procedures performed at Lenox Hill, the
Guidant Off-Pump Vessel Stabilizing Systems have enabled us to perform presently over
98% of our CABG procedures without cardiopulmonary bypass. Hemodynamic stability
and excellent vessel stabilization can be achieved in the majority of multivessel OPCAB
patients, allowing for proper anastomotic suture placement and achievement of complete
revascularization.
Further development in enabling off-pump technologies are needed to obtain even
greater access and stabilization to deep proximal obtuse marginal circumflex vessels and
to prevent conversion to cardiopulmonary bypass. As technology develops, greater sur-
geon acceptance will increase and CABG patient outcomes will improve. Anastomotic
devices will most likely play an important part in this future development.
REFERENCES
1. Subramanian V, Sani G, Benetti F, et al. Minimally invasive coronary artery bypass surgery: a multi-
center report of preliminary clinical experience. Circulation 1995;92(suppl):645.
2. Calafiore A, DiGiammarco G, Teodori G, et al. Left anterior descending coronary artery grafting via left
anterior thoractomy without cardiopulmonary bypass. Ann Thorac Surg 1996;61:1658–1665.
3. Connolly M, Subramanian V, Patel N. Multivessel coronary artery bypass grafting without cardiopul-
monary bypass. Op Tech Thoracic 2000;5:166–175.
4. Dullum M, Resano F. Xpose™: A new device that provides reproducible and easy access for multivessel
beating heart bypass grafts. Heart Surg Forum 2000;3(2):113–118.
Chapter 9 / The MIDCAB Operation 113
9 The MIDCAB Operation
Mercedes K. C. Dullum, MD
and Albert J. Pfister, MD
CONTENTS
DEFINITION
HISTORY
INDICATIONS AND CONTRAINDICATIONS
SURGICAL TECHNIQUE
POSTOPERATIVE RESULTS
REFERENCES
DEFINITION
For purposes of this discussion, a minimally invasive direct coronary artery bypass
(MIDCAB) is considered to be any surgical coronary revascularization performed on the
beating heart (“off-pump”) through an incision other than a median sternotomy.
HISTORY
Interestingly, the initial attempts at surgical myocardial revascularization were per-
formed on the beating heart via a thoracotomy. Vineberg (1) in 1946 implanted the
internal mammary artery directly into the cardiac muscle. In 1967, Kolessov (2) reported
direct coronary revascularization of the LAD and obtuse marginal via a thoracotomy
without cardiopulmonary bypass. Sabiston (3) is often credited with performing the first
coronary artery bypass graft of a saphenous vein graft to the right coronary artery via a
median sternotomy. The ease of this incision and the excellent exposure that it afforded
quickly made median sternotomy the incision of choice for coronary revascularization.
Other surgeons including Favaloro, Garrett, Trapp, Bysana, and Ankeny (4–7) continued
to demonstrate the safety of off-pump coronary bypass.
The improvements in cardiopulmonary bypass technology combined with the safety
and efficacy of cardioplegia in the 1960s changed the approach to coronary bypass. The

113
ability to perform distal anastomoses in a bloodless field on a motionless target dramati-

cally increased the reproducibility of coronary artery bypass grafting (CABG). Coronary
bypass via median sternotomy done on cardiopulmonary bypass with cardioplegic arrest
became (and some would argue still is) the gold standard for coronary revascularization.
In the 1980s, Benetti (8) in Argentina began a resurgence of off-pump coronary
surgery, reporting his early experiences in 1985 and performing his first left anterior
thoracotomy for CABG in 1988. He reported the results of his series in 1990 (9). Pfister
(10) in 1992 demonstrated the safety of off-pump bypass surgery (most via median
sternotomy) in the United States. MIDCAB was introduced to the United States in 1994,
and in 1995 Subramanian and colleagues (11) reported the first MIDCAB multicenter
study of over 150 patients.
Following these early reports, great momentum and enthusiasm for learning and per-
forming MIDCABs ensued. After this early expansion in the late 1990s, interest in the
procedure reached a plateau for several reasons: (1) the limited applicability, because the
procedure was used only for single-vessel bypass; (2) the steep learning curve required
to become skillful at the operation; (3) the development of off-pump multivessel bypass
techniques via median sternotomy; and (4) improved percutaneous therapies.
INDICATIONS AND CONTRAINDICATIONS

The surgeon’s decision to perform the MIDCAB procedure should depend on his or
her belief that it is the procedure of choice to produce the best clinical outcome for the
patient. The procedure should be tailored to the patient and not vice versa.
Several patient populations can be considered for MIDCAB approaches. First are
patients with isolated LAD and/or diagonal lesions. These usually represent restenosis
following transcatheter therapy or patients with complex lesions not suitable for percu-
taneous revascularization. Second are patients with more than single-vessel disease and
a contraindication to sternotomy in whom the non-LAD lesion is to be treated percuta-
neously, a so-called hybrid approach (12,13) (see Chapter 15). Third is the group of
patients who can be adequately revascularized via a nonsternotomy approach and who
have a high risk of complications resulting from cardiopulmonary bypass. These would
include patients with a recent or high risk for stroke, uncontrolled diabetes, renal insuf-
ficiency, hematological disorders, advanced age, lung disease, calcified ascending aor-
tas, redos, and patients whose religious conviction preclude transfusion. Finally, patient
preference must be considered, for example, those patients who wish to return to manual
labor or strenuous physical activity earlier. MIDCAB is an excellent alternative in patients
who require a redo CABG to the LAD only, whether the LIMA has been used previously
or not. In the latter case a “homegrown LIMA” can be made using a radial artery or
segment of saphenous vein with proximal inflow via the axillary artery (14–16) (as will
be explained below). A MIDCAB via a posterolateral thoracotomy can be performed on
patients who are redos with patent LIMA to LAD and who require grafting of the lateral
wall (17).
Contraindications for this procedure are based on a variety of anatomical, technical,
and clinical considerations. We consider the MIDCAB operation (via anterior thorac-
otomy) to be a single-vessel bypass to the LAD. Initially, our group attempted to perform
multivessel bypass via an anterior thoracotomy, but found that it was less painful for the
patient and more expeditious to perform multivessel bypass through a median sterno-
tomy. Therefore, we consider multivessel disease to be a relative contraindication for

MIDCAB.
Patients with deep chest walls or breasts too large to accommodate the instruments for
access, and those with small, deeply intramyocardial and/or calcified target vessels should
also be considered unsuitable candidates for MIDCAB. Patients with pulmonary insuf-
ficiency who are unable to tolerate single-lung ventilation may represent a relative con-
traindication to MIDCAB. While it is possible to perform the operation with both lungs
being ventilated, it can be tedious and time-consuming. Chronic obstructive pulmonary
disease (COPD) by itself is not a contraindication to MIDCAB. Interestingly, COPD
frequently makes it easier to harvest the LIMA and to access the heart, since the COPD
enlarges the anterior–posterior dimension of the chest cavity. The ideal first patient for
MIDCAB has the following characteristics: single-vessel coronary artery disease repre-
sented by total occlusion of the LAD with good collaterals; a 2 mm or greater diameter
distal vessel that is epicardial and not calcified; normal ventricular function; a thin chest
wall; and mild to moderate COPD. We believe that patients with decompensated heart
failure or recalcitrant arrhythmias represent contraindications to MIDCAB. Clearly,
access is much better through a sternotomy in these situations. Finally, conversion from
a MIDCAB approach to a sternotomy, either on or off bypass, is rare, but not unheard of.
Situations that require conversion from a thoracotomy approach include patients with an
inaccessible LAD (deeply intramyocardial, or technically difficult to graft on the beating
heart), inadequate conduit, uncontrolled arrhythmias, or hemodynamic compromise.
SURGICAL TECHNIQUE
Different variations of the MIDCAB procedure are used, depending on the patient’s
specific need for revascularization. These various approaches include: (1) the left ante-
rior thoracotomy performed through the fourth or fifth intercostal space (standard
MIDCAB); (2) the xiphoid approach performed through the xiphisternum; (3) the left
anterior thoracotomy combined with an infraclavicular incision for axillary artery
inflow; (4) the endoscopic approach; (5) the posterior lateral thoracotomy, particularly
in redo patients and those who need circumflex grafting only; and (6) the transabdominal
approach as described by Subramanian (18). Regardless of approach, it is of utmost
importance to have an anesthesiologist who is proficient with anesthesia for off-pump
surgery and can maintain a comfortable, safe working environment for the surgeon. Each
of these approaches will be discussed separately.
Left Anterior Thoracotomy

The patient is positioned supine on the table. In our experience, we initially elevated
the left chest with a rolled sheet, but have now abandoned this, as with current instrumen-
tation it is easier to harvest the left internal mammary artery. External defibrillator pads
are placed, particularly in redo patients. The positioning of these pads should allow for
a sternotomy access. Routine coronary artery bypass graft prepping and draping is per-
formed with exposure of the entire left the chest laterally over to the axilla. A 6- to
10-cm incision is made over the fourth intercostal space (ICS) (Fig. 1). This incision is
usually just below the nipple, but may be placed above in asthenic men. The incision is
started two fingerbreadths lateral to the midline. In women, performing the incision
Fig. 1. The incision is performed at the fourth intercostal space, below the nipple.
1–2 cm below the inframammary crease allows for better healing, as there is not constant
friction and maceration at the incision site. The pleural space is entered and the LIMA
retractor (Guidant Corporation, Cupertino, CA) is inserted. A double-lumen endotra-
cheal tube or bronchial blocker is used to collapse the ipsilateral lung; however, if nec-
essary, the lung can be packed out of the way with warm lap packs. Care is taken to
mobilize any pleural or cardiac adhesions (particularly in redos) prior to insertion and
opening of the retractor, to prevent inadvertent damage to these organs. Dissection of the
internal mammary artery under direct vision is the same as through the sternotomy except
that it is from a mirror-image lateral location. Identification of the internal mammary
artery medially is important prior to extensive dissection or traction, due to the possibility
of injuring it before the pedicle has been mobilized. The endothoracic fascia is scored and
mobilized. The collateral branches to the intercostal vessels are identified, clipped, and
divided with electrocautery or scissors. The harmonic scalpel can also be used for the
dissection. Mobilizing a short length of the LIMA pedicle without too much traction is
important at the beginning of the dissection to prevent avulsion of the branches. The
LIMA should be mobilized proximally to the subclavian vein and distally to one inter-
costal space below the incision (Fig. 2). A LIMA pedicle length of approx 15–17 cm is
required for a tension-free anastomosis. The highest intercostal branch of the LIMA
should be clipped and divided to prevent coronary steal. This phenomenon has seldom
been seen in our experience; however, if identified in the postoperative period, the inter-
Fig. 2. The endothoracic fascia is incised and the LIMA dissected. The pedicle is freed proximally
to the subclavian vein and to one intercostal space distal to the incision.
costal branch can be embolized percutaneously. The distal end of the internal mammary
artery is freed and can be clipped after heparin has been administered. To prevent damage
to the LIMA, we prefer harvesting the internal mammary artery as a pedicled rather than
a skeletonized graft. Identification of the LAD can be performed prior to the LIMA
dissection by opening the pericardium before starting the dissection. The pericardium is
opened over the pulmonary artery and should be dissected distally to the apex and proxi-
mally to the mediastinal tissue. It is important to have studied the angiogram carefully so
that the diagonal branches can be used as landmarks to locate the LAD and prevent
inadvertently grafting the wrong vessel. The LAD can usually be identified as it emerges
lateral to the pulmonary artery and courses medially to the apex. It behooves the surgeon
to take a few moments to identify the LAD correctly, as ventricular enlargement or
adhesions can alter the cardiac position and rotate the LAD out of view. Tension on the
pericardium allows the surgeon to reposition the heart and bring the LAD into the surgical
site. We currently use the Guidant MIDCAB system with separate retractors for the
LIMA harvest and anastomosis. The LIMA retractor should be opened slowly, as this
reduces the risk of rib fracture and pain. Once the LIMA has been harvested, the patient
is systemically heparinized (3 mg/kg) and the LIMA is divided distally and prepared for
grafting. Pericardial traction sutures are placed pulling the upper ones toward the left
shoulder and the lower ones laterally, as this rotates the heart into the midline of the
incision. The compression stabilizer is placed and the immobility of the graft site and
presence of stable hemodynamics are confirmed prior to proceeding. Proximal occlusion
Fig. 3. The anastomosis is performed with local stabilization and proximal occlusion. A shunt can
be used if necessary. The LAD is rotated into view with pericardial sutures.
of the LAD is obtained by passing a pledgeted silastic tape around the vessel (Fig. 3). If
deemed necessary, a shunt may be placed. Preischemic conditioning is not routinely
performed, but the anesthesiologist is notified prior to occlusion of the vessel. The LAD
is then opened and the anastomosis is performed with running 7 or 8 prolene. The use of
anastomotic clips may facilitate creation of this anastomosis. Once the anastomosis is
completed, the LAD and LIMA should be flushed prior to tying the last suture or placing
the last clip. We recommend tacking the LIMA pedicle to the epicardium. The pedicle
should be attached medially to the mediastinal tissue without tension. It is necessary to
reexpand the lung under direct vision to ensure that the lung rises over and not under the
LIMA pedicle, so as to prevent the risk of avulsion of the pedicle. The heparin is fully
reversed with protamine. Once hemostasis is adequate, the mediastinal fat pad is brought
back over to cover the anastomosis. One Blake drain is placed in the left pleural space.
Local anesthetic block in the intercostal spaces is performed with marcaine. The incision
is closed in the usual fashion, first with pericostal sutures and then completing the other
layers of tissue.
Left Anterior Thoracotomy with Axillary Artery Inflow

In patients in whom the LIMA cannot be used, i.e., in which the LIMA has been
damaged in the harvest, or in redo patients with a previously used and occluded LIMA,
or for LIMAs that have sustained radiation damage, a radial artery or saphenous vein can
be used with axillary artery inflow. In these patients, once the LAD is identified and found
Fig. 4. The axillary artery is accessed through an infraclavicular incision. Once the proximal end
of the graft is anastomosed to the axillary artery, it is tunneled in to the chest along the mediastinum
to the LAD.
to be graftable, a segment of radial artery or saphenous vein is harvested. The axillary

artery is then accessed through an infraclavicular incision. The proximal end of the
conduit is then sewn to the axillary artery, then brought into the chest through the pectoral
fascia in the first or second intercostal space. The conduit is then brought to the LAD with
care to tack it to the mediastinal tissue to prevent avulsion from the lung (Fig. 4). The rest
of the operation follows as described above for the left anterior thoracotomy.
T MIDCAB or H Graft
A variation of the left anterior thoracotomy MIDCAB known as the (T MIDCAB) or
the (H Graft) has been described by Cohn, Caulson, and Wolfe (19–21). In these cases,
a limited anterior thoracotomy is performed and a short length of the LIMA is exposed.
A segment of radial artery graft is then anastomosed between the LIMA and LAD. This
allows for limited dissection, specifically in high-risk patients or those in whom dissec-
tion is difficult. Favorable results have been reported with this technique, but it is impor-
tant to ligate the LIMA distally to prevent coronary steal.
Xiphoid MIDCAB
Calafiore has further validated the “standard “ MIDCAB approach with the results of
his limited-access small thoracotomy (LAST) operation (22). It is important to note that
patients have experienced more early postoperative pain than originally expected. Alter-
natively, another variation of the original MIDCAB is one in which a xiphoid approach
Fig. 5. A 7- to 8-cm incision is made over the xiphoid.
is taken to the distal LIMA and LAD (23,24). The xiphoid MIDCAB appears to be a
simpler, less painful approach than that through a left anterior thoracotomy.
A vertical midline 6- to 8-cm skin incision is made over the xiphoid (Fig. 5). The
xiphoid process is then divided and the incision is extended into the lower sternum for
one or two interspaces if necessary. A Rultract® mammary artery retractor is used to
elevate the left side of the costal margin and the distal LIMA is identified. All efforts are
made to leave the pleura intact during the dissection. The LIMA is harvested at a level
proximal to the third to fourth interspace (Fig. 6). Heparin (10,000 units) is administered
and the mammary artery is divided distally and prepared for grafting.
The pericardium is then opened and the edges retracted with sutures. These sutures
allow the heart to be elevated and rotated medially with the assistance of a saline-filled
glove placed in the pericardial well (Fig. 7). The LAD is then stabilized with a mechanical
stabilizer and an arteriotomy is performed. Proximal occlusion may be obtained by the
use of a pledgeted silastic tape surrounding the proximal LAD, or, alternatively, an
appropriately sized intracoronary shunt is used to allow distal perfusion during the anas-
tomosis (Fig. 8). Distal coronary occlusion is never used. A bloodless field is maintained
with the use of a CO2 blower. The LIMA-to-LAD anastomosis is performed with running
7-0 Prolene® and the pedicle is tacked to the epicardium (Fig. 9). The ventricle is then
allowed to fall back into the pericardial well, and a trough is cut in the pericardium to
Fig. 6. The LIMA is harvested proximally to the third or fourth ICS.
Fig. 7. The LAD is brought into view with the glove and pericardial sutures.
Fig. 8. The vessel is stabilized for the anastomosis.
Fig. 9. Completed xiphoid MIDCAB anastomosis.

prevent kinking of the LIMA pedicle. A small soft drain is then placed in the pericardial
well extending into the pleura (if it is open). Patients are usually extubated in the oper-
ating room and are kept in the intensive care unit overnight.
In a recent report of xiphoid MIDCAB (24), 10 patients underwent LIMA to LAD
grafting by this approach. Mean age was 73 yr (range 52–86). There were no reops, and
most patients had multiple comorbidities with Parsonnet scores ranging from 3 to 41.5
(25). Three patients underwent hybrid percutaneous intervention on the first postopera-
tive day. Postoperative length of stay ranged from 3 d to 29 d, with the older and hybrid
patients requiring longer stays. There were no deaths, but 2 of the hybrid patients had
major complications resulting in a prolonged hospital stay. The first had pericardial
tamponade and low cardiac output after receiving heparin on the first postoperative day
for the placement of an intracoronary stent. This complication required reexploration and
drainage. After the initial experience with this patient who suffered delayed tamponade,
a soft drain was left in the pericardial well until the day after percutaneous revascularization
in all subsequent hybrid patients. The second patient also had bleeding and low cardiac
output after percutaneous intervention, requiring placement of an intra-aortic balloon pump.
Postoperative angiograms in the hybrid patients showed that the LIMA-to-LAD anas-
tomoses were patent. Transthoracic Doppler evaluation was done in three nonhybrid
patients and showed good diastolic augmentation, suggesting anastomotic patency. In
this initial group of patients reported there were no mortalities and the postoperative
complications were not compounded by the xiphoid incision.
Xiphoid MIDCAB is a safe and effective approach for grafting the LIMA to the LAD.
The proximity of the distal LAD and distal LIMA to the left costal margin makes the
anatomy favorable for this approach. By avoiding chest wall incisions and intercostal
retraction, pain is lessened. We believe that the xiphoid approach to isolated LIMA-to-
LAD anastomosis deserves further pursuit with controlled studies, comparing the mor-
bidity and mortality from this procedure to alternative approaches for hybrid or isolated
surgical revascularization.
Left Posterior Lateral Thoracotomy MIDCAB

The left lateral thoracotomy is the best approach when grafting vessels of the circum-
flex system (18). It is a safe approach in redos, especially with patent LIMAs. The patient
should be positioned to allow for femoral cannulation if necessary. External defibrillation
pads should also be placed prior to starting the procedure, and the radial artery or saphe-
nous vein conduits may need to be harvested before placing the patient in the right lateral
decubitus position. A 10-cm thoracotomy is performed and the left lung is freed, mobi-
lizing all adhesions and releasing the inferior pulmonary ligament up to the pulmonary
vein. The circumflex vessel can be identified once the pericardium is opened, and can
frequently be found by following the previous graft down to the target coronary vessel.
The conduit can then be anastomosed to the descending thoracic aorta after hepariniza-
tion. One note of caution with this proximal anastomosis is that there may be significant
atherosclerotic disease in the descending aorta, which could affect the inflow. Proximal
occlusion of the coronary artery is then obtained and the distal anastomosis can be
performed. The compression stabilizers work best for this approach. Usually a shorter
length of conduit will be needed for grafting the circumflex system through a left poste-
rolateral thoracotomy rather than through a median sternotomy. The graft should be
brought underneath the pulmonary ligament and measured prior to grafting, being careful
Fig. 10. The proximal end of the graft is attached to the descending thoracic aorta. The conduit is
the passed below the inferior pulmonary ligament and anastomosed to the coronary artery. (Cour-
tesy of Guidant Corporation, modification of original illustration by Michael Morejohn.)
to prevent kinking once the lung is reexpanded (Fig. 10). A pleural drainage tube is placed
and then the chest closed in the usual fashion.
Endoscopic MIDCAB
Mack reported on the early use of thoracoscopically assisted coronary artery surgery
for harvesting of the LIMA (26). Other reports described thoracoscopically harvesting
both LIMAs and then performing bilateral anterior small thoracotomy incisions for graft-
ing the LAD and RCA systems (27). The use of video assistance for minimally invasive
coronary bypass has enjoyed great success as shown by Wolf (28), and has advanced at
many centers with excellent results.
This field has now progressed to robotic assistance, “RAVECAB,” as described by
Boyd in Ontario (29,30), and TECAB, performed by Diegler’s group in Germany (31,32).
Closed-chest totally endoscopic coronary bypass is no longer a fantasy and has now
evolved into a reality with good early results (33,34). Endoscopic and robotic surgery are
described in detail in other chapters of this book.
POSTOPERATIVE RESULTS
Pain management and the development of multivessel beating-heart surgery have
restricted the growth of conventional MIDCAB. The use of thoracic blocks, epidurals,
and intrathecal morphine has not had the beneficial results expected in MIDCAB surgery,
and many patients still experience early pain. However, these patients usually recover
more quickly than sternotomy patients and are able to resume regular activities sooner.
The majority of patients who have MIDCABs are extubated in the operating room or soon
after in the intensive care unit. Early and late clinical outcomes have been excellent, and
the angiographic results in these patients have shown durable patency. Calafiore (35)
reported on 540 patients who were scheduled to undergo LIMA to LAD using his left
anterior small thoracotomy (LAST) approach. In 5.2% of the patients, the LAST opera-
tion could not be performed for technical reasons. Thirty-day results included blood
transfusions in 3.3%, reoperation for bleeding in 1.9%, atrial fibrillation in 9%; peri-
carditis in 1.2%, and delayed chest wound healing 2.9%. There was a 30-d 1% mortality
and 1.2% late mortality. A total of 26 patients (5.1%) required reoperation, 3.5% within
30 d, and 1.6% later than 30 days. Three patients required repeat angioplasty, one patient
within 30 d after the MIDCAB and two thereafter. Calafiore found that with increasing
experience and improved instrumentation, the incidence of reoperation after his initial
group decreased in his last 100 patients. Reoperation was required in one patient due to
conduit occlusion. Mack and Magovern reported on a group of 103 patients undergoing
MIDCAB operations from December 1996 through December 1997 (36). Most of these
patients (97%) underwent angiography, with 99% patency and 91% with Fitz-Gibbons
grade A patency (37). There were two operative noncardiac deaths, and in both patients
patent grafts were documented. Three patients who underwent intraoperative angiogra-
phy had graphs revised with improved flow. Incidence of reoperation for bleeding, stroke,
and atrial fibrillation were 3.9%, 9%, and 1.7%, respectively. Interestingly, early angiog-
raphy after MIDCAB has demonstrated that early stenosis (not occlusion) often resolves
within 6–10 mo on angiographic follow-up. A recent report by Diegler and his group (38)
showed early and 6-mo patency rates of the LIMA-to-LAD anastomosis of 97% and 95%,
respectively, with a 3% reintervention rate. Mehran et al. (39) reported on our single-
center experience with MIDCABs in 274 patients. The mortality was lower in the latter
174 patients than in the initial 100; however, this was not statistically significant. There
was a 4% occurrence of major adverse cardiac events in the initial group and 1% in the
last group. At 1 yr, the reintervention rate was approximately 2.9% and the actuarial
survival rate was 98% in both groups. In the year after surgery, none of the patients who
had angiographic follow-up in the study showed a restenosis of the arterial bypass conduit.
Undoubtedly, there is a learning curve associated with the MIDCAB operation.
Despite this, it has been shown that this procedure can be performed with excellent
clinical results and, in an effort to tailor the bypass operation to our patients, MIDCAB
should be part of the armamentarium of every coronary surgeon.
REFERENCES
1. Vineberg AM. Development of anastomosis between coronary vessels and transplanted mammary
artery. Med Assoc J 1954;71:594.
2. Kolessov VI. Mammary artery-coronary anastomosis as method of treatment for angina pectoris.
3. Sabiston DC. The coronary circulation. Johns Hopkins Med J 1974:314.
4. Ankeny JL. To use or not to use the pump oxygenator in coronary bypass operations. Ann Thorac Surg
1975;19:108–109.
5. Favoloro RG, et al. Current status of coronary artery bypass graft (CABG) surgery. Semin Thorac
6. Garrett HE, Dennid EW, DeBakey ME. Aorto-coronary bypass with saphenous vein graft:seven year
follow up. JAMA 1973;223:792–794.
7. Trapp WG, Bisarya R. Replacement of coronary artery bypass graft without pump oxygenator. Ann
Thorac Surg 1975;19:1–8.
8. Benetti FJ. Direct coronary surgery with saphenous vein bypass without either cardiopulmonary bypass
or cardiac arrest. J Cardiovasc Surg 1985;26:217–222.
9. Benetti FJ, Nasellli G, Wood M, Geffner L. Direct myocardial revascularization without extracorporeal
circulation. Experience in 700 patients. Chest 1991;100:312–316.
10. Pfister AJ, Zaki MS, Garcia JM, et al. Coronary artery bypass without cardiopulmonary bypass. Ann
Thorac Surg 1992;54:1085–1091.
11. Subramanian VA, Sani G, Benetti FJ, Calafiore AM. Minimally invasive coronary bypass surgery: a
multicenter report of preliminary clinical experience. Circulation 1995(suppl 8);92:S1645.
12. Cohen HA, Zenati M. Integrated coronary revascularization J Invasive Cardiol 1999;11(3):184–191.
13. Farahat F, Depuydt F, VanPraet F, et al. Hybrid cardiac revascularization using a totally closed-chest
robotic technology and a percutaneous transluminal coronary dilatation. Heart Surg Forum 2000;9403.
14. Bonatti J, Coulson A, Bakhshay S, Posch L, Sloan T. The subclavian and axillary arteries as inflow
vessels for coronary artery bypass grafts—combined experience from three cardiac centers. Heart Surg
Forum 2000;741718.
15. Flege JB, Wolf RK, Minimally invasive axillary—coronary artery bypass. Heart Surg Forum
2000;3(3):238–240.
16. Shabb B, Khalil I. Minimally invasive axillary–LAD saphenous vein bypass. Heart Surg Forum
1999;2(3):254–255.
17. Baumgartner FJ, Gheissari A, Panagiotides GP, et al. Off-pump obtuse marginal grafting with local
stabilization: thoracotomy approach. Ann Thorac Surg 1999(Sep);68(3):946–948.
(MIDCAB). Eur J Cardiothorac Surg 2000;17(4):485–487.
19. Cohn W, Suen H, Weintraub R, Johnson R. The “H” graft: an alternative approach for performing
minimally invasive direct coronary artery bypass. J Thorac Cardiovasc Surg 1998;115:148–151.
20. Coulson A, Bakshay S. The “T-MIDCAB” procedure. Use of extension grafts from the undisturbed
LIMA in high risk patients. Heart Surg Forum 1998;1(1):54–59.
21. Miyaji K, Wolf R, Flege J. Minimally invasive direct coronary artery bypass using H graft for pleural
symphysis. Ann Thorac Surg 1999;68(1):234–235.
22. Calafiore AM, DeGiammarco GD, Teodori G, et al. Left anterior descending coronary artery grafting
via left anterior small thoracotomy without cardiopulmonary bypass. Ann Thorac Surg 1996;61:
1658–1665.
23. Benetti F. Minimally invasive coronary surgery (the xyphoid approach). Eur J Cardiothorac Surg
1999;16(suppl 12):S10–S11.
24. Dullum M, , Block J, Benetti F, et al. Xiphoid MIDCAB: report of the technique and experience with
a less invasive MIDCAB. Heart Surg Forum 1999;2(1):77–81.
25. Parsonnet V, Dean D, Bernstein AD. A method of uniform stratification of risk for evaluating the results
of surgery in acquired adult heart disease. Circulation 1989;79:(suppl 1):I3–I12.
26. Mack M, Acuff T, Yong P, Jett G, Carter D. Minimally invasive thoracoscopically assisted coronary
artery bypass surgery. Eur J Cardiothorac Surg 1997;12(1):20–24.
27. Watanabe G, Misaki T, Kotoh K, Yamashita A, Ueyama K. Bilateral thoracoscopic minimally invasive
direct coronary artery bypass grafting using internal thoracic arteries. Ann Thorac Surg
1998;65(6):1673–1675.
28. Kagami M, Wolf R, Flege J. Surgical results of video-assisted minimally invasive direct coronary artery
bypass. Ann Thorac Surg 1998;67:1018–1021.
29. Boyd W, Kiaii B, Novick R, et al. RAVECAB: improving outcome in off-pump minimal access surgery
with robotic assistance and video enhancement. Can J Surg 2001;44(1):45–50.
30. Kiaii B, Boyd W, Rayman R, et al. Robot-assisted computer enhanced closed-chest coronary surgery:
preliminary experience using a harmonic scalpel (R) and Zeus (TM). Heart Surg Forum 2000;3(3):194–197.
31. Falk V, Diegler A, Walther T, et al. Total endoscopic computer enhanced coronary artery bypass
grafting. Eur J Cardiothorac Surg 2000;17(1):38–45.
32. Falk V, Diegler A, Walther T, et al. Total endoscopic off-pump coronary artery bypass grafting. Heart
Surg Forum 2000;3 (1):29–31.
33. Kappert U, Schneider J, Cichon R, et al. Closed chest totally endoscopic coronary artery bypass surgery:
fantasy or reality? Curr Card Rep 2000;2(6):558–563.
34. Kappert U, Schneider J, Cichon R, et al. Robotic coronary artery surgery—the evolution of a new
minimally invasive approach in coronary artery surgery. Thorac Cardiovasc Surg 2000;48(4):193–197.
35. Calafiore A, Contini M, et al. Minimally invasive coronary artery bypass grafting on the beating heart:
the European experience. In: Oz MC, Goldstein DJ, eds. Minimally Invasive Cardiac Surgery. Totowa,
NJ: Humana Press, 1999:110–113.
36. Mack M, Magovern J, Acuff T, et al Results of graft patency by immediate angiography in minimally
invasive coronary artery surgery. Ann Thorac Surg 1999;68(1):383–390.
37. FitzGibbon G, Kafka H, Leach A, et al. Coronary artery bypass graft fate and patient outcome:
angiographic follow-up of 5,065 grafts related to survival and reoperation in 1,388 patients during
25 years. J Am Coll Cardiol 1996;28:616–626.
38. Diegler A, Matin M, Falk V, et al. Coronary bypass grafting without cardiopulmonary bypass: technical
considerations, clinical results, and follow-up. Thorac Cardiovasc Surg 1999;47(1):14–18.
39. Mehran R, Dangas G, Stamou SC, et al. One-year clinical outcome after minimally invasive direct
coronary artery bypass. Circulation 2000;102:2799–2802.
Chapter 10 / MICABG: The European Experience 129
10 Minimally Invasive Coronary Artery

Bypass Grafting on the Beating Heart
The European Experience
Antonio M. Calafiore, MD,

Michele Di Mauro, MD, Alessandro Pardini, MD,
Antonio Bivona, MD,
and Stefano D’Alessandro, MD
CONTENTS
INTRODUCTION
LEFT ANTERIOR SMALL THORACOTOMY (LAST OPERATION)
MEDIAN STERNOTOMY
DISCUSSION
REFERENCES
INTRODUCTION
The prospect of grafting the internal mammary artery (IMA) to the left anterior
descending (LAD) artery via a thoracotomy without the aid of cardiopulmonary bypass
(CPB) was first explored by Kolessov in 1967 (1), and further applied by Favaloro (2),
Garrett (3), Trapp (4), and others. The early wave of enthusiasm for this technique soon
wavered with the widespread availability of CPB and cardioplegia, which allowed for a
motionless and bloodless operative field. The unequivocal and widespread success of
conventional coronary artery bypass grafting (CABG) limited the use of unsupported
bypass grafting. Two developments in the early 1990s revived the technique of myocar-
dial revascularization without CPB: (1) the emergence of minimally invasive technology
applicable to the chest, and (2) the promising results of “pumpless” bypass grafting
reported by a number of authors (5–7).
In our opinion, the definition of minimally invasive cardiac surgery encompasses all
approaches that avoid cardiopulmonary bypass, regardless of mode of surgical access

129
Fig. 1. Angiographic demonstration of an intramyocardial LAD. This constitutes a contraindica-

tion to off-pump grafting.
(i.e., full sternotomy, thoracotomy, etc.). In this chapter, we report our experience with
myocardial revascularization without CPB utlilizing two different approaches: (1) the
left anterior small thoracotomy and (2) the median sternotomy.
LEFT ANTERIOR SMALL THORACOTOMY (LAST OPERATION)

The promising clinical reports of minimally invasive coronary artery bypass grafting
(MICABG) demonstrating excellent early patency, low morbidity and mortality, and
shortened hospital stays stimulated a worldwide interest in these techniques. Our group
popularized MICABG through a left anterior small thoracotomy, coining the term the
“LAST” operation (8).
Surgical Indications
Candidates for the LAST operation include patients with isolated LAD disease in
whom a percutaneous transluminal coronary angioplasty (PTCA) was unsuccessful,
impossible (occluded LAD), or contraindicated (proximal and/or complex stenoses).
Patients with two-vessel disease (right coronary or circumflex plus LAD) in which the
non-LAD vessel is occluded and recanalized or with a mild stenosis or stenosis that could
be dilated are also considered. In addition, patients with multivessel disease in which the
lesion in the other vessels can be addressed percutaneously (hybrid approach explored
elsewhere in this book) are suitable candidates. Finally, patients for whom extracorporeal
circulation is deemed a high-risk undertaking, including patients with malignancies,

renal failure, generalized vasculopathy, coagulation disorders, and advanced age, are
also considered.
Contraindications to the LAST operation are limited to anatomic considerations and
are strictly related to the impossibility of performing the left internal mammary artery
(LIMA) to left anterior descending artery (LAD) anastomosis because of an unsuitable
or unreachable LAD. In some patients, unfavorable conditions can be detected preopera-
tively. Presence of left subclavian stenosis or occlusion precludes the LAST operation
because it can result in a coronary steal syndrome (9,10). An intramyocardial vessel,
often discernible at angiography (Fig. 1), is also an absolute contraindication. The latter
is best diagnosed on angiographic oblique right anterior projections in which the vessel
is seen to progress downward and then, after a few centimeters, turn upward (toward the
epicardium). An LAD of <1.5 mm or a calcified LAD preclude this approach. Exquisite
attention must be given to the area 2–4 cm distal to the second diagonal branch because
this is the usual anastomotic site. We rely on the evaluation of different angiographic
projections to show the internal size of the distal LAD, the quality of its walls, and its
position relative to the epicardial surface. In the majority of cases, however, the final
decision to proceed or not with the LAST approach is made at operation.
Surgical Technique
After establishing hemodynamic monitoring, general endotracheal anesthesia with a
single-lumen endotracheal tube is instituted. Early in our experienee, we relied on the use
of double-lumen endotracheal tubes to optimize left-sided exposure. With increased
experience, we have abandoned its routine use.
In our most recent experience, anesthesia was induced with fentanyl (2 µg/kg) and
propofol (2.0–2.5 mg/kg) and was maintained with fentanyl in bolus (0.8 µg/kg) and
propofol (0.4–1.2 µg/kg) in continuous infusion. Muscular relaxation was obtained with
atracurium besilatum. During the closure of the chest, the infusion of propofol was
reduced and then stopped, to achieve a rapid awakening of the patient and, if possible, his
or her disconnection from mechanical ventilation. As soon as the chest was closed, a
continuous infusion of agents (90 mg/24 h ketoralac, 10 mg/24 h metoclopramide, and
200 mg/24 h tramadol) was started for at least 24 h.
Access to the heart is obtained via LAST in the fourth or fifth intercostal space,
depending on angiographic criteria. The pleural cavity is entered, the ribs are retracted,
and the pericardium is incised parallel to the sternum. The LAD is inspected and the
feasibility of the operation is considered. In the presence of any contraindication (as
described above), the chest is closed and median sternotomy is performed. In the first part
of our experience the LIMA was harvested for a short length (4–5 cm), usually extending
from the superior intercostal space to the level of the inferior rib. In our early experience,
if the LAD lay excessively lateral, the inferior epigastric artery was used to prolong the
LAD in an end-to-end fashion (Fig. 2). Occasionally, a saphenous vein graft would be
used to achieve the necessary length (Fig. 3). This technique has been reported previously
(11). Since April 1997, the availability of better instruments has allowed us to procure a
longer segment of the LAD. Indeed, the IMA Access Retractor (CardioThoracic Sys-
tems, Cupertino, CA) has allowed extended IMA dissection to the level of the bifurcation
inferiorly and first rib superiorly. Following systemic heparinization (1 mg/kg), 3 mL of
papaverine (1 mg/mL) are injected into the LIMA, and the vessel is clipped distally (11).
Fig. 2. The inferior epigastric artery is anastomosed end to end to the LIMA to allow grafting of
an abnormally lateral LAD.
Fig. 3. Use of an interposition saphenous vein graft to achieve LIMA-LAD grafting in a laterally
placed LAD.
Fig. 4. Flow meter: mean flow and the shape of the curve.
The LAD is then occluded proximally and distally using a 4-0 Prolene suture with a
25-mm needle, snared on a small piece of silicone tubing, to avoid direct compression of
the coronary artery. A stabilizer with two feet connected to the retractor is positioned
parallel to the LAD and pushed down gently. This maneuver minimizes the effect of the
beating heart on the operative field, making the LAD virtually motionless. No precon-
ditioning of the distal LAD territory is performed.
The distal LIMA is prepared in the usual fashion. The anastomotic site of the LAD is
dissected and incised with a scalpel for a distance of 4–5 mm. The anastomosis is created
using two running sutures of 8-0 Prolene. The anastomosis is created by “parachuting”
the stitches at the heel and apex, and the anastomosis is completed by running the Prolene
sutures in standard vascular fashion. The LIMA and LAD are unclamped and meticulous
hemostasis is obtained. The flow inside the graft is then measured with a flowmeter
(Transonic, Ithaca, NY, USA, and Medi-Stim, Oslo, Norway) evaluating both the mean
flow and the shape of the curve (Fig. 4).
During the first part of our experience, we routinely reversed the heparin with a 1:1
dose of protamine, but we have abandoned heparin reversal more recently. A thorac-
otomy tube is positioned in the chest along with a small catheter, which is used to infuse
analgesics. The wound is closed in the usual manner.
Postoperative Course
The patient is extubated in the operating room, or shortly after arrival at the intensive
care unit. Routine blood samples, electrocardiogram, and chest roentgenogram are
obtained. After a few hours of observation, the patient is transferred to the ward. The chest
tube and pleural catheter are removed on the morning of the first postoperative day.
Evaluation of LIMA flow is repeated at rest and after acutely induced hypervolemia, by
lifting the patient’s legs and instructing him or her to perform an isometric exercise (12).
This maneuver induces tachycardia and increases cardiac output. In the absence of a
restrictive anastomosis, diastolic blood flow velocity increases owing to the larger amount
of blood needed by the coronary territory. Early in our experience, most patients under-
went angiography. At present, every patient suspected of having a restrictive anastomosis
undergoes angiography. If no problems are found, the majority of patients are discharged
on the second or third postoperative day.
All patients are followed up at our outpatient clinic and at the end of the first and sixth
postoperative months. All patients are subjected to a stress test and, if possible, myocar-
dial scintigraphy is obtained on their second visit. In addition, Doppler evaluation of the
LIMA–LAD anastomosis is repeated at rest and during hypervolemia, as described previously.
Results
Between November 21, 1994, and June 30, 2001, 829 patients underwent the LAST
procedure at the Division of Cardiac Surgery of the University of Chieti. The first 244
cases were operated on before the stabilization era (up to April 20, 1997) and the remain-
ing 585 thereafter. The differences between the two periods were reported previously
(13). Herein results of the latter group of patients are shown. The clinical profile of this
group of patients is depicted in Table 1.
The LIMA was directly anastomosed to the LAD in 562 patients, with an inferior
epigastric artery or saphenous vein interposition graft between the LIMA and the LAD
in 18 patients. In two patients the LIMA was used to graft a diagonal branch; in four redo
cases, a saphenous vein was used from axillary artery to LAD, and in two patients a radial
artery was used as a bridge from the in situ LIMA to the LAD. Mean occlusion time was
17.2 ± 6.2 min, distal anastomotic time was 9.5 ± 8.1 min, and mean operation time was
2.0 ± 0.6 h. Postoperative data are displayed in Table 2. All patients had Doppler flow
evaluation, at rest or during Azoulay maneuver, angiography, or both. Globally, the
patency rate was 99.1% and the incidence of perfect anastomosis was 98.1%.
Among the 585 patients who underwent the LAST procedure, early and late mortality
were 2.9% and 1.9%, respectively. Early deaths were due to cardiac causes in four cases
(sudden death due to infero-lateral myocardial infarction with widely patent graft at
necropsy in two cases; low output syndrome, preoperatively present and not reversed by
the graft, widely patent at necropsy; cardiac tamponade in a patient who had a successful
stent placement on the right coronary artery 3 d after the LAST procedure), and not
cardiac-related in two cases (cerebral hemorrhage, abdominal complication).
Late deaths were cardiac in six instances (three sudden deaths and two acute myocar-
dial infarction) and not cardiac-related in five cases (one cerebrovascular accident, three
malignancies, and one car accident).
Eight patients (1.4%) were reoperated on, due to conduit occlusion (seven) or anasto-
motic malfunction (one). All patients underwent an uneventful reoperation, via median
sternotomy. Nine patients (1.5%) required percutaneous intervention, four due to pro-
gression of LAD disease progression, and five on another coronary territory.
At a mean follow-up of 38.5 ± 15.1 mo, 569 patients (97.3%) are alive and asymptom-
atic with or without medical treatment; 554 patients (94.7%) are alive, asymptomatic, and
without a repeated surgical or cardiological intervention on the LAD.
Five-year survival and event-free survival were 96.1 ± 1.0 and 93.2 ± 1.2, respec-
tively (Fig. 5).
With increasing experience and new instruments, angiographic results are improving.
We reported recently our early angiographic results. In a series of 190 patients, patency
Table 1
Clinical Profile of 585 Patients
Undergoing the LAST Operation
Age
Mean 62 ± 10 yr
Range 32–88
≥ 75 yr 56 (10.8%)
Female 74 (12.6%)
Urgent 75 (12.8%)
Single-LAD disease 295 (50.4%)
EF (%)
Mean 61 ± 12
Range 15–90
≤ 35% 13 (2.2%)
Diabetes 101 (17.3%)
Previous MI 262 (44.8%)
Redo 34 (5.8%)
Other comorbiditiesa 126 (21.5%)
aAge ≥ 75, malignancy, coagulopathy, HIV, diffuse
cerebrovascular disease, chronic renal failure.
Table 2
Postoperative Data
for 585 Patients Undergoing LAST Operation
Extubation by 2 h 495 (84.6%)
Reop for bleeding 17 (2.9%)
Transfusion 22 (3.8%)
ICU length of stay (h) 6.6 ± 8.3
Stroke 0
Perioperative MI 2 (0.3%)
Atrial fibrillation 50 (8.3%)
Postoperative length of stay (d) 3.1 ± 1.2
Discharged by postop day 187 (32%)
Fig. 5. Five-year survival and event-free survival (LAST operation).

rate was 98.9% and perfect patency rate was 97.4 (14). After a mean follow-up of 11.1 ±
9.5 mo, patency rate and perfect patency rate were 94.4% (34/36) and 91.7% (33/36),
respectively.
An unexpected finding occurred during our experience. Six patients with anastomotic
or conduit anomalies on postoperative evaluation showed reversal to normal anatomy
after a mean period of 94 ± 56 d from the first angiography (Fig. 6A,B). We feel this is
due to adventitial hematomas or clots that resolve with time. At present, decisions to
reoperate on the basis of an imperfect angiographic appearance of the anastomosis or of
the conduit are delayed for 2 mo, at which time the angiography is repeated (8).
MEDIAN STERNOTOMY
Between January 1995 and May 20, 1997, only patients with favorable anatomy were
considered for off-pump coronary bypass grafting. However, among them, the procedure
was performed only in patients with high risk factors for institution of cardiopulmonary
bypass (CPB), as marginal branch or posterior descending artery grafting was difficult
and cumbersome. The remaining patients had only grafts to the LAD, diagonal branches,
or right coronary artery proximal to the crux. During this period, only 129 patients were
operated on without CPB (13.3% of the patients operated on with median sternotomy),
with 2 conversions to CPB (1.6%).
In May 1997, we introduced a technique for heart verticalization that allowed exposure
of the lateral and inferior walls (15). Thereafter, the incidence of patients operated on
without CPB rose through June 30, 2001, to 50% (1134/2268, with 54 conversions or 4.8%).
Clinical and angiographic election criteria utilized in this group of patients included:
(1) the coronary vessels had to be epicardial, with an internal size diameter >1.2 mm and
with no calcifications at the level of the anastomotic site; (2) the marginal branches had
to be long, with a uniform internal size up to the midportion of the vessel, as this is often
the anastomotic site; (3) electric stability had to be present, as the risk of severe ventricu-
lar arrhythmias during manipulation is high; (4) while ejection fraction per se was not a
contraindication for beating-heart revascularization, an enlarged heart can make the
exposure of the lateral wall difficult; (5) need for multiple sequential grafts was consid-
ered a contraindication up to the end of December 1997, as we concentrated our attention
in obtaining a good end-to-side anastomosis. Since January 1998, sequential grafts have
been routinely performed without CPB if necessary.
Surgical Technique
The protocol of anesthesia was the same as reported in the first part of this chapter. The
mammary arteries are harvested as skeletonized grafts; the remaining conduits are har-
vested as described previously (12). After full heparinization (3 mg/kg), all the arterial
grafts are prepared as reported previously (11).The target coronary vessels are explored
and the surgical strategy is confirmed.
Techniques to verticalize of the heart have been modified with time. At first, four
gauze slings were used, passed behind the inferior vena cava (two) and through the
transverse sinus (two) (15). In 1999 only one gauze sling was passed through the trans-
verse sinus and behind the inferior vena cava. In addition, one deep pericardial suture was
Fig. 6. (Opposite page) Anastomotic anomalies, at first angiography on postoperative day 2 (A),
and 4 mo postoperatively (B). Note reversal to normal-appearing anatomy.
Fig. 7. Verticalization of the heart as obtained with the Xpose.
placed, as described by Lima (16), between the left inferior pulmonary vein and the
inferior vena cava, with a third sling in the same place. More recently, the verticalization
of the heart was obtained with the Xpose™ (Guidant Corporation, Cupertino, CA, USA)
(Fig. 7) and a single Lima stitch with a single additional sling, if necessary. Hemody-
namic stability was obtained with Trendelenburg position, addition of volume and with
small boluses of a vasopressor when needed (metaraminol or diluted norepinephrine),
and mechanical stabilization was obtained with CTS stabilizers (Acces Ultima™ Sys-
tem, Guidant Corporation, Cupertino, CA, USA) (17). At the end of every anastomosis
the flow in the graft is measured using a flow meter (Cardiomed, Medi-Stim Oslo,
Norway; Transonic System, Ithaca, NY, USA).
Blood lost during the procedure is reinfused in the patient using a cell saver (DIDECO,
Mirandola, Modena, Italy). Protamine is reversed 0.5:1 and the wound is closed in the
usual fashion.
Postoperative Course
All patients were admitted to the ICU, where they remained for a mean of 13.7 ± 17.8 h.
Globally, 41.3% of the patients were discharged from the ICU the same day of the
operation. This percentage rises to 66.7% if only the procedures performed in the morn-
ing are considered. The patients were discharged from the hospital after a mean of 4.2 ±
2.6 d. Forty-seven percent were sent directly home and the remaining wre transferred to
another hospital.
Table 3
Comparison of Pre- and Intraoperative Data Among Patients Undergoing
Coronary Revascularization with and Without CPB Support
Off-Pump CABG Conventional CABG p Value
n 1134 1134
Age (yr) 65 ± 10 64 ± 10 ns
Age ≥ 75 yr 160 (14.1%) 166 (14.6%) ns
Female 210 (18.5%) 212 (18.7%) ns
LVEF (%) 58 ± 13.5 56 ± 13.6 < 0.001
LVEF ≤ 35% 70 (6.2%) 75 (6.6%) ns
Urgent operation 259 (22.8%) 251 (22.1%) ns
Diabetes 250 (22%) 284 (25%) ns
Left main disease 135 (11.9%) 134 (11.8%) ns
Redo 19 (1.7%) 87 (7.7%) < 0.01
Anastomoses/patient 2.4 ± 0.8 3±1 < 0.001
Sequential grafts 217 574 < 0.001
Carotid surgery 24 (2.1%) 24 (2.1%) ns
Table 4
Comparison of Clinical Outcomes Between OPCAB (n = 1134)
and Conventional CABG) n = 1134) Patients
Off-Pump CABG Conventional CABG p Value
1134 1134
Conversion to CPB 53 (4.8%) NA
Mortality 17 (1.5%) 37 (3.3%) 0.005
Inotrope use >24 h 32 (2.8%) 69 (6.1%) 0.002
Perioperative MI 10 (0.9%) 33 (2.9%) <0.001
CK-MB peak (IU/L) 25 ± 42 48 ± 82 <0.001
Stroke 9 (0.8%) 13 (1.1%) ns
Bleeding mL/24 h 399 ± 431 435 ± 409 0.041
Transfusion 263 (23.2%) 388 (34.2%) <0.001
Atrial fibrillation 117 (10.3%) 135 (11.9%) ns
Early major events 59 (5.2%) 110 (9.7%) <0.001
Results
Between January 1995 and July 2001, 1263 patients underwent myocardial
revascularization on a beating heart via a median sternotomy at our institution. Up to May
20, 1997, patients were strictly selected, lacking any technique for heart verticalization.
In 129 patients, 211 distal anastomoses were performed (1.7 ± 0.7/patient). Among them,
only 12 were performed on the lateral wall and 5 on the inferior wall.
Following introduction and application of verticalization techniques, selection criteria
were relaxed and more patients were operated upon without CPB.
From May 21, 1997 to June 30, 2001, 1134 patients (50% of patients operated on via
a median sternotomy in this period) had myocardial revascularization on a beating heart
without CPB. Comparative data with patients operated on with CPB in the same time
frame are listed in Tables 3 and 4.
Fig. 8. Sequential grafts according to target territories. White bars depict conventional bypass
patients and dark bars represent beating-heart revascularization patients.
Patients in the off-pump group had a higher ejection fraction, a lower incidence of
redo, and fewer anastomoses per patient, with less sequential grafting (Fig. 8). However,
both groups had the same expected risk according to EuroSCORE, 3.6 vs 3.7 (p = ns).
Mortality was lower in the off-pump group, as was the incidence of acute myocardial
infarction, need for inotropes, and need for transfusion. Incidence of cerebrovascular
accidents and atrial fibrillation was similar in both groups.
After a mean follow up of 12.7 ± 11.7 mo, 20 patients who underwent myocardial
revascularization without CPB died, 10 due to cardiac causes (2 due to heart failure,
2 due to acute myocardial infarction, and 6 sudden deaths) and 10 due to other causes
(3 due to cerebrovascular accident, 1 due to sepsis, 2 after abdominal aneurysm surgery,
2 due to malignancy, 1 from intestinal ischemia, and 1 from respiratory failure associated
with pulmonary fibrosis).
In the conventional cardiopulmonary bypass group, nine patients died after a mean
follow-up of 4.0 ± 3.0 months. Six deaths were due to cardiac causes.
After a mean of 7.0 ± 3.4 mo, seven off-pump recipients needed a repeat coronary
intervention. Of these, five required repeat bypass grafting (two graft failures, three with
progression of the disease) and two underwent percutaneous interventions (2 graft fail-
ures). After a mean of 13.5 ± 10.3 mo, two patients operated on with conventional CPB
required reop CABG for graft failure. Four-year survival for the off-pump and conven-
tional groups was 96.0 ± 0.6 and 97.1 ± 0.5, respectively (p = ns) (Fig. 9). Event-free
survival was 96.3 ± 0.6 for patients who had beating-heart revascularization and 95.7 ±
0.6 for patients operated on with CPB support (p = ns) (Fig. 10).
Early angiographic results, shown in a previous report (18), were similar to those
obtained with CPB support. A high patency rate (98.2%) of marginal branch grafts is a
very important issue, as lateral wall grafting is a major problem in myocardial
revascularization without CPB.
Fig. 9. Four-year survival of patients undergoing OPCAB (dashed line) and conventional CABG
(solid line).
Fig. 10. Four-year event-free survival for patients undergoing beating-heart revascularization
(solid line) and conventional CPB-supported bypass grafting (dashed line).
DISCUSSION
Myocardial revascularization without CPB is an accepted surgical strategy. Its benefit
becomes evident in patients for whom CPB may have potential complications, as in the
elderly or in patients with severe preoperative organ dysfunction.
The benefit of the LAST operation in particular is obvious during the early
psotoperative recovery. These patients have short lengths of stay and faster return to
social activities. The use of a LAST approach vs percutaneous interventions in patients
with single-vessel disease is an interesting question. The ability to offer a LIMA-to-LAD
anastomosis with very high early and midterm patency, the most important factor for
long-term survival (19), and with a safe, minimally invasive operation offers significant
advantages over coronary artery stenting.
Despite this, stenting of the proximal LAD remains the first treatment for these lesions.
However, application of coronary artery stenting to patients with type C lesions is asso-
ciated with high failure rates and worse midterm outcomes than MIDCAB (20). Even if
the new coated-stent technology confirms a low risk of restenosis, the LAST operation
is a well-standardized operation with good early and long-term results and remains a valid
surgical option for single LAD disease.
One of the criticisms to the LAST operation was related to the possibility of flow
competition if some of the side branches of the LIMA were left undivided. Indeed, some
anecdotal reports (21) where single cases of ligature or embolization of huge side branches
of the internal mammary artery to reverse ischemia in the LAD territory have been reported.
The concept that the LIMA can only be partially harvested with the minithoracotomy
LAST approach is not true. During conventional median sternotomy, the LIMA is har-
vested up to the subclavian vein. The most proximal side branch is the first intercostal
artery, the same branch that is divided during LIMA harvesting via a LAST approach. The
very first part of the LIMA is never harvested, and lies the costolateral branch in this
portion and some other subclavian artery branches. It is important to note that in 30% of
the cases, these branches have a common origin with the LIMA (22,23). If the flow
competition between the side branches of the LIMA and the LAD were a clinical prob-
lem, it would have been present in all instances of myocardial revascularization regard-
less of approach.
Moreover, the physiology of coronary circulation is such that the nutritive flow hap-
pens mainly during diastole, whereas in the muscular territories the nutritive flow is
mainly systolic (24,25). An undivided side branch could cause flow competition, but only
if it is addressed to a territory perfused during diastole, as the lungs, but this situation is
very rare.
When multiple grafts are needed, off-pump revascularization via median sternotomy
is the approach of choice. Although myocardial revascularization without CPB was the
initial approach in the early era of coronary surgery (1,25), the advent of safe CPB and
cardioplegic arrest limited the interest in a technique that was more technically demand-
ing. In fact, even though several groups continued to perform such a procedure (5,6,26),
the surgical technique was far from being well established and grafting the lateral wall
was rarely done (27).
In recent years, a new way of performing the operation was developed. The introduc-
tion of stabilizers and different techniques to verticalize the heart (15,16,28–30) were
striking improvements that modified the feasibility of the procedure. The hemodynamic
changes during the verticalization, when present (31), are small enought to have no effect
on the surgical outcome (32).
Although some studies showed that, in patients in whom CPB was not used, there was
significant reduction of the inflammatory response (33), improvement in neurocognitive
outcome (34), and no increase in extracerebral water (35) in comparison with patients
operated on with CPB; primary end points (mortality, CVA, and acute myocardial inf-
arction incidence) were never reported to be lower in off-pump recipients.
In our recent report, we were able to demonstrate with the aid of stepwise logistic
regression that CPB is an independent variable for early death, perioperative myocardial
infarction, and early major events (the composite of death, low output syndrome,
perioperative AMI, cerebrovascular accident, acute renal failure, acute respiratory fail-
ure, and acute abdominal complication) (Table 5) (17). We believe that, if myocardial
revascularization can be done without CPB, patients will experience lower morbidity
and mortality.
Lack of angiographic studies in patients who underwent coronary surgery on a beating
heart had led surgeons to question the safety and efficacy of the operation. Our
angiographic studies demonstrated satisfying patency rate; even if some loss of patency
occurs during the first year related to perioperative technical inadequacies, it holds true
for every coronary anastomosis independent of the surgical technique (18).
Chapter 10 / MICABG: The European Experience
Table 5
Stepwise Logistic Regression Analysis of Risk Factors for Untoward Clinical Outcomes
Early Death Stroke Early MI EMEd
Risk Factor OR (95% CI) p Value OR (95% CI) p Value OR (95% CI) p Value OR (95% CI) p Value
Age ≥ 75 yr — — — — 2.2 (1.0–5.0) 0.0465 — —

AMIa > 24 h 18.7 (2.8–123.7) 0.002 31.8 (2.7–365) 0.0055 — — 18 (3.0–108) 0.0016
Chronic renal failureb — — — — — — 2.9 (1.4–6.4) 0.0058
Diabetes 2.2 (1.1–4.2) 0.016 — — — — — —
LVEF ≤ 35% 2.9 (1.3–6.5) 0.011 9.3 (3.2–27.1) <0.00001 3.0 (1.2–7.3) 0.017 3.5 (2.1-5.90) 0.0001
143
Heart failure 4.0 (1.5–10.4) 0.005 — — 6.7 (2.4–18.7) 0.0003 3.4 (1.7–6.90) 0.0006
Preop IABPc 21.9 (2.2–222.4) 0.009 — — 17.2 (1.5–192.1) 0.0208 — —
Left main disease — — 3.9 (1.3–11.8) 0.0151 — — — —
Unstable angina — — — — 2.8 (1.4–5.6) 0.0049 2.0 (1.4–3.0) 0.0002
Ventricular arrhythmia — — — — — — 2.9 (1.3–6.8) 0.013
Cardiopulmonary bypass 2.2 (1.1–4.3) 0.022 — — 2.5 (1.2–5.2) 0.0185 1.8 (1.2–2.6) 0.0034
Carotid endarterectomy — — 9.0 (1.9–43.7) 0.006 — — — —
aAcute myocardial infarction.
bDefined as serum creatinine ≥ 2.0 mg/dL.
cIntra-aortic balloon pump.
dEarly major events (composite of death, low-output syndrome, acute myocardial infarction, cerebrovascular accident, acute respiratory dailure, acute
renal failure, abdominal complication).
143
Myocardial revascularization without CPB is a surgical strategy that, in selected cases,

can have satisfying early and midterm results. Selection of patients, in our experience, is
the key to the success of the procedure. Tailoring the strategy to the patient will improve
the already excellent outcomes currently achieved with coronary bypass grafting.
REFERENCES
1. Kolessov VI. Mammary artery-coronary artery anastomosis as method of treatment for angina pectoris.
2. Favaloro RG. Saphenous vein autograft replacement of severe segmental coronary artery occlusion. Ann
Thorac Surg 1968;5:334–339.
3. Garrett HE, Dennid EW, DeBakey ME. Aortocoronary bypass with saphenous vein graft. Seven year
follow-up. JAMA 1973;223:792–794.
4. Trapp WG, Bisarya R. Placement of coronary artery bypass graft without pump oxygenator. Ann Thorac
Surg 1975;19:1–9.
5. Benetti FJ, Naselli G, Wood M, Geffner L. Direct myocardial revascularization without extracorporeal
circulation: experience in 700 patients. Chest 1991;100:312–316.
6. Buffolo E, de Andrade CS, Branco JN, Teles CA, Aguilar LF, Gomes WJ. Coronary artery bypass
7. Moshkovitz Y, Lusky A, Mohr R. Coronary artery bypass without cardiopulmonary bypass: analysis of
short-term and mid-term outcomes in 220 patients. J Thorac Cardiovasc Surg 1995;1(10):979–987.
8. Calafiore AM, Di Giammarco G, Teodori G, et al. Left anterior descending coronary artery grafting via
left anterior small thoracotomy without cardiopulmonary bypass. Ann Thorac Surg 1996;671:
1658–1665.
9. Bryan FC, Allen RC, Lumsden AB. Coronary-subclavian steal syndrome: report of five cases. Ann Vasc
Surg 1995;9(1):115–122.
10. Breall JA, Grossman W, Stillman IE, Gianturco LE, Kim D. Atherectomy of the subclavian artery for
patients with symptomatic coronary-subclavian steal syndrome. J Am Coll Cardiol 1993;21(7):
1564–1567.
11. Calafiore AM, Di Giammarco G, Luciani N, et al. Composite arterial conduits for a wider myocardial
revascularization. Ann Thorac Surg 1994;58:185–190.
12. Calafiore AM, Teodori G, Di Giammarco G, et al. Minimally invasive coronary artery bypass grafting
on a beating heart. Ann Thorac Surg 1997;63:S72–S75.
13. Calafiore AM, Vitolla G, Mazzei V, et al. The LAST Operation: techniques and results before and after
the stabilization era. Ann Thorac Surg 1998;66:998–1001.
14. Calafiore AM, Di Giammarco G, Teodori G, et al. Midterm results after minimally invasive coronary
surgery (LAST operation). J Thorac Cardiovasc Surg 1998;115:763–771.
15. Calafiore AM, Di Giammarco G, Teodori G, Mazzei V, Vitolla G. Recent advances in multivessel
coronary grafting without cardiopulmonary bypass. Heart Surg Forum 1998; 33589.
16. Lima R Revascularizacion a o da artèria circunflexa sem auxilio da CEC. In: XII encontro dos discipulos
do dr. EJ Zerbini, Curitiba, 1995. Sessao de videos. Curitiba, Parana, Sociedade dos discipulos do dr.
EJ Zerbini Outtubro de 1995: 6.b.
17. Calafiore AM, Di Mauro M, Contini M, et al. Myocardial revascularization with and without cardiop-
ulmonary bypass in multivessel disease. Impact of the strategy on early outcome. Ann Thorac Surg
2001;72:456–463.
18. Calafiore AM, Teodori G, Di Giammarco G, et al. Multiple arterial conduits without cardiopulmonary
bypass. Early angiographic results. Ann Thorac Surg 1999;67:450–456.
19. Loop FD. Internal thoracic artery grafts—biologically better coronary arteries. N Engl J Med
1996;334:263–265.
20. Mariani MA, Boonstra PW, Grandjean JG, et al. Minimally invasive coronary artery bypass grafting
versus coronary angioplasty for isolated type C stenosis of left anterior descending artery. J Thorac
21. Hartz RS, Heuser RR. Embolization of IMA side branches for post CABG ischemia. Ann Thorac Surg
1997;63:1765–1766.
22. Henriquez-Pino JA, Gomes WJ, Prates JC , Buffolo E. Surgical anatomy of the internal thoracic artery.
Ann Thorac Surg 1997;64:1041–1045.
23. Calafiore AM, Contini M, Iacò AL, et al. Angiographic anatomy of the left internal mammary artery.
Ann Thorac Surg 1999;68:1636–1639.
24. Kern MJ, Bach RG, Donohue TJ, et al. Role of large pectoralis branch artery in flow through a patent
left internal mammary artery conduit. Cathet Cardiovasc Diagn 1995;34:240–244.
25. Favaloro RG, Effler DB, Groves LK, Sheldon WC, Sones FM. Direct myocardial revascularization by
saphenous vein graft. Present operative technique and indications. Ann Thorac Surg 1970;10:97–111.
26. Ankeney JL. To use or not to use the pump oxigenator in coronary bypass operation. Ann Thorac Surg
1975;19:108–109.
27. Tasdemir O, Vural KM, Karagoz H, Bayazit K. Coronary artery bypass grafting on the beating heart
without the use of the extracorporeal circulation: review of 2052 cases. J Thorac Cardiovasc Surg
1998;116:68–73.
28. Spooner TH, Dyrud TE, Monson BK, Dixon GE, Robinson LD. Coronary artery bypass on the beating
heart with the Octopus: a North American experience. Ann Thorac Surg 1998;66:1032–1035.
29. Angelini GD, Lucchetti V. An inexpensive method of heart stabilization during coronary artery opera-
tion without cardiopulmonary bypass. Ann Thorac Surg 1998;1477–1478.
30. Bergsland J, Karamanoukian HL, Soltosky PR, Salerno TA. “Single suture” for circumflex exposure in
off pump coronary artery bypass grafting. Ann Thorac Surg 1999;67:1653–1658.
31. Mathison M, Edgerton JR, Horswell JL, Akin JJ, Mack MJ. Analisys of haemodynamic changes during
beating heart surgical procedures. Ann Thorac Surg 2000;70:1355–1361.
32. Nierich AP, Diephuis J, Jansen EWL, Borst C, Knape JTA. Heart displacement during off pump CABG:
how well is it tolerated?. Ann Thorac Surg 2000;70:466–472.
33. Ascione R, Lloyd CT, Underwood MJ, Lotto AA, Pitsis AA, Angelini GD. Inflammatory response after
coronary revascularization with or without cardiopulmonary bypass. Ann Thorac Surg 2000;69:
1198–1204.
34. Diegeler A, Hirsch R, Schneider F, et al. Neuromonitoring and and neurocognitive outcome in off pump
35. Anderson RE, Li TQ, Hindmarsh T, Settergren G, Vaage J. Increased extracellular brain water after
coronary artery bypass grafting is avoided by off-pump surgery. J Cardiothorac Vasc Anesth
1999;13:698–702.
Chapter 11 / MICABG: The South American Experience 147
11 Minimally Invasive Coronary Artery

Bypass Grafting
The South American Experience
Federico J. Benetti MD, PHD

and Maximo Guida MD
CONTENTS
INTRODUCTION
MINIMALLY INVASIVE CORONARY SURGERY
SURGICAL TECHNIQUE
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
INTRODUCTION
It is widely accepted that the single most important development in cardiac surgery
was the introduction and refinement of extracorporeal circulation via cardiopulmonary
bypass (CPB). The ensuing advances in cardioplegia and myocardial protection provided
a motionless and bloodless field for accurate reparative surgery and improved event-free
survival for patients undergoing a variety of cardiac procedures. Further advances in
myocardial protection along with developments in percutaneous catheter-based inter-
vention fostered a decade of debate regarding the definitive treatment for ischemic coro-
nary artery disease.
Prior to the introduction of percutaneous transluminal coronary angioplasty (PTCA),
coronary artery bypass grafting (CABG), rather than an initial strategy of medical therapy,
resulted in longer survival and better quality of life in specific subgroups of patients with
multivessel disease (1–3). With the introduction of PTCA in 1977, the use of less invasive

147
methods for coronary revascularization rapidly expanded (4). The well-publicized and
much-quoted BARI trial, in which patients with multivessel disease were randomized to
CABG or PTCA, concluded that there was no statistically significant difference in late
survival between the two treatment strategies. Of note, however, the rate of repeat per-
cutaneous intervention or subsequent revascularization was 42% in the PTCA group vs
3% in the CABG group. Indeed, 31% of the patients initially randomized to PTCA
ultimately underwent CABG procedure anyway. Of great interest was the finding that the
subset of patients with treated diabetes who underwent CABG was found to have a
significant survival advantage over those receiving PTCA (5).
The fact that the left internal mammary artery (LIMA) connected to the left anterior
descending artery (LAD) is the single most powerful predictor of long-term survival has
not been refuted even in cardiology circles (6). In fact, despite the advances in catheter-
based treatment methods, proximal and diffuse LAD disease still remain problematic and
continue to have restenosis or reintervention rates much higher than bypass grafting of
this vessel (7). The advantages of PTCA include decreased trauma, shorter hospital stay
and recovery, and, though somewhat debatable, lower costs. It is clear that patients are
often willing to take the less invasive option despite being educated to the risks for
restenosis and reintervention rates in order to avoid anesthesia, sternotomy, and cardiop-
ulmonary bypass (8).
As reviewed elsewhere in this book, the earliest coronary operations were done with-
out the use of the extracorporeal circulation (9–16). These initial experiences were largely
ignored after the advent of precise, reproducible, bloodless, and motionless heart surgery;
despite this, these pioneering experiences demonstrated the feasibility of beating-heart
revascularization.
The era of cardiopulmonary bypass coronary surgery flourished and results continued
to improve. However, it became increasingly obvious that extracorporeal circulation has
deleterious effects in all body systems (17–26).
MINIMALLY INVASIVE CORONARY SURGERY

Benetti (27,28) and Buffolo (29) repopularized beating-heart coronary revascularization
(OPCAB) by expanding the technique, addressing lesions of the circumflex system, and
applying it to diverse clinical scenarios. Experimentation with several surgical access
approaches (other than full sternotomy) were used, including left anterolateral, postero-
lateral, and right anterolateral thoracotomies as well as partial sternotomy. Thoracoscopy
was used for the first time to dissect the LIMA without opening the pleural cavity. The
LIMA was then anastomosed to the LAD through a small left anterior thoracotomy in
1994 (30–32), marking the era of the minimal-access cardiac surgery, and the eponym
MIDCAB (minimally invasive direct coronary artery bypass) was coined.
A new series of technological developments allowed widespread application of the
MIDCAB technique (33–36) as detailed in Chapter 9. A prospective multicenter study
evaluating MIDCAB vs conventional bypass grafting demonstrated equivalent LIMA to
LAD patency rates (37). Today, coronary surgery in the beating heart (OPCAB and/or
MIDCAB) is routine in many hospitals around the world, confirming the advantages of
these procedures (38,39).
In 1997 we began to examine the possibility of ambulatory coronary surgery. The
approach was based on access via a xiphoid (or lower sternotomy) approach. Using three-
dimensional visualization, the LIMA was harvested and anastomosed to the LAD
Fig. 1. Lower midsternal incision for the xiphoid approach.
(40–42). Important advantages of this approach include reduced postoperative pain and
preservation of intact pleural cavities, increasing the possibility of early discharge (43,44).
We have now applied current technology through a lower sternotomy approach for
multivessel revascularization in several patients with the aim of discharging them within
24 h of operation (40–42). The following paragraphs describe our technique and results
with ambulatory coronary surgery.
SURGICAL TECHNIQUE
The original technique was developed in 1997 (40–42). Currently, we do not use the
videoscope routinely and we always extend the sternal incision to the left, right, or both
to fit the retractor adequately (Figs. 1–3). Preoperative measurements with the cardioscan
are very helpful to evaluate the internal mammary artery diameter and presence of
collaterals. It is also very important to measure the distance to potential anastomotic sites,
as it will dictate the length of mammary needed to reach the target comfortably (Figs. 4
and 5). Following lower sternal incision, a mammary retractor (Favaloro type) is used to
dissect one or both mammary arteries (Fig. 6). Additional conduits are harvested and
anastomosed to the LIMA if necessary (Fig. 7). Care is taken not to enter either pleural
cavity. When the LIMA is the only conduit needed, we harvest approx 7 cm of length. We
Fig. 2. Midsternal incision with extension to the left.
Fig. 3. Midsternal incision with bilateral extensions.

Fig. 4. Preoperative Cardioscan demonstrating the internal mammary arteries.
Fig. 5. Preoperative measurements with the Cardioscan system.

Fig. 6. Dissection of the mammary artery.
Fig. 7. Conduits used to create the y-grafts from the LIMA.

Fig. 8. Use of the Xpose apical system.
do not believe the steal phenomenon is clinically relevant and hence we do not bother
clipping LIMA branches. It is critical to measure the exact distance between the mam-
mary and the potential site in the LAD with the heart in a normal position to avoid any
kinking or angulation, particularly at the level of the sternum. Because the mammary is
attached to a fixed point, we take these measurements after the pericardium has been
opened and with the retractor in place.
The pericardium is opened to the left of the pulmonary artery and toward the left
ventricular apex and a flap is created to the right. A stitch is then placed in the left edge
of the pericardium. We position the Xpose™ Apical Positioning Device (Fig. 8), trying
to minimize cardiac manipulation to avoid atrial fibrillation (30). We routinely use the
Guidant stabilization system to perform the anastomosis (Fig. 9). Upon completion of the
anastomosis, and after satisfactory hemostasis has been achieved, drains are carefully
placed to avoid any damage to the conduits.
After closing the sternum with one to three wires, we infiltrate all intercostals spaces
and the path of the drainage with local anesthesia to enhance the chance of early extubation.
MATERIALS AND METHODS

Between October 1997 and June 2001, 85 patients were operated through the xiphoid
(lower sternotomy). Among this group of patients, 15 (12 men, 3 women) with a mean
age of 65 yr (range 45–78) were discharged within 24 h after the operation, although some
of the patients stayed for a few more hours in the facilities for practical reasons (42).
Fig. 9. Use of stabilizer to perform the anastomoses.
RESULTS
Operative mortality was 1% (one patient). Conversion to complete sternotomy oc-
curred in one patient (1%) and Q-wave perioperative infarction occurred in one (1%)
patient. In all 16 patients who consented to postoperative angiography, the patency rate
was 100%. In the 15 ambulatory patients the mortality was 0% and the perioperative
infarction 0%. The postoperatative follow-up ranged between 12 and 44 mo, with an
average of 22 mo. All patients were asymptomatic. Five patients have had stress thallium
testing, which has demonstrated no active ischemia.
DISCUSSION
Coronary bypass surgery without CPB in the beating heart via different incisions is a
proven and reproducible technique (38–42) that has the distinct advantages of avoiding
the morbidity associated with extracorporeal circulation (8,17,26).
In our mind, the goal of the MIDCAB operation (31) was to perform coronary surgery
in an ambulatory fashion, based on the time-proven concept that the left internal mam-
mary connected to the LAD is the best long-term treatment for lesions of this artery.
We soon realized that the thoracotomy incision required for the MIDCAB approach
created two obstacles to early discharge: significant postoperative pain and invasion of
the left pleura. The former often mandates the use of intravenous narcotics, while the
latter requires chest tube placement, serial radiographs, and removal of the chest tube,
with concerns about pneumothoraces and pleural effusions. In view of these limitations,
we refocused our approach to the xiphoid approach. This strategy avoids pleural entry
and is associated with much lesser and more treatable postoperative pain, which together
permit the application of coronary bypass grafting to the ambulatory setting. We are now
focusing on developing newer methodologies and instrumentation to expand this
approach to multivessel grafting. We envision that, as technological improvements are
made, multivessel coronary bypass grafting on the beating heart, with multiple arterial
conduits, will be feasible via a small incision and permit ambulatory or very early
hospital discharge with results that mimic those currently obtained by conventional full
sternotomy.
REFERENCES
1. Jones R, Kesler K, Phillips H, et al. Long-term survival benefits of coronary artery bypass grafting and
percutaneous transluminal angioplasty in patients with coronary artery disease. J. Thorac Cardiovasc
Surg 1996;111:1013–1025.
2. Jones EL, Craver JM, Guyton RA, et al. Importance of complete revascularization in performance of the
coronary bypass operation. Am J Cardiol 1983;51:7–12.
3. Califf RM, Mark DB. Percutaneous intervention, surgery and medical therapy: a perspective from the
Duke databank for cardiovascular diseases. Semin Thorac Cardiovasc Surg 1994;6:120–128.
4. King SB III, Lembo NJ, Weintraub WS, et al. Emory angioplasty versus surgery trial (EAST): design,
recruitment and baseline description of patients. Am J Cardiol 1995;75:42C–59C.
5. Feit F, Brooks MM, Sopko G, et al. Long-term clinical outcome in the Bypass Angioplasty
Revascularization Investigation Registry: comparison with the randomized trial. BARI Investigators.
Circulation 2000;101:2795–2802.
6. Loop FD, Internal thoracic artery grafts: biologically better coronary arteries. N Engl J Med
1996;334:263–265.
7. Fischman DL, Lenon MB, Baim DS, et al. A randomized comparison of coronary-stent placement and
balloon angioplasty in the treatment of coronary artery disease. N Engl J Med 1994;331:496–501.
8. Roach GW, Kanchuger M, Mangano CM, et al. Adverse cerebral outcomes after coronary bypass
surgery. N Engl J Med 1996;335:1857–1863.
9. Vineberg AM. Development of anastomosis between coronary vessels and transplanted mammary
artery. Med Assoc J 1954;71:594.
10. Westaby S, Benetti FJ. Less invasive coronary surgery: consensus from the Oxford meeting. Ann Thorac
Surg 1996;62:924–931.
11. Goetz RH, Rohman M, Haller JD, et al. Internal mammary-coronary anastomosis: a moisture method
employing tantalum rings. J Thorac Cardiovasc Surg 1961;41:378–386.
12. Sabiston DC. The coronary circulation. Johns Hopkins Med J 1974;134:314–329.
13. Kolessov, VI. Mammary artery-coronary artery anastomosis as method of treatment for angina pectoris.
14. Ankeney JL. To use or not to use the pump oxygenation in coronary bypass operation. Ann Thorac Surg
1975;19:108–109.
15. Garret HE, Dennis EW, DeBakey Me. Aortocoronary bypass with saphenous vein graft. Seven-year
follow-up. JAMA 1973;223:792–794.
16. Trapp WG, Bisarya R. Placement of coronary artery bypass graft without pump oxygenator. Ann Thorac
Surg 1975;19:1–8.
17. Westby S, Johnson P, Parry AJ, et al. Serum S100 protein: a potential marker for cerebral events during
18. Robin ED, McCaurley RF, Notkin H. Long- term cognitive abnormalities associated with cardiopulmo-
nary bypass (CPB) and the Babel effect. Chest 1994;106:278–281.
19. Murkin JM, Martzke JS, Buchan AM, et al. A randomized study of the influence of perfusion technique
and pH management strategy in 316 patients undergoing coronary artery bypass surgery. II. Neurologic
and cognitive outcomes. J Thorac Cardiovasc Surg 1995;110:349–362.
20. Hiberman M, Derby GC, Spenser RJ, Stinson EB. Sequential path physiological changes characterizing
the progression from renal dysfunction to acute failure following cardiac operation. J Thorac Surg
1980;79:838–844.
21. Zenardo G, Michielon P, Paccagnella A, et al. Acute renal failure in the patient undergoing cardiac
operation: prevalence, mortality rate an main risk factors. J Thorac Cardiovasc Surg 1994;107:
1489–1495.
22. Corwin HL Sprague SM, DeLaria GA, Norusis MJ. Acute renal failure associated with cardiac opera-
tions. J Thorac Cardiovasc Surg 1989;98:1107–1112.
23. Butler J, Rocker GM, Westby S. Inflammatory response to cardiopulmonary bypass. Ann Thorac Surg
1993;55:552–559.
24. Gailiunas P Jr, Chawla R, Lazarus JM, et al. Acute renal failure following cardiac operations. J Thorac
25. Bhat JG, Gluck MC, Lowenstein J, et al. Renal failure after heart surgery. Ann Intern Med 1976;84:
677–682.
26. Emunds L H Why cardiopulmonary bypass make patients sick. Strategies to control the blood-synthetic
surface interface. Adv Card Surg 1996;6:88–92.
27. Benetti FJ. Cirugía cardiaca a sin CEC o parada cardiaca. Rev de la Federación Argentina de Cardiología
1980;8:3.
28. Benetti FJ. Direct coronary surgery with saphenous vein bypass without either cardiopulmonary bypass
or cardiac arrest. J Cardiovasc Surg 1985;26:217–222.
29. Buffolo E, Andrade JC, Succi J, Leao LE, Gallucci C. Direct myocardial revascularization without
cardiopulmonary bypass. Thorac Cardiovasc Surg 1985;33:26–29.
30. Benetti FJ, et al. Uso de la toracoscopía en cirugía coronaria para disección de la arteria mamaria interna.
Prensa Medica Argentina 1994;81:877–879.
31. Benetti FJ, Ballester C. Use of thoracoscophy and minimal thoracotomy in mammary to coronary bypass
to left anterior descending artery, without extracorporeal circulation. J Cardiovasc Surg 1995;36:159–161
32. Benetti FJ, et al. Coronary revascularization with arterial conduits via a small thoracotomy and assisted
by thoracoscopy, although without cardiopulmonary bypass. Coronary Revasc 1995;4(1):22–24.
33. Benett FJ. Method for Coronary Artery Bypass. US Patent 5,888,247.
34. Benetti FJ, et al. Access Platform for Internal Mammary Dissection. US Patent 5,730,757.
35. Benetti FJ, et al. Surgical Method for Stabilizing the Beating Heart During Coronary Bypass Surgery.
US Patent 5,894,843.
36. Benetti FJ, et al. Surgical Devices for Imposing a Negative Pressure to Fix the Position of Cardiac Tissue
During Surgery. US Patent 5,727,569.
37. Meharan R. Long-term patency of LIMA-LAD beating heart anastomosis (POEM) trial The 6th annual
Cardiothoracic Techniques & Technologies meeting 2000. January 27–29, 2000. Bal Harbour, FL.
38. Mitchell J, et al. Elimination of cardiopulmonary bypass improves early survival in multivessel coronary
artery bypass patients. Society of Thoracic Surgeons, 2001 (Jan), New Orleans, LA, USA.
39. Benetti FJ. Minimally invasive coronary surgery. Seventieth Scientific Sessions of the American Heart
Association, Orlando, FL, USA, 1997 (Nov).
40. Benetti FJ. Minimally invasive coronary surgery (the xiphoid approach). Eur J Cardiothorac Surg
1999;16(suppl 2):S10–S11.
41. Guida M, Torrealba C, Rivera J, et al. CTT meeting, New Orleans, LA, USA, 2001.
42. Subramanian VA, Sani G, Benetti FJ, Calafiore A. Minimally invasive coronary bypass surgery: a
multicenter-report of preliminary clinical experience. Circulation 1995;92(8):I-645.
43. Diegeler A, et al Minimally invasive coronary bypass experience. Eur J Cardiothorac Surg 2000;
17:501–504.
44. Benetti FJ. Cirugía coronaria directa sin circulación extracorporeal. Buenos Aires: Ed Akadia, 1993.
Chapter 12 / Reoperative Off-Pump CABG 157
12 Reoperative Off-Pump CABG
Valavanur A. Subramanian, MD,

James D. Fonger, MD,
and Nilesh U. Patel, MD
CONTENTS
INTRODUCTION
INDICATION FOR PREOPERATIVE PLANNING
PREOPERATIVE PLANNING FOR REOPERATION
ANESTHESIA AND INTRAOPERATIVE MONITORING
SURGICAL TECHNIQUE
OPERATIVE APPROACHES
INSTITUTIONAL EXPERIENCE
REFERENCES
INTRODUCTION
Coronary reoperations continue to play an increasing role in the practice of coronary
bypass surgery (1,2). At present, approx 10% of isolated coronary revascularization
operations in the United States are reoperations (3). Operative mortality and morbidity
are increased for patients undergoing reoperative or redo coronary artery bypass grafting
(CABG), with an operative mortality in most series of reoperations three to five times that
of primary CABG (1,2,4–6).
Reoperative procedures have several technical obstacles that differentiate them from
primary procedures. These problems include: (1) difficulties with sternal reentry, with
potential for cardiac and conduit injury during dissection when the surgical plane is
obscured by dense adhesions and normal anatomical landmarks are obliterated; (2) avail-
ability of conduit; (3) management of patent vein grafts with severe atherosclerosis; (4)
myocardial protection in situations where there are complex routes of myocardial perfu-
sion, depending on the status of the native coronary circulation and the patency of the vein
or internal mammary artery (IMA) grafts.

157
Alternative strategies currently used include different techniques for sternal reentry,
strict avoidance of graft manipulation to minimize the risk of graft atheroembolism, and
numerous modifications of myocardial protection, including performing coronary artery
bypass grafting without cardiopulmonary bypass through the standard sternal reentry
approach (7–10).
Since 1994, alternative nonsternal, minimal-access off-pump approaches (i.e., left
anterior thoracotomy, subxyphoid, short lateral thoracotomy, and transabdominal) have
been increasingly used in reoperation either alone or in combination (11–14).
INDICATION FOR REOPERATION

Some indications for coronary reoperation are clear and are based on clinical evidence,
while other indications must be based on logic. Clear indications include:
1. The presence of late stenosis (≥50%) in vein grafts supplying the LAD or large circum-
flex and/or right coronary systems that supply viable myocardium.
2. Patients with left main or triple vessel disease and abnormal left ventricular function, or
a large LAD coronary artery with a proximal stenosis and multivessel disease who do not
have any patent grafts. These indications are based on data from the randomized trials of
primary coronary bypass surgery (15,16), and meta-analysis of those data.
3. For patients with severe disabling angina and previous bypass surgery, coronary reoperation
is one option that can be considered for the purpose of relieving symptoms (17).
Noninvasive testing can also add information to the anatomic data. When patients
exhibit a decrease in left ventricular function associated with stress testing, that situation
predicts a worse outcome without surgery, an observation that can aid in decision making.
PREOPERATIVE PLANNING FOR REOPERATION

Since most reoperative candidates have had symptomatic atherosclerosis for at least
a decade, they often have far advanced diffuse coronary and noncoronary atherosclerosis.
Routine preoperative or intraoperative echocardiography is used to detect the presence
of severe ascending and descending aortic atherosclerosis, a problem that may alter
strategies for grafts and graft inflow. Preoperative computed tomography (CT) scan of
the chest in addition to TEE in patients requiring circumflex CABG to evaluate the
descending aorta as a site for proximal anastomosis is helpful. Patients with carotid bruits
undergo preoperative noninvasive screening tests for carotid stenosis for possible staged
or combined carotid endarterectomy.
Limited availability of bypass conduits is a common problem during reoperation,
particularly if the patient has had multiple previous operations. The availability of lower
extremity veins with preoperative venous Doppler studies of the greater and lesser saphe-
nous vein systems can be helpful. The surgeon needs to be aware of previous abdominal
procedures that may have made the inferior epigastric arteries or the right gastroepiploic
artery unusable. If radial artery grafts are a possibility, Doppler studies can demonstrate
the integrity of the collateral arterial supply to the hand. The presence of patent inferior
epigastric arteries can also be documented noninvasively. If conduits are extremely
limited, it is also wise to perform a preoperative angiogram of the internal mammary
arteries to make certain they are not atherosclerotic and were not damaged during the
previous sternal closure. Lastly, while not ideal, cryopreserved veins are commercially
available for use.
ANESTHESIA AND INTRAOPERATIVE MONITORING

Routine hemodynamic monitoring includes a radial artery (sometimes femoral artery)
line and a continuous mixed venous oxygen saturation pulmonary artery catheter. After
routine double-lumen endotracheal tube placement, a transesophageal echocardio-
graphic probe is inserted. External defibrillator pads (R2 stat·<i>padz™, Zoll, Inc.,
Burlington, MA) are routinely placed prior to skin prepping. Cardiopulmonary bypass
machine is always available in the room but not primed.
SURGICAL TECHNIQUE
After exposure of the target coronary vessels by a variety of incisions and operative
exposure techniques described below, the patient is routinely heparnized before the local
coronary occlusion. The dose of intravenous heparin consists of a bolus intravenous
administration of 10,000 IU to keep the activated clotting time (ACT) within a range of
300–350 s. ACT is repeatedly measured every 20–30 min in multivessel grafting prior
to each coronary occlusion to make sure that it is in the 300–350-s range, and additional
bolus of heparin is given if it is required. This is important in off-pump surgery since there
is a hypercoagulable tendency in some patients as the length of the operation increases.
For coronary stabilization, both mechanical stabilizer and epicardial radial sutures on
either side of the coronary target site are used routinely. CTS (Cardio Thoracic Systems,
Cupertino, CA) MIDCAB Stabilizer platform for anterior thoracotomy and CTS Ultima™
Offset Stabilizer are used routinely in our institution in all other incisional
approaches. Local coronary occlusion of a short segment of the artery (2 cm) is obtained
with silastic sutures on a blunt needle encircling the entire artery, epicardial fat, and veins
placed both proximal and distal to the anastomotic site.
A blower device (Axius™ CO2 blower, Guidant Corp., Cupertino, CA) is used to keep
the field clear of blood and incised edges of the coronary artery separated during the
anastomosis. Use of intracoronary shunt is not routine, and contingent on the size of
the target and the severity of anatomical obstruction. More frequently, the shunt is used
if the size of the coronary artery is ≥2 mm and the degree of coronary artery obstruction
is < 75%. Coronary anastomosis is performed using 7-0 or 8-0 continuous Prolene suture.
Heparin is routinely neutralized with protamine at the end of anastomosis. No intrave-
nous antiarrhythmics are given routinely. If the coronary artery is <2 mm, diffusely
diseased, and calcified, and if the patient has a patent intracoronary stent, clopidogrel
(Bristol Myers Squibb, New York, NY) is administered via nasogastric tube prior to the
coronary artery occlusion and postoperatively and continued daily for 4 wk. Patients are
usually extubated either in the operating room before transfer to the intensive care unit
(ICU) or within 1–2 h after arrival in the ICU.
Strategies for perioperative pain control included intercostal nerve block (0.5%
bupivacaine), epidermal catheter, and, recently, intercostal nerve protectors and intercos-
tal cryoanalgesia. Discomfort due to rib retraction in thoracotomy approaches usually
subsides within 2 d compared to the sternotomy approaches. This discomfort during the
initial recovery period has been dramatically reduced with the use of new disposable
intercostal nerve protectors (Deflector™, ViaMedics, White Bear Lake, MN) and appli-
cation of cryoanalgesia (Cryohit, Galil Medical, Israel) to the neighboring intercostal
nerves before closure. Cryoanalgesia technique results in temporary damage to the nerve’s
myelin sheath without significant damage to the underlying nerve axon. Interrupting the
myelin sheath stops the transmission of pain stimuli, and the sheath regenerates quickly
(2–3 wk) without incident. Leaving the axon intact minimizes the nerve axon regenera-
tion and the chance of developing a subsequent painful neuroma. The pain control with
this method extends over 2–3 wk postoperatively.
OPERATIVE APPROACHES
Standard Resternotomy
Standard resternotomy is ideal for patients with no prior patent anterior (i.e.,
LIMA→LAD, SVG→RCA) grafts and who need multivessel grafting.
STERNAL REENTRY
To perform the median sternotomy, the skin and subcutaneous tissues are divided
sharply, the sternal wires are exposed and cut anterior to the sternum but are not removed.
While an assistant elevates both costal margins with retractors, an oscillating saw is used
to divide the sternum. The position of the wires posterior to the sternum helps to avoid
damage to the underlying structures.
MEDIASTINAL DISSECTION
Once the sternum is reopened, the two sides of the sternum are dissected away from
the underlying structures. On the right side, the pleura usually can be entered safely,
and this is also possible on the left side unless a patent ITA graft is present. Complete
dissection of the cardiac structures and old bypass grafts is not needed at this time. It is
safest to enter the right and left pleural spaces by dissection along the diaphragm, because
at that level only a right coronary graft is exposed to possible injury. Once entry into the
left pleural space is accomplished at the level of the diaphragm, the mediastinal struc-
tures, including a patent left ITA, can be separated from the chest wall by dissection
proceeding superiorly.
EXPOSURE OF LATERAL AND INFERIOR WALL DURING RESTERNOTOMY
The following technique greatly facilitates a good exposure for performance of coro-
nary anastomosis to circumflex branches and distal right coronary artery branches.
1. Limiting the posterior pericardial dissection to anterior to the A-V groove. Posterior
pericardial sutures are placed at the junction of the pericardium with its attachment to the
A-V groove.
2. Opening of the right pleura widely with dissection of right ventricle and right atrium. This
allows the heart to fall into the right pleural cavity during exposure of the posterior wall
of the left ventricle.
3. Liberal use of the new apical suction cup device (Xpose Access Device™, Guidant/CTS,
Cupertino, CA) (Fig. 1) to elongate and pull the heart out of the mediastinal cavity.
4. Division of the attachments of the diaphragm to the anterior costal margin bilaterally.
This maneuver creates more space in the chest to move around the heart freely.
5. Placing the sternal retractor from the suprasternal notch down. Then the handle of the
retractor is no longer in the way during operation on the inferior and posterior surface of
the heart.
6. Partial vertical dissection of central tendon of the diaphragm posteriorly and downward
traction on the diaphragm with heavy traction sutures fixed to the abdominal wall. This
naturally moves the posterior part of the heart to the right and out of the pericardial cavity.
7. Trendelenburg position.
Fig. 1. Use of the Xpose™ apical device to facilitate posterior exposure.
These steps provide an excellent exposure for all the branches of circumflex artery and
posterior descending branch of the right coronary artery.
Routine skeletonization of the internal mammary artery (right or left) grafts, use of
composite radial artery grafts, and increasing use of the right gastroepiploic artery have
made it possible to use multiple arterial grafting in most of the reoperative procedures.
With bilateral mammary artery use in reoperative patients, the radial artery composite
grafts are performed end to end to the right mammary artery to revascularize the circum-
flex branches. The RIMA composite graft is frequently routed anterior to the aorta.
Sequential anastomoses are used liberally.
If saphenous vein grafts (SVG) are used in reoperation, the proximal anastomosis is
performed to the old hood of the proximal vein graft. Recently, St. Jude Symmetry™
Aortic Proximal Connector (St. Jude Medical, St. Paul, MN) has been used frequently to
avoid any clamping of the aorta.
Fig. 2. Classical anterior MIDCAB approach. Angiogram (sequential LIMA to LAD and diagonal
arteries) and size of incision are depicted.
For patients who need single or double bypass, minimally invasive direct coronary
artery bypass (MIDCAB) approaches through anterior minithoracotomy, short lateral
thoracotomy, transabdominal incision, and subxyphoid ministernotomy are used.
Anterior MIDCAB
Anterior MIDCAB is most frequently used for LAD and diagonal revascularization in
patients with closed or compromized prior SVG graft to the LAD. Patient is placed in a
left semianterolateral decubitus position with a roll under the left scapula. An 8-cm
incision over the fourth left intercostal space, with two-thirds of the incision medial and
one-third lateral to the nipple is made (Fig. 2). The pleura is routinely entered. The left
lung is collapsed and the ventilation continues with single right lung ventilation. Routine
use of a reusable IMA access retractor system (Thoralift, U.S. Surgical, Norwalk, CT)
(Fig. 3) creates a wide visual tunnel for exposure of the entire LIMA, allowing its mo-
bilization up to its origin under direct vision. This retractor has several different length
blades adaptable to different shapes of the chest. Moreover, by reversing the retractor to
the lower chest, the LIMA can be mobilized to its bifurcation at the sixth intercostal space,
thus allowing routine sequential anastomoses to the LAD, diagonal and marginal (Fig. 2).
In reoperative patients, there is very little endothoracic fascial scarring on the left
lateral side of the LIMA pedicle, which makes routine skeletonization of the LIMA easier
than in the sternal approach.
PREPARATION OF THE LIMA
Prior to the division of the LIMA at the sixth intercostal space, 5000 U of intravenous
heparin are given. The LIMA graft is prepared with separate intraluminal injection of
verapamil (5 mg in 30 cc of heparnized saline—4 cc injection) and papaverine hydrochlo-
Fig. 3. Thoralift MIDCAB retractor.
ride. Injections are administered without clamping the IMA graft, and no hydrostatic
dilatation is used. The distal end of the IMA graft is clipped and allowed to auto-dilate
while the coronary artery target site is prepared. This technique for preparation of the
IMA graft has been used at our institution over the past decade and has been shown to
eliminate any IMA spasm in more than 2000 patients undergoing multiple arterial bypass
grafting. Indeed, the arterial conduits are quite large by the time the coronary anastomosis
is ready to be performed.
In situations when composite T- or Y-grafting from the LIMA is chosen for grafting
of diagonal or ramus intermedius branches through this operative approach (Fig. 4), it is
essential to construct these anastomoses to the LIMA prior to creation of the LIMA–LAD
anastomosis for proper assessment of the length of these grafts.
EXPOSURE OF THE LAD AND ITS BRANCHES
After the LIMA is prepared, the IMA retractor system is replaced by the CTS MIDCAB
stabilizer platform. Exposure of the target coronary artery (LAD) is easy in reoperation
since one can feel the old graft through the pericardium, which easily leads one to the
coronary artery just distal to the old anastomotic site. No extensive intrapericardial dis-
Fig. 4. Midlateral MIDCAB approach.
section is necessary. More extensive dissection will result in troublesome venous bleed-
ing and reduced natural stabilization provided by adhesions. Pericardial traction sutures
are rarely needed.
A pericardial incision is made anterior to the phrenic nerve. With an intramyocardial
LAD, the intramyocardial tunnel is incised to expose the vessel after antegrade stabili-
zation of the heart with a mechanical stabilizer is achieved. The LAD is then “unearthed”
by using two lateral epicardial sutures with tiny Teflon pledgets placed on the sides of the
intramyocardial tunnel. In patients with a prior closed IMA graft to LAD, the right
gastroepiploic artery graft is harvested via a small upper laparotomy incision and brought
through the diaphragm into the left chest to be anastomosed to the LAD.
Subclavian MIDCAB
Subclavian MIDCAB consists of using the radial artery or saphenous vein graft arising
from the left axillary artery to graft the LAD or diagonal and ramus intermedius targets
(Fig. 5). This approach is excellent for patients with prior patent IMA who need diagonal
and/or ramus grafts and also in patients with prior occluded IMA who need grafting of
the LAD and its branches.
The anterior thoracotomy incision is made again over the left fourth intercostal space,
but placed lateral to the nipple to expose the LAD, diagonal and ramus intermedius
branches. It is essential to free the adhesions of the medial aspect of the left upper lobe
from the apex down to the left hilum, to make sure the graft lies along the medial surface
of the hilum to avoid kinking. The coronary target site is exposed as in the anterior
MIDCAB. A separate incision is made in intraclavicular fossa one to two fingerbreadths
below the left clavicle. The left axillary artery is identified after separating the pectoralis
major and pectoralis minor muscle fibers (sometimes division of pectoralis minor muscle).
Fig. 5. Subclavian MIDCAB approach.
After division and retraction of these muscles, the clavipectoral fascia is identified and
incised. The axillary vein is identified. The axillary artery is inferior to the vein and in
most instances the axillary vein has to be retracted away. Heparin is given intravenously
(5000 U) and the axillary artery is clamped proximally and distally. A vertical incision
from anterior to posterior is made in the inferior aspect of the axillary artery.
The anastomosis of radial artery or SVG graft is performed as a T-graft with the heel
of the graft placed posteriorly and inferiorly. The graft is brought in the left chest through
Fig. 6. Lateral MIDCAB approach.
a sizable hole made in the third intercostal space membrane of the muscle. The rest of the
operation is done as in the anterior MIDCAB LIMA-to-LAD grafting. Sequential com-
posite (SVG) grafting is used liberally in this approach.
Lateral MIDCAB
Lateral MIDCAB is a limited (3-in.-) access sternal-sparing thoracotomy to the lateral
wall of the heart (Fig. 6). This is an operation well suited for patients who present for
revascularization of circumflex branches with prior patent IMA graft. These patients
generally fall into three groups:
1. Prior patent LIMA with closed SVG graft to circumflex branches
2. Prior patent LIMA with SVG graft stenosis that has failed PTCA
3. Prior patent LIMA with progressive stenosis of native circumflex branches that are
unsuitable for or failed PTCA
Standard anesthetic techniques and monitoring is used with the patient initially in a
supine position. Double-lumen endotracheal intubation is routinely used. Radial artery
and SVGs are harvested by endoscopic technique, the grafts are prepared, and the incision
is closed. The patient is then turned to the right lateral decubitus standard left posterolat-
eral thoracotomy position. The pelvis and legs are externally rotated (45%) for any
emergent access to femoral vessels (i.e., IABP, CPB.) The 3-in. skin incision is made two
fingerbreadths below the tip of the scapula, with two-thirds of the incision anterior and
one-third posterior to the scapula. Extension of this incision anteriorly is required if
grafting of the anterior ramus intermedius branches is necessary in addition to the mar-
ginal branches. Posterior extension is required rarely except in instances when the pre-
operative graft inflow site is determined to be in the proximal left subclavian or proximal
descending thoracic aorta and the aortic arch. Most commonly, for grafting of the first and
second marginal branches of the circumflex, the fifth intercostal space is entered. The left
lung is deflated. In patients with prior multiple operations (more than one coronary
reoperation), there may be extensive adhesions between the lung and the chest wall. In
these patients, the ribs are retracted and lifted up with a table-mounted IMA retractor
hook (Rultract) and sharp dissection of the lung from the chest wall is performed under
direct vision, just enough to get a small Finnochieto chest retractor placed in the fifth
intercostal space. Further dissection is completed within the chest.
The fourth intercostal space is entered for grafting of the ramus intermedius and the
sixth for grafting of the posterolateral and left posterior descending coronary artery
branches from the circumflex system. Marsupialization of the serratus anterior and latis-
simus dorsi muscle with heavy retraction sutures placed through the muscle and brought
to the skin and anchored well above and below the skin incision is routine and improves
the operative exposure, allowing maximal working space within the chest cavity with the
chest retractor in place. Excision of small portions of the posterior part of the fifth and
sixth ribs is optional, and occasionally the entire rib is excised. At our institution we have
routinely used the CTS Ultima™ Offset Stabilizer, which has excellent retractor blades
and gives maximum exposure. This retractor is placed with its handle in the medial aspect
of the incision. With the lung clamp retracting the left lower lobe, the inferior pulmonary
ligament is incised with cautery, the left lower lobe is dissected from the anterior surface
of the distal descending thoracic aorta, and the anteromedial portion of the left lower lobe
is dissected away from the pericardium up to the left hilum until the left pulmonary vein
becomes visible. In most instances, the lung adhesions to the pericardium are loose and
thin and easily dissected away. In patients with prior multiple reoperations with dense
adhesions, only the anteromedial surface of the left lower lobe is dissected away from the
pericardium and the lateral aspect of the left lower lobe is dissected from the descending
thoracic aorta. In this situation the graft is then routed through a small hole in the adhesion
between the left lower lobe and the anterior surface of the descending thoracic aorta to
be brought to the coronary artery target site. Once the left lower lobe is liberated, this part
of the lobe is then packed superiorly into the apex of the thorax with laparotomy gauze
pads, thereby providing uninterrupted exposure to the entire surface of the heart poste-
riorly (Fig. 7). The diaphragm is pulled downward with heavy traction sutures placed
over the dome of the diaphragm, pulled through the skin at the seventh intercostal space
in the anterior axillary line, and tied over a small rubber bolster. Any long nerve retractor
hook available in any operating room or the Heartport retractor hook is ideal for this
maneuver. This operative maneuver increases the exposure to the posterolateral branch
and the left posterior descending coronary artery branch from the circumflex. The peri-
cardium is palpated to locate the old vein grafts when present. In general, the pericardium
is opened posterior to the phrenic nerve for grafting of all the circumflex branches except
for the ramus intermedius branch, which is exposed via pericardial incision made anterior
to the phrenic nerve. Limited dissection of the pericardium is all that is needed to locate
the target vessel, usually tracing the old grafts to its anastomotic site. More extensive
dissection results in troublesome venous bleeding and reduces the natural stabilization
provided by the adhesions. Dissections are rarely carried out anteriorly, and the left
internal mammary artery graft is usually not encountered in this operation. After identi-
fying the target vessel, the attention is usually turned to the proximal anastomosis.
Advantages of performing the proximal anastomosis first include more accurate mea-
surements and routing of the distended graft conduit, the opportunity to flush atheroscle-
rotic debris from the graft before the completion of the distal anastomosis, and immediate
Fig. 7. Exposure of the pericardium on MIDCAB approach.
establishment of flow upon the completion of distal anastomosis. The descending tho-
racic aorta is assessed below the hilum further by transesophageal or epiaortic
echocardiography. Frequently, the proximal anastomosis is performed in the distal
descending thoracic aorta midway between the lung hilum and the diaphragm. The heel
of the anastomosis is placed proximally, with the toe pointing toward the diaphragm. The
proximal anastomosis is completed with the aid of a specially made side-biting clamp on
the descending thoracic aorta. This clamp (Scanlan, Inc., St. Paul, MN) has a long handle
with a short jaw and is ideally suited for a lateral MIDCAB because of the deep position
of the descending thoracic aorta in this approach (Fig. 8). The handle of the clamp stays
well away from the chest wall and the anastomosis proceeds uninterruptedly with con-
tinuous 6-0 Prolene sutures. Recently, a St. Jude Symmetry™ Aortic Connector (St. Jude
Medical, St. Paul, MN) has been frequently used for nonclamp technique and is especially
useful in this operation due to the deep position of the descending thoracic aorta. If the
descending thoracic aorta is atherosclerotic, and there is not a suitable site to be obtained
for inflow, the left subclavian artery is used as an inflow. To expose the left subclavian
artery, posterior extension of the skin incision is necessary. The proximal portion of this
vessel is dissected circumferentially, a side-biting clamp is placed on the artery, or the
vessel is occluded with vessel loops. In patients with a patent LIMA graft, a 5-min test
occlusion of the subclavian artery is undertaken before opening the artery. During the test
occlusion, if there are no electrocardiographic or hemodynamic signs of ischemia, the
artery is opened and the proximal graft anastomosis is completed with 6-0 Prolene
sutures. If the test occlusion is not tolerated, alternative sites for graft inflow in the aortic
arch or in the proximal descending thoracic aorta are chosen. If the arch and the proximal
descending thoracic aorta are also heavily atherosclerotic, the left lateral thoracodorsal
artery or the left axillary artery have been used for inflow without changing the position
of the patient. The lateral thoracodorsal artery is exposed by making a vertical anterior
Fig. 8. Aortic clamping in midlateral MIDCAB.
incision in front of the anterior border of the latissimus dorsi and the left thoracodorsal
artery. The nerve and vein are identified and carefully separated. End-to-end anastomosis
of the graft to the left thoracodorsal artery is performed using fine 7-0 Prolene sutures,
and this composite graft is then dropped through a sizable hole in the third intercostal
membrane in the anterior axillary line and routed in the major fissure between the upper
and the lower lobe of the left lung anterior to the hilum to be anastomosed to the target
vessel of the circumflex system. The left axillary artery is approached via an incision
posterior to the anterior axillary fold. The axillary artery and vein are identified and
anastomosis of the graft to the axillary artery is performed; the graft is then dropped
through the third intercostal membrane, coming anterior to the left hilum to be anasto-
mosed to the circumflex branch. In this instance, it is important to mobilize the left upper
lobe of the lung from the anterior medial surface of the pericardium, the aortic arch, and
the subclavian artery to make the graft lie in a gentle straight route without any kinking.
Distal coronary artery anastomosis is completed in a standard fashion running 7-0 polypro-
pylene sutures. The use of intracoronary shunts is optional; the orientation of the graft in
most instances is in antegrade fashion with the heel of the anastomosis placed proximally
and the toe distally in the coronary artery target site. Occasionally, when the descending
thoracic aorta is atherosclerotic and only the lower part of descending aorta is found to be
a good inflow site, then anastomosis of the coronary artery target site is performed in
retrograde fashion, utilizing the toe pointing toward the proximal and the heel pointing to
the distal end of the coronary target site. In all instances the graft has to be tacked very gently
over the surrounding tissues to form a nice gentle curve, and the lung is usually inflated after
the completion of the anastomosis to assess the final position of the graft. Sequential
anastomoses are liberally used in this operation for multivessel grafting.
A recently popularized variation of the lateral MIDCAB is the sternal sparing Thoracab
approach (18). The left thoracotomy incision for a Thoracab is longer (4 in.), higher on
the chest wall, and more anterior. The proximal source for graft inflow is the ascending
aorta with either a side-biting clamp or now a facilitated anastomotic connector. All the
native coronary arteries can be visualized and grafted on a beating heart, including the
harvesting of the LIMA with grafting to the LAD. With innovative apical suction cup-
like devices, the heart can be rotated through this small incision to expose the inferior
surface for multivessel CABG on a beating heart entirely through a nonsternal incision.
The thoracab approach is being used more often and is particularly good when multivessel
beating-heart grafting involves extensive grafting of sites on the lateral wall. The skin
incision can also be placed cosmetically under the left breast fold in female patients.
Subxiphoid MIDCAB
The subxiphoid MIDCAB operative approach is ideally suited for grafting of the mid
and distal RCA prior to the crux and the proximal part of posterior descending coronary
artery with a high origin from the RCA. The right gastroepiploic artery is used as a graft
(Fig. 9). With the patient in supine position, a vertical incision (3 in.) is made from the
xiphisternum to midway between the umbilicus and the sternum (Fig. 9). The linea alba
is incised and a small V-shaped piece of the xiphisternum is excised. A table-mounted
IMA retractor (Rultract) with hooks is placed on the left side of the table and the thoracic
cavity is partially lifted with the hook placed underneath the xiphisternum. Partial detach-
ment of the costal attachment of the diaphragm on both sides gives exposure to the mid-
right coronary artery and the inferior surface of the right ventricle. Downward sutures on
the diaphragm tied to the abdominal wall further increase the exposure. The peritoneum
is entered and the gastroepiploic artery is harvested as a pedicle, using a Harmonic
scalpel. A vertical incision of the central part of the diaphragm is made and the gastro-
epiploic artery is brought over the diaphragm into the mediastinal cavity. A mechanical
stabilizer (CTS Ultima™ Stabilizer) is placed vertically to expose the target vessel for
coronary artery anastomosis. The stabilizer is used with the handle placed from either
the superior or inferior aspect of the incision and the coronary artery is stabilized. The
anastomosis is then made in a standard fashion, again using 7-0 Prolene sutures. In a large
heart with increased anteroposterior dimension with difficulty in exposure of the distal
coronary artery target, the incision needs to be extended into the lower part of the sternum
with lateral division of the sternum into the right fifth intercostal space. The distal right
coronary artery and posterolateral branch are difficult to access in most instances in this
approach.
Transabdominal MIDCAB
Transabdominal MIDCAB is a newer minimally invasive direct coronary artery by-
pass operation specifically designed to achieve single and multivessel coronary
revascularization in patients including those with distal disease and “full metal jacket”
syndrome following prior PTCA in the LAD and RCA, reoperative coronary surgery, and
to perform simultaneous bypass via a single incision to the left anterior descending and
the right coronary artery and its branches. Our preliminary anatomic observations indi-
cated the following:
1. The rectus abdominus muscle has the strongest downward pull on the lower part of the
sternum, so if it is divided the lower sternum can be lifted up farther than the usual
position by retractors.
2. When the rib cage is elevated during normal respiration, the lower rib projects directly
forward so that the sternum also moves away from the spine, increasing the anteropos-
terior diameter of the chest by 20%.
3. By dividing the rectus muscle bilaterally and allowing the costal arch to be lifted ante-
riorly and dividing both costal attachments of the diaphragm, the entire abdominal vis-
cera fall posteriorly, thus increasing the operative exposure and the operative angle to
about 60° from the vertical midline of the abdominal incision.
Fig. 9. Subxyphoid MIDCAB approach.

Fig. 10. Transabdominal exposure.
The operation is performed with the patient in the supine position. A 3-in. curvilinear
epigastric incision is made below the xiphoid and the costal margins. Both rectus
abdominus muscles with their anterior and posterior sheaths are divided, preserving the
lateral neurovascular bundles. The diaphragmatic costal attachments are sharply released
with cautery on both sides to further facilitate exposure of the heart. A V-shaped piece
of xiphoid is excised, a table-mounted left internal mammary artery (Rultract) retractor
is placed on the left side, and the two hooks are placed underneath the xiphoid and the
costal arch. The thoracic cage is lifted to expose the heart (Fig. 10). Pericardial adhesions
to the dome of the diaphragm are released. The sternal pericardial adhesion underneath
the surface of the sternum and the fibrous adhesion between the heart and the left chest
wall are released with cautery, thus dropping the entire mediastinal structure away from
the chest wall posteriorly. Because of the bucket-handle movement of the lower ribs, it
is possible to lift the lower sternum and the lower costal arch to enable the dissection of
the bilateral mammary arteries up to the second or third space under direct vision.
Recently we have used bilateral thoracoscopic techniques from both sides of the chest to
facilitate mammary harvesting, and during the last 6 mo we have added robotic
telemanipulation to isolate both mammary arteries through the left chest and then com-
plete the dissection of the inferior part of the bilateral mammary pedicle via the transab-
dominal incision for multivessel arterial CABG. The right gastroepiploic artery is
harvested after opening the peritoneum. The exposure of the left anterior descending
coronary artery is facilitated by deep lateral pericardial retractor sutures with traction
downward and to the right side of the abdominal wall. Multiple deep posterior pericardial
and diaphragmatic sutures with downward traction to the abdominal wall provide excel-
lent exposure for the posterior descending and posterolateral branches and the distal right
coronary artery prior and after the crux. A vertical 2-in. incision in the central tendon of
the diaphragm with traction sutures on both sides toward the abdominal wall further
improves exposure of the target vessels, with the heart dislocated in a vertical position
with the apex of the heart pointing toward the sternum. In some patients with diffuse
in-stent restenosis of the entire RCA with previous bypass graft failure, transperitoneal
exposure of the right coronary artery is performed. The triangular ligament is incised and
the left lobe of the liver is displaced to the right and packed with a laparotomy pad, and
then a transverse diaphragmatic incision is made. The stent is palpated and the incision
Fig. 11. Transdiaphragmatic incision with stabilization in place.
is then carried out over the coronary artery target site beyond the limit of the stent
(Fig. 11). The CTS OPCAB retractor is placed either vertically or transversely and the
incision is spread. The stabilizer is always placed on the left side and on the inferior part
of the retractor. Coronary artery anastomosis is performed in a standard fashion as
described in the surgical techniques section. The pleural cavity is rarely entered in these
patients, and a small pericardial Blake drainage tube is placed in the posterior part of
the mediastinum over the diaphragm, brought through the skin at the lateral aspect of the
incision. The incision is closed in layers and the patient is extubated intraoperatively or
within the first hour in the ICU. The postoperative invasive monitoring lines are
removed within 6 h, and ambulation starts within 24 h. Most patients stay in the hospital
for 1–2 d prior to discharge (Fig. 12). In addition to the use of bilateral mammary and right
gastroepiploic arteries, composite grafting with saphenous vein graft or radial artery
conduit anastomosed to the arterial pedicled grafts are sometimes necessary.
RESULTS OF REOPERATIVE OFF-PUMP

CABG AT LENOX HILL HOSPITAL
Between April 1994 and December 2001, 499 patients underwent reoperative CABG
at our institution. This represents 12.5% of all isolated coronary reoperations during this
period. One hundred of these patients had their CABG done on-pump, while the remain-
ing 399 patients had off-pump surgery. Since April 1999, intent to treat all isolated CABG
Fig. 12. Postoperative transabdominal MIDCAB incision.
(primary of reoperative) patients with an off-pump approach was initiated at our institu-
tion. No reoperative CABGs have been performed on-pump since that time. Of the total
399 off-pump cases, 346 (87%) were performed via nonsternotomy, minimal-access
approaches. During the last 18 mo, rare patients have had sternotomy reoperative OPCAB.
The demographics of patient population who had reoperative off-pump CABG via
MIDCAB and CABG are shown in Table 1. With experience in minimal access off-pump
surgery, the percentage of patients receiving double grafts has increased to 25%. The
lowest risk-adjusted mortality is seen in the MIDCAB group, which is better than most
series reported on pump reoperative CABG. Postoperative complications and morbidity
have also significantly decreased with reoperative off-pump CABG. Strikingly, stroke is
practically eliminated with off-pump approaches, especially in view of the high preop-
erative risk factors associated with this patient population.
The beneficial effect of off-pump approaches is seen even in subgroups including
various MIDCAB approaches (Tables 2–4). Of note is the increased percentage of patients
receiving double grafts (29.3%) and the absence of stroke in transabdominal and lateral
MIDCAB. With increased use of LIMA-to-LAD grafting in the last decade and a half and
the continued deterioration of SVG grafts to the circumflex, there has been aggressive use
of stenting and PTCA procedures for these patients. Unfortunately the results of SVG
graft intervention by PTCA or stenting have been dismal. These patients with prior patent
LIMA are frequently referred for reoperative CABG. Most of them need one or two grafts
to the circumflex branches in the setting of a patent LIMA and with several occluded vein
grafts to other regions of the heart. Hemodynamical stability may depend on the integrity
Table 1
Demographics and Outcomes Among 346 Patients
Undergoing Reoperative MIDCAB Approaches (1994–2001)
Preoperative demographics
Peripheral vascular disease 194 56.1
Multiple previous MI 71 20.5
History of CHF 207 59.8
COPD 41 11.8
Extensively calcified aorta 46 13.3
LVEF ≤ 30 92 26.6
Postoperative complications
None 316 91.3
Stroke 2 0.6
Q-wave MI 0 0
Deep sternal wound infection 0 0
Renal failure 1 0.3
Risk-adjusted operative mortality 0.76
Number of grafts
Single 260 75.1
Double 79 22.8
Triple 7 2
Grafts/patient 1.33
Table 2
Undergoing Reoperative Anterior MIDCAB (1994–2001)
COPD 22 13.2
LVEF ≤ 52 31.1
None 144 86.2
Stroke 2 1.2
Q-wave MI 0 0
Renal failure 1 0.6
Number of grafts
Single 126 75.4
Double 37 22.2
Triple 4 2.4
Grafts/patients 1.21
of the LIMA during reoperative dissection. Even with careful transsternal approach for
reoperative grafting in experienced institutions, others have documented that the LIMA
is still damaged 5% of the time. When this happens, operative mortality rises threefold,
176 Table 3 Part II / Minimally Invasive Bypass Grafting
Undergoing Reoperative Transabdominal MIDCAB (1998–2001)
Peripheral vascular disease 39 52
Multiple previous MI 15 20
Hypertension 51 68
History of CHF 45 60
COPD 9 12
LVEF ≤ 30 18 24
None 66 88
Stroke 0 0
Q-wave MI 0 0
Renal failure 0 0
Number of grafts
Single 52 69.3
Double 22 29.3
Triple 1 1.3
Grafts/patient 1.3
Table 4
Undergoing Reoperative Lateral MIDCAB (1996–2001)
COPD 10 8.8
Extensively calcified aorta 16 14
LVEF ≤ 30 22 19.3
None 106 93
Stroke 0 0
Q-wave MI 0 0
Renal failure 0 0
Number of grafts
Single 82 71.9
Double 30 26.3
Triple 2 1.8
Grafts/patient 1.31
from the expected 3% to 9%. When a limited-access approach, such as lateral MIDCAB,
is used, the patent graft is not disturbed and the operative mortality is reduced to 0.5% as
shown by our experience.
Multivessel coronary artery disease after prior CABG may present with more than
one ischemic myocardial region that cannot be fully addressed via a single limited-access
incision. To resolve this, we have been increasingly using two separate minimal-access
incisions to address both regions. In some instances we have operated at two different
times separated by 1–2 mo with different access approaches.
The experience with evolving facilitated anastomotic technology will further enhance
the utility of minimal reoperative CABG. The St. Jude Symmetry™ Aortic Connector
(St. Jude Medical, St. Paul, MN) in lateral MIDCAB has already simplified this opera-
tion. Novel magnetic connectors (19) are being evaluated for distal coronary arterial
anastomosis in clinical trials in Europe (see Chapter 30). These devices combined with
limited-access incisions and robotic assistance, we believe, will pave the way for outpa-
tient reoperative CABG.
REFERENCES
1. Loop FD, Lytle BW, Cosgrove DM, et al. Influence of the internal-mammary-artery graft on 10-year
survival and other cardiac events. N Engl J Med 1986;314:1–6.
2. Cameron A, Kemp HG Jr, Green GE. Bypass surgery with the internal mammary artery graft: 15 year
follow-up. Circulation 1986;74(suppl III):III-30-6.
3. Lytle BW. Coronary reoperations. In Franco KL, Verrier ED, eds. Advanced Therapy in Cardiac Surgery.
B. C. Decker, 1999;84–99.
4. Rosengart TK. Risk analysis of primary versus reoperative coronary artery bypass grafting. Ann Thorac
Surg 1993;56:S74–S77.
5. He GW, Acuff TE, Ryan WH, et al. Determinants of operative mortality in reoperative coronary artery
bypass grafting. J Thorac Cardiovasc Surg 1995;110:971–978.
6. Loop FD, Lytle BW, Cosgrove DM, et al. Reoperation for coronary atherosclerosis: changing practice
in 2509 consecutive patients. Ann Surg 1990;212:378–386.
7. Fanning WJ, Kakos GS, Williams TE Jr. Reoperative coronary artery bypass grafting without
8. Mohr R, Moshkovitz Y, Gurevitch J, Benetti FJ. Reoperative coronary artery bypass without cardiop-
ulmonary bypass. Ann Thorac Surg 1997;63(suppl 6):S40–S43.
9. Bergsland J, Hasnain S, Lajos TZ, Salerno TA. Elimination of cardiopulmonary bypass: a prime goal
in reoperative coronary artery bypass surgery. Eur J Cardiothorac Surg 1998;14:59–63.
10. Pfister AJ, Zaki S, Garcia JM, et al. Coronary artery bypass without cardiopulmonary bypass. Ann
Thorac Surg 1992;54:1085–1092.
11. Subramanian VA. Clinical experience with minimally invasive reoperative coronary bypass surgery.
Eur J Cardiothorac Surg 1996;10:1058–1063.
12. Grandjean JG, Mariani MA, Ebels T. Coronary reoperation via small laparotomy using right gastroepi-
ploic artery without cardiopulmonary bypass. Ann Thorac Surg 1996;61:1853–1855.
13. Fonger JD, Doty JR, Sussman MS, Salomon NW. Lateral MIDCAB grafting via limited posterior
thoracotomy. Eur J Cardiothorac Surg 1997;12:399–405.
(MIDCAB). Eur J Cardiothorac Surg 2000;17(4):485–487.
15. Varnauskas E, the European Coronary Study Group. Twelve-year follow-up of survival in the random-
ized European Coronary Surgery Study. N Engl J Med 1988;319:332.
16. CASS Principle Investigators, et al. Myocardial infarction and mortality in the Coronary Artery Surgery
Study (CASS) randomized trial. N Engl J Med 1984;310:750.
17. Yusuf S, Zucker D, Peduzzi P, et al. Effect of coronary bypass surgery on survival: overview of 10 year
results from randomized trials by The Coronary Artery Bypass Graft Surgery Trialists Collaboration.
Lancet 1994;344:563–570.
18. Srivastava SP, Kirit NP, Tummala P, et al. Thoracab: an innovative approach to total revascularization
without cardiopulmonary bypass and median sternotomy—Srivastava approach. A report of first 160
consecutive cases. Heart Surgery Forum 2001;4(suppl 2):S80.
19. Subramanian VA. Overview of magnetic technology used to facilitate proximal and distal anastomoses.
Presented at Euro-College on ATCS: The Key to Advanced Techniques in Cardiac Surgery, Sep 15,
2001, Lisbon, Portugal.
Chapter 13 / SPerioperative Evaluation of Graft Patency in OPCAB 179
13 Perioperative Evaluation
of Graft Patency in OPCAB
Vinod H. Thourani, MD
and John D. Puskas, MD, MSC
CONTENTS
INTRODUCTION
INTROPERATIVE GRAFT ASSESSMENT
POSTOPERATIVE GRAFT ASSESSMENT
FUTURE TECHNOLOGIES TO EVALUATE POSTOPERATIVE
GRAFT PATENCY
SUMMARY
REFERENCES
INTRODUCTION
With presently available instrumentation, off-pump coronary artery bypass (OPCAB)
grafting via a median sternotomy can now be performed for lesions in virtually any
coronary artery with a high degree of patient safety and surgeon comfort. Recent reports
have documented excellent short-term outcomes for patients undergoing OPCAB (1,2).
Despite multiple clinical studies evaluating clinical outcomes, there remains a paucity of
literature evaluating the patency of grafts constructed by OPCAB techniques. Although
the introduction of epicardial stabilizers has improved the accuracy with which distal
anastomoses on the beating heart can be constructed, there has been concern that the
technical difficulties of performing OPCAB could possibly increase the risk of technical
anastomotic failure (3). This chapter reviews both the intraoperative and postoperative
methods and outcomes of anastomotic evaluation for patients undergoing off-pump
coronary artery surgery.

179
INTRAOPERATIVE GRAFT ASSESSMENT

Intraoperative assessment of graft patency has not been commonly performed after
conventional coronary artery bypass surgery. Most cardiac surgeons have relied on simple
clinical signs (e.g., electrographic tracings or hemodynamic stability) or manual palpa-
tion (e.g., fingertips or direct probing) to make a diagnosis of coronary graft occlusion.
With the increasing sophistication and popularity of OPCAB, together with the introduc-
tion and improvement in tools to measure intraoperative coronary graft flow, there has
been a revived interest in documenting graft patency. While angiography remains the
definitive test of graft patency, it is time-consuming, often requires cardiology assistance,
and is not without risk. A rapid, safe test of graft patency that can be easily performed by
the operating surgeon has been sought.
Initially, electromagnetic devices were used to measure the intensity of the electro-
magnetic field generated by electrically charged red blood cells flowing within the ves-
sels (4). The intensity of the electromagnetic field generated was used to calculate the
actual blood flow. However, this technology has been abandoned due to the numerous
sources of potential error. These flow measurements were often difficult to obtain,
required perfect placement of probes, and necessitated careful calibration of instruments.
Finally, the measured values are influenced by the hematocrit and the thickness of the
vessel wall (5). Generally, ultrasound technology—Doppler and transit time flow mea-
surement (TTFM)—has replaced electromagnetic techniques (5–7).
Transit Time Flow Measurement

Although the first transit time flow meter was described in 1962 (8), it was not until
1983 that the first flow meter became commercially available. The flow probe consists
of two small piezoelectric crystals, one upstream and one downstream, mounted on the
same side of the vessel. Opposite the crystal is a small metallic reflector. Each crystal
produces a wide pulsed ultrasound beam covering the entire vessel width. The area of the
transducers and the distance the beam has to travel between the two transducers are
known and are used to calculate the flow of blood through the graft. The probe is con-
nected to a computer and the necessary time for an ultrasound beam, emitted from the
upstream crystal to arrive at the downstream crystal after being reflected, and for a signal
from the downstream crystal to reach the upstream crystal is measured. Because ultra-
sound travels faster if transmitted in the same direction as flow, a small time difference
between the two beams is calculated as the transit time of flow (TTF). Thus, the blood
flow in the graft is proportional to the transit time. All calculations are made automati-
cally by the flow meter and are displayed as milliliters per minute. The level of acoustical
coupling is expressed by a color-coded square and as a percentage of the optimal contact.
Measurements are not dependent on the angle between the vessel and the probe. Measure-
ments are also independent of the hematocrit level, heart rate, and thickness of the vessel
wall. Flow curves, together with flow and pulsatile index (PI) values, are displayed in real
time on a video screen.
The PI is a good indicator of the blood flow pattern and, consequently, of the quality
of the anastomosis. This number is obtained by dividing the difference between the
maximum and minimum flow by the value of the mean flow. The optimum PI should be
between 1 and 5; the higher the number, the more suggestive of anastomotic imperfec-
tions (9,10). D’Ancano and colleagues suggest revision of any distal coronary anastomo-
sis with a measured intraoperative pulsatile index ≥ 5 (9,10).
The ability of TTFM to detect less than critical stenosis has not been clearly defined.
Another limit of TTFM is the lack of standard or nominal curves and flow values for
different types of grafts and revascularized vessels. Interpretation of flow curves and
TTFM findings is still empirical and dependent on the surgeon’s personal experience.
Jaber et al. (11) reported that differences in flow tracing morphology were virtually
indistinguishable from fully patent to moderately stenotic anastomoses. Moreover, grafts
with up to 75% stenosis still had predominantly diastolic flow, and only grafts with
greater than 75% stenosis exhibited significantly reduced diastolic flow in a canine model.
In order to improve the objectivity of TTFM in detecting significant differences
between mild-to-moderate and more severe stenosis, Koening et al. (12) have utilized
magnitude and phase-component spectral analysis. Furthermore, Cerrito et al. (13) de-
signed a neural network to represent an “acceptable” anastomosis of less than 50%
stenosis, or an anastomosis that “should be redone” with a stenosis greater than 50%. The
future for clinical application of neural network analysis will require a database of graft
flow measurements correlated to varying degrees of anastomotic stenosis, validated by
angiography. In humans, Salerno and Bergsland have suggested that flow patterns, PI
values, flow values, and clinical findings (e.g., electrocardiographic tracing, hemody-
namic values) should always be evaluated simultaneously with TTFM to complement
important clinical decisions.
Since 1996, Salerno and Bergsland have consistently utilized this technology in
patients undergoing OPCAB and have reported on their extensive experience (6,10,14–17).
In a series of articles on off-pump coronary surgery, they reported 161 patients undergo-
ing a total of 323 distal anastomoses (15). All completed grafts were tested intraopera-
tively with TTFM, and the decision to accept or revise any individual graft was based on
a decision nomogram using key values readily available from the TTFM output. They
noted that 32 grafts (10%) were surgically revised based on unsatisfactory flow curves,
the pulsatile index (PI), or both. All revised grafts were found to have a significant
technical error, such as an intimal flap, thrombus, conduit kinking, or dissection. In a
follow-up study by the same group, the authors (16) reported on 464 patients undergoing
a total of 1002 flow assessments. A total of 57 grafts (6%) of the distal grafts required
revision. The authors (14) have eloquently summarized their clinical recommendations
in the following seven dictums:
1. Flow must be measured with and without proximal occlusion. A graft obstructed at the
toe may show perfectly normal flow if there is significant outflow in the proximal direction.
2. The diastolic flow pattern must be evaluated, as the graft may be compromized with
relatively high mean flow. Correspondingly, a graft can be normal with low mean flow.
3. When in doubt, reevaluation of TTFM at a higher blood pressure and/or through an
intraoperative pharmacological stress test should be performed with nitroglycerin,
papaverine, or adenosine.
4. The presence of air bubbles in the graft may simulate organic obstruction. If this is
suspected, the graft should be de-aired.
5. When a malfunctioning graft is diagnosed, it should be immediately revised to prevent
ischemia and hemodynamic problems during the remaining procedure.
6. Measurement of the graft flow should be repeated after heparin reversal.
7. Final graft verification should be done just before chest closure, since grafts that are too
short or too long may kink on chest closure.
A limited number of studies have determined the validity of the intraoperative transit
time flow measurements of grafts in coronary bypass grafting by correlating TTFM with
intraoperative or postoperative quantitative angiography. In a small series of 35 patients
(28 patients underwent conventional “on-pump” CABG and 7 underwent OPCAB) hav-
ing 82 distal anastomoses, Takami et al. (18) noted that intraoperative flow measurement
reflected precisely the short-term (14 ± 5 d) postoperative angiographic quality of the
distal anastomoses. They note that the calculations based on the fast Fourier transforma-
tion (FFT) of the flow curve allowed distinction of patent from nonpatent grafts. A FFT
ratio (the ratio of powers of the fundamental frequency and its first harmonic) greater than
1.0 corresponded with patent grafts (stenosis less than 20%) and those with FFT ratio less
than 1.0 corresponded with nonpatent grafts. They conclude that power spectral analysis
of the TTFM flow waveform using FFT is useful for intraoperative prediction of anas-
tomotic patency (19).
In contrast, Hol and colleages (20) studied 72 patients (21 patients underwent conven-
tional “on-pump” CABG and 51 underwent OPCAB) undergoing 124 grafts. All patients
underwent intraoperative TTFM, followed by coronary angiogram in the operating room
after chest closure while the patients were under general anesthesia. Follow-up angiog-
raphy of both grafts and the native coronary arteries was carried out at 3 mo in all cases.
Forty-eight grafts were assessed by an additional angiography at 12 mo. The authors
concluded that blood flow measurements (mean flow, PI, and waveform) performed
intraoperatively could not identify significant stenoses in arterial or vein grafts and were
not predictive of long-term graft patency. Furthermore, they believe that blood flow
measurements should be interpreted cautiously and viewed as additional evidence to
other indicators of graft dysfunction.
Even if interpretation of TTFM findings is still based on personal experience and
empirical values, many researchers are focusing their attention on trying to develop
nominal TTFM curves and objective mathematical values to improve the applicability of
this technology. At the present time, TTFM remains the most popular modality among
an armamentarium of intraoperative techniques used to assess OPCAB graft patency.
Doppler Ultrasound
Pulse-wave Doppler ultrasound is another technique commonly employed to assess
intraoperative coronary artery graft patency. The ultrasound pencil probe (generally
5–20 MHz) is acoustically coupled to the vessel by a small amount of sterile gel. Sharp
dissection is not necessary to measure flow in most conduits. The parameters commonly
determined by the pulsed Doppler are flow (L/min), velocity (cm/s), and internal diam-
eter (mm) of the vessel. Furthermore, a total resistance of the graft and coronary bed,
along with a pulsatility index (PI) can also be calculated. Although a variety of investi-
gators have evaluated the use of Doppler ultrasound as a tool of flow assessment in
patients undergoing on-pump conventional coronary artery bypass grafting (4,21–26),
only a few have investigated this tool in OPCAB patients (27–30).
Dr. Chitwood’s group (27) reported the use of simple continuous-wave Doppler flow
assessment of the LIMA graft with concurrent intraoperative angiography via the radial
artery in 50 consecutive patients. They used a nonquantitative and subjective methodol-
ogy for analysis of the Doppler flow signal. Flow was graded as 0 = no flow, 1+ = poor
or questionable signal, and 2+ = satisfactory flow with diastolic augmentation. To avoid

mistaking retrograde for antegrade flow, assessment was made with a proximal tourni-
quet still in place. They demonstrated a 75% sensitivity of the Doppler ultrasound in
detecting actual problems and 94% specificity. Despite the advent of more sophisticated
duplex ultrasound technology, the authors contend that all noninvasive testing may be
subject to technical problems with equipment, user error, and subjective interpretation.
More recently, Magovern and his associates (28) studied intraoperative color pulsed-
Doppler flow methods of graft analysis in 35 patients undergoing elective left internal
mammary artery (LIMA) anastomosis to the left anterior descending (LAD) coronary
artery via OPCAB. In addition, immediate graft patency was determined with intraopera-
tive angiography using selective injection of the left internal mammary artery from a
femoral approach. A normal Doppler study was defined as a diastolic predominant flow
pattern with a pan-diastolic flow velocity of greater than 15 cm/s. There was immediate
perfect patency with brisk flow in 32 patients (91%), all confirmed by intraoperative
angiograms. All patients with abnormal angiograms also had abnormal Doppler flow. In
this study, the three patients with imperfect anastomosis were revised. Interestingly, none
of these patients demonstrated echocardiographic evidence of ischemia, hemodynamic
instability, or any obvious abnormality of the graft. At 2 yr follow-up, all surviving
patients were without angina, and 97% were free from reintervention on the LAD. The
authors conclude that color pulsed-Doppler analysis was 100% accurate for confirming
graft patency and for detecting failed grafts intraoperatively.
Calafiore et al. (29,30) have reported that the pulsed-Doppler flow velocity performed
perioperatively at rest and after the adjunctive Azoulay maneuver (which transiently
augments venous return and increases cardiac output and graft flow) is a reliable tech-
nique to follow up patients who undergo OPCAB. They described 100 patients who
underwent early postoperative Doppler flow velocity assessment and coronary angiog-
raphy. They concluded that normal grafts show an increase in diastolic flow velocity
whereas stenotic grafts do not. These results were confirmed by angiography.
In general, Doppler flow measurements are not as familiar to cardiac surgeons as
angiograms, but they provide useful information and avoid the risks of contrast dye and
catheter-related injury. Doppler technology may be expeditiously utilized by experi-
enced OPCAB surgeons wishing to document the patency of distal coronary anasto-
moses.
INTRAOPERATIVE ANGIOGRAPHY
Angiography is considered the “gold standard” for assessment of anastomotic quality.
However, angiography is not readily available in most operating rooms and is invasive,
costly, and time-consuming. It has therefore been utilized sparingly in the intraoperative
assessment of graft patency during OPCAB (27,31–37).
Chitwood and his colleagues reported that, with some experience, cardiac surgeons
can perform intraoperative LIMA arteriography via the left radial artery in 10–15 min
(31). The LAD–LIMA patency was visualized under direct fluoroscopy in a series of 20
patients without complications. Intervention based on angiography was required in three
patients (15%), two of whom had received sequential LIMA anastomoses. Mild LIMA
or LAD spasm was a common finding, which generally resolved after a second contrast
injection. The authors felt that intraoperative arteriography offered several advantages
over routine postoperative cardiac catheterization prior to hospital discharge: (1) the
results were obtained immediately and interventions performed prior to the patient leav-
ing the operating suite; (2) same-day catheterization requiring transfer of the patient from
the operating room to a catheterization lab was eliminated; and (3) the cost to the patient
for a postoperative coronary catheterization was reduced.
In a follow-up study, Chitwood (27) compared the use of continuous-wave Doppler
flow assessment of the LIMA graft with concurrent intraoperative surgeon-performed
angiography via a radial artery technique in 50 consecutive patients. They suggested that
intraoperative arteriographic assessment of grafts be performed (1) in any patient with
equivocal Doppler flow signals; (2) in any patient with severely diseased native coronary
vessels; (3) in all sequential or complex anastomotic cases; and (4) in any case in which
the surgeon is not entirely satisfied with the anastomosis. Furthermore, they noted that
intraoperative angiography may be helpful to a surgeon early in the OPCAB learning curve.
Other Techniques for Intraoperative Graft Assessment

Takayama reported that intraoperative coronary angiography using fluorescein in 29
consecutive cases had sufficient resolution only to demonstrate whether an anastomotic
stenosis was critical and whether the distribution of graft flow was normal (38). Intraop-
erative fluorescein angiography provided the cardiac surgeon with insufficient objective
data to adequately assess the quality of coronary anastomoses and to make decisions
regarding graft revision. However, since this technique does not require selective cath-
eterization of coronary arteries, it is easy to perform.
Thermal coronary angiography (TCA), or infrared thermography, is a noninvasive
method that requires no catheter, contrast medium, radiation, or interference with the
surgical procedure (39–42). Yet it is similar to angiography in that it gives visual repre-
sentation of grafts and coronaries arteries. Falk and colleagues (42) used this method in
370 on-pump coronary artery bypass patients and successfully documented patency rates
for saphenous vein grafts (90.6%) and LIMA grafts (96.2%). However, owing to ana-
tomical reasons, certain coronary arteries could not be visualized and thermal imaging
was compromised by excessive epicardial fat pads (42). More recently, Suma et al. (37)
used thermal coronary artery imaging with a newer-generation infrared camera (IRIS III)
in 12 normothermic OPCAB patients undergoing 18 distal anastomoses. One LIMA graft
(5.6%) was shown to be closed and was revised. All grafts were restudied by conventional
catheter angiography postoperatively, and all were patent. In this small cohort of patients,
these authors showed that intraoperative noninvasive coronary imaging with a highly
sensitive infrared camera had adequate resolution to detect graft closure, but this tech-
nique still offers inadequate resolution to assess coronary anastomoses qualitatively.
At present no alternative method provides the same information and resolution as
angiography. Nonetheless, only a small minority of surgeons perform routine intraopera-
tive angiography, owing to technical and logistical concerns. It is foreseeable that a new,
less invasive, and equally informative technique of graft assessment might become a
widely accepted tool for intraoperative assessment of anastomoses performed during
OPCAB, but none presently exists.
POSTOPERATIVE GRAFT ASSESSMENT

Although there is an extensive body of literature containing the results of postoperative
angiographic patency rates in conventional coronary artery bypass grafting, only a lim-
ited number of series contain information on postoperative angiographic findings follow-
ing OPCAB.
Results of Early (< 1 mo) Angiography After OPCAB

Schaff and his colleagues (43) from the Mayo Clinic studied 15 of 16 patients with
angiography immediately following minimally invasive direct coronary artery bypass
grafting (MIDCAB). Three patients (20%) required reexploration, and one patient (7%)
required reconstruction of the distal anastomosis.
Subramanian et al. (44) have shown that mechanical stabilizers can profoundly impact
angiographic patency rate. In an off-pump series of 199 patients, 111 patients had post-
operative angiographic evaluation and were divided into two groups. In 44 patients
MIDCAB operations were performed prior to April 1996, utilizing β-blockers and
calcium-channel blockers and intermittent transient cardiac standstill accomplished with
5- to 10-mg intravenous boluses of adenosine. In 67 patients, MIDCAB was performed
after April 1996, using a mechanical stabilizer. Early overall graft patency (less than
36 h postoperatively) was 93% (103 of 111 patients). They found that the patency rate
for the patients in which MIDCAB was performed with a mechanical stabilizer was
significantly improved compared to those performed without a mechanical stabilizer
(97% vs 86%, respectively, p = 0.028).
Similarly, Cartier and his colleagues in Montreal (45) evaluated the quality of LIMA
anastomosis to the LAD in 20 patients who underwent beating-heart coronary bypass
surgery without stabilization and 14 patients who had OPCAB with a stabilization device.
Eight patients in whom the anastomoses were performed without stabilization (8/20,
40%) had stenoses of more than 50%. There was only one stenosis (7%) of more than 50%
of coronary luminal diameter among the patients in whom the operation was performed
with a stabilizer (p = 0.02). In a multi-institutional study, Mack et al. (46) reported
intraoperative (38 patients) or immediate postoperative (62 patients) angiographic evalu-
ation in 100 of 103 patients undergoing LIMA to LAD by MIDCAB. They found
angiographic graft patency of 99%, with Grade A graft patency (no stenosis greater than
50%) being 91%. A total of three grafts (8%) were revised in the operating room; one
patient (2%) underwent reoperation and three more (5%) underwent PTCA. Similarly, in
a small series of 25 patients, Gill et al. (47) completed angiographic follow-up in all
patients within 4–6 h postoperatively and noted a 97.5% (28 of 29 grafts) overall patency rate.
At our institution, one surgeon (J.D.P.) routinely performed early postoperative
angiography for all OPCAB patients who would consent since November 1996 (1,2,48).
A recent review of 378 consecutive patients at Emory University before September 2000
revealed that 287 patients (76%) underwent postoperative cardiac catheterization prior
to hospital discharge (49). The average number of distal anastomoses was 2.8 ± 1.1. The
angiograms were reviewed by a panel of three cardiologists, and quantitative measure-
ments were performed on all grafts and target vessels.
Postoperative graft angiography was performed within 3 d in 95% of the 287 patients.
A total of 737 conduits and 785 distal anastomoses were evaluated. The majority of
conduits utilized were the reverse saphenous vein graft and internal mammary graft
(Table 1). A variety of target vessels were grafted in this series and are listed in Table 2.
At cardiac catheterization, 728 of 737 grafts (98.8%) were patent utilizing the Fitzgibbon
classification (Table 3). Eight of 423 saphenous vein grafts were occluded, one of 278
IMA grafts was occluded, and no radial grafts were occluded. There were no differences
in patency rates of grafts supplying different regions of the heart (Table 4).
At a mean interval of 15 mo (range 2–42 mo), 90% follow-up was obtained in this
population of OPCAB patients. There were no cardiac deaths and no myocardial
Table 1
OPCAB Conduits Studied
Conduit No.
LIMA 269
Saphenous vein graft 423
Radial 36
RIMA 8
Free RIMA 1
Table 2
OPCAB Fitsgibbon Scores by Conduit
Conduit A B A+B O Total
LIMA 247 21 99.6% 1 263
Saphenous vein graft 408 7 98.1% 8 423
Radial 33 3 100.0% 0 36
RIMA 8 0 100.0% 0 8
Free RIMA 1 0 100.0% 0 1
Table 3
OPCAB Target Vessels Studies (n = 785)
Target No. %
LAD 278 35.4
Diagonal 13 16.6
D1 114 14.5
D2 16 2
Ramus 13 1.7
OM1 36 4.6
OM2 95 12.1
OM3 18 2.3
PLOM 21 2.7
RCA 44 5.6
PDA 127 16.2
Table 4
OPCAB Fitzgibbon Scores by Target Vessel
Target A B A+B O Total
LAD 256 21 99.6% 1 278
Diagonal 126 2 98.5% 2 130
Ramus 13 0 100.0% 0 13
Marginal 159 7 97.6% 4 170
RCA 82 2 97.7% 2 86
PDA 85 0 100.0% 0 85
Other 22 0 97.7% 1 23
Total 723 32 98.7 10 785
infarctions. Seven patients (1.8%) required cardiac catheterization, showing a total of

five occluded grafts. Four patients (1.1%) had PTCA, two of which were for new lesions
in unbypassed coronary arteries. Six patients (1.6%) had late non-cardiac-related deaths,
1–18 mo after discharge. All other patients were alive and well without angina.
Results of Intermediate-Term (1–12 mo) Angiography After OPCAB

In one of the earliest accounts of angiographic evaluation of patients undergoing
OPCAB, Calafiore and his colleagues (30) described angiographic patency in 271 of 434
patients of (62%) undergoing left internal thoracic artery (LITA) to left anterior descend-
ing (LAD) artery grafting via MIDCAB. Postoperative angiography was performed
during the first year after surgery with an overall angiographic patency rate of 94%.
Furthermore, they note that anastomotic quality improved over the course of this series
of patients, such that in the last 190 patients, 134 (71%) underwent postoperative coro-
nary angiography and had a 99% patency rate.
Possati et al. (50) studied 76 of 77 patients (99%) with postoperative angiograms at a
mean of 1.7 ± 2.8 months. In 66 cases (87%) the LIMA graft, the LIMA–LAD anasto-
mosis, and the LAD adjacent to the anastomotic site were normal. In one case (1%), the
LIMA was occluded. In the remaining nine cases (12%) the LIMA was patent, but
anomalies of the LIMA course, the anastomosis, or the adjacent LAD were present. Of
these nine patients, three required reoperation and subsequent repair of the anastomosis
using conventional coronary surgery and one patient underwent successful PTCA of an
anastomotic stenosis. When they analyzed their results with respect to the type of instru-
mentation used at surgery (either specifically designed for beating-heart surgery or not),
the authors noted a 100% patency rate for those patients with dedicated chest retractors
and coronary stabilizers compared to 82% for those without dedicated OPCAB instru-
mentation.
In 1998, Mohr and his colleagues (51) reviewed their experience in 195 patients
undergoing single LIMA-to-LAD coronary artery bypass via a left anterolateral thorac-
otomy incision (MIDCAB). Postoperative angiography prior to discharge was completed
in 191 patients (91%) and revealed an overall patency rate of 97%. The results of a
midterm (6 mo) follow-up coronary angiogram in the first 58 patients revealed a patency
rate of 98%.
While most preceding series evaluated single-vessel coronary artery bypass (LIMA-
to-LAD) via the thoracotomy incision, Calafiore (52) described his initial experience
using a median sternotomy with two or more arterial conduits in 122 primary coronary
patients. Sixty-seven patients (55%) underwent 185 distal anastomoses and postopera-
tive angiography was performed at a mean of 33 ± 35 d after the operation. The overall
patency rate was 98.9% (183 of 185 grafts); the (FitzGibbon grade A patency rate was
98% (182 of 185 grafts). These data demonstrated that precise arterial grafting of all the
regions of the heart is possible without cardiopulmonary bypass.
In an innovative series of robotic-assisted IMA harvesting and direct CABG through
a 5-cm thoracotomy incision without the use of cardiopulmonary bypass, Vassiliades
et al. (53) assessed angiographic graft patency in 45 of 66 consecutive patients 6 mo
(range 2–15 mo) postoperatively. The overall patency rate for the study group was 98%,
demonstrating that beating-heart coronary artery bypass surgery using thoracoscopic
IMA harvesting can achieve effective intermediate-term revascularization.
In a study evaluating early and mid-term angiographic patency of 55 coronary anas-
tomosis constructed on 51 patients undergoing beating-heart surgery without the benefit
of mechanical stabilization, Gill et al. (54) noted that overall patency was 96% (53 of 55
sites) at 4–6 h following surgery. Follow-up angiography at a mean of 9.6 ± 4.5 mo (range
3–19 mo) in 32 of 51 patients (64%) revealed a 94% (30 of 32 patients) anastomotic
patency rate. Although these data recount anastomotic patency prior to the widespread
utilization of mechanical stabilization, they do describe acceptable longer-term
angiographic follow-up for beating-heart coronary grafting.
Results of Late (> 1 yr) Angiography After OPCAB

In a controversial paper, Ömero{gg}lu and his colleagues (55) reported the long-term
angiographic patency of 70 randomly chosen patients (10%) from a total population of
696 patients who had beating-heart surgery at their institution. The interval from opera-
tion to angiography varied from 24 to 61 mo (mean, 36 ± 11 mo). They noted a 96%
patency (65 of 68 patients) of the LIMA-to-LAD anastomoses, but only a 47% patency
(16 of 34 patients) in saphenous vein grafts (SVG). The angiography study population
consisted of only 10% of the overall beating-heart surgery population and 3% of the total
hospital CABG experience over 7 yr. Nevertheless, the low saphenous vein graft 3-yr
angiographic patency rate is of concern. In conventional coronary bypass, FitzGibbon et al.
(56) have previously described a patency rate in 1170 saphenous vein grafts of 81% at 1 yr,
75% at 5 yr, and 50% at greater than 15 yr.
More recently, Kim and associates (57) have analyzed the results of 122 consecutive,
nonrandomized OPCAB cases compared with those of 65 consecutive conventional
CABG cases. The average number of distal anastomoses between groups was compa-
rable (OPCAB: 3.1 ± 1.1 vs conventional CABG: 3.7 ± 0.9). In the OPCAB group,
coronary angiograms prior to discharge were performed in 92% (121 of 122) of patients
revealing a 96% (162 of 168 grafts) patency rate for arterial grafts and 86% (160 of 187
grafts) for SVG. One-year follow-up coronary angiograms in the OPCAB group were
performed in 74% of patients (90 of 122) and the patency rate was 98% (132 of 135 grafts)
for arterial grafts and 68% (106 of 156 grafts) for SVG. One-year follow-up in conven-
tional bypass patients was performed in 65% of patients (42 of 65). The patency rate was
94%, in the authors’ hands (43 of 46 grafts) for arterial grafts and 88% (98 of 111 grafts)
for SVG. This small series suggests that early patency of vein grafts after OPCAB is less
than that for arterial grafts, and also suggests a lower patency rate in 1-yr postoperative
angiograms for OPCAB vein grafts compared to vein grafts constructed utilizing conven-
tional CABG on CPB. In contrast, our series of postoperative coronary angiograms prior
to hospital discharge (Table 3) does not reveal a significant difference between arterial
and vein graft patency rates.
Currently, there is little information and no standardization of the angiographic
findings of OPCAB bypass grafts in the postoperative period (3). The most commonly
used classification to assess the quality of grafts following cardiac surgery is the afore-
mentioned FitzGibbon classification (Fitz O grafts are occluded; Fitz B are grafts with
stenosis resulting in a diameter less than 50% of the target vessel diameter; and Fitz A are
grafts without stenoses) (58). Although this classification alone does not adequately
describe all the findings and its lack of robustness hinders its use as a prognostic tool, it
is the most commonly used angiographic classification.
FUTURE TECHNOLOGIES TO EVALUATE

POSTOPERATIVE GRAFT PATENCY
Despite its “gold standard” status for patency analysis, limitations of angiography
have spawned great interest in new technologies to assess patency of postoperative
coronary artery bypass grafts. Refinements and advancements in electron-beam com-

puted tomogaphy (EBT) have effectively detected coronary calcifications and stenoses
and evaluated coronary artery bypass graft patency (59–63). More recently, a novel,
multislice, helical computed tomography (CT) scanner with four detector rows for 3-D
reconstruction has been introduced for visualization of postoperative coronary grafts
(64). Magnetic resonance angiography (65,66) and high-frequency transthoracic
echocardiography (67) have also been evaluated as alternatives to standard contrast
angiography and cardiac catheterization. None of these techniques is presently able to
provide the resolution and detail of angiography, but technological advances may allow
of these to supplant angiography as the technique of choice for graft assessment in
the future.
SUMMARY
Surgical techniques for multivessel OPCAB continue to evolve and improve. The
techniques utilized in performing multivessel OPCAB remain diverse (68–70). Careful
maintenance of myocardial protection during off-pump coronary revascularization is of
central importance and will lead to reduced morbidity and mortality. Intraoperative or
postoperative coronary angiography remains the “gold standard” in the evaluation of
graft patency. Notwithstanding, controversies remain regarding performance, cost, and
clinical decision making with this technique. The advantage of obtaining an intraopera-
tive angiogram is the ability to document graft patency before leaving the operating room.
However, the presence of an angiographic abnormality may lead to an unnecessary
revision, since some early angiographic lesions may be due to transient intramural he-
matomas or spasm. Postoperative angiography offers the benefit of being able to docu-
ment graft patency before discharge, but involves an additional procedure with the
associated additional morbidity and expense. Advances in standardization and ease with
which noninvasive techniques can be utilized by cardiac surgeons intraoperatively and
postoperatively may improve the assessment of coronary distal anastomoses following
beating-heart coronary surgery.
REFERENCES
1. Puskas JD, Wright CE, Ronson RS, Brown WM, Gott JP, Guyton RA. Clinical outcomes and
angiographic patency in 125 consecutive off-pump coronary bypass patients. Heart Surg Forum
1999;2:216–221.
utilization in 200 consecutive off-pump coronary bypass patients. Ann Thorac Surg 2001;71:1477–1483.
3. Mack MJ, Osborne JA, Shennib H. Arterial graft patency in coronary artery bypass grafting: what do
we really know? Ann Thorac Surg 1998;66:1055–1059.
4. Louagie YAG, Haxhe JP, Jamart J, Buche M, Schoevaerdts JC. Doppler flow measurement in coronary-
artery bypass grafts and early postoperative clinical outcome. Thorac Cardiovasc Surg 1994;42:175–181.
5. Walpoth BH, Bosshard A, Genyk I, et al. Transit-time flow measurement for detection of early graft
failure during myocardial revascularization. Ann Thorac Surg 1998;66:1097–1100.
6. D’Ancona G, Karamanoukian HL, Ricci M, Bergsland J, Salerno TA. Graft patency verification in
coronary artery bypass grafting: principles and clinical applications of transit time flow measurement.
Angiology 2000;51:725–731.
7. VanHimbergen DJ, Koenig SC, Jaber SF, Cerrito PB, Spence PA. A review of transit-time flow mea-
surement for assessing graft patency. Heart Surg Forum 1999;2:226–229.
8. Franklin DL, Ellis RS, Rushmir RF. Ultrasonic transit time flowmeter. IRE Trans Biomed Engl
1962;9:44–49.
9. D’Ancona G, Karamanoukian HL, Ricci M, et al. Graft revision after transit time flow measurements
in off-pump coronary artery bypass grafting. Eur J Cardiothorac Surg 2000;17:287–293.
10. Ricci M, Karamanoukian HL, Salerno TA, D’Ancona G, Bergsland J. Role of coronary graft flow
measurement during reoperations for early graft failure after off-pump coronary revascularization.
J Card Surg 1999;14:342–347.
11. Jaber SF, Koenig SC, BhaskerRao B, VanHimbergen DJ, Spence PA. Can visual assessment of flow
waveform morphology detect anastomotic error in off-pump coronary artery bypass grafting? Eur J
12. Koenig S, VanHimbergen DJ, Jaber SF, Ewert D, Cerrito P, Spence PA. Spectral analysis of graft flow
for anastomotic error detection in off-pump CABG. Eur J Cardiothoracic Surg 1999;16:S83–S87.
13. Cerrito P, Koenig SC, VanHimbergen DJ, Jaber SF, Ewert DL, Spence PA. Neural network pattern
recognition analysis of graft flow characteristics improves intra-operative anastomotic error detection
in minimally invasive CABG. Eur J Cardiothoracic Surg 1999;16:88–89.
14. Bergsland J, D’Ancona G, Karamanoukian HL, Ricci M, Schmid S, Salerno TA. Technical tips and
pitfalls in OPCAB surgery: the Buffalo experience. Heart Surg Forum 2000;3:189–193.
15. D’Ancona G, Karamanoukian HL, Salerno TA, Schmid S, Bergsland J. Flow measurement in coronary
surgery. Heart Surg Forum 1999;2:121–124.
16. D’Ancona G, Karamanoukian HL, Soltoski P, Salerno TA, Bergsland J. Changing referral pattern in
off-pump coronary artery bypass surgery: a strategy for improving surgical results. Heart Surg Forum
1999;2:246–249.
17. D’Ancona G, Karamanoukian H, Ricci M, Bergsland J, Salerno TA. Preoperative angiography and
intraoperative transit time flow measurement to detect coronary graft patency in reoperations: an inte-
grated approach. Angiology 2000;51:777–780.
18. Takami Y, Ina H. Relation of intraoperative flow measurement with postoperative quantitative
angiographic assessment of coronary artery bypass grafting. Ann Thorac Surg 2001;72:1270–1274.
19. Takami Y, Ina H. A simple method to determine anastomotic quality of coronary artery bypass grafting
in the operating room. Cardiovasc Surg 2001;9:499–503.
20. Hol PK, Fosse E, Mørk BE, et al. Graft control by transit time flow measurement and intraoperative
angiography in coronary artery bypass surgery. Heart Surg Forum 2001;4:254–258.
21. Louagie YAG, Haxhe JP, Jamart J, Buche M, Schoevaerdts JC. Intraoperative assessment of coronary
artery bypass grafts using a pulsed Doppler flowmeter. Ann Thorac Surg 1994;58:742–749.
22. Louagie YAG, Brockmann CE, Jamart J, et al. Pulsed Doppler intraoperative flow assessment and
midterm coronary graft patency. Ann Thorac Surg 1998;66:1282–1288.
23. Louagie YAG, Haxhe JP, Buche M, Schoevaerdts JC. Intraoperative electromagnetic flowmeter mea-
surements in coronary artery bypass grafts. Ann Thorac Surg 1994;57:357–364.
24. Bandyk DF, Galbraith TA, Haasler GB, Almassi GH. Blood flow velocity of internal mammary artery
and saphenous vein grafts to the coronary arteries. J Surg Res 1988;44:342–351.
25. Oda K, Hirose K, Nishimori H, Sato K, Yamashiro T, Ogoshi S. Assessment of internal thoracic artery
graft with intraoperative color Doppler ultrasonography. Ann Thorac Surg 1998;66:79–81.
26. Takayama T, Suma H, Wanibuchi Y, Tohda E, Matsunaka T, Yamashita S. Physiological and pharma-
cological responses of arterial graft flow after coronary-artery bypass-grafting measured with an im-
plantable ultrasonic Doppler miniprobe. Circulation 1992;86:217–223.
27. Elbeery JR, Brown PM, Chitwood WR. Intraoperative MIDCABG arteriography via the left radial
artery: a comparison with Doppler ultrasound for assessment of graft patency. Ann Thorac Surg
1998;66:51–55.
28. Lin JC, Fisher DL, Szwerc MF, Magovern JA. Evaluation of graft patency during minimally invasive
coronary artery bypass grafting with Doppler flow analysis. Ann Thorac Surg 2000;70:1350–1354.
29. Calafiore AM, Gallina S, Iaco A, et al. Minimally invasive mammary artery Doppler flow velocity
evaluation in minimally invasive coronary operations. Ann Thorac Surg 1998;66:1236–1241.
30. Calafiore AM, Di Giammarco G, Teodori G, et al. Midterm results after minimally invasive coronary
surgery (LAST operation). J Thorac Cardiovasc Surg 1998;115:763–771.
31. Elbeery JR, Chitwood WR. Intraoperative catheterization of the left internal mammary artery via the left
radial artery. Ann Thorac Surg 1997;64:1840–1842.
32. Barstad RM, Fosse E, Vatne K, et al. Intraoperative angiography in minimally invasive direct coronary
33. Izzat MB, Khaw KS, Atassi W, Yim APC, Wan Wan S, El-Zufari MH. Routine intraoperative angiogra-
phy improves the early patency of coronary grafts performed on the beating heart. Chest 1999;115:987–990.
34. Izzat MB, Yim APC. MIDCABG: lessons learned from routing “on-table” angiography. Ann Thorac
Surg 1997;64:1872–1874.
35. Lazzara RR, Kidwell FE, Griffith R. A new technique for intraoperative greaft angiography utilizing the
radial artery stump. Heart Surg Forum 2000;3:123–126.
36. Lazzara RR, McLellan BA, Kidwel FE, et al. Intraoperative angiography during minimally invasive
direct coronary artery bypass operations. Ann Thorac Surg 1997;64:1725–1727.
37. Suma H, Isomura T, Horii T, Sato T. Intraoperative coronary artery imaging with infrared camera in off-
pump CABG. Ann Thorac Surg 2000;70:1741–1742.
38. Takayama T, Wanibuchi Y, Suma H, et al. Intraoperative coronary angiography using fluorescein. Ann
Thorac Surg 1991;51:140–143.
39. Shabbo FP, Rees GM. Thermography in assessing coronary artery saphenous graft patency and blood
flow. Cardiovasc Res 1982;16:158–162.
40. Mohr FW, Matloff J, Grundfest W, et al. Thermal coronary angiography: a method for assessing graft
aptency and coronary anatomy in coronary bypass surgery. Ann Thorac Surg 1989;47:441–449.
41. Lawson W, BenEliyahu D, Meinken L, et al. Infrared thermography in the detection and management
of coronary artery disease. Am J Cardiol 1993;72:894–896.
42. Falk V, Walther T, Philippi A, et al. Thermal coronary angiography for intraoperative patency control
of arterial and saphenous vein coronary artery bypass grafts: results in 370 patients. J Card Surg
1995;10:147–160.
43. Schaff HV, Cable DG, Rihal CS, Daly RC, Orszulak TA. Minimial thoracotomy for coronary artery
bypass: value of immediate postprocedure graft angiography. Circulation 1996;94(suppl 1):51.
44. Subramanian VA, McCabe JC, Geller CM. Minimally invasive direct coronary artery bypass grafting:
two-year clinical experience. Ann Thorac Surg 1997;64:1648–1655.
45. Poirier NC, Carrier M, Lespérance J, et al. Quantitative angiographic assessment of coronary anasto-
moses performed without cardiopulmonary bypass. J Thorac Cardiovasc Surg 1999;117:292–297.
46. Mack MJ, Magovern JA, Acuff TA, et al. Results of graft patency by immediate angiography in mini-
mally invasive coronary artery surgery. Ann Thorac Surg 1999;68:383–390.
47. Gill IS, FitzGibbon GM, Higginson LAJ, Valji A, Keon WJ. Minimally invasive coronary artery bypass:
a series with early qualitative angiographic follow-up. Ann Thorac Surg 1997;64:710–714.
48. Puskas JD, Wright CE, Ronson RS, Brown WM, Gott JP, Guyton RA. Off-pump multivessel coronary
bypass via sternotomy is safe and effective. Ann Thorac Surg 1998;66:1068–1072.
49. Thourani VH, Puskas JD, Marshall JJ, et al. Postoperative graft patency and clinical outcomes in 378
consecutive off-pump coronary bypass patients. Ann Thorac Surg, to be published.
50. Possati G, Gaudino M, Alessandrini F, Zimarino M, Glieca F, Luciani N. Systemic clinical and
angiographic follow-up of patients undergoing minimally invasive coronary artery bypass. J Thorac
51. Diegeler A, Falk V, Matin M, et al. Minimally invasive coronary artery bypass grafting without cardiop-
ulmonary bypass: early experience and follow-up. Ann Thorac Surg 1998;66:1022–1025.
52. Calafiore AM, Teodori G, Di Giammarco G, et al. Multiple arterial conduits without cardiopulmonary
bypass: early angiographic results. Ann Thorac Surg 1999;67:450-6.
53. Vassiliades TA Jr, Rogers EW, Nielson JL, Lonquist JL. Minimally invasive direct coronary artery
bypass grafting: intermediate-term results. Ann Thorac Surg 2000;70:1063–1065.
54. Gill IS, Higginson LA, Maharajh GS, Keon WJ. Early and follow-up angiography in minimally invasive
coronary bypass without mechanical stabilization. Ann Thorac Surg 2000;69:56–60.
55. Ömero{gg}lu SN, Kirali K, Güler M, et al. Midterm angiographic assessment of coronary artery bypass
56. FitzGibbon GM, Leach AJ, Keon WJ, Burton JR, Kafka HP. Coronary bypass fate. J Thorac Cardiovasc
Surg 1986;91:773–778.
57. Kim K-B, Lim C, Lee C, et al. Off-pump coronary artery bypass may decrease the patency of saphenous
ven grafts. Ann Thorac Surg 2001;72:S1033–S1037.
58. FitzGibbon GM, Kafka HP, Leach AJ. Coronary bypass graft fate and patient outcome: angiographic
follow-up of 5,065 grafts related to survival and reoperation in 1,388 patients during 25 years. J Am Coll
Cardiol 1996;28:616–626.
59. Gulbins H, Reichenspurner H, Becker C, et al. Preoperative 3D-reconstructions of ultrafast-CT images
for the planning of minimally invasive direct coronary artery bypass operation (MIDCAB). Heart Surg
Forum 1998;1:111–115.
60. Hernigou A, Challande P, Boudeville JC, Sene V, Grataloup C, Planfosse MC. Reproducibility of
coronary calcification detection with electron-beam computed tomography. Eur Radiol 1996;6:
210–216.
61. Moshage W, Achenbach S, Seese B, Bachman K. Non-invasive coronary diagnosis with EBT (electon
beam tomography). Population screening for coronary heart disease (reliability imaging of coronary
stenoses). Fortschr Med 1997;115:45–49.
62. Ruping D, Shaoxiong Z, Bin L, et al. Three-dimensional reconstruction of electron beam computed
tomograhy angiography for evaluating coronary artery bypass grafts. Chinese Med J 1998;111:588–592.
63. Lu B, Dai R-P, Jing B-L, et al. Evaluation of coronary artery bypass graft patency using three-dimen-
sional reconstruction and flow study on electron beam tomography. J Comput Assisted Tomogr
2000;24:663–670.
64. Sawamura Y, Takase K, Saito H, Kikuchi S, Ito T. Noninvasive postoperative angiography for internal
mammary artery grafts. Circulation 2001;104:373–374.
65. Boehm DH, Wintersperger BJ, Reichenspurner H, et al. Contrast-enhanced magnetic resonance angiog-
raphy for control of minimally invasive coronary artery bypass conduits (MIDCAB/OPCAB). Heart
Surg Forum 1999;2:222–225.
66. Vetter HO, Driever R, Mertens H, Kempkes U, Cramer BM. Contrast-enhanced magnetic resonance
angiography of mammary artery grafts after minimally invasive coronary bypass surgery. Ann Thorac
Surg 2001;71:1229–1232.
67. De Simone L, Caso P, Severino S, et al. Noninvasive assessment of left and right internal mammary
artery graft patency with high-frequency transthoracic echocardiography. J Am Soc Echocardiogr
1999;12:841–849.
68. Flameng W. Role of myocardial protection for coronary artery bypass grafting on the beating heart. Ann
Thorac Surg 1997;63:S18–S22.
69. Puskas JP, Vinten-Johansen J, Muraki S, Guyton RA. Myocardial protection for off-pump coronary
artery bypass surgery. Sem Thorac Cardiovasc Surg 2001;13(1):82–88.
70. Chitwood WR Jr, Wixon CL, Elbeery JR, et al. Minimally invasive cardiac operation: adapting
cardioprotective strategies. Ann Thorac Surg 1999;68:1974–1977.
Chapter 14 / Minimally Invasive Conduit Harvesting 193
14 Minimally Invasive Conduit Harvesting
Kevin D. Accola, MD, Mike Butkus, PA-C,

and Brenda Dickey, RN
CONTENTS
INTRODUCTION
DIRECT LIMITED-ACCESS SAPHENOUS VEIN HARVEST
ENDOSCOPIC SAPHENOUS VEIN HARVEST
TECHNIQUES OF MINIMALLY INVASIVE RADIAL ARTERY HARVEST
SUMMARY
REFERENCES
INTRODUCTION
More than 500,000 coronary artery bypass grafting (CABG) procedures were per-
formed last year in the United States. The greater saphenous vein (GSV) was used as a
bypass conduit in approx 95% of these cases (1). The traditional method of harvesting the
GSV for either CABG or for peripheral arterial bypass involves making a long incision
along the length of the patient’s lower extremity. Much literature has been written about
the pain associated with the vein harvest site incision as compared to the lesser discomfort
of the primary sternotomy incision. The associated pain and morbidity that results from
harvesting of the vein via the conventional approach may begin shortly after surgery and
may last for years. Numerous reports have also documented potential major and minor
complications that may result from GSV harvesting (2–6). Major complications include
infection, dehiscence, or necrosis requiring debridement, skin grafting, arterial
revascularization, or even amputation (6) (Figs. 1 and 2). Minor complications, reported
to occur in 0.5–31% of patients, include cellulitis, hematoma, seroma, and edema.
Minimally invasive saphenous vein harvesting has been shown to be associated with
decreased postoperative pain, earlier ambulation, and decreased hospital stay as com-
pared to conventional vein harvesting (1,7–14). It also provides patients with an
improved cosmetic result secondary to the smaller incision. With the availability of the
internet and increased access to information, patients have become better informed
193
Fig. 1. Saphenous wound site dehiscence.
Fig. 2. Severe infection of saphenous wound site with erythema, lymphedema, and cellulitis.
regarding minimally invasive harvesting techniques and often present to their surgeon
requesting a minimally invasive approach.
This chapter discusses direct limited-access and video-assisted endoscopic vein har-
vesting, two minimally invasive techniques used for harvesting the GSV. Each of these
techniques will be outlined separately, with comments on their advantages and limitations.
Fig. 3. The Genzyme BioSurgical SaphLite hand-held lighted retractor.
Fig. 4. The Genzyme BioSurgical SaphLite console.
DIRECT LIMITED-ACCESS SAPHENOUS VEIN HARVEST

The technique of skin bridging for saphenous vein harvesting represents an attempt to
minimize overall incision length, in which a series of multiple incisions is interspersed
with “bridges” of skin. Direct limited-access harvesting of saphenous veins is a modifi-
cation of this technique. A retractor with a light source provides a “lighted” tunnel and
an opportunity to decrease the number of skin incisions necessary (Genzyme BioSurgical
SaphLite) (Figs. 3 and 4). With this approach, the primary objectives are to avoid inci-
Fig. 5. Creation of mid-calf incision overlying saphenous vein.
sions over the medial malleolus, groin, and knee. Avoiding the ankle and groin reduces
infection and healing complications. Avoiding the knee improves healing, joint mobility,
and patient comfort. The advantages of this technique are the ease and simplicity of
instrumentation and mobility of the system. Because there is no tower, the retractor and
light source are easily movable. One can modify the direct limited technique depending
on the anatomy of the vein. When necessary, incisions can be extended, while still
avoiding the ankle, groin, and knee.
Techniques of Direct Limited-Access Saphenous Vein Harvest

A 10- to 12-cm incision is made above the ankle to locate the saphenous vein (Fig. 5).
A tunnel is then constructed proximal and distal to this incision (Fig. 6). A retractor is
placed and aids in clipping of side branches with conventional instrumentation (Figs. 7
and 8). A second incision is then made distal to the knee as the process is repeated both
distally and proximally across the knee joint. A third incision is made above the knee, and
the tunnel is extended proximally. Once the desired length of vein is achieved, a stab
incision is made near the ankle and the groin. The saphenous vein is then clipped and tied
(Fig. 9). Hemostasis is achieved during this dissection using electrocautery.
The wound is closed with running subcuticular suture and a topical dressing is placed
over it (Fig. 10). Circumferential wrapping of the harvest site is recommended during the
bypass procedure to ensure hemostasis (Fig. 11). Hemostasis should be rechecked once
the patient is separated from cardiopulmonary bypass and heparin has been reversed.
Some centers recommend a leg wrap for 24 h, although the reported incidence of post-
operative hematomas and seromas in the perioperative period has not been shown to be
increased if a compressive wrap is not used.
Fig. 6. Digital creation of saphenous tunnel.
Fig. 7. Lighted retractor is placed to localize branches.
ENDOSCOPIC SAPHENOUS VEIN HARVEST

The technology for endoscopic saphenous vein harvesting continues to evolve (15).
There is a significant learning curve in becoming proficient in this technique. Once
Fig. 8. With retractor in place, conventional instrumentation is used to control saphenous tributaries.
Fig. 9. The vein harvest has been completed through three small incisions.
proficiency is established, however, this technique allows the entire length of saphenous
vein to be harvested from a single incision site above the knee.
Techniques of Endoscopic Saphenous Vein Harvest

The instrumentation that we employ is the Guidant Vasoview Uniport Plus System
(Fig. 12), which involves videoscopic viewing and CO2 insufflation. The incision is
Fig. 10. Subcuticular suture and topical dressing on wound site.
Fig. 11. Circumferential wrapping to reduce chance of postoperative hematomas.
typically made above the knee, through which the dissection takes place. A seal is made
with the skin around the insufflator port. The insufflator is placed in the incision, followed
by an injection, enabling visualization of the vein. Continuous insufflation of CO2 main-
tains a “window” to circumferentially dissect the vein and has not been found to nega-
tively affect the quality of the vein (16). The insertion site of the dissection cannula must
remain airtight to maintain CO2 insufflation pressure. Dissection is performed under
direct endoscopic guidance using endoscopic instruments (Figs. 13–16). An endoscopic
Fig. 12. The Guidant Vasoview Uniport Plus system. (Reproduced with permission of Guidant
Corporation.)
Fig. 13. The videoscope is introduced into the wound.
view seen through the videoscope is demonstrated in Fig. 17. The vein is typically
dissected proximally and then distally, depending on the length of conduit that is
required. Cautery scissors can be placed into the insufflator port to maintain adequate
Fig. 14. Insufflator is enabled to create tunnel and allow visualization.
Fig. 15. Dissection progresses distally.
hemostasis as well as to aid in the dissection process. Small clip applicators can be used
to control side branches during dissection.
Proximal and distal division of the vein is performed with an endoscopic scissors and
vein-clipping apparatus. Division of the vein can also be performed through a small
incision over the most proximal and distal sites with ligation of the vein. Once the vein
is removed from the leg, conventional operative techniques are utilized.
Results of Minimally Invasive Saphenous Vein Harvest

Many studies in the literature have demonstrated a variety of benefits associated with
a minimally invasive harvesting approach as compared to open (7–14). A prospective,
randomized trial by Allen revealed a decreased wound complication rate of 4% for
endoscopic harvest as compared to 19% for open, as well as decreased hospital stay for
Fig. 16. Endoscopic clip applicator is inserted to control side branches.
Fig. 17. Endoscopic view of saphenous vein.
endoscopic patients (p < 0.05) (7). Puskas also prospectively randomized open and
endoscopic patients and demonstrated improvement in postoperative discomfort with
an endoscopic approach (8).
In our series, the incidence of leg wound infections was less than 2%, with no deep
wound infections or any cases requiring operative intervention or debridement. Other
studies have also shown a decreased wound infection rate in endoscopic as compared to
open harvesting (Table 1).
Several studies have compared endoscopic and bridged approaches (1,10). Patel et al.
reported on 200 CABG patients who were prospectively studied after undergoing either
bridged or endoscopic GSV harvest. There was a significantly lower incidence of wound
complications in the endoscopic group—2% vs 12% (p < 0.05). Endoscopic patients also
ambulated earlier and had a significantly decreased hospital stay (p < 0.05) (1). Some
Table 1
Comparison of Infection Rates between Endoscopically
and Coinventionally Harvested SVG
Wound Infection Rate (%)
Author Year Endoscopic Conventional p-value
Bitondo et al. 2002 6.8 28.3 <0.001
Allen et al. 1998 4 19 <0.02
Table 2
Comparison of Infection Rates between Endoscopically
and Bridge Approaches
Wound Infection Rate (%)
Author Year Endoscopic Bridge p-value
Horvath et al. 1998 32 3 <0.005
Patel et al. 2001 2 12 <0.05
studies have confirmed these conclusions, while others have shown a bridged approach
to be associated with an improved outcome (Table 2).
Although minimally invasive harvesting techniques have associated benefits, until
recently, many questioned the quality and integrity of the vein with such an approach.
Several studies have addressed this concern (17,18). Griffith et al. studied 178 patients
who underwent either open or endoscopic vein harvest (17). Veins were analyzed for
histological abnormalities using hematoxylin–eosin, trichrome, and elastin staining. No
differences were found between veins harvested via an open or endoscopic approach with
regard to intimal endothelial continuity (p = 0.45), elastic lamina continuity (p = 0.27),
medial connective tissue uniformity (p = 0.91), medial smooth muscle continuity (p = 0.16),
or adventitial connective tissue uniformity (p = 0.92).
This observation was also noted by Meyer et al., who compared veins harvested via
an endoscopic approach with those harvested via an open approach (18). No significant
abnormalities were found in the intimal, medial, or adventitial layers of the veins in either
of the groups. Additional immunohistochemical studies for factors VIII:vWF and CD 34
also showed no difference between the groups.
Studies have also investigated whether there is a difference in smooth muscle reactiv-
ity and vasomotor function between veins harvested via an open vs an endoscopic
approach (19–21). Black et al. prospectively randomized 40 patients into open or endo-
scopic vein harvest and studied smooth muscle function of the harvested veins (19).
Smooth muscle baseline contractile function was assessed by responses to potassium
chloride and phenylephrine, and relaxation was assessed by receptor-dependent agonists
(acetylcholine and bradykinin) as well as receptor-independent agonists (calcium iono-
phore and sodium nitroprusside) (p = 0.3). No difference was found between the two
groups, implying that there is preservation of endothelial and smooth muscle cell func-
tion. Cable et al. also evaluated the functional state of the endothelium in veins harvested
via an endoscopic approach and found similar results (20).
In a randomized study performed in our institution comparing direct-lighted retractor

vs conventional harvest techniques, subjective benefits were noted using minimally
invasive techniques. Patients ambulated earlier and with considerably less pain. The
study was terminated prematurely due to an overwhelming patient satisfaction both in
comfort and cosmesis with minimally invasive harvesting.
Disadvantages of Minimally Invasive Saphenous Vein Harvest

Another possible disadvantage associated with a minimally invasive approach may be
inadequate exposure and lighting. The lighted retractor or video-assisted equipment does
not always yield optimal visualization of the vein. This occurs most often in obese
patients, when the anatomy of the vein is abnormal, and when there is extensive branching
of the vein. When there is inadequate visualization or lighting, it is often necessary to
extend the initial incision to obtain adequate visualization and to ensure that the vein
quality is not compromised. It may be necessary to make a mid-thigh incision to better
delineate anomalous anatomy. A direct-lighted retractor technique may have a benefit in
these patients, as it does not rely on insufflated venous tunnels. If the harvest requires
opening the leg partially, the remainder of the vein can be removed utilizing the lighted
retractor in a more “limited access” fashion.
Several other factors can limit the use of minimally invasive vein harvest procedures.
Varicosities cause a significant amount of bleeding and obscure visualization through the
tunnel. Superficial veins, close to the dermal edge, will not accommodate an adequate-
sized tunnel to work within. This not only obscures visualization but also limits the
space for optimal vein dissection. Additionally, delicate or friable saphenous veins must
be approached with caution and may, in some cases, be a contraindication to a minimally
invasive approach. When these situations are encountered, the decision must be made
whether the limiting factors will compromise the quality of the vein.
In emergency cases, when it is necessary to harvest the vein rapidly, we typically
pursue a conventional leg incision. We utilize the direct-lighted retractor across the knee
joint and then extend the incision up into the groin to limit postoperative infection and
knee pain.
Possible Postoperative Complications of Minimally

Invasive Saphenous Vein Harvest
Although wound infections are uncommon with a minimally invasive approach, when
they occur, they may be associated with an infected tunnel hematoma. This may require
opening of the tunnel with irrigation, although there are reports of effective closed-space
irrigation (22).
Postoperative hematomas need to be evacuated from the tunnel site. This can typically
be accomplished with a small stab incision over the hematoma site and application of
pressure. A pressure dressing is then applied and maintained as a wrap for 24–36 h.
Seromas occur in 1–2% of patients and typically occur 2–3 wk postoperatively in
patients with previous hematomas. They are generally treated with needle aspiration and
a pressure dressing, although there are occasions when a small incision with packing of
the seroma site may be required.
Direct injury to the saphenous vein can also occur, although not more frequently as
compared to a conventional approach. In a prospective randomized study at our institu-
tion, the necessity of conduit repair was less in minimally invasive harvested conduits as
compared to those harvested by conventional techniques.
Table 3
Scoring System for Saphenous Vein Graft Intraluminal
Lesions
Score Grading
0 No endoluminal surface lesions
1 A. Isolated simple fibrous strands < 5
B. Hemorrhagic staining < 3
C. Grade 0 with spasm
2 A. Multiple simple fibrous strands 5–10
B. Complex fibrous strands < 5
C. Hemorrhagic staining >3
D. Grade 1 with spasm
3 A. Diffuse strands > 10
B. Complex fibrous strands > 5
C. Thrombus
D. Grade 2 with spasm
Angioscopy for Monitoring of Intraluminal Lesions

We recently began performing intraoperative angioscopy on harvested veins to evalu-
ate for intraluminal pathology. Using a 2.2-mm Olympus angioscope, a variety of lesions
have been identified that include simple fibrous strands, complex fibrous strands, hem-
orrhagic staining, and thrombus. We devised a scoring system ranging from 0 to 3, with
3 being associated with the greatest number of intraluminal lesions (Table 3). Intraopera-
tive angioscopy is recorded and viewed by three surgeons after the operation, and veins
are scored based on the number and severity of intraluminal lesions. Each vein is scored
before and after saline irrigation. A ratio of the number of lesions postirrigation
divided by the number of lesions preirrigation is calculated. A ratio of the number of
lesions before and after irrigation divided by the length of vein is also calculated. We
recently began comparing scores of open and endoscopic veins. Preliminary data indicate
that the majority of lesions tend to flush out with saline irrigation, with an improvement
in the postirrigation score. There seems to be no significant difference in the number of
lesions postirrigation or the postirrigation scores of open vs endoscopic.
TECHNIQUES OF MINIMALLY INVASIVE RADIAL ARTERY HARVEST

Minimally invasive techniques have been applied to harvesting of the radial artery (RA)
as well. Several studies evaluating endoscopic RA harvest have demonstrated favorable
results with regard to arterial patency. However, this approach is not widely used because
there is a relatively high incidence of complications. Possible complications include direct
injury to the artery, dissection, and hematoma with subsequent compartment syndrome.
Other complications include cellulitis and dorsal thenar sensory numbness. In a series of
300 patients who underwent endoscopic RA harvest, the incidence of cellulitis and dorsal
thenar sensory numbness was 1.6% and 8.7%, respectively (23,24).
SUMMARY
Saphenous vein harvesting can be performed safely using minimally invasive tech-
niques. These minimally invasive approaches are associated with improvement in cos-
metic results, decreased wound complications, shorter hospital stay, and patient satisfac-
tion. We believe that these techniques will continue to evolve as technology continues to
improve.
REFERENCES
1. Patel AN, Hebeler RF, Hamman BL, et al. Prospective analysis of endoscopic vein harvesting. Am J Surg
2001;182:716–719.
2. DeLaria GA, Hunter JA, Goldin MD, Serry C, Javid H, Najafi H. Leg wound complications associated
with coronary revascularization. J Thorac Cardiovasc Surg 1981;81:403–407.
3. Baddour, LM, Bisno AL. Recurrent cellulitis after saphenous venectomy for coronary bypass surgery.
Ann Intern Med 1982;97:493–496.
4. Reifsnyder T, Bandyk D, Seabrook G, et al. Wound complications of the in-situ saphenous vein bypass
technique. J Vasc Surg 1992;15:843–850.
5. Wipke-Tevis DD, Stotts NA, Skov P, Carrieri-Kohlman V. Frequency, manifestations, and correlates
of impaired healing of saphenous vein harvest incisions. Heart Lung 1996;25:108–116.
6. Lee KS, Reinstein L. Lower limb amputation of the donor site extremity after coronary artery bypass
graft surgery. Arch Phys Med Rehabil 1986;67:564–565.
7. Allen KB, Griffith GL, Heimansohn DA, et al. Endoscopic versus traditional saphenous vein harvesting:
a prospective randomized trial. Ann Thorac Surg 1998;66:26–32.
8. Puskas J, Wright C, Miller P, et al. A randomized trial of endoscopic versus open saphenous vein harvest
in coronary bypass surgery. Ann Thorac Surg 1999;68:1509–1512.
9. Crouch JD, O’Hair DP, Keuler JP, Barragry TP, Wermer PN, Kleinman LN. Open versus endoscopic
saphenous vein harvesting: wound complications and vein quality. Ann Thorac Surg 1999;68:1513–1516.
10. Horvath KD, Gray D, Benton L, Hill J, Swanstrom LL. Operative outcomes of minimally invasive
saphenous vein harvest. Am J Surg 1998;175:391–395.
11. Davis Z, Jacobs HK, Zhang M, Thomas C, Castellanos Y. Endoscopic vein harvest for coronary artery
bypass grafting: technique and outcomes. J Thorac Cardiovasc Surg 1998:116:228–235.
12. Morris RJ, Butler MT, Samuels LE. Minimally invasive saphenous vein harvesting. Ann Thorac Surg
1998;66:1026–1028.
13. Hayward TZ, Hey LA, Newman LL, et al. Endoscopic versus open saphenous vein harvest: the effect
on postoperative outcomes. Ann Thorac Surg 1999;68:2107–2111.
14. Kan C, Luo C. Endoscopic saphenous vein harvest decreases leg wound complication in coronary artery
bypass grafting patients. J Card Surg 1999;14:157–162.
15. Allen KB, Shaar CJ. Endoscopic saphenous vein harvesting. Ann Thor Surg 1997;64:265–266.
16. Vitali RM, Reddy RC, Molinaro PJ, Sabado MF, Jacobowitz IJ. Hemodynamic effects of carbon dioxide
insufflation during endoscopic vein harvesting. Ann Thorac Surg 2000;70:1098–1099.
17. Griffith GL, Allen KB, Waller BF, et al. Endoscopic and traditional saphenous vein harvest: a histologic
comparison. Ann Thorac Surg 2000;69:520–523.
18. Meyer DM, Rogers TE, Jessen ME, Estrera AS, Chin AK. Histologic evidence of the safety of endo-
scopic saphenous vein graft preparation. Ann Thorac Surg 200;70:487–491.
19. Black EA, Guzik TJ, West NRJ, et al. Minimally invasive saphenous vein harvesting: effects on endot-
helial and smooth muscle function . Ann Thorac Surg 2001;71:1503–1507.
20. Cable DG, Dearani JA, Pfeifer EA, Daly RC, Schaff HV. Minimally invasive saphenous vein harvesting:
endothelial integrity and early clinical results. Ann Thorac Surg 1998;66:139–143.
21. Fabricius AM, Oser A, Diegeler A, Rauch T, Mohr FW. Endothelial function of human vena saphena
magna prepared with different minimally invasive harvesting techniques. Eur J Cardiothorac Surg
2000;18:400–403.
22. Allen KB, Fitzgerald EB, Heimansohn DA, Shaar CJ. Management of closed space infections associated
with endoscopic vein harvest. Ann Thorac Surg 2000;69:960–961.
23. Connolly MW, Torrillo MD, Stauder MJ, et al. Endoscopic radial artery harvesting: results of the first
300 patients. Ann Thorac Surg 2002;74(2):502–505.
24. Terada Y, Uchida A, Fukuda I, Hochberg J, Mitsui T, Sato F. Endoscopic harvesting of the radial artery
as a coronary artery bypass graft. Ann Thorac Surg 1998;66:2123–2124.
Chapter 15 / Hybrid Revascularization 207
15 Hybrid Revascularization
Uwe Klima, MD and Axel Haverich, MD

CONTENTS
INTRODUCTION
PATIENTS AND METHODS
OPERATIVE PROCEDURE: MIDCAB
PERCUTANEOUS INTERVENTION
CASE SCENARIOS
RESULTS
DISCUSSION
SUMMARY
REFERENCES
INTRODUCTION
Revascularization with the left internal mammary artery (LIMA) and additional radial
artery and/or vein grafts via a median sternotomy is the generally accepted surgical
approach to multivessel coronary artery disease. Cardiopulmonary bypass (CPB) with
cardioplegic arrest still represents the standard technique for coronary artery surgery in
most centers worldwide. However, as this book underscores, there is a strong trend
toward beating-heart surgery in an effort to avoid potential complications caused by CPB
(1,2). Minimally invasive direct coronary artery bypass (MIDCAB) grafting without
cardiopulmonary bypass through an anterolateral minithoracotomy has become a thera-
peutic option, especially in multimorbid, elderly, and redo patients with single-vessel
disease (3). However, owing to limited access through the minithoracotomy, this
approach cannot be applied to multivessel revascularization without additional incisions
or use of CPB. To expand the benefits of the MIDCAB approach to patients with
multivessel disease, a “hybrid” procedure combining MIDCAB surgical revascularization
of the left anterior descending artery (LAD) with catheter-based interventional proce-
dures (PTCA; percutanous transluminal coronary angioplasty ± stent) for additional
coronary lesions is increasingly being considered as an attractive therapeutic option (4).

207
Table 1
Demographics and Extent of Coronary Artery
Disease Among 33 Patients Undergoing
Hybrid Revascularization
Demographic
Age (yr) 62.1 ± 8.5
Sex (M/F) 27/6
Unstable angina 6 (18.2%)
Previous MIa 15 (45.5%)
One–vessel diseaseb 6 (18.2%)
Two–vessel disease 18 (54.5%)
Three–vessel disease 9 (27.3%)
Previous PTCAc/stent 3 (9.1%)
Proximal LADd disease 33 (100%)
aMyocardial infarction.
bDiseased vessel is the LAD and PTCA is to the
diagonal branch of the LAD.
cPercutaneous coronary transluminal angioplasty.
dLeft anterior descending.
Our preliminary results using a “hybrid” approach to myocardial revascularization sug-

gest that this concept is a safe and effective method of complete revascularization for
selected patients with multivessel involvement. Very old patients and redo patients with
significant comorbidity are likely to benefit most from this approach. However, even
younger patients with aggressive disease, for whom repeat coronary revascularization
procedures seem likely in the future, should also benefit. Such patients will avoid the risk
of CPB, and the discomfort and possible complications of median sternotomy.
In this chapter, we present our preliminary results in 33 consecutive cases undergoing
a “hybrid” procedure consisting of an initial MIDCAB surgical revascularization of the
LAD with subsequent angioplasty of additional coronary lesions.

Between June 1996 and September 2001 a total of 530 MIDCAB procedures were
performed at Hannover Medical School. Thirty-three (6.2%) of these patients had an
elective hybrid procedure. Following MIDCAB revascularization of the LAD, 33
patients had PTCA of an additional 48 coronary lesions after a median of 7 d. Patient
characteristics, cardiac performance, and comorbidities of those 33 patients are summa-
rized in Tables 1 and 2.
OPERATIVE PROCEDURE: MIDCAB

After induction of general anesthesia with standard intravenous agents, the left chest
is exposed by placing the patient on the right side in the 30° decubitus position. The left
lung is collapsed by use of a bronchial blocker. An indicator dilution catheter is placed
for hemodynamic measurements. A short (8–10 cm), left fourth- or fifth-intercostal space
incision is made using the chest radiograph, for estimating the better level for exposing
the anterolateral surface of the heart. The LIMA pedicle is then dissected from the chest
wall. It is important to dissect a sufficient length to obviate tension on the anastomosis.
Table 2
Comorbidities Among 33 Patients
Undergoing Hybrid Revascularization
Comorbidity No. (%) of patients
Peripheral vascular disease 2 (6.1%)
Deep venous thrombosis 1 (3%)
Left ventricular aneurysm 1 (3%)
Chronic renal failure 2 (6.1%)
Cerebrovascular accident 2 (6.1%)
Diabetes 9 (27.3%)
COPDa 3 (9.2%)
Tuberculosis 2 (6.1%)
Hyperthyroidism 1 (3%)
Melanoma 1 (3%)
Morbid obesity 2 (6.1%)
aChronic obstructive pulmonary disease.
At least 8 cm and, with a large heart or a very distal anastomosis, as much as 15 cm may
be required. Proximal dissection may reach the subclavian origin of the LIMA. Dilute
papaverine is sprayed onto the pedicle frequently to avert conduit spasm.
After dissection of the LIMA is completed, the patient is systemically heparinized
(100 IU/kg body weight) and the distal end of the LIMA pedicle is divided between clips.
The pericardium is opened with a “T”-incision and a pericardial sling is formed with
strategically placed stay sutures to elevate and position the heart for optimum exposure
of the LAD. The appropriate site for an arteriotomy is selected and that region of the heart
is fixed and rendered relatively motionless by use of a semielliptical, atraumatic, stabi-
lizing retractor. To provide a bloodless field, we have used a commercially available mist
machine and 4-0 polypropylene snares looped around the LIMA. For the first 200
MIDCAB patients we used snares both proximal and distal to the arteriotomy.
It was later shown that snare trauma may presage arterial stenosis in the native vessel
at the snare site; thereafter we modified our technique to omit the distal snare. In the
present study group of 33 patients, the first 15 patients had both proximal and distal
snares. Hemodynamics were evaluated for 2 min after tightening the snare. If the heart
tolerated this, the anastomosis was undertaken using 8-0 polypropylene sutures. During
the snaring period, two patients in the study had hemodynamic instability unresponsive
to inotropes and a temporary intraluminal shunt was inserted to bridge the arteriotomy
and afford distal myocardial perfusion during the performance of the anastomosis. At the
completion of the anastomosis the snare is released and removed. Protamine is then given
at a dose calculated to yield a 50% reversal of the administered heparin. Fibrin glue is
applied to the anastomosis for further hemostatic assurance, and is used to fix the pedicle
to the epicardium in a gentle curve without tension, torsion, or angulations. The pericar-
dium is loosely closed. A single chest tube is placed in the costophrenic sulcus. The
interspace incision is closed in layers in standard fashion.
PERCUTANEOUS INTERVENTION
Each patient was recatheterized between postoperative d 3 and 9. Patients were
instructed to take 500 mg of aspirin at 24 h and again 1 h before recatheterization. The
Fig. 1. Completed LIMA to LAD anastomosis.
transfemoral approach was used. All LIMA grafts were demonstrated to show good to
excellent flow (Fig. 1). Assured of good flow in the LIMA–LAD distribution, attention
was then directed to the additional known coronary lesions, which were confirmed by
reinjection. Each patient was then given 15,000 IU of heparin through the intra-arterial
catheter. If a stent was used, 250 mg of ticlopidine was also administered. A guide
catheter (7 or 8 F), suitable for angioplasty, was positioned appropriately for each addi-
tional lesion of concern. Balloon dilatations were then performed over a 0.0014 guide
wire. If any lesion had a residual stenosis of over 35%, the lesion was stented. A final
coronary arteriogram documented the success of the PTCA/stenting procedure (Fig. 2).
CASE SCENARIOS
Several clinical scenarios led to our decision to perform hybrid procedures in our
patients. In the following paragraphs, three “typical” patients are described in whom we
utilized the two-pronged surgical and interventional approach.
Case 1
A 58-yr-old male patient had suffered recurrent angina after previous surgical myo-
cardial revascularization. At cardiac evaluation the vein graft to the right coronary artery
was patent and well functioning. However, vein grafts to both the circumflex and left
anterior descending artery were closed, and both native coronaries showed significant
proximal stenosis. Because the LIMA was still available, the patient was reoperated upon
by MIDCAB technique. On the fifth postoperative day, a repeat angiogram showed a
Fig. 2. Successful PTCA of right coronary artery: (A) Stenosis before and (B) after intervention.
patent LIMA-to-LAD anastomosis. The proximal circumflex stenosis was dilated and
stented successfully. Three years after the hybrid procedure the patient remains free of
angina.
Case 2
A 49-year-old female patient underwent cardiac transplantation 2 yr before for dilated
cardiomyopathy. Although she remained free of cardiac failure, on her second annual
follow-up catheterization severe triple-vessel transplant vasculopathy was diagnosed.
The right coronary artery supplied no significant area of myocardium and was not treated.
A dominant obtuse marginal branch was significantly stenosed, as was the mid-LAD.
After performing a LIMA-to-LAD MIDCAB procedure, the stenosis in the obtuse mar-
ginal branch of the circumflex artery was successfully stented 5 d after the MIDCAB.
Case 3
A 79-year-old female patient with unstable angina had significant comorbidity. In
addition to angina, her cardiac evaluation showed that her left ventricular ejection frac-
tion was less than 25%. She had advanced pulmonary obstructive disease with a forced
expiratory volume at 1 s (FEV1) of 600 mL. Her CT scan showed a porcelain aorta. She
had insulin-dependent diabetes. These findings excluded her from consideration for
conventional revascularization. The preoperative cardiac angiogram revealed severe
triple-vessel disease with a severe obstruction in the proximal portion of the right coro-
nary artery, a subtotal proximal LAD occlusion, and multiple stenoses in the circumflex.
Medical management was unsuccessful, as the patient could not be weaned from continu-
ous intravenous nitroglycerin and heparin. Hence, we performed a MIDCAB procedure,
followed by stenting of the proximal right coronary artery on the seventh day postopera-
tively. Both procedures were successful and the patient was discharged to rehabilitation
on the ninth postoperative day.
RESULTS
There were no intra- or perioperative deaths. There were no conversions to a sterno-
tomy. There were no major complications (i.e., myocardial infarction, stroke, other neu-
rological deficit, bleeding sufficient to require reoperation, or respiratory failure requiring
prolonged mechanical ventilation) during hospitalization. These patients did not require
inotropic support. Blood loss was minimal, with a mean of 450 ± 120 mL per patient.
There was no major (> 1 unit of blood) transfusion requirement. Average postoperative
ventilation time was 8.5 ± 4.1 h.
After the MIDCAB procedures, coronary reangiography revealed patent and function-
ing LIMA grafts in all patients. There was one mal-insertion of the LIMA into a second
diagonal branch instead of the main trunk of the LAD, but there was excellent backflow
into the LAD and perfusion of its distal distribution. One LIMA pedicle was judged to
be under tension but was functioning. One new stenosis was noted in the LAD distal
to the LIMA–LAD anastomosis and two new stenoses were noted in the LAD proximal
to the LIMA–LAD anastomosis. All these stenoses were less than 50%. Two LIMA
pedicles were found to be fixed to the chest wall but were functioning. Angiography data
are summarized in Table 3.
No further deterioration of organ function occurred in patients with preexisting
comorbidity of lungs, kidneys, or central nervous system. Subsequent PTCA, and stenting
were required (n = 14). A total of 45 additional lesions were treated successfully. There
were no major complications related to these procedures. Two patients had groin hemato-
mas that did not require surgical management. Mean follow-up time was 54 ± 9.2 mo.
Thirty-one patients (94 %) are alive and free of angina. One patient died of a myocardial
infarction 16 mo after the hybrid procedure. The cause of death in the one additional
patient is unknown; one patient suffered a myocardial infarction (posterior wall) during
follow-up 12 mo after the operation. There have been seven nonelective percutaneous
interventions during the long-term follow-up. One of these was a repeat PTCA (12 mo
postoperatively) of a lesion previously dilated as part of an initial hybrid procedure. Six
were dilatations of new lesions in lateral or posterior wall coronary branches.
DISCUSSION
MIDCAB has become an accepted therapeutic option worldwide for the management
of single-vessel disease, especially in patients with high proximal stenosis of the left
anterior descending coronary artery (5). Major advantages have been realized because of
the avoidance of extracorporeal circulation and sternotomy incision. This has been par-
ticularly true for elderly patients and patients with extensive comorbidities. Because the
results have been so favorable for MIDCAB alone, additional applications have been
sought. As a result, the combined surgical (MIDCAB) and catheter-type intervention
emerged (6,7), called the “hybrid” approach. This strategy is gaining increasing accep-
tance as a method to perform a protected left main coronary stenosis angioplasty (8).
In our series, the patients had numerous comorbid factors. All these comorbidities are
by themselves, or in combination, risk factors for increased perioperative morbidity and
mortality. However, in our patients undergoing a hybrid procedure, no intra- or
Table 3
Angiographic Results After MIDCAB LIMA-to-LAD Anastomosis
LIMA–LAD Anastomosis n Native LAD n
Tension 1 New proximal stenosis (< 50%) 2
Dissection 0 New distal stenosis (< 50%) 1
Competitive flow 0
Occlusion 0
Fixation to chest wall 2
Narrowing 2
Stenosis 0
Malinsertion to D1 1
perioperative death occurred and none of our patients suffered a major complication
during hospitalization. Neuropsychiatric complications, including perioperative stroke,
are frequently seen complications following coronary artery bypass grafting. Trehan and
coworkers (9) evaluated the use of an individualized surgical approach including the
hybrid procedure for reducing neurological injury in patients undergoing CABG who
were at high risk of stroke because of aortic atherosclerosis or carotid artery disease. They
concluded that selective surgical techniques, including hybrid procedures, can prevent
adverse neurological sequelae while achieving complete myocardial revascularization,
even in a group of high-risk patients. Our patients had no postoperative strokes, other
neurological complications, or psychoses.
At the present time, MIDCAB is used predominantly for grafting of vessels on the
anterior surface of the heart. Vessels on the lateral or posterior left ventricular walls are
operated on through a mid-sternotomy using a beating-heart technique or conventional
CPB-supported surgery. Can we select certain patients with multiple lesions who might
benefit more from a hybrid procedure than from conventional surgery or PTCA-stenting
alone? A patient with a specific contraindication to CPB, who in the past might have had
only a MIDCAB to the LAD, would be left with incomplete revascularization. Because
it has been well demonstrated that incomplete revascularization carries an increased risk
of myocardial infarction, repeated angioplasties or stenting procedures, and redo surgery,
we believe that additional efforts should be made to do as complete a revascularization
as possible. Patients with an LAD not amenable to PTCA-stenting who in addition have
an unacceptable risk for CPB are clearly candidates for the hybrid procedure.
Two factors determine the long-term efficacy of the hybrid procedure: the long-term
patency of the MIDCAB LIMA–LAD graft and the duration of benefit from the PTCA-
stenting procedure. Previous experience has shown LIMA–LAD surgery to be superior
in long-term patency, freedom from significant cardiac events, survival, and relief of
symptoms, to catheter intervention techniques. Also PTCA-stenting has a better record
regarding restenosis when applied to the right coronary artery or to lesions in the circum-
flex system than the 20% first-year failure rate, even with stenting, when similar inter-
ventions are applied to the LAD. The hybrid procedure combines the best features of both
the surgical and the interventional techniques. It permits the benefits of MIDCAB surgery
to be extended to a wider spectrum of patients. Certain patients with two-vessel disease
in which the LAD stenosis is considered to be a type C lesion are known to be unfavorable
candidates for PTCA-stenting and are also recognized to have a progressive form of
coronary disease in which repeated interventions and redo surgery are all too often the
rule. An initial MIDCAB LIMA–LAD would avoid redo surgery through a
midsternotomy whenever the likely recurrence of symptoms develops. A particularly
elegant application of the hybrid principle is in the patient with a left main coronary artery
stenosis who is also an unacceptable risk for conventional surgery. A preliminary MIDCAB
LIMA–LAD affords protection from myocardial ischemia during the subsequent cath-
eter intervention.
There is current debate regarding the proper sequence of the two segments of the
hybrid procedure. If the percutaneous intervention is done first, conventional surgery
could possibly treat the actual coronary artery disease and manage the complication at the
same time. Supporters of that sequence also mention that the extensive anticoagulation
required after catheterization complications and after stenting might lead to bleeding
difficulties in the previous MIDCAB operative site. However, complications of
interventional techniques such as intimal tears, plaque disruption, acute thrombosis, and
distal embolization are rare. If the catheter procedure requires the use of high-dose,
multiagent anticoagulant and antiplatelet therapy, there may be a significant and detri-
mental delay in performing the MIDCAB. We believe that the LAD lesion is likely to be
the most significant ischemic lesion and should be treated first by MIDCAB LIMA–
LAD. The catheter intervention for additional lesions should follow at a convenient early
date thereafter. We also believe that the ultimate goal should be the performance of the
two phases of the hybrid procedure on the same day, in the same operative theater, with
the MIDCAB preceding the catheter intervention. For a variety of reasons this is imprac-
tical in most hospitals today.
Mid- and long-term patency and an uneventful clinical course are important factors in
evaluating the efficacy of this relatively new procedure. After a mean follow-up time of
54 ± 9.2 mo, all patients had a clinical reevaluation or, if that was not possible, answered
a standardized questionnaire. Analysis of the data was encouraging: 31 patients (94%)
were alive and free of angina. Considering that this subset of patients has severe
comorbidity, the results are very promising. Similar results have been reported by other
groups (7,10,11) with 11.4–24 mo follow-up. Table 4 delineates the outcomes of five
most recent series investigating hybrid approaches to revascularization (12–16).
SUMMARY
The combination of an initial MIDCAB LIMA–LAD followed, in 3–9 d, by PTCA-
stenting for additional coronary obstructive lesions, is called the hybrid approach to
myocardial revasularization. We have found the hybrid approach to be safe, with low
morbidity, and effective in establishing complete revascularization in patients with
double- and triple-vessel coronary disease. It has been particularly helpful in the very old
patient, the patient with extensive associated disease, and in those patients who are at high
risk for CPB or mid-sternotomy. Younger patients whose coronary anatomy or type of
atherosclerotic coronary artery disease makes them unsuitable for standard interventional
or standard surgical procedures may also be considered for the hybrid approach.
REFERENCES
1. Puskas JD, Wright CE, Ronson RS, Brown WM 3rd, Gott JP, Guyton RA. Clinical outcomes and
angiographic patency in 125 consecutive off-pump coronary bypass patients. Heart Surg Forum
1999;2(3):216–221.
Chapter 15 / Hybrid Revascularization
Table 4
Most Recent Series Investigating the Results of Hybrid Revascularization
Author (Ref.) No. of Order of Mean Early Reintervention
Patients Intervention Age (yr) Mortality Graft Patency (%) PTCA or CABG
215
Matsumoto (12) 11 PTCA first 70 ± 9 0 100.0 27% of PTCA lesions

Presbitero (13) 42 na na 4.70% 92.3 4.7% of LIMA–LAD grafts
Riess (14) 57 MIDCAB first 65 ± 8 0 98.0 23.5% of PTCA lesions
Cisowski (15) 50 MIDCAB first 55 ± 20 0 100.0 10% of PTCA lesions
2% of LIMA–LAD grafts
215
2. Isomura T, Suma H, Horii T, Sato T, Kobashi T, Kanemitsu H. Minimally invasive coronary artery
revascularization: off-pump bypass grafting and the hybrid procedure. Ann Thorac Surg
2000;70(6):2017–2022.
3. Cremer JT, Wittwer T, Boning A, et al. Minimally invasive coronary artery revascularization on the
beating heart. Ann Thorac Surg 2000;69(6):1787–1791.
4. Wittwer T, Cremer J, Boonstra P, et al. Myocardial “hybrid” revascularisation with minimally invasive
direct coronary artery bypass grafting combined with coronary angioplasty: preliminary results of a
multicentre study. Heart 2000;83(1):58–63.
5. Diegeler A. Left internal mammary artery grafting to left anterior descending coronary artery by mini-
mally invasive direct coronary artery bypass approach. Curr Cardiol Rep 1999;1(4):323–330.
6. Riess FC, Schofer J, Kremer P, et al. Beating heart operations including hybrid revascularization: initial
experiences. Ann Thorac Surg 1998;66(3):1076–1081.
7. De Canniere D, Jansens JL, Goldschmidt-Clermont P, Barvais L, Decrol P, Stoupel E. Combination of
minimally invasive coronary bypass and percutaneous transluminal coronary angioplasty in the treat-
ment of double-vessel coronary disease: two-year follow-up of a new hybrid procedure compared with
“on-pump” double bypass grafting Am Heart J 2001;142(4):563–570.
8. Mack MJ, Brown DL, Sankaran A. Minimallly invasive coronary bypass for protected left main coro-
nary stenosis angioplasty. Ann Thorac Surg 1998;64(5):1313–1315.
9. Trehan N, Mishra M, Kasliwal RR, Mishra A. Surgical strategies in patients at high risk for stroke
undergoing coronary artery bypass grafting. Ann Thorac Surg 2000;70(3):1037–1045.
10. Wittwer T, Haverich A, Cremer J, Boonstra P, Franke U, Wahlers T. Follow-up experience with coro-
nary hybrid-revascularisation. Thorac Cardiovasc Surg 2000;48(6):356–359.
11. Wittwer T, Haverich A, Cremer JT, Boonstra PW. The hybrid procedure for myocardial revascularisation:
intermediate results. Ann Thorac Surg 1999;69(3):975.
12. Matsumoto Y, Endo M, Kasashima F, et al. Hybrid revascularization feasibility in minimally invasive
direct coronary artery bypass grafting combined with percutaneous transluminal coronary angioplasty
in patients with acute coronary syndrome and multivessel disease. Jpn J Cardiovasc Surg 2001;49:700–705.
13. Presbitero P, Nicolini F, Maiello L, et al. “Hybrid” percutaneous and surgical coronary revascularization:
selection criteria from a single-center experience. Ital Heart J 2001;2:369–371.
14. Riess FC, Bader R, Kremer P, et al. Coronary hybrid revascularization from January 1997 to January
2001: a clinical follow up. Ann Thorac Surg 2002;73:1849–1855.
15. Cisowski M, Morawski W, Drzewiecki J, et al. Integrated minimally invasive direct coronary artery
bypass grafting and angioplasty for coronary artery revascularization. Eur J Cardiothorac Surg
2002;22:261–265.
Chapter 16 / Neurocognitive Issues in Off-Pump CABG 217
16 Neurocognitive Issues
in Off-Pump CABG
Ronald M. Lazar, PHD

and Daniel F. Heitjan, PHD
CONTENTS
INTRODUCTION
NEUROLOGICAL ASPECTS
NEUROCOGNITIVE ASPECTS
NEUROBEHAVIORAL OUTCOMES AFTER OFF-PUMP CORONARY
ARTERY SURGERY
CLINICAL MANAGEMENT
REFERENCES
INTRODUCTION
Since its introduction in 1968, coronary artery bypass grafting (CABG) has resulted
in superior survival and a better quality of life for specific subgroups of patients with
coronary artery disease when compared to medical therapy (1). For most patients, CABG
and the combined use of cardiopulmonary bypass (CPB) have resulted in clinically
undetectable deficits, but in a significant minority the results have been more serious (2).
Of all the adverse consequences associated with bypass grafting, neurological outcomes
represent an important proportion (3). Given that more than 650,000 people in the United
States and 800,000 worldwide undergo CABG (4), improvements in surgical techniques
and patient management stand to have an impact on significant numbers of patients with
respect to both medical costs and quality of life.
The purpose of this chapter is to demonstrate that one impetus for the trend toward the
use of off-pump CABG is the neurogenic morbidity that can be associated with on-pump
surgery. The following material thus represents a discussion of the broad neurological

217
and neurocognitive issues. Suggestions regarding pre- and postoperative evaluations will
be made.
NEUROLOGICAL ASPECTS
Cerebrovascular morbidity is a well-documented complication of (on-pump) CABG,
with the rate of perioperative cerebral infarction ranging from 2% to 6% in studies
conducted over the past 20 years (5–10). In a large, multicenter prospective study, Roach
and colleagues studied 2108 patients in which the overall stroke rate following CABG
was 3.0%, with an additional 3.1% suffering prolonged unconsciousness, seizures, or
encephalopathy (4). The mortality rate of patients with stroke, stupor, or coma was 21%
compared to 2% for those without. Age and duration of cardiopulmonary bypass were the
factors most significantly correlated with adverse neurological events. Tuman’s group
found an overall stroke rate of 2.8% among 2000 prospective patients undergoing CABG;
patients under the age of 65 had a rate of 0.9% while those over 75 had an incidence of
9.0% (11). Like other studies, duration of CPB and prior stroke were also significant
risk factors.
In one of the largest studies to date, Stamou and his group at the Washington Hospital
Center found that among 16,528 consecutive patients undergoing on-pump CABG over
a 10-yr period, 333 (2%) had postoperative stroke (12). Patients who suffered these
events were older, but several important operative characteristics were predictive. Indeed,
the crossclamp and cardiopulmonary bypass times were significantly longer in those who
developed postoperative stroke than in those who did not.
Two pathophysiological mechanisms have been proposed to underlie major neuro-
logical events arising from CABG associated with CPB: multiple emboli and cerebral
hypoperfusion arising from intraoperative hypotension. With respect to macroembolization,
possible mechanisms include air, valve, and aortic atheroma (13). Wijdicks and Jack
evaluated the clinical and brain-imaging (CT) features of 25 patients with ischemic stroke
following CABG, with events occurring from 1 to 22 d after surgery (14). No patient
had watershed infarcts. Rather, the imaging and clinical course of these patients sug-
gested that post-CABG stroke was consistent with embolic phenomena, related to retained
cardiac thrombi, new-onset atrial fibrillation, or intimal damage of the aorta during
clamping. Indeed, there are new data supporting the importance of premorbid aortic
disease in CABG that makes off-pump surgery an attractive alternative. Trehan et al.
prospectively screened 3660 patients undergoing CABG with transesophageal
echocardiography (TEE) and found 104 patients with plaque with a mobile element
(grade III) (15). Using intraoperative TEE, they found no embolic events in the subgroup
of 88 patients who underwent off-pump CABG. The other possible mechanism, hypoten-
sion, is thought to occur in some cases, but is considered much less common (16). It is
not yet clear whether embolism in the brain circulation in the context of a transient,
hypoperfused state and/or during a postsurgical inflammatory period accelerates the
ischemic cascade.
The coexistence of coronary and carotid atherosclerosis continues to be documented,
yet controversy still surrounds the prophylactic application of carotid endarterectomy
prior to CABG (17,18). Nevertheless, Yoon et al. found among 201 prospectively studied
(on-pump) CABG patients that 52.4% had extracranial carotid disease (≥70% stenosis)
and/or intracranial carotid disease (≥ 50% stenosis) (19). More important, on multivariate
analysis, intracranial carotid disease had an independent association with the develop-
ment of CNS complications. Such studies have yet to be performed with off-pump
patients.
NEUROCOGNITIVE ASPECTS
Neuropsychological function refers to the complex set of cognitive tasks subserved by
the brain involving such spheres as language, memory, attention, visual perception, and
executive skills (20). The terms “neuropsychological” and “neurocognitive” are gener-
ally used interchangeably. It has been suggested that the assessment of neurocognition
is probably a more sensitive index of brain integrity with cerebrovascular disease than
sensorimotor function or coordination (21–23), and is dissociable from depression (24).
It is therefore not surprising that, in contrast to major neurological complications, the
reported incidence of cognitive changes following CABG, especially with CPB, is much
higher. Except in cases of frank ischemic stroke, cortical deficits such as aphasia, apraxia,
agnosia, and amnesia are rarely found. Rather, alteration of cognitive function following
CABG resembles that found in chronic, small-vessel cerebrovascular disease in the
periventricular and subcortical white matter (25).
Both short- and long-term deficits have been identified, but there is considerable
variation across studies (26). Early studies demosntrated that the incidence of decline in
the first 2 wk postoperatively ranged from 30% to 79%, and when extended to 6-mo
follow-up, decline ranged from 24% to 57% (27). Mahanna et al. summarized the litera-
ture from 1980 to 1994, encompassing 21 studies with 1864 CABG patients, and found
that the variability occurred from factors such as surgical technique, type of anesthesia,
cognitive instruments used, severity of illness, and patient demographics (28). In addi-
tion, the operational definition of cognitive impairment also affected the reported inci-
dence of postsurgical dysfunction. These studies, however, led to two important consensus
conferences outlining such needs as having a baseline, presurgical testing; an evaluation
of a comprehensive range of function; an assessment at least 3 mo postoperatively; the
importance of identifying change scores in individual patients; and consideration of
“practice effects” in the analysis of findings (29,30).
McKann et al. studied eight cognitive domains in 127 CABG patients at 1 mo and 1 yr
after surgery (27) . All but 12% of patients demonstrated decline in cognitive function.
While functions such as language declined and then improved, approx 10% had persistent
declines in verbal memory, visual memory, attention, and visuoconstruction, and 24% had
late, further decline in visuoconstruction. More recently, another prospective study of 127
patients from the same investigator group showed that surgery-related factors correlated
with cognitive impairment at 3 mo, but at 1 yr, only medical history variables were related
to such dysfunction, all of which suggested that there may be different determinants of
cognitive change at different postoperative time points (31). It was pointed out, however,
that only 127 of the original 172 patients were successfully evaluated at all study points.
The longest-term study to date assessing the neurocognitive consequences of on-pump
CABG comes from Newman and his group at Duke (32). They followed 261 patients who
underwent CABG by administering a neurocognitive battery preoperatively, before dis-
charge, 6 wk, 6 mo, and 5 yr after surgery. The authors performed a statistical analysis
of the baseline (presurgical) scores to eliminate concern about the redundancy of tests and
derived four “noncorrelated” domains of function: (1) verbal memory and language
comprehension; (2) abstraction and visual–spatial orientation; (3) attention, psychomo-

tor processing speed, and concentration; and (4) visual memory. Cognitive decline was
defined as a loss of one standard deviation in performance in any one of the four domains.
It was found among the 172 patients who completed the study that 52% had cognitive
decline at discharge, 36% at 6 wk, 24% at 6 mo, and 42% at 5 yr. Predictors of decline
at 5 yr included older age, higher baseline neurocognitive function, lower educational
level, and cognitive decline at discharge. Interestingly, duration of cardiopulmonary
bypass and duration of aortic crossclamping were not significant predictors in this study.
Analogous to major neurological events following conventional CABG, microembolism
is said to account for most of the cognitive changes seen after surgery (33). Moody and
colleagues found fatty material containing aluminum in the brain microvasculature from
autopsy specimens of patients who had died shortly after CABG with CPB and who had
cognitive impairment (34). Using intraoperative transcranial Doppler (TCD), Jacobs et al.
found that high-intensity transient signals (HITS) in 18 CABG patients occurred prima-
rily during cardiopulmonary bypass and to a lesser extent during aortic manipulation and
were correlated with cognitive changes 8–12 d postoperatively (35). It was also seen that
such neuropsychological deficits were not correlated with the total number of HITS,
suggesting that the location of brain injury was more important for the development of
such deficits, a notion that is also supported by brain SPECT (36).
One method for detecting brain ischemic disease is magnetic resonance imaging (MRI)
(37). Sylivris and colleagues studied 41 patients undergoing CABG both with intraopera-
tive TCD and pre- and postsurgery MRI (38). TCD confirmed that most microemboli
occurred during cardiopulmonary bypass, but MRI failed to show a relationship between
new-onset small-vessel disease and subsequent neuropsychological deficits. They pro-
posed that neuropsychological testing may be more sensitive than radiological detection
of vascular events as a result of the limit in the resolution of conventional MRI to detect
very small strokes.
A new method for the early detection of brain ischemia is diffusion-weighted MRI
(DWI), which has been shown capable of detecting neurovascular changes as quickly as
105 min after the onset of a clinical event, with the ability to discriminate old vs new
lesions (39). Moreover, it has been found that DWI is not only more sensitive than
conventional MRI in detecting acute lesions, but that lesion volume on early DWI cor-
relates significantly with clinical outcome ratings on the NIH Stroke Scale, the Barthel
Index, and the Rankin Scale (40). Recently, DWI was reported in 14 post-CABG
(on-pump) cases of acute stroke (41). Whereas CT detected 5/12 ischemic events, infarcts
were seen in 10/14 patients with DWI, with radiographic findings suggestive of multiple
embolic infarcts involving multiple vascular territories.
To illustrate the sensitivity of this new radiographic technique, we present in Fig. 1 two
cases of patients who underwent CABG, one off-pump (above) and one on-pump (below),
as part of an ongoing clinical trial. The T2-weighted image obtained shortly after surgery in
both cases shows the presence of old strokes. In the off-pump case, the DWI shows no new
event, but there is a new lesion in the left thalamus of the patient who underwent on-pump
surgery. To our knowledge, this new imaging technology has the potential to identify highly
subtle changes in brain integrity that could elucidate further the underlying mechanisms.
One of the important as yet unsettled issues in the detection of neurocognitive change
after CABG is the operational definition of abnormality: What standard should be used
as an index of an alteration of function? There seem to be two major approaches in this
regard. Both have in common the notion that there must be a presurgical baseline, and that
Fig. 1. Magnetic resonance imaging of the brain on two patients within 72 h after CABG surgery,
with T2-weighted images on the left and diffusion-weighted images (DWI) on the right. Patient
1 had off-pump CABG and Patient 2 had on-pump CABG. The black arrow points to a new infarct
in the left thalamus.
with repeated testing over time there is at least some degree of practice effect. One line
of investigation has used a one-standard-deviation (SD) decline on either a particular test
or a group of tests clustered into a domain of function. As pointed out in the review by
Symes et al., problems with this form of measurement include the likelihood of “floor
effects” in patients whose scores are so low at baseline that further decline is difficult to
detect and that declines in high baseline scores require a greater degree change (42).
Within this framework, there has never been a consensus as to the number of tests or
domains that need to change. The other definition has been a 20% change on 20% of the
tests employed, which some consider more sensitive in the identification of cognitive loss
(28). The problem with this method, however, is that it does not take into account the
greater degree of normal variation as patients get older, especially on tests that are timed
and/or require manual dexterity (43).
A major implication of the way in which decline is defined bears on the statistical
analysis needed in a clinical trial. As indicated above, some investigators have used a
decline of 1 SD as the benchmark for defining change. In all cases that we have seen, the
standard deviation was estimated from a pool of subjects (study participants or normal
subjects in the population), each evaluated on a single occasion. The problem arises when
this estimated standard deviation is used to evaluate longitudinal changes within subjects.
When one observes repeated measurements on a population, the total variability in the
measurements has two components: one representing within-subject variability and
the other representing between-subject variability. That is, some subjects will gener-
ally give high values, with some variability around their long-term means, whereas
others will give low values, again with some variability around their long-term means.
The between-subject SD measures variability between the long-term means of different
subjects, and the within-subject SD measures variability within subjects around their
long-term means. When the within-subject variance is low relative to the between-sub-
ject variance, we say that the measure has good reliability (44), because a single test
provides a good idea of that subject’s long-term performance. When the within-subject
variance is high relative to the between-subject variance, a single measurement is a poor
indicator of performance, and we say that the reliability is low. Thus, one reason for
discrepant findings across neurocognitive studies may lie in the failure to establish in
advance how such patients perform on the same test over time.
When evaluating change over time within subjects, as we intend in studying treatment
effects on neurocognition, it is preferable to use only within-subject variability, because
it gauges the expected level of variation within a subject and allows us to distinguish
between normal temporal variability and real treatment effects. For example, a change
from baseline of 3 within-subject SDs or more would occur only about 5% of the time in
stable individuals. We cannot calibrate changes based on the SD that we commonly use,
which includes both within- and between-subject components, without also knowing the
reliability: For example, even a 1 SD change may be large if the reliability is sufficiently
high. Proper calibration of changes requires creating norms from populations that are
measured more than once, so that an estimate of the within-subject SD is available.
A second issue that arises in establishing standards of variability for measuring change
within subjects is the source of the estimated SD. Some studies (e.g., Newman, et al. [32])
have estimated the SD from the study’s own baseline observations. Aside from the issues
discussed above, this can create a practical problem for data analysis. In clinical trials,
it is common to perform one or more interim analyses with the objective of stopping early
should the data lean strongly one way or the other. If the primary outcome is, say, the
percentage of patients who experience a decline of 1 SD or more, and the SD is based on
baseline data from the study itself, then this SD will change as more subjects are accrued.
So unless accrual is complete by the time of the first interim analysis, the SD is bound to
change from one interim analysis to the next. This creates an ambiguity in the data
analysis that is not desirable in a clinical trial. From this perspective, it is clearly prefer-
able to use a prespecified set of norms, rather than the accruing study data itself, to
determine the measure of variability for a neurocognitive endpoint in a clinical trial.
While previous neurocognitive studies have suggested possible mechanisms for alter-
ations of functioning, the clinical importance of detecting these cognitive changes, some-
times referred to as “ecological validity,” is unresolved (45,46). A major contribution to
our understanding of the short- and long-term effects of CABG with cardiopulmonary
bypass, or to compare it with other surgical techniques, would be the relationship between
specific neuropsychological deficits and the outcomes of validated quality-of-life instru-
ments. The paucity of such correlational data to date may also account for the lack of
agreement in the definition of cognitive abnormality found in the literature.
Another group of relatively new markers of neuronal injury associated with on-pump
CABG is the specific brain-originated proteins that include protein S-100B and neuron-
specific enolase (NSE), easily assessed from blood samples (47,48). Kilminster et al.
found a statistical association between increased S-100 release during and 5 h after the
onset of cardiopulmonary bypass and poorer neuropsychological performance 6–8 wk
following surgery (49). More recently, Herrmann et al., studying patients undergoing
either on-pump CABG or valve-replacement surgery, found similar early correlations
between increased NSE and S-100B and neuropsychological impairment. Unfortunately,
only two patients were seen at the planned 6-mo follow-up. In contrast, Wimmer-
Greinecker et al. found that increases in NSE and S-100B protein during and immediately
after surgery did not predict impairment in neuropsychological scores 5 d or 2 mo
postsurgery. Thus, the brain protein data to date are suggestive but not yet convincing as
predictors of cognitive problems, but the effects may only be related to short-term,
transient impairment.
NEUROBEHAVIORAL OUTCOMES AFTER OFF-PUMP

CORONARY ARTERY SURGERY
The presence of central nervous system complications, whether gross or subtle, has led
to a few recent preliminary efforts to compare coronary artery surgery with and without
CPB. Hernandez and his colleagues compared outcomes between 445 patients undergo-
ing conventional CABG with 322 off-pump patients. There were no differences in
perioperative stroke or in-hospital mortality, but cognitive status was not reported (50).
In a Brazilian study, 81 patients (48 with CPB and 33 without) were evaluated neurologi-
cally and cognitively before and 5–7 d after surgery (51). Of the 81 patients, 5 died. There
were otherwise no statistical differences in outcomes between the two groups. It should
be noted, however, that only patients with diagonal, left anterior descending, and right
coronary artery lesions were eligible for the non-CPB group, thus precluding a random-
ized design. In addition, 49 patients were either illiterate or had less than a fifth-grade
education, so it was unlikely that significant deficits in cognition function could be detected.
Andrew and colleagues compared pre- and postoperative neuropsychological dys-
function among three groups of patients: minimally invasive direct coronary bypass
grafting (MIDCAB) patients (n = 7), single-graft CABG/CPB recipients (n = 9) patients,
and multiple-graft CABG/CPB (n = 27) (52). Postsurgery follow-up occurred before
discharge. It was found that the multiple-graft CPB group had a significantly higher
incidence of cognitive decline than the MIDCAB or single-graft groups. There was,
however, no longer-term follow-up.
Murkin’s group in Western Ontario published results of an ongoing study comparing
33 on-pump CABG patients with 35 off-pump CABG patients (53). The preliminary data
showed no clinically apparent strokes in either group. At the fifth postoperative day, there
was a significantly lower incidence of cognitive dysfunction in the off-pump cases. At
3 mo postoperatively, 50% of the on-pump patients had cognitive deficits, compared to
5% in the off-pump group.
Finally, Diegeler et al. randomized 40 patients to either on- or off-pump CABG (54).
During surgery, there were a significantly greater number of high-intensity transient
224 Table 1 Part II / Minimally Invasive Bypass Grafting
Domains of Neurocognitive Function and Representative Tests
That Can Comprise a Neuropsychological Test Battery
Domain Test (Ref.)
Global mental status Mini-Mental State Examination (55)
Verbal memory Rey Auditory Verbal Learning Test (57)
California Verbal Learning Test (58)
Visual memory Wechsler Memory Scale—Third Edition (56)
Rey Complex Figure Test (Recall) (59)
Language Boston Diagnostic Aphasia Examination (60)
Boston Naming Test (61)
Attention Digit Span (62)
Visuoconstruction Rey Complex Figure (Copy) (54)
Block Design (57)
Executive function Trail Making Test (63)
Stroop Test (64)
Motor speed Purdue Pegboard (65)
Grooved Pegboard (66)
signals (HITS) in the conventional group compared to the off-pump patients. Using a
cognitive function scale validated for Alzheimer’s disease, they found that on d 7 after
surgery, 90% of the on-pump patients had abnormal cognitive scores, with no abnormal-
ity in the off-pump group. There was, however, no long-term follow-up of these patients,
and it was acknowledged that perioperative mental and physical stress can have an early effect.
CLINICAL MANAGEMENT
The data are highly suggestive that conventional coronary artery bypass surgery poses
neurological and neurocognitive risk for patients, mainly due to embolism. Although the
incidence of frank stroke has diminished with advances in surgical equipment and technol-
ogy, patients can be left with injury that is less physically apparent but equally disabling.
The patients at risk for neurological/neuropsychological insult appear, in particular,
to be older individuals with a prior history of stroke or TIA, carotid artery disease, and
atherosclerosis of the ascending aorta. Of course, the presence of other risk factors for
stroke and cardiac disease, in general, must be taken into consideration, such as hyper-
tension, diabetes, hypercholesterolemia, smoking, atrial fibrillation, and so forth. It would
therefore seem prudent that all patients with elevated cerebrovascular risk undergo neu-
rological examination, carotid duplex Doppler, and transcranial Doppler (TCD) prior to
surgery. Conventional MRI can detect the presence of old strokes. The onset of strokelike
symptoms should be evaluated by diffusion-weighted MRI (DWI), which is capable of
differentiating between acute and chronic lesions. CT can show larger strokes but is less
sensitive to small-vessel disease.
With respect to neurocognitive function, patients who have any degree of premorbid
decline of higher cognitive function or have significant risk for stroke should undergo
baseline neuropsychological examination, not simply the administration of a single
screening test. Table 1 shows the broad spectrum of domains that should be evaluated,
and representative tests for each sphere of function that should be administered by a
qualified neuropsychologist. Lezak’s review of neuropsychological tests (20) is among
the most comprehensive compendia of well-validated, standardized measures in the
field. Depending on the index of suspicion of premorbid deficits and the time available
for testing, such batteries can last from 1 to 8 h. It is not advisable to have patients perform
such tests late in the evening prior to the day of surgery. There can be nonspecific
cognitive changes just after surgery, so follow-up should not be done until the time of
discharge, only if specifically indicated, with a more valid evaluation occurring no earlier
than 4 wk postsurgery, and preferrably at 8–12 wk. If depression appears to be a part of
the clinical picture, then a psychiatric consultation is indicated.
REFERENCES
1. The Veterans Administration Coronary Artery Bypass Surgery Cooperative Study Group. Eleven-year
survival in the Veterans Administration randomized trial of coronary bypass surgery for stable angina.
N Engl J Med 1984;311:1333–1339.
2. Murkin JM, Boyd WD, Ganapathy S, et al. Beating heart surgery: why expect less central system
morbidity? Ann Thorac Surg 1999;68:1498–1501.
3. Brillman J. Central nervous system complications in coronary artery bypass graft surgery. Neurol Clin
1993;11:475–495.
surgery. N Engl J Med 1996;335;1857–1863.
5. Gardner TJ, Horneffer PJ, Manolio TA, et al. Major stroke after coronary artery bypass graft surgery:
changing magnitude of the problem. J Vasc Surg 1986;3:684–694.
6. Loop FD, Cosgrove CM, Lytle BW, et al. An 11-year evolution of coronary arterial surgery (1968–1978).
Ann Surg 1979;190:444–455.
7. Reed GL, Singer DE, Picard EH, et al. Stroke following coronary artery bypass surgery. N Eng J Med
1988;319:1246–1250.
8. Bull DA, Neumayer LA, Hunter GC, et al. Risk factors for stroke in patients undergoing coronary artery
bypass grafting. Cardiovasc Surg 1993;1:182–185.
9. Breuer AC, Furlan AJ, Hanson MR, et al. Central nervous system complications of coronary artery
bypass graft surgery: prospective analysis of 421 patients. Stroke 1983;14(5):682.
10. Murkin JM, Martzke JS, Buchan AM, et al. A randomized study of the influence of perfusion technique
and pH management strategy in 316 patients undergoing coronary artery bypass surgery. II. Neurologic
and cognitive outcomes. J Thorac Cardiovasc Surg 1995;110:349–362.
11. Tuman KJ, McCarthy RJ, Najafi H, et al. Differential effects of advanced age on neurologic and cardiac
risks of coronary operations. J Thorac Cardiovasc Surg 1992;104:1510–1517.
12. Stamou SC, Hill PC, Dangas G. Stroke after coronary artery bypass: incidence, predictors, and clinical
outcome. Stroke 2001;32:1508–1513.
13. Shaw PJ, Bates D, Cartlidge NEF, et al. Neurologic and neuropsychological morbidity following major
surgery-comparison of coronary artery bypass and peripheral vascular surgery. Stroke 1987;18:700–707.
14. Wijdicks EFM, Jack CR. Coronary artery bypass grafting-associated ischemic stroke. J Neuroimag
1996;6:20–22.
15. Trehan N, Mishra M, Kasliwal RR, et al. Reduced neurological injury during CABG in patients with
mobile aortic atheromas: a five-year follow-up study. Ann Thorac Surg 2000;70:1558–1564.
16. Sotaniemi KA. Long-term neurologic outcome after cardiac operation. Ann Thorac Surg
1995;59:1336–1339.
17. Mackey WC, Khabbaz K, Bojar R, et al. Simultaneous carotid endarterectomy and coronary bypass:
perioperative risk and long term survival. J Vasc Surg 1996;24:58–64.
18. Jahangiri M, Rees M, Edmondson SJ, et al. A surgical approach to coexistent coronary and carotid artery
disease. Heart 1997:77;164–167.
19. Yoon B-Y, Bae H-J, Kang D-W, et al. Intracranial cerebral artery disease as a risk factor for central
nervous system complications of coronary artery bypass graft surgery. Stroke 2001;32:94–99.
20. Lezak M. Neuropsychological Assessment. 3rd ed. New York: Oxford, 1995.
21. Lazar RM, Marshall RS, Pile-Spellman J, Young WL, Sloan RP, Mohr JP. Continuous time estimation
as a behavioral index of human cerebral ischemia during temporary occlusion of the internal carotid
artery. J Neurol Neurosurg Psychiatr 1996; 60:559–563.
22. Lazar RM, Connaire C, Marshall RS, Pile-Spelman J, Hacein-Bey L, Solomon RA, Sisti MB, Young
WL, Mohr JP. Developmental learning disorders in adult patients with cerebral arteriovenous malfor-
mations. Arch Neurology 1999;56:103–106.
23. Mohr JP. Acute clinical trials: an expression of concern Cerebrovasc Dis 1999(suppl);3:45–50.
24. McKhann GM, Borowicz LM, Goldborough MA, et al. Depression and cognitive decline after coronary
artery bypass grafting. Lancet 1997;349:1282–1284.
25. de Groot JC, de Leeuw FE, Oudkerk M, et al. Cerebral white matter lesions and cognitive function: the
Rotterdam Scan Study. Ann Neurol 2000(Feb);47:145–151.
26. Selnes OA, Goldsborough MA, Borowicz LM, McKann GM. Neurobehavioural sequelae of cardiop-
ulmonary bypass. Lancet 1999;353:1601–1606.
27. McKann GM, Goldsborough MA, Borowicz LM, et al. Cognitive outcome after coronary artery bypass:
A one-year prospective study. Ann Thorac Surg 1997;63:510–515.
28. Mahanna EP, Blumenthal JA, White WD, et al. Defining neuropsychological dysfunction after coronary
artery bypass grafting. Ann Thorac Surg 1996;61:1342–1347
29. Murkin JM, Newman SP, Stump DA, et al. Statement of consensus on assessment of neurobehavioral
outcomes after cardiac surgery. Ann Thorac Surg 1995;59:1289–1295.
30. Murkin JM, Stump DA, Blumenthal JA, et al. Defining dysfunction: group means versus incidence
analysis—a statement of consensus. Ann Thorac Surg 1997;64:904–905.
31. Selnes OA, Goldborough MA, Borowicz LM, et al. Determinants of cognitive change after coronary
artery bypass surgery: a multifactorial problem. Ann Thor Surg 1999;67:1669–1676.
32. Newman MF, Kirchner JL, Jerry L, et al. Longitudinal assessment of neurocognitive function after
coronary-artery bypass surgery. N Engl J Med 2001;344:395–402.
33. Stump DA, Rogers AT, Hammon JW, et al. Cerebral emboli and cognitive outcome after cardiac surgery.
J Cardio Vasc Anesth 1996;10:113–119.
34. Moody DM, Brown WR, Challa VR, et al. Brain microemboli associated with cardiopulmonary bypass:
a histologic and magnetic resonance imaging study. Ann Thorac Surg 1995;59;1304–1307.
35. Jacobs A, Neveling M, Horst M, et al. Alterations of neuropsychological function and cerebral glucose
metabolism after cardiac surgery are not related only to intraoperative microembolic events. Stroke
1998;29:660–667.
36. Degirmenci B, Durak H, Hazan E, et al. The effect of coronary artery bypass surgery on brain perfusion.
J Nucl Med 1998;39:587–591.
37. DeLaPaz RL, Mohr JP. Magnetic resonance imaging. In Barnett HJM, Mohr JP, Stein BM, Yatsu FM,
eds. Stroke: Pathophysiology, Diagnosis, and Management. 3rd ed. New York: Churchill-Livingstone,
1998;227–256.
38. Sylivris S, Levi C, Matalanis G, et al. Pattern and significance of cerebral microemboli during coronary
39. Warach S, Chen D, Li W, Ronthal M, Edelman RR. Fast magnetic resonance imaging of acute human
stroke. Neurology 1992;42:1717–1723.
40. van Everdingen KJ, van der Grond J, Kappelle LJ, et al. Diffusion-weighted magnetic resonance imag-
ing in acute stroke. Stroke 1998;29:1783–1790.
41. Wityk RJ, Goldsborough MA, Hillis A. Diffusion and perfusion-weighted magnetic resonance imaging
in patients with neurological complications after cardiac surgery. Arch Neurol 2001;58:571–576.
42. Symes ES, Maruff P, Ajani A, et al. Issues associated with the identification of cognitive change
following coronary artery bypass grafting. Austral N Z J Psychiatr 2000;34:770–784.
43. Lazar RM, Heitjan DH, Kurlansky P, et al. Randomized trial of on- vs off-pump CABG reveals baseline
memory dysfunction. Circulation 2001;104:II-815.
44. Fleiss J. The Design and Analysis of Clinical Experiments. New York: Wiley, 1986.
45. Makatura TJ, Lam CS, Leahy BJ, Castillo MT, Kalpakjian CZ. Standardized memory tests and the
appraisal of everyday memory. Brain Injury 1999;13:355–367.
46. Burgess PW. Alderman N. Evans J. Emslie H. Wilson BA. The ecological validity of tests of executive
function. J Int Neuropsychol Soc 1998;4:547–558.
47. Johnsson P, Lundqvist C, Lindgren A, et al. Cerebral complications after cardiac surgery assessed by
S100 and NSE levels in blood. J Cardiothorac Vasc Anesth 1995;9:694–699.
48. Kumar P, Dhital K. Hossein-Nia M, et al. S-100 protein release in a range of cardiothoracic surgical
procedures. J Thorac Cardiovasc Surg 1997;113:953–954.
49. Kilminster S, Treasure T, McMillan T, et al. Neuropsychological change and S-100 protein release in
130 unselected patients undergoing cardiac surgery. Stroke 1999;30:1869–1874.
50. Hernandez F, Clough RA, Klemperer JD, et al. Off-pump coronary artery bypass grafting: initial expe-
rience at one community hospital. Ann Thorac Surg 2000;70:1070–1072.
51. Malheiros SMF, Brucki SMD, Gabbai AA, et al. Neurological outcome in coronary artery surgery with
and without cardiopulmonary bypass. Acta Neurol Scand 1995;92:256–260.
52. Andrew MJ, Baker RA, Kneebone AC, et al. Neuropsychological dysfunction after minimally invasive
direct coronary artery bypass grafting. Ann Thorac Surg 1998;66:1611–1617.
53. Murkin JM, Boyd WD, Ganapathy S, et al. Beating heart surgery: why expect less central nervous system
morbidity? Ann Thorac Surg 1999;68:1498–1501.
54. Diegeler A, Hirsh R., Schneider F, et al. Neuromonitoring and neurocognitive outcome in off-pump
55. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state.” J Psychiat Res 1975;12:189–198.
56. Wechsler D. WMS-III Manual. San Antonio, TX: The Psychological Corporation, 1997.
57. Schmidt M. Rey auditory verbal learning test. Los Angeles: Western Psychological Services, 1996.
58. Delis DC, Kramer JH, Kaplan E, et al. California Verbal Learning Test—second edition. San Antonio,
TX: The Psychological Corporation, 2000.
59. Meyers JE, Meyers KR. Rey Complex Figure Test and Recognition Trial. Odessa, FL, 1995.
60. Goodglass H, Kaplan E, Barresi B. The Assessment of Aphasia and Related Disorders. 3rd ed.
Hagerstown, MD: Lippincott Williams & Wilkins, 2000.
61. Kaplan E, Goodglass H, Weintraub S. The Boston Naming Test. 2nd ed. Hagerstown, MD: Lippincott
Williams & Wilkins, 2000.
62. Wechsler D. WAIS-III Manual. San Antonio, TX: The Psychological Corporation, 1997.
63. Reitan RM, Wolfson D. The Halstead-Reitan Neuropsychological Test Battery: theory and clinical
interpretation. Tucson, AZ: Neuropsychology Press, 1985.
64. Sachs TL, Clark CR, Pols RG, et al. Comparability and stability of performance on six alternate forms
of the Dodrill-Stroop color-word test. Clin Neuropsychol 1991;5:220–225.
65. Lafayette Instrument Company. Lafayette, IN.
66. Lafayette Instrument Company. Lafayett, IN.
Chapter 17 / Multivessel Off-Pump Revascularization 229
17 Multivessel Off-Pump
Revascularization in High-Risk Patients
Severe Left Ventricular Dysfunction
Daniel J. Goldstein, MD,

Robert B. Beauford, MD, Patricia Garland, MD,
and Craig R. Saunders, MD
CONTENTS
INTRODUCTION
METHODS
PATIENTS
OPERATIVE TECHNIQUE
RESULTS
QUALITY OF LIFE
DISCUSSION
REFERENCES
INTRODUCTION
Off-pump coronary artery bypass grafting (OPCAB) is increasingly being applied as
an alternative to conventional cardiopulmonary bypass-supported myocardial
revascularization. Indeed, the last harvest of the Society of Thoracic Surgeons database
demonstrates that in the year 2001, nearly 19% of patients who had isolated coronary
revascularization underwent a beating-heart procedure (Table 1). Despite the rising
enthusiasm for the clinical use of OPCAB, there has been a reluctance to apply this
approach to patients with severe left ventricular dysfunction, for fear of hemodynamic
instability and increased morbidity and mortality.

229
Table 1
Proportion of Isolated Coronary Bypass Surgery Procedures Performed Off-Pump
According to Society of Thoracic Surgeons Database, 1998–2001
1998 1999 2000 2001
No. of isolated coronary bypass procedures 181,774 155,581 138,826 128,530
Percent off-pump 2.5 6.9 13.7 18.6
It is well known that patients with impaired left ventricular function undergoing
revascularization on cardiopulmonary bypass have increased mortality and morbidity
when compared to patients with normal left ventricular function (1). It has been specu-
lated that extracorporeal circulation may exacerbate myocardial damage in compromised
left ventricles as a result of: (1) activation of inflammatory mediators (2); (2)
nonphysiological ventricular geometry of the empty heart impeding collateral flow to
ischemic areas (3); and (3) worsened preservation of interventricular septal movement
(4). In view of these factors and because patients with depressed left ventricular function
often have several accompanying comorbidities, we hypothesized that beating-heart
revascularization may impact favorably on the suboptimal early and intermediate out-
comes associated with cardiopulmonary bypass-supported revascularization in this ill
group of patients. To this effect, we review our institutional experience and that in the
literature to shed light on the safety and feasibility of this approach in this high-risk cohort
of patients.
METHODS
We interrogated our prospectively (daily) updated database (CAOS, Intelligent Busi-
ness Solutions, Clemmons, NC) to identify all patients who underwent OPCAB at our
institution between January 1, 1999, and July 31, 2001, and who had a preoperative left
ventricular ejection fraction (LVEF) of 30% or less as measured by echocardiography,
nuclear imaging, and/or ventriculography. The chosen dates ensured that all surgeons’
learning curves had been overcome and that the technique was standardized among
all surgeons.
The incidence of adverse events among this high-risk group were compared to the
Society of Thoracic Surgeons (STS) benchmarks for all CABG patients, as those for
patients with severe left ventricular dysfunction in the database were not available to us.
National STS guidelines and definitions were used to define adverse events and out-
comes. Patients and referring cardiologists were contacted for follow-up data.
While no formal quality-of-life instruments (MLHF, SF-36) were used, we contacted
each patient to assess for surrogates of quality-of-life factors. In particular, we investi-
gated the incidence of recurrent angina, hospital readmission, hospital readmission for
cardiac causes, need for recatheterization, and need for repeat CABG. We also inquired
of the patients whether they would undergo surgery again knowing what it entailed.
PATIENTS
During the study period, 1624 patients underwent coronary artery bypass grafting as
an isolated procedure or accompanied by transmyocardial laser revascularization or
carotid endarterectomy. Of these, 911 (56%) were performed as off-pump procedures.
Among these, 811 (89%) patients had an LVEF of greater than 30% and 100 (11%) had
an LVEF less or equal to 30%. The latter constitutes the study group.
Patients’ demographic profile is depicted in Table 2. The mean age was 67 yr and the
typical patient was a Caucasian male. Hypertension, previous myocardial infarction, and
a history of smoking were prevalent comorbidities. Approximately half of the
patients were receiving β-blockers, angiotensin-converting-enzyme inhibitors, and
diuretics, but the duration of treatment with these agents was not available. Mean LVEF
was 26 ± 4.6%. The majority of patients presented with symptoms of congestive heart
failure. For those patients without angina, efforts were made to determine viability on
nuclear imaging studies or dobutamine stress echocardiography. Left main disease,
defined as >50% stenosis, and triple-vessel disease, were rather prevalent in this population.
OPERATIVE TECHNIQUE
All patients underwent general endotracheal anesthesia with placement of continuous
output Swan–Ganz catheter monitoring, transesophageal echocardiography (TEE), and
arterial pressure monitoring. A thorough TEE evaluation was undertaken in each patient
to assess wall-motion abnormalities, presence and degree of mitral regurgitation, severity
of atherosclerotic disease of the aorta, and right and left ventricular function. Significant
atherosclerotic disease of the aortic arch and/or descending aorta prompted evaluation of
the ascending aorta with epicardial echo probe. Presence of more than 2+ mitral insuf-
ficiency was a contraindication to off-pump revascularization. Those patients were placed
on cardiopulmonary bypass and underwent conventional revascularization and mitral
valve repair.
After harvest of conduits, the pericardium was opened. Extent of disease, epicardial
anatomy (i.e., intramuscular?), and size of target vessels as well as size of the heart were
assessed. Presence of extensive calcification, deep intramyocardial vessel, small target
(<1.5 mm) and/or significant cardiomegaly were relative contraindications to proceed
with off pump grafting.
After the decision had been made to undertake off-pump grafting, four deep pericar-
dial sutures with snare protectors were placed in a straight line between the left inferior
pulmonary vein and the inferior vena cava. The snare protectors served to avoid erosion
of the taut sutures through the epicardium. The right pleura was routinely opened, and
pericardial sutures were not used on the right side of the heart.
In general, distal anastomoses were performed first and the aorta was partially
occluded only once. On rare occasions, a proximal anastomosis was performed first if a
severely disease aorta required use of a facilitating anastomotic device. In most instances,
the left internal mammary artery-to-LAD anastomosis was performed first, unless a
totally occluded right coronary artery was present, in which case the latter was
revascularized first. Full systemic heparinization and complete protamine reversal were
used in most instances. No antifibrinolytic therapy (i.e., aminocaproic acid, aprotinin)
was used. Intracoronary shunts were rarely used. No partial bypass circuits or adjunctive
retracting devices were used.
Intraoperative Doppler graft flow assessment (Medi-Stim Butterly Flowmeter,
Medtronic, MN) was performed on a selective basis at the discretion of the surgeon. EKG
tracing, transesophageal wall motion assessment, visual and manual inspection of the
graft as well as Doppler flow interrogation were all used to determine need to revise an
anastomosis.
Table 2
Demographic Profile of Patients (n = 100) with Severe Left
Ventricular Dysfunction Undergoing OPCAB
Age 67 ± 10
% Female 21
Race (%)
Caucasian 69
Black 21
Hispanic 7
Asian 3
CADa risk factors (%)
Hypertension 77
Diabetes 37
Family history 14
Smoking
Current 27
Former 50
Never 22
Unknown 1
Previous MIb 49
Obesity 14
Medication use (%)
β-blockers 57
Diuretics 55
ACE inhibitors 48
Intravenous nitrates 32
Digoxin 19
Inotropes 11
Renal insufficiencyc (%) 16
LVEFd (%) 26 ± 4.6
NYHAe class 2.4 ± 0.6
Congestive heart failure (%) 65
Left main disease > 50% 37
Surgical priority (%)
Elective 24
Urgent 67
Emergent 8
Salvage 1
aCoronary artery disease.
bMyocardial infarction.
cDefined as preoperative creatinine greater than 1.5 mg/dL.
dLeft ventricular ejection fraction.
eNew York Heart Association.
RESULTS
One hundred consecutive patients with LVEF ≤ 30% underwent OPCAB. One addi-
tional patient was converted to conventional CABG because of severe cardiomegaly.
Intraoperative details are summarized in Table 3. Most patients underwent isolated
OPCAB, but a few patients had concomitant transmyocardial laser revascularization or
Table 3
Intraoperative Variables
Prior CABG (%) 4
Hemodynamic instability (%) 10
Intraaortic balloon pump (%)
Preoperative 16
Intraoperative 6
Operation (%)
OPCABa 89
Redo OPCAB 4
OPCAB + TMRb 5
OPCAB + CEAc 2
Number of grafts
Mean 3.5 ± 1.3
Range 1–7
% Internal mammary artery use 83
Mean skin-to-skin OR time 3 h, 57 min
aOff-pump coronary artery bypass.
bTransmyocardial revascularization.
cCarotid endarterectomy.
Table 4
Incidence of Major Complications and Comparison to STS
Database Benchmark Figures for all CABG Patients
STS National Database
Complication Study Benchmark for all CABG
Reoperation for bleeding 4% 2.32%
Atrial fibrillation 20% 19.37%
Q-wave MI 0% 1.08%
Stroke 1% 1.65%
Transient ischemic attack 1% 0.74%
Prolonged ventilationa 26% 5.46%
Mediastinis 1% 0.70%
Renal failureb 3% 3.14%
Reoperation for graft occlusion 0% 0.15%
Readmission within 30 d 6% 5.22%
aDefined as greater than 24 h.
bDefined as 50% rise in baseline creatinine or new need for dialysis.
carotid endarterectomy. Intra-aortic balloon support was prevalent. Sixteen percent of

patients had a balloon placed preoperatively, and 6% intraoperatively. Emphasis was
placed on the use of arterial grafts (internal mammary and radial arteries) and on complete
revascularization.
The incidence of adverse events is displayed in Table 4. These are compared to the STS
database benchmarks for all CABG patients regardless of left ventricular function. Median
postoperative length of stay was 7 d (range 4–88 days). Six patients were readmitted
234
234
Part II / Minimally Invasive Bypass Grafting

Fig. 1. Kaplan–Meier estimates of survival for patients with severe left ventricular dysfunction undergoing multivessel OPCAB.
Table 5
Quality-of-Life Events Following OPCAB
Quality-of-Life Event
Rehospitalization (%)
None 63
Cardiac-related 7
Non-cardiac-related 6
Unknown 10
Angina (%)
Yes 16
No 75
Unknown 10
Recatheterization for recurrent angina (%)
Yes 1
No 89
Unknown 10
Would have surgery again?
Yes 77
No 8
Not Sure 5
Unknown 10
Figures are rounded to nearest number and may
not add up to 100%.
within 30 d of surgery. Reasons included sternal dehiscence (2), wound infection (2),
near-syncope (1), and recurrent angina (1). Observed mortality was 3%, while the pre-
dicted (STS) mortality was 5.3%, with an observed-to-expected (O/E) ratio of 0.56.
Individual predicted STS mortalities ranged from 0.41% to 50.9%.
The three deaths were due to low-output syndromes, one on the day of surgery and two
on postoperative d 4. There were 13 late deaths over a mean follow-up time of 13.5 mo.
Causes of late death were cardiac-related in 11, a stroke in 1 patient, and unknown in 1
patient. Kaplan–Meier estimates of survival are depicted in Fig. 1. Six- and 12-mo
survivals were 94% and 88%, respectively.
QUALITY OF LIFE
Follow-up was 90% complete. Mean follow-up was 13.5 ± 9 mo. Freedom from
recurrent angina was 83%, freedom from hospital readmission was 69%, and freedom
from readmission for cardiac causes was 92%. One patient underwent recatheterization.
No patients required repeat surgical revascularization. When asked if they would have
surgery again knowing what the process entailed, 77% answered yes, 8% answered no,
and 5% were unsure. Ten patients could not be reached (see Table 5).
DISCUSSION
The introduction of enabling technologies in the mid-1990s paved the way for the
dissemination of beating-heart revascularization. Prior to this, OPCAB procedures had
been largely in the realm of a few pioneering groups in South America (5,6) and limited
to patients with one- or two-vessel (noncircumflex) disease.
Most of the literature addressing beating-heart revascularization is focused on the

application of this technology to low-risk patients. In fact, the six published randomized
trials comparing coronary bypass grafting with and without cardiopulmonary bypass
(7–12) did not include any patients with severe left ventricular dysfunction. This, we
believe, reflects a skepticism regarding the safety of off-pump surgery in patients with
compromised left ventricular function. Many investigators, including our group, feel that
the greatest benefits derived from avoidance of extracorporeal circulation will become
manifest not in low-risk individuals, but in patients with significant comorbidities.
Extensive literature exists evaluating the outcomes of conventional bypass grafting in
patients with severe left ventricular dysfunction (13–15), but a limited number of such
patients has been submitted to off-pump revascularization. Sternik et al. (16) retrospec-
tively compared 64 patients operated on without cardiopulmonary bypass to 53 who
underwent conventional operation. Mean number of grafts was much lower in the off-
pump group (1.9 grafts/patient) vs the conventional group (3.5 grafts/patient), with
almost no patients receiving circumflex revascularization in the former group. The early
and midterm survival were better for the off-pump cohort, but this probably reflects the
lesser severity of disease present in the off-pump population. Another report from the
same group (17), again investigating the safety of beating-heart approaches in patients
with compromised left ventricular function, described 75 patients with ejection fraction
less than 35%. There was a 2.7% mortality rate and 1.3% stroke rate. Again, a mean of
1.9 grafts were performed, suggesting less extensive disease or incomplete revascularization.
Arom and colleagues (18) compared in retrospective fashion 45 OPCAB vs 132 con-
ventional CABG recipients with ejection fraction less than 30%. The authors highlighted,
as we do, that cardiac displacement is well tolerated in these patients. Fewer grafts were
performed in the OPCAB group, which benefited from reduced postoperative blood loss.
The incidence of adverse events in both groups was comparable for all variables studied.
Meharwal and colleagues (19) published the largest study to date. The authors reported
on 355 patients with ejection fractions of 30% or less, with an average age of 58 yr, who
underwent OPCAB. Mean number of grafts was 2.8 (vs 3.3 in the on-pump cohort) and
mortality was 3.9%. The incidence of atrial fibrillation, prolonged ventilation, and hos-
pital stay was reduced in off-pump recipients. This group of patients from India appear
to be younger and with less triple-vessel disease than that reported in most CABG series
in the recent literature.
In an effort to enhance the limited experience available describing patients with com-
promised left ventricular function undergoing beating-heart coronary bypass, we report
on the use of off-pump coronary bypass in 100 consecutive patients with severe (LVEF
< 30%) ventricular dysfunction. Unlike the aforementioned studies, our population of
patients had a very high incidence of triple-vessel disease. Moreover, the mean number
of grafts performed in our patients (3.5 per patient) is much higher than that achieved in
the literature and is highly indicative of complete revascularization. In contrast to these
earlier studies, revascularization of the circumflex territory was routine. We were very
pleased with the low incidence of adverse events, which parallel the benchmark incidence
of these events for all CABG patients, regardless of ejection fraction, in the STS database.
We presume that the incidence of adverse events for patients with LVEF ≤ 30 in the
database would be much higher, but these data were not available to us. We did not fare
so well with early extubation, probably a result of the high use of balloon pumps and
inotropes in the early postoperative setting in combination with a high prevalence of

active or recent smoking (77%) in our population of patients.
Only three perioperative deaths (3% mortality) were observed despite an expected
5.3% mortality. The incidence of early hospital readmission was remarkably low and was
mostly related to wound complications. No instances of acute graft occlusion occurred
and no patients have required a reoperation.
The low perioperative mortality and incidence of adverse events in this series can be
attributed to several factors. First is the achievement of complete revascularization.
Completeness of revascularization has been found to improve early survival in young
(20) and elderly (21) patients, and we believe it to be a critical factor in patients with left
ventricular dysfunction. Second, the liberal use of balloon pump support stabilizes hemo-
dynamics and allows the cardiac mobilization necessary for exposure of lateral and
inferior wall targets. It is of interest that Dietl et al. documented improved survival for
the cohort patients with ejection fractions less than 25% who underwent conventional
CABG with preoperative balloon insertion as compared to those who did not receive
counterpulsation support (22). Finally, and perhaps most important, is the presence of a
dedicated cardiac anesthesiologist constantly vigilant of hemodynamics and progress of
the operation, who can alert the surgeon to the need to return the verticalized heart to a
more physiological position and to rapidly create a proximal anastomosis or to insert a
shunt for hemodynamically compromising ischemia. Moreover, the experienced off-
pump anesthesiologist is adept at using inotropes and pressors as necessary based on
hemodynamics and continuous TEE evaluation of the degree of mitral regurgitation and
can anticipate the next step in the operation and adjust vasoactive drips and bed position-
ing as necessary to ensure a smooth operation.
On follow-up, most of the patients were free from angina, rehospitalization, and
recatheterization at a mean of 13.5 mo postoperatively. When asked if they would go
through the procedure again, the majority of patients responded affirmatively. The 88%
1-yr survival and good quality of life achieved in this ill group of patients lends validity
to the safety and efficacy of beating-heart revascularization for patients with severely
compromised left ventricular function.
While the present study represents one of the largest series of consecutive patients with
severe left ventricular dysfunction undergoing OPCAB with multivessel revascularization,
several limitations of the study deserve mention. First, because of its retrospective
design, no concurrent conventional CABG cohort exists for comparison. A concurrent
retrospective cohort comparison was not available. A second and probably most signifi-
cant limitation is the introduction of selection bias. The decision to proceed with OPCAB
or conventional CABG was the sole decision of the operating surgeon, made at the time
of operation, based usually on anatomy or clinical findings. Recently, however, we
approach all patients for revascularization as potential off-pump candidates and tend to
select patients for OPCAB preferentially when we think they would benefit most from
elimination of CPB. These tend to be the sicker, more morbid patients. As with many
other surgical undertakings, we doubt that a randomized trial evaluating the safety and
efficacy of OPCAB in severe left ventricular dysfunction will ever be realized, and the
best alternative is the accumulation of large series of consecutive patients with prospec-
tively collected data. Second, no long-term follow-up is available to confirm the encour-
aging early results. The addition of formal instruments to measure quality of life would
be of great interest to enhance our understanding of the benefits of off-pump

revascularization in this ill group of patients.
In our hands, OPCAB is a safe and effective approach for revascularization of the
patient with severe left ventricular dysfunction. Liberal use of intra-aortic balloon pump
is recommended, and complete revascularization is achievable and desirable. To confirm
the benefit of this approach, long-term follow-up is mandatory, but we believe that every
coronary surgeon should have equal expertise in on- and off-pump revascularization so
that the bypass operation can be optimally tailored to any individual patient, regardless
of anatomy or comorbidities.
REFERENCES
1. O’Connor GT, Pume SK, Olmstead EM, et al. Multivariate prediction of in-hospital mortality associated
with coronary artery bypass graft surgery. Circulation 1992;85:2110–2118.
2. Moshkovitz Y, Sternik L, Paz Y, et al. Primary coronary artery bypass grafting without cardiopulmonary
bypass in impaired left ventricular function. Ann Thorac Surg 1997;63:S44–S47.
3. Kirklin JW, Barrat-Boyes BG, eds. Cardiac Surgery, 2nd ed. New York: Churchill-Livingstone,
1993:83–97.
4. Akins CW, Boucher CA, Pohost GM. Preservation of interventricular septal function in patients having
coronary artery bypass with cardiopulmonary bypass. Am Heart J 1984;107:304–309.
5. Benetti FJ, Naselli G, Wood M, Gefner L. Direct myocardial revascularization without extracorporeal
circulation. Chest 1991;100:312–316.
6. Gomes LF. Coronary artery bypass grafting without cardiopulmonary bypass. Ann Thorac Surg
1996;61:63–66.
7. van Dijk D, Jansen EW, Hijman R, et al. Cognitive outcome after off-pump and on-pump coronary artery
bypass graft surgery. JAMA 2002;287:1405–1412.
8. Diegeler A, Hirsch R, Schneider F, et al. Neuromonitoring and neurocognitive outcome in off-pump
9. Ascione R, Lloyd CT, Underwood MJ, Lotto AA, Pitsis AA, Angelini GD. Inflammatory response after
coronary revascularization with of without cardiopulmonary bypass. Ann Thorac Surg 2000;69:1198–1204.
10. Ascione R, Caputo M, Calori G, Lloyd CT, Underwood MJ, Angelini GD. Predictors of atrial fibrillation
after conventional and beating heart coronary surgery. Circulation 2000;102;1530–1535.
11. Wandschneider W, Thalmann M, Trampitsch E, Ziervogel G, Kobinia G. Off pump coronary bypass
operations significantly reduce S100 release: an indicatory for less cerebral damage? Ann Thorac Surg
2000;70:1577–1579.
12. van Dijk D, Nierich AP, Jansen EW, et al. Early outcome after off-pump versus on-pump coronary
bypass surgery. Circulation 2001;104:1761–1766.
13. Mickleborough LL, Carson S, Tamariz M, Ivanov J. Results of revascularization in patients with severe
left ventricular dysfunction. J Thorac Cardiovasc Surg 2000;119(3):550–557.
14. Trachiotis GD, Weintraub WS, Johnston TS, Jones EL, Guyton RA, Craver JM. Coronary artery bypass
grafting in patients with advanced left ventricular dysfunction. Ann Thorac Surg 1998;66:1632–1639.
15. Elefteriades JA, Morales DM, Gradel C, Tollis G Jr., Levi E, Zaret BL. Results of coronary artery bypass
grafting by a single surgeon in patients with left ventricular ejection fractions < or = 30%. Am J Cardiol
1997;79:1573–1578.
16. Sternik L, Moshkovitz Y, Hod H, Mohr R. Comparison of myocardial revascularization without cardiop-
ulmonary bypass to standard open heart technique in patients with left ventricular dysfunction. Eur J
17. Moshkovitz Y, Sternik L, Paz Y, et al. Primary coronary artery bypass grafting without cardiopulmonary
bypass in impaired left ventricular function. Ann Thorac Surg 1997;63:S44–S47.
18. Arom KV, Flavin TF, Emery RW, Kshettry VR, Petersen RJ, Janey PA. Is low ejection fraction safe for
off-pump coronary bypass operation ? Ann Thorac Surg 2000;70:1021–1025.
19. Meharwal ZS, Trehan N. Off-pump coronary artery bypass grafting in patients with left ventricular
dysfunction. Heart Surg Forum 2002;5(1):41–45.
20. Osswald BR, Tochtermann U, Schweiger P, Thomas G, Vahl CF, Hagl S. Does the completeness of
revascularization contribute to an improved early survival in patients up to 70 years of age? Thorac
21. Osswald BR, Blackstone EH, Tochtermann U, et al. Does the completeness of revascularization affect
early survival after coronary artery bypass grafting in elderly patients? Eur J Cardiothorac Surg
2001;20:120–125.
22. Dietl CA, Berkheimer MD, Woods EL, Gilbert CL, Pharr WF, Benoit CH. Efficacy and cost-effective-
ness of preoperative IABP in patients with ejection fraction of 0.25 or less. Ann Thorac Surg
1996;62:401–408.
18 Multivessel Off-Pump
Revascularization in High-Risk
Populations
Octogenarians
Frederic Sardari, MD,

Robert B. Beauford, MD,
and Daniel J. Goldstein, MD
CONTENTS
INTRODUCTION
RESULTS
SUMMARY
REFERENCES
INTRODUCTION
The elderly population in the United States is growing exponentially. It is estimated
that by the year 2050 there will be 38 million octogenarians (up from 9 million in 2002)
(1). Increasingly, government initiatives are being created in an attempt to qualify their
health status, ascertain their functional limitations, and identify their morbidity and
mortality. It is well known that 25% of octogenarians have severe functional limitations
secondary to cardiovascular disease (2). Furthermore, cardiovascular disease has been,
and continues to be, the number one cause of death in people over 65 yr of age (3). Many
of these patients have disease refractory to medical therapy, which has led to a marked
increase in the number of elderly patients being referred for coronary revascularization.
Over the last decade, numerous investigators have reported on the feasibility and efficacy
of cardiac surgery in this population. Most remarkably, the preponderance of studies
underscore the high-risk profile of elderly patients with their higher prevalence of

241
comorbidities and left ventricular dysfunction as well as the greater severity of coronary
artery disease (4–10). Several reports indicate that coronary revascularization can be
performed with an acceptable mortality in octogenarians (4,6,7,11) and efforts are now
aimed at further decreasing morbidity and mortality after coronary bypass in this older
population.
The growing popularity of off-pump coronary artery bypass (OPCAB) has been fueled
by developments in enabling technologies and by early data suggesting improved coro-
nary surgery outcomes by avoidance of extracorporeal circulation. Several comparative
studies have documented reduced transfusion requirements, shortened length of stay, and
reduced hospital costs for OPCAB recipients when compared to those undergoing con-
ventional coronary artery bypass (12–18).
Octogenarian patients represent a particularly attractive target for application of
OPCAB. The prevalence of comorbidities and the propensity for neurological dysfunc-
tion place octogenarians at higher risk for cardiopulmonary bypass (CPB)-induced
morbidity and mortality. Despite this, scarce data exist documenting the safety and
efficacy of OPCAB in these elderly patients. To this effect we report our recent experi-
ence with a large series of octogenarian patients undergoing beating-heart revascularization.

Our prospectively (daily) updated database (CAOS, Intelligent Business Solutions,
Clemmons, NC) was queried to identify all patients who underwent bypass grafting at
Newark Beth Israel Medical Center and Saint Barnabas Hospital. Between January 1,
1999, and July 31, 2001, a total of 1624 were identified. Of these, 911 (56%) had OPCAB
procedures, including 113 octogenarians. The latter represents our study group.
Demographics, clinical profiles, and adverse outcomes were collected. Complica-
tions were compared to STS benchmarks for all CABG patients. Follow-up was obtained
by telephone interviews with patients and referring cardiologists. Information was gath-
ered regarding quality-of-life surrogates including anginal symptoms, subsequent
hospitalizations and/or procedures, and date and cause of death where applicable. Addi-
tionally, the patients were also asked whether, in retrospect, they would undergo the
procedure again.
Operative Technique
All patients underwent OPCAB under general endotracheal anesthesia. Continuous
output Swan–Ganz catheter, transesophageal echocardiography (TEE) probe, and arte-
rial pressure monitoring lines were placed. A thorough TEE exam was performed to
assess wall-motion abnormalities, presence and degree of mitral regurgitation, severity
of atherosclerotic disease of the aorta, and right and left ventricular function. Significant
atherosclerotic disease (i.e., mobile atheromata, wall thickening, etc.) of the aortic arch
and/or descending aorta prompted evaluation of the ascending aorta with epicardial echo
probe. Presence of more than 2+ mitral insufficiency precluded an off-pump procedure,
and conventional cardiopulmonary bypass was undertaken with mitral valve repair. Other
relative contraindications to OPCAB included extensively calcified, deep intramyocardial,
or small target (<1.5 mm) vessels, as well as significant cardiomegaly.
All operations were performed through a median sternotomy incision. Briefly, the
conduits (left internal mammary artery, radial artery, or saphenous vein) were harvested
and the pericardium was opened widely. Elevation and stabilization of the heart was
accomplished using four deep pericardial sutures with snare protectors. The right pleura
was opened widely. Pericardial sutures were not routinely used on the right side of the
heart. In general, distal anastomoses were performed first and the aorta was partially
occluded only once. The anastomoses were performed with the aid of the Octopus 2
(Medtronic, Inc., Minneapolis, MN) stabilizing system, and their order was flexible. Full
systemic heparinization and complete protamine reversal were used in most instances.
No antifibrinolytic therapy (i.e., aminocaproic acid, aprotinin) was used. Intracoronary
shunts were rarely used. No partial bypass circuits or adjunctive apical retracting devices
were used.
Intraoperative Doppler graft flow assessment (Medi-Stim Butterly Flowmeter,
Medtronic, Minneapolis, MN) was performed at the discretion of the operating surgeon.
Transesophageal wall motion, EKG tracing, and visual and manual inspection of the
grafts were all considered in evaluation of the conduits. Failure to obtain pulsatility
indices between 1 and 5 or flows greater than 15 cc/min usually led to immediate graft
revision.
RESULTS
One hundred thirteen consecutive octogenarians underwent OPCAB. The demographic
profile and clinical characteristics are depicted in Table 1. Mean age was 83 ± 2.5 yr. The
majority (81%) of patients had a history of hypertension. Previous myocardial infarct,
tobacco use, diabetes, and congestive heart failure were also prevalent comorbidities.
The mean left ventricular ejection fraction (LVEF) was 51 ± 11%, and 11% had a history
of previous neurological event.
The operative profile is summarized in Table 2. Three patients underwent concomitant
transmyocardial laser revascularization or carotid endarterectomy; all other patients
underwent isolated OPCAB. Intra-aortic balloon support was rarely used. Emphasis was
placed on internal mammary artery use and on complete revascularization as illustrated
by the mean number of grafts performed.
Adverse events are depicted in Table 3 and are compared to the STS database bench-
marks for all CABG patients, regardless of age. The most prevalent complication was
development of atrial fibrillation postoperatively, which occurred in 43% of the patients.
Postoperative cerebrovascular accident occurred in four patients (3.6%).
The postoperative length of stay ranged from 4 to 33 d with a mean of 9 d. There were
nine readmissions within 30 d. Of these, three were cardiac-related, two were pulmonary-
related, and two were wound-related. Altered mental status and a toe amputation com-
prised the other two.
In this series, there was only one postoperative death. Our observed mortality was 0.9,
which compares favorably to the STS predicted mortality of 6.0%, resulting in an
observed-to-expected (O/E) ratio of 0.15. This death occurred on postoperative d 33 and
was due to a cerebrovascular accident. There were three late deaths over a mean follow-
up time of 13.2 mo. One death was cardiac-related, the other two were secondary to
malignancy. Kaplan–Meier estimates of survival are depicted in Fig. 1. Six-, 12-, and
24-mo survivals were 99%, 96%, and 90%, respectively.
Quality of Life
Mean follow-up was 13.2 ± 7 mo and was complete in 90% of the study group
(Table 4). Eighty-five of 98 patients (87%) available for follow-up were free from
Table 1
Demographic Profile and Clinical Characteristics
of Octogenarians Undergoing OPCABa (n = 113)
Variable N (%)
Age (yr)
Mean 83 ± 2
Gender
Male 67 (59)
Female 46 (41)
Race
Caucasian 97 (86)
African-American 8 (7)
Hispanic 3 (3)
Asian 5 (4)
Comorbidities
Hypertension 92 (81)
Previous MIb 45 (40)
Smoker 43 (38)
Diabetes 39 (35)
CHFc 35 (31)
Obesity 17 (15)
Previous CVAd 12 (11)
Renal insufficiencye 16 (14)
Family history 6 (5)
Medication use
β-Blockers 74 (65)
ACE inhibitors 42 (37)
Diuretics 37 (33)
Intravenous nitrates 31 (27)
Digoxin 16 (14)
Inotropes 5 (4)
Mean LVEFf 51 ± 11%
Left main disease > 50% 53 (47)
Priority
Elective 21 (19)
Urgent 85 (75)
Emergency 7 (6)
aOff-pump coronary artery bypass.
bMyocardial Infarct.
cCongestive heart failure.
dCerebrovascular accident.
eDefined and creatinine >1.5.
fLeft ventricular ejection fraction.
recurrent angina at the time of telephone survey. Sixty-six percent of contacted patients
were free from hospital readmission for any reason, and 90% were free from readmission
for cardiac causes. Recatheterization was required in seven patients, but none required
surgical revascularization. When asked if, in retrospect, they would undergo the surgery
again knowing what it entailed, most of the patients answered in the affirmative.
Table 2
Operative Profile for Study Population
Variable N (%)
Operation
OPCABa 110 (97)
OPCAB/TMRb 1 (1)
Redo OPCAB/TMR 1 (1)
OPCAB/CEAc 1 (1)
Intra-aortic balloon pump
Pre-op/intra-op 5/2 (6)
Internal mammary artery use 94 (83)
Number of grafts
Mean 3.3 ± 1
Range 2 to 6
Mean skin-to-skin OR time 3 h, 34 min
aOff pump coronary artery bypass.
bTransmyocardial revascularization.
cCarotid endarterectomy.
Table 3
Incidence of Major Complications in the Study Population and Comparison
to STS National Database Figures for All CABG Patients
Study STS National Database
Complication N (%) Benchmark for All CABGb (%)
Atrial fibrillation 49 43% 19.4%
Readmission within 30 d 9 8% 5.2%
New renal failurea 6 5% 3.1%
Reexploration for bleeding 5 4% 2.3%
Neurological complication (CVA) 4 4% 1.7%
Q-wave myocardial infarction 1 1% 1.1%
Mediastinitis 1 1% 0.7%
Reexploration for graft occlusion 0 0% 0.2%
aDefined as 50% rise in baseline creatinine or new need for dialysis.
bRegardless of age.
SUMMARY
The explosive growth of off-pump bypass grafting parallels the growth of elderly
patients being referred for revascularization. Most of the literature, however, addresses
the use of OPCAB in younger patients with lower risk profiles, highlighting the trend to
exclude this population of patients when considering newer technologies. As the learning
curve with OPCAB has been overcome and technical skill with the procedure has been
enhanced, many investigators, including our group, feel that the greatest benefit derived
from avoidance of cardiopulmonary bypass will be realized not in low-risk individuals
but in those patients who fall into a higher-risk profile, including those with multiple
comorbidities and the elderly.
Table 4
Quality of Life
Mean follow-up 13.2 ± 7 mo (n = 98)a
Rehospitalization
None 66%
Cardiac-related 10%
Non-cardiac-related 23%
Angina
Yes 13%
No 87%
Recatheterization
Yes 7%
No 93%
Would have surgery againb
Yes 66%
No 26%
Undecided 9%
a15 patientslost to follow-up, including 4 deaths.
b(n = 93) 5 patients were followed up exclusively
through the cardiologist’s office; this subjective
information was therefore not available.
Figure 1
An extensive body of literature exists documenting the experience with conventional

CABG in octogenarians. The largest study to date was conducted by Peterson et al. (19).
The authors surveyed the Medicare Provider Analysis and Review file from 1987 to 1990
and returned 24,461 octogenarians undergoing cardiac revascularization. This study
documented a 67% increase in the national use of bypass surgery for octogenarians.
Noted was a significant escalation in the lengths of stay and hospital costs as well as
in-hospital and short-term (3-yr) mortality when compared with younger patients (ages
65–70). Their 30-d, 1-yr, and 3-yr mortality rates were 11.5%, 19.3%, and 28.8%, respec-
tively. Interestingly, the long-term mortality for patients in this population was similar
to that of the general octogenarian population. Predictors of mortality prior to this study
were depressed left ventricular function, diabetes mellitus in combination with ventricu-
lar dysfunction, and presence of congestive heart failure or need for urgent surgery.
Peterson and colleagues noted that these predictions were based on single-institution
experiences with relatively small numbers of patients (maximum of 159). Using national
Medicare data, this group had access to more than 24,000 patients, from which predictors
of mortality could be extrapolated. They found that increasing age, female sex, preopera-
tive myocardial infarction, and congestive heart failure identified patients with higher
30-d and 3-yr mortality. Chronic renal disease, peripheral vascular disease, and cerebral
vascular disease also independently predicted patients with higher short- and long-term
mortality. Diabetes mellitus and pulmonary disease, though not predictive of increased
30-d mortality, were predictive of octogenarians with worse long-term outcomes.
Morris and colleagues (20) investigated coronary artery bypass grafting in octogenar-
ians at Hahnemann University between 1987 and 1994 (n = 474). Though their focus was
the efficacy of conduits, this still represents one of the largest octogenarian study popu-
lations since the benchmark study by Peterson and colleagues. They reported a mortality
rate of 7.8%. Comparatively, a cohort of younger patients (<80 yr old) had a mortality rate
of 3.5%. This further documents the age-associated increase in mortality. In their
patients, as for most octogenarians referred for coronary revascularization, they saw a
trend of increased comorbidities and greater severity of disease. Their most prevalent
morbidity was respiratory failure (9.8% patients). Postoperative cerebrovascular acci-
dent was seen in 6.9% of patients and myocardial infarct was noted in the postoperative
period in 4.4%. Follow-up in this study averaged 46 mo and was complete in 92% of
patients. The long-term survival was approx 68% at 5 yr.
Pliam et al. (4) scrutinized their recent improvements in the results of coronary bypass
surgery in octogenarians. Their search for all octogenarians undergoing CABG at the San
Francisco Heart Institute between 1985 and 1996 yielded 404 patients. They divided this
study group into two groups of 202 to compare their old and more recent experience. They
report statistically significant decreases in postoperative days to extubation, days in the
intensive care unit, and lengths of stay in hospital. Also significantly decreased were rates
of complications [34.2% vs 22.8% (p = 0.03)] and mortality [12.9% vs 5.4% (p = 0.01)].
Williams and colleagues (6) specifically investigated determinants of mortality in
octogenarians undergoing coronary bypass. Their investigation at Miami’s Mount Sinai
Medical Center yielded 300 octogenarians undergoing isolated CABG. Their major
morbidities included pulmonary complications (20%), renal dysfunction (13%), and
cerebrovascular accident (CVA) (2.3%). Their principal determinants of mortality were
preoperative renal dysfunction and postoperative pulmonary insufficiency, renal dys-
function, intraoperative aortic balloon pump use, and sternal wound infection. Surgical
priority was also a major determinant of mortality, as their elective surgery mortality rate
was 9.6% and that of their emergent operations was 33%. Their overall mortality rate was
11%. Their actuarial survival for patients discharged from the hospital was 74.6% ± 5.6%
at 54 mo.
In short, there is no debate that the number of octogenarians being referred for
revascularization has increased significantly in recent years. Additionally, these patients
tend to have more comorbidities, more significant left ventricular dysfunction, and more
Table 5
Published Series of Mortality Rates
and CVA Rates in Octogenarians Who Had OPCAB
Author n Mean No. of Grafts Mortality CVA %
Ricci et al. 2000 (23) 97 1.8 10.3% 0.0%
Stamou et al. 2000 (24) 71 1.6 6.0% 3.0%
Yokoyama et al. 2000 (26) 28 3.2 0.0% 7.1%
Present study, 2002 113 3.3 0.9% 3.5%
severe symptomatology. The literature also collectively suggests that CABG in octoge-
narians is a safe operation with acceptable mortality and long-term survival.
Scarce data exist, however, on the use of OPCAB in octogenarians (Table 5). Ricci
et al. (21) retrospectively reviewed myocardial revascularization in octogenarians with
and without bypass. Of 269 octogenarians who had coronary revascularization at SUNY
Buffalo between 1995 and 1999, 197 were done conventionally and 97 without cardiop-
ulmonary bypass. This study noted a significant incidence of postoperative complica-
tions in elderly patients compared to younger patients, but noted that elderly patients
displayed a significantly higher rate of freedom from complications after OPCAB. Eighty-
five percent of off-pump octogenarians were free from postoperative complications,
compared with 75% of the CPB group. Most notable, however, was a 0% stroke rate in
the OPCAB group (vs 9.3%). They did, however, note that there was a trend toward a
higher mortality rate (10.3%).
Stamou and colleagues (22) reported on their experience with all patients undergoing
OPCAB who were 60 years of age and older. Among their 71 octogenarians, they report
a CVA rate of 3% as compared to 1% in patients 70–79 yr old and 0.3% in patients and
60–69. More significant findings were higher incidence of pneumonia, postoperative
atrial fibrillation, and the need for inotropic support in the older age group. Similar to that
published in the conventional CABG literature for this age group, the postoperative
length of stay is prolonged compared to younger patients. They reported predictors of
prolonged length of stay including preoperative congestive heart failure (CHF), previous
CVA, ejection fraction less than 34%, and postoperative atrial fibrillation and inotropic
support. Octogenarians in this study had a mortality rate of 6%, which was statistically
significant compared to 3% and 0.3% seen in the 70–79-yr-olds and the 60–69-yr-olds,
respectively.
It is notable that both Ricci and colleagues and Stamou et al. included MIDCAB
patients in their patient population. Additionally, their mean number of grafts was less
than 2, which in our opinion is a different study group from the octogenarians we studied,
who had a mean number of grafts over 3.
Finally, Yokoyama et al. (23) investigated coronary bypass in high-risk subgroups.
Though this series was not designed specifically for octogenarians, they did report 28
octogenarians who underwent OPCAB who had a mean of 3.2 grafts per patient. The
authors reported reductions in renal complications, postoperative bleeding, and days in
the ICU or on the ventilator. This study documented a 7.1% stroke rate and a 0% mortality
rate. They concluded that eliminating cardiopulmonary bypass reduced the overall inci-
dence of postoperative complications.
To our knowledge, our series of 113 patients represents the largest series of consecu-
tive octogenarians undergoing multivessel OPCAB reported in the literature. In analyz-
ing our results it is worth pointing out that surgical intervention was undertaken in an
urgent or emergency manner in 92 patients (81%). This factor alone is considered an
independent risk factor for morbidity in cardiac surgery patients (24–26). Moreover, the
mean number of grafts was 3.3 per patient, which is commensurate with data for conven-
tional coronary artery bypass and significantly higher than that reported in the literature
with OPCAB.
Our most prevalent complication was the development of postoperative atrial fibril-
lation. It is well known that the overall incidence of atrial fibrillation increases incremen-
tally with age, and this factor is multiplied in patients with coronary artery disease
(27,28). With the exception of atrial fibrillation, the rate of complications was very low.
These results (Table 3) parallel the STS benchmark data for all CABG patients, regard-
less of age. Although not available, we presume that the morbidity in octogenarians in
the STS database would likely be magnified given their higher risk profile.
Stroke is reported to occur in 1–9% of patients undergoing myocardial revascularization,
and increasing age is reported to be the leading risk factor (29–32). In a 24-institution
randomized trial, Roach et al. (29) report the incidence of stroke after revascularization
was approximately 6% in all patients. In this study they note an incidence of neurological
events in octogenarians to be approx 8%. In the small retrospective studies referenced in
the present study, the general consensus is that there is a significant reduction in stroke
rate after OPCAB compared with conventional revascularization with cardiopulmonary
bypass (9,21–23,31–34). Our results further support these findings. A previous history
of neurological event, seen in 11% (12) of our patients preoperatively, and the incidence
of atrial fibrillation postoperatively, seen in 43% of our patients (49), have both been
reported as independent risk factors for CVA (29), yet our postoperative stroke rate was
3.5% (4 patients).
Despite an expected high mortality, only one patient died in the perioperative period
(0.9% mortality rate). During sternal closure, this patient was noted to have ischemic
changes on EKG and was hemodynamically unstable, ultimately requiring cardiopulmo-
nary bypass support and redo of one of the grafts. Though the patient recovered from the
acute episode, she suffered a CVA later in the hospital course and ultimately died on
postoperative d 33. There were only 3 late deaths (1 cardiac), which brings the 1-yr
mortality rate to 3.6%.
There are limited data regarding follow up in octogenarians after OPCAB. Our early-
outcome analysis correlates strongly with the literature for conventional CABG in that
octogenarians can lead event-free lifestyles after this major operation, evidenced by 80%
of our patients living chest-pain-free and only 9% necessitating readmission for reasons
related to the surgery or for chest pain. The best indicator of our octogenarians’ satisfac-
tion with the surgery is the fact that the majority would have the surgery again knowing
what was involved.
In spite of our satisfying results, several limitations exist that warrant mention. This
study was a retrospective review, and while a randomized study comparing conventional
CABG to OPCAB would be ideal, it is unlikely to occur. Additionally, serial neurological
exams, head imaging, or formal neurocognitive evaluations were not conducted on our
series of patients, and hence, minor events, e.g., clinically silent strokes, could have
gone unrecognized.
The results of the present study strongly suggest that off-pump multivessel
revascularization in octogenarians is associated with excellent early and intermediate
outcomes and provides a highly satisfactory quality of life. While extended follow-up is
mandatory to confirm these encouraging early findings, we preferentially approach all
octogenarians as potential off-pump candidates.
REFERENCES
1. Health United States: 2001; Current Population Reports, “Americans with Disabilities, 1997” P70-73,
February 2001, and related Internet data; Internet releases of the Census Bureau and the National Center
on Health Statistics; and unpublished tables from the Bureau of Labor Statistics.
2. Van Nostrand JF, Furrier SE, Suzman R, eds. Health data on older Americans in the United States, 1992.
National Center for Health Statistics. Vital Health Stat 1993;3(27).
3. Sahyoun NR, Lentzner H, Hoyert D, Robinson KN. Trends in causes of death among the elderly. Aging
Trends; No.1. Hyattsville, MD: National Center for Health Statistics, 2001.
4. Pliam MB, Zapolanski A, Ryan CJ, Shaw RE, Mengarelli L. Recent improvement in results of coronary
bypass surgery in octogenarians. J Invas Cardiol 1999;11:281–289.
5. Sollano JA, Rose EA, Williams DL, et al. Cost-effectiveness of coronary artery bypass surgery in
octogenarians. Ann Surg 1998;228(3):297–306.
6. Williams DB, Carrillo RG, Traad EA, et al. Determinants of operative mortality in octogenarians
undergoing coronary bypass. Ann Thorac Surg 1995;60(4):1038–1043.
7. Deiwick M, Tandler R, Mollhoff T, et al. Heart surgery in patients aged eighty years and above:
determinants of morbidity and mortality. Thorac Cardiovasc Surgeon 1997;45(3):119–126.
8. Naunheim KS, Dean PA, Fiore AC, et al. Cardiac surgery in the octogenarian. Eur J Cardiothorac Surg
1990;4(3):130–135.
9. D’Ancona G, Karamanoukian H, Kawaguchi AT, Ricci M, Salerno TA, Bergsland J. Myocardial
revascularization of the beating heart in high-risk patients. J Card Surg 2001;16(2):132–139.
10. Batchelor WB, Anstrom KJ, Muhlbaier LH, et al. Contemporary outcome trends in the elderly under-
going percutaneous coronary interventions: results in 7,472 octogenarians. National Cardiovascular
Network Collaboration. J Am College Cardiol 2000;36(3):723–730.
11. Mullany CJ, Darling GE, Pluth JR, et al. Early and late results after isolated coronary artery bypass
surgery in 159 patients aged 80 years and older. Circulation 1990;82(5 Suppl):IV229–236.
12. Hart JC. A review of 140 Octopus off-pump bypass patients over the age of seventy: procedure of choice?
Heart Surg Forum 2001;4 Suppl 1:S24–29.
13. Prifti E, Bonacchi M, Giunti G, et al. Does on-pump/beating-heart coronary artery bypass grafting offer
better outcome in end-stage coronary artery disease patients? J Card Surg 2000;15(6):403–410.
14. Cleveland JC Jr, Shroyer AL, Chen AY, Peterson E, Grover FL. Off-pump coronary artery bypass
grafting decreases risk-adjusted mortality and morbidity. J Ann Thorac Surg 2001;72(4):1282–1288.
15. Weintraub WS, Craver JM, Cohen CL, Jones EL, Guyton RA. Influence of age on results of coronary
artery surgery. Circulation 1991;84(Suppl 3):226–235.
16. Kilger E, Weis FC, Goetz AE, et al. Intensive care after minimally invasive and conventional coronary
surgery: a prospective comparison. Intensive Care Med 2001;27(3):534–539.
17. Bittner HB, Savitt MA, McKeown PP, Lucke JC. Off-pump coronary artery bypass grafting. Excellent
results in a group of selected high-risk patients. J Cardiovasc Surg 2001;42(4):451–456.
18. Demers P, Cartier R. Multivessel off-pump coronary artery bypass surgery in the elderly. Eur J Cardio-
Thorac Surg 2001;20(5):908–912.
19. Peterson ED, Cowper PA, Jollis JG, et al. Outcomes of coronary artery bypass graft surgery in 24,461
patients aged 80 years or older. Circulation 1995;92(9 Suppl):II85–91.
20. Morris RJ, Strong MD, Grunewald KE, et al. Internal thoracic artery for coronary artery grafting in
octogenarians. Ann Thorac Surg 1996;62(1):16–22.
surgery with and without cardiopulmonary bypass. Ann Thorac Surg 2000;69(5):1471–1475.
22. Stamou SC, Dangas G, Dullum MK, et al. Beating heart surgery in octogenarians: perioperative outcome
and comparison with younger age groups. Ann Thorac Surg 2000;69(4):1140–1145.
23. Yokoyama T, Baumgartner FJ, Gheissari A, Capouya ER, Panagiotides GP, Declusin RJ. Off-pump
versus on-pump coronary bypass in high-risk subgroups. Ann Thorac Surg 2000;70(5):1546–1550.
24. Gersh BJ, Kronmal RA, Frye RL, et al. Coronary arteriography and coronary artery bypass surgery:
morbidity and mortality in patients ages 65 years or older. A report from the Coronary Artery Surgery
Study. Circulation 1983;67(3):483–491.
25. Tasdemir O, Vural K, Karagoz H, Bayazit K. Coronary artery bypass grafting on the beating heart
without the use of extracorporeal circulation: review of 2052 cases. J Thorac Cardiovasc Surg
1998;116:68–73.
26. Hirose H, Amano A, Takahashi A. Coronary artery bypass grafting for octogenarians: Experience in a
private hospital and review of literature. Ann Thorac Cardiovasc Surg 2001;7:282–291.
27. Aranki SF, Shaw DP, Adams DH, et al. Predictors of atrial fibrillation after coronary artery surgery.
Current trends and impact on hospital resources. Circulation. 1997;96(6):2084–2085.
28. Furberg CD, Psaty BM, Manolio TA, Gardin JM, Smith VE, Rautaharju PM. Prevalence of atrial
fibrillation in elderly subjects (the Cardiovascular Health Study). Am J Cardiol 1994;74(3):236–241.
surgery. Multicenter study of perioperative ischemia research group and the Ischemia Research and
Education Foundation Investigators. N Engl J Med 1996;335(25):1857–1863.
30. Tuman KJ, McCarthy RJ, Najafi H, Ivankovich AD. Differential effects of advanced age on neurologic
and cardiac risks of coronary artery operations. J Thorac Cardiovasc Surg 1992;104(6):1510–1517.
31. Gardner TJ, Horneffer PJ, Manolio TA, Hoff SJ, Pearson TA. Major stroke after coronary bypass
surgery: changing magnitude of the problem. J Vasc Surg 1986;3684–3687.
32. Iglesias I, Murkin JM. Beating heart surgery or conventional CABG: are neurologic outcomes different?
Semin Thorac Cardiovasc Surg 2001;13(2):158–169.
33. Hirose H, Amano A, Takahashi A. Off-pump coronary artery bypass grafting for elderly patients. Ann
Thorac Surg 2001;72(6):2013–2019.
34. Koutlas TC, Elbeery JR, Williams JM, Moran JF, Francalancia NA, Chitwood WR Jr. Myocardial
revascularization in the elderly using beating heart coronary artery bypass surgery. Ann Thorac Surg
2000;69(4):1042–1047.
Chapter 19 / Minimally Invasive Mitral Valve Surgery 253
III MINIMALLY INVASIVE VALVULAR SURGERY

19 Minimally Invasive Mitral Valve Surgery
Victor F. Chu, MD, L. Wiley Nifong, MD,

and W. Randolph Chitwood, Jr., MD
CONTENTS
INTRODUCTION
MINIMALLY INVASIVE MITRAL SURGERY: EVOLUTION
THE “MICRO-MITRAL OPERATION”
CLINICAL EXPERIENCE
DISCUSSION
REFERENCES
INTRODUCTION
Interest in minimally invasive cardiac surgery in general, and mitral valve surgery in
particular, has grown exponentially among cardiac surgeons and their patients. This
enthusiasm can be attributed to improved enabling technologies, excellent clinical
results, and enhanced communication. The authors believe that a truly “minimally inva-
sive approach” is a philosophy and it consists of: (1) limited or no cardiopulmonary
perfusion; (2) very small incisions; (3) minimal musculoskeletal retraction; (4) careful
hemostasis; and (5) meticulous blood conservation. At present, mitral valve operations
require cardiopulmonary bypass (CPB) support and thus, we have directed our focus
toward “minimalism” by the use of small incisions, videoscopic vision, robotic assis-
tance, modified perfusion techniques, simplified aortic occlusion, and optimal myocar-
dial preservation during mitral valve surgery. This chapter provides a synopsis of our
mid-term minimally invasive mitral valve surgery experience along with a description of
new methodologies.
MINIMALLY INVASIVE MITRAL SURGERY: EVOLUTION

Developments in minimally invasive mitral surgery began in the mid-1990s with the
pioneering work of Cohn (1), Cosgrove (2), Navia (3), and others. Technological

255
256 Part III / Minimally Invasive Valvular Surgery
Table 1
Carpentier-Loulmet Classificaion of Degrees of Surgical Invasiveness
Level I
Mini-incision (10–12 cm)
Direct vision
Level II
Micro-incision (4–6 cm)
Video-assisted
Level III
Micro or port incision (1–2 cm)
Video-directed
Level IV
Port incision with robotic instruments
Video-directed
advancements in instrumentation, assisted vision, and CPB support have followed closely
and have expedited this evolutionary process. Within a few short years we have gone
from simple modifications of conventional techniques to near totally endoscopic opera-
tions, using computer-based telemanipulation, and assisted vision. Although a myriad of
approaches have arisen, these procedures can be best classified using the Carpentier/
Loulmet nomenclature, which is based on incision size and visualization methods (Table 1).
In this scheme, higher classification levels denote smaller incision sizes and increased
reliance on visual assistance technology. Of course, the uncompromised objective is to
preserve the same quality of valve repair or replacement as is achieved with traditional
sternotomy, full-access operations.
Level I: Mini-Incision and Direct Vision

Level I operations differ from traditional “open” approaches in that mini-incisions
(10–12 cm) are used instead of either a full median sternotomy or thoracotomy (20–30 cm).
A variety of reduced incisions, including mini-sternotomies, para-sternotomies, and mini-
thoracotomies, have been shown to provide adequate mitral valve exposure. Using “mini-
incisions” and direct vision, simple modifications of standard techniques and instruments
alone are required, thus facilitating a surgeon’s transition from the traditional sterno-
tomy. Numerous groups (1,2,4,5) independently have reported encouraging results with
low mortalities (1–3%) and little morbidity using these methods. In 1994, surgeons at
Stanford University first deployed a peripheral perfusion system with intra-aortic occlu-
sion balloon for mitral valve surgery (6). Thereafter, this Port-Access™ system was used
successfully by a number of surgeons to perform Level I mitral repairs and replacements.
Early reports suggested more surgical complications (retrograde aortic dissection) than
expected with this device; however, operator experience and improved guidewires have
reduced this risk for later models. Many centers, ours included, believe that intra-aortic
balloon occlusion is associated with unnecessary increases in cost and complexity.
Instead, our group has favored transthoracic aortic clamp occlusion as a safe, economical,
and simple method for performing routine limited-access procedures (7,8).
Level I operations, with or without the Port-Access method, have produced results
comparable to that of the “open” or conventional sternotomy operations (9–11). Impor-
tantly, the success of Level I procedures has served as a new “comfort zone” and helped
launch more advanced techniques in minimization.
Level II: Micro-Incisions and Video-Assisted Vision

Level II operations rely on video-assisted vision to facilitate exposure through even
smaller incisions (4–6 cm). Interestingly, this idea is not new and, in fact, Cutler (12) and
Sakakibara (13) used cardioscopy clinically earlier in the twentieth century, but technical
limitations at the time precluded widespread adoption. In 1996, Kaneko first used video
assistance during a sternotomy-based mitral valve replacement (14). Carpentier per-
formed the first true video-assisted mitral valve repair in early 1996, through a mini-
thoracotomy using cold fibrillatory arrest (15). A few months later, our group in East
Carolina University (ECU) performed a mitral valve replacement using a transthoracic
aortic clamp and retrograde cardioplegia (8). In these early videoscopic operations an
assistant manipulated the telescopic camera.
Level III and IV: Micro- or Port-Incision

and Video-Directed, Robotic Operation
The addition of a voice-activated robot (AESOP 3000™) for camera control now
enables smoother, more precise video control, which further facilitates these operations.
The clinical series from ECU (7) and other groups (16,17) have demonstrated safety and
efficacy in using video assistance with micro-incisions for valve repairs. Moreover,
complex repairs with good results have become routine. Repair techniques have included
quadrangular resections, sliding valvuloplasties, chordal replacements, and transfers, as
well as edge-to-edge (Alfieri) leaflet approximations. After operators gain experience,
both anterior and posterior, as well as annular, pathology become repairable with this
technique. Currently, the AESOP 3000 video-assisted mini-thoracotomy approach is our
standard operation for primary and reoperative mitral valve operations.
Computer-assisted robotic systems potentially enable surgeons to overcome the limi-
tations of operating through small ports with long instruments. Telemanipulation is the
last step toward a totally endoscopic mitral valve operation. Mitral repairs using the da
Vinci™ Surgical System (Intuitive Surgical, Inc., Sunnyvale, CA) were first performed
in May 1997 by Carpentier in Paris and by Mohr in Leipzig (15,17). The first complete
robotic mitral repair with an annuloplasty ring in North America was performed at our
center using the da Vinci system (18). A somewhat similar Zeus™ system (Computer
Motion, Inc., Santa Barbara, CA) has been used by the New York University group to
perform partial mitral repairs (19). This system is reviewed in detail in Chapter 30. We
recently completed a 20-patient Phase I FDA mitral valve repair trial using the da Vinci
system. Moreover, a 10-center, FDA-approved robotic mitral surgery trial is well under-
way using this system. Over 75 completely robotic mitral operations have been per-
formed successfully in this trial to date. This study should elucidate both the benefits and
difficulties associated with robotic mitral surgery done by multiple surgeons in their
home institutions. With the aid of surgical robotic systems, it is now possible to perform
complex mitral operations using near-port incisions. With continuous development of
adjunctive innovations, it is possible to foresee totally endoscopic mitral operations in the
near future.
THE “MICRO-MITRAL OPERATION”

Since 1996, ECU surgeons have used a standardized minimally invasive video-
assisted approach, or micro-mitral operation (MMO), for mitral valve procedures. With
the exception of the da Vinci FDA trial patients, all isolated mitral valve surgery is
performed by this method. Key elements of MMO include: (1) mini-thoracotomy access;
Fig. 1. Minimally invasive technique for cannulating right femoral artery and vein.
(2) direct transthoracic aortic clamping; (3) telescopic video assistance; and (4) periph-
eral arterial perfusion and assisted venous return. Specialized surgical instruments have
been either developed or modified from other ones to perform these operations.
Preoperative Preparation
Independent left lung ventilation is obtained using either a double-lumen endotracheal
tube or a right-sided bronchial blocker. Subsequently, the patient is positioned on the
operating table with the right chest elevated by 40° and the shoulders tilted back. The right
arm is suspended carefully over the head on a padded holder. Alternatively, the right arm
can be positioned along the side, residing behind the posterior axillary line. External
defibrillator pads are positioned posterior to the right scapula and at the left anterior
axillary line near the fifth interspace, thereby subtending the greatest cardiac mass. A
transesophageal echocardiographic (TEE) probe is inserted and a baseline study
obtained. Patients are prepared in a similar manner for either the MMO or the robotic
operations.
Technical Considerations
A 3-cm incision is made just below the right inguinal fold and the anterior femoral
artery and vein are exposed with minimal dissection. Circumferential dissection and
clamping of both vessels are avoided and oval purse-string sutures (4-0 monofilament)
are placed longitudinally. Following heparinization, a 17–19 fr. Biomedicus™
(Medtronic, Minneapolis, MN) arterial cannula is advanced into the proximal iliac artery
using the Seldinger technique (Fig. 1). When femoral cannulation is contraindicated or
impossible, central aortic cannulation is performed. In this instance, a purse-string suture
Fig. 2. Sub mammary right mini-thoracotomy in the fourth intercostal space.
is placed just proximal to the innominate artery and the arterial cannula introduced under
videoscopic assistance using the Seldinger technique, either via the incision or through
a thoracoport. Specialized transthoracic cannulas (Straightshot™ and EndoDirect™,
Cardiovations, Inc., Sommerville, NJ) have been developed for this purpose. Both are
deployed without a guidewire, and this maneuver is facilitated by a retractable intralu-
minal blade tip. Alternatively, the right axillary artery can be exposed, and a short seg-
ment of 8-mm synthetic graft attached end-to-side. A conventional straight arterial
cannula is then inserted into the graft.
The anesthesia team inserts a 17 fr percutaneous cannula into the right internal jugular
vein to augment superior vena caval drainage. For the majority of cases, a 19–23 fr
Biomedicus femoral venous cannula is passed over a guide wire to the right atrium under
TEE visualization. Echo guidance is essential to confirm correct luminal passage and
destination of the venous cannula. A Biomedicus centrifugal pump is used for kinetic
bicaval drainage. When femoral vein insertion is contraindicated or catheter passage is
difficult, the right atrium can be cannulated directly.
A right 5-cm anterior axillary line submammary skin incision is made and the pecto-
ralis muscles are mobilized for fourth intercostal space thoracic entry (Fig. 2). A low-
Fig. 3. Cross-section view of the videoscopic micro-mitral operation setup. Note the presence of
transthoracic retraction suture for the pericardium.
profile thoracotomy retractor is used to deflect the soft tissues, while providing minimal
rib spreading. Vanermen has emphasized that videoscopic mitral surgery can be done
without any rib spreading, using only a soft tissue restraint (20). The pericardium is
opened 2-cm ventral to the phrenic nerve under direct vision and carried cephalad to the
aortic reflection. The anterior edge of the pericardium is tacked to incision edges using
silk sutures, while the posterior edge is distracted posterolaterally using transthoracic
sutures (Fig. 3). This maneuver rotates the heart counterclockwise, effectively displacing
the left atrium laterally and ventrad. This arrangement provides direct-vision exposure
and access to the aortic origin, atriocaval junction, and right superior pulmonary vein.
For camera placement, a 5-mm thoracoport (Genzyme-DSP, Boston, MA) is inserted
through the fourth intercostal space just dorsal to the incision. The port should be posi-
tioned with the tip ventral to the pericardial edge and parallel to the superior pulmonary
vein. A 5-mm 0° telescopic camera is then passed through the port and attached to the
Aesop 3000 robotic arm (Figs. 4 and 5). For da Vinci cases the camera is inserted through
the incision and the arm ports are placed in the second and fourth interspaces.
With videoscopic assistance, a purse-string suture is placed along the anterior ascend-
ing aorta, just distal to the right coronary origin. This is easier to accomplish with the
patient on full CPB and the right atrium decompressed. A suction-vent cardioplegia
catheter is inserted either through the thoracotomy or via a separate port incision. The
transthoracic clamp is inserted directly through the third intercostal space along the mid-
axillary line. Under videoscopic guidance, and with care not to injure either the right
pulmonary artery or the left atrial appendage, the clamp is positioned with the posterior
fixed-tine, directed through the transverse sinus (Fig. 6). Proper positioning of the clamp
Fig. 4. Patient setup for Aesop 3000-assisted videoscopic micro-mitral operation.
should be confirmed with the video camera. Compared to the traditional approach, tran-
sthoracic aortic clamping minimizes aortic and cardiac distortion and maintains aortic
valve competency. Consequently, intermittent antegrade cardioplegia can be used and
provides superb myocardial protection. If desired, a retrograde cardioplegia catheter can
be positioned into the coronary sinus under TEE guidance, either directly through the
atrial wall or percutaneously via the jugular vein.
Intrathoracic CO2 insufflation is begun before the atriotomy is performed, and is
maintained until the atriotomy is closed. Sondergaard’s interatrial groove is dissected
over only 1–2 cm, and the small left atriotomy should be made just medial to the right
superior pulmonary vein entrance. A fourth interspace transthoracic retractor (Heartport™
Fig. 5. Aesop 3000-controlled videoscopic assistance.
Inc.) is used to provide valve exposure by “toeing in,” which elevates the interatrial
septum toward the sternum, encouraging the anterior mitral leaflet to hang freely. In the
presence of large left atria, the lateral walls tend to collapse. Hence, early placement of
commissural sutures reestablishes anatomical orientation and exposure of the valve.
For mitral valve repairs, we prefer using both Carpentier–Edwards Physio™
annuloplasty rings and Cosgrove–Edwards™ bands. As the latter prosthesis extends only
posteriorly between the fibrous trigones, fewer sutures are required. Often all annular
sutures are placed before beginning leaflet or chordal repairs as visualization becomes
progressively improved. Sutures are placed in a counterclockwise fashion, beginning at
the right (posterior) fibrous trigone. The sutures are arranged serially and maintained in
Fig. 6. Transthoracic (Chitwood) aortic clamp. (Inset) Videoscopic confirmation of clamp place-
ment through transverse sinus.
Fig. 7. Endoscopic view of completed quadrangular resection with a posterior sliding plasty ring
annuloplasty.
external suture guides. Figure 7 depicts a completed P1 and P3 sliding plasty after a
posterior leaflet quadrangular resection with chordal transfers to A2. For mitral valve
replacements, subannular pledgeted 2-0 sutures are placed serially, while the anterior
leaflet is progressively “snipped” in a counterclockwise fashion. The posterior leaflet and
subvalvar apparatus are always left intact to optimize ventricular function. Along some
annular segments we may transition to supra-annular everting mattress sutures. The

combination of these suture techniques has resulted in no early perivalvular leakage. This
method facilitates videoscopic anterior annular suture placement while minimizing the
risk of mechanical prostheses impingement by the posterior leaflet.
Placement of anterior sutures often requires video assistance, whereas posterior su-
tures can usually be placed either by direct vision or endoscopically. Suturing maneuvers
at the left trigone and anterior commissure can be difficult and may require the aid of a
30° telescope. Optimal needle stability is essential while placing sutures through the
mitral annulus. This can be best achieved by the use of a short grip, shafted “Heartport™-
type” needle holder. When all sutures have been passed through the annulus, they are
inserted into the prosthesis extrathoracically before it is lowered into position through
the small atriotomy. The left trigone suture is tied first, followed by the right trigone and
the posterior middle sutures. A modified valve positioner/knot tier is used to seat the
valve and secure the knots. It is critical that the first two knots slip in order to achieve ideal
prosthesis to annular tissue apposition. Using a guillotine-type suture cutter, the remain-
ing sutures are cut to a uniform length.
The left atriotomy is closed under direct vision using 3-0 monofilament sutures. A
transvalvular vent is used routinely following valve replacements. Prior to aortic clamp
release, venous return is decreased, while both lungs are ventilated, and the intracardiac
air is evacuated through both the atriotomy suture line and aortic vent. All de-airing
maneuvers are performed with the patient in the Trendelenburg position and under TEE
guidance. Partial release of aortic occlusion and direct ventricular massage are carried out
until complete evacuation of air is documented. With continuous CO2 insufflation, we
have found it much easier to achieve complete cardiac deairing. After the transthoracic
aortic clamp is removed, the posterior aorta is examined with the videoscope. Thereafter,
the patient is weaned from CPB, and two small thoracostomy tubes are placed through
the existing stab incisions. Temporary epicardial pacing wires are best placed before
weaning from bypass.
Robotic Assistance with the da Vinci Surgical System

Over the last few years, computerized surgical “robotic” systems have been developed
and introduced into cardiac surgery. The current da Vinci surgical robotic system is a
master–slave (console–effector)-type telemanipulation device that consists of a 3-D high-
resolution camera and two robotic arms with interchangeable instrument tips (Fig. 8).
This system is described in detail in Chapter 31. The advantages of surgical robotic
systems include enhanced 3-D visualization with “tele-presence,” precise micromanipu-
lation with motion scaling and tremor “filtration,” along with an ergonomic operating
position for the surgeon. As of May, 2003, a total of 81 patients had undergone robotic
mitral operations in ECU, representing the largest clinical series in the world.
Patients are prepared as described above for the MMO, except that a smaller, 4-cm
chest incision is used. After the valve is inspected, ergonomic trajectories from the chest
wall to valve annulus are determined prior to insertion of left and right robotic arms. Most
frequently, the right trocar is placed in the fourth or fifth intercostal space, posterior-
lateral to the incision and parallel to the right superior pulmonary vein. The left trocar
generally is placed 6 cm cephalad and medial to the right trocar, ensuring sufficient
clearance between arms to avoid both external and internal arm conflicts. Optimal geo-
metric positioning avoids obtuse converging angles between arms, which decreases left
atrial wall tearing with instrument manipulation. The 3-D endoscope is placed through
Fig. 8. Operating room floor plan for da Vinci robotic mitral operations.
the medial portion of the mini-thoracotomy, and the patient-side assistant uses the
remainder of the incision as a working port (Fig. 9). A magnetic suture retriever and large-
bore suture vacuum enable the assistant in delivering supplies and removing refuse
rapidly. Surgeons operate from the console, placed 10 ft from the patient, which houses
the 3-D video screen and the master-control hand pieces (Fig. 10 and inset). In addition,
foot controls enable ergonomic hand repositioning and dynamic camera manipulations.
A variety of mitral repairs have been performed, including complex anterior leaflet
reconstructive procedures. In all da Vinci cases, a Cosgrove annuloplasty band was used.
Upon completion of the repair, robotic devices are removed from the operating table and
the atrium is closed under direct vision. Standard deairing and weaning procedures are
identical to that of the micro-mitral operations. Figure 11 depicts a da Vinci mitral valve
repair employing a posterior leaflet quadrangular resection.
CLINICAL EXPERIENCE
Our most recently published videoscopic mitral surgery experience includes 127
micro-mitral operations, of which 72 employed the Aesop 3000 voice-activated robotic
camera and 38 used the complete robotic (da Vinci) system (7,21). Table 2 shows the
demographic data for both groups. For comparison, retrospective analyses of 100
consecutive patients who underwent traditional mitral operations at our institution are
included. Preoperative gender, New York Heart Association functional class (NYHA),
left ventricular ejection function (EF), and cardiac risk factors were similar among all
groups (p = ns). Operative techniques and perioperative data for all cohorts are shown in
Fig. 9. Cross-section view of da Vinci robotic mitral operations. Note that the scope is inserted via
the mini-thoracotomy incision.
Fig. 10. Surgeon operating at the console during da Vinci robotic mitral operation. (Inset) Close-
up view of the hand control unit.
Fig. 11. Close-up view of the mitral valve leaflet and subvalvar apparatus during da Vinci robotic
operation.
Table 2
Comparison of Patient Demographics Among the Three
Groups of Patients Undergoing Mitral Valve
Conventional Videoscopic Robotic (da Vinci)
Demographics (n = 100) (n = 127) (n = 38)
Age (yr) 62 57 56
% Female 48 50 31
NYHA I (%) 3 7 N/Aa
NYHA II (%) 41 46 N/A
NYHA III (%) 49 42 N/A
NYHA IV (%) 7 5 N/A
LVEF (%) 54 ± 1.2 55 ± 1.2 53 ± 1.3
Reop (%) 0 8.6 0
aNot available.
Tables 3 and 4, respectively. The only difference between the MMO and the da Vinci
groups was defined by the 39 prosthetic valve replacements in the former series. For both
videoscopic and da Vinci groups there was a progressive decline in both operative and
Table 3
Comparison of Surgical Techniques of Mitral Valve
Surgery Utilized Among Three Groups of Patients
Surgical Technique (n = 100) (n = 127) (n = 38)
Replacement 46 (46%) 39 (31%) 0
Repair 54 (54%) 88 (69%) 38 (100%)
Annuloplasty only 10 (10%) 43 (34%) 6 (16%)
Quadrangular resection 29 (29%) 33 (26%) 18 (47%)
Sliding plasty 0 3 (2%) 5 (13%)
Chordal transfer 6 (6%) 0 2 (5%)
Chordal replacement 0 2 (1.6%) 3 (8%)
Alfieri (edge-to-edge) repair 0 2 (1.6%) 3 (8%)
Other 9 (9%) 3 (2%) 1 (3%)
Table 4
Comparison of Operative Data and Length of Stay Among
the Three Surgical Strategies for Mitral Valve Surgery
Operative Data (n = 100) (n = 127) (n = 38)
Crossclamp time (min) 84.4 ± 2.7 106.5 ± 4.5 143 ± 5
Cardiopulmonary bypass time (min) 108.4 ± 3.6 155.9 ± 5.7 178 ± 6.5
Operation time (h) 3.1 ± 0.9 4.5 ± 0.1 4.8 ± 0.1
Ventilator time (h) 23 ± 4.7 15 ± 3.5 10.4 ± 1.1
ICU length of stay (d) 1.6 ± 0.2 1.1 ± 0.1 0.9 ± 0.1
Hospital length of stay (d) 7.9 ± 0.6 4.7 ± 0.4 3.8 ± 0.6
Patients transfused (%) 32 34 16
Units PRBCa transfused/patient 1.1 1 3.5
Units FFPb transfused/patient 0.7 0.2 N/Ac
Units platelets transfused/patient 1.2 0.1 N/A
Units cryoprecipitate transfused/patient 0.1 0 N/A
aPacked red blood cells.
bFresh frozen plasma.
cNot applicable.
crossclamp times. Our latest operative times for the videoscopic group are approaching
that of the conventional group (7). We now have performed over 200 video-assisted
mitral operations using long instruments.
Postoperative chest tube drainage was significantly less in patients who underwent
videoscopic operations. However, no group differences existed, either in percentage of
patients receiving transfusions or total amount of packed red blood cells, fresh frozen
plasma, or platelets transfused. Rank-order analysis of variance of ventilator dependency
revealed a significant difference between conventional and videoscopic cases, but not
between the later cohort and the robotic group. Both minimally invasive groups had
similar intensive care periods and shorter hospital lengths of stay compared to the cohort
of patients who underwent conventional mitral surgery. Comparative crossclamp and
perfusion times are shown in Figs. 12 and 13.
Fig. 12. The 127 videoscopic micro-mitral patients were separated into 5 consecutive cohorts of
25 patients. There was a consistent decline of CPB time and crossclamp time from the earliest
cohort (Group 1) to the latest cohort (Group 2).
Fig. 13. Comparison of CPB time and crossclamp time between the first 19 da Vinci robotic
patients (Group 1) and the last 19 patients (Group 2).
Seventeen patients in this cohort were reoperations, and there was one perioperative
death (5.8%) from postoperative pneumonia. Thirteen of these operations were per-
formed under hypothermic fibrillatory arrest without aortic crossclamping. The average
operative time and CPB times were 4.6 and 2.7 h, respectively. There were no injuries to
Table 5
Comparison of Adverse Events Among Three Operative Strategies for Mitral Valve Surgery
Adverse Event (n = 100) (n = 127) (n = 38)
Conversion to sternotomy 0 2 (1.5%) 0
Atrial fibrillation 27 (27%) 27 (21%) 7 (18%)
Sternal infection 1 (1%) 1 (0.8%) 0
Groin infection 0 1 (0.8%) 0
Transient ischemic attack 2 (2%) 1 (0.8%) 0
Stroke 1 (1%) 1 (0.8%) 1 (2.6%)
Bleeding 7 (7%) 1 (0.8%) 1 (2.6%)
Prolonged ventilation 13 (13%) 1 (0.8%) 0
Pneumonia 2 (2%) 1 (0.8%) 0
Renal failure 2 (2%) 1 (0.8%) 0
Deep vein thrombosis 0 1 (0.8%) 0
Myocardial infarction 1 (1%) 0 0
existing coronary grafts, strokes, or incision conversions in the entire cohort. This method
is quite effective when mitral valve operations are required in patients having prior
coronary surgery.
Table 5 shows comparative adverse everts. Total videoscopic MMO patient follow-
up was 90% complete at 25.1 ± 3.6 mo. Most recently, 97 patients (85.1%) were NYHA
class 0-I and 9 patients (7.9%) were class II. Five patients (5.1%) had initial improve-
ments of their symptoms but later returned with heart failure and residual mitral regur-
gitation. Of these, three had late (>38 mo) recurrences and required valve replacements.
The other two patients were dialysis-dependent and developed bacterial endocarditis
requiring valve replacements at 1 and 2 mo following repair. Overall 30-d mortality for
the videoscopic group was 2.3% (3/127 patients). One death was in-hospital on postop-
erative d 28 from pneumonia and the other two were home deaths on postoperative d 15
and d 29 from presumed arrhythmias. There were two conversions to sternotomy for
bleeding. One permanent stroke occurred, and there were two transient neurological
sequelae (brachial plexus, ataxia). One patient required prolonged ventilator support (32
d) from a phrenic nerve palsy.
Patient follow-up was 100% for the 38 patients in the da Vinci group, with a mean of
10.7 ± 0.9 mo. Postoperative echocardiograms showed grade 1 or 0 residual mitral
regurgitation (MR) in 34 patients (89%) and grade 2 MR in four patients (11%). There
were no operative deaths or device related complications. One patient was reexplored
through the same incision 6 h after surgery for pacing wire site bleeding but was dis-
charged from the hospital on postoperative d 3. One patient required mechanical valve
replacement on postoperative d 19 for residual leak hemolysis. This patient suffered a
fatal stroke later while on adequate warfarin therapy. We have found this device to
facilitate complex mitral reconstructions and operative times are beginning to fall progres-
sively. In the last 20 da Vinci mitral valve repairs, crossclamp time has averaged 126 min.
DISCUSSION
The authors believe that smaller incisions are beneficial to patients, and can provide
an uncompromised, quality operation. This has been demonstrated by our experiences
both with the video-assisted micro-mitral operations as well as with the da Vinci-per-
formed procedures. Using video assistance, we have been able to achieve superior mitral
valve exposure with a 4–5-cm skin incision without resecting ribs or cartilage. A small
skin incision also limits soft tissue dissection and rib retraction, which in turn minimizes
overall surgical trauma. Compared to patients with traditional sternotomy approaches,
our patients generally require less blood product transfusions, have less postoperative
pain, recover faster, and are discharged earlier.
We have identified several technical elements that are crucial for the success of these
videoscopic operations. Peripheral cannulation and assisted venous drainage in combi-
nation with transthoracic direct aortic clamping have provided excellent exposure and
cardiac protection. With the addition of the SVC drainage catheter via the internal jugular
vein, we have had no problems with cardiac decompression, even with smaller percuta-
neous femoral catheters. Transthoracic aortic clamping is simple, safe, and reliable.
Using a mini-thoracotomy and videoscope, the mitral valve can be exposed through a
small atriotomy without tense retraction on the heart. Not only is the visualization of the
valve and the subvalvar apparatus superior, the aortic valve remains relatively compe-
tent, permitting effective antegrade cardioplegia delivery in most cases.
Specially designed minimally invasive surgical instruments, such as the custom atrial
retractors, long-shafted forceps and needle drivers, as well as the valve pusher/knot tier,
are required to facilitate the MMO. As stated previously, the micro-mitral videoscopic
approach is our standard choice for isolated mitral operations. The da Vinci device
obviates the need for these long instruments, as the operative action is at the valvular site.
We believe that there are very few absolute contraindications to using these innovative
approaches. However, relative contraindications include: (1) a previous right thorac-
otomy, (2) previous radiation to right thorax, (3) severe pulmonary dysfunction, and (4)
severe pulmonary hypertension. A full sternotomy still affords the most options for
dealing with potential complications of severe pulmonary hypertension. Generally we do
not use either minimally invasive approach in patients with pulmonary systolic pressures
either over half-systemic or greater than 70 mmHg.
Based on our experience, as well as inference from past general surgical endoscopic
history, there is clearly a learning curve in the transition from conventional to videoscopic
mitral operations. Our operative and perfusion times in the latest videoscopic cases are
comparable to that of our conventional approach (7). Above all, obtaining the best opera-
tion should take primacy over the approach, and quality should never be compromised
for any reason. It is our belief that postoperative quality of life, including satisfaction with
the procedure, return to work, level of discomfort, and performance of daily activities,
are improved by these minimally invasive approaches. However, large-scale prospective
studies are needed to confirm our impressions. Only time and cumulative clinical volume
will be able to prove irrefutably the benefits of an operative strategy aimed at reducing
surgical trauma. However, it is clear that progress is being made daily. The concept of a
totally endoscopic mitral repair for complex mitral pathology seemingly is becoming a
reality.
REFERENCES
1. Cohn LH, et al. Minimally invasive cardiac valve surgery improves patient satisfaction while reducing
costs of cardiac valve replacement and repair. Ann Surg 1997;226(4):421–426; discussion 427–428.
2. Cosgrove DM, Sabik JF, Navia JL. Minimally invasive valve operations. Ann Thorac Surg
1998;65(6):1535–1538; discussion 1538–1539.
3. Navia JL, Cosgrove DM. Minimally invasive mitral valve operations. Ann Thorac Surg
1996.;62(5):1542–1544.
4. Arom KV, Emery RW. Minimally invasive mitral operations. Ann Thorac Surg 1997;63(4):1219–1220.
5. Gundry SR, et al. Facile minimally invasive cardiac surgery via ministernotomy. Ann Thorac Surg
1998;65(4):1100–1104.
6. Pompili MF, et al. Port-access mitral valve replacement in dogs. J Thorac Cardiovasc Surg
1996;112(5):1268–1274.
7. Felger JE, et al. Evolution of mitral valve surgery: toward a totally endoscopic approach. Ann Thorac
Surg 2001;72(4):1203–1208; discussion 1208–1209.
8. Chitwood WR Jr, et al. Video-assisted minimally invasive mitral valve surgery. J Thorac Cardiovasc
Surg 1997;114(5):773–780; discussion 780–782.
9. Gillinov AM, Banbury MK, Cosgrove DM, et al. Hemisternotomy approach for aortic and mitral valve
surgery. J Card Surg 2000;15(1):15–20.
10. Glower DD, et al. Predictors of outcome in a multicenter port-access valve registry. Ann Thorac Surg
2000;70(3):1054–1059.
11. Byrne JG, et al. Minimally invasive direct access heart valve surgery. J Card Surg 2000;15(1):21–34.
12. Cutler E, Beck C. Surgery of the heart and pericardium. In (Nelson, ed.) Loose Leaf Surgery, New York:
Thos. Nelson & Sons, 1927.
13. Sakakibara S, et al. Direct visual operation of aortic stenosis with a cardioscope: studies on cardioscope
no. 2. Bull Heart Inst Jap 1958;2:1–21.
14. Kaneko Y, et al. Video-assisted observation in mitral valve surgery. J Thorac Cardiovasc Surg
1996;111(1):279–280.
15. Carpentier A, et al. [Computer assisted open heart surgery. First case operated on with success]. C R
Acad Sci III 1998;321(5):437–442.
16. Reichenspurner H, et al. Three-dimensional video and robot-assisted port-access mitral valve operation.
Ann Thorac Surg 2000;69(4):1176–1181; discussion 1181–1182.
17. Falk V, et al. Robot-assisted minimally invasive solo mitral valve operation. J Thorac Cardiovasc Surg
1998;115(2):470–471.
18. Chitwood WR Jr, et al. Robotic mitral valve repair: trapezoidal resection and prosthetic annuloplasty
with the da vinci surgical system. J Thorac Cardiovasc Surg 2000;120(6):1171–1172.
19. Grossi EA, et al. Case report of robotic instrument-enhanced mitral valve surgery. J Thorac Cardiovasc
Surg 2000;120(6):1169–1171.
20. Vanermen H, et al. Video-assisted Port-Access mitral valve surgery: from debut to routine surgery, Will
Trocar-Port-Access cardiac surgery lead to robotic surgery? Semin Thorac Cardiovasc Surg
1999;11:223–234.
21. Nifong LW, Chu VF, Bailey BM, et al. Robotic mitral valve repair: experience with the da Vinci system.
Ann Thorac Surg 2003;75(2):438–442.
Chapter 20 / Port-Access Mitral Valve Surgery 273
20 Port-Access Mitral Valve Surgery
Ashish S. Shah, MD and Donald D. Glower, MD

CONTENTS
INTRODUCTION
PATIENT SELECTION
OPERATIVE TECHNIQUE
INCISIONS
FEMORAL ARTERY CANNULATION
CENTRAL AORTIC CANNULATION
AORTIC OCCLUSION/CARDIOPLEGIA
COMPLETION OF PROCEDURE
POSTOPERATIVE CARE
POTENTIAL PITFALLS
RESULTS
CONCLUSIONS
FUTURE
REFERENCES
INTRODUCTION
Port access, a term originally coined by the founders of Heartport, Inc. (Redwood City,
CA), was intended to describe a totally endoscopic approach to cardiac surgery using
cardiopulmonary bypass (1,2). While the founders of Heartport ultimately concluded in
the mid-1990s that totally endoscopic cardiac surgery was not technically feasible at the
time, the term port access was retained and applied to minimally invasive cardiac surgical
procedures using a small thoracotomy and any combination of catheters developed by
Heartport. The system consists of endoclamps, catheters for percutaneous retrograde
cardioplegia and pulmonary artery venting, femoral venous cannulation, direct aortic
cannulation, and femoral arterial cannulation.
Once animal studies at Stanford University and New York University demonstrated
the feasibility of applying port access to mitral valve operations (1,3,4), port-access

273
mitral valve procedures were performed at Stanford University and New York University
in 1996 as part of a U.S. Food and Drug Administration (FDA)-approved clinical trial
(5,6). Simultaneously, port-access mitral valve procedures were performed in Germany,
Singapore, and several other sites worldwide (7). In October 1996, the FDA approved
port-access devices for clinical use in the United States, and over 100 sites in the United
States and Europe began to apply port-access techniques to mitral valve surgery. Initial
users of port access underwent training on animals and proctoring (8). Subsequent users
adopted port access after visiting established sites and by proctoring. As of the end of the
year 2000, over 18,000 port-access valve and coronary procedures had been performed
worldwide.
Other approaches to the mitral valve using a small thoracotomy were developed in
parallel with port access (9–13). The best-known approach is the Chitwood mini-mitral
procedure (14–16), described by the author in Chapter 19. These approaches all include
a small right thoracotomy and an external aortic clamp, generally with femoral arterial
and venous cannulation. Other investigators favor a slightly larger right thoracotomy
using standard equipment (17–20).
Now that port access has been established clinically, port access continues to be
applied with a growing number of modifications to the original technique. Appropriately,
the port-access approach (which was originally intended for totally endoscopic robotic
procedures) is once again being applied to robotic mitral valve surgery (21–23). To date,
a small number of totally endoscopic robotic mitral valve procedures have been per-
formed in the United States and Europe using port-access techniques. The viability of the
port-access approach today (despite diverse economic and technological pressures)
attests to the flexibility and validity of the original concept behind port access.
PATIENT SELECTION
When given the choice, most patients will choose port access or a small thoracotomy
over standard median sternotomy. In the experience of one surgeon at Duke University,
252 of 268 (94%) of all isolated mitral valve patients were considered candidates for the
port-access approach and underwent mitral operation using port access.
Relative indications for port access include desire for improved cosmesis (especially
in young women), previous median sternotomy (especially with patent bypass grafts),
and any factors that increase risk of sternotomy complications such as obesity, chronic
steroid use, lung disease, and diabetes. Data supporting these relative indications for port
access are presented later in this chapter and are based primarily on the lack of sternal
wound complications and the limited mediastinal dissection associated with port access.
Relative contraindications include previous right thoracotomy, severe pectus
excavatum, and need for concurrent coronary or aortic valve operation. Of those patients
excluded from port access at Duke University, 13 of 16 (81%) were excluded because of
possibly needing concurrent aortic or coronary operation.
OPERATIVE TECHNIQUE
Patients are positioned supine with the right shoulder elevated and with the right arm
at the patient’s side (6,24). Alternatively, the right arm may be supported over the head
(25). External defibrillator pads should be placed prior to skin preparation to allow
defibrillation in the presence of severe pericardial adhesions. One defibrillator lead is
placed over the left anterolateral chest, and the other lead is placed under the right shoulder.
If central aortic cannulation is anticipated, the ability to deflate the right lung is gen-
erally necessary. These patients should be intubated with a dual-lumen endotracheal tube
or with an endobronchial blocker placed in the right mainstem bronchus. In those patients
in whom the right lung cannot be deflated for technical reasons or because of significant
lung disease, cannulation of the femoral artery and femoral vein will be necessary to
initiate cardiopulmonary bypass prior to dissecting the mediastinum. The combination of
lung disease precluding one-lung ventilation and the inability to cannulate the femoral
artery would be a contraindication to the port-access or thoracotomy approach.
Anesthetic management of port-access mitral patients differs from that of sternotomy
patients in several regards (26). First is the need to obtain single-lung ventilation if aortic
cannulation is planned. Second, if the endoclamp is placed from the femoral artery with
the possibility of endoclamp balloon migration, bilateral radial artery arterial lines should
be placed to detect occlusion of the innominate artery should the endoclamp migrate
distally. If central aortic cannulation is planned, only a single radial arterial line is needed.
Third, several other pressure transducers are needed for use of the endoclamp, and these
include lines for aortic root pressure, endoclamp balloon pressure, and coronary sinus
pressure. Finally, if desired, the anesthesiologist may place a percutaneous catheter (EP,
CardioVations, Sommerville, NJ) into the coronary sinus during line placement (27).
This technique has the advantage of providing retrograde cardioplegia without having an
additional catheter in the surgical field. The disadvantages to percutaneous coronary
sinus catheter placement include the learning curve in catheter placement, the variability
of coronary sinus anatomy, and the additional cost of the coronary sinus catheter. In
experienced hands, the coronary sinus catheter can be placed percutaneously 80% of the
time. Also available is a pulmonary artery vent catheter (CardioVations, Sommerville,
NJ), which can be inserted by the anesthesiologist into the pulmonary artery through a
venous sheath. In the future, anesthetic management of port-access patients may include
regional analgesia such as paravertebral blocks to assist in the postoperative course.
INCISIONS
Several different incisions have been used for mitral and tricuspid valve operations
using port access. The first incision is placed just lateral to the nipple over the fourth
intercostal space (above the nipple in men and in the inframammary crease in most
women) (Fig. 1). This anterolateral incision has the advantage of providing excellent
access for central aortic cannulation and also provides the most direct and most lateral
view of the mitral valve. The disadvantages of this incision are that it places the surgeon
at a greater distance from the mitral valve and that more lateral cutaneous nerves may be
injured, providing a somewhat wider area of medial numbness and paresthesia in some
patients. The second commonly used incision is placed in the inframammary crease
directly inferior to the nipple and just lateral to the mammary artery. This incision has the
advantages of placing the surgeon closer to the mitral valve and of causing less medial
paresthesia. Disadvantages include less access to the ascending aorta and a less favorable
angle to view the mitral valve and subvalvular apparatus. The more medial incision is
often used with femoral arterial cannulation and with greater use of the video camera to
visualize the mitral valve (video-directed surgery) (28) rather than direct vision through
the incision. Incisions can be made as small as 4 cm if the femoral artery is cannulated,
and cannulation of the ascending aorta generally requires a 6-cm incision. An additional
adjunct to provide exposure of the ascending aorta is to divide the fourth rib medially at
the costochondral junction and then to enter the bed of the fourth rib. The rib is repaired
Fig. 1. Four different incisions applicable to port access mitral procedures: (A) inframammary, (B)
infralateral, (C) supralateral, and (D) supramedial. The supramedial incision (D) is in the third
intercostal space, and all others are in the fourth intercostal space.
at the end of the procedure. A third approach to the mitral valve is to enter the right third
intercostal space just lateral to the sternum if combined mitral and aortic valve procedures
are planned (29).
The pericardium is opened vertically from the diaphragm to the ascending aorta,
keeping the incision at least 1–2 cm anterior to the phrenic nerve. If the ascending aorta
is to be cannulated, the incision should be carried up to the innominate vein. Three lateral
retraction sutures are placed on the posterior pericardial edge. The first is placed over the
right superior pulmonary vein and is secured to the lateral corner of the skin incision. The
second is placed halfway to the diaphragm and is passed through the chest wall using a
12-gage needle, a small hook, and a small clamp to secure the suture. The third suture is
Fig. 2. Venous cannula with serial dilators suitable for percutaneous access (above), and aortic
cannula with knife-tipped introducer (below). Note the side arm on the aortic cannula for insertion
of the endoclamp into the arterial line.
placed at the level of the superior vena cava and is passed through the third intercostal
space as laterally as possible. If the ascending aorta is to be cannulated, the medial
pericardium at the mid-ascending aorta is secured to the posterior sternum to provide
aortic exposure.
FEMORAL ARTERY CANNULATION

Femoral arterial cannulation is performed through a small 3- to 4-cm transverse inci-
sion in the groin between the inguinal crease and the inguinal ligament. The femoral
artery and femoral vein are exposed and encircled with tapes, although tape placement
is not necessary in a scarred and previously operated groin. Two concentric purse strings
are placed in the femoral vein and artery using 5-0 polypropylene suture secured using
tourniquets (30).The femoral venous cannula is generally placed first. After hepariniza-
tion, a guidewire is passed up the femoral vein into the superior vena cava using either
echocardiography, direct palpation, or direct vision. The 22 or 25 fr femoral venous
cannula (QD-22 or QD-25, CardioVations, Sommerville, NJ ) (Fig. 2) is then passed over
the wire and through the purse strings to place the tip of the cannula 2 cm into the superior
vena cava. The two 5-0 polypropylene suture tourniquets are then secured, and tapes on
the proximal and distal vein are released, allowing continuous venous drainage of the leg
and excellent hemostasis. Two concentric 5-0 polypropylene sutures are similarly placed
in the common femoral artery, being careful that the diameter of the purse strings is less
than one-half the diameter of the vessel to avoid vessel stenosis. After controlling the
artery proximally and distally, an arteriotomy is made within the purse strings and is
dilated to a diameter large enough to pass the arterial cannula. A 21 or 23 fr arterial
cannula (ER-23 or ER-21, CardioVations Sommerville, NJ) is then passed over a wire
into the femoral artery, placing the cannula tip at least 2 cm into the femoral artery and
away from any plaques or bends in the femoral artery. The 5-0 polypropylene tourniquets
are then secured, and all tapes on the femoral artery are released so that the leg will be
continuously perfused around the cannula. Both the arterial and venous cannulas can be
placed as above if no difficulties are apparent, otherwise fluoroscopy may be necessary.
Fluoroscopy is rarely needed for port access in experienced centers today.
Fig. 3. Three different options to obain aortic occlusion: (A) transfemoral endoclamp, (B) external
aortic clamp, and (C) transaortic endoclamp.
CENTRAL AORTIC CANNULATION

Central aortic cannulation has the advantages of eliminating endoclamp balloon mi-
gration, avoiding embolization and/or dissection due to aortic or peripheral vascular
disease, avoiding femoral arterial injury, limb ischemia, and groin incision complications
(seroma, infection, hematoma) (31). Central aortic cannulation has the disadvantage of
requiring a somewhat larger incision and requiring the additional hardware of a specifi-
cally designed aortic cannula. Once the thoracotomy incision is made, the chest wall
retractor is placed. A reusable-type retractor may be desirable here, as opposed to the
earlier soft tissue retractor, to provide greater exposure of the ascending aorta by lifting
upward on the superior chest wall. A 16-gage 6-in. needle is then passed through the first
intercostal space at the midclavicular level, angling toward the desired cannulation site
on ascending aorta to identify the optimal placement of the arterial line port. Care is
needed to avoid the right internal mammary artery and vein. The arterial cannula should
enter the aorta 1–2 cm proximal to the innominate artery with the aortic cannula directed
toward the aortic valve. Angulation toward the aortic valve ensures that the endoclamp
will pass toward the aortic valve and not down the aortic arch. Once this approach line
is demonstrated with the 16-gage needle, a 2.5-cm transverse skin incision is made at
the 16-gage needle puncture site, and an 11.5-mm trocar is passed through the former
16-gage needle track to provide a direct port for the aortic cannula. This port should point
directly at the desired aortic cannulation site. The aortic cannula (DF-K-24, CardioVations,
Sommerville, NJ) (Fig. 3A–C) is then passed through the 11.5-mm port to ascertain that
direct access to the desired cannulation site is possible without torque on the cannula (Fig. 4).
Fig. 4. Surgeon’s view of the aortic cannula placed through a port in the first intercostal space.
Tourniquets on the aortic purse-string sutures pass through the trocar for stability.
Attention is then turned to the femoral vein, which can be accessed percutaneously to
avoid a groin incision. After heparinization, the common femoral vein is cannulated as
described previously. Once the femoral vein is cannulated, a standard aortic purse string
is placed at the desired cannulation site on the ascending aorta using two concentric
2-0 polyester sutures. The aortic cannula (DF-K-24, CardioVations, Sommerville, NJ)
(Fig. 3) is passed through the 11.5-mm trocar in the first intercostal space so that the
cannula introducer is centered in the aortic purse-string suture. The knife tip on the aortic
cannula introducer is deployed only long enough to allow the introducer to reach the
aortic wall. The introducer knife tip is then immediately withdrawn (to avoid injury to the
back wall of the aorta), and the introducer and cannula are easily passed with a twisting
motion into the ascending aorta. The 2-0 purse-string sutures are then passed through the
11.5-mm trocar and secured using plastic tourniquets placed through the trocar and
secured with a heavy suture. Excellent aortic cannula stability is obtained (Fig. 4). The
placement of the venous cannula prior to aortic cannulation provides an additional source
of vascular access should any bleeding be encountered. If an endoclamp is not to be used,
then alternative aortic cannulas (SS-23-A-K or SS-23-S-K, CardioVations, Sommerville,
NJ) can be used that do not have a side arm for an endoclamp but that still have the knife-
tipped introducer.
Additional alternatives in venous cannulation include placement of a second venous
catheter into the superior vena cava via the right internal jugular vein (25,28). This is
generally a 17 fr arterial-type cannula placed over a wire. Otherwise, a 28 fr angled plastic
Fig. 5. A flexible Cosgrove external aortic clamp that may be used as an alternative to an external
Chitwood clamp or an endoclamp.
venous cannula can be placed through the chest wall or through the thoracotomy incision
to provide standard cannulation of the superior vena cava. In this instance, a standard
purse-string suture would be placed in the right atrium or in the superior vena cava after
withdrawing the femoral venous cannula into the body of the right atrium. In those
patients in whom femoral venous cannulation is not possible or desirable, the inferior
vena cava can be cannulated through a right atrial purse-string suture with the cannula
passed through the chest wall or through the thoracotomy incision. It is important that
venous cannulation provide adequate drainage of the superior vena cava when the left
atrium is retracted anteriorly, so that obstruction of the superior vena cava does not result.
Assisted venous return should generally be used either via vacuum assist or by use of a
Biomedicus centrifugal pump in the venous line.
AORTIC OCCLUSION/CARDIOPLEGIA
Several alternatives exist for aortic occlusion. First, ventricular fibrillation may be
used when the ascending aorta is not easily occluded. This may be desirable when open
bypass grafts (especially left internal mammary grafts) are present and are not easily or
safely accessible. Second, the ascending aorta can be occluded with an external clamp of
either the Cosgrove (Fig. 5) or Chitwood (14–16) variety (Figs. 3 and 6). These aortic
clamps can be passed through the incision or through a separate stab wound lateral to the
incision. If an external clamp is used, anterograde cardioplegia may be delivered through
a standard cardioplegia cannula secured with purse-string sutures in the ascending aorta.
Alternatively, a single dose of anterograde cardioplegia can be given using a spinal
needle passed through the chest wall into the ascending aorta.
Fig. 6. External clamping (A) of the ascending aorta. Note the cannula (B) in the aortic root for
cardioplegia/venting. The ascending aorta may be cannulated (C) for arterial access as seen here,
or the femoral artery may be used.
Finally, the aorta can be occluded using an endoclamp passed through the aortic or
femoral arterial cannula (Fig. 3). Once the arterial cannula is placed, an endoclamp
(EC-65 or EC-100, CardioVations, Sommerville, NJ) (Fig. 7) is passed through the
“Y”-limb in the arterial cannula to position the end of clamp in the ascending aorta just
above the aortic valve. Transesophageal echo or fluoroscopy can be used. If a femoral
arterial cannula is employed, the endoclamp is passed over a wire. For direct cannulation
of the ascending aorta, a wire is unnecessary. Once in the endoclamp is positioned,
cardiopulmonary bypass is initiated using assisted venous drainage. With the transaortic
endoclamp, the balloon is partially inflated, pulled back snugly against the tip of the aortic
cannula, and then inflated to a volume sufficient to occlude the ascending aorta as dem-
onstrated by transesophageal echo and/or direct palpation and inspection. With femoral
cannulation, the endoclamp is inflated using transesophageal echo or fluoroscopy to
ascertain positioning at least 1 cm above the aortic valve but proximal to the innominate
artery. Once in place, slack on the femoral endoclamp catheter is tightened enough to
prevent proximal balloon migration, without being so tight as to cause distal endoclamp
balloon migration.
Similarly, cardioplegia delivery can be handled in several ways. Ventricular fibrilla-
tion is an option when aortic occlusion cannot be obtained or when anterograde and
retrograde cardioplegia are not feasible. Anterograde cardioplegia can be delivered
through the endoclamp catheter or through a needle in the aortic root if an external clamp
is used. Finally, retrograde cardioplegia can be delivered through a percutaneous coro-
nary sinus catheter or through a retrograde coronary sinus catheter placed directly through
a purse string in the right atrium.
Fig. 7. Transaortic balloon endoclamp (65 cm). The transfemoral endoclamp is similar but longer
(100 cm). Note the three catheter lumens for cardioplegia/venting (proximal Y connector), balloon
inflation (dark stopcock), and monitoring of aortic root pressure (white stopcock).
Fig. 8. Opening the left atrium (LA) through the interatrial groove. Suckers are in the left atrium.
The right atrium (RA) is retracted toward the aorta (Ao) with forceps. The pericardium (Peric) and
phrenic nerve are below.
COMPLETION OF PROCEDURE
With the heart arrested on cardiopulmonary bypass, the left atrium is opened adjacent
to the interatrial groove (Fig. 8). If needed, the left atriotomy can be extended superiorly
behind the superior vena cava and inferiorly behind the inferior vena cava. The view of
Fig. 9. Operative view of the mitral valve with annular sutures being placed with endoscopic
instruments. The view of the anterior and posterior mitral leaflets (AML and PML) and subvalvular
apparatus is excellent.
Fig. 10. Typical endoscopic instruments (forceps above and scissors below) used to reach the
mitral valve.
the left atrium and mitral valve is generally excellent and sufficient to perform complex
repairs of the mitral valve and subvalvular apparatus. The mitral procedure itself is
performed in a manner identical to that with sternotomy, with the exception that endo-
scopic instruments are required due to the distance to the mitral valve and due to the
limited incision (Fig. 9). These endoscopic instruments include endoscopic forceps,
endoscopic needle drivers, endoscopic scissors, and endoscopic knot pushers (Fig. 10).
Upon completion of the mitral procedure, the left atrium is closed in a standard fashion
with a left ventricular vent passed through the left atrial incision, through the mitral valve,
and into the left ventricle. Several important adjuncts are used to air the heart. First, the
entire thoracotomy field may be flooded with carbon dioxide throughout the case (32).
Second, with the left atrium closed around the left ventricular vent by a tourniquet,
suction is applied to the left ventricular vent with the aorta is still clamped. Ideally, some
of the vent cannula holes should be in the left ventricle, and some holes should also be
in the left atrium. Third, the patient is turned side to side while the perfusionist fills
the heart with blood. Once the heart appears to be adequately de-aired on transesophageal
echo, the patient is placed in Trendelenberg position, suction is applied to the aortic root
vent, and the aorta is unclamped. The patient is maintained in Trendelenberg position for
2 min after a weaning from cardiopulmonary bypass. Any residual air on echo can be
aspirated through the aortic root vent or cleared down the right coronary artery. Once the
heart is well de-aired on echo, the ventricular vent is removed, closure of the left atriotomy
is completed, and the patient is weaned from cardiopulmonary bypass.
Aortic decannulation is performed by tying the dual aortic purse-string sutures directly
through the thoracotomy incision. The percutaneous venous cannula is simply with-
drawn and the subcutaneous tissue closed with absorbable suture. If a groin incision is
used, the arterial and venous purse strings are tied, and the groin incision is closed in a
standard fashion. The surgeon should have a low threshold to repair or patch any small,
diseased, or narrowed femoral artery. A single 19 fr rubber drain may be placed in the
pericardium and brought through the chest wall. The drain is placed so that both the
pericardial space and the pleural space are drained postoperatively. A single 36 fr chest
tube is left in the pleural space for the first 12–24 h. The fourth rib is repaired with a #4
sternal wire if the rib was divided. The skin incision and any other port sites are closed with
absorbable sutures.
POSTOPERATIVE CARE
Postoperative management differs little from that of standard sternotomy patients.
Because bleeding is generally minimal, the pleural drain is usually removed within 12 h
and the 19 fr drain is removed after 3–5 d. Postoperative pain appears to be similar to
sternotomy for the first 1–3 d, and thereafter appears to be less than with sternotomy (33).
Patients appear to have greater arm use and earlier mobility than with sternotomy (24).
Thoracotomy pain in the first 48 h responds well to anti-inflammatory agents, but at the
risk of renal complications. All patients are given an intravenous bolus of 750 mg of
procainamide while on cardiopulmonary bypass and are maintained on a procainamide
(or quinidine) drip until oral antiarrhythmic agents can be started on the second postop-
erative day. If patients are discharged in sinus rhythm, the antiarrhythmic therapy is
continued for 2–3 wk or longer, depending on the likelihood of postoperative atrial
fibrillation. Anticoagulation is used only for those patients at risk for atrial fibrillation
after discharge or with mechanical prostheses. All patients are discharged on aspirin.
Median hospital stay in our experience is 5 d. Patients are advised to return to driving in
2 wk and may return to working in 3–6 wk depending on their overall condition.
POTENTIAL PITFALLS
As mentioned previously, body habitus can preclude the port-access approach or can
require a modified approach. Body habitus can preclude either aortic or femoral cannu-
lation, requiring the alternative technique. In patients with lung disease who cannot
tolerate single-lung ventilation, or in patients in whom one-lung ventilation cannot be
obtained, femoral arterial cannulation would be the next alternative. Patients with disease
of the femoral artery, iliac artery, or abdominal or thoracic aorta may be better candidates
for central aortic cannulation.
The endoclamp may have difficulty occluding an ascending aorta of 3.5 cm or larger
diameter. In these patients, ventricular fibrillation or an external aortic clamp should be
used. If the femoral venous cannula cannot be placed into the superior vena cava on
cardiopulmonary bypass, a separate cannula should be placed in the superior vena cava
to guarantee drainage of the superior vena cava and right atrium while retracting on the
left atrium. In patients with preexisting pericardial adhesions, placement of temporary
ventricular pacing wires may be difficult or impossible. On rare occasions, a small left
thoracotomy can be performed in the left fourth intercostal space to place permanent
epicardial pacing leads on the left ventricle.
A number of techniques are available to handle intraoperative difficulties. Coronary
sinus perforation can generally be repaired while on cardiopulmonary bypass by expos-
ing the coronary sinus posterior to the inferior vena cava through the right thoracotomy
incision. In the event of more serious cardiac injury or problems inaccessible from a small
right thoracotomy, the thoracotomy incision can be extended medially and laterally into
a full right anterolateral thoracotomy. The incision can also be extended across the
midline into a transverse sternotomy, or the thoracotomy can be abandoned and a stan-
dard median sternotomy performed. Bleeding most commonly comes from the right
chest wall, so all accessible wounds and puncture sites should be examined closely for
hemostasis prior to closing the chest. Ascending aortas 4 cm or more in diameter ideally
should not be cannulated through the thoracotomy due to the difficulty repairing any
injury to a dilated and abnormal aorta.
RESULTS
Early feasibility studies generated initial enthusiasm for port-access techniques. By
the year 2001, over 18,000 port-access procedures had been performed worldwide. There
has since been a progression to evaluate this technology critically, comparing port access
to conventional surgery with respect to patient outcomes. As of 2001, a small number of
totally endoscopic mitral valve operations have been performed using robotics to perform
the mitral valve procedure, while the port-access platform provides femoral vein-to-
femoral artery cardiopulmonary bypass and cardioplegic arrest with an endoclamp.
Intraoperative Factors
Recent reviews of port-access mitral valve surgery at single institutions and the Port
Access International Registry (PAIR) have looked at intraoperative and patients out-
comes following port-access mitral surgery, with excellent results (Table 1). Both mitral
valve repair and replacement have been performed with cardiopulmonary bypass times
that range from 135 to 212 min and crossclamp times from 95 to 133 min. Despite the age
of this technology, conversion to a median sternotomy has had reported rates of 0–9.5%.
Hospital Outcomes
Similarly, in-hospital results following port-access mitral surgery have been outstand-
ing in light of the age of this technology. Thirty-day mortality ranged from 1.1 to 5.0%
with ICU stays of 1–1.8 d and mean hospital length of stay between 5 and 11 d (Table 1).
Chapter 20 / Port-Access Mitral Valve Surgery
Table 1
Outcomes of Mitral Operations in Largest Clinical Series Using Port-Access Approach
No. of Mitral Valve CPBa Crossclamp Conversion Hospital Atrial

Author, Year (Ref.) b
Patients Mean Age Repair Replacement Time (min) Time (min) Rate (%) Mortality ICU LOS c LOS Fibril. (%) Stroke (%)
Glower, 1998 (24) 21 60 13 8 212 133 5 0 NAd 6 28 0

Mohr, 1998 (7) 51 58 28 23 133 72 12 9.8 2 13 NA NA
Gulielmos, 2000 (34,35) 33 64 9 24 NA 108 10 3 1 10 NA NA
Glower, 2000 (36) 1059 58 491 568 127 92 4 1.6–5.5 0.9 5 10 2.6
Chaney, 2000 (37) 19 57 NA NA 145 105 NA NA 1.3 6.1 11 NA
Reichenspurner, 2000 (39,40) 50 61 26 24 125 83 NA 0 1.5 9 NA NA
Vanerman, 2000 (28) 121 60 75 46 140 101 4 2 2.1 8.7 NA 1
Grossi, 2001 (38) 100 56 100 NA 135 NA 0 0 NA NA NA 2
aCardiopulmonary bypass time.
bConversion to full sternotomy.
cIntensive care unit length of stay.
dNot available/reported.
287
288 Table Part
2 III / Minimally Invasive Valvular Surgery
Institutional Experience with Port-Access Mitral Valve Surgery
Mitral Valve
Repair Replacement Total
No. of patients 191 111 302
Age (yr) 58 ± 14 58 ± 14 58 ± 15
Intraoperative date
Crossclamp time (min) 104 ± 57 90 ± 65 99 ± 60
Cardiopulmonary bypass time (min) 181 ± 49 194 ± 54 186 ± 51
Surgery time (min) 332 ± 62 353 ± 75 339 ± 67
Postoperative data
Transfusion (units) 2±4 2±2 2±3
Chest tube output (mL) 422 ± 470 330 ± 274 389 ± 413
Intubation time (h) 8±5 11 ± 12 9±8
Length of stay (d) 7 ± 10 7±4 7±8
Adverse events
Aortic dissection 0 (%) 0 (0%) 0 (0%)
Wound infection 0 (0%) 1 (1%) 1 (0.3%)
New pacemaker 2 (1%) 3 (3%) 5 (2%)
Stroke 1 (1%) 3 (3%) 4 (1%)
Reop for bleeding 10 (5%) 1 (1%) 11 (4%)
Conversion to sternotomy 3 (2%) 2 (2%0 5 (2%)
Reop on mitral valve 4 (2%) 1 (1%) 5 (2%)
Mortality 1 (1%) 0 (0%) 1 (0.3%)
In the PAIR study, the incidence of stroke was 2.6% and 2.8% for mitral valve repair and
replacement, respectively. The incidence of renal failure and new atrial fibrillation was
2.1% and 10.4% for mitral repair and 3.3% and 10.0% for mitral replacement (36). Early
reports of aortic dissection have dramatically decreased with the use of direct aortic
cannulation (7,36,41). A multivariable analysis of the PAIR data showed that the major
predictors of death were reoperation, older age, and mitral valve replacement (36).
Importantly, low case volume did not appear to be significantly associated with adverse
outcomes. Using port access, complex repairs of the anterior mitral chordeae yields initial
and early (2 yr) success rates equivalent to those after simple annuloplasty using port
access (42,43). Although difficult to quantify, cosmesis is generally superior with a small
thoracotomy than with standard sternotomy, and cosmesis may be more relevant to the
patient than to the physician.
Blood Transfusion
Transfusion practices vary among institutions, but several reports have documented
median blood transfusion of 1–3 units (Table 2) (24,38). When compared to conventional
mitral valve repair or replacement, two case-controlled series demonstrated significantly
less blood use in port access patients (24,38). Trends have been for less chest tube output
with port access relative to sternotomy (24).
Comparison to Other Approaches

Port-access mitral valve surgery has been compared to conventional approaches in
several case series. Postoperative pain scores were found to be significantly less between
postoperative d 3 and 7 with port access relative to sternotomy (33). Hospital stay has
been similar or slightly decreased with port access (24,38), with fewer wound or septic
complications in all patients (38) and in elderly patients (44). Port-access patients have
returned to normal activity four weeks earlier than did sternotomy patients (24). Alter-
natively, cardiopulmonary bypass, cross-clamp, and operative times are universally
longer with port access as compared to conventional mitral surgery (24,38,45) (Table 1).
The frequency of cerebral microemboli has not been different between techniques (46).
Arom et al. (47,48) compared port access to partial sternotomy in 65 patients having
aortic or mitral valve operation. Port-access patients had longer clamp time and longer
surgery time. However, port-access patients had less atrial fibrillation and returned to
work 4 wk earlier. In a comparison of port access to the Chitwood mini-mitral procedure,
Aybek et al. (49) found that port access had longer clamp time, longer pump time, longer
procedure time, and more blood loss.
Learning Curves
As described above, operative times are consistently longer for port-access proce-
dures, and as with any new technology there is significant interest in learning curves and
the effect of institutional volumes on outcomes. A recent study from PAIR data examined
the determinants of operative times looking at institutional experience, case volumes, and
procedure type (36). Interestingly, institutional case volume was not a determinant of
operative time. Instead, procedure type (mitral valve repairs particularly) was an impor-
tant factor. A multivariable analysis of registry data also showed that institutional volume
was not significantly associated with death, stroke, or reoperation for bleeding (36,50).
Interestingly, time spent off-pump accounted for significant variation among institu-
tions. Furthermore, operative times continued to decrease beyond 100 cases, suggesting
that learning in port-access surgery is a continuing process (50).
CONCLUSIONS
Port-access approaches to the mitral valve have been shown to be safe and effective.
Relative to sternotomy, port access may have the advantages of better cosmesis, earlier
postoperative patient mobilization, less sepsis and wound complications, and possibly
less blood loss and use. Port access may be especially advantageous in patients (1) with
prior sternotomy, (2) at risk for sternal complications, (3) with impaired mobility, and (4)
placing a high value on cosmesis. These advantages are at the expense of longer operating
time, longer learning curve, and greater equipment costs.
Port access is applicable in over 90% of isolated mitral procedures, and even complex
mitral repairs can be performed with excellent short-term outcome. The driving forces at
institutions applying port access to the mitral valve include patient biases toward mini-
mally invasive approaches, a desire to improve patient outcome and further technology,
and perhaps a desire to obtain a marketing advantage. Ultimately, broader application of
port-access technology may be limited. There are certainly specific applications for port-
access techniques, particularly in reoperative situations where a right chest approach is
easier. Furthermore, port-access techniques have been considered a transition to total
endoscopic or robotic repairs of the mitral valve. Nonetheless, port-access approaches to
mitral reconstruction offer important alternatives to conventional techniques and are
associated with excellent outcomes.
FUTURE
In a few short years, port-access technology has proven to be a safe and effective
method to operate on the mitral valve. Benchmark comparisons to conventional surgery
have shown similar efficacy and modest improvements in patient recovery and satisfac-
tion. Importantly, as experience grows, specific indications for port access (such as
reoperative surgery) will be better refined. Nonetheless, the future of port-access mitral
surgery is unclear. Many continue to advocate that port access is simply a stepping stone
to a totally endoscopic repair of the mitral valve. The goal of totally endoscopic mitral
surgery is becoming possible using port access, but only with robotic assistance
(22,28,51–54). At several institutions, port access has been used to perform mitral opera-
tions through a small thoracotomy using robotic assistance to manipulate the camera or
surgical instruments (39–41,51–53). Robotics in turn could greatly facilitate training of
younger surgeons in minimally invasive techniques. Robotics can for the first time allow
remote surgery with the operating surgeon and the patient in two different locations (55).
However, current robotic procedures have disadvantages of expensive and bulky instru-
mentation, longer operating times, need for a groin incision, and remaining technical
issues such as limited facility with knot tying.
The ultimate market share enjoyed by port access for isolated mitral valve surgery will
be a balance between patient demand for the procedure vs limited time, limited resources,
and limited experience of providing physicians and hospitals. Efforts to simplify the port-
access platform will continue over the next few years as robotic technology evolves. As
instrumentation and imaging improve, port-access approaches to the mitral valve may
have broader appeal. It is clear that subsets of patients can benefit from the port-access
approach to mitral operations, and continued development of techniques, equipment, and
education will help define these cohorts.
REFERENCES
1. Pompili MF, Stevens JH, Burdon TA, et al. Port-access mitral valve replacement in dogs. J Thorac
2. Pompili MF, Yakub A, Siegel LC, Stevens JH, Awang Y, Burdon TA. Port-access mitral valve replace-
ment: initial clinical experience [abstr]. Circulation 1996;94:I-533.
3. Schwartz DS, Ribakove GH, Grossi EA, et al. Minimally invasive mitral valve replacement: port-access
technique, feasibility, and myocardial functional preservation. J Thorac Cardiovasc Surg
1997;113:1022–1031.
4. Schwartz DS, Ribakove GH, Grossi EA, et al. Minimally invasive cardiopulmonary bypass with
cardioplegic arrest: a closed chest technique with equivalent myocardial protection. J Thorac Cardiovasc
Surg 1996;111:556–566.
5. Galloway AC, Ribakove GH, Miller JS, et al. Minimally invasive port-access valvular surgery: initial
clinical experience [abstr]. Circulation 1997; I-508.
6. Fann JI, Pompili MF, Burdon TA, Stevens JH, St. Ghoar FG, Reitz BA. Minimally invasive mitral valve
surgery. Semin Thorac Cardiovasc Surg 1997;9:320–330.
7. Mohr FW, Falk V, Diegeler A, Walther T, van Son JAM, Autschbach R. Minimally invasive port-access
mitral valve surgery. J Thorac Cardiovasc Surg 1998;115:567–576.
8. Galloway AC, Shemin RJ, Glower DD, et al. First report of the Port Access International Registry. Ann
Thorac Surg 1999;67:51–58.
9. Carpentier A, Loulmet D, Carpentier A, LeBret E, Haugades B, Petal D. Chirurgie a coeur ouvert par
video-chirurgie et minithoracotomie—premier cas (valvuloplastie mitrale) opere avec success.
L’Academie des Sciences: Sciences de la vie 1996;319:219–223.
10. Loulmet DF, Carpentier A, Cho PW, et al. Less invasive techniques for mitral valve surgery. J Thorac
11. Zapolanski A, Korver K, Pliam MB, Mengarelli L. Mitral valve surgery via a right anterior mini-
thoracotomy with central aortic cannulation and no endoscopic assistance [abstr]. Heart Surg Forum
2001;4:S80.
12. Mishra YK, Malhotra R, Mehta Y, Sharma KK, Kasliwal RR, Trehan N. Minimally invasive mitral
valve surgery through right anterolateral minithoracotomy. Ann Thorac Surg 1999;68:1520–1524.
13. El-Fiky MM, El-Sayegh T, El-Beishry AS, et al. Limited right anterolateral thoracotomy for mitral valve
surgery. Eur J Cardiothorac Surg 2000;17:710–713.
14. Chitwood WR Jr, Elbeery JR, Chapman WHH, et al. Video-assisted minimally invasive mitral valve
surgery: the “micro-mitral” operation. J Thorac Cardiovasc Surg 1997;113:413–414.
15. Chitwood WR Jr, Elbeery JR, Moran JF. Minimally invasive mitral valve repair using transthoracic
aortic occlusion. Ann Thorac Surg 1997;63:1477–1479.
16. Chitwood WR Jr, Wixon CL, Elbeery JR, Moran JF, Chapman WHH, Lust RM. Video-assisted mini-
mally invasive mitral valve surgery. J Thorac Cardiovasc Surg 1997;114:773–782.
17. Holman WL, Goldberg SP, Early LJ, et al. Right thoracotomy for mitral reoperation: analysis of tech-
nique and outcome. Ann Thorac Surg 2000;70:1970–1973.
18. Braxton JH, Higgins RS, Schwann TA, et al. Reoperative mitral valve surgery via right thoracotomy:
decreased blood loss and improved hemodynamics. J Heart Valve Dis 1996;5:169–173.
19. Cohn LH. Right thoracotomy, femorofemoral bypass, and deep hypothermia for re-replacement of the
mitral valve. Ann Thorac Surg 1997;64:578–579.
20. Srivastava AK, Garg SK, Ganjoo AK. Approach for primary mitral valve surgery: right anterolateral
thoracotomy or median sternotomy. J Heart Valve Dis 1998;7:370–375.
21. Reichenspurner H, Boehm D, Reichart B. Minimally invasive mitral valve surgery using three-dimen-
sional video and robotic assistance. Semin Thorac Cardiovasc Surg 1999;11:235–243.
22. Chitwood WR Jr. Video-assisted and robotic mitral valve surgery: toward an endoscopic surgery. Semin
Thor Cardiovasc Surg 1999;11:191–205.
23. Vanerman H, Wellens F, DeGeest R, Degrieck I, VanPraet F. Video-assisted port-access mitral valve
surgery: from debut to routine surgery. Will trocar-port- access cardiac surgery ultimately lead to robotic
cardiac surgery? Semin Thorac Cardiovasc Surg 1999;11:223–234.
24. Glower DD, Landolfo KP, Clements F, et al. Mitral valve operation via Port Access versus median
sternotomy. Eur J Cardiothorac Surg 1998;14:S143–S147.
25. Chitwood WR Jr. Video-assisted mitral valve surgery: using the Chitwood Clamp. Op Tech Thorac
26. Peters WS, Burdon TA, Pompili MF. Port-access cardiac surgery: a system analysis. Perfusion
1998;13:253–258.
27. Grossi EA, Ribakove G, Galloway AC, Colvin SB. Minimally invasive mitral valve surgery with
endovascular balloon technique. Op Tech Cardiovasc Surg 2000;5:176–189.
28. Vanerman H, Farhat F, Wellens F, et al. Minimally invasive video-assisted mitral valve surgery: from
port-access towards a totally endoscopic procedure. J Card Surg 2000;15:51–60.
29. Colvin SB, Grossi EA, Ribakove G, Galloway AC. Minimally invasive aortic and mitral valve operation.
Op Tech Cardiovasc Surg 2000;5:212–220.
30. Chitwood WR Jr, Nifong LW. Minimally invasive videoscopic mitral valve surgery: the current role of
surgical robotics. J Card Surg 2000;15:61–75.
31. Glower DD, Komtebedde J, Clements FM, Debruijn NP, Stafford-Smith M, Newman MF. Direct aortic
cannulation for Port-Access mitral or coronary bypass operations. Ann Thorac Surg 1999;68:1878–1880.
32. Gillinov AM, Cosgrove DM. Minimally invasive mitral valve surgery: mini- sternotomy with extended
transseptal approach. Semin Thorac Cardiovasc Surg 1999;11:206–211.
33. Walther T, Falk V, Metz S, et al. Pain and quality of life after minimally invasive versus conventional
cardiac surgery. Ann Thorac Surg 1999;67:1643–1647.
34. Gulielmos V, Tugtekin SM, Kappert U, et al. Three-year follow-up after port access mitral valve surgery.
J Card Surg 2000;15:43–50.
35. Gulielmos V, Wunderlich J, Dangel M, et al. Minimally invasive mitral valve surgery—clinical expe-
riences with a Port Access system. Eur J Cardiothorac Surg 1998;14:S148–S153.
36. Glower DD, Siegel LC, Frischmeyer KJ, et al. Predictors of outcome in a multicenter Port-Access valve
registry. Ann Thorac Surg 2000;70:1054–1059.
37. Chaney MA, Durazo-Arvizu RA, Fluder EM, et al. Port-access minimally invasive cardiac surgery
increases surgical complexity, increases operating room time, and facilitates early postoperative hospi-
tal discharge. Anesthesiology 2000;92:1637–1645.
38. Grossi EA, Galloway AC, Ribakove GH, et al. Impact of minimally invasive valvular heart surgery: a
case-control study. Ann Thorac Surg 2001;71:807–810.
39. Reichenspurner H, Boehm DH, Gulbins H, et al. Three-dimensional video and robot-assisted port-
access mitral valve operation. Ann Thorac Surg 2000;69:1181–1182.
40. Reichenspurner H, Weltz A, GulielmosV, Boehm DH, Reichart B. Port-Access cardiac surgery using
endovascular cardiolpumonary bypass: theory, practice, and results. J Card Surg 1999;14:275–280.
41. Mohr FW, Onnasch JF, Falk V, et al. The evolution of minimally invasive mitral valve surgery - 2 year
experience. Eur J Cardiovasc Surg 1999;17:233–239.
42. Glower DD, Siegel LC, Galloway AC, et al. Predictors of operative time in multicenter Port-Access
valve registry: institutional differences in learning. Heart Surg Forum 2001;4:40–46.
43. Grossi EA, LaPietra A, Ribakove GH, et al. Minimally invasive versus sternotomy approaches for mitral
reconstruction: comparison of intermediate-term results. J Thorac Cardiovasc Surg 2001;1221:708–713.
44. Grossi EA, Galloway AC, Ribakove GH, et al. Minimally invasive port access surgery reduces operative
morbidity for valve replacement in the elderly. Heart Surg Forum 1999;2:212–215.
45. Ferdinand FD, Trace C, Priest BP, Sutter FP, Kowey PR. Minimally invasive mitral valve surgery: less
may be more [abstr]. Heart Surg Forum 2001;4:S81.
46. Schneider F, Onnasch JF, Falk V, Walther T, Autschbach R, Mohr FW. Cerebral microemboli during
minimally invasive and conventional mitral valve operations. Ann Thorac Surg 2000;70:1094–1097.
47. Arom KV, Emery RW, Kshettry VR, Dubois KA. Evaluation of two new heart valve surgery techniques:
partial sternotomy and port-access approaches. Eur J Cardiothorac Surg 1999;16:S99–S102.
48. Arom KV, Emery RW, Kshettry VR, Janey PA. Comparison between port-access and less invasive valve
surgery. Ann Thorac Surg 2000;68:1525–1528.
49. Aybek T, Dogan S, Wimmer-Greinecker G, Westphal K, Moritz A. The micro- mitral operation compar-
ing the Port-Access technique and the transthoracic clamp technique. J Card Surg 2000;15:76–81.
50. Glower D, Landolfo K, Kypson A, Bashore T, Harrison K, Wang A. Is port- access applicable to complex
mitral valve repair? [abstr]. Heart Surg Forum 2001;4:S73.
51. Onnasch JF, Falk V, Schneider F, Diegler A, Autschbach R, Mohr FW. Minimally invasive mitral valve
surgery—experience in 361 patients [abstr]. Heart Surg Forum 2001;4:S81.
52. Chitwood WR Jr, Nifong LW, Elbeery JE, et al. Robotic mitral valve repair: trapezoidal resection and
prosthetic annuloplasty with the da Vinci surgical system. J Thorac Cardiovasc Surg 2000;120:1171–1172.
53. Falk V, Autschbach R, Krakor R, et al. Computer-enhanced mitral valve surgery: toward a total endo-
scopic procedure. Semin Thorac Cardiovasc Surg 1999;11:244–249.
54. Carpentier A, Loulmet D, Aupecle B, Berrebi A, Relland J. Computer-assisted cardiac surgery. Lancet
1999;353:379–380.
55. Autschbach R, Onnasch JF, Falk V, et al. The Liepzig experience with robotic valve surgery. J Card Surg
2000;15:82–87.
Chapter 21 / Minimally Invasive Aortic Valve Surgery 293
21 Minimally Invasive Aortic Valve Surgery
Jerome Sepic, MD and Lawrence H. Cohn, MD

CONTENTS
INTRODUCTION
OPERATIVE TECHNIQUES
CONCLUSION
REFERENCES
INTRODUCTION
Corrective aortic valve surgery, both replacement and repair, has been a significant
advancement in modern medical science. These operations have given countless patients
with aortic stenosis and aortic regurgitation longer and more productive lives. Starting
with the aortic prosthetic valves developed by Harken and Starr and extending to the
current era of mechanical, bioprosthetic, and biological valves (including homografts
and autografts), aortic valve replacement surgery has evolved considerably and has proven
to be efficacious (1,2). In addition, numerous reparative techniques such as commissuro-
tomy, debridement, and prolapse reduction have been developed to, when possible,
preserve the native aortic valve.
In the last 30 years, aortic valve surgery typically has been performed using a full
median sternotomy with cardiopulmonary bypass (CPB), with direct aortic and right
atrial or bicaval cannulation. Outcomes have improved dramatically with the use of
varying degrees of systemic hypothermia and myocardial protection with antegrade
cardioplegia and, more recently, retrograde cardioplegia via coronary sinus catheters.
Since the first report by Cosgrove in 1996, minimally invasive aortic valve surgery has
continued to be an evolving concept (3). Its goals include reducing incision size, decreas-
ing surgical trauma and pain, and improving cosmesis, patient satisfaction, and recovery
times. However, the most important goal of minimally invasive aortic valve surgery must
be to maintain or improve the efficacy and safety of conventional aortic valve surgery.
Unlike minimally invasive coronary artery bypass grafting, which includes off-pump

293
coronary artery bypass (OPCAB), all aortic valve surgery to date requires the use of CPB.
Therefore, minimally invasive aortic valve replacement and repair (MIAVR) technique
developments have centered on different operative exposures and cannulation strategies.
Although numerous different MIAVR techniques have been proposed, characteristics
of some techniques make them more feasible and more easily adopted then others.
Ideally, low-morbidity techniques that have a relatively quick learning curve, use con-
ventional instrumentation or economical variations, and provide versatile exposure and
cannulation combinations are preferable. Additionally, these techniques should be appli-
cable to a wide variety of first-time and reoperative patient candidates and should allow
for all types of aortic valve replacements and repairs, as well as possible aortic root
procedures when needed. If the necessity arises, conversion to full sternotomy exposure
should be practical.
In this chapter, we will review the different operative techniques for MIAVR. Empha-
sis will be placed on detailed descriptions of varying surgical approaches and instrumen-
tation. In addition, we will report our 365-patient experience with MIAVR at the Brigham
and Women’s Hospital from July 1996 to July 2001.
OPERATIVE TECHNIQUES
Preoperative Preparation
Each patient must be assessed on an individual basis when considering a minimally
invasive approach. The patient’s body habitus, operative history, and posteroanterior and
lateral chest X-ray films should be carefully reviewed. Although not routinely necessary,
spiral computed tomography (CT) can be an accurate preoperative study to assess aortic
annulus position (4). Obtaining adequate exposure can be complicated by marked obe-
sity, significant chest wall abnormalities (pectus excavatum, deep anteroposterior diam-
eter), and unusual cardiac orientation (extremely vertical or horizontal, left shift) (5). The
need for aortic valve surgery and concomitant procedures, such as coronary grafting or
additional valve procedures, may not be amenable to some or all minimally invasive
approaches. Transesophageal echocardiography is recommended for guiding cannula-
tion, clearing intracardiac air, and assessing the technical results. Pulmonary artery cath-
eter use is varied but, in general, recommended in elderly patients, those with decreased
left ventricular function, and those with severe pulmonary hypertension. Transcutaneous
defibrillator pads placed on patients may be useful.
Incisions
In general, there have been four operative approaches presented for minimally inva-
sive aortic valve surgery. All four approaches utilize a less-than-10-cm incision and
include the right parasternal incision, the right anterolateral thoracotomy, the transverse
sternotomy, and variations of the mini-sternotomy. Numerous cannulation combinations
are possible, including the aorta, atrium, axillary artery/vein, innominate vein, and femo-
ral artery/vein.
RIGHT PARASTERNAL INCISION
The right parasternal incision, as presented by Cosgrove in 1996, was the first reported
MIAVR technique (Fig. 1) (3). A 5–10-cm incision is made, extending from the lower
border of the second costal cartilage to the upper border of the fifth costal cartilage. Once
the pectoralis major muscle is divided, the second and third costal cartilages are exposed
Fig. 1. A right parasternal skin incision over the second and third costal cartilages for aortic valve
exposure is illustrated. The femoral vessels are exposed through an incision made 1 cm above the
groin crease.
and totally excised. Of note, a costal cartilage sparing variation of this procedure,
reported by Minale, involves cutting and retracting the third and fourth costal cartilages
for exposure without excision (6). Next, the right pleura is opened to prevent postopera-
tive pericardial effusions. The intact right internal thoracic artery is retracted laterally and
the pericardium is incised and marsupialized to the wound edges with multiple retraction
sutures (7). At this point the aorta is exposed along with the right atrium, which can be
used for direct cannulation if needed, and the superior pulmonary vein, which can be used
for venting if needed. In most cases, after heparinization, either percutaneous or open
right femoral artery and vein cannulation is utilized. Open femoral access is achieved by
making a 3–5-cm inguinal crease incision. Using transesophageal echocardiographic
guidance, the long femoral venous cannula is positioned with the tip in the right atrium
for CPB venous drainage. Vacuum-assisted venous drainage via regulated wall suction
or a centrifugal pump enables the use of smaller cannulas while ensuring complete
drainage of the right side of the heart. A relatively small percentage of aortic valve surgery
patients may have significant peripheral atherosclerotic disease, making cannulation of
the femoral artery risky. In such cases, the axillary artery and vein can be used as alter-
native cannulation sites (8).
After CPB is initiated, the systemic temperature is lowered to 28°C and the aorta is
crossclamped above the sinotubular junction through this incision. Once the heart
decompresses, the aortic root is easily displayed. Antegrade cardioplegia can be delivered
in the ascending aorta when isolated aortic stenosis is present. At this point, an oblique
incision is made in the aorta, extending to the noncoronary cusp. In cases with aortic insuf-
ficiency, antegrade cardioplegia can now be administered directly into the coronary ostia.
Retrograde cardioplegia delivered via a coronary sinus catheter may offer additional
myocardial protection in all cases. In order to optimize exposure after the aortotomy, four
retraction sutures are placed to keep the aorta open and tension sutures are placed above
each commissure to elevate the aortic annulus into the center of the operative field. It is
now possible to perform a full spectrum of aortic valve repair or replacement procedures.
Adequate venting is usually possible through the aortic annulus, and de-airing is achieved
via an ascending aortic vent. Transesophageal echocardiography probe placement is
recommended for monitoring of intracardiac air. If needed, ventricular pacing wires are
more easily placed at this point, while the heart is decompressed. Once the aortotomy
closure is complete, the crossclamp is removed and the patient is rewarmed, transfused,
and separated from CPB. After decannulation, two right pleural chest tubes are placed
and the wounds are closed.
The right parasternal incision for MIAVR has reported benefits when compared to full
sternotomy, such as decreased blood loss, less postoperative pain, and a reduction of both
hospital length of stay and direct hospital costs (9,10). Another advantage of this approach
is that the pericardium is not opened over the right ventricular outflow tract, thereby
reducing the risk of injury to this area should reoperation be necessary. Since its incep-
tion, the right parasternal incision has had shortcomings limiting its clinical use, includ-
ing instances of suboptimal aortic valve exposure, anterior chest wall instability, and
impractical conversion to full sternotomy when necessary (9). Other disadvantages of
this approach include complications associated with femorofemoral cannulation, fre-
quently necessary due to incision constraints limiting direct cannulation. Owing to the
above-mentioned disadvantages and the subsequent development of more versatile
approaches, the right parasternal incision is no longer extensively used for MIAVR.
RIGHT ANTEROLATERAL THORACOTOMY
A right anterolateral thoracotomy for MIAVR was first described by Benetti in 1997
(11). For this approach, the patient is either positioned on the operating table in a 30° left
lateral decubitus position with the right arm elevated above the head or in the standard
supine position. A 5–7-cm incision is made through the third intercostal space and an
adapted rib spreader (Access Platform, CTS, Inc., Menlo Park, CA) is used to expose the
aortic root, the right atrium, and the right superior pulmonary vein. CPB cannulation is
established via the right femoral artery and the right atrium. Alternatively, a long femoral
venous cannula positioned in the right atrium can be used, as previously described.
Venting is performed through the right superior pulmonary vein. Upon completion of a
transverse aortotomy and placement of commissural sutures, aortic valve replacement or
repair is completed.
Avoidance of rib fractures when establishing exposure with this technique limits
postoperative pain. Benetti contends that the right anterolateral thoracotomy is a less
traumatic and painful approach because removal of rib or cartilage fragments is not
required (11). This approach is useful when combined aortic and mitral valve procedures
are necessary (12). However, the right anterolateral thoracotomy for isolated aortic valve
procedures provides some exposure limitations and is not practical for conversion to full
sternotomy, if necessary. Overall, this approach has not been widely adopted for MIAVR.
TRANSVERSE STERNOTOMY
The transverse sternotomy for MIAVR consists of a small vertical or horizontal skin
incision with the sternum being transected at the level of the third intercostal space. Both
internal thoracic arteries are ligated and divided. A small retractor allows access to aortic
root, the superior vena cava, and the right atrium. CPB is established by directly cannu-
lating the aorta and the right atrium (a superior vena caval cannula can be added). Alter-
natively, venous drainage can be performed using a long femoral vein cannula, as
previously described. Once the diseased aortic valve is repaired or replaced, the sterno-
tomy is closed with two interrupted steel wires.
The benefits of the transverse sternotomy for MIAVR include good exposure of the
aortic valve and the possibility of direct cannulation for CPB through the incision.
Although only a small percentage of isolated aortic valve surgery patients will eventually
require coronary revascularization, the sacrifice of both internal thoracic arteries is
potentially detrimental should coronary artery bypass grafting ever be necessary (13).
This approach can result in sternal instability in some cases. Should full sternotomy be
necessary, the combination of these two approaches would result in four-piece fragmen-
tation of the sternum and a more complex sternal closure. In at least one small series,
comparing the transverse sternotomy to median sternotomy for aortic valve replacement,
the authors found unacceptably high mortality and morbidity rates with the transverse
sternotomy (14). In general, the transverse sternotomy approach has not been adopted on
a wide scale for MIAVR.
MINI-STERNOTOMY
Since the T mini-sternotomy was first popularized by Gundry, numerous variations of
the mini-sternotomy approach for MIAVR have been proposed (15). The T mini-sterno-
tomy consists of a 6–10-cm midline skin incision and a reverse T sternal midline incision
made into the second, third, or fourth intercostal space (Fig. 2). Other commonly used
variations of the mini-sternotomy include the “j,” the “J” or reverse “C,” and the hemi-
sternotomy without horizontal transection. Svensson described the “j” sternotomy as
consisting of an upper midline sternotomy that extends to a right intercostal space, pre-
serving the left sternum. This approach is especially useful when good exposure of the
ascending aorta or aortic arch is required. The “J” or reverse “C” sternotomy, also
described by Svensson, preserves the upper sternum by using an incision of the right
sternum at the first intercostal space (16). Tam has described a hemi-sternotomy
approach that consists of a partial upper sternotomy without transection of the sternum
(17). With these exposures, it is relatively easy to cannulate the right atrium or innominate
Fig. 2. An upper T mini-sternotomy for aortic valve exposure.
vein and the distal ascending aorta directly. Vents can be placed either transannular or in
the right superior pulmonary vein.
The mini-sternotomy approach allows for good exposure during MIAVR and a variety
of aortic root procedures, including full root, hemi-root, and subcoronary techniques for
implantation of homografts or stentless bioprostheses, as well as valved conduits (18).
This approach, compared to full sternotomy for aortic valve operations, has been reported
to reduce incisional trauma and duration of ventilation, decrease blood loss and postop-
erative pain, and result in a cosmetically superior wound (Fig. 3) (19,20). The mini-
sternotomy is particularly useful in reoperative aortic valve operations. In this group,
peripheral cannulation should be done before the mini-sternotomy is performed, in case
of injury to vital structures adherent to the underside of the sternum. In a comparison
between reoperative aortic valve replacement with partial upper mini-sternotomy vs full
sternotomy, the minimal incision avoided unnecessary lower mediastinal dissection,
thereby reducing blood loss, transfusion requirements, and total operative times (21).
Additional benefits of mini-sternotomy include direct cannulation (avoiding possible
complications associated with femoral cannulation) and easy conversion to full sterno-
tomy, if necessary. The mini-sternotomy is currently the most popular approach for
minimally invasive aortic valve and aortic root procedures.
Fig. 3. A mini-sternotomy incision scar after aortic valve replacement.

MINIMALLY INVASIVE AORTIC VALVE SURGERY:

A 5-YR EXPERIENCE
In this section, we will summarize our 5-yr experience with minimally invasive aortic
valve replacement and repair at the Brigham and Women’s Hospital through July 2001.
Methods
Between July 1996 and July 2001, 365 patients (mean age 63.4 ± 14.5 yr) under-
went minimally invasive aortic valve surgery through either an upper mini-sternotomy
(n = 326, 89%), a right parasternal incision (n = 37, 10.1%), or a right thoracotomy (n = 2,
0.5%). Two hundred twenty-six (63.7%) were males and 47 (12.9%) were reoperations.
Of the 365 patients, 363 (99.5%) underwent aortic valve replacement and 2 (0.5%)
underwent aortic valve repair. The most common concomitant procedures performed
included 17 (4.7%) ascending aorta repairs and 4 (1.1%) single-vessel coronary artery
bypass graftings. In the 363 patient aortic valve replacement group, 174 (47.7%) received
Carpentier–Edwards (Baxter Healthcare Corp., Santa Ana, CA) pericardial valves, 109
(29.9%) received St. Jude (St. Paul, MN) mechanical valves, 55 (15%) received
homografts, and 23 (6.3%) received Hancock (Medtronic, Inc., Minneapolis, MN) por-
cine valves. The only operation that we have not attempted through a minimally invasive
incision is the pulmonary autograft aortic valve replacement (Ross procedure).
The indication for surgery in this patient group was isolated aortic stenosis in 183
(50.3%), isolated aortic regurgitation in 99 (27.1%), and mixed aortic stenosis and regur-
gitation in 83 (22.7%). In order of frequency, the most common causes of aortic valve
pathology were calcific aortic stenosis (n = 244, 66.8%), congenital (n = 66, 18.1%),
rheumatic disease (n = 17, 4.7%), and endocarditis (n = 15, 4.1%). A majority of patients
had New York Heart Association Functional Class II or III symptoms (mean 2.4 ± 0.6).
Left ventricular ejection fraction (EF) was preserved or mildly depressed in most patients
(mean EF 57.1 ± 11.1%). Preoperative rhythm was normal sinus in 324 (88.8%) patients
and atrial fibrillation in 32 (8.8%) patients. Most patients underwent coronary angiogra-
phy (n = 296, 81.1%), 238 (65.2%) had no significant coronary disease, 51 (14.0%) had
one- to two-vessel disease requiring no intervention, and 7 (1.9%) had one- to two-vessel
disease with preoperative angioplasty or stenting. Preoperative patient characteristics
and surgical procedure details are summarized in Tables 1, 2, and 3.
Operative Techniques
Each patient’s body habitus, operative history, and chest X-ray were carefully
reviewed prior to proceeding with minimally invasive aortic valve surgery. High-risk and
elderly patients were generally approached via a full sternotomy to expedite bypass and
operative times. Significant multivessel coronary artery disease was a contraindication
to the minimally invasive approach. Transesophageal echocardiography and transcuta-
neous defibrillator pads were used in all patients. A pulmonary artery catheter was
reserved for elderly patients, those with depressed left ventricular function, and those
with severe pulmonary hypertension.
Early in our experience, we used the right parasternal (n = 37, 10.1%) approach, as
described previously in this chapter. The majority of patients were cannulated for CPB
through a small (2–4-cm) incision above and parallel to the right groin crease. The most
common cannulae we used included a 20F cannula (Sarns, Inc., Ann Arbor, MI) in the
right femoral artery, and a long 23 to 27 fr Bio-Medicus cannula (Medtronic Bio-Medicus,
Table 1
Demographics Among Patients Undergoing Minimally
Invasive Aortic Valve Surgery at One Institution
Demographics n (%)
Number of patients 365
Age (yr)
Mean 63.4 ± 14.5
Range 25–93
Men 226 63.7
NYHAa functional class 2.4 ± 0.6
Aortic valve pathophysiology
Aortic regurgitation 99 27.1
Aortic stenosis 183 50.3
Mixed 83 22.7
Aortic valve pathology
Calcific 244 66.8
Congenital 66 18.1
Rheumatic 17 4.7
Endocarditis 15 4.1
Structural valve degeneration (reop) 11 3.0
Nonstructural valve degeneration (reop) 4 1.1
Periprosthetic Leak (reop) 2 0.5
Annulo-aortic ectasia 3 0.8
Dissection 1 0.3
Myxomatous 2 0.5
Left ventricular function
Mean ejection fraction (EF, %) 57.1 ± 11.1
Preserved function (EF > 50%) 294 80.5
Mild dysfunction (EF = 40–50%) 41 11.2
Moderate dysfunction (EF = 30–39%) 18 4.9
Severe dysfunction (EF < 30%) 12 3.3
Preoperatrive rhythm
Normal sinus rhythm 324 88.8
Atrial fibrillation 32 8.8
Paced 9 2.5
Coronary artery disease
None 238 65.2
One- to two-vessel, no intervention 51 14.0
One- to two-vessel, preoperative PTCAb 7 1.9
Coronary angiogram not performed 69 18.9
aNew York Heart Association.
bPercutaneous transluminal coronary angioplasty.
Eden Prairie, MN) inserted into the right atrium via the right femoral vein. Later, we
found it possible to avoid a groin incision by cannulating the ascending aorta with a wire-
guided, thin-walled 20F arterial cannula equipped with a tapered, pliable dilator
(Medtronic DLP, Grand Rapids, MI) and the right atrium via the right femoral vein using
a 21F percutaneous Bio-Medicus cannula. However, we abandoned the right parasternal
approach for aortic valve surgery due to limited adaptability and wound complications.
Table 2
Details of Surgical Procedures
Among 35 Patients Undergoing Aortic Valve Surgery
Operative details (n = 365) n %
Incision
Hemisternotomy 326 89.3
Right parasternal 37 10.1
Right thoracotomy 2 0.0
Cannulation for cardiopulmonary bypass
Arterial cannulation
Aorta 307 0.8
Femoral artery 58 15.9
Venous cannulation
Right atrium/SVCa 163 44.7
Femoral vein/SVC 156 42.7
Innominate vein 35 9.6
Innominate + femoral 11 3.0
Cardioplegia
Antegrade only 162 44.4
Retrograde 6 2.0
Antegrade + retrograde 197 54.0
Procedure
Aortic valve repair 2 0.5
Aortic valve replacement 363 99.5
aSuperior vena cava.
Currently, we mainly use an upper mini-sternotomy (n = 326, 89.3%), carried into the
third or fourth intercostal space on the right (Fig. 4a). A Koros sliding coronary artery
bypass retractor (Baxter Corp., Chicago, IL) is used for exposure. Unless there is an
ascending aortic aneurysm that must be resected, the distal aorta is usually cannulated.
Alternatively, the femoral artery can be used for cannulation. Direct venous cannulation
is most commonly used via a right-angle 24 fr cannula (Research Medical, Midvale, UT)
positioned in the right atrium. Other possibilities for venous cannulation include innomi-
nate vein or femoral vein access (Fig. 4b) (22). The use of smaller cannula sizes is feasible
with vacuum-assisted venous drainage on CPB.
Both combination antegrade and retrograde (n = 197, 54.0%) and antegrade only
(n = 162, 44.4%) cold blood potassium cardioplegia was used in the vast majority of these
MIAVR cases. After crossclamping the aorta, antegrade cardioplegia was administered
through the aortic root and/or by Spencer cannula into the coronary ostia; additional
intermittent doses were given through the coronary ostia as necessary. Retrograde car-
dioplegia was delivered by a Heartport (Redwood City, CA) transjugular catheter to
avoid wound clutter (23). Additional myocardial protection was provided by moderate
systemic hypothermia (28°C), which was used in the majority of cases.
Upon completion of either the aortic valve replacement or repair, a ventricular pacing
wire was placed on the decompressed right ventricle prior to aortic crossclamp removal.
Intracardiac air was evacuated through the aortic root and the left atrium with
transesophageal echocardiographic confirmation. After discontinuing CPB and achiev-
Table 3
Intraoperative Details Among 365 Patients Undergoing
Minimally Invasive Aortic Valve Surgery
Intraoperative details (n=365) n %
Operative number
First time 320 87.7
Second time 39 10.7
Third time 5 1.4
Fifth time 1 0.3
Aortic valve repair 2 0.5
Aortic valve replacement 363 99.5
Carpentier Edwards pericardial 174 47.7
St Jude mechanical 109 29.9
Homograft 55 15.1
Hancock 23 6.3
Freestyle 1 0.3
Toronto SPV 1 0.3
Secondary procedures
Ascending aortic aneurysm repair 17 4.7
One-vessel bypass grafting 4 1.1
Atrial septal defect repair 2 0.5
Ascending aorta endarterectomy 3 0.8
Left ventricular outflow tract repair 1 0.3
Left ventricular myectomy 1 0.3
Sinus valsalva aneurysm repair 1 0.3
Sub-aortic membrane repair 1 0.3
Excision of anterior mediastinal tumor 1 0.3
Closure of patent foramen ovale 1 0.3
Operative duration (min)
Ischemic (aortic crossclamp)
Mean ± SD 91.3 ± 40.5
Median 80
Cardiopulmonary bypass
Mean ± SD 124.0 ± 50.5
Median 112
ing hemostasis, the incisions were closed. In the parasternal patients, the pectoralis
muscles were reapproximated to the sternum. In several cases a Gor-Tex patch (W.L.
Gore and Assoc., Inc., Elkton, MD) was used to reinforce the closure. In the mini-
sternotomy patients, the sternum was closed with five or six stainless-steel sternal wires.
Two mediastinal chest drains were routinely placed through previously placed stab incisions.
Results
OPERATIVE DATA AND RESULTS
For the 365 MIAVR patients, mean aortic crossclamp and total CPB times were 91.3
± 40.5 and 124.0 ± 50.5 min, respectively. The mean transfusion rate was 2.1 ± 5.1 U per
patient. No blood products were transfused in 136 (37.3%) patients, and 39 (10.7%)
patients required more than 4 U of packed red blood cells. Fifteen (4.1%) patients were
Fig. 4. (a) An upper mini-sternotomy consists of a 5–7 cm midline skin incision and a partial upper
sternotomy extended to the second or third intercostal space. (b) Aortic valve exposure through
an upper mini-sternotomy is illustrated. The venous (innominate—right atrial) and aortic cannulae
exit from the superior aspect of the incision to aid exposure.
reexplored for bleeding. Fifty-four (14.8%) patients spent more than 48 h in the postop-
erative intensive care unit.
There were 9 (2.5%) operative deaths from arrhythmia (n = 3), stroke (n = 2), sepsis
(n = 2), myocardial infarction (n = 1), and respiratory failure (n = 1). Early cardiac
complications included new atrial fibrillation in 84 (23.0%) patients, conduction block
requiring permanent pacemaker implantation in 23 (6.3%) patients, and myocardial in-
farction in 2 (0.5%) patients. Other significant early complications included strokes (n =
10, 2.7%) and deep sternal infections (n = 3, 0.8%). The mean hospital length of stay
(LOS) was 7.2 ± d; 79 (21.6%) patients had less than 5-d LOS and 104 (28.5%) had greater
than 7-d LOS. A majority of patients were discharged to home, while 79 (21.6%) were
transferred to a rehabilitation facility. Early postoperative data are summarized in
Tables 4 and 5.
Late postoperative outcomes include 22 (6.0%) deaths from causes such as cancer,
congestive heart failure, systemic infections, and respiratory failure. A total of five
(1.4%) reoperative aortic valve replacements have been performed for valve en-
docarditis (n = 4, 1.1%) and scarification after aortic valve repair (n = 1, 0.3%). Two
(0.5%) required reoperation to repair paravalvular leaks.
CONCLUSION
Over the past 5 years, minimally invasive aortic valve and root surgery has become
feasible in a majority of patients. Currently at the Brigham and Women’s Hospital, we
use the mini sternotomy in aortic valve replacements and repairs, aortic root procedures,
and reoperations. Contraindications to the mini-sternotomy include individual patient
characteristics such as chest wall deformities, cardiac malposition, and obesity, concomi-
tant multivessel coronary artery bypass grafting, and autograft aortic valve replacement
(Ross procedure). Often, full sternotomy exposure is chosen in elderly patients or those
with significant comorbidities to reduce crossclamp times and total operative durations.
The mini-sternotomy, extended into a right intercostal space, has many benefits
including good exposure, direct cannulation, and practical conversion to full sternotomy,
if necessary. The exposure it provides is familiar to cardiac surgeons and requires few,
if any, specialized surgical instruments. With this approach, both internal thoracic arter-
ies are preserved.
Table 4
Adverse Events Among 365 Patients
Undergoing Minimally Invasive Aortic Valve Surgery
Adverse events (n = 365) No. %
Mortality 9 2.5
Cardiac complications
New atrial fibrillation 84 23.0
Permanent pacemaker 23 6.3
Balloon pump required 1 0.3
Myocardial infarction 2 0.5
Other complications
Deep vein thrombosis/pulmonary embolism 1 0.3
Pericadial effusion requiring drainage 1 0.3
Reexploration for bleeding 15 4.1
Neurological complications 12 3.3
Transient ischemic attack 2 0.5
Stroke 10 2.7
Wound complications 8 2.2
Deep chest infection 3 0.8
Superficial chest infection 5 1.4
Groin lymphocele 0 0.0
Vascular complications 1 0.3
Ascending aortic dissection or disruption 0 0.0
Femoral artery complications 1 0.3
Table 5
Postoperative Data Among 365 Recipients
of Minimally Invasive Aortic Valve Surgery
Postoperative data n %
Intensive care unit length of stay > 48 h 54 14.8
Hospital length of stay
Mean (d) 7.2 ± 4.3
<5 d 79 21.6
>7 d 104 28.5
Red blood cell transfusion
Mean (U) 2.1 ± 5.1
None 136 37.3
>4 U 39 10.7
Transfer to rehabilitation facility 79 21.6
The advent of minimally invasive techniques for aortic valve surgery has successfully
minimized associated surgical trauma and morbidity while maintaining or improving the
surgical efficacy of conventional aortic valve surgery. Undoubtedly, future surgical and
technological advances will allow continued improvements in minimally invasive aortic
valve surgery techniques.
REFERENCES
1. Harken DE, Soroff HS, Taylor WJ, et al. Partial and complete prostheses in aortic insufficiency. J Thorac
2. Starr A. Total mitral replacement: fixation and thrombosis. Surg Forum 1960;11:258–260.
3. Cosgrove DM III, Sabik JF. A minimally invasive approach for aortic valve operations. Ann Thorc Surg
1996;62:596–597.
4. Ammar R, Porat E, Eisenberg DS, Uretzky G. Utility of spiral CT in minimally invasive approach for
aortic valve replacement. Eur J Cardio-thorac Surg 1998;14(suppl 1):S130–S133.
5. Aklog L, Adams DH, Couper GS, Gobezie R, Sears S, Cohn LH. Techniques and results of direct access
minimally invasive mitral surgery: a paradigm for the future. J Thorac Cardiovasc Surg 1998;116:705–715.
6. Minale C, Reifschneider HJ, Schmitz E, Uckmann FP. Minimally invasive aortic valve replacement
without sternotomy. Experience with the first 50 cases. Eur J Cardio-Thorac Surg 1998;14(suppl
1):S126–S129.
7. Cohn LH, Adams DH, Couper GS, Bichell DP. Minimally invasive aortic valve replacement. Semin
Thorac Cardiovasc Surg 1997;9:331–336.
8. Bichell DP, Balaguer JM, Aranki SF, et al. Axilloaxillary cardiopulmonary bypass: a practical alterna-
tive to femorofemoral bypass. Ann Thorac Surg 1997;64:702–705.
9. Cosgrove DM III, Sabik JF, Navia JL. Minimally invasive valve operations. Ann Thorac Surg
1998;65:1535–1539.
10. Frazier BL, Derrick MJ, Purewal SS, Sowka LR, Johna S. Minimally invasive aortic valve replacement.
Eur J Cardio-Thorac Surg 1998;14(suppl 1):S122–S125.
11. Benetti FJ, Mariani MA, Rizzardi JL, Benetti I. Minimally invasive aortic valve replacement. J Thorac
12. Colvin SB, Grossi EA, Ribakove G, Galloway AC. Minimally invasive aortic and mitral operation. Oper
Tech Thorac Cardiovasc Surg 2000;5(3):212–220.
13. Lytle BW, Cosgrove DM III, Taylor PC, et al. Primary isolated aortic valve replacement. Early and late
results. J Thorac Cardiovasc Surg 1989;97:675–694.
14. Bridgewater B, Steyn RS, Ray S, Hooper T. Minimally invasive aortic valve replacement through a
transverse sternotomy: a word of caution. Heart 1998;79:605–607.
15. Gundry SR, Shattuck OH, Razzouk AJ, del Rio MJ, Sardari FF, Bailey LL. Facile minimally invasive
cardiac surgery via ministernotomy. Ann Thorac Surg 1998;65:1100–1104.
16. Svensson LG. Minimal-access “J” or “j” sternotomy for valvular, aortic, and coronary operations or
reoperations. Ann Thorac Surg 1997;64:1501–1503.
17. Tam RKW, Almeida AA. Minimally invasive aortic valve replacement via hemi-sternotomy: a prelimi-
nary report. Eur J Cardio-Thorac Surg 1998;14(suppl 1):S134–S137.
18. Byrne JG, Adams DH, Couper GS, Rizzo RJ, Cohn LH, Aranki SF. Minimally invasive aortic root
replacement. Heart Surg Forum 1999;2(4):326–329.
19. Cohn LH, Adams DH, Couper GS, et al. Minimally invasive cardiac valve surgery improves patient
satisfaction while reducing costs of cardiac valve replacement and repair. Ann Surg 1997;226:421–428.
20. Machler HE, Bergmann P, Anelli-Monti M, et al. Minimally invasive versus conventional aortic valve
operations: a prospective study in 120 patients. Ann Thorac Surg 1999;67:1001–1005.
21. Byrne JG, Aranki SF, Couper GS, Adams DH, Allred EN, Cohn LH. Reoperative aortic valve replace-
ment: partial upper hemisternotomy versus conventional full sternotomy. J Thorac Cardiovasc Surg
1999;118:991–997.
22. Zlotnick AY, Gilfeather MS, Adams DH, Cohn LH, Couper GS. Innominate vein cannulation for venous
drainage in minimally invasive aortic valve replacement. Ann Thorac Surg 1999;67:864–865.
23. Byrne JG, Hsin MK, Adams DH, et al. Minimally invasive direct access heart valve surgery. J Card Surg
2000;15:21–34.
Chapter 22 / Strategies for Reducing Trauma in Congenital Heart Surgery 309
IV MINIMALLY INVASIVE CONGENITAL,

PERICARDIAL, AND ARRHYTHMIA SURGERY
22 Strategies for Reducing Trauma

in Congenital Heart Surgery
Redmond P. Burke, MD
and Robert L. Hannan, MD
CONTENTS
INTRODUCTION
SIMPLE CONGENITAL CARDIAC LESIONS
VATS PROCEDURES FOR OTHER EXTRACARDIAC LESIONS
AVOIDING THORACOTOMY FOR PALLIATIVE SHUNTS
ENHANCED OPERATIVE IMAGING AND DOCUMENTATION
ATTEMPTS TO REDUCE INCISIONAL TRAUMA
FOR CONGENITAL SURGERY
ALTERNATIVE INCISIONS FOR OTHER CONGENITAL HEART
PROCEDURES
EFFORTS TO REDUCE CARDIOPULMONARY BYPASS TRAUMA
ATTEMPS TO REDUCE SURGICAL TRAUMA WITH ROBOTICS
FUSING INTERVENTIONAL CARDIOLOGY AND CARDIAC SURGERY
TO REDUCE TRAUMA
HYBRID CLOSURE OF SEPTAL DEFECTS
HYBRID THERAPY FOR PULMONARY ARTERY STENOSIS
CONCLUSION
REFERENCES
INTRODUCTION
Strategies designed specifically to reduce trauma in congenital heart surgery have
emerged over the past decade. Parallel advances in endoscopic imaging and cardiopul-
monary bypass technology have resulted in new surgical approaches to a variety of
congenital heart lesions. The relentless demands for extreme technical precision, speed,
and gentleness in congenital heart repairs place profound pressures on these “minimally
311
312 Part IV / Congenital, Pericardial, and Arrhythmia Surgery
invasive” techniques, and the surgeons performing them. Consequently, adoption has
proceeded at a slow pace and has been limited to straightforward repairs. Advances in
interventional cardiology have created striking new opportunities for completely avoid-
ing surgical trauma in patients with simple lesions, and have been adopted at a much more
rapid pace. It is possible that combined or hybrid procedures utilizing the technology and
skills of both interventional cardiologist and minimally invasive congenital heart surgeon
may produce the most important reductions in therapeutic trauma for children with even
the most complex forms of congenital heart disease.
SIMPLE CONGENITAL CARDIAC LESIONS

Patent Ductus Arteriosus
Multiple therapeutic options are available for patent ductus arteriosus, the simplest
congenital heart lesion. Indeed, thoracotomy or median sternotomy, with ligation or
ductal division, has been largely supplanted by transcatheter occlusion techniques. Sev-
eral limited thoracotomy incisions (1), and more sophisticated video-assisted thoraco-
scopic surgery (VATS) techniques (2), have been designed to minimize incisional trauma
in patients requiring this surgery.
Transcatheter coil occlusion is now a routine outpatient procedure, requiring no
incisions, which usually results in complete ductal closure. Residual flow has been
reported as a potential limitation requiring repeat procedures (3). Six-month closure rates
for the Rashkind device have been reported at 77%, vs a 90% closure rate for Gianturco
coils (4). Small femoral vessel size limits the applicability of the transcatheter techniques
in premature newborns, and patients with larger ducts (over 4 mm) consistently have had
a higher incidence of residual flow. Device migrations, hemolysis, femoral vessel dam-
age, radiation, aortic and pulmonary artery obstructions, and the infectious risk of an
intravascular foreign body are other potential complications related to device closure of
the patent ductus (5,6).
VATS evolved in response to the striking chest wall morbidity produced by the tho-
racotomy incision (7). Endoscopic applications in adult general and thoracic surgery
soon led to the first pediatric applications. Laborde described a pediatric cardiac VATS
procedure for ligation of patent ductus arteriosus (PDA). Using three instrument ports,
the duct is dissected with a one-handed cautery technique, and a vascular clip is applied
for occlusion. In his first clinical report in 38 patients with patent ductus arteriosus (2),
patients ranged from 1.5 to 90 mo in age, and from 2.4 to 25 kg in weight. Successful
ductal closure was achieved in every patient. One patient had a permanent recurrent nerve
injury, and there were no deaths.
Our technique for VATS occlusion of patent ductus utilizes four thoracostomy
incisions, allowing the surgeon to operate with two hands. Routine endotracheal intuba-
tion and general anesthesia are sufficient and single-lung ventilation is not necessary.
Visualization is achieved with an endoscope (4-mm, 30° angled endoscope [Smith and
Nephew, Dyonics, Andover, MA]), and a retractor (Pilling-Weck, Fort Washington, PA).
The remaining two ports are used for the endoscopic grasper and cautery, scissors, or clip
applier (Fig. 1). Under 4× magnification, the surgeon is able to dissect precisely the duct,
and apply either vascular clips or intracorporeal ligatures (10). DeCampli described his
experience with VATS for patent ductus arteriosus, with no residual flow, recurrent nerve
injury, chylothorax, or transfusion in full-term infants and children (11).
VATS for patent ductus arteriosus has also been used effectively in premature new-
borns (12). In 34 premature newborns weighing 575 g to 2.5 kg, the median procedure
Fig. 1. Instrument positions for VATS PDA ligation.
time was 60 min. Operative mortality was zero, and there was trace residual Doppler flow
in two patients. Conversion to thoracotomy was performed in four patients (12%) to
improve exposure. Hines et al. described similar results in their group of 21 neonates
undergoing VATS PDA interruption (13).
Compared to transcatheter device occlusion techniques, VATS allows highly effec-
tive ductal closure with no restrictions based on patient size, hemodynamic condition, or
ductal anatomy. VATS does not expose patients to the risks of radiation, device embo-
lization, left pulmonary artery obstruction, aortic obstruction, or the long-term presence
of an intravascular foreign body. The learning curve is steep for surgeons lacking endo-
scopic skills, suggesting that preparation in an animal or simulation laboratory might be
beneficial. We are confident that long-term assessment of the VATS technique will show
a reduction in the incidence of scoliosis (7–9), postthoracotomy pain syndromes (14), and
chest wall deformities (15), described in pediatric patients after posterolateral thorac-
otomy. Numerous studies demonstrate no convincing differences in cost or efficacy for
either transcatheter or surgical closure (3,16–18). The techniques each have strengths and
weaknesses, which vary among institutions. At Miami Children’s Hospital, we compared
the results of VATS and coil occlusion for PDA in a consecutive series of patients and
found no significant clinical differences in the techniques. We offer families and refer-
ring doctors the option of a VATS approach or transcatheter device, and share the feeling
that full thoracotomy for patent ductus can and should almost always be avoided.
Vascular Ring
The first VATS vascular ring division was described in 1993 (19), using the same
technique and instrumentation developed for VATS PDA interruption. The left subcla-
vian artery serves as a landmark to initiate endoscopic exposure and dissection of the ring
elements. The atretic arch segments, or ligamentum, are identified, ligated proximally
and distally, and divided, as in an open approach. In a report comparing VATS vascular
ring division to conventional division via thoracotomy (20), VATS was used in eight
pediatric patients ranging in age from 40 d to 5.5 yr and in weight from 1.8 to 17.1 kg.
Clinical success was achieved in all but one patient, who required reexploration via
thoracotomy to divide a residual obstruction created by an undivided ligamentum
arteriosum. Complications included transfusion (one patient), and chylothorax (one
patient). When compared to a historical control group of patients undergoing vascular
ring division via thoracotomy, the VATS patients did not differ in age, weight, intensive
care unit or postoperative hospital stay, duration of intubation, or hospital charges.
To date, the authors have performed VATS vascular ring division in 25 patients,
ranging in age from 8 d to 5.5 yr. In the past year, procedure times have averaged 2 h.
Length of stay has ranged from 1 to 6 d. Conversion to thoracotomy is not considered a
complication, but rather, the exercise of good clinical judgment, particularly when the
ring is formed by large, patent vessels. In such cases, a 3-cm lateral thoracotomy is used
to gain proximal and distal vascular control. There is good evidence that pulmonary
function abnormalities will persist in vascular ring patients repaired after the onset of
symptoms (21). The decreased chest wall incisional trauma of the VATS approach may
justify earlier intervention in these patients with asymptomatic vascular rings, and per-
haps improve their long-/term outcomes.
VATS PROCEDURES FOR OTHER EXTRACARDIAC LESIONS

VATS techniques have been used for a variety of other extracardiac congenital lesions
and applications. These include pericardial window, epicardial pacemaker insertion,
treatment of chylothorax, diaphragm plication (22), interruption of arterial and venous
collaterals, and thoracic explorations for esoteric lesions (i.e., absent left pericardium
syndrome) (23). Recently, VATS for early treatment of pediatric empyema has been
described as a method of decreasing the duration of hospital stay—average postoperative
stay was 4.9 d, total stay was 7.3 d (24).
AVOIDING THORACOTOMY FOR PALLIATIVE SHUNTS

Thoracotomy for modified Blalock–Tausig shunt palliation can be avoided in all
patients by using a median sternotomy approach (25). It is our opinion that subjecting
patients to a thoracotomy incision for palliative shunting is unjustifiable, and patients
subjected to bilateral thoracotomies and repeated median sternotomy (frequently seen in
patients with pulmonary atresia, ventricular septal defect, and multiple aortopulmonary
collaterals) have clearly suffered unnecessary cumulative chest wall trauma. When
necessary, staged management can be performed effectively through repeated median
sternotomy with minimal risk related to the reoperative sternotomy. Other advantages to
the median sternotomy approach for shunts have been well documented and include
improved patency, decreased phrenic nerve injury, decreased pulmonary artery stenosis,
decreased accidental shunts to the pulmonary vein, better access for conversion to bypass,
easier access to the shunt site at reoperation, and decreased chest wall collateral formation
in cyanotic patients. Justification for a thoracotomy approach to the modified Blalock–
Taussig shunt seems to have been reduced to the expression, “That’s the way we’ve
always done it,” and should no longer be condoned.
ENHANCED OPERATIVE IMAGING AND DOCUMENTATION

Routine video-assisted procedures for extracardiac lesions led naturally to video-
assisted endoscopic applications during open-heart operations (26). We define intraop-
erative cardioscopy as the use of endoscopic imaging tools during open-heart procedures
to facilitate visualization and repair of remote intracardiac structures, and to create visual
documentation of cardiac lesions before and after repair. This technique allows surgeons
to achieve anatomic visualization without resorting to excessive retraction or extended
cardiac incisions. This is a particular advantage when exposure has been limited by the
use of small incisions.
Over the past 7 yr at Miami Children’s Hospital, this technique has evolved into the
routine use of a 4-mm operating endoscope in the operative field as a third eye for the
surgical team, exposing remote areas within the ventricles, the ventricular outflow tracts,
and the subvalvular apparatus. The pleural spaces are also easily explored with the
endoscope. Cardioscopy has been used to facilitate complex valve surgery, ventricular
septal defect repair, left ventricular outflow tract resections, placement of septal occlu-
sion devices, and even left ventricular thrombectomy (27).
Video-assisted endoscopic technology also allows collection and storage of digital
video and still images, for immediate and future correlation with other imaging modali-
ties (angiography, MRI, and echocardiography). The images can also be digitally stored
for long-term reference, allowing surgeons to assess the outcomes of various surgical
techniques more accurately, and to prepare for reoperations. We have used routine
intraoperative cardioscopy for the past 7 years in 2000 open-heart procedures, with no
complications related to the endoscopy (28). Other investigators have described similar
experiences with routine intraoperative cardioscopy (29,30).
ATTEMPTS TO REDUCE INCISIONAL TRAUMA

FOR CONGENITAL SURGERY
Coincident with our efforts to reduce the use of thoracotomy in congenital heart
therapy, publications on alternatives to median sternotomy have proliferated in the tho-
racic literature. One of the first alternatives to full median sternotomy was reported on
the Internet by Levinson (31). In an adult patient, an atrial septal defect was repaired
through a lower sternotomy and xiphoid resection, combined with femoral cannulation
and fibrillatory arrest. This report was followed by a flurry of alternate incisions.
Thoracotomy
Ironically, despite increasing documentation of the complications of thoracotomy
incisions for extracardiac procedures, numerous groups have adopted this incision for
simple congenital open-heart repairs. The obvious historical intent of this approach has
been to hide the incision under the right breast in female patients. Kappert described a
right lateral chest incision and femoral cannulation using port-access equipment
(Heartport, Inc, Redwood City, CA). Thirteen adults, from 17 to 61 yr of age, were
repaired through 4–8-cm incisions in the fourth intercostal space. Bypass was achieved
with femoral cannulation. There was no mortality, and the median hospital stay was 8 d
(32). Widespread use of the port-access system has been limited because of complica-
tions including aortic dissection related to the aortic cannulae. The large size of these
cannulae has also limited the use of port-access technology in the pediatric population,
in whom the risk of this approach might be lessened by absence of aortic atherosclerotic lesions.
In India, 37 patients, ranging from 18 to 67 yr in age were approached with a 7-cm right
anterior thoracotomy and femoral cannulation, with a centrifugal pump to facilitate car-
diac drainage. Aortic clamping was achieved with a special clamp placed through a small
midclavicular incision. Hospital stay averaged 4.2 d, and there were no complications or
mortality (33). Surgeons in Singapore described a similar technique for atrial septal
defects repair via a right anterior thoracotomy with femoral cannulation (34). In Hannover,
Germany, three approaches have been compared, including right parasternal incision
with central cannulation, anterior submammary mini-incision with femoral arterial and
central caval cannulation, and percutaneous cervical cannulation (35). In 24 patients, this
technique was used with no neurological complications or mortality, and cosmetic accep-
tance was described as “high.” In Japan, both parasternal and partial sternotomy
approaches have been advocated (36,37), and a group in China has also advocated repair
of cardiac defects through right lateral thoracotomy (38). In this uncontrolled study of
683 patients, the authors described transecting the sternum when visualization was inad-
equate, but did not document how often this was necessary. There was no follow-up
assessment of chest wall appearance or function to justify the conclusion that lateral
thoracotomy is either less invasive than, or cosmetically superior to, median sternotomy.
For numerous reasons, we have not adopted the thoracotomy approach for simple
congenital open-heart repairs. In our opinion, the anterior thoracotomy is a cosmetic risk
in infants, as the breast tissue is unpredictable in location, and vulnerable to injury and
subsequent deformity. In adults, where the incision can be placed to avoid the breast, we
counsel patients that the risks of postthoracotomy pain syndrome, breast dysesthesia,
femoral vessel cannulation injury, and distance from the operative target outweigh the
cosmetic benefit of the submammary incision. Most patients choose a partial lower
sternotomy. Cautious observers have rightfully described the many limitations of these
variations from established technique for atrial septal defect repair, contending that the
only real benefit of a “minimally invasive” approach is cosmetic (39).
Partial Sternotomy
There have been many original descriptions of the partial lower sternotomy for simple
congenital heart repairs. Byrne described a series of 59 adult patients repaired through
either a right parasternal, submammary, or upper hemisternotomy incision. There was no
mortality, and four major complications were seen (40). Del Nido described a similar
approach for pediatric atrial septal defect repairs (41). A consecutive series of patients
was divided into sternotomy and partial lower sternotomy groups. Within the latter
group, patients were cannulated either centrally or femorally. The “minimal access group”
experienced no conversions to full sternotomy, no difficulties placing femoral or aortic
cannulas, no reexplorations for bleeding, and no significant wound infections. Echo
follow-up on 19 of 54 partial sternotomy patients showed no residual atrial shunts. There
were no differences between the partial sternotomy group and the full sternotomy group
in cardiopulmonary bypass times. There were no remarkable differences between the
groups in the incidence of postoperative pericardial effusions, or hospital stay, which
averaged 3 d after partial sternotomy, and 4 d after full sternotomy. This experience has
now grown (42) and been expanded to selected patients with more complex congenital
defects (43). Khan also described a lower sternotomy approach sparing the manubrium.
Using conventional aortic and bicaval cannulation, with fibrillatory arrest, 115 patients
were repaired through a partial sternotomy. Patients had a median hospital stay of 4 d, and
there was no mortality (44). There are some data to suggest that the partial lower sterno-
tomy results in less postoperative mediastinal drainage and shorter hospital stays than full
sternotomy (45), however, measures of postoperative pain and respiratory mechanics
appear to be equal when comparing these two incisions (46).
We share the general impression that a full median sternotomy is neither esthetic nor
necessary to expose the atrial septum safely, and have adopted a lower partial sternotomy
for atrial septal defect repair. Our limited-incision technique for atrial septal defect repair
was designed to utilize standard cardiopulmonary bypass equipment and technique,
myocardial protection, neuroprotection (central aortic deairing and superior vena caval
drainage), and operative precision. Our rationale for a midline approach is multifactorial:
(1) central cannulation avoids femoral vessel injury, (2) midline incision avoids breast
tissue and denervation (47) and intercostal and phrenic nerve trauma, (3) cardioplegic
arrest protects the myocardium and minimizes the risk of air embolism, and (4) rapid
conversion to median sternotomy is possible in the event of unexpected anatomy or
pathophysiology (such as left superior vena cava, or post-repair pulmonary hypertension
and right ventricular dysfunction).
Cardiopulmonary bypass techniques can be modified to facilitate these alternate
approaches to median sternotomy. An assisted venous drainage circuit utilizing a con-
strained vortex cone (Biocone, Medtronics, Minneapolis, MN) allows smaller cannulae
to be used without compromising bypass flows (48). Routine carbon dioxide flooding of
the operative field at 3 L/min displaces air from the pericardial well and may decrease the
risk of air embolism. During longer procedures, sweep flows are increased to avoid
increases in circulating carbon dioxide.
ALTERNATIVE INCISIONS FOR OTHER

CONGENITAL HEART PROCEDURES
The merits of limited-access incisions in most complex congenital repairs are dubious
at best. Strategies to reduce or eliminate circulatory arrest times should supercede con-
cerns about incision size or location. Despite this obvious misplacement of priorities,
there have been several reports of complex congenital heart lesions being performed
through partial sternotomy incisions. Gundry (49) described partial upper and lower
sternotomy in a series of 57 pediatric patients, including neonates, for operations includ-
ing arterial switch, tetralogy of Fallot repair, atrial and ventricular septal defect repair,
aortic and mitral valve replacement, and complex arch reconstruction. Using a T incision
in the upper or lower sternum, with aortic and atrial cannulation and conventional bypass
and myocardial preservation, direct visualization of the heart is achieved and repairs are
performed using conventional instrumentation. Potential weaknesses to this approach
include the lack of effective chest tube drainage of the mediastinum, prolonged bypass
times, the possibility of sternal instability from a T incision, and loss of myocardial
protection from rewarming by the chest wall.
At Miami Children’s Hospital, we have never used partial sternotomy for complex
neonatal surgery. We have used an upper sternotomy routed into the right third interspace
for simple aortic lesions, such as subaortic membrane resection (Fig. 2). This approach
Fig. 2. Operative field during partial sternotomy for subaortic membrane resection.
allows central aortic and venous cannulation and effective deairing. Video-assisted en-
doscopy improves visualization of the left ventricular outflow tract, and operative pre-
cision is not compromised. Partial sternotomy patients tend to have shorter hospital stays
to complement their shorter incisions (50).
Alternate incisions for repair of ventricular septal defects have been described, includ-
ing lower sternotomy and right anterolateral thoracotomy (38,51,52). This approach has
not been widely accepted. The disadvantages of these reported techniques for ventricular
septal defect repair include: potentially traumatic femoral cannulation, fibrillatory arrest
without cardioplegic myocardial protection, loss of precision, highly visible anterior
thoracotomy incisions, ventricular incisions to improve exposure, and prolonged oper-

ating and bypass times. While these approaches are technically possible, no data are
presented to suggest that they are in any way “minimally invasive” or superior to conven-
tional surgical techniques.
An alarming report described a “minimally invasive” approach to right ventricular
outflow tract reconstruction (53). The authors chose to avoid repeat sternotomy by using
a “left anterior small thoracotomy” with femoral cannulation for bypass, to replace con-
duits in 4 adolescent patients. No data were presented to support the hypothesis that the
technique is less traumatic than a repeat median sternotomy. Given the obvious anatomi-
cal tendency for right ventricular-to-pulmonary artery conduits to adhere to the left
anterior chest wall, we are concerned that this approach actually increases the risk of
conduit entry upon opening the chest.
EFFORTS TO REDUCE CARDIOPULMONARY BYPASS TRAUMA

Research in cardiopulmonary bypass technology, blood conservation, neuroprotection,
myocardial protection, inflammation, hypothermia, and acid–base management strategy
have constituted the true foundation for minimizing the trauma of congenital heart sur-
gery. We contend that reduction or elimination of circulatory arrest, ischemic arrest, and
cardiopulmonary bypass times, while preserving operative precision, constitutes the best
ongoing philosophy to further the goal of reducing surgical trauma.
A good example of this philosophy in action is the effort to perform complete single-
ventricle palliation for selected patients without any cardiopulmonary bypass. Initial
palliation to increase pulmonary blood flow can sometimes be achieved with interventional
catheterization. Subsequent bidirectional cavopulmonary anastomosis can be performed
without bypass (54), although we strongly believe that maintaining cerebral perfusion by
effective superior vena caval drainage is critical. Extracardiac Fontan operations can also
be performed without the use of cardiopulmonary bypass (55), by shunting from the
inferior vena cava to the common atrium as the extracardiac anastomoses are completed.
Patients with more than one source of pulmonary blood flow are perhaps the safest
candidates for the off-bypass Fontan procedure. These include patients with bilateral
superior vena cava, or with native pulmonary artery flow in addition to their bidirectional
cavopulmonary anastomosis. Understanding that obstructed caval blood flow could
compromise central nervous system perfusion, we have used these approaches sparingly,
and agree with others that careful neurological outcome measurements should form the
basis for wider application (56).
Ventricular septal defect repair without cardiopulmonary bypass has been described,
although the limitations in precision are cause for concern. Recent studies suggest that
some muscular ventricular septal defects might be closed using a device inserted through
the right atrium without the need for cardiopulmonary bypass. This technique would
allow the use of larger septal occlusion devices, without the risk of vascular trauma from
peripheral access, and eliminate the need for cardiopulmonary bypass (57).
ATTEMPTS TO REDUCE SURGICAL TRAUMA WITH ROBOTICS

Building on military applications for robotic battlefield medics, and marketing strat-
egies for personal computers, robotic systems were rapidly developed for cardiac sur-
gery. After a frantic race, voice-activated and computer-assisted systems were used to
perform endoscopic coronary artery bypass grafting, mitral valve repair, and atrial septal
defect repair.
A truly remote, computer controlled, robotic surgery system (daVinci™ System,
Intuitive Surgical, Inc., Mountain View, CA) has been used to perform coronary artery
bypass grafting in humans (58) (see Chapter 31). Robotic arms control the surgical
instrumentation and the camera. The surgeon manipulates the instruments through a
remote virtual manipulator. These techniques have relied on peripheral access bypass
systems (Heartport system, Heartport, Inc., Redwood City, CA), which are simply too
large for infant femoral vessels. Adult atrial septal defect repair and mitral valvuloplasty
have been performed with varying degrees of robotic assistance (59–61). The other
available robotic system uses a voice-controlled robotic arm (Aesop™, Computer
Motion, Inc., Goleta, CA) to control a 3-D camera system (Vista system, Vista, Inc.,
Westborough, MA), and a telemanipulator (Zeus™, Computer Motion, Goleta, CA) for
endoscopic suturing (62) (see Chapter 30).
Purported advantages of these systems include “scaling of the surgeon’s hand move-
ments” to increase precision, and filtering to minimize tremor. This may actually be
creative marketing of the built-in lag between operator activation and instrument
response created by yards of software and hardware between the surgeon and patient. By
mounting endoscopic instruments on universal joints that can be advanced through small
trocars into the thorax, rotational freedom at the tip of the instruments is improved
compared to conventional endoscopic instrumentation; however, in the authors’ experi-
ence, this capability has never reached that of the human wrist. The virtually complete
loss of tactile feedback and the inability to regulate force applied to tissues comprise the
two most fundamental weaknesses of the robotic technology.
Pediatric congenital heart applications await resolution of several other limitations in
the technology and technique. Current 3-D imaging endoscopes (10-mm) are too large
to pass through the infant intercostal space without trauma. Available robotic systems
require relatively large (7-mm) ports for instrument insertion. Four such incisions would
be as large as a standard median sternotomy in an infant, negating any cosmetic benefit.
Furthermore, the pediatric thorax is also too small for adequate separation of the robotic
arms controlling the individual instruments.
Current robotic systems were designed to complete very simple anastomotic suture
lines, as required for coronary bypass grafting and valve replacement or repair (60,62,63).
In training simulators, these anastomoses may require 30 min to finish 10 suture throws
(64). A typical repair for tetralogy of Fallot requires approx 100 suture throws, giving us
a projected robotic suture time of 300 min. This would produce prohibitively long myo-
cardial ischemic times. It is anticipated that as the instruments are refined and downsized,
pediatric applications might be enabled, particularly for simple septal defects. The more
daunting technical hurdle will be the absolute necessity for tactile feedback to avoid
tissue trauma.
FUSING INTERVENTIONAL CARDIOLOGY AND CARDIAC

SURGERY TO REDUCE TRAUMA
A competitive relationship between interventional cardiologists and congenital heart
surgeons may hinder the development of less traumatic therapy for congenital heart
disease. In contrast, a unified field approach, emphasizing the technological capabilities
of both disciplines, might accelerate innovation. To achieve this, we envision a
multidisciplinary intervention/surgery/diagnostic theater incorporating all existing thera-

peutic technology into a single procedure suite.
HYBRID CLOSURE OF SEPTAL DEFECTS

Transcatheter device occlusion for septal defects is evolving at a ferocious pace.
Devices and techniques are now available for device closure of atrial (65) and ventricular
septal defects, patent ductus arteriosus, collateral vessels, and fenestrations. Drawbacks
of device occlusion include moderate residual flow rates (66–68); infectious, abrasive,
and corrosive risks of persistent intravascular foreign bodies; and vascular trauma related
to insertion. Device fractures have been a persistent problem in the mid-term assessments
of septal occlusion devices using both Nitinol and stainless steel. To minimize the trauma
of using these techniques in our center, surgical backup is provided so that patients found
in the laboratory to be unsuitable for device closure can be immediately converted to a
surgical approach, sparing the child a second anesthetic, and the family the stress of a
failed intervention.
Hybrid procedures are evolving. Building on early cooperative ventures between
interventional cardiology and surgery for the management of complex muscular ven-
tricular septal defects (69), a variety of combined procedures have evolved. Video-
assisted cardioscopy has been used to guide placement of transcatheter septal occlusion
devices (23). Other hybrid procedures include transcatheter closure of extracardiac Fontan
fenestrations (70) and emergent placements of transcatheter stents for occluded sys-
temic-pulmonary shunts where patients have been stabilized with planned or emergent
cardiopulmonary support in the catheterization laboratory (71). A hybrid repair of a
pulmonary artery pseudoaneurysm with endovascular stent grafting was recently
described (72).
HYBRID THERAPY FOR PULMONARY ARTERY STENOSIS

Hybrid techniques may produce the most effective strategies for managing complex
pulmonary artery stenoses. In the operating room, surgical exposure can be used to
facilitate transcatheter stent placement. We have developed several variations of this
technique at Miami Children’s Hospital in the past 2 yr. Most commonly, transcatheter
stents have been positioned intraoperatively, on bypass, with video-assisted endoscopic
guidance (Fig. 3). Conversely, in the catheterization laboratory, angiographically guided
stent placement can be facilitated with surgical assistance. Surgeons can place sheaths
into central vessels or directly into the right ventricular outflow tract with or without
cardiopulmonary bypass. We have recognized several advantages to these hybrid
approaches for pulmonary artery angioplasty. Bypass support minimizes hemodynamic
instability, central vascular access minimizes peripheral vascular trauma, large stents can
be placed in small patients, stents can be removed and repositioned easily and rapidly, and
extensive surgical dissection in the hilum can be avoided. Completion angiography
allows confirmation of effective surgical and transcatheter interventions, and ensures
viable physiology before transfer to the cardiac intensive care unit.
CONCLUSION
Morbidity, mortality, and physiological endpoints continue to form the foundation for
therapeutic plans in congenital heart care; however, the potential for emerging technol-
Fig. 3. Video-assisted endoscopic stent implantation. A transcatheter stent is positioned in the left
pulmonary artery using endoscopic visualization on cardiopulmonary bypass.
ogy to reduce the trauma of these plans remains tantalizing. We believe that a cumulative
measure of trauma to individual patients over time is the best index to develop and
maintain the most minimally invasive approach possible at any given time with existing
technology. Such a measurement strategy has not yet been realized, but within this
technique will lie the justification of our attempts to reduce incision size, procedure times,
hospital stay, and the myriad other inadequate measures of our attempts to heal.
REFERENCES
1. Yan D, Xie Q, Zhang Z, Gu C, Kawada S. Surgical treatment of patent ductus arteriosus (PDA) through
mini subaxillary extrapleural approach. Ann Thorac Cardiovasc Surg 1999;5(4):233–236.
2. Laborde F, Noirhomme P, Karam J, Batisse A, Bourel P, Saint MO. A new video-assisted thoracoscopic
surgical technique for interruption of patient ductus arteriosus in infants and children [see comments].
J Thorac Cardiovasc Surg 1993;105(2):278–280.
3. Hawkins JA, Minich LL, Tani LY, StiMevant JE, Orsmond GS, McGough EC. Cost and efficacy of
surgical ligation versus transeatheter coil occlusion of patent ductus arteriosus. J Thorac Cardiovasc
Surg 1996;112(6):1634–1638.
4. Janorkar S, Goh T, Wilkinson J. Transcatheter closure of ventricular septal defects using the Rashkind
device: initial experience [see comments]. Catheter Cardiovasc Interv 1999;46(l):43–48.
5. Patel HT, Cao QL, Rhodes J, Hijazi ZM. Long-term outcome of transcatheter coil closure of small to
large patent ductus arteriosus. Catheter Cardiovasc Interv 1999;47(4):457–461.
6. Duke C, Chan KC. Aortic obstruction caused by device occlusion of patent arterial duct. Heart
1999;82(i):109–111.
7. Van Biezen FC, Bakx PA, De Villeneuve VH, Hop WC. Scoliosis in children after thoracotomy for aortic
coarctation. J Bone Joint Surg Am 1993;75(4):514–518.
8. Westfelt JN, Nordwall A. Thoracotomy and scoliosis. Spine 1991;16(9):1124–1125.
9. Wong-Chung J, France J, Gillespie R. Scoliosis caused by rib fusion after thoracotomy for esophageal
atresia. Report of a case and review of the literature. Spine 1992;17(7):851–854.
10. Burke RP. Video-assisted thoracoscopic surgery for patent ductus arteriosus. Pediatrics 1994;93(5):823–825.
11. DeCampli WM. Video-assisted thoracic surgical procedures in children. Semin Thorac Cardiovasc Surg
1998;1:61–73.
12. Burke RP, Jacobs JP, Cheng W, Trento A, Fontana GP. Video-assisted thoracoscopic surgery for patent
ductus arteriosus in low birth weight neonates and infants. Pediatrics 1999;104(2 pt 1):227–230.
13. Hines MH, Bensky AS, Hammon JW Jr, Pennington DG. Video-assisted thoracoscopic ligation of patent
ductus arteriosus: safe and outpatient. Am Thorac Surg 1998;66(3):853–858.
14. Dajczman E, Gordon A, Kreisman H, Wolkove N. Long-term postthoracotomy pain [see comments].
Chest 1991;99(2):270–274.
15. Jaureguizar E, Vazquez J, Murcia J, Diez Pardo JA. Morbid musculoskeletal sequelae of thoracotomy
for tracheoesophageal fistula. J Pediatr Surg 1985;20(5):511–514.
16. Galal O, Nehgme R, al Fadley F, et al. The role of surgical ligation of patent ductus arteriosus in the era
of the Rashkind device. Ann Thorac Surg 1997;63(2):434–437.
17. Fedderly RT, Beekman RH 3rd, Mosea RS, Bove EL, Lloyd TR. Comparison of hospital charges for
closure of patent ductus arteriosus by surgery and by transcatheter coil occlusion. Am J Cardiol
1996;77(9):776–779.
18. Gray DT, Fyler DC, Walker AM, Weinstein MC, Chalmers TC. Clinical outcomes and costs of
transcatheter as compared with surgical closure of patent ductus arteriosus. The Patient Ductus Arterio-
sus Closure Comparative Study Group [see comments]. N Engl J Med 1993;329(21):1517–1523.
19. Burke RP, Chang AC. Video-assisted thoracoscopic division of a vascular ring in an infant: a new
operative technique [see comments]. J Card Surg 1993;8(5):537–540.
20. Burke RP, Rosenfeld HM, Wemovsky G, Jonas RA. Video-assisted thoracoscopic vascular ring division
in infants and children. J Am Coll Cardiol 1995;25(4):943–947.
21. Marmon LM, Bye MR, Haas JM, Balsara RK, Dunn JM. Vascular rings and slings: long-term follow-
up of pulmonary function. J Pediatr Surg 1984;19(6):683–692.
22. Van Smith C, Jacobs JP, Burke RP. Minimally invasive diaphragm placation in an infant. Ann Thorac
Surg 1998;65(3):842–844.
23. Burke RP, Wemovsky G, Van Der Velde M, Hansen D, Castaneda AR. Video-assisted thoracoscopic
surgery for congenital heart disease. J Thorac Cardiovasc Surg 1995;109(3):499–507; discussion 508.
24. Grewal H, Jackson RJ, Wagner CW, Smith SD. Early video-assisted thoracic surgery in the management
of empyema. Pediatrics 1999;103(5):e63.
25. Odim J, Portzky M, Zurakowski D, et al. Sternotomy approach for the modified Blalock-Taussig shunt.
Circulation 1995;92(9 suppl):11,256–11,261.
26. Burke RP, Michielon G, Wemovsky G. Video-assisted cardioscopy in congenital heart operations. Ann
Thorac Surg 1994;58(3):864–868.
27. Mazza IL, Jacobs JP, Aidousany A, Chang AC, Burke RP. Video-assisted cardioscopy for left ventricu-
lar thrombectomy in a child. Ann Thorac Surg 1998;66(l):248–250.
28. Miyaji K, Hannan RL, Ojito J, Dygert JM, White JA, Burke RP. Video-assisted cardioscopy for intra-
ventricular repair in congenital heart disease. Ann Thorac Surg 2000;70(3):730–737.
29. Rao V, Freedom RM, Black MD. Minimally invasive surgery with cardioscopy for congenital heart
defects. Ann Thorac Surg 1999;68(5):1742–1745.
30. Miyaji K, Murakami A, Kobayashi J, Suematsu Y, Takamoto S. Transxiphoid approach for intracardiac
repair using video-assisted cardioscopy. Ann Thorac Surg 2001;71(5):1716–1718.
31. Levinson M. Heart Surgery Forum 1995. http://www.hsforum.com/heartsurgery/homehsf html.
32. Kappert U, Wagner FM, Gulielmos V, Taha M, Schneider J, Schueler S. Port Access surgery for
congenital heart disease. Eur J Cardiothorac Surg 1999;16(suppl 2):S86–S88.
33. Malhotra R, Mishra Y, Sharma KK, Mehta Y, Trehan N. Minimally invasive atrial septal defects repair.
Indian Heart J 1999;51(2):193–197.
34. Sim EK, Gob IJ, Grignani RT, Chen XZ, Duranni A. Minimally invasive surgical closure of atrial septal
defects via a right anterior thoracotomy. Singapore Med J 1999;40(4):271–272.
35. Cremer JT, Boning A, Anssar MB, et al. Different approaches for minimally invasive closure of atrial
septal defects. Ann Thorac Surg 1999;67(6):1648–1652.
36. Murashita T, Hatta E, Miyatake T, Kubota T, Sasaki S, Yasuda K. Partial median sternotomy as a
minimal access for the closure of subarterial ventricular septal defect. Feasibility of transpulmonary
approach [In Process Citation]. Jpn J Thorac Cardiovasc Surg 1999;47(9):440–444.
37. Sawa Y, Matsuda H. Minimally invasive cardiac surgery—the efficacy of right parasternal approach.
Nippon Geka Gakkai Zasshi 1998;99(12):825–830.
38. Liu YL, Zhang HJ, Sun HS, et al. Repair of cardiac defects through a shorter right lateral thoracotomy
in children. Ann Thorac Surg 2000;70(3):738–741.
39. Massetti M, Nataf P, Babatasi G, Khayat A. Cosmetic aspects in minimally invasive cardiac surgery. Eur
J Cardiothorac Surg 1999;16(suppl 2):S73–S75.
40. Byme JG, Adanis DH, Mitchell ME, Cohn LH. Minimally invasive direct access for repair of atrial septal
defect in adults. Am J Cardiol 1999;84(8):919–922.
41. del Nido PJ, Bichell DP. Minimal-access surgery for congenital heart defects. Semin Thorac Cardiovasc
Surg 1998;1:75–80.
42. Bichell DP, Geva T, Bacha EA, Mayer JE, Jonas RA, del Nido PJ. Minimal access approach for the repair
of atrial septal defect: the initial 135 patients. Ann Thorac Surg 2000;70(l):115–118.
43. Nicholson IA, Bichell DP, Bacha EA, del Nido PJ. Minimal sternotomy approach for congenital heart
operations. Ann Thorac Surg 2001;71(2):469–472.
44. Khan JH, McEUiftmey DB, Reddy VM, Hanley FL. A 5-year experience with surgical repair of atrial
septal defect employing limited exposure. Cardiol Young 1999;9(6):572–576.
45. Luo W, Chang C, Chen S. Mini sternotomy versus full sternotomy in congenital heart defects: a prospec-
tive randomized study. Ann Thorac Surg 2001;71(2):473–475.
46. Laussen PC, Bichell DP, McGowan FX, Zurakowski D, DeMaso DR, del Nido PJ. Postoperative recov-
ery in children after minimum versus full-length sternotomy. Ann Thorac Surg 2000;69(2):591–596.
47. Cherup LL, Siewers RD, Futrell JW. Breast and pectoral muscle maldevelopment after anterolateral and
posterolateral thoracotomies in children. Ann Thorac Surg 1986;41(5):492–497.
48. Ojito JW, Hannan RL, Miyaji K, et al. Assisted venous drainage cardiopulmonary bypass in congenital
heart surgery. Ann Thorac Surg 2001;71(4):1267–1271.
49. Gundry SR, Shattuck OH, Razzouk AJ, del Rio MJ, Sardari FF, Bailey LL. Facile minimally invasive
cardiac surgery via a ministernotomy. Ann Thorac Surg 1998;65(4):1100–1104.
50. Miyaji K, Hannan RL, Ojito JW, White JA, Burke RP. Minimally invasive resection of congenital
subaortic stenosis. Ann Thorac Surg 2000;69(4):1273–1275.
51. Lin PJ, Chang CH, Chu IJ, et al. Minimally invasive cardiac surgical techniques in the closure of
ventricular septal defect: an alternative approach. Ann Thorac Surg 1998;65(l):165–169.
52. Chu SH, Chou NK, Chou TF, Wang SS. Left anterolateral thoracotomy for simultaneous correction of
ventricular septal defect and coarctation of the aorta. Tex Heart Inst J 1994;21(2):158–160.
53. Inianaka K, Takamoto S, Murakami A, Kaneko Y. Minimally invasive extracardiac conduit replacement
via a left anterior small thoracotomy. J Thorac Cardiovasc Surg 1999;118(6):1124–1125.
54. Jahangiri M, Keogh B, Shineboume EA, Lincoln C. Should the bidirectional Glenn procedure be per-
formed through a thoracotomy without cardiopulmonary bypass? J Thorac Cardiovasc Surg
1999;118(2):367–368.
55. Burke RP, Jacobs JP, Ashraf MH, Aidousany A, Chang AC. Extracardiac Fontan operation without
cardiopulmonary bypass. Ann Thorac Surg 1997;63(4):1175–1177.
56. Jonas RA. Commentary [In Process Citation]. J Thorac Cardiovasc Surg 1999;118(5):957.
57. Amin Z, Gu X, Berry JM, Titus JL, Gidding SS, Rocchini AP. Periventricular closure of ventricular
septal defects without cardiopulmonary bypass. Ann Thorac Surg 1999;68(l):149–153.
58. Loulmet D, Carpentier A, d’Attellis N, et al. Endoscopic coronary artery bypass grafting with the aid
of robotic assisted instruments [see comments]. J Thorac Cardiovasc Surg 1999;118(l):4–10.
Thorac Cardiovasc Surg 1999;11(3):194–205.
60. Reichenspurner H, Boehm D, Reichart B. Minimally invasive mitral valve surgery using three-dimen-
sional video and robotic assistance. Semin Thorac Cardiovasc Surg 1999;11(3):235–243.
61. Falk V, Walther T, Autschbach R, Diegeler A, Battellini R, Mohr FW. Robot-assisted minimally inva-
sive solo mitral valve operation. J Thorac Cardiovasc Surg 1998;115(2):470–471.
62. Reichenspumer H, Damiano RJ, Mack M, et al. Use of the voice-controlled and computer-assisted
surgical system ZEUS for endoscopic coronary artery bypass grafting [see comments]. J Thorac
Cardiovasc Surg 1999;118(l):11–16.
63. Shennib H, Bastawisy A, McLoughlin J, Moll F. Robotic computer-assisted telemanipulation enhances
coronary artery bypass. J Thorac Cardiovasc Surg 1999;117(2):310–313.
64. Gulbins H, Boehm D. 3-D visualization improves the dry-lab coronary anastomosis using the Zeus
Robotics System. Heart Surg Forum 1999;4:318–325.
65. Hausdorf G, Schneider M, Franzbach B, Kampmann C, Kargus K, Goeldner B. Transcatheter closure
of secundum atrial septal defects with the atrial septal defect occlusion system (ASDOS): initial expe-
rience in children. Heart 1996;75(i):83–88.
66. Hekmat K, Mehlhom U, Rainer V. Surgical repair of a large residual atrial septal defect after transcatheter
closure. Ann Thorac Surg 1997;63(5):1456–1458.
67. Das GS. Failure of devices used for the closure of atrial septal defects [letter; comment]. J Thorac
Cardiovasc Surg 1997;113(2):426–427.
68. Agarwal SK, Ghosh PK, Mittal PK. Failure of devices used for closure of atrial septal defects: mecha-
nisms and management [see comments]. J Thorac Cardiovasc Surg 1996;112(i):21–26.
69. Bridges ND, Perry SB, Keane JF, et al. Preoperative transcatheter closure of congenital muscular
ventricular septal defects. N Engl J Med 1991; 324(19):1312–1317.
70. Zahn EM, Chang AC, Burke RP, Jacobs JP. Transcatheter closure of an extracardiac Fontan fenestration.
Ann Thorac Surg 1998;66(l):260–262.
71. Zahn EM, Chang AC, Aldousany A, Burke RP. Emergent stent placement for acute Blalock-Taussig
shunt obstruction after stage I Norwood surgery [see comments]. Cathet Cardiovasc Diagn
1997;42(2):191–194.
72. Hannan RL, Miyaji K, Burke RP, Zahn EM. Endovascular stent graft treatment of a pulmonary artery
pseudoaneurysm. Ann Thorac Surg 2001;71(2):727–729.
Chapter 23 / Thoracoscopic Pericardial Surgery 327
23 Thoracoscopic Pericardial Surgery
P. Michael McFadden, MD
CONTENTS
INTRODUCTION
HISTORY
PERICARDIAL ANATOMY AND PHYSIOLOGY
PATIENT SELECTION AND PREPARATION
ANESTHETIC MANAGEMENT
SURGICAL TECHNIQUE
PERICARDIAL EFFUSION AND TAMPONADE
PERICARDIAL CONSTRICTION
PERICARDIAL TUMORS
CONGENITAL PERICARDIAL DEFECTS
PERICARDIAL CYSTS
EPICARDIAL PACEMAKER IMPLANTATION
DIAPHRAGM PACING
SUMMARY
REFERENCES
INTRODUCTION
The pericardium is a unique organ because of the strategic position it occupies within
the chest and its close proximity to other vital structures. The pericardium is subject to
many of the pathological conditions that affect other organs, such as congenital
abnormalities, inflammation, infection, trauma, systemic metabolic disorders, and neo-
plastic disease. The anatomical and functional effects of these conditions on the pericar-
dium generally result in effusive, restrictive, or constrictive conditions that may
compromise cardiac function. In most patients, pericardial disease is asymptomatic, self-
limited, and responsive to medical therapy. Symptomatic patients who do not respond to
traditional medical therapy or who become physiologically compromised may require
surgery. A variety of surgical procedures, requiring subxiphoid, thoracic, and sternal

327
incisions, have been proposed and used effectively to diagnose and treat pericardial
diseases and their sequelae. The development of video-assisted thoracoscopy and
remarkable advances in endoscopic instrumentation over the last two decades has ushered
a paradigm shift from open to minimally invasive approaches to treat pericardial disease.
HISTORY
The first anatomical description of the human pericardium is found in the writings of
Hippocrates in the fourth century BC (1). Observations in animals recorded by Galen (AD
131–201) five centuries later first described inflammatory pericarditis and the effects of
pericardial effusion and constriction on the movement of the heart. For the next 2000 yr,
observations on pericardial disease were limited to postmortem anatomical
descriptions of pericarditis, pericardial effusion, adhesions, pericardial constriction, and
calcific pericarditis. It was not until the eighteenth century that physicians began describ-
ing the physical findings associated with pericardial effusion and constriction such as
pain, percussion dullness, enlargement of the cardiac silhouette, and pericardial friction
rub. Recognition of the physiological effects of pericardial disease began in the 1800s
with observations by Kussmaul and others. Treatment of pericardial effusion by
pericardiocenthesis, subxiphoid drainage, and pericardial resection soon followed and
ushered in the era of surgical management of pericardial disease (1–3).
Thoracoscopy for the diagnosis and treatment of diseases of the chest found its origins
in the early 1900s (4–7). However, most surgeons failed to embrace thoracoscopy at that
time because its utility was limited due to rudimentary instrumentation and the absence
of endotracheal anesthesia. Despite these limitations, a few entrepreneurial surgeons
continued to develop the field of thoracoscopy during the twentieth century (8–13).
Thoracotomy, sternotomy, and subxiphoid approaches to manage pericardial disease
were adopted by the majority of surgeons with excellent results. The introduction of the
double-lumen endobronchial tube in the 1970s made single-lung ventilation possible.
This innovation not only enhanced open thoracotomy but also rekindled an interest in
thoracoscopy. The subsequent development of video-assisted thoracoscopic instrumen-
tation in the 1980s was a major stimulus for the enthusiastic expansion of thoracoscopy
as it is practiced today. Video-assisted thoracoscopy has in many circumstances sup-
planted open surgical procedures in the diagnosis and treatment of many diseases of the
chest and pericardium. Essentially every known malady of the pericardium that has been
treated by open pericardial window, pericardiectomy, biopsy, or resection may with
proper indication be preferentially managed now by video-assisted thoracoscopy.
PERICARDIAL ANATOMY AND PHYSIOLOGY

The pericardium arises from mesenchymal lacunae that develop between the pleural
spaces and the diaphragm in the first trimester of fetal life. Fusion of the lacunae result
in the formation of a sac called the pericardium, which encompasses the heart from the
diaphragm to the base of great vessels and laterally to the hilum of each lung. The fibrous
outer layer of the pericardium is composed of dense collagen tissue. The inner layer is a
thin serosal surface that lines the entire pericardial cavity and covers the heart. The serosal
surface around the heart is termed the visceral pericardium or epicardium. The parietal
pericardium is composed of the fibrous sac and the serosal surface lining it (14).
A neurovascular bundle comprised of the phrenic nerve, the pericardiophrenic artery

and vein, and a pericardial fat pad transverses the parietal pericardium bilaterally. Func-
tionally, the pericardium protects and maintains the heart in a stable midthoracic position.
It prevents overdistention, torsion, and pendulous motion that may occur with volume
overload or change in body position (15). It also serves as a barrier for the spread of
infection (14). The inner serosal surface is comprised of cuboidal cells that contain
microvilli, which are responsible for both the secretion and absorption of a clear
plasmalike fluid (16). The 15–30 mL of pericardial fluid normally found in the pericardial
space lubricate the surface and reduce friction to the beating heart. Disease or conditions
that alter either the secretion or reabsorption of pericardial fluid may result in pericardial
dysfunction and lead to compromised cardiac function.
PATIENT SELECTION AND PREPARATION

Selection of appropriate candidates for a video-assisted thoracoscopic approach for
the diagnosis and management of pericardial disease depends on the complexity of the
problem to be addressed and the physical condition and limitations of the individual
patient (17). In general, the patient must be able to tolerate general anesthesia and con-
tralateral single-lung ventilation. Patients with a single lung, pulmonary hypertension, or
advanced interstitial lung disease who are unable to maintain acceptable arterial oxygen
saturations, blood pH, and systemic blood pressure due to poor pulmonary reserve are not
considered candidates for a thoracoscopic pericardial procedure. The thoracoscopic
approach is directed by the type, location, and complexity of the pericardial disease.
Inflammatory conditions that cause concomitant pleuropericarditis may result in adhe-
sions that limit access to the pleural space. Likewise, patients with a history of pleurodesis,
irradiation, emphysema, or previous thoracoscopy must be selected carefully. Ventilator-
dependent patients are generally poor candidates due to their underlying lung disease and
inability to sustain single-lung ventilation. The physiological effects of pericardial effu-
sion, constriction or restriction under the conditions of general anesthesia, a lateral
decubitus position, and single-lung ventilation must be considered and individualized for
each patient.
The thoracoscopic procedure along with its objective, anticipated results, and possible
complications are reviewed with the patient preoperatively. If the procedure cannot be
successfully accomplished thoracoscopically or should complications arise, the possibil-
ity of conversion to open thoracotomy must be included in the consent. The cardiothoracic
surgeon must be familiar with intrathoracic anatomy and the pathophysiology of pericar-
dial disease. The surgeon must be thoroughly trained and experienced in thoracoscopic
technique and capable of managing emergent surgical complications (18–20).
ANESTHETIC MANAGEMENT
The thoracoscopic approach to the pericardium requires general endotracheal anesthe-
sia. Continuous monitoring of the oxygen saturation and arterial blood pressure is man-
datory. Patients with cardiac embarrassment from effusion, tamponade, or constriction
may require more invasive central venous or pulmonary arterial monitoring in order to
better evaluate and maintain hemodynamic stability. Rebound hypertension can be
anticipated following the relief of cardiac tamponade or effusion. The hypertensive
response encountered after pericardiectomy for constriction is often short-lived, and

hypotension from poor contractility due to developing interstitial myocardial edema may
result. Bladder catherization is usually reserved for those patients in whom more difficult
hemodynamic management is anticipated or when the duration of the thoracoscopic
pericardial procedure is expected to be prolonged.
A double-lumen endobronchial tube is required to allow decompression of the lung on
the side of the procedure. Although either a right- or left-sided double-lumen endobron-
chial tube may be used, a left-sided tube is preferred because of its ease of placement and
stability (21,22). Single-lumen endotracheal tubes with bronchial blockers are com-
monly used in children and patients with small tracheal diameters, in whom the larger-
profile double-lumen tubes would be contraindicated (23). We have found the bronchial
blocker tubes less reliable, as they frequently dislodge and are difficult to reposition.
Proper position and function of the endobronchial tube are essential and are confirmed
by auscultation and fiber-optic bronchoscopy before and after turning the patient to the
appropriate lateral decubitus position for thoracoscopy. Flexion of the table assists the
surgeon by widening the intercostal spaces to facilitate thoracoscopy. A supine position
with arm extension is utilized to provide access to either side of the pericardium when
more extensive procedures are required, such as bilateral pericardial windows or total
pericardiectomy.
Anesthetic agents that delay immediate postoperative extubation are avoided. Signifi-
cant postoperative pain is rare. If oral analgesics do not suffice, consideration may be
given to the use of intercostal nerve blocks or epidural pain catheters. Nonsteroidal anti-
inflammatory agents are often beneficial when combined with oral analgesics for post-
operative intercostal pain (17).
SURGICAL TECHNIQUE
A video-assisted thoracoscopic approach to the pericardium is performed, with the
patient prepared in a lateral decubitus position as for thoracotomy. The lung on the
operative side is decompressed and a 2-cm thoracoport incision is made in the fifth or
sixth intercostal space in the mid-auxiliary or posterior-axillary line. The pleural space
is digitally explored for adhesions and to ensure deflation of the lung. If a coexisting
pleural effusion is encountered, it should be aspirated and sent for appropriate studies
when indicated. A 15-mm thoracoport is placed through the incision into the pleural
space. A 0° thoracoscope with video camera attachment is introduced through the
thoracoport and the entire pleural cavity and its contents are thoroughly evaluated.
Particular attention should be given to the pericardial anatomy and the location and
course of the phrenic nerve and the accompanying pericardiophrenic artery and vein. On
the right, the phrenic nerve courses laterally on the vena cava superiorly and traverses the
pericardial reflection, which closely approximates the right lung hilum. On the left, the
phrenic nerve courses more anteriorly and away from the hilum and reaches the dia-
phragm just posterior to the pericardial apex. This left-sided anatomical relationship is
important, as it allows for both anterior and posterior approaches to the left pericardium
(Fig. 1). Two additional thoracoport sites are usually necessary to allow for the introduc-
tion of instruments to perform the indicated procedure. These sites are best positioned one
or two interspaces lateral to the endoscopic camera port and in the anterior-axillary line
to avoid “dueling” of equipment and to facilitate operative exposure (24) (Fig. 2).
Thoracoport sites are individualized depending on the anatomy and particular character-
Fig. 1. Thoracosopic exposure of the left pericardium. Note relationship of the phrenic nerve (PN),
pericardium (P), chest wall (C), diaphragm (D), and lung (L).
istics of the patient. Video monitors are oriented so that both the surgeon and assistant
have good visualization. During the procedure, close attention is given to homeostasis,
as the pericardium is a highly vascular structure, particularly in disease. At the termina-
tion of the procedure a chest tube is placed posterolateral to the lung through the most
inferior incision to provide adequate, dependent postoperative pleural drainage. The
remaining thoracoport incisions are closed with absorbable sutures. The patient is extu-
bated in the operating room unless there is an indication for continued ventilatory
support. An erect anterior–posterior chest film is obtained in the recovery room to con-
firm full expansion of the lung and appropriate chest tube position.
PERICARDIAL EFFUSION AND TAMPONADE

An accumulation of fluid within the pericardial sac may result from a variety of benign
or malignant conditions. Acute inflammation or systemic conditions that involve the
pericardium and alter the balance between secretion and reabsorption of pericardial fluid
may result in pericardial effusion. Causes of pericarditis and pericardial effusion include
inflammatory, infectious, neoplastic, postsurgical, systemic metabolic, and traumatic
etiologies (Table 1).
Acute pericardial effusions are often symptomatic, due either to direct effects of
pericardial inflammation and irritation or to the adverse physiological effects of tampon-
ade on cardiac function. Since the pericardium is essentially noncompliant, cardiac com-
pression may occur with relatively small amounts of acute pericardial effusion. The
character of the effusion may be serous, serosanguinous, hemorrhagic, or chylous,
Fig. 2. Video-assisted thoracoscopic approach to the pericardium. Working thoracoports are

placed lateral and slightly anterior to video-thoracoscopic camera ports to avoid “dueling” of the
instruments during the procedure.
depending on the cause. Initial treatment consists of bed rest, pain control, and pharma-
cological treatment of the specific etiology, such as administration of anti-inflammatory
agents for nonspecific pericarditis, antibiotics for infectious pericarditis, dialysis for
uremia, and thyroid replacement for myxedema. Some cases may require pericardiocentesis
to determine the cause and treat the effusion. These measures alone are effective in
Table 1
Causes of Pericarditis and Pericardial Effusion
Inflammatory
Acute nonspecific pericarditis
Dressler’s syndrome
Infectious
Viral
Bacterial
Tuberculous
Parasitic
Neoplastic
Metastatic carcinoma
Lymphoma/leukemia
Primary pericardial/cardiac malignancy
Postirradiation
Postsurgical
Postpericardiotomy syndrome
Postthoracotomy
Chylopericardium
Hemothorax
Postoperative
Anticoagulation (delayed)
Systemic metabolic
Uremia
Cirrhosis
Amylodosis
Hypothyroidism
Connective tissue disorders
Lupus erythematosis
Rheumatoid arthritis
Rheumatic fever
Traumatic
Penetrating
Blunt
resolving the inflammation and effusion in most patients. However, in cases of acute
cardiac tamponade or when accelerated reaccumulation of effusion occurs following
pericardiocentesis, rapid and effective drainage is indicated. Effective drainage aids not
only in establishing the diagnosis and resolving the symptoms of effusion but also in
preventing the recurrence of effusion or the development of late pericardial constriction.
Simple bedside subxiphoid pericardial drainage performed under local anesthesia and
sterile preparation is quick and usually effective. This is especially true in the case of
cardiac tamponade following open-heart surgery or catheter-induced complications
occurring after percutaneous cardiac intervention. Open pericardial window and peri-
cardiectomy also provide excellent pericardial drainage, but they require time for patient
mobilization and preparation, thoracotomy, and general anesthesia. These procedures
may also be performed thoracoscopically. Thoracoscopic drainage is less morbid than
surgical thoracotomy but shares the same constraints as thoracotomy with regard to
preparation, mobilization, and general anesthesia. Severely compromised patients ben-
efit from the most expeditious approach to pericardial drainage (25). Thoracoscopic
pericardiectomy and pericardial window appear to be best suited for drainage of chronic
effusions in hemodynamically stable patients. Visualization of the pericardium is excel-
lent, and wide resection of the pericardium both anterior and posterior to the phrenic
nerve on the left is possible. In the presence of a highly vascular pericardium, an endo-
scopic stapler may be useful to ensure hemostasis when opening or resecting the pericardium.
Thoracoscopic drainage has demonstrated excellent results, with a low incidence of
recurrence, morbidity, and mortality (15). Advocates of the procedure report better
visualization, avoidance of the morbidity of thoracotomy, and shorter hospitalization.
Despite these distinct advantages, long-term survival with video-assisted thoracoscopic
pericardiectomy has not been affected (15,26,27). This is likely the result of frequent
application of thoracoscopic pericardial resection in patients with malignant effusions,
a condition that portends a poor survival. Thoracoscopic pericardiectomy for adherent
pericardial constriction has been proposed but has yet to be adopted as a satisfactory
alternative to managing this late sequelae of pericardial disease.
PERICARDIAL CONSTRICTION
Pericardial constriction is the result of a chronic constrictive inflammatory process of
the pericardium that may lead to significant morbidity. The etiology is usually idiopathic,
but other frequent causes are mediastinal irradiation, connective tissue disorders, tuber-
culosis, neoplastic disease, and trauma (28,29). Since the advent of open-heart surgery
in the 1950s, postoperative pericarditis has become another important cause of pericar-
dial constriction (30). Essentially, constriction may result from any process that causes
pericarditis. Fibrosis and pericardial thickening constricts the ventricles, impedes dias-
tolic filling, and leads to reduced cardiac output. The process is insidious and the clinical
course is that of progressive deterioration. Hemodynamic evidence of impaired diastolic
ventricular filling at cardiac catherization along with other supportive studies such as
computerized tomography and magnetic resonance imaging are helpful in making the
diagnosis. Definitive treatment of pericardial constriction requires pericardiectomy.
Although the technique of thoracoscopic pericardiectomy has advanced, its use is
limited to the management of pure effusive pericardial disease or in patients with peri-
cardial constriction who demonstrate a combined effusive–constrictive component
(15,31–34). Thoracoscopic pericardiectomy for dense, adhesive pericardial constriction
has not been advocated. Adequate exposure for control of possible hemorrhagic compli-
cations or for the institution of cardiopulmonary bypass, should it be required, is limited.
Thoracoscopic pericardiectomy in the presence of adherent and patent coronary bypass
grafts will also be dangerous. Additionally, a thoracoscopic approach is not possible in
patients who require concomitant cardiac procedures, such as valve replacement or coro-
nary artery revascularization. For these reasons, median sternotomy remains the
approach of choice for pericardial resection in pericardial constriction, as it has demon-
strated excellent results, lesser morbidity, and shorter hospital stays while allowing greater
flexibility for the performance of concomitant procedures (35–39). In a rather limited
number of patients, thoracoscopic pericardiectomy for pericardial constriction may be
possible. However, the indications and limitations of this approach for pericardial con-
striction have yet to be defined.
PERICARDIAL TUMORS
The majority of pericardial neoplasms are metastatic malignancies. The pericardium
is involved in approx 8.5% of patients with malignant tumors. The heart is also involved
in the majority of patients with pericardial metastases (1). Malignant pericardial effusion
and hemopericardium are frequent sequelae of pericardial malignancy and result from
lymphatic obstruction or bleeding from the tumor. Carcinoma of the lung and breast are
the two malignancies that most frequently involve the pericardium (40,41). Owing to the
proximity of these two organs to the pericardium, local extension and lymphatic spread
are common routes to pericardial involvement. Melanoma, leukemia, and sarcoma are
metastatic tumors that have a particular propensity for hematogenous spread to the peri-
cardium (1). Other tumors, such as malignant thymoma, thymic carcinoma, teratoma, and
pleural mesothelioma, which arise in proximity, also often involve the pericardium. The
associated pericardial effusion and bulky extent of the metastatic implants often result in
cardiac compromise from tamponade or cardiac compression. Therefore, the role of
thoracoscopy in pericardial tumor management is often limited to diagnostic biopsy in
occult disease or the creation of a pericardial window for drainage of a known symptom-
atic malignant pericardial effusion.
Primary pericardial malignancies, tumors that arise directly from the pericardium or
its associated structures, are very rare. Angiosarcoma constitutes the most common
sarcoma of pericardial origin and often involves the right atrium (42). Thoracoscopic
biopsy of an angiosarcomatous tumor must be undertaken with caution, as this tumor is
highly vascular and prone to hemorrhage. Owing to its extent, vascular nature, and
frequent cardiac involvement, thoracoscopic resection of angiosarcoma is usually not
possible. Management of these tumors typically requires extensive open surgical resec-
tion or even cardiac transplantation (43,44). Primary pericardial mesothelioma is exceed-
ingly rare and comprises only 0.7% of all mesotheliomas (45). Liposarcoma of the
pericardial fat and malignant neurogenic tumors of the autonomic or phrenic nerves may
also be encountered. If these tumors are small, localized, and not associated with a
malignant pericardial or pleural effusion, thoracoscopic resection may be considered.
Benign tumors of the pericardium are also uncommon. However, thoracoscopic resec-
tion is the procedure of choice when a benign diagnosis is confirmed and the tumor can
be removed technically without compromising the surrounding cardiac, vascular, or
neural structures.
CONGENITAL PERICARDIAL DEFECTS

Defects of the pericardium are rare and occur in fetal life as a result of compromised
circulation to the developing pleuropericardial membrane. The majority of congenital
pericardial defects are found on the left. The diagnosis is usually made serendipitously
while the patient is undergoing evaluation for more common symptomatic cardiac dis-
orders. A prominence of the left heart border or pulmonary artery on plain chest film may
suggest a pericardial defect (Fig. 3). However, computerized axial tomography,
cinemagnetic resonance imaging, and cardiac catherization best establish the diagnosis
(46–48). Complete absence of the pericardium and defects smaller than 2 cm are usually
asymptomatic. However, defects that are large enough to allow partial herniation of the
Fig. 3. Preoperative chest roentgenogram showing convex prominence of left heart and border
(arrow). From: Risher WH, Rees AP, Ochsner JL, et al. Thoracoscopic resection of pericardium
from symptomatic congenital pericardial defect. Reprinted with permission from the Society of
Thoracic Surgeons. Annals of Thoracic Surgery 1993;56:1390–1391.
left atrial appendage or a portion of the left ventricle may predispose to cardiac symptoms
such as palpitations, dyspnea, syncope, and chest discomfort (Fig. 4). Angina has resulted
from partial cardiac herniation and dynamic compression of the coronary arteries by the
lower border of the pericardial defect (46) (Fig. 5). Symptomatic pericardial defects
require surgical management to alleviate symptoms and prevent complications. Wide
thoracoscopic pericardial resection is usually sufficient to relieve constriction and
resolve the cardiac compromise. Endostapling of the left atrial appendage base may even
be required in some patients with infarction of the atrial appendage or when systemic
emboli result from thrombi with the atrial appendage.
PERICARDIAL CYSTS
Pericardial cysts, like pericardial defects, develop in embryonic life. Cysts are derived
from mesenchymal origin and may arise either from a diverticulum of the developing
pericardium or an abnormal fusion of the pericardial lacunae. Unlike pericardial defects,
cysts occur most often on the right and usually do not communicate with the pericardial
cavity (32). The radiographic appearance is classically teardrop in shape, due to the
frequent location within the anterior costophrenic angle (49). Pericardial cysts are benign
and usually unilocular and contain a clear watery fluid. They are found in persons of all
Fig. 4. Thoracoscopic appearance of a congenital left pericardial defect demonstrating partial

cardiac herniation (V = ventricle, A = left atrium, L = lung, arrow = pericardial border).
Fig. 5. Migratory coronary obstructions of left anterior descending and diagonal arteries due to
impingement of congenital left pericardial defect (arrows). From: Risher WH, Rees AP, Ochsner
JL, et al. Thoracoscopic resection of pericardium from symptomatic congenital pericardial defect.
Reprinted with permission from the Society of Thoracic Surgeons. Annals of Thoracic Surgery
1993;56:1390–1391.
ages. Radiographic studies are usually diagnostic, but aspiration or resection is indicated
if the diagnosis is uncertain or if the cyst is symptomatic.
Most pericardial cysts may be easily approached thoracoscopically, where they may
be aspirated or completely resected (50–53). Large or thick-walled cysts may require
marsupialization and fulguration of the base with electrocautery or laser (54,55). The cyst
fluid and wall are submitted for pathological confirmation of the diagnosis. Complete
resection of a symptomatic cyst is important, since recurrence of cysts has been observed (53).
EPICARDIAL PACEMAKER IMPLANTATION

Consideration may also be given to a transpericardial thoracoscopic approach to the
heart for implantation of permanent epicardial pacemaker leads or internal cardiac
defibrillator pads. Most patients requiring cardiac pacing or defibrillator implantation
have adequate venous access and will preferentially undergo transvenous lead implan-
tation under local anesthesia with excellent results (56). In the absence of central venous
access due to either venous compromise or preexisting transvenous hardware, a thoraco-
scopic approach to lead implantation provides an alternative to open thoracotomy (57–59).
When the pericardium and the pericardial space are normal, a pericardial window on the
left may be created to allow access to the myocardium for electrophysiological testing by
the use of a hand-held wand electrode to determine what may be appropriate sites for lead
implantation. “Screw-in” bipolar epicardial leads may then be implanted in the areas that
demonstrate the best thresholds for capture, sensing, and impedance. Hemostasis must
be ensured at the time of lead implantation, as intrapericardial bleeding may lead to
tamponade. To avoid this complication the pericardial window through which the leads
are placed must be opened widely to allow for pericardial decompression should bleeding
occur. The leads are passed medial to the lung in the anterior–superior mediastinum and
tunneled transthoracically into a subcutaneous pocket anterior to the pectoralis muscle.
The leads are secured to the pectoral fascia to prevent migration and then connected to
the pacing pulse generator or defibrillator. If the pericardial space is found to be obliter-
ated and a thin pericardium adherent to the heart is encountered, placement of the leads
into the heart may still be possible. Most recently, biventricular pacing therapy has gained
acceptance for the treatment of advanced congestive heart failure in patients with pro-
longed QRS intervals. Because of the difficulties associated with the placement of the
coronary sinus lead for capture of the left ventricle, thoracic surgeons are increasingly
being asked to position the left ventricular lead on the lateral wall of the heart. This can
easily be accomplished with thoracoscopic approaches.
DIAPHRAGM PACING
A discussion of thoracoscopic pericardial surgery would be incomplete without men-
tioning diaphragmatic pacing, a surgical procedure with limited but important applica-
tion. Patients with neurological injury or conditions that result in central respiratory
paralysis and who require ventilatory support may be candidates for diaphragmatic pac-
ing. Tetraplegia, central sleep apnea, congenital hypoventilation syndrome, and certain
brainstem lesions are common causes of central hypoventilation and apnea. An intact
phrenic nerve (lower motor neuron) and functional diaphragmatic muscle are necessary
requisites for diaphragm pacing (60). The site of phrenic nerve stimulation may be
selected anywhere along its cervical or intrathoracic course. Because the majority of
these patients have undergone tracheostomy for mechanical ventilation and prevention
Fig. 6. Illustration of intra-thoracic placement of left phrenic nerve stimulator by video-assisted

thoracoscopy.
of aspiration, the once-popular cervical approach to phrenic nerve stimulation has gen-
erally been abandoned in order to avoid infection of the implantable hardware and
improve patient comfort (61). Median sternotomy, thoracotomy through the anterior
second intercostal space or axilla, and paramediastinotomy have each provided excellent
exposure to the intrathoracic phrenic nerve. However, a less invasive thoracoscopic
approach has recently been advocated, which provides excellent visualization and expo-
sure of the phrenic nerve and application of the pacing electrode (62).
Phrenic nerve stimulation requires successive pulse currents to effect diaphragmatic
contraction. A biphasic, alternating, positive and negative current is preferred over a
monophasic current because it does not require as much energy and is less likely to cause
phrenic nerve injury. Although a variety of devices have been developed for phrenic
nerve stimulation, the basic component of each consists of nerve electrode strips, elec-
trode wires, a receiver, and an external stimulator (pacer). Longitudinal windows anterior
and posterior to the phrenic nerve are created thoracoscopically, and bipolar strips are
applied to either side of the nerve (Fig. 6). Because electrocautery poses a potential to
nerve injury, the pericardial windows are best created with sharp dissection. Once the
strips are secured into position, the electrode wires are tunneled into a subcutaneous
position in the anterior chest wall and attached to a receiver. Pacer function is verified
prior to closure. Patients undergoing phrenic nerve stimulation require a progressive

program of diaphragm conditioning, which may last for months. Ventilator support is
continued until respiratory parameters indicate adequate independent ventilation and
diaphragmatic function with phrenic nerve stimulation (60,63).
SUMMARY
A video-assisted thoracoscopic surgical approach for the diagnosis and management
of most diseases of the pericardium and their sequelae has become the procedure of
choice. Provided that candidates for this surgical approach are well selected and the
indications and limitations of the procedure to be performed are recognized, a good
clinical result can be expected. It is no longer a matter of determining if a particular
thoracoscopic surgical procedure can be performed on the pericardium. Thoracic sur-
geons are currently defining when and in whom a specific thoracoscopic procedure is
appropriate. As with other developing technologies in the area of video-assisted endo-
scopic surgery, the use of thoracoscopy for the management of pericardial disease is
likely to increase.
REFERENCES
1. Darsee JR, Braunwald E. Diseases of the pericardium. In: Braunwald E, ed. Heart Disease: A Textbook
of Cardiovascular Medicine. Philadelphia: Saunders, 1980:1517–1582.
2. Spodick DH. Medical history of the pericardium. The hairy hearts of hoary heroes. Am J Cardiol
1970;26:447–454.
3. Rogers FB. Historical review of pericardial disease. In: Cortes FM, ed. The Pericardium and Its Disor-
ders. Springhill, IL: Charles C Thomas, 1971:3.
4. Jacobaeus HC. Possibility of the use of the cystoscope for the investigation of serous cavities. Munch
Med Wochenschr 1910;57:2090.
5. Jacobaeus HC. Endopheural operations by means of a thoracoscope. Beitr Klin Tuberk 1915;35:1.
6. Jacobaeus HC. The cauterization of adhesions in pneumothorax treatment of tuberculosis. Surg Gynecol
Obstet 1921;32:493.
7. Jacobaeus HC. The proactive importance of thoracoscopy in surgery of the chest. Surg Gynecol Obstet
1922;34:209.
8. Fisher J. Instruments for illuminating dark cavities. Phila J Med Phys Sci 1927;14:409.
9. Lloyd MS. Thoracoscopy and biopsy in the diagnosis of pleurisy with effusion. Q Bull Sea View Hosp
1953;14:128.
10. Hatch HB Jr, DeCamp PT. Diagnostic thoracoscopy. Surg Clin North Am 1966;46:1405–1410.
11. Ochsner A Jr, Ochsner A. Cancer of the lung: recognition and management. Surg Clin North Am
1966;46:1411–1425.
12. DeCamp PT, Moseby PW, Scott ML, Hatch HB Jr. Diagnostic thoracoscopy. Ann Thorac Surg
1973;16:79–84.
13. Miller JI, Hatcher CR Jr. Thoracoscopy: a useful tool in the diagnosis of thoracic disease. Ann Thorac
Surg 1978;26:68–72.
14. Jones WG II. Pericardial disease. In: Pearson FG, Deslaurier J, Ginsberg RJ, Hiebert CA, McKneally
MF, Urschel HC Jr, eds. Thoracic Surgery. New York: Churchill Livingston, 1995:1373–1383.
15. Caccavale RJ. Video-assisted thoracic surgery for pericardial disease. In: Lewis RJ, ed. Video-Assisted
Thoracic Surgery. Chest Surg Clin N Am 1993;3:271–281.
16. Elias H, Boyd LJ. Notes on the anatomy, embryology, and histology of the pericardium. II. J NY Med
Coll 1960;2:50–75.
17. McFadden PM, Robbins RJ. Thoracoscopic surgery. Surg Clin North Am 1998;78:763–772.
18. McKneally MF. Video-asssited thoracic surgery. Standards and guidelines. In: Lewis RJ, ed. Video-
Assisted Thoracic Surgery. Chest Surg Clin N Am 1993;3:345–351.
19. McKneally MF, Lewis RJ, Anderson RJ, et al. Statement of the AATS/STS Joint Committee on Tho-
racoscopy and Video Assisted Thoracic Surgery. J Thorac Cardiovasc Surg 1992;104:1.
20. McKneally MF, Lewis RJ, Anderson RJ, et al. Statement of the AATS/STS Joint Committee on Tho-
racoscopy and Video Assisted Thoracic Surgery. Ann Thorac Surg 1992;54:1.
21. Kraenzler EJ, Hearn CJ. Anesthetic considerations for video-assisted thoracic surgery. Semin Thorac
22. Weiss SJ, Cheung AT. Anesthesia for thoracoscopic surgery. In: Kaiser LR, Daniel TM, eds. Thoraco-
scopic Surgery. Boston: Little, Brown, 1993:17–36.
23. Lobe TE. Pediatric thoracoscopy. Semin Thorac Cardiovasc Surg 1993;5:298–302.
24. Landreneau RJ, Hazelrigg SR, Mack MJ, Keenan RJ, Ferson PF. Video-assisted thoracic surgery for
diseases within the mediastinum. In: Shields TW, ed. General Thoracic Surgery. Vol. 2, 4th ed., Balti-
more: Williams & Wilkins, 1994:1810–1816.
25. Ginsburg RJ. Thoracoscopy: a cautionary note. Ann Thorac Surg 1993;56:801–803.
26. Tuna IC, Danielson GK. Surgical management of pericardial diseases. Cardiol Clin 1990;8:683–696.
27. Piehler JM, Pluth JR, Schaff HV, Danielson GK, Orszulak TA, Puga FJ. Surgical management of
effusive pericardial disease. Influence of extent of pericardial resection on clinical course. J Thorac
Cardiovasc Surg 1985;90;506–516.
28. Cameron J, Oesterle SN, Baldwin JC, Hancock EW. The etiologic spectrum of constrictive pericarditis.
Am Heart J 1987;113:354–360.
29. Fowler NO. Constrictive pericarditis: its history and current status. Clin Cardiol 1995;18:341–350.
30. Copeland JG, Riley JE, Fuller J. Pericardiectomy for effusive constrictive pericarditis after heart trans-
plantation. J Heart Transplant 1986;5:171–172.
31. Krasna M, Fiocco M. Thoracoscopic pericardiectomy. Surg Laparosc Endosc 1995;5:202–204.
32. Ferguson MK. Thoracoscopic management of pericardial disease. Semin Thorac Cardiovasc Surg
1993;5:310–315.
33. Robles R, Pinero A, Lujan JA, et al. Thoracoscopic partial pericardiectomy in the diagnosis and man-
agement of pericardial effusion. Surg Endosc 1997;11:253–256.
34. Hazelrigg SR, Mack MJ, Landreneau RJ, Acuff TE, Seifert PE, Auer JE. Thoracoscopic pericardiectomy
for effusive pericardial disease. Ann Thorac Surg 1993;56:792–795.
35. Trotter MC, Chung KC, Ochsner JL, McFadden PM. Pericardiectomy for pericardial constriction. Am
Surg 1996;62:304–307.
36. Copeland JG, Stinson EB, Griepp RB, Shumway NE. Surgical treatment of chronic constrictive peri-
carditis using cardiopulmonary bypass. J Thorac Cardiovasc Surg 1975;69:236–238.
37. McCaughan BC, Schaff HV, Pichler JM, et al. Early and late results of pericardiectomy for constrictive
pericarditis. J Thorac Cardiovasc Surg 1985;89:340–350.
38. Seifert FC, Miller DC, Oesterle SN, Oyer PE, Stinson EB, Shumway NE. Surgical treatment of constric-
tive pericarditis: analysis of outcome and diagnostic error. Circulation 1985;72:II264–II273.
39. Tuna IC, Danielson GK. Surgical management of pericardial diseases. Cardiol Clin 1990;8:683–696.
40. Young JM, Goldman IR. Tumor metastases to the heart. Circulation 1954;9:220–229.
41. Onuigbo, WIB. The spread of lung cancer to the heart, pericardium and great vessels. Jpn Heart J
1974;15:234–238.
42. Burke A, Virmani R. Primary cardiac sarcomas. In: Tumors of the Heart and Great Vessels. Atlas of
Tumor Pathology, Third series, Fascicle 16. Washington, DC: Armed Forces Institute of Pathology,
1996:127–170.
43. McFadden PM, Ochsner JL. Atrial replacement and tricuspid valve reconstruction after angiosarcoma
resection. Ann Thorac Surg 1997;64:1164–1166.
44. Dein JR, Frist WH, Stinson EB, et al. Primary cardiac neoplasms: early and late results of surgical
treatment in 42 patients. J Thorac Cardiovasc Surg 1987;93:502–511.
45. Burke A, Virmani R. Malignant mesothelioma of the pericardium. In: Tumors of the Heart and Great
Vessels. Atlas of Tumor Pathology, Third series, Fascicle 16. Washington, DC: Armed Forces Institute
of Pathology, 1996:181–194.
46. Risher WH, Rees AP, Ochsner JL, McFadden PM. Thoracoscopic resection of pericardium for symp-
tomatic congenital pericardial defect. Ann Thorac Surg 1993;56:1390–1391.
47. Baim RS, MacDonald IL, Wise DJ, Leukei SC. Computed tomography of absent left pericardium.
Radiology 1980;135:127–128.
48. Schiavone WA, O’Donnell JK. Congeintal absence of the left portion of parietal pericardium demon-
strated by nuclear magnetic response inaging. Am J Cardiol 1985;55:1439–1440.
49. Ochsner JL, Ochsner SF. Congenital cysts of the mediastinum: 20-year experience with 42 cases. Ann
Surg 1966;163:909–920.
50. Mack MJ, Aronoff RJ, Acuff TE, Douthit MB, Bowman RT, Ryan WH. Present role of thoracoscopy
in the diagnosis and treatment of diseases of the chest. Ann Thorac Surg 1992;54:403–409.
51. Michel JL, Revillon Y, Montupet P, Sauvat F, Sarnacki S, Sayegh N, N-Fekete C. Thoracoscopic
treatment of mediastinal cysts in children. J Pediatr Surg 1998;33:1745–1748.
52. Horita K, Sakao Y, Itoh T. Excision of recurrant pericardial cyst using video-assisted thoracic surgery.
Chest 1998;114:1203–1204.
53. Satur CM, Hsin MK, Dussek JE. Giant pericardial cysts. Ann Thorac Surg 1996;61:208–210.
54. Lewis RJ, Caccavale RJ, Sisler GE. Imaged thoracoscopic surgery: a new thoracic technique for resec-
tion of mediastinal cysts. Ann Thorac Surg 1992;53:318–320.
55. Naunheim KS, Andrus CH. Thoracoscopic drainage and resection of giant mediastinal cyst. Ann Thorac
Surg 1993;55:156–158.
56. Mack MJ, Acuff TE, Ryan WH. Implantable cardioverter defibrillator: the role of thoracoscopy. Ann
Thorac Surg 1993;56:739–740.
57. Robles R, Pinero A, Lujan JA, Parrilla P. Thoracoscopic implantation of an epicardial pacemaker.
Br J Surg 1996;83:400.
58. Furrer M, Fuhrer J, Altermatt HJ, et al. VATS-guided epicardial pacemaker implantation. Hand-sutured
fixation of atrioventricular leads in an experimental setting. Surg Endosc 1997;11:1167–1170.
59. Obadia JF, Kirkorian G, Rescigno G, el Farra M, Chassignoelle JF, Touboul P. Thoracoscopic approach
to implantable cardioverter defibrillator patch electrode implantation. Pacing Clin Electrophysiol
1996;19:955–959.
60. Elefteriades JA, Quin JA. Diaphragm pacing. Chest Surg Clin N Am 1998;8:331–357.
61. Weese-Mayer DE, Morrow AS, Brouillette RT, Ilbawi MN, Hunt CE. Diaphragm pacing in infants and
children: a life-table analysis of implanted components. Am Rev Respir Dis 1989;139:974–979.
62. Nye JD, Brown WT. Phrenic nerve stimulator. In WT Brown, ed. Atlas of Video-Assisted Thoracic
Surgery. Philadelphia: Saunders, 1994:274–276.
63. Chervin RD, Guilleminault C. Diaphragm pacing for respiratory insufficiency. J Clin Neurophysiol
1997;14:369–377.
Chapter 24 / Less Invasive Surgical Treatment 343
24 Less Invasive Surgical Treatment

of Atrial Fibrillation
Mathew R. Williams, MD
and Michael Argenziano, MD
CONTENTS
INTRODUCTION
ENABLING TECHNOLOGY
HYPOTHERMIC ABLATION
HYPERTHERMIC ABLATION
RADIOFREQUENCY
MICROWAVE
LASER
ULTRASOUND
NONTHERMAL ABLATION
CONCLUSIONS
REFERENCES
INTRODUCTION
Atrial fibrillation (AF) remains the most prevalent arrhythmia, with as many as 2.2
million cases in the United States alone (1), and it accounts for over 3.1 million office
visits a year (2). Despite often being labeled as a benign rhythm, AF plays a role in as
many as 15% of all strokes (3) and is associated with an increased odds ratio for death
(1.5 in men and 1.9 in women) (4). In general, AF is rarely treated successfully by the
medical community, with the mainstay of therapy being aimed at ventricular rate control
and anticoagulation. New understanding of AF mechanisms combined with technologi-
cal advances have led to a resurgence of anatomically based curative treatment of AF in
both the catheter-based and the surgical arenas.

343
Surgical treatment of AF began in 1980 with the introduction of the left atrial isolation
procedure (5), which was successful at restoring sinus rhythm in the right atrium with
subsequent conversion of both a normal ventricular response and improved hemodynam-
ics. This approach was abandoned because the left atrium continued to fibrillate and thus
the theoretical risk for stasis of blood in the fibrillating left atrium persisted, with a
subsequent stroke risk. In 1985 Guiradon introduced the corridor procedure that created
an isolated discrete pathway—a corridor—between the sinoatrial node and the atrioven-
tricular node to restore a normal rate (6). This procedure did not receive much acceptance
due to both limited success and the fact that both atria continued to fibrillate with unal-
tered hemodynamics. In 1986 Cox performed his first clinical attempt at treating AF with
the atrial transection procedure, which was initially promising but ultimately failed (7).
This failure lead Dr. Cox and his colleagues to better characterize the electrophysiologi-
cal basis for AF (8), which eventually resulted in the introduction of the Maze Procedure
(9). Based on clinical results, the procedure underwent revisions to become the Cox-
Maze III procedure (10), regarded today as the gold standard surgical treatment for AF.
The Maze III has been applied by Dr. Cox and others (11,12) with success rates as high
as 98%, though not all groups have been able to duplicate these results (13). Despite
excellent results, the procedure has not been applied widely. Reasons for this reluctance
are multifactorial but include the invasive nature of the procedure with the need for
cardiopulmonary bypass, multiple atriotomies with their associated bleeding risk, and the
complex basis of the procedure. These concerns as well as new understandings of AF
mechanisms have spawned attempts at less invasive surgical cure of AF. In assessing any
new curative treatment of AF it is important that the determination of a successful pro-
cedure be well defined. In particular, a procedure to cure AF should not only achieve sinus
rhythm but also restore atrial contractility with its associated improvement in hemody-
namics and decreased incidence of thromboembolic complications. Finally, a less inva-
sive cure must also address the problems with previous approaches with an acceptable
success rate.
The transition to a less invasive approach for curing AF will involve using alternative
modalities in lieu of the “cut and sew” technique to create lesions and will ideally be
applied from the epicardial surface on the beating heart. It will also undergo modification
to be applied through smaller incisions, perhaps with the aid of thoracoscopic or robotic
devices. Finally, while subject to much debate, there will likely be an alteration of the
lesion set to simplify it from the Maze III. This chapter will discuss the current and
forthcoming devices, approaches, and lesion sets that will facilitate the transition of AF
surgery to a minimally invasive cure.
ENABLING TECHNOLOGY
In an effort to reduce bleeding, reduce operative time, and ultimately perform ablations
from the epicardial surface, multiple technologies have been created to ablate atrial
tissue. These technologies can be broken into three categories: hypothermic sources,
hyperthermic sources, and nonthermal sources (Table 1). These energy sources can be
applied from either the epicardium or the endocardium, and each approach is associated
with very different physics.
Surgical approaches from the endocardium generally are on the arrested and opened
heart, and successful ablation is easier to accomplish due to the lack of the convective
heating or cooling effects of flowing blood. The primary impeding factors to ablating
tissue are conductive and convective thermal alterations from ambient air, body tempera-
Table 1
Energy Sources for Atrial Ablation
Mechanism Energy Source Company FDA Approval Status
Hypothermic
Cryo EndoCare, Cryocath, Treatment of arrhythmias
Frigitronics
Hyperthermic
Radiofrequency
Unipolar Boston Scientific Soft tissue ablation
Bipolar Atricure Soft tissue ablation
Irrigated Medtronic, Osypka Pending
Microwave AFx Cardiac ablation
Laser Cardiofocus/Edwards Cardiac ablation
USG Spectrasonics, Epicor Experimental
Nonthermal
Chemical Microheart Experimental
Fig. 1. Trichrome slide of canine atrial tissue obtained 4 wk after unipolar RF ablation. The lesion
created is dome-shaped and relatively wide, though it falls short of transmural near the endocardium.
ture, tissue thickness, and perhaps remaining stagnant blood in the atrium, but these are
generally easily overcome. Other impeding factors include the degree of contact at the
tissue–device interface, intrinsic tissue thermal conductivity (which can be affected by
increased deposition of fat or fibrosis in the myocardium), and acquired tissue thermal
conductivity alterations such as char formation.
Surgical approaches from the epicardium result in very different ablative physics,
since the goal of these lesions is to perform them on the beating heart. In this setting, the
flowing blood inside the heart acts as a near-infinite thermal sink, making it difficult to
achieve appropriate ablating temperatures at the endocardium while keeping the epicar-
dial temperatures in a safe range (Fig. 1). Other limiting factors include the thermal
insulation and added thickness from epicardial fat, conduction to surrounding structures,
and limited exposure. In some cases the addition of all these factors can make epicardial
ablation impossible.
HYPOTHERMIC ABLATION
Cryoablation has the longest history of surgical atrial ablation and is an important
component in Dr. Cox’s Maze procedures. Cryo is currently the only source that has an
arrhythmia-specific indication from the Food and Drug Administration (FDA) in the
United States. Cryoablation involves three stages: freeze/thaw, hemorrhagic and inflam-
matory, and replacement fibrosis (14). The mechanism of cell injury involves mitochon-
drial and other organelle destruction from ice crystal formation during the first stage and
further damage from edema with subsequent necrosis during the second stage. For suc-
cessful ablation to occur, the tissue must reach at least –55°C for 2 min, and in some cases
repeat ablations must be performed. Other methods of optimizing ablation include
increasing the probe size, increasing the ablation time, increasing the number of freeze/
thaw cycles, increasing the tank pressure, or changing the gas to allow colder tempera-
tures to be reached. Current modifications have been aimed at creating flexible linear
probes for use in epicardial and endocardial ablation. At this time there is a lack of
evidence to demonstrate the feasibility of epicardial cryoablation, though it is expected
that the warming effect of the endocardial blood will make successful ablation
extremely difficult.
Cryoablation has extensive clinical use and has now even been used by some to
complete an entire Maze procedure (15). In general, cyroablation has an excellent clinical
safety record, though its use in AF surgery has typically been reserved for creating spot
lesions over the tricuspid and mitral valve annuli. This practice has resulted in a few cases
of coronary artery stenosis, though this is a rare complication (16).
Owing to its generally safe record and arrhythmia-specific indication, cryoablation
will continue to play a role in endocardial ablation. Its future in minimally invasive
approaches has yet to be determined, and although it is an attractive energy source, its
bulky nature and limited flexibility may limit its widespread use.
HYPERTHERMIC ABLATION
The majority of available energy sources rely on heating tissue to create ablation. The
goal of a hyperthermic lesion is to obtain a tissue temperature of 50°C, which has been
shown to cause irreversible loss of cellular excitability (17). It is important, however, to
ensure that the tissue does not achieve temperatures of above 100°C, since that will cause
water to boil with a risk for cavitation or potential wall disruption. The mechanism for
cell injury is not entirely understood but likely results from injury to the sarcolemmal
membrane with subsequent calcium overload of the cytosol.
The available hyperthermic sources span the electromagnetic spectrum from the
massive radiofrequency waves to fine near-infrared laser. Each behaves differently in
terms of tissue penetration and the mechanism by which heat is produced, but ultimately
the end result is simply to heat the tissue to the point of cell death.
RADIOFREQUENCY
Radiofrequency (RF) ablation has been available for many years, though it has not
been used extensively by surgeons until recently. RF is the source most familiar to
Table 2
Comparison of Different Types of Radiofrequency
Type Advantages Disadvantages
Unipolar
Most flexible Poor epicardial
Generally safe Comparitvely longer lesion times
Good clinical history Possibility for collateral injury
Less expensive Some char formation
Long lesions Inefficient
Irrigated
More efficient Less flexible
Good clinical history Nonspecific ablation parameters
Less expensive ? Epicardial
Prevents char formation Possibility for collateral injury
Irrigation required
Bipolar
Most efficient No flexibility
No collateral injury Little clinical use
Effective from epicardium More expensive
Feedback mechanism
cardiologists performing catheter-based arrhythmia treatments. RF is essentially

unmodulated alternating current at 500–1000 kHz, which is not dissimilar to conven-
tional electrocautery. The frequency of RF is so rapid that it is unable to capture cardiac
tissue and thus there are not problems with stimulating the heart or inducing fibrillation.
The mechanism of heat generation with RF is termed resistive or ohmic heating. Essen-
tially, the RF energy is emitted from the probe over a very small area and thus has a high
current density. This high current encounters the tissue, which acts as a resistor, and heat
is generated. The true resistive heating occurs only around 1 mm deep into the tissue and
the remainder of ablation occurs via conductive heating from the area of resistive heating.
This results in a relatively inefficient heating process, yet it provides for well-controlled
and generally safe ablation. In reality these lesions rarely take more than 1 min to create.
Although there are multiple modifications of RF, there are three general methods:
unipolar, bipolar, and irrigated. Each has its own relative advantages and disadvantages
(Table 2).
Unipolar ablation relies on grounding pads to act as the other pole and is the simplest
way to apply the energy. With unipolar ablation, the energy is focused at the ablating
surface (highest current density) and disperses throughout the body to exit the ground.
This is the slowest and most inefficient of the RF modalities but is also the most controlled
method. It can be applied using two different methods: energy control and temperature
control. Energy control regulates the amount of power delivered and does not monitor the
temperature. Temperature control relies on setting a goal temperature and the RF energy
is regulated such that an appropriate power is provided to maintain the goal temperature.
The current unipolar surgical device uses temperature-control ablation. The ideal goal
temperature is not entirely clear. For endocardial ablation, experimental data have dem-
onstrated reliable and effective ablation if ablation is performed for 60 s at 70°C. How-
ever, because there is greater variability in humans (particularly those with multiple
pathologies), we and many others use 80°C for 60 s. It is also important to note that the
hottest temperature will actually be achieved just below the surface. As a result, the goal
temperature is usually a few degrees cooler than the hottest tissue temperature and there-
fore the goal temperature should never be set at more than 95°C to avoid potential tissue
disruption. Ablations must be performed for at least 60 s, because a steady state is not
achieved until 40–50 s of energy application.
The use of standard RF from the epicardial surface is unfortunately not as simple as
from the endocardium. Because the primary method of ablation with RF is through
conductive heating, it is more prone to energy losses from other variables. In the case of
epicardial ablation, the convective cooling of the endocardial blood makes transmural
ablation nearly impossible in tissue greater thicker than 4 mm, though the flow dynamics
are probably more important than tissue thickness. This is particularly true if epicardial
fat is present, because the fat serves as an excellent insulator of RF energy. In general,
epicardial RF lesions will become wide but will fall just short of transmurality on the
endocardial side (Fig. 1). It is important to note, as will be discussed later, that lesions
may not need to be transmural in order to be electrically effective.
Some of the inefficiencies of unipolar ablation can be overcome by irrigating the active
electrode with a saline solution. Irrigated or “cooled tip” RF ablation improves efficiency
by various means. The cooling effect on the surface of the tissue actually drives the focus
(hottest point) of energy deeper into the tissue, providing for both faster and deeper
ablation. The irrigation also prevents the accumulation of char on the ablating surface,
as can occur with standard unipolar ablation. This char can serve as an insulator and
prevent the creation of an optimal lesion. With irrigated designs (Fig. 2), it is important
that the entire ablating element be uniformly irrigated and thus longer or flexible devices
are more difficult to create.
With both irrigated and standard RF, the energy disperses from a single element and
there is the possibility to ablate adjacent noncardiac tissue such as the esophagus. Adja-
cent damage is generally avoided provided appropriate ablation parameters are followed.
Bipolar RF is another modality that has the ability to make very fast and discrete
lesions. This modality relies simply on having a pole on each side of the tissue to be ablated.
This focuses all of the energy between the two poles, and lesions can be made in less than
10 s. The current bipolar product also has impedance sensors that detect when transmural
ablation has occurred. The bipolar application coupled to the sensing mechanism pro-
vides for reliable transmural and electrophysiologically effective lesions. These lesions
can also be created from the epicardium, provided the lesion is in an area of the heart that
can be opposed to itself or one of the poles can be inserted into the heart through a purse
string. Since both poles must be perfectly opposed to each other, there is very limited
flexibility with the device.
MICROWAVE
Microwave energy has been used successfully in Europe and recently became
available in the United States. Microwave ablation can be performed at either 915 or
2450 MHz, which are the frequencies allowed by the Federal Communications Commis-
sion (FCC) for medical microwave use. The current device uses 2450 MHz and works
through dielectric heating. Essentially, the microwave causes oscillation of dipoles (which
happen to be water molecules), producing kinetic energy and resulting heat generation.
Not all of the ablation occurs through dielectric effects, and there is a component of
conductive heating.
Fig. 2. Example of radiofrequency ablation systems. (A) The Cobra Ablation system, a unipolar,
mutiple-element device that allows creation of linear RF lesions up to 9 cm in length. (Courtesy
EPT, San Jose, CA.) (B) The Cardioblate system, a unipolar, irrigated design. The tip is continu-
ously irrigated and the device is used like a pen to slowly draw the lesion. (Courtesy Medtronic
Corp., Minneapolis, MN.)
Fig. 3. Example of the microwave ablation system. (Top) Flex 10 device: the antennae sit within
a sheath that is placed on the outside of the heart. The device can be placed using minimally
invasive approaches. (Middle) Flex 4 device: used for both endocardial and epicardial ablation.
Consists of a linear and malleable unidirectional ablating element 4 cm in length. (Bottom) Flex
2 device: a rigid 2-cm-long ablating device used predominantly in endocardial ablation. (Courtesy
AFx, Fremont, CA.)
Microwave has the advantage of creating deeper lesions in a shorter amount of time
than RF, though this difference is partially negated by the shorter lesion length. In
benchtop work using 40 W for 25 s, microwave reliably ablates at 6 mm depth. Thermal
profiles of these lesions demonstrated efficient and uniform penetration depth without
areas of overheating and thus no char formation (18). This effectiveness has been cor-
roborated by good clinical experience.
There are currently three types of devices available, all made by the same company.
One is intended primarily for endocardial ablation, and while the handle is flexible, the
ablating element is not. The second is longer and flexible and can be used for both
epicardial and endocardial application. The third device is modified for minimally inva-
sive application on the beating heart. This device has been applied successfully in our
laboratory on animals using only ports and a minithoracotomy. Recent clinical work also
shows it to be a valuable device in the minimally invasive arena (Fig. 3).
Because of its better penetration, microwave energy has more potential to be success-
ful for epicardial ablation, and the early clinical experience has been promising.
LASER
Laser energy has recently received FDA approval for cardiac ablation in the United
States. The primary enabling technology for laser ablation is the fiber-optic delivery
device rather than the laser itself. The delivery device has a diffusing tip that contains
silicon particles in the distal part. The silicon causes the laser to be emitted perpendicular
to the fiber direction. A gold foil is then placed halfway around the circumference of the
diffusing tip, which reflects the energy back in the other direction. The end result is
unidirectional linear ablation of 2–5 cm with a flexible configuration. The mechanism of
laser ablation is wavelength-dependent but works predominantly by inducing a harmonic
Fig. 4. The Optimaze Ablation System. The device consists of a 980-nm diode laser and a diffus-
ing-tip hand piece that permits linear ablation up to 5 cm in length. (Courtesy Edwards Lifesciences,
Irvine, CA.)
oscillation in water molecules, with resulting kinetic energy and heat generation. The
currently used wavelength for cardiac ablation utilizes a 980-nm diode laser. This wave-
length was chosen because of its good penetration of cardiac tissue with minimal pigment
absorption. This wavelength reliably ablates tissue with absorption of actual laser energy
as deep as 4 mm into the tissue, with further ablation occurring via conductive heating
mechanisms. The amount of heat absorbed by the tissue and the ablation depth can be
altered by changing the power and/or the wavelength of the laser, and this may be advan-
tageous in optimizing epicardial ablation, especially through epicardial fat. Preliminary
work in animals with epicardial ablation is very promising, though more data are required
prior to reaching any conclusions.
In preclinical work in our laboratory we have been able to create 100% transmural
lesions in canines from an endocardial approach. These lesions were found to be electro-
physiologically effective when subjected to pacing (19). This technology has now been
applied clinically, with encouraging early results. With the current clinical device,
lesions are created in 36 s for a length of 5 cm (Fig. 4).
ULTRASOUND
Ultrasound ablation is currently a preclinical modality in the cardiac surgical field,
though it has been used clinically in other surgical applications and by catheters in the
electrophysiology laboratory. The mechanism for ultrasound ablation is mechanical
hyperthermia. The ultrasound wave is emitted from the transducer and the resulting wave
travels through the tissue causing compression, refraction, and particle movement,
resulting in kinetic energy and heat. Because the mechanism is mechanical, if the wave
is applied too aggressively, shear stress and tissue disruption can occur. However,
because the wave can be controlled via a number of different parameters, it is simple to
regulate and to avoid this complication. Ultrasound can be applied in either a focused
(HIFU, high-intensity focused ultrasound) or nonfocused manner. HIFU allows for rapid,
high-concentration energy in a confined space, and in our experience, produces transmu-
ral epicardial lesions through epicardial fat in less than 2 s (20). Nonfocused ultrasound
is a slower process, but the transducer is simpler to create and may have more flexibility.
A potential advantage of ultrasound (though not yet studied extensively) will be the
ability to image lesions. The transducer can be used both to image and ablate and thus it
may be possible, particularly with HIFU, to determine atrial wall thickness, set the focus
to ablate the appropriate amount of tissue, and then confirm that the ablation is transmural.
NONTHERMAL ABLATION
Not all ablation modalities involve thermal ablation. The most widely applied method,
utilizing the “cut and sew” technique, obviously is purely mechanical, though, as men-
tioned previously, it is limited by the inability to perform epicardial ablation, increased
bleeding risk, and longer operative time.
Chemical ablation provides for a possible method of creating cardiac lesions. It relies
on the idea of toxic chemicals inducing cell death, with subsequent formation of scar
tissue. Chemical ablation has been used in the catheter laboratory for ablation of ventricu-
lar arrhythmias by infusing alcohol selectively into a coronary artery that feeds the
arrhythmogenic area (21). This principle could not be used in the atria due to the widely
varied and collateralized blood supply. Thus, for atrial ablation the toxic chemicals must
be placed directly into the atrial wall. Our attempts at doing this with a needle have been
variably successful, and it has been difficult to localize the ablation. Utilization of needle-
free jet injection technology has overcome this problem and allows well-controlled and
very rapid transmural ablation in a linear fashion. The needle-free technology essentially
propels the chemical out a small orifice at a high velocity, resulting in penetration of the
epicardium. If the correct pressure parameters are used, the chemical jet rapidly loses its
velocity upon entering the tissue, distributes relatively evenly throughout the atrial wall,
and creates a transmural lesion. While much of these data are still preliminary, we have
been able to perform linear transmural ablation from the epicardium in animals, and have
found these to be electrophysiologically effective. Ethanol was used as the chemical and
the ablation time per each lesion (2-cm length) was a fraction of a second. Further work,
primarily in survival models, is required to determine if chemical ablation will be a useful
modality, but it has several advantages because it does not depend on thermal mecha-
nisms, ablates regardless of tissue type (i.e., is not hindered by epicardial fat), and is a
relatively inexpensive technology.
MINIMALLY INVASIVE APPROACHES AND LESION

SETS FOR SURGICAL TREATMENT OF AF
As with all surgery, calling a procedure “minimally invasive” can be a subjective
matter. The same is true for surgical treatment of atrial fibrillation. Many authors have
described techniques and approaches, all of which are in a sense less invasive ways of
treating AF, compared to the traditional technique of the Maze III. When evaluating such
techniques, it is important to understand that a less invasive technique may also be a less
effective technique. In terms of AF surgery, a less invasive operation may simply involve
utilizing one of the above-described energy sources to create Maze III lesions or to reduce
the number of lesions. Taken to an extreme, it might involve creating fewer lesions, with
an alternative energy source, on the beating heart with endoscopic techniques.
Many authors have presented clinical data with alternative energy modalities, with
generally similar results. This includes authors performing the lesions outlined by the
Maze III using an alternative energy source. There are two recent publications in which
authors have substituted radiofrequency energy for “cut and sew” lesions and essentially
followed the lesions described by Dr. Cox’s Maze III. Both of these groups used an
irrigated-design catheter for the ablations. One of the reports is a randomized study
comparing the addition of the AF procedure to no AF procedure in patients having
concurrent mitral valve surgery. The study found that the group having the RF–Maze III
in addition to mitral valve surgery had an 80% success in achieving sinus rhythm com-
pared to 27% in those having only mitral valve surgery (22). The other study involved 122
patients and achieved a success of 78.5% in obtaining sinus rhythm (23).
More recently, many investigators have been limiting their approaches to the left
atrium, driven largely by data suggesting that focal sources of automaticity originate in
the pulmonary veins (24) and that microrotors of reentry anchor themselves near the
pulmonary veins and produce fibrillatory conduction (25). Thus, isolating the pulmonary
veins and the posterior left atrium may be sufficient to cure AF. There are many different
lesion sets proposed to achieve this isolation (Fig. 5), and at the present time there is no
way of determining which, if any, is superior. There are multiple active debates as to
which lesions if any are absolutely necessary, though all agree that the pulmonary veins
must be addressed in some manner. The predominant controversies include the necessity
for the mitral valve lesion and the necessity of removing or isolating the left atrial append-
age. The mitral annular lesion is felt by Dr. Cox to be absolutely necessary to control atrial
fibrillation, while others feel the lesion is important only to prevent a left atrial flutter,
which can be addressed by other lesions such as the one used to remove left atrial
appendage. Ideally this lesion will not be necessary, as it will be a difficult lesion to
perform on the beating heart from the epicardial surface, since it will require ablating
directly over the circumflex artery and coronary sinus. However, if the lesion is found to
be necessary, it may require either placement of a device in the atrium through a purse
string or the assistance of our interventional cardiology colleagues. The left atrial
appendage is felt by some to play a role in the origin and initiation of AF, though this is
not a commonly held view. Its importance likely relies in the fact that most documented
intracardiac thrombi in patients with AF are found in the left atrial appendage. For this
reason, removing or at least anatomically excluding the left atrial appendage may reduce
the stroke rate even if the patient remains in AF. It may be in the future that an ideal lesion
Fig. 5. Examples of some of the advocated left atrial lesion sets. (A) The lesion set currently
advocated by our group. It involves complete encirclement of the posterior left atrium containing
all the pulmonary veins. The left atrial appendage is also either removed or electrically and
anatomically isolated. A lesion is then made from the pulmonary vein lesion to the left atrial
appendage lesion. This lesion set can easily be accomplished from the epicardium with minimal
dissection. (B) The initial procedure described by Melo, in which only the left and right veins are
isolated. This approach has a theoretical advantage in that the posterior left atrium is preserved to
contribute to left atrial contractility. This approach can also be performed from the epicardium. (C)
The approach advocated by Benussi. Part of this lesion set is currently performed from the epicar-
dium and part from the endocardium. The left atrial appendage is also addressed. A lesion in also
performed to the mitral valve annulus, which may be difficult from the epicardial surface. (D) The
approach advocated by Knaut and several other European investigators. This lesion set has been
termed the “7.” In most groups’ experience the left atrial appendage is not addressed, though it can
be easily accomplished if desired. This lesion set is interesting in that it does not actually isolate
the pulmonary veins but probably disrupts microrotors of reentry. This lesion set shares similar
success with those that actually isolate the pulmonary veins. A lesion to the mitral annulus is
required and thus epicardial performance is limited.
set will be patient-specific, based on intraoperative variables that are as yet poorly under-
stood. Another possibility will be a multispecialty approach in which the cardiac surgeon
creates the left-sided lesions and, if unsuccessful, the cardiologist follows up by creating
right-sided lesions.
The clinical results of left-sided-only ablations are varied, but again the definition of
success is different in each series. Some centers require bi-atrial contraction before
success is achieved, whereas others rely solely on electrophysiological data. These
approaches have been performed predominantly with RF energy, though microwave,
cryo, and “cut and sew” techniques have also been utilized (26–30). The reported success
rates in these studies vary between 70% and 90%, with most centers reporting rates
between 75% and 80%. Comparison among these studies is rather difficult due to varia-
tions in lesion sets, baseline patient characteristics, energy sources, and the type and
number of concomitant procedures. The great majority of the patients in all of these
studies had concurrent mitral valve surgery. This is promising for the success of a limited
left atrial procedure for a couple of reasons. First, it is likely that patients with mitral valve
disease will be more difficult to treat due to concurrent atrial pathology and the general
state of health of the patients. Additionally, the results for left atrial procedures during
mitral valve surgery are quite similar to the largest reported series of Maze III patients
during concurrent mitral valve surgery, with a success rate of 76% (31), suggesting that
a full Maze III may not be of additional benefit.
While the results of these studies are extremely promising, the success of surgical
treatment of AF will rely on the ability to create lesions from the epicardium and to that
end there have been several investigators pursing this goal. On the clinical side there is
growing experience with epicardial ablation, though most investigators have too little
experience to report meaningful results. The two largest series of epicardial ablation have
been with RF, and both still utilized cardiopulmonary bypass. They demonstrated success
rates of 76.9% and 95%, though the latter also utilized right atrial lesions (32,33). Another
series of isolated pulmonary vein ablation on the beating heart without mitral valve
surgery had extremely promising results, suggesting that limited procedures may be even
more successful in patients without valvular disease (34). There have been anecdotal
reports of successful beating-heart epicardial microwave and bipolar RF ablations, though
at this time there have been no published series.
The epicardial field has advanced more on the preclinical front and there should be
more widespread clinical application in the upcoming year. In our laboratory, we have
achieved 100% electrical isolation of the posterior left atrium (containing all pulmonary
veins) in a beating-heart canine model. This approach utilized a specialized microwave
device and was performed using a 3-cm thoracotomy and three thoracoscopy ports in
the left chest that allowed removal of the left atrial appendage with a stapling device (35). We
have also developed a totally endoscopic version of this operation, using robotic technology,
which we have performed in a few patients with lone AF.
Another important issue that has come out of epicardial ablation is the question of
whether lesions need to be transmural to be effective. Most surgeons feel a lesion must
be transmural to be effective, and indeed this is probably a reasonable goal. However,
there is increasing data from both surgical and catheter-based investigations suggesting
that lesions may not need to be transmural to be effective. Perhaps, then, discussions
about lesion quality should be based on electrophysiological effectiveness rather than
just transmurality. In reality, most (though not all) transmural lesions are electrophysi-
ologically effective, but many nontransmural lesions are also effective (35,36). The
reasons for this are uncertain but may be related to the lesion width. It may also be that
simply inducing a conduction delay may be sufficient to prevent or disrupt maintenance

of AF.
CONCLUSIONS
Surgical treatment of AF has received renewed attention. This is due partly to the
realization that AF is a morbid arrhythmia and one that is poorly managed by traditional
medical therapy. Interest in these procedures has been further fueled by the development
of new energy sources, minimally invasive techniques, and an understanding of AF
mechanisms. The future of the field has yet to be realized but ultimately will involve a
procedure on the beating heart from the epicardial surface utilizing endoscopic tech-
niques. Indeed, surgical treatment for AF may provide a definitive cure for many of the
thousands of patients with AF and at the same time represent the first cardiac operation
that may be performed as an outpatient procedure.
REFERENCES
1. Feinberg WM, Blackshear JL, Laupacis A, Kronmal R, Hart RG. Prevalence, age distribution, and
gender of patients with atrial fibrillation; analysis and implications. Arch Intern Med 1995;155:469–473.
2. Stafford RS, Singer DE. National patterns of warfarin use in atrial fibrillation. Arch Intern Med
1996;156:2537–2541.
3. American Heart Association. 1998 Heart and Stroke Statistical Update. American Heart Association, 1997.
4. Gasjewski J, Singer RB. Mortality in an insured population with atrial fibrillation. JAMA
1981;245:1540–1544.
5. Williams JM, Ungerleider RM, Logland GK, et al. Left atrial isolation: new technique for the treatment
of supraventricular arrhythmias. J Thorac Cardiovasc Surg 1980;80:373–380.
6. Guiradon GM, Campbell CS, Jones DL, et al. Combined sino-atrial node atrio-ventircular node isola-
tion: a surgical alternative to His bundle ablation in patients with atrial fibrillation. Circulation
1985;72:III–220.
7. Cox JL, Schuessler RB, Boineau JP. The development of the maze procedure for the treatment of atrial
fibrillation. Sem Thorac Cardiovasc Surg 2000;12:2–14.
8. Cox JL, Canavan TE, Schuessler RB, et al. The surgical tretment of atrial fibrillation: II. Intraoperative
electrophysiologic mapping and description of the electrophysiologic basis of atrial flutter and atrial
fibrillation. J Thorac Cardiovasc Surg 1991;101:406–426.
9. Cox JL. The surgical treatment of atrial fibrillation: IV. Surgical Technique. J Thorac Cardiovasc Surg
1991;101:584–592.
10. Cox JL, Jaquiss RD, Schuessler RB, et al. Modification of the maze procedure for atrial flutter and
fibrillation. II. Surgical technique of the Maze III procedure. J Thorac Cardiovasc Surg 1995;110:485–495.
11. Cox JL, Schuessler RB, Lappas, DG, et al. An 8-1/2 year clinical experience with surgery for atrial
fibrillation. Ann Surg 1996;224:267–275.
12. McCarthy PM, Gillinov AM, Castle L, Chung M, Cosgrove D. The Cox-Maze Procedure: the Cleveland
Clinic experience. Semin Thorac Cardiovasc Surg 2000;12:25–29.
13. Kosakai Y. Treatment of atrial fibrillation using the Maze Procedure: the Japanese experience. Semin
14. Lustgarten DL, Keane D, Ruskin J. Cryothermal ablation: mechanism of tissue injury and current
experience in the treatment of tachyarrhythmias. Prog Cardiovasc Dis 1999;41:481–498.
15. Cox JL, Ad N, Palazzzo T, et al. Current status of the Maze Procedure for the treatment of atrial
fibrillation. Semin Thorac Cardiovasc Surg 2000;12:15–19.
16. Holman WL, Ikeshita M, Ungerleider RM, Smith PK, Ideker RE, Cox JL. Cryosurgery for cardiac
arrhythmias: acute and chronic effects on coronary arteries. Am J Cardiol 1983;51:149–155.
17. Nath S, Lynch C, Whayne JG, Haines DE. Cellular electrophysiological effects of hyperthermia on
isolated guinea pig papillary muscle. Implications for catheter ablation. Circulation 1993;88:1826–1831.
18. Williams MR, Knaut M, Berube D, Oz MC. Application of microwave energy in cardiac tissue ablation:
from in vitro analyses to clinical use. Ann Thorac Surg 2002;74:1500–1505.
19. Williams MR, Casher J, Klein KA, et al. Linear atrial ablation using laser energy. Submitted.
20. Williams MR, Kourpanidis S, Casher J, et al. Epicardial atrial ablation with high intensity focused
ultrasound on the beating heart. Circulation 2001;104(17):II-409.
21. Weismuller P, Mayer U, Richter P, et al. Chemical ablation by subendocardial injection of ethanol via
catheter—preliminary results in the pig heart. Eur Heart J 1991;12(11):1234–1239.
22. Khargi K, Deneke T, Haardt H, et al. Saline-irrigated, cooled-tip radiofrequency ablation is an effective
technique to perform the maze procedure. Ann Thorac Surg 2001;72:S1090–S1095.
23. Sie HT, Beukema WP, Misier AR, et al. Radiofrequency modified Maze in patients with atrial fibrilla-
tion undergoing concomitant cardiac surgery. J Thorac Cardiovasc Surg 2001;122:249–256.
24. Haissaguerre M, Jais P, Shah DC, et al. Spontaneous initiation of atrial fibrillation by ectopic beats
originating in the pulmonary veins. N Engl J Med 1998;339:659–666.
25. Jalife J, Berenfeld O, Skanes A, Mandapati R. Mechanisms of atrial fibrillation: mother rotors or
multiple daughter wavelets, or both? J Cardiovasc Electrophysiol 1998;9:S2–S12.
26. Sueda T, Nagata H, Shikata H, et al. Simple left atrial procedure for chronic atrial fibrillation associated
with mitral valve disease. Ann Thorac Surg 1996;62:1796–1800.
27. Graffigna A, Pagani F, Minzioni G, Salerno J, Vigano M. Left atrial isolation associated with mitral
valve operations. Ann Thorac Surg 1992;54:1093–1097.
28. Kottkamp H, Kindricks G, Hammel D, et al. Intraoperative radiofrequency ablation of chronic atrial
fibrillation: a left atrial curative approach by elimination of anatomic “anchor” reentrant circuits.
J Cardiovasc Electrophysiol 1999;10:772–780.
29. Williams MR, Stewart JR, Bolling SF, et al. Surgical treatment of atrial fibrillation using radiofrequency
energy. Ann Thorac Surg 2001;71:1939–1944.
30. Knaut M, Spitzer SG, Karolyi L, et al. Intraoperative microwave ablation for curative treatment of atrial
fibrillation in open heart surgery—the MICRO-STAF and MICRO-PASS pilot trial. Microwave appli-
cation in surgical treatment of atrial fibrillation. Microwave application for the treatment of atrial
fibrillation in bypass-surgery. Thorac Cardiovasc Surg 1999;47(S3):379–384.
31. Kosakai Y. Treatment of atrial fibrillation using the Maze Procedure: the Japanese experience. Semin
32. Benussi S, Pappone C, Nascimbene S, et al. A simple way to treat chronic atrial fibrillation during mitral
vavle surgery: the epicardial radiofrequency approach. Eur J Cardiothorac Surg 2000;17:524–529.
33. Raman JS, Seevanayagam S, Storer M, Power JM. Combined endocardial and epicardial radiofrequency
ablation of right and left atria in the treatment of atrial fibrillation. Ann Thorac Surg 2001;72:S1096–S1099.
34. Williams MR, Garrido M, Casher JM, et al. A minimally invasive beating heart approach to pulmonary
vein isolation for the treatment of atrial fibrillation. Proceedings of the Cardiovascular Therapeutics and
Technologies Meeting, Miami Beach, FL, 2001.
35. Taylor GW, Walcott GP, Hall JA, et al. High-resolution and histologic examination of long
radiofrequency lesions in canine atria. J Cardiovasc Electrophysiol 1999;10:1467–1477.
36. Williams MR, JA Sanchez, Barbone A, et al. Simultaneous endocardial contact activation mapping
and beating heart epicardial radiofrequency pulmonary vein isolation. Circulation 2000;102(18):
II-443–II-444.
Chapter 25 / The Economic Impact of Minimally Invasive Cardiac Surgery 359
V MISCELLANEOUS ASPECTS
OF MINIMALLY INVASIVE CARDIAC SURGERY
25 The Economic Impact

of Minimally Invasive Cardiac Surgery
Aftab R. Kherani, MD, Elizabeth H. Burton, BA,

and Mehmet C. Oz, MD
CONTENTS
INTRODUCTION
MINIMALLY INVASIVE VALVE SURGERY
MINIMALLY INVASIVE CORONARY ARTERY BYPASS GRAFTING
CONCLUSIONS
REFERENCES
INTRODUCTION
As the focus of cardiac surgery moves from simple mortality endpoints to more qual-
ity-of-life metrics, the importance of minimally invasive surgery has grown. Yet to assess
the true value that these innovations bring to people, we must determine the cost efficacy
of the technology. Proponents of this technology claim that these advanced technologies
translate to decreased cost and decreased morbidity. Numerous studies have been con-
ducted in an attempt to clarify this issue.
MINIMALLY INVASIVE VALVE SURGERY

In 1997, Cohn and colleagues described their experience with minimally invasive
valve procedures at Brigham and Women’s Hospital. Their study included a comparison
of charges associated with minimally invasive techniques and traditional sternotomy.
They found that the former was associated with approx 20% lower hospital charges.
Additionally, these patients returned to work nearly 2 wk earlier, translating to further
economic benefit on a societal scale. Their pain medication requirement was less, and
they returned to normal activity in less than half the time (1).

361
362 Part V / Miscellaneous Aspects of MICS
Cosgrove and colleagues recently compared their minimally invasive valve experi-
ence to the traditional median sternotomy approach at the Cleveland Clinic Foundation.
They compared their observations of the first 40 patients to undergo aortic valve proce-
dures via a right parasternal incision with 40 patients who underwent the standard median
sternotomy. They found that postoperative length of stay was 5.8 ± 3.0 d compared to 7.8
± 6.5 d, p < 0.001. This, in part, contributed to a 19% reduction in direct costs. This group
similarly examined mitral valve procedures. Here, no statistical significance was seen in
postoperative length of stay, while direct costs were 7% lower for the minimally invasive
group (2).
Comparisons among the various types of minimally invasive procedures have also
been made. Early data suggest that port-access valve surgery via a small anterior thora-
cotomy is associated with statistically significantly longer operating room time. The
difference was nearly 2 h, theoretically adding an extra $1500 to $3000 in cost per case.
The charges associated with the disposable Endo CPB System further adds to the cost
difference between the two methods (3). Inferences can certainly be made with data from
this study as well as the others described; however, to address definitively the economic
question surrounding different approaches to valve procedures requires a prospective,
controlled, randomized studying comparing the various minimally invasive approaches
to each other as well as to the traditional median sternotomy. Cost should be among the
variables analyzed.
MINIMALLY INVASIVE CORONARY ARTERY BYPASS GRAFTING

In recent years, several groups have compared minimally invasive direct coronary
artery bypass grafting (MIDCABG) and full sternotomy off-pump methods (OPCAB) to
traditional on-pump coronary artery bypass grafting (CABG). One factor studied by
many of these institutions is how costs compare. The issue of cost in this area is of great
interest especially in the United States, where Medicare alone spends roughly $4 billion
annually on coronary artery bypass grafting procedures (4). In this field, there does exist
a prospective randomized study examining the economic outcome of off-pump coronary
artery bypass surgery. This research is from the United Kingdom and involves 200
patients undergoing first-time coronary artery bypass grafting prospectively randomized
to either conventional CABG or off-pump surgery. Pre- and intraoperative variables were
similar between the two groups. This study observed a significant difference (p < 0.001)
in cost, defined as the sum of charges involving operating equipment, bed occupancy, and
transfusion requirement, between the two groups. The hospital costs of the on- and off-
pump groups were $3731.6 ± $1169.7 and $2615.1 ± $953.6, respectively. The increased
cost reflects increased morbidity seen in the group undergoing conventional treatment (5).
In the United States, several retrospective studies exist. One such study examined
MIDCABG, OPCAB, and conventional CABG costs. This study demonstrated hospital
costs of $15,000, $17,000, and $19,000, respectively. This represents a statistically sig-
nificant difference between the conventional and less invasive techniques (6). In addition
to an apparent cost advantage, a multicenter trial involving several Veterans Affairs
hospitals demonstrated significant improvement in terms of risk-adjusted morbidity and
mortality (0.52 vs 0.56 multivariable odds ratios for off-pump and on-pump, respec-
tively) (7).
A German study examined cost of conventional CABG, MIDCABG, OPCAB, and
Port-Access™ CABG (PACABG) (8). Compared to total cost of traditional CABG
($16,230, n = 60), OPCAB ($14,060, n = 5), and MIDCAB ($14,050, n = 5) were
associated with lower cost, but PACABG ($17,230, n = 5) was more expensive. This
study was hampered by its nonrandomized design, small sample size, and lack of data on
cost as it pertains to duration of rehabilitation and length of time away from work. Also
lacking long-term follow-up data was Magovern’s retrospective comparison between
MIDCABG (n = 60) and traditional CABG (n = 55) (9). The former group had a decreased
incidence of transfusion, shorter intubation time, and lower estimated hospital costs
($11,200 ± $3100 vs $15,600 ± $4200, p < 0.001). The decreased morbidity and cost
enjoyed by MIDCABG patients were particularly evident in high-risk patients. The
advantages experienced in the short run by the MIDCABG group, however, were balanced
by the fact that at 6 mo, five patients in this group had recurrent ischemia in the left
anterior descending artery distribution, mostly due to anastomotic stricture. The costs
associated with treating these patients were not included in this analysis.
The morbidity and cost advantages observed in high-risk patients in the Magovern
study were the focus of Del Rizzo’s study, which compared safety and cost-effectiveness
of MIDCABG in high-risk (HR) and low-risk (LR) patient populations (10). HR patients
were significantly older than their LR counterparts (72.2 ± 11.6 yr vs 63.3 ± 9.7 yr,
p = 0.006); HR patients also were significantly (p < 0.05) more likely to be female, have
a history of stroke, chronic obstructive pulmonary diseases (COPD), peripheral vascular
disease, prior cardiac surgery, congestive heart failure, and require urgent operation. HR
patients revascularized via MIDCABG enjoyed a 50% cost saving over traditional CABG
patients. However, the costs associated with the latter group were not actual cost data;
rather, they were determined from a cost regression model based on Ontario Ministry of
Health funding. Professional fees, capital depreciation, and long-term follow-up of these
patients were not considered in this study.
There are data that lend support to Del Rizzo’s findings. One study out of the Univer-
sity of Pittsburgh found an average savings of $8375 (p < 0.0001) enjoyed by MIDCABG
patients over CABG patients (11). Doty and colleagues further compared MIDCABG to
stenting, percutaneous transluminal angioplasty (PTCA), and conventional CABG (for
single-vessel disease) (12). Costs associated with the MIDCABG procedure were lower
than for conventional CABG and stenting but greater than for PTCA.
Additional support for the utilization of minimally invasive techniques can be found
in certain subsets of the patient population. Boyd and colleagues demonstrated differ-
ences in several areas in their retrospective review of their OPCAB experience in elderly
(age greater than 70) patients supporting the utilization of off-pump surgery in this
population. They observed significant differences (p < 0.05) between patients undergo-
ing off-pump and on-pump procedures in mean hospital stay (6.3 ± 1.8 d vs 7.7 ± 3.9 d),
average intensive care unit stay (24.0 ± 10.9 h vs 36.6 ± 33.5 h), and morbidity (in terms
of postoperative atrial fibrillation or low-output syndrome). The cost associated with off-
pump cases also was significantly lower at $6702 ± $1047 compared to $7784 ± $2846
(p < 0.05) (13). The Christiana Hospital (Medical Center of Delaware) similarly observed
that in this same patient population, postoperative length of stay was significantly lower
in their MIDCABG group compared to their CABG group (6.09 vs 8.39 d, p = 0.013).
Interestingly, the same group observed a significantly higher postoperative length of stay
in their MIDCABG group compared to their CABG group among their patients under the
age of 70 (14) (Table 1).
The most recent study of the elderly was performed retrospectively on octogenarians
undergoing either OPCAB or conventional CABG (15). There was no statistically sig-
nificant survival difference between the two groups. OPCAB patients enjoyed advan-
tages in terms of perioperative stroke (0% OPCAB vs 7.1% CABG, p = 0.04), prolonged
Table 1
Summary of Retrospective Conventional vs Minimally
Invasive Coronary Artery Bypass Graft Studies
Cost ($) LOS (d)
CABG Cost
and LOS On-pump Off-pump On-pump Off-pump p
Minneapolis Heart
Institute (8) 19,000 17,000 — — <0.05
Canada
(Multicentered) (10)a 7784 ± 2846 6702 ± 1047 7.7 ± 3.9 6.3 ± 1.8 <0.05
Christiana Hospital (11)a — — 6.09 8.39 0.013
aPatient population age > 70.
ventilation (1.7% OPCAB vs 11.8% CABG, p = 0.02), and transfusion rate (33% OPCAB
vs 70.4% CABG, p < 0.001). A shorter hospital stay (6.3 d OPCAB vs 11.5 d CABG, p
< 0.001) translated to cost savings for the OPCAB group ($9,363 OPCAB vs $12,312
CABG, p < 0.001).
During January and February of 2000, 52 on-pump CABGs and 42 off-pump CABGs
were performed at New York Presbyterian Hospital–Columbia. A retrospective review
of this small sample of patients showed off-pump total costs to be significantly different
than on-pump total costs (p = 0.015). On average, off-pump cases cost approx $22,000
less than on-pump cases. The average postoperative length of stay was 13 d ± 20.4 vs 7.8
± 7 for off-pump (p = 0.078). We further broke our patient population into subgroups
based on age (<70 and ≥70), gender, and number of grafts (three or four). In this
subanalysis, significant cost differences between on-pump and off-pump cases (the
former being associated with higher costs) in our <70 patients and our male patients
were observed. With regard to postoperative length of stay, we saw a significant differ-
ence only in our male patients, again in favor of the off-pump method (Table 2).
In our patient pool, as would be expected, a handful of patients had protracted hospital
stays. Thus, we also analyzed our data eliminating outliers that fell beyond two standard
deviations (Table 3). Again, significant cost differences favoring the off-pump method
among all patients as well as our <70 and male subgroups were observed. Additionally,
our subgroups of three and four grafts now demonstrated significant difference. Regard-
ing length of stay, again only our male subgroup demonstrated a significant difference.
Overall, our data favor the less invasive off-pump procedure; however, since the data are
derived from a nonrandomized population, we could control for the severity of illness in
the two cohorts.
In our review of the literature, there is some evidence favoring off-pump surgery from
prospective studies. For instance, Lee and colleagues performed a case-controlled study
comparing outcomes and costs associated with 100 consecutive patients undergoing
OPCAB procedures and those associated with 100 contemporary matched conventional
CABG patients. Three important demographic differences existed between the two
groups. First, on-pump cases averaged 3.8 ± 1.0 grafts, significantly higher than the 3.1
± 1.0 grafts for the off-pump cases (p < 0.001). Second, all renal-failure patients who were
dialysis-dependent underwent the OPCAB procedure. Finally, five times as many con-
ventional CABG patients were undergoing reoperation (10% vs 2%, p = 0.03). This being
said, this group had numerous findings seemingly favoring the off-pump technique. They
TableInvasive
Chapter 25 / The Economic Impact of Minimally 2 Cardiac Surgery 365
The Columbia Experience, January–February 2000
Columbia CABG Data, Postoperative Length

January–February Total cost ($) of Stay (d)
2000 (On vs Off) On-pump Off-pump p On-pump Off-pump p
Total experience 52,130.58 ± 30,530.98 ± 13.00 ± 7.85 ±
0.015 0.078
(n = 57 vs 48) 59,504.19 25,569.50 20.38 6.95
< 70 61,041.69 ± 30,489.94 ± 15.76 ± 7.74 ±
0.047 0.13
(n = 29 vs 34) 75,027.96 29,690.93 26.74 7.80
Age
≥ 70 42,902.25 ± 30,630.64 ± 10.00 ± 8.14 ±
(n = 28 vs 14) 36,505.63 11,222.10 0.11 9.51 4.47 0.40
Female 70,526.75 ± 45,483.85 ± 22.63 ± 11.85 ±

(n = 8 vs 13) 94.242.89 44,274.41 0.50 41.05 11.52 0.49
Gender
Male 49,127.12 ± 24,977.06 ± 11.43 ± 6.37 ±
0.0028 0.025
(n = 49 vs 35) 52,630.89 9731.11 14.85 3.40
3 vessels 34,472.11 ± 32,455.60 ± 7.61 ± 7.76 ±
No. of (n = 18 vs 25) 8657.36 23,571.65 0.70 2.23 3.95 0.88
grafts 4 vessels 60,438.05 ± 36,473.89 ± 14.33 ± 10.67 ±
0.27 0.56
(n = 21 vs 9) 70,719.38 43,016.12 19.56 13.71
Table 3
The Columbia Experience, Excluding Outliers Beyond Two Standard Deviations
Columbia CABG Data, Postoperative Length

January–February Total cost ($) of Stay (d)
2000 (On vs Off) On-pump Off-pump p On-pump Off-pump p
Total experience 39,558.85 ± 25,641.87 ± 8.78 ± 6.72 ±
0.00007 0.060
(n = 54 vs 46) 21,897.53 9684.39 7.068 3.35
< 70 41,894.04 ± 23,459.28 ± 9.11 ± 6.094 ±
0.0004 0.097
(n = 27 vs 32) 22,799.72 8195.99 8.84 2.57
Age
≥ 70 37,223.67 ± 30,630.64 ± 8.44± 8.14 ±
(n = 27 vs 14) 21,125.66 11,222.10 0.20 4.85 4.47 0.84
Female 37,380.71 ± 27,757.18 ± 8.14 ± 7.82 ±

(n = 7 vs 11) 10,385.24 9675.37 0.072 3.13 3.09 0.83
Gender
Male 39,883.26 ± 24,977.06 ± 8.87 ± 6.37 ±
0.00019 0.047
(n = 47 vs 35) 23,185.61 9731.11 7.50 3.40
3 vessels 34,472.11 ± 28,171.50 ± 7.61 ± 7.21 ±
No. of (n = 18 vs 25) 8657.36 10,049.51 0.036 2.23 2.89 0.61
grafts 4 vessels 46,635.35 ± 22,197.25 ± 10.70 ± 6.13 ±
0.0035 0.73
(n = 20 vs 8) 32,451.20 4274.10 10.53 1.64
observed a significantly reduced postoperative length of stay (6.1 ± 2.5 d vs 7.1 ± 3.3 d,
p = 0.003) and a 29% lower variable direct cost per case (p < 0.001) (16). Selection biases
likely play a role in this prospective study.
When addressing the outstanding economic question of the costs of minimally inva-
sive cardiac surgery, the majority of data are found in nonrandomized studies done at
various institutions throughout the United States, Canada, and Europe. Many of these
studies suggest that minimally invasive cardiac procedures hold various advantages over
conventional surgery in terms of cost and morbidity. However, recent prospective data
do not definitively support these findings. In 2001, the University of Utah published its
OPCAB experiences and showed no significant difference in terms of complication rate,
length of hospital stay, and hospital costs. Costs were comparable aggregately and when
divided into surgical, intensive care, and ward-care costs (17).
CONCLUSIONS
To summarize, there is no clear answer to whether minimally invasive cardiac surgery
confers a significant cost benefit over traditional methods. This technology, which is ever
changing, is still in its infancy. For instance, robotic surgery shows promise in ASD
repair, mitral valve repair, and even coronary artery bypass grafting. It will be some time
before we will know how long-term savings through improved outcomes may justify the
substantial fixed costs associated with acquiring and learning this new technology.
Today, we lack the support of controlled, randomized studies to compare the two
accurately with regard to the bottom line. This is the first issue that must be addressed.
To this end, there is promising new research on the horizon. The Octopus Study from
Europe is comprised of two multicenter randomized clinical trials (18). The first is known
as the OctoStent trial and compares the medical effectiveness of intracoronary stent
placement with off-pump coronary bypass grafting. The second, termed the OctoPump
trial, looks at neurological status following on- and off-pump bypass grafting. Secondary
endpoints for both trials will be cost-effectiveness.
Our institution is in the midst of a Comparative On and Off Pump (COOP) CABG trial.
The trial is a prospective, randomized pilot study designed to examine costs and mortality
and morbidity for on-pump vs off-pump patients as well as determine whether
neurocognitive differences exist between the two groups. Neurocognitive function is
evaluated with simple neurocognitive tests that are administered preoperatively and then
again at 1 wk, 1 mo, and 6 mo postoperatively. The tests are blindly scored and will help
determine if the deleterious effects of the cardiopulmonary bypass system translate into
any significant neurocognitive differences.
Preliminary data from our COOP CABG study is in line with the other current prospec-
tive data available. With approximately half of the total patients enrolled (n = 45), initial
cost analysis shows no significant difference between on- and off-pump procedures.
Specifically, current COOP CABG cost data show on-pump costs of $30,716 ± $11,071
vs $30,299 ± $11,597 for off-pump procedures (p = ns). Although these data suggest little
difference in overall costs, they do reveal a significantly lower utilization of operating-
room resources for the off-pump group. These data raise interesting questions about cost
differences between the on- and off-pump groups. We hope that our completed random-
ized study will clearly delineate how the two procedures compare from an economic
perspective (19).
The results of these trials will help define what niche minimally invasive cardiac
surgeries will fill. However, when comparing them to conventional procedures, results
such as duration of time away from work postoperatively must be noted to get an
accurate sense of cost-effectiveness on a societal scale. Upcoming randomized con-
trolled studies will evaluate and probably validate the utilization of minimally invasive
surgery for patients with certain comorbidities that may not tolerate cardiopulmonary
bypass well. As minimally invasive technologies mature, assessment of cost-efficacy
may become the major driving force to adoption of these procedures.
REFERENCES
1. Cohn LH, Adams DH, Couper GS, et al. minimally invasive cardiac valve surgery improves patient
satisfaction while reducing costs of cardiac valve replacement and repair. Ann Surg 1997;226:421–428.
2. Cosgrove DM, Sabik JF, Navia JL. Minimally invasive bypass operations. Ann Thorac Surg
1998;65:1535–1539.
3. Arom KV, Emery RW, Kshettry VR, Jancy PA. Comparison between Port-Access and less invasive
valve surgery. Ann Thorac Surg 1999;68:1525–1528.
4. Subramanian S, Liu C-F, Cromwell J, Thestrup-Nielsen S. Preoperative correlates of the cost of coro-
nary artery bypass surgery: comparison of results from three hospitals. Am J Med Qual 2001;16:87–91.
5. Ascione R, Lloyd CT, Underwood MJ, Lotto AA, Pitsis AA, Angelini GD. Economic outcome of off-
pump coronary artery bypass surgery: a prospective randomized study. Ann Thorac Surg
1999;68:2237–2242.
6. Arom KV, Emery RW, Flavin TF, Petersen RJ. Cost effectiveness of minimally invasive coronary artery
bypass surgery. Ann Thorac Surg 1999;68:1562–1566.
7. Plomondon ME, Cleveland JC Jr, Ludwig ST, et al. Off puimp coronary artery bypass is associated with
improved risk-adjusted outcomes. Ann Thorac Surg 2001;72:114–119.
8. Reichenspurner H, Boehm D, Detter C, Schiller W, Reichart B. Economic evaluations of different
minimally invasive procedures for the treatment of coronary artery disease. Eur J Cardiothorac Surg
1999;16(suppl 2):S76–S79.
9. Magovern JA, Benckart DH, Landreneau RJ, Sakert T, Magovern GJ Jr. Morbidity, cost, and six-month
outcome of minimally invasive direct coronary artery bypass grafting. Ann Thorac Surg 1998;66:1224–1229.
10. Del Rizzo DF, Boyd WD, Novick RJ, McCkenzie FN, Desai ND, Menkis AH. Safety and cost-effec-
tiveness of MIDCABG in high-risk CABG patients. Ann Thorac Surg 1998;66:1002–1007.
11. Zenati M, Domit TM, Saul M, et al. Resource utilization for minimally invasive direct and standard
coronary artery bypass grafting. Ann Thorac Surg 1997;63:S84–S87.
12. Doty JR, Fonger JD, Nicholson CF, Sussman MS, Salomon NW. Cost analysis of current therapies for
limited coronary artery revascularization. Circulation 1997;96(suppl II):II16–II20.
13. Boyd WD, Desai ND, Del Rizzo DF, Novick RJ, McCkenzie N, Menkis AH. Off pump surgery decreases
postoperative complications and resource utilization in the elderly. Ann Thorac Surg 1999;68:1490–1493.
14. Lemole GM, Choudri AF, Oz MC, Goldstein DJ, Gianguzzi R, Nguyen HC. Economic impact o fless
invasive cardiac operations. In: Oz MC, Goldstein DJ, eds. Contemporary Cardiology: Minimally
Invasive Cardiac Surgery. Totwa, NJ: Humana Press, 1998.
15. Hoff SJ, Ball SK, Coltharp WH, Glasfford DM. Lea JW 4th, Petracek MR. Coronary artery bypass in
patients 80 years and over: is off pump the operation of choice? Ann Thorac Surg 2002;74:S1340–S1343.
16. Lee JH, Abdelhady K, Capdeville M. Clinical outcomes and resource usage in 100 consecutive patients
after off-pump coronary bypass procedures. Surgery 2000;128:548–555.
17. Bull DA, Neumayer LA, Stringham JC, Meldrum P, Affleck DG, Karwande SV. Coronary artery bypass
grafting with cardiopulmonary bypass versus off-pump coronary bypass grafting: does eliminating the
pump reduce morbidity and cost? Ann Thorac Surg 2001;71:170–173.
18. van Dijk D, Nierich AP, Eefting FD, et al. The Octopus study: rationale and design of two randomized
trials on medical effectiveness, safety, and cost-effectiveness of bypass surgery on the beating ehart.
Control Clin Trials 2000;21:595–609.
19. Gelijns A, Gupta L, Kherani A, Burton E, Oz MC. Comparative on and off-pump CABg pilot trial.
Unpublished data.
Chapter 26 / Quality-of-Life Issues 369
26 Minimally Invasive Cardiac Surgery
Quality-of-Life Issues
Giulio Pompilio, MD, PHD,

Francesco Alamanni, MD, and Paolo Biglioli, MD
CONTENTS
INTRODUCTION
QUALITY OF LIFE AND MINIMALLY INVASIVE CARDIAC SURGERY:
DOES IT REALLY MATTER?
MINIMALLY INVASIVE CARDIAC SURGERY AND QUALITY-OF-LIFE:
MEASUREMENT TOOLS
MINIMALLY INVASIVE CARDIAC SURGERY AND QUALITY-OF-LIFE ISSUES:
LITERATURE REVIEW
CONCLUSIONS
REFERENCES
INTRODUCTION
Since 1948, when the Word Health Organization defined health not only as the absence
of disease, but also as the presence of physical, social, and mental well-being, studies on
the quality of life (QoL) in medicine have become increasingly important in daily clinical
practice and in medical research. The term “quality of life” refers to the physical, psycho-
logical, and social aspects of an individual’s health, aspects that should be considered as
a separate entity as they are closely related to the personal experience, expectations,
hopes, and impressions the patient has concerning his or her own health (1).
This is therefore a complex subject and any attempt to conceive of indices with which
to measure QoL, thus adopting a “scientific” approach, must inevitably take into account
its analytical, personal, and empirical nature. Essentially, the only reliable control of QoL
studies is the individual. To identify the general aim of a study on the quality of life in
the medical field, Testa and Simonson (1) proposed the following definition: “measure-
369
ment of the quality of life should address, that is, cover, each objective and subjective
component (symptom, condition or social role) that is important to members of the patient
population and susceptible to being affected positively or negatively by interventions.”
Despite the objective difficulty of the analysis, in various branches of medicine, stud-
ies on QoL have increased exponentially, not least in cardiac surgery, especially over the
last 5 yr. In this chapter we analyze the impact QoL has had on minimally invasive cardiac
surgery (MICS), probably the most important innovation in cardiac surgery in recent years.
It is not easy to give a brief definition of MICS, as there is considerable literature on
various surgical strategies and interventions described under this “umbrella” term. Gen-
erally, MICS involves cardiac surgery techniques and strategies that together reduce
surgical trauma as much as possible and that increase patient acceptance of, and satisfac-
tion with, the operation. This said, it seems obvious that MICS is something that is closely
linked to the QoL.
However, a review of the literature dedicated to QoL in MICS does not exactly reflect
this point of view. From 1995 to the present day, the PubMed database has listed more
than 180 articles that discuss various aspects of MICS; of these, only five (less than 3%)
address issues related to QoL.
QUALITY OF LIFE AND MINIMALLY INVASIVE CARDIAC

SURGERY: DOES IT REALLY MATTER?
It is therefore opportune to ask which aspects of MICS concern QoL. QoL is, in fact,
a generic term that encompasses the total well-being of the person, including functional
capacity, symptoms, and the patient’s perception of well-being (2). The concept of
“minimally invasive” should theoretically imply benefits derived from reduced surgical
trauma. The hope of using MICS approaches is that patients benefit by having less
postoperative physical discomfort, earlier return to physical fitness, and expedited
resumption of their normal daily activities. Patients are also expected to be more satisfied
with the operation from a cosmetic point of view than patients who have had conventional
techniques.
Moreover, MICS may also reduce the impact of surgery on the patient’s psychological
status. It is well known that the patient’s emotional status (anxiety, depression, self-
control) plays an important role on the QoL of heart surgery patients. Theoretically,
variables related to the patient’s psychology, such as resuming a normal family, social,
and sexual life and getting back to work, may benefit in the early and long-term period
from a minimally invasive approach.
Furthermore, the interesting hypothesis that various factors linked to QoL may influ-
ence the “quod vitam et valitudinem” long-term prognosis of heart surgery patients has
been recently suggested. Hertz (3) and Rumsfeld (4) have carried out elegant studies on
the impact of preoperative QoL on long-term survival (5 yr) of patients who had had
coronary artery bypass (CAB). Their interesting conclusions were in agreement: condi-
tions linked to low QoL before surgery are an independent risk factor for long-term
survival. It would therefore be useful, for instance, to establish if variables that are
theoretically linked to the techniques of minimally invasive surgery, such as long-term
patient satisfaction with the operation, in terms of physical and psychological well-being,
can significantly improve the prognosis of those patients in whom preoperative indices
suggest an unfavorable impact of QoL on life expectancy. This factor may then lead the
surgeon to consider minimally invasive techniques.
It is clearly not easy to provide definitive answers to such complex questions. This is
because randomized and prospective studies are needed on traditional surgical tech-
niques and because of the many different types of surgery described as MICS. We strongly
believe, however, that any surgeon practicing minimally invasive cardiac surgery should
be cognizant and sensitive to QoL issues.
Recent studies have attempted to answer the many questions raised. Before reviewing
the pertinent literature, it is necessary to briefly consider the instruments used in cardiac
surgery to analyze QoL.
MINIMALLY INVASIVE CARDIAC SURGERY AND QUALITY

OF LIFE: MEASUREMENT TOOLS
QoL cannot be measured directly, but rather through a series of items in a questionnaire
that are awarded a numerical score. Various questionnaires have been used to measure
QoL in heart surgery. Table 1 reports the main studies and questionnaires used to date in
cardiac surgery. It is immediately evident that there are many methods for assessing QoL.
Each author clearly used the questionnaires he or she believed most appropriate to the aim
of the study. It is just as obvious that there is no questionnaire that is universally accepted
as measuring the main issues of QoL in cardiac surgery. However, from a careful review
of the methods used, it appears that a combination of certain questionnaires could
adequately cover the majority of issues inherent to the QoL in a prospective study on
minimally invasive cardiac surgery.
Physical Activity Score

The Physical Activity Score (PAS) represents one dimension of an angina-specific
questionnaire (5), the Angina Pectoris Quality of Life questionnaire, which contains six
questions, for the self-estimation of physical abilities and limitations. Each response is
graded from 1 to 6 and the mean value for all six questions is calculated. The higher the
total value, the greater is the degree of disability.
The PAS is a useful instrument for quantifying the benefits of MICS in postoperative
recovery, with special reference to the resumption of normal daily activities. In the long
term, it can be used to measure and compare the degree of physical well-being achieved
after MICS and conventional surgery. It should be noted that the questionnaire indi-
rectly also refers to the clinical, situation of the patient as the reappearance of, for
example, angina, or a worsening of the patient’s New York Heart Association (NYHA)
class, is directly reflected in their degree of physical activity and therefore in their QoL.
Nottingham Health Profile

The Nottingham Health Profile (NHP) is divided in two parts (6). Part I consists of 38
statements that convey limitations of activity or aspects of distress in six dimensions:
physical mobility, pain, sleep, energy, social isolation, and emotional reactions. Patients
are required to indicate by a yes/no answer which of the problems they are experiencing
at the time they complete the questionnaire. A score ranging from 0 to 100 can be
calculated for each dimension of this part of the profile; the higher the score, the greater
the limitations in the activity or the distressing social and emotional problems. Parts II
lists the seven aspects of life that are found to be most affected by a person’s state of
health: occupation, ability to perform jobs around the house, social life, home relation-
ships, sexual life, hobbies, and holidays; a yes/no answer indicates which areas are
affected by the respondent’s present state of health.
Table 1
372
Main Studies and Questionnaires Used to Date in Cardiac Surgerya
Reference Type of procedure Questionnaire Type of QL study
Chocron S et al.: CAB, valve or combined NHP Comparative study, outcome
Ann Thorac Surg 1996 procedures and QL
Nielsen D et al.: Miscellaneous NHP QL in complicated patients
Crit Care Med 1997
Sjöland H et al.: CAB PAS, NHP, PGWBI Hypertensive patients
J Hyperten 1997
Hlatky MA et al.: CAB and PTCA Duke ASI, RAND MHI Comparative study
N Engl J Med 1997
Shapira OM et al.: Valve surgery Duke MOS Elderly patients
Chest 1997
Westin L et al.: CAB and PTCA Self-made Comparative study
J Int Med 1997
Soklano JA et al.: CAB Euro QL Octogenarians
372
Ann Surg 1998

Fruitman DS, et al.: CAB, AVR, MVR, SAQ, RAND SF-36 Octogenarians
Ann Thorac Surg 1999 and combined
Part V / Miscellaneous Aspects of MICS

Rumsfeld JS et al.: CAB SF-36 Long-term outcome
JAMA 1999 and prognosis
Herlitz J et al.: 1999 CAB PAS, NHP, PGWBI Long-term outcome
and prognosis
Sjöland H et al.: CAB PAS, NHP, PGWBI Gender influence
J Int Med 1999
Yun KL, et al.: CAB, valve HSQ Time-related QL
Ann Thorac Surg 1999 or combined procedures
Smith HJ et al.: CAB QLI, SF-36, QLMI, SEIQL Comparative study
Heart 2000
aNHP=Nottingham Health Profile; PAS=Physical Activity Score; PGWBI=Physiological General Well-Being Index; Duke ASI=Duke
University Activity Status Index; RAND MHI=RAND Mental Health Inventory; Duke MOS=Duke University Medical Outcomes Study;
HSQ=Health Status Questionnaire; Euro QL-Euro Quality of Life Questionnaire; SF36=Short Form 36; QLI=Quality of Life Index-Cardiac
Version; QLMI=Quality of Life after Myocardial Infarction Questionnaire; SEIQL=Schedule for the Evaluation of Individual Quality of Life.
The NHP partially completes the PAS for those aspects related to the resumption of
physical activity after the operation. Part I concerns aspects of QoL that often deteriorate
in the postoperative period of conventional surgery, mainly concerned with the patient’s
physical and mental well-being, as, for example, the normal rhythms of sleeping and
waking or the degree of anxiety. MICS might, we believe, have a positive influence on
these variables, compared to conventional surgery. Part II is useful to assess the markers
that indicate the patient has returned to normal social and family life, and therefore to
assess the long-term benefits of MICS on the QoL of the patient who has had surgery.
The Psychological General Well-Being Index

The Psychological General Well-Being Index (PGWBI) contains 22 questions, deal-
ing with six sections of well-being: anxiety, depressed mood, vitality, general health,
self-control, and well-being (7). The response format is graded from 1 to 6 (total score
range 22–132), with the highest value corresponding to superior well-being.
The PGWBI completes both parts of the NHP. The emphasis on the state of well-being
may be of interest for patients who have MICS, especially because of the beneficial
influence this type of surgery has on the level of patient satisfaction and acceptance of surgery.
MINIMALLY INVASIVE CARDIAC SURGERY AND QUALITY-OF-LIFE

ISSUES: LITERATURE REVIEW
The impact of minimally invasive cardiac surgery on QoL has received little attention
in the literature. Table 2 highlights the principal articles covering the subject.
Pain
The Leipzig group has addressed the problem of quantifying postoperative pain after
MICS and comparing it with that after conventional surgery in two studies. Walther and
colleagues (8) carried out a prospective study on 338 patients who had had a coronary
bypass, aortic, and/or mitral valve replacement with both the minimally invasive method
and with conventional surgery using median sternotomy. The MICS techniques used
were left anterior minithoracotomy with beating heart (minimally invasive direct coro-
nary artery bypass or MIDCAB) (see Chapter 9); right lateral minithoracotomy using
CPB and endoaortic clamping of the aorta to replace the mitral valve (see Chapters 19 and
20), and partial sternotomy with the usual CPB arrangement for aortic valve replacement
(see Chapter 21).
The assessment of postoperative pain was made using a standard “pain questionnaire,”
which provides information on the intensity of the pain, the nature of the pain, where it
is, how it changes over time, how it changes with physical activity, and the most effective
doses of the most efficient painkillers. Two different scoring tests were used: a verbal
rating scale, which allows differentiation between no pain, mild pain, moderate pain,
severe pain, and unbearable pain; and a visual analog scale, which quantifies pain on a
scale from 1 (no pain) to 10 (the worst pain the patient has ever experienced).
The level of postoperative pain diminishes in all patients during the first postoperative
week, with no difference between MICS and conventional patients in their use of pain-
killers. Recipients of the MIDCAB operation and those who underwent minimally inva-
sive approaches to the mitral valve benefited from less pain after postoperative d 3 than
those who have had conventional sternotomy, which causes pain and tension in the back
muscles for some time. Interestingly, however, patients who underwent the MIDCAB
Table 2
Principal Recent Papersa
Reference Type of procedure Questionnaire Type of QoL study
Massetti M et al.: Miscellaneous MICS Home-made Cosmetic aspects
Eur J Cardiothorc
Surg 1999
Grossi EA et al.: MIDCAB DASI Pain, stress response, QL
Eur J Cardiothorac
Surg 1999
Walther T et al.: Miscellaneous MICS Pain Q, NHP Pain and QL
Ann Thorac
Surg 1999
Biglioli P et al.: MIDCAB PAS, PGWBI QL
Ann Thorac
Surg 1999
Diegeler A et al.: MIDCAB HNP Pain, QL
Heart Surg
Forum 2000
aQL=Quality of Life; MIDCAB=minimally invasive coronary bypass grafting; Pain Q=pain questionnaire;
DASI=Duke Activity Status Index; NHP=Nottingham Health Profile; PAS=Physical Activity Score;
PGWBI=Psychological General Well-being Index.
operation experienced more severe incisional pain for 48 h after the operation, making
local anesthetic and/or epidural catheter useful in relieving pain.
These results are confirmed elsewhere by the same authors (9) in a case series that
included only patients who underwent a MIDCAB. The difference in the postoperative
levels of pain was close to reaching statistical significance until postoperative d 7. In this
study, using a modified version of the NHP questionnaire, they also assessed the aspects
of QoL that are linked directly to postoperative pain (degree of limitation of mobility and
physical activity by pain). They observed a net improvement in MIDCAB patients in
resuming their normal physical activities after the operation compared to those who had
had conventional surgery. These results were confirmed 3 mo after surgery.
Cosmetic Issues
The cosmetic effects of a midline thoracic scar may have a negative impact on patients’
self-concepts and therefore on their QoL. Massetti et al. (10) investigated the long-term
cosmetic impact of MICS in a series of 56 young female patients who, in a 17-yr time
span, underwent atrial septal defect closure through a right anterolateral thoracotomy.
Patients completed a multiple-choice questionnaire focusing on self-evaluation of the
esthetic result and its psychological influence. In particular, the patients were asked
about their impression regarding: (1) the volume and the symmetry of the breasts; (2) a
description of the quality (color, dimension, and visibility) of the scar; (3) their feelings
about the scar when dressings were removed, appearance in a bathing suit, playing sports,
buying bras, and during intimate relations with partners; and (4) their satisfaction or
dissatisfaction resulting from surgery.
The answers of the questionnaires suggested that the patient’s subjective impressions
are at least commensurate with the objective findings. More than 90% of the patients self-
evaluated the cosmetic result of the operation as excellent or good. Regarding the psy-
chological influence of the surgical scar, absence of troubles as described in point 3 are
found in more than 80% of patients. Hence, the minimally invasive approach to atrial
septal defect through small right thoracotomy appears to have a positive long-term
impact on patients’ QoL. The inferences of the authors were that, once surgeons perfected
the surgical techniques, the challenge remained to reduce the impact of the cosmetic
blemish that median sternotomy may leave on the patient’s chest.
Quality of Life
The Leipzig group (8,9) assessed QoL in patients undergoing MICS using a modifi-
cation of the Nottingham Health Questionnaire. The latter consisted of the inclusion of
one possible positive answer in each category. Different aspects of QoL were analyzed,
including mobility, social status, level of activities, emotional state, and sleeping disor-
ders. Changes in QoL were evaluated preoperatively, early in the postoperative period,
and during interviews at 3 mo postoperatively.
When considering miscellaneous MICS patients (different MICS approaches in coro-
nary and valve diseases), the authors (8) were not able to assess at 3 mo relevant differ-
ences in postoperative QoL for MICS in comparison with conventional surgery. On the
other hand, using the same QoL measurements, if MIDCAB patients only are considered
(9), a statistically improved outcome regarding QoL aspects such as mobility, and a trend
toward less limitations in terms of activities and sleep, were observed. No differences
between surgical groups could be evaluated regarding emotional and social aspects.
Grossi and colleagues (11) evaluated QoL in port-access (PA) vs standard sternotomy
coronary bypass patients (STD). The Duke Activity Status Index questionnaire was
administered to patients preoperatively and at the end of postoperative wk 1, 2, 4, and 8.
This questionnaire measured return to normal functional activity in terms of 12 common
activities of daily living, including the patient’s ability to walk, climb stairs, work around
the house, and perform related activities. In addition, as an indication of rapidity of
recovery, patients were asked to rate their health compared to the preoperative level at
the same postoperative time periods.
A significantly higher percentage of PA patients compared with STD patients were
able to walk one or two blocks at 1 wk, climb stairs at 1 and 2 wk, and engage in moderate
recreational activity and heavy housework at 8 wk. The PA group reported a significantly
higher percentage of return to preoperative activity level than did the STD group.
CENTRO CARDIOLOGICO MONZINO IRCCS EXPERIENCE

Patients and Methods
Between March 1995 and March 2001, 108 patients underwent MIDCAB at our insti-
tution. Mean age was 59 ± 9.7 yr, 72 (66.6%) were male, 37 (34.2%) had unstable angina,
48 (44.4%) had a history of hypertension, 7 (6.4%) had insulin-dependent diabetes
mellitus, 9 (8.3%) had chronic renal failure, 3 (2.7%) had a previous stroke, 3 (2.7%) had
a previous transient ischemic attack, 43 (39.8%) had a previous remote myocardial
infarct (>3 mo), 17 (15.7%) had a recent myocardial infarct (<3 mo), and 21 (19.4%) had
previously undergone a percutaneous intervention.
All patients completed three self-administered questionnaires for the evaluation of
their quality of life, including the Nottingham Health Profile (Parts 1 and 2), the Physical
Activity Score, and the Psychological General Well-Being Index questionnaires before
Fig. 1. Kaplan-Meyer survival 48 mo after MIDCAB.
surgery. The same questionnaires were then mailed to patients 6 (n = 108) and 12 (n = 108)
mo after surgery (12). At 48 mo (n = 63 pts.) postoperatively, QoL information was
obtained by direct interview. All the questionnaires were validated carefully and tested
for reliability.
Clinical follow-up information, available for 100% of the patients, was obtained every
year postoperatively by means of mailed questionnaires and at the fourth postoperative
year by direct examination of the patient. The date of the last inquiry was between January
and April 2001. The following clinical endpoints were collected for each patient: (1)
survival; (2) acute myocardial infarction-free survival; (3) angina-free survival; (4) redo-
free survival; (5) PTCA-free survival; (6) PTCA to LAD-free survival. Survival rates of
the entire patient population were determined by Kaplan–Meier survival analysis, and the
estimated survival proportions are reported ± the standard error of the estimates.
Early Results
All patients underwent MIDCAB through a left anterior thoracotomy and in all cases
a LIMA-to-LAD anastomosis was performed on a beating heart; no conversions to ster-
notomy were performed. No perioperative deaths occurred. Two patients (1.8%) showed
suffered a perioperative myocardial infarction, without hemodynamic deterioration. Four
patients (3.7%) underwent reoperation for bleeding. The mean ventilatory support time
and the mean ICU stay were 7 ± 7.4 h and 15 ± 10 h, respectively, while postoperative
length of stay was 5.4 ± 1.5 d.
Late Results
Average follow-up was 39 ± 13.4 mo (median 41). Survival and survival-free esti-
mates of the patient population are reported in Figs. 1, 2, and 3; of note, 4-yr survival was
98% ± 1.8%, redo-free survival was 97% ± 0.9%, acute myocardial infarction-free sur-
vival was 95% ± 1.9%, and angina-free survival was 90% ± 4.4%.
Measurement of Quality of Life

The Physical Activity Score improved significantly after the operation, with improve-
ments of 26.8% and 34.2% at 12 and 48 mo, respectively (Table 3). The same behavior
was shared by the Psychological General Well-Being Index, which improved 25.6% and
Fig. 2. Freedom from angina and acute myocardial infarction (AMI) 48 mo after MIDCAB.
Fig. 3. Freedom from reoperation, percutaneous coronary angioplasty (PTCA) to the left
anteriordescending coronary artery (LAD), 48 mo after MIDCAB.
Table 3
Average Physical Activity Scores Before and After MIDCABa
All patients population (n = 108)
Before surgery 12 mo after surgery
Average score 4.3 ± 0.8 [4.5] 3.1 ± 0.7 [3.3]*
% change –27.9%
Patients who reached 48 mo follow-up interval (n = 63)
Before surgery 12 mo after surgery 48 mo after surgery
Average score 4.1 ± 1.2 [4.4] 3.0 ± 1.1 [3.1]* 2.83 ± 1.3 [3.4]*
% change –26.8% –31%
aScores are reported as mean ± s.d. [median in brackets]. * = p < 0.01 vs before surgery.
Negative % change values indicate an improvement.
Table 4
Average Total Scores for the Psychological General
Well-Being Index Before and After MIDCABGa
All patients population (n = 108)
Before surgery 12 mo after surgery
Average score 82.9 ± 8.9 [82.0] 103.5 ± 7.3 [104.0]*
% change +24.8%
Patients who reached 48 mo follow-up interval (n = 63)
Before surgery 12 mo after surgery 48 mo after surgery
Average score 83.1 ± 8.4 [81.5] 104.4 ± 9.1 [104]* 111.6 ± 6.8 [102.5]*
% change +25.6% +34.2%
aScores are reported as mean ± s.d. [median in brackets].
* = p <0.01 vs before surgery. Positive % change values indicate an improvement.
Fig. 4. Bar graph of the six different domains of the Psychological General Well-Being Index
(PGWBI) as assessed before, 1 yr, and 4 yr after MIDCAB.
26.2% at 12 and 48 mo, respectively (Table 4). The analysis of the six different subscales
of the test revealed further improvements at 12 and 48 mo in anxiety, positive well-being,
vitality, and general health domains (Fig. 4). Improvements seem to be reached in all
Fig. 5. Bar graph of the six different domains of the Notthingham Health Profile (NHP) Part I as
assessed before, 1 yr, and 4 yr after MIDCAB.
subscales. In particular, the positive well-being and the vitality scales show a favorable
trend from 12 to 48 mo postoperatively.
Figure 5 reports the average scores of Part 1 of the Nottingham Health Profile, with
lower scores indicating better health status. One year after surgery there were important
improvements in all the dimensions of this score. A further improvement when compar-
ing 1- and 4-yr results was seen mainly in mobility, social, and emotional subscales.
With regard to Part 2 of the Nottingham Health Profile (Fig. 6), the percentage of
patients who were experiencing problems in daily living was also markedly improved 1 yr
after surgery. At 4 yr, further improvements were observed in home, work recovery,
hobbies, and social life in general, indicating a long-term benefit of MIDCAB in daily living.
CONCLUSIONS
Despite the difficulty of analyzing a subject so closely related to varying subjective
and social conditions, our data and the literature indicate that MICS seems to have a
positive impact on the QoL of heart surgery patients. The concept of QoL after surgery,
however, cannot disregard clinical results (both in short postoperative and long-term
periods), as the primary expectation of surgery lies in its outcome. Results of MICS are
Fig. 6. Bar graph of the seven different domains of the Notthingham Health Profile (NHP) Part II
as assessed before, 1 yr, and 4 yr after MIDCAB.
now accepted as satisfactory in terms of mortality and morbidity (8–13); patients can
therefore enjoy further advantages in terms of QoL because of lower levels of invasive-
ness and surgical trauma.
Although the available data are still preliminary, MICS seems to have a good effect
on both the physical and psychological aspects of the patient’s status. Our findings
indicate that 4 yr after MIDCAB, patients experience a constant improvement in physical
and psychological well-being and social quality of life.
Specific questionnaires are required to measure QoL in MICS rather than the various
tested generic and disease-specific questionnaires for QoL available in heart surgery
literature. Condition-specific questionnaires (QL questionnaires and functional-living
indexes for MICS) and situation-specific ones (trial-specific modules and checklists)
able to strengthen reliability and validity of studies dedicated to QoL in MICS (13) are
needed. These questionnaires would supply useful additional information for the selec-
tion of those patients who may better benefit from a mini-invasive approach (depending
on sex, age, social position, etc.).
On the other hand, many questions remain unanswered. First, the effective long-term
superiority of MICS to improve QoL (or graft patency or survival) needs further confir-
mation. Second, long-term prognosis of MICS patients has to be analyzed and compared
to conventional techniques with prospective studies to identify those patients whose QoL
may have a negative impact on life expectancy. These data are not available today because
of the young age of clinical trials in MICS; it is hoped that they will become known as
the follow-up period increases. Finally, there is the need to prospectively match many of
the interventions labeled as MICS for their impact on QoL, especially in cases of different
approaches to the same pathology (for instance mini-sternotomy vs mini-thoracotomy
for the revascularization of LAD coronary artery). It is possible that some surgical inter-
ventions described as “minimally invasive” do not provide any real benefit in terms of
QoL. Once again, such a selection can be carried out as the survey and follow-up period
is extended.
It has been said that “in addition to relieving clinical symptoms and prolonging survival,
a primary objective of any health care intervention is the enhancement of quality of life
and wellbeing” (14). There is a well-founded hope that minimally invasive heart surgery
has the potential to meet this laudable objective.
REFERENCES
1. Testa MA, Simonson DC. Current concepts: assessment of quality-of-life outcomes. N Engl J Med
1996;334(13):835–840.
2. Stewart AL, Ware JE. Measuring functioning and well being—the Medical Outcomes Study approach.
Durham, NC: Duke University, 1992;1–449.
3. Herlitz J, Wiklund I, Caidahl K, et al. Determinants of an impaired quality of life five years after coronary
artery bypass surgery. Heart 1999;81:342–346.
4. Rumsfeld JS, Mc Whinney S, McCarthy M, et al. Health-related quality of life as a predictor of mortality
following coronary artery bypass graft surgery. JAMA 1999;281:1298–1303.
5. Wilson A, Wiklund I, Lahtl T, Wahl M. A summary index for the assessment of quality of life in angina
pectoris. J Clin Epidemiol 1991;44:981–988.
6. Hunt S, Wiklund I. Cross-cultural variation in the weighting of health statements: a comparison of
English and Swedish valuations. Health Policy 1987;8:227–235.
7. Dupuy H. The Psychological General Well-being (PGWB) Index. In: Wenger NK, Mattson ME, Furberg
CD, Elison J, eds. Assessment of Quality of Life in Clinical Trials of Cardiovascular Therapies. Green-
wich, CT: Le Jacq Publishers, 1985:170–183.
8. Walther T, Falk V, Metz S, Diegeler A, Battellini R, Autschbach R, Mohr FW. Pain and quality of life
after minimally invasive versus conventional cardiac surgery. Ann Thorac Surg 1999;67:1643–1647.
9. Diegeler A, Walther T, Metz S, et al. Comparison of MIDCAB versus conventional CAB surgery
regarding pain and quality of life. Heart Surg Forum 1999;2(4):290–296.
10. Massetti M, Nataf P, Babatasi G, Khayat A. Cosmetic aspects in minimally invasive cardiac surgery. Eur
J Cardiothorac Surg 1999;16(suppl 2):S73–S75.
11. Grossi EA, Zakow PK, Ribakove G, et al. Comparison of postoperative pain, stress response, and quality
of life in port access vs. standard sternotomy coronary bypass patients. Eur J Cardiothorac Surg
1999;16(suppl 2):S39–S42.
12. Biglioli P, Antona C, Alamanni F, et al. Minimally invasive direct coronary artery bypass grafting: mid-
term results and quality of life. Ann Thorac Surg 2000;70(2):456–460.
13. Osoba D. Guidelines for measuring health-related quality of life in clinical trials. In: Staquet MJ, Hays
RD, Fayers P, eds. Quality of Life Assessment in Clinical Trials. Methods and Practice. Oxford Univ
Press, 1998:19–35.
14. Berzon RA. Understanding and using health-related quality of life instruments within clinical research
studies. In: Staquet MJ, Hays RD, Fayers P, eds. Quality of Life Assessment in Clinical Trials. Methods
and Practice. Oxford Univ Press, 1998:1–15.
Chapter 27 / Experimental Percutaneous Mitral Valve Repair 383
27 Experimental Percutaneous
Mitral Valve Repair
Juan P. Umaña, MD and Peter Fitzgerald, MD

CONTENTS
INTRODUCTION
MANAGEMENT OF MITRAL REGURGITATION
THE EDGE-TO-EDGE TECHNIQUE
DEVELOPMENT OF AN OFF-PUMP MITRAL VALVE REPAIR
FUTURE TRENDS
REFERENCES
The dogmas of the quiet past are inadequate for the stormy present. The occasion is
piled high with difficulty, and we must rise with the occasion....We must think anew and
act anew—Abraham Lincoln
INTRODUCTION
Over the past few years we have witnessed an impressive increase in the number of
percutaneous cardiac procedures, in parallel with the development of new technology.
Drug-impregnated stents implanted in the coronary circulation promise significant
decreases in restenosis rates. Covered endovascular stent-grafts are being used success-
fully to treat descending thoracic aortic aneurysms and dissections, significantly lessen-
ing the morbidity and mortality of surgical intervention, and ongoing trials are evaluating
the safety and efficacy of percutaneous device closure of atrial septal defects.
Contrary to other cardiovascular structures, the mitral valve (MV) is less appealing for
percutaneous intervention, owing to its location (difficult access and risk of embolism)
and the complexity of its subvalvular apparatus. The concept of intervening on the mitral
valve without the use of cardiopulmonary bypass was introduced by Harken and Bailey
in 1948 with the use of closed mitral commissurotomy for mitral stenosis. More recently,
Reyes et al. compared percutaneous balloon valvuloplasty to open surgical commissuro-

383
tomy (1,2) and found on long-term follow-up that durability was better for the percuta-
neous procedure, while there was no difference in the incidence of embolic complica-
tions. Application of similar technologies to repair incompetent mitral valves has only
recently been entertained, with improved understanding of the pathophysiology of mitral
regurgitation and the subvalvular apparatus.
MANAGEMENT OF MITRAL REGURGITATION

The mitral valve and its subvalvular apparatus play a crucial role in maintaining left
ventricular geometry and optimizing the mechanical efficiency of the left ventricle (LV).
An intact subvalvular apparatus helps maintain the ellipsoid shape of the LV, decreasing
wall stress at end-systole and preserving a normal ratio of wall thickness to ventricular
diameter (3). As a consequence, treatment of mitral regurgitation (MR) should aim to
preserve this mechanism while using autologous materials to avoid thromboembolic
complications. Mitral valve repair methods developed by Carpentier et al. have been used
successfully for the last three decades with excellent reproducibility and 98% freedom
from reoperation at 10 yr (4). These techniques preserve ventricular function, are asso-
ciated with lower mortality rates, and have fewer thromboembolic complications when
compared to MV replacement (5–7).
Optimal timing for operation in mitral regurgitation remains controversial owing to
the very nature of the condition, which progresses insidiously due to loading conditions
causing left ventricular function to appear normal on echocardiogram until late in the
disease process. As a consequence, patients are referred for surgery when compensatory
mechanisms have been overrun and poor prognostic indicators are present, i.e., decreased
ejection fraction (<60%), increased end-systolic dimension (≥45 mm), and pulmonary
hypertension (8–12). This delayed referral pattern is in part based on evidence that
medically treated patients with severe MR who are in New York Heart Association
(NYHA) Class I or II have a 5-yr survival rate in excess of 85%, which is only slightly
lower than the survival rate in an age-matched population (13). Another factor leading
to untimely referral is the reluctance on the part of cardiologists to send asymptomatic
patients for “open-heart surgery,” with its attendant morbidity and mortality.
Some of these referral biases could be overcome with the development of a minimally
invasive method of valve repair that would allow cardiac surgeons to perform a less
morbid procedure early in the course of mitral regurgitation, perhaps altering the natural
history of the disease. A subset of patients who would benefit from a percutaneous or
closed approach are those with ischemic mitral regurgitation due to inferior/lateral wall
myocardial ischemia. This patient population is particularly challenging, as the addition
of mitral repair to routine myocardial revascularization can increase operative mortality
to 10–15%. Nonetheless, studies suggest that long-term prognosis for these patients is
better if revascularization is accompanied by repair of the mitral valve. An off-pump or
“closed-heart” technique would facilitate addressing the valve in these situations,
decreasing operative mortality/morbidity and improving long-term outcome. In more
advanced cases, where pulmonary hypertension has developed and ventricular function
is depressed, the benefits of avoiding cardiopulmonary bypass altogether are self-evident.
THE EDGE-TO-EDGE TECHNIQUE

The idea of minimally invasive cardiac surgery emerged around 1995, when surgeons
began to realize the benefits of smaller incisions and shorter cardiopulmonary bypass
times. An improved understanding of the pathophysiology of cardiopulmonary bypass,

its deleterious effects on neurocognitive function, and its proinflammatory effect was
followed by a resurrection of off-pump coronary bypass operations. With it, methods
were developed to perform coronary revascularization and valve surgery thoracoscopically
and with the help of robotics (14). Nevertheless, conventional mitral valve repair tech-
niques can only be performed on cardiopulmonary bypass, and although potentially
amenable to a totally endoscopic approach, an off-pump approach would not be feasible
in most circumstances. Alfieri et al. in 1998 reported their experience using a simplified
method of mitral valve repair in 121 patients, which consisted of suturing the prolapsing
portion of the diseased leaflet to the corresponding free edge of the opposing leaflet
(Fig. 1). In their earlier series, 93% of patients had a concomitant ring annuloplasty,
except for those patients with restricted leaflet motion or small annuli (15). Short-term
results were impressive, with an overall survival of 92 ± 3.1% at 6 yr, 95 ± 4.8% freedom
from reoperation, and no instances of mitral stenosis. In a recent update (16), the same
group corroborated their results with a mean follow-up of 2 yr (range, 1 mo to 7 yr).
Operative mortality was 0.7%, and actuarial survival at 5 yr was 94% ± 3%. Freedom
from reoperation was significantly better when a ring was used concomitantly with the
leaflet suture (92% vs 72%). Although the numbers are impressive, most patients who
had an edge-to-edge repair without a ring had extensive calcification of the posterior
annulus, which in itself compromises durability of any type of repair.
Umaña et al. in 1998 published the Columbia experience using the edge-to-edge
technique (“bow-tie” repair) in 10 patients with ischemic regurgitation. The bow-tie
repair was used as an adjunct to ring annuloplasty in cases in which mitral valve replace-
ment was felt to have resulted in unacceptably high morbidity or mortality due to poor
left ventricular function. Mitral regurgitation was significantly reduced from 4+ to 1+ on
average, and left ventricular ejection fraction improved from 33% ± 13% to 45% ± 11%
(p = 0.016) prior to discharge (17). Mean follow-up was 337 d (range 85–554 d), and
reoperation rate was 40%, all in patients with underlying structural leaflet pathology or
papillary muscle rupture. A particularly interesting observation was the significant
improvement in ejection fraction seen in this small patient cohort. We hypothesized that
the bow-tie stitch could have an “anchoring” effect on the dysfunctional posterior pap-
illary muscle, therefore decreasing paradoxical motion of the posterolateral segment and
improving overall ventricular function. To test this hypothesis we used an ovine model
of ischemic mitral regurgitation, and studied the effect of the bow-tie repair on ventricular
geometry, degree of regurgitation, and mitral annular function (18). Animals were inf-
arcted by surgically ligating the obtuse marginal branches 2 and 3 of the circumflex
coronary artery, which leads to posterior papillary muscle tethering and development of
chronic regurgitation as described by Llaneras et al. (19). Using three-dimensional
sonomicrometry array localization (3D-SAL), the mitral valve annulus, papillary
muscles, and ventricular long and short axes were studied in three different conditions:
baseline (with existing regurgitation), during snaring of posterior mitral suture
annuloplasty (deVega type), and during snaring of the bow-tie stitch.
We found that both techniques achieved similar control of regurgitation (3+ to 1+; p
< 0.05) (Fig. 2). Mitral annular contractility defined as [(maximum area – minimum
area)/maximum area] × 100, was significantly impaired by conventional posterior
annuloplasty, while it did not change with the bow-tie repair (Fig. 3). To determine
changes in ventricular geometry and ventriculo-mitral interaction, we measured the dis-
tance from the valvular plane to the tips and bases of the papillary muscles, as well as
Fig. 1. “Bow-tie” repair—Atrial view. (A) Diastole. (B) Systole.
between papillary muscle bases throughout the cardiac cycle (Fig. 4). It was of interest
to find that the edge-to-edge repair led to an increase in distance from the base of the
papillary muscles to the plane of the valve. On the other hand, the distance between the
bases of the papillary muscles decreased with the use of the bow-tie repair. These findings
suggest that this simple repair may cause an accentuation of the elliptical shape of the
ventricle even after an ischemic insult. This, in turn, may improve hemodynamic perfor-
mance and facilitate ventricular remodeling, as the wall tension of the left ventricle decreases.
Fig. 2. Degree of mitral regurgitation.
Fig. 3. Mitral annular contractility.
Fig. 4. Ventricular geometry—measurements. PPM, posterior papillary muscle; APM, anterior

papillary muscle.
Fig. 5. Coil-fastening device. (A) Grasper and undeployed coil. (B) Coil fastener holding leaflets
together.
DEVELOPMENT OF AN OFF-PUMP MITRAL VALVE REPAIR

Once the safety and efficacy of the bow-tie repair was demonstrated clinically and
experimentally, we set out to develop a device that could be inserted through the apex of
the left ventricle to grasp, approximate, and suture or staple the mitral valve leaflets
together, on or off cardiopulmonary bypass (Fig. 5). The initial prototype was a stainless
steel tool with a grasper consisting of two jaws that secured the leaflets and adjusted them
to minimize the amount of regurgitation prior to deployment of the screw found between
the jaws (Fig. 6). The screw was made of stainless steel with decreasing rung diameter that
allowed the leaflet tissue to be drawn tighter and closer together as the fastener was
advanced. Morales et al. tested this model in a mock loop using explanted human mitral
valves mounted on a ring and made to simulate regurgitation secondary to posterior
leaflet prolapse or restriction (20). The screw was successfully deployed in this model,
finding a significant decrease in mitral regurgitant flow from 71.6% ± 7.2% to 34.4% ±
17.3% (p = 0.0025), with no change in transvalvular flow (1.1 ± 0.3 to 1.2 ± 0.4 L/min;
p = 0.7).
In vivo, the concept was tested in nine mongrel dogs, inserting the device through the
apex of the left ventricle by means of a left-sided thoracotomy and using transesophageal
echocardiographic guidance. Given that this was the first iteration of the device being
applied to normal tissue, the animals were placed on cardiopulmonary bypass to observe
the leaflets being grasped and adjusted. The leaflets were torn in one animal during this
process. All eight other animals had intact tissue and no instances of screw migration or
dislodgement up to 12 wk. This study established the feasibility of the concept and also
demonstrated some of the possible shortcomings if the apparatus is used on a beating
heart, e.g., a stiff 21 fr shaft inserted through the apex of the left ventricle is unlikely to
be well tolerated off-pump. Equally, grasping of the leaflets from the ventricular side of
the valve using transesophageal echocardiography (TEE) guidance may be too imprecise
to guarantee a good repair. Potential improvements for the future include a flexible shaft
inserted transatrially through the left atrial appendage, or better yet, a totally percutane-
ous intravenous version introduced transeptally into the left atrium. The instrument can
Fig. 6. Grasper—fastener device in place.
then be placed across the plane of the annulus, apposing and immobilizing the leaflets
temporarily while the staple, screw, or suture is applied. Alfieri et al. recently reported
experience with a similar device that is introduced through the left atrial appendage
and uses suction to immobilize the leaflets while a central suture is applied. An integrated
knot pusher and cutter is subsequently used to secure the stitch. This technique was
reportedly successful in eight beating-heart sheep models, and the authors postulate that
it may be applicable to the treatment of ischemic regurgitation in conjunction with
revascularization procedures or mitral regurgitation in the setting of heart failure.
Although the bow-tie repair may be beneficial in ischemic mitral regurgitation, we
doubt that it will be effective in cases of regurgitation secondary to annular dilatation,
where the effective area of coaptation is severely reduced. This cohort of patients requires
an annular reduction procedure in order to restore the normal mitral leaflet area-to-mitral
valve orifice ratio of 2.5:1 (21). Recently, Timek and colleagues (22) determined that
certain echocardiographic predictors help in assigning patients who will benefit from
annular reduction procedures. Measurements in the four-chamber view relating to the
coaptation distance from the annular plane may separate those patients better suited for
repair vs replacement. Specifically, a distance less than 4–5 mm may indicate pure
annular dilation with little chordal apparatus abnormalities that may be efficiently cor-
rected by ring annuloplasty. Bolling and colleagues have treated functional mitral regur-
Fig. 7. Coronary sinus anatomy in relation to mitral valve annulus with tightening rod in situ.
gitation in patients with dilated cardiomyopathy with an undersized, complete, rigid

annuloplasty ring. They have observed that, in addition to unloading the ventricle, there
is “acute” remodeling of the base of the heart, with restoration of the ellipsoid shape of
the left ventricular cavity (21). Alternate strategies to reduce annular dimensions in
cardiomyopathic patients have been proposed by Buchanan and co-workers, who
hypothesized that an external banding procedure could reduce mitral regurgitation. They
showed that by positioning a tightening external circumferential band, annular morphol-
ogy could be changed and regurgitation improved in a mongrel dog model.
In the mid- to late 1990s others began to use routine cannulation and navigation of the
coronary venous system for in situ bypass and drug delivery, which raised the possibility
of intervening on certain adjacent structures such as the pericardium and mitral valve.
The coronary venous anatomy approaches most regions of the myocardium without
significant valvular obstruction or presence of atherosclerotic disease. Although these
venous conduits can be variable in their exact location and drainage pattern, the coronary
sinus, as it courses the AV grove to the great cardiac vein, is fairly predictable. In this
location, it circumnavigates a large portion of the outside of the mitral valve annulus
(Fig. 7). Using this conduit for the placement of a device, either percutaneously or
surgically, may provide a way to grip the annulus of the mitral valve and change its
geometry. Such a change could, in theory, help reduce mitral regurgitation in certain
pathological conditions. In congestive cardiomyopathy, the anterior, lateral, and poste-
rior segments of the annulus are at risk for dilatation. A mechanical rod that can be placed
into the coronary sinus extending to the great cardiac vein could be adjusted and “crimped”
from the vein, to tighten these three segments of the annulus. A percutaneous tightening
rod has been prototyped and delivered percutaneously through the internal jugular vein.
It has been placed into the coronary sinus and advanced into the great cardiac vein, near
the transition of the anterior interventricular vein but not extending into the interventricu-
lar groove (Fig. 7). The transition to the great cardiac vein can, in some instances, run
close to (and often cross over) the circumflex artery. This region needs to be avoided in
order not to transmit compression onto adjacent arterial structures. Navigation and place-
ment of such a device can be guided fluoroscopically with the use of routine cath lab
equipment. It is important to use transcutaneous echocardiography to characterize and
assess mechanical manipulation of the annulus—specifically, understanding the impact
on leaflet coaptation and reduction of mitral regurgitation as the device is tightened. If
this device favorably affects the anatomy and regurgitant volume, then disconnection can
be accomplished and the partial annuloplasty ring left in place permanently. Based on
chronic evaluation of many animal studies, there is no need for anticoagulation, as endot-
helium completely covers the device at 30 d. Additionally, no erosion or migration of the
device has been observed long term. Initial human feasibility studies are presently under-
way to determine if certain patient subsets plagued with large amounts of mitral regur-
gitation as part of their congestive heart failure constellation can be improved either
permanently or as a bridge to definitive surgical correction.
FUTURE TRENDS
Although these techniques may represent the future of mitral valve repair, use of either
one alone is unlikely to yield results comparable to conventional methods such as “the
French correction.” We postulate that at this stage in the development of this technology,
the best way to address valvular as well as subvalvular dysfunction, is by using both
techniques in combination. This could be done with an intravenous device placed in the
right atrium for insertion of the transcoronary sinus apparatus, which could then be
inserted transeptally across the mitral to apply an edge-to-edge suture.
REFERENCES
1. Reyes VP, Raju BS, Wynne J, et al. Percutaneous balloon valvuloplasty compared with open surgical
commissurotomy for mitral stenosis. N Engl J Med 1994;331:961–967.
2. Carabello BA, Crawford FA. Therapy for mitral stenosis comes full circle. N Engl J Med
1994;331:1014–1015.
3. David TE, Armstrong S, Sun Z. Left ventricular function after mitral valve surgery. J Heart Valve Dis
1995;4(suppl 2):S175–S180.
4. Carpentier A. Cardiac valve surgery—the “French correction.” J Thorac Cardiovasc Surg 1983;86:323–337.
5. Yun KL, Miller DC. Mitral valve repair versus replacement. Cardiol Clin 1991;9:315–327.
6. Gillinov AM, Cosgrove DM, Lytle BW, et al. Reoperation for failure of mitral valve repair. J Thorac
7. Gillinov AM, Cosgrove DM, Blackstone EH, et al. Durability of mitral valve repair for degenerative
disease. J Thorac Cardiovasc Surg 1998;116:734–743.
8. Enriquez-Sarano M, Tajik AJ, Schaff HV, Orszulak TA, Bailey KR, Frye RL. Echocardiographic
prediction of survival after surgical correction of organic mitral regurgitation. Circulation 1994;90:830–837.
9. Zile MR, Gaasch WH, Carroll JD, Levine HJ. Chronic mitral regurgitation: predictive value of preop-
erative echocardiographic indexes of left ventricular function and wall stress. J Am Coll Cardiol
1984;3:235–242.
10. Wisenbaugh T, Skudicky D, Sareli P. Prediction of outcome after valve replacement for rheumatic mitral
regurgitation in the era of chordal preservation. Circulation 1994;89:191–197.
11. Crawford MH, Souchek J, Oprian CA, et al. Determinants of survival and left ventricular performance
after mitral valve replacement: Department of Veterans Affairs Cooperative Study on Valvular Heart
Disease. Circulation 1990;81:1173–1181.
12. Hochreiter C, Niles N, Devereux RB, Kligfield P, Borer JS. Mitral regurgitation: relationship of
noninvasive descriptors of right and left ventricular performance to clinical and hemodynamic findings
and to prognosis in medically and surgically treated patients. Circulation 1986;73:900–912.
13. Ross J.The timing of surgery for severe mitral regurgitation. N Engl J Med 1996;335:1456–1458.
14. Felger JE, Chitwood RW, Nifong W, Holbert D. Evolution of mitral valve surgery: toward a totally
endoscopic approach. Ann Thorac Surg 2001;72:1203–1209.
15. Maisano F, Torraca L, Oppizzi M, et al. The edge-to-edge technique: a simplified method to correct
mitral insufficiency. Eur J Cardiothorac Surg 1998;13:240–246.
16. Alfieri O, Maisano F, De Bonis M, et al. The double-orifice technique in mitral valve repair: a simple
solution for complex problems. J Thorac Cardiovasc Surg 2001;122:674–681.
17. Umaña JP, Salehizadeh B, DeRose JJ, et al. “Bow-tie” mitral valve repair: an adjuvant technique for
ischemic mitral regurgitation. Ann Thorac Surg 1998;66:1640–1646.
18. Umaña JP, DeRose JJ, Choudhri A, et al. “Bow-tie” mitral valve repair successfully addresses subvalvular
dysfunction in ischemic mitral regurgitation. Surg Forum 1997;48:279–280.
19. Llaneras MR, Nance ML, Streicher JT, et al. Large animal model of ischemic mitral regurgitation. Ann
Thorac Surg 1994;57:432–439.
20. Morales DLS, Madigan JD, Coudhri AF, et al. Development of an off bypass mitral valve repair. Heart
Surg Forum 1999;2:115–120.
21. Bolling SF, Smolens IA, Pagani FD. Surgical alternatives for heart failure. J Heart Lung Transplant
2001;20:729–733.
22. Timek TA, Dagum P, Lai DT, et al. Pathogenesis of mitral regurgitation in tachycardia-induced cardi-
omyopathy. Circulation 2001;104(suppl I):I-47–I-53.
Chapter 28 / Alternative Anastomotic Techniques 393
28 Alternative Anastomotic Techniques
David A. D’Alessandro, MD
and Mehmet C. Oz, MD
CONTENTS
INTRODUCTION
MECHANICAL ANASTOMOTIC SYSTEMS
VASCULAR CONNECTORS
SYMMETRY ™ BYPASS SYSTEM
AUTOMATIC ANASTOMOTIC DEVICE
ONE SHOT SYSTEM
HEARTFLO™
GRAFTCONNECTOR
VENTRICA
VASCONNECT
ANASTOMOTIC ADHESIVES
BIOGLUE
LASER WELDING
ROBOTIC SURGERY
CONCLUSION
REFERENCES
INTRODUCTION
Technical advances in percutaneous techniques for coronary artery disease have less-
ened the need for surgical revascularization. Although the superior patency rates achieved
by surgical approaches are widely recognized, the morbidity associated with surgery
limits the overall outcome. For this reason, surgeons have continued to refine techniques
to minimize morbidity. Examples include the development of minimally invasive direct
coronary artery bypass (MIDCAB), off-pump coronary artery bypass (OPCAB), and,
most recently, robotics. This evolution in cardiac therapies represents the focus of this book.

393
Current methods of surgical graft anastomoses necessitate time-consuming hand-

sewn techniques and generous exposure. Many investigators believe that the future of
coronary bypass surgery will involve increased use of robotics and minimalist approaches.
Various alternative means to traditional suturing have been proposed and tested to achieve
patent, hemostatic anastomoses. These include use of lasers, glues, clips, and mechanical
devices. These strategies have been used by themselves or in combination to improve or
advance the art and technique of coronary revascularization surgery and are the focus of
this chapter.
The technologies described herein may provide the following potential benefits:
1. They may reduce operative time and time spent on cardiopulmonary bypass (CPB).
2. They may improve the quality of coronary anastomoses while limiting intersurgeon
disparity.
3. They may reduce the need for aortic manipulations.
4. They may minimize needed operative exposure.
5. They may create superior vascular connections translating into improved long-term
outcomes.
In order to achieve these goals and to become commercially successful, these tech-
nologies must meet all the following requirements:
1. They must demonstrate reproducible results regardless of both surgeon and patient vari-
ability.
2. They must demonstrate long-term patency results equivalent or superior to those of
suture techniques.
3. They must create hemostatic anastomoses.
4. They must not add exorbitant cost.
The remainder of this chapter will describe the evolution of the various anastomotic
strategies and the more prominent emerging commercial systems. One should evaluate
each technology based on its theoretic ability to achieve the above-stated benefits while
meeting these requirements.
MECHANICAL ANASTOMOTIC SYSTEMS

The surgical stapler is perhaps the most widely used tool for mechanical anastomoses.
All staplers have the same principal design. A pusher system advances a U-shaped wire
through two or more objects to be joined and into an anvil that bends the wire in some
fashion to secure them. Over 200 years ago, the French literature reported the use of a
stapler for large-vessel repair (1). While staples became commonplace in other indus-
tries, they did not gain surgical notoriety until much later.
Humer Hültl, a Hungarian surgeon, pioneered surgical stapling, introducing a gas-
trointestinal stapler that he used in a series of gastrectomies in 1906 (2). Von Brücke later
introduced a device that placed individual staples in 1935 (3). Another Hungarian group
was among the first to develop a vascular anastomotic stapler, as presented by Bikfalvi
and Dubecz in 1953 (4). Intended for larger (6–8-mm) vessels, this device joined everted
vessels using U-shaped silver clasps. While this represented a technical milestone, it was
not immediately applicable to small vessel anastomoses.
About this time, a Russian group of engineers and surgeons was refining a similar but
more sophisticated device. Established in 1952, the Soviet Scientific Research Institute
for Experimental Surgical Apparatus and Instruments brought together multiple disci-
Fig. 1. Androsov stapling device pictured assembled and in component halves (Reprinted with
permission from Surgery.)
plines to develop a vascular stapler based on the work done by Gudov beginning in 1941
(5). Resulting from this collaboration, in 1956, Androsov described the use of an auto-
mated, circular stapling device for joining damaged blood vessels in an end-to-end fash-
ion (6). In contrast to previous designs, this device joined vessels from 1.3 to 15 mm in
diameter using U-shaped clips constructed of tantalum wire (Fig. 1). These early inves-
tigators appreciated the importance of intimal apposition and the advantage of avoiding
an intimal foreign body. Nevertheless, their device, although elaborate, was cumbersome
and costly (7), and it never achieved commercial success.
In 1958, Inokuchi, a Japanese surgeon, described a simplified apparatus employing the
same principles (8). He further compared the stapled anastomosis with current suture
technique, demonstrating reduced early thrombogenicity and improved late histological
characteristics of the automated anastomoses. Also in 1958, Vogelfanger and Beattie
reported a similarly simplified Canadian stapler (9). A U.S. version was later developed
with disposable staple inserts to simplify operative use (10). This device was intended to
give general surgeons the ability to perform small-vessel anastomoses and even nerve
repairs with an easy-to-use, affordable device. Finally, in 1960, Takaro (11) introduced
a simple stapling device that he used to anastomose Dacron interposition grafts to dog
aortas. This anastomotic system was unique in that the Dacron graft was stapled within
the lumen of the recipient vessel, thus avoiding the need to evert the edges of the native
artery. While Takaro’s device showed foresight into the difficulty of everting diseased
vessels, his experimental work demonstrated these stapled anastomoses to be inferior in
quality to those that were hand-sewn.
Automated vascular staplers were developed primarily for use in traumatic vascular
injury and limb amputation, offering a more rapid and precise repair of relatively
nondiseased vessels. Some also foresaw their use in the emerging field of organ and limb
transplantation, in which rapid reestablishment of blood flow was paramount in the era
before the establishment of preservation techniques. Enthusiasm for these early vascular
stapling devices was in part fueled by the results achieved with hand-sewn methods in the
era of catgut- or braided silk-sutured anastomoses and prior to the widespread availability
of magnification. Subsequent advances in microvascular technique, loupe magnifica-
tion, and suture materials such as polypropylene lessened the benefit of vascular staplers.
Additionally, hand-sewn techniques were more adaptable to the diseased vessels com-
monly encountered in vascular surgery. As a result, vascular staplers have played little
historical role in the field of cardiac surgery.
The modern era of vascular stapling perhaps began in 1992 with the introduction by
Kirsch and colleagues of an arcuate-legged clip (12). These clips were originally devel-
oped for use in neurosurgical vascular anastomoses, but the benefits of this technology,
which created vascular connections with minimal endothelial injury, were widely recog-
nized. The VCS Auto Suture device (US Surgical Corp., Norwalk, CT) is a later-genera-
tion clip dispenser. This device places titanium arcuate-legged clips, which come in four
different sizes and can dispense 25 clips in quick succession (Fig. 2). Nataf et al. (13)
reported their experience with these clips in 10 patients undergoing conventional coro-
nary artery bypass grafting (CABG). Nonpenetrating clips have theoretical advantages
over sutures (14,15), but these authors note that they are difficult to apply in patients with
calcified coronaries. Calcifications complicate vessel edge eversion, thus compromising
anastomotic integrity. Despite the availability of this device, few surgeons have adopted
these stapling techniques, and the current role of this technology in cardiac surgery is limited.
VASCULAR CONNECTORS
Concurrent with the development of stapling devices, intra- and extraluminal coupling
and stent devices were also under development. In 1900, Payr reported the use of an
intraluminal, absorbable magnesium ring for vascular end-to-end coupling (16). A
decade later, Lespinasse and colleagues (17) published their development of a magne-
sium ring anastomosis that provided vascular coupling without the presence of an endot-
helial foreign body. Nakayama et al. (18) later described the clinical use of a tantalum ring
anastomotic system for coupling small (1.5–4-mm) arteries in either end-to-end or end-
to-side fashion.
Blakemore and Lord (19) introduced a Vitallium-alloy cannulae system for perform-
ing vein interposition arterial anastomoses. These authors targeted their system for the
repair of wartime extremity injuries, where rapid restoration of blood flow is vital for limb
salvage. They reported a 90% patency rate when bridging femoral artery defects in dogs.
Fig. 2. Technique for performing a stapled microvascular anastomosis is illustrated. (A) Everting
forceps are used to stabilize the vessels while clips are applied form right to left. (B) The vessels
are rotated 180 degrees and the posterior walls are similarly joined. (Reprinted with permission
from the Annals of Thoracic Surgery.)
Inspired by this work and others, Carter and Roth (20) later reported their experience with
polyethylene rings for use in nonsutured anastomoses of the left internal mammary artery
to the circumflex coronary artery in dogs. These authors achieved an 83%, 2-mo patency
rate. The main advantage of their technique was speed, as they could perform these
anastomoses in less than 3 min, a marked improvement over standard suture technique.
In 1960, Rohman and others (21) described the use of a tantalum ring for internal
mammary-to-coronary anastomoses in dogs, achieving 18-mo graft patency in some
animals. Several others demonstrated experimental success with ringed or ring-pin sys-
tems (22–24). Lemole (25) recently described a collar-and-punch method designed to
facilitate end-to-side or side-to-side anastomoses. This system allows the creation of an
anastomosis without the need to occlude blood flow in the native vessel. A sleeve-and-
collar combination attached to the graft vessel is sutured to the target vessel prior to
making an arteriotomy. Once secured, a punch system is employed to create the arteri-
otomy and thus complete the anastomosis. This system seems particularly suited for off-
pump coronary artery bypass (OPCAB) procedures, in which temporary occlusion of
coronary flow may be detrimental. It awaits commercial development.
As with vascular staplers, historical interest in these and other anastomotic systems
waned concurrent with improvements in hand-sewn techniques. Early devices were often
cumbersome and hampered by intolerable thrombosis rates, despite sometimes achieving
long-term successes. More recent adaptations have had difficulty gaining acceptance and
commercial success in an arena dominated by standardized suture techniques with proven
efficacy and durability. Current anastomotic devices combine stapling and intraluminal
stent technology to perform assisted or total automated anastomoses. The leading designs
in or approaching the marketplace will be described subsequently. The present impetus
driving these automated anastomotic systems is the need to perform vascular connections
using minimal exposure, with a thrust toward total endoscopic approaches. While per-
haps none of these devices has achieved this goal, as a group they represent a necessary
step in the evolution of minimal-access surgery. Clinical experience with these and future
designs will lead to technological refinements necessary for a total endoscopic anasto-
motic tool.
Fig. 3. Schematic of the Symmetry™ Bypass System deliver system with a loaded vein graft.
(Printed with permission from St. Jude Medical, Inc.)
SYMMETRY™ BYPASS SYSTEM

The St. Jude Medical Anastomotic Technology Group (Minneapolis, MN) has invested
considerably in systems designed for creating both proximal aortosaphenous and distal
coronary anastomoses. Among these, the Symmetry™ Bypass System, Aortic Connector,
is currently the only U.S. Food and Drug Administration (FDA)-approved and commer-
cially available automated device for proximal anastomoses. This device facilitates the
creation of a sutureless anastomosis via a Nitinol, self-expanding connector. When using
this system, the operator places the harvested saphenous vein conduit in the delivery
system (Fig. 3). One of four stent sizes is selected. The vein is guided into the delivery
system and the ends everted over the connector hooks. An aortic cutter is next used to
make a clean, circular aortotomy, essential for proper stent deployment and hemostasis.
Finally, the graft is delivered and a 90° stented anastomosis is performed in seconds,
completing the process (Fig. 4). The obvious advantage of this and other similar devices
is their ability to perform aortic anastomosis without the need for clamping, a maneuver
that has been associated with atheroemboli and stroke (26). Eckstein et al. (27) reported
the first clinical series using this device in 20 consecutive patients undergoing OPCAB.
This device is currently gaining popularity among surgeons in the United States, but its
efficacy and effects on postoperative neurological morbidity remain to be established. In
the current era of cost containment, the cost of this device may prove too high for many
clinicians.
Currently, St. Jude is developing a stented system for sutureless side-to-side coronary
anastomoses designed for use with saphenous vein conduits. This delivery system
employs a balloon expandable intraluminal stainless steel stent that facilitates a hemo-
static anastomosis in less than 3 min. Eckstein et al. (28) recently reported the first clinical
use of this anastomotic device. This system awaits FDA approval.
Fig. 4. (Right) Cartoon depicting the Symmetry™ Bypass System’s cut then connect sequence to
a complete anastomosis. (Printed with permission from St. Jude Medical, Inc.).
Fig. 5. Cartoon depicting Ethicon’s Automatic Anastomotic Device deployment sequence. (A)
Punching device creates a 3.2 mm hole in the unclamped aortic wall. (B) Aortic punch is with-
drawn and an internal sealing ring prevents aortic backflow. (C) The delivery system is advanced
into the aortic lumen. (D) Mounted vein graft is delivered into the aortic lumen. Once in place (not
shown), inner pins partially penetrate the aortic intima while outer pins stabilize the graft to the
adventitia. (Printed with permission from Ethicon, Inc.)
AUTOMATIC ANASTOMOTIC DEVICE

Marketed by Ethicon, Inc. (Somerville, NJ), the Automatic Anastomotic Device is a
Nitinol extraluminal stent system that combines aortotomy and aortosaphenous anasto-
mosis in a single step. Use of this device requires previous mounting of the conduit vein
on the delivery system, everting the vein edges over the distal pins. Next, a punching
device is inserted through the delivery handle, creating an aortotomy, and is withdrawn
through a sealing ring that prevents excessive bleeding. The loaded vein graft is then
inserted through the aortotomy and deployed (Fig. 5). The self-expanding Nitinol stent
system creates a hemostatic, intima-to-intima seal. Once the vein graft is loaded, a
maneuver that can be performed by support staff on the back table, the entire aortotomy
and delivery maneuver can be completed in less than 2 min. As with the St. Jude device,
this system does not require aortic side clamping and can be performed with minimal
aortic manipulation. The unique aspect of this device is a delivery system that creates an
aortotomy and places the vein conduit in a single maneuver. Once the aortotomy is
created, the applicator is never removed from the aorta. This will facilitate placement of
Fig. 6. Close-up of the one-shot anastomotic stapler. The graft vessel is hooked and pulled through
the clip housing. The vessel end is then everted across the anvil and secured over the ends of the
clips. The loaded cartridge is next placed in the firing handle and deployed. (Reprinted with
permission from the Journal of Thoracic and Cardiovascular Surgery.)
veins from a distance. Calafiore et al. (29) reported limited clinical experience with the
first generation of this device, with promising results. They noted several limitations that
will be addressed in later versions.
ONE SHOT SYSTEM

Developed by United States Surgical Corporation (Norwalk, CT), the One Shot Sys-
tem places 10 or 12 evenly spaced titanium clips to everted and approximated endothe-
lium. The system utilizes arcuate-legged titanium clips which are nonpenetrating and
which autoregulate closing pressure depending on the thickness of interposed tissue. The
application device instantaneously applies a ring of clips that are both hemostatic and
equivalent to conventional sutures with respect to tensile and burst strengths. The system
can be used to create both proximal and distal anastomoses in either an end-to-side or end-to-
end fashion. The donor graft is positioned through a disposable cartridge housing and the
edges are everted over the distal clips (Fig. 6). The donor graft is then positioned on the
target site at a 45° angle and the clips are fired simultaneously, creating an anastomosis
with direct intimal apposition and an uninterrupted endothelium. Disposable cartridges
are equipped with variable clip sizes ranging from 10 to 12 in number and can be used
on target vessels less than 2 mm in diameter.
This device has been tested extensively in animal models and in cadaveric studies and
more recently in a clinical trial creating arteriovenous access fistulas (30). Heijmen et al.
(31) reported their experience with a device prototype creating distal anastomoses in a
porcine OPCAB model, noting several shortcomings. While the One Shot device received
FDA marketing approval in 1997, technical difficulties have prevented its commercial
use in cardiac surgery. US Surgical recently suspended research and development of the
One Shot System due to ongoing design concerns.
HEARTFLO™
Developed by Perclose/ Abbott Labs (Redwood, CA), the Heartflo™ device simulta-
neously deploys 10 interrupted 7-0 polypropylene sutures (Figs. 7 and 8). Once placed,
the sutures are then hand-tied in the conventional fashion. This device can facilitate both
Fig. 7. The Heartflo automated anastomotic device. (Reprinted with permission from the Annals
of Thoracic Surgery.)
side-to-side and end-to-side anastomoses. Tozzi and co-workers (32) demonstrated

superior anastomotic properties using the Heartflo system compared to running sutured
anastomoses. Shennib et al. (33) described their initial experience with this device in a
porcine model, performing IMA-to-LAD anastomoses in a side-to-side fashion. While
the authors found this prototype design cumbersome and time-consuming to operate,
they noted that the next-generation Heartflo addresses these limitations and is modified
for use in the total endoscopic setting.
GRAFTCONNECTOR
Developed by Jomed International AB (Helsinborg, Sweden), the GraftConnector™
is a Nitinol-stented anastomosis system that combines polytetrafluorethylene (PTFE)
and vascular conduits, and facilitates a sutureless anastomosis. This system, designed for
target site, end-to-side anastomoses, requires a four-step delivery sequence (Fig. 9).
Solem et al. (34) evaluated this device in an animal OPCAB model and showed similar
patency rates and shorter anastomotic times when compared to standard surgical tech-
nique. More recently, Tozzi and colleagues (35) presented 6-mo animal data that docu-
mented increased anastomotic diameters and conduit blood flow when using this device,
compared to running suture technique. Furthermore, these authors demonstrated this
device’s ease of use, requiring less manual dexterity than hand-sewing methods, an
important requirement of any system with applicability in minimally invasive cardiac
surgery. These studies, however, compare GraftConnector anastomoses with hand-sewn
anastomoses in an OPCAB animal model without the use of a cardiac stabilizer. Although
this experimental design highlights the ease of use of this system, one may question the
precision of hand sewing in such a scenario and therefore its clinical relevance. While this
Fig. 8. External and internal views of the automated anastomosis. (Reprinted with permission from
the Annals of Thoracic Surgery.)
system is intriguing and deserving of further study, it has not been evaluated in the clinical
arena and remains in the development phase.
VENTRICA
The Magnetic Vascular Positioner (MVP™), developed by Ventrica®, Inc. (Freemont,
CA), represents one of the more novel approaches to alternative anastomoses. Introduced
in 2001 (36), this system utilizes the self-aligning and self-sealing characteristics of
magnetic attraction to achieve rapid, hemostatic vascular connections. The MVP system
Fig. 9. Schematic of the GraftConnector (GC) anastomotic sequence. (1) The graft vessel is
positioned in the sidearm of the GC. (2–3) An arteriotomy is created and the stent-system is
inserted into the target artery. (4) The device is deployed and the stented graft self-expands to
secure and complete the anastomosis. (Reprinted with permission from the Annals of Thoracic
Surgery.)
accomplishes this by a delivery mechanism that inserts elliptical magnetic couplers in

both conduit and recipient vessels. The magnets are joined to create an instantaneous
anastomosis (Figs. 10 and 11).
Ventrica is concurrently developing the MagnePort™ system, which utilizes the same
technology to provide proximal aortic vascular connections without the need for aortic
clamping. These systems await FDA approval and are currently undergoing preclinical
and clinical testing. Their design simplicity, however, seems particularly well suited for
future, minimally invasive cardiac revascularization procedures.
VASCONNECT
Coalescent Surgical, Inc. (Sunnyvale, CA), has developed another unique and inno-
vative concept for vascular anastomoses. Their nitinol U-Clip Device combines the speed
and versatility of Nitinol with the flexibility of hand-sewing technique. This system
consists of a self-closing Nitinol wire attached to a needle delivery member (Fig. 12).
Once detached, the Nitinol clip assumes its preformed shape, approximating tissue simi-
lar to suture but obviating knot tying. The U-Clip device was evaluated in a bovine model
by Hill et al. (37) and proved effective for use in an OPCAB model. These authors point
out that such a device, which allows rapid deployment of interrupted clips using conven-
tional sewing technique, may prove superior to continuous suture technique, as it allows
anastomotic expansion. The simplified anastomotic technique that this system provides
also seems well suited for minimally invasive and robotic-assisted approaches.
Fig. 10. Delivery system with close up and schematic of the magnetic connection. (Printed with
permission form Ventrica, Inc.)
Fig. 11. (A) The MVP™ is inserted into standard arteriotomy. (B) The MVP™ delivery device
is activated to create the anastomotic port. (C) The identically prepared graft is approximated to
the artery. (D) A virtually instantaneous anastomotic connection is created. (Printed with permis-
sion form Ventrica, Inc.)
Fig. 12. The Nitinol U-Clip device. The U-Clip is placed like a conventional suture via a tapered
needle and utilizes nitinol alloy shape memory to execute a rapid vascular connection. (Reprinted
with permission form the Journal of Thoracic and Cardiovascular Surgery.)
ANASTOMOTIC ADHESIVES
Recognizing the limitations of hand-sewn anastomoses, several groups have sought a
sutureless alternative to joining vessels using adhesives. Braunwald and others first
developed the gelatin–resorcin–formaldehyde (GRF) glue for use in vascular surgery
(38). Since its development, GRF glue has been used to treat acute aortic dissection (39)
and more recently has been applied to PTCA balloon-assisted small vessel anastomoses
(40). Application of other biological glues such as cyanoacrylates (41), fibrin glues (42),
and urethane polymers has been reported with similar results. While a glued anastomosis
has theoretic advantages, several limitations have stymied their clinical adoption. These
include concerns over tissue toxicity, lack of flexibility, inadequate tensile strength,
intraluminal thrombosis, and pseudo-aneurysm formation.
BIOGLUE
BioGlue (CryoLife, Inc., Marietta, GA) is a gluteraldehyde compound containing
concentrated bovine serum albumin. Once mixed in the applicator tip, this compound
forms a strong adhesive with a setting time of under 2 min. The glue is biologically inert
and is under investigation for treatment of aortic dissection. Gundry and colleagues (43)
reported on the use of this glue for constructing coronary anastomoses with a catheter-
assisted procedure (Fig. 13). They further demonstrated long-term patency in 2 animals
(10 mo and 1 yr) without evidence of intraluminal thrombosis or anastomotic aneurysms,
the usual pitfalls with glued anastomoses.
While totally glued anastomoses have yet to become a clinical reality, such agents
might become important adjuncts to alternative anastomotic approaches. Glue can be
easily applied via catheters and may prove invaluable in hemostasis in a limited endo-
scopic surgical field.
LASER WELDING
Jain and Gorisch first reported the use of lasers for vascular repair in 1979 (44). The
following year, Morris and Carter (45) described use of the CO2 laser for microanastomoses.
Since that time, a myriad of energy sources have been employed in both vascular repair
and in anastomotic construction (46–48). The mechanism by which lasers function
involves the breaking and re-forming of protein bonds in a process that has been termed
“welding.” The theoretical advantages of laser-assisted anastomoses are speed, their
ability to grow, the minimal foreign body reaction, and improved long-term patency.
Fig. 13. Schematic of BioGlue anastomosis. (1) Graft and target vessels are approximated and
arteriotomies are performed. (2) Balloon catheters are placed as shown and BioGlue is applied and
allowed to set for 2 min. (3) Catheters are removed restoring flow and the distal graft vessel is
clipped completing the anastomosis. (Reprinted with permission form the Journal of Thoracic and
Cardiovascular Surgery.)
Critics have raised concern over the long-term potential for anastomotic disruption or
aneurysm formation when applying laser techniques to larger vessels (49). Nakata and
co-workers (50), however, demonstrated CO2 laser anastomotic bursting pressures
equivalent to that of standard suture anastomoses without propensity for aneurysm for-
mation in rabbit carotid arteries. They further demonstrated reduced long-term intimal
hyperplasia, suggesting a role for lasers in coronary bypass surgery.
The efficiency of laser welding has since been improved by the addition of dye solders
(51–53) to the vessel surface, which increases the absorbance of the laser light and thus
increases heat deposition. Utilizing this effect, Maitz and colleagues (54) adapted the
original Payr technique, substituting a protein tube for the absorbable magnesium ring,
and successfully performed sutureless anastomoses.
Although these techniques have been utilized in clinical microvascular surgery appli-
cations, they have not gained widespread acceptance in cardiovascular surgery. This is
perhaps due to ongoing concern over the short- and long-term durability of these anas-
tomoses in a setting where sudden disruption could be catastrophic. Furthermore, the
laser energy sources are cumbersome and necessitate the use of protective eyewear.
Lastly, laser-assisted anastomoses are difficult to perform in a limited surgical field.
ROBOTIC SURGERY
Many believe that the future of surgical practice will rely heavily on robotics. Robotic
equipment affords several advantages to traditional open or laparoscopic procedures.
These include binocular three-dimensional viewing, smaller access port requirements,
movement filters that improve manual dexterity, and improved vision from multiple
camera angles. Robotic surgery has only recently entered the cardiac surgical arena, but
several centers have amassed an early experience (55–58). The trend toward minimally
invasive coronary artery bypass surgery has led first to the MIDCAB, then the OPCAB,
and now the total endoscopic coronary artery bypass (TECAB), which can be performed
off-pump. While computer-assisted robotics enable hand-sewing technique, exposure is
limited and generally only single-vessel bypasses can be currently performed. Improved
instrumentation and refinements in technique will undoubtedly extend the reach of this
technology to include patients with multivessel disease. Systems that facilitate the anas-
tomotic process will certainly find use in this burgeoning arena.
CONCLUSION
Automated graft anastomotic systems have been used successfully, but their overall
effectiveness has not been fully evaluated. Ultimately, such systems will become neces-
sary as cardiac surgery continues its evolution toward less invasive techniques. As inves-
tigators and industry continue the quest for a usable anastomotic device, several challenges
must be surmounted. Importantly, the effectiveness of any device on the research bench
or in animals with healthy arterial anatomy should be interpreted with caution. For
clinical utility, a device must create reproducible anastomoses in varied target site loca-
tions and on diseased, calcified vessels that are characteristic in this patient population.
The prevailing anastomotic technology may become as fundamental to coronary surgery
as polypropylene has been over the past several decades. This financial incentive will
drive these technologies and reshape the field of coronary bypass surgery.
REFERENCES
1. Carrel A. Results of transplantation of blood vessels, organs, and limbs. JAMA 1908;51:1662.
2. Cooper P, Christie SG. Develpoment of the surgical stapler with emphasis on vascular anastomosis.
Trans N Y Acad Sci 1963;25:365–377.
3. Von Brücke H, Über ein neuartiges chirurgisches Nähinstrument. Zentralbl Chir 1935;62:1684–1686.
4. Bikfalvi A, Dubecz S. Observations in animal experiments with mechanized vessel suture. J Int Chir
1953;5:481–495.
5. Mallina RF, Miller TR, Cooper P, Christie SG. Surgical stapling. Sci Am 1962;207:48–56.
6. Androsov PI. New method of surgical treatment of blood vessel lesions. AMA Arch Surg 1956;73:902–920.
7. Miller TR, Corso PF, Mallina RF. Preliminary evaluation of the Adrosov stapling device for the circular
suture of blood vessels. Surgery 1962;51:216–219.
8. Inokuchi. A new tyope of vessel-suturing apparatus. AMA Arch Surg 1958;77:954–957.
9. Vogelfanger IJ, Beattie WG. A concept of automation in vascular surgery: a preliminary report on a
mechanical instrument for arterial anastomosis. Can J Surg 1958;1:262–265.
10. Mallina RF. The technical aspects of the vascular stapler. Trans N Y Acad Sci 1963;253–264.
11. Takaro T. A simple stapling device for the anastomosis of blood vessels. J Thorac Cardiovasc Surg
1960;40:673–684.
12. Kirsch WM, Zhu YH, Hardesty RA, Chapolini R. A new method for microvascular anastomosis. Am
Surg 1992;58:722–727.
13. Nataf P, Kirsch W, Hill AC, et al. Nonpenetrating clips for the coronary anastomosis. Ann Thorac Surg
1997;63:S135–S137.
14. Boeckx WD, Darius O, van den hof B, van Holder C. Scanning electron microscope analysis of the
stapled microvascular anastomosis in the rabbit. Ann Thorac Surg 1997;63:S128–S134.
15. Zeebregts CJ, van den Dungen JJ, Kalicharan D, Cromheecke M, van der Want J, van Schilfgaarde R.
Nonpenetrating vascular clips for small-caliber anastomosis. Microsurgery 2000;20:131–138.
16. Payr E. Beitrage zur Technique der Blutgefass- und Nervennaht nebst Mittheilungen über die
Verwendung eines resorbirbaren Metalles in de Chirurgie. Arch Klin Chir 1900;62:67–93.
17. Lespinasse VD, Fisher GC, Eisenstaedt J. A practical mechanical method of end-to-end anastomosis of
blood-vessels. JAMA 1910;55:1785–1790.
18. Nakayama K, Tamiya T, Yamamoto K, Akimoto S. A simple new apparatus for small vessel anastomosis
(free autograft of the signmoid included). Surgery 1962;52:918–931.
19. Blakemore AH, Lord JW. A nonsuture method of blood vessel anastomosis. JAMA 1945;127:685–691.
20. Carter EL, Roth EJ. Direct nonsuture coronary artery anastomosis in the dog. Ann Surg 1958;148:212–218.
21. Rohman M, Goetz RH, Dee R. Double coronary artery-internal mammary tantalum ring technique. Surg
Forum 1960;11:236–237.
22. Carton CA, Kessler LA, Seidenberg B, Hurwitt ES. Experimental studies in the surgery of small blood
vessels. Nonsuture anastomosis of arteries and veins using flanged ring prosthesis and plastic adhesive.
Surg Forum 1960;11:238–239.
23. Ratan RS, Leon M, Lovette JB, Levowitz BS, Magovern GJ, Kent EM. Modified nonsuture anastomosis
of coronary artery and internal mammary artery in dogs. Surg Forum 1960;11:239–241.
24. Holt GP, Lewis FJ. A new technique of end-to-end anastomosis of small arteries. Surg Forum
1960;11:242–243.
25. Lemole GM. Procedure for bypassing an occlusion in a blood vessel. US Patent 5,893,369. 1999 Apr 13.
26. Sylivris S, Levi C, Matalanis G, et al. Pattern and significance of cerebral microemboli during coronary
27. Eckstein FS, Bonilla LF, Englberger L, et al. Minimizing aortic manipulation during OPCAB using the
Symmetry Aortic Connector System for proximal vein graft anastomoses. Ann Thorac Surg
2001;72:S995–S998.
28. Eckstein FS, Bonilla LF, Meyer B, et al. Sutureless mechanical anastomosis of a saphenous vein graft
to a coronary artery with a new connector device. Lancet 2001;357:931–932.
29. Calafiore AM, Bar-El Y, Vitolla G, et al. Early clinical experience with a new surtureless anastomotic
device for proximal anastomosis of the saphenous vein to the aorta. J Thorac Cardiovasc Surg
2001;121:854–858.
30. Nataf P, Hinchliffe P, Manzo S, et al. Facilitated vascular anastomoses: The one-shot device. Ann
Thorac Surg 1998;66:1041–1044.
31. Heijmen RH, Hinchliffe P, Borst C, et al. A novel one-shot anastomotic stapler prototype for the
coronary bypass grafting on the beating heart: feasibility on the pig. J Thorac Cardiovasc Surg
1999;117:117–125.
32. Tozzi P, Hayoz D, Ruchat P, et al. Animal model to compare the effects of suture technique on cross-
sectional compliance on end-to-side anastomoses. Eur J Cardiothorac Surg 2001;19:477–481.
33. Shennib H, Korkola SJ, Bousette N, Giaid A. An automated interrupted suturing device for coronary
artery bypass grafting: automated coronary anastomosis. Ann Thorac Surg 2000;70:1046–1048.
34. Solem JO, Boumzebra D, Al-Buraiki J, Nakeeb S, Rafeh W, Al-Halees Z. Evaluation of a new device
for quick sutureless coronary artery anastomosis in surviving sheep. Eur J Cardiothorac Surg
2000;17:312–318.
35. Tozzi P, Solem JO, Boumzebra D, et al. Is the GraftConnector a valid alternative to running suture in
end-to-side coronary artery anastomoses? Ann Thorac Surg 2001;72:S999–S1003.
36. Presentation. Advanced Cardiac Techniques Meeting. New York, NY. May 2001.
37. Hill AC, Maroney TP, Virmani R. Facilitated coronary anastomosis using a nitinol U-Clip device:
bovine model. J Thorac Cardiovasc Surg 2001;121:859–870.
38. Braunwald NS, Gay W, Tatooles CJ. Evaluation of cross-linked gelatin as a tissue adhesive and hemo-
static agent: an experimental study. Surgery 1966;59:1024–1030.
39. Bachet J, Goudot B, Teodori G, et al. Surgery of type A acute aortic dissection with gelatin-resorcin-
formol biological glue: a twelve-year experience. J Cardiovasc Surg 1990;31:263–273.
40. Oiwa H, Takamoto S, Murakami A, Tanaka O, Furuse A. Experimental study of small arterial anasto-
mosis with gelatin-resorcon-formaldehyde glue and collagen sheet. Artif Organs 2001;25(4):281–291.
41. Takenaka H. Sutureless anastomosis of blood vessels using cyanoacrylate adhesives. Jpn J Surg
1992;22:46.
42. Dowbak GM, Rohrich RJ, Robinson JB, Peden E. Effectiveness of a new non-thrombogenic bio-adhe-
sive in microvascular anastomoses. J Reconstruct Microsurg 1994;10(6):383–386.
43. Gundry SR, Black K, Izutani H. Sutureless coronary artery bypass with biologic glued anastomoses:
preliminary in vivo and in vitro results. J Thorac Cardiovasc Surg 2000;120:473–477.
44. Jain KK, Gorisch W. Repair of small blood vessels with the neodynium-YAG laser: a preliminary report.
Surgery 1979;85:684–688.
45. Morris JR, Carter M. Laser assisted microvascular anastomosis (LAMA) (abstr). Orthopedic Research
Society. Las Vegas, NV, 1980.
46. Gomes OM, Macruz Z, Armelin E, Aerbini EJ. Vascular anastomosis by argon laser beam. Texas Heart
Instit J 1983;10:145–149.
47. Frazier OH, Painvin GA, Morris JR, Thomsen S, Neblett CR. Laser-assisted microvascular anastomosis:
angiographic and anatomopathologic studies on growing microvascular anastomoses: preliminary
report. Surgery 1985;97:585–589.
48. Lewis WJ, Uribe A. Contact diode laser microvascular anastomosis. Laryngoscope 1993;103:850–853.
49. Quigley MR, Bailes JE, Kwaan HC, Cerullo LJ, Brown JT. Aneurysm formation following low power
CO2 laser-assisted vascular anastomosis. Neurosurgery 1986;18:292–299.
50. Nakata S, Campbell CD, Pick R, Replogle RL. End-to-side and end-to-end vascular anastomoses with
a carbon dioxide laser. J Thorac Cardiovasc Surg 1989;98:57–62.
51. Reali UM, Gelli R, Giannotti V, Gori F, Pratesi R, Pini R. Experimental diode laser-assisted microvas-
cular anastomosis. J Reconstr Microsurg 1993;9:203–210.
52. Oz MC, Johnson JP, Parangi S, et al. Tissue soldering by use of indocyanine green dye-enhanced
fibrinogen with near infrared diode laser. J Vasc Surg 1990;11:718–725.
53. Chuck RS, Oz MC, Delohery TM, et al. Dye-enhanced laser tissue welding. Lasers Surg Med
1989;9:471–477.
54. Maitz PKM, Tricket RI, Dekker P, et al. Sutureless microvascular anastomoses by a biodegradable laser-
activated solid protein solder. Plastic Reconstr Surg 1999;104:1726–1731.
55. Falk V, Diegeler A, Walther T, et al. Total endoscopic coronary artery bypass grafting. Eur J Cardiothorac
Surg 2000;17:38–45.
56. Falk V, Autschbach R, Walther T, Diegeler A, Chitwood WR, Mohr FW. Computer enhanced mitral
valve surgery—towards a total endoscopic procedure. Sem Thorac Surg 1999;11:244–249.
57. Reuchenspurner H, Damiano RJ, Mack M, et al. Use of the voice-controlled and computer-assisted
surgical system Zeus for endoscopic coronary artery bypass grafting. J Thorac Cardiovasc Surg
1999;118:11–16.
58. Mohr FW, Falk V, Diegeler A, et al. Computer-enhanced “robotic” cardiac surgery: experience in 148
patients. J Thorac Cardiovasc Surg 2001;121(5)842–853.
Chapter 29 / Making Cardiopulmonary Bypass Less Invasive 411
29 Making Cardiopulmonary Bypass

Less Invasive
James R. Beck, CCP, Linda B. Mongero, CCP,

and A. Kenneth Litzie
CONTENTS
INTRODUCTION
MAJOR MILESTONES IN ECC DEVELOPMENT
PARADIGM CONSTRAINT: WHAT IS ECC?
DISCONNECT BETWEEN KNOWLEDGE AND PRACTICE
SETTING FOR PARADIGM SHIFT
EXTRACORPOREAL CIRCUIT COMPONENTS
CLINICAL APPLICATION: CLOSED EXTRACORPOREAL CIRCUITS
SAFETY AND EFFICACY: CLOSED ECCS
SUMMARY
REFERENCES
INTRODUCTION
Most physicians believe that cardiopulmonary bypass (CPB) is the most invasive
aspect of open-heart surgery. Modification of current CPB circuits to reduce the trauma
induced by these systems is a rational approach to improving clinical results, and recent
developments have reinforced the value of these changes. Of all the characteristics of
CPB circuits, problems associated with less-than-optimal postoperative patient outcomes
fall into two general areas: (1) the nonendothelial foreign surface area (NEFSA) causing
the systemic inflammatory response syndrome (SIRS) (1,2); and (2) a constellation of
extracorporeal circuit (ECC) characteristics (size, hemodilution, emboli production, etc.),
causally related to a number of deleterious postoperative patient sequelae (3–9).
MAJOR MILESTONES IN ECC DEVELOPMENT

Membrane oxygenators and myocardial preservation are the most significant recent
achievements in ECC. These two technological innovations have afforded countless
411
cardiac surgical patients the opportunity to avoid key hematological dyscrasias that affect
the major organ systems.
Chemical treatments for the NEFSA of the ECC have also become available recently.
The goal of surface modification is to present a more uniform surface for blood contact
in an attempt to moderate the pathological inflammatory response (10–12). Although
their chemistries differ greatly, each approach appears to afford some degree of diminu-
tion of some indices of cellular- or plasma-protein-mediated stimulation of SIRS second-
ary to contact with NEFSAs. The maximal expression of the benefits of these
“biocompatible” surface treatments, as assessed by sensitive hematological activation
markers, is thought to be masked by the multifold contribution of other traditional ECC
technologies and practices. There is now good evidence that the very nature of cardiop-
ulmonary bypass technology may be the most fundamental and principal contributor to
postbypass patient morbidity.
PARADIGM CONSTRAINT: WHAT IS ECC?

The current technology and practice of cardiopulmonary bypass routinely conjures up
an image similar to the one on the right in Fig. 1. Extracorporeal circulation for support
of the cardiac surgical patient is inextricably associated with the image of a 700-kg
machine, 30 m of plastic tubing, and a host of discrete extracorporeal components.
Given the voluminous literature documenting patient morbidity attributable to this
traditional technology, it is curious to note that developments in ECC over the last 15 yr
have continued to be constrained within the physical and functional paradigms embodied
in that old technology. Commodity-type products of differing shapes, sizes, colors, with
a panoply of user-configured ports, connections, and conveniences have dominated the
new-product offerings available to clinicians. The focus of ECC development over these
years appears to have been user-oriented rather than patient-oriented, suggesting a some-
what fatalistic belief that the morbidity associated with traditional CPB has reached a
tolerable plateau. The acceptability of patient morbidity attributable to traditional CPB
justifying the absence of meaningful developments in extracorporeal equipment has, in
recent years, been challenged with the advent of off-pump coronary artery bypass
(OPCAB) procedures, as described throughout this book.
DISCONNECT BETWEEN KNOWLEDGE AND PRACTICE

Compounding the complicated landscape of moderating perfusion outcomes are time-
honored patient management practices that have been shown to contribute to patient
morbidity. The tolerance of deleterious patient outcomes directly attributable to ECC
components and techniques is perhaps best exemplified by the continuing use of tradi-
tional cardiotomy suction techniques. The impact on patient outcome indices (e.g., SIRS,
coagulation disorders, neurobehavioral deficits) that results from the practice of directly
returning shed blood from the surgical wound has been clearly documented in numerous
scientific publications (5,7–9,12–17). The use of cardiotomy suction, and more impor-
tantly cardiotomy-based reservoirs, exposes patients to two of the most damaging aspects
of ECC design, a massive blood–air interface and abundant silica/silicon antifoam
agents (18–21).
Other examples of tolerance for characteristics of traditional ECC that are causally
related to increasing patient morbidity and cost of care are hemodilution and homologous
Fig. 1. New and traditional cardiopulmonary bypass circuit.
blood use. In light of the consensus of scientific opinion regarding the physiological
desirability of obtaining and maintaining higher perioperative hemoglobin and hemat-
ocrit levels in cardiac surgical patients (3,5,22–26) and a similar consensus of opinion
regarding the morbidity and cost impact of allogenic blood use, 20–40% reductions in
baseline hematocrit levels are still routine during cardiac surgery.
Several technologies have emerged over the past several decades that attempt to
moderate these hemodilutional effects. Autologous blood processing techniques and
ultrafiltration technologies have emerged based on the clinical need to moderate the
excessive hemodilutional volume characteristic of contemporary CPB techniques. There
are over 240 studies investigating the use of ultrafiltration in routine CPB. Recent com-
mercial popularity for assisted venous drainage technologies, most notably vacuum-
assisted venous return (VAVD), have allowed modification of traditional ECC
components in an attempt to reduce the hemodilutional effect (27–34). Although the
application of assisted drainage was originally driven by the need for smaller cannulae
in light of port-access and minimally invasive surgical techniques, a modest 20–30%
reduction in hemodilutional volume can be realized in the most aggressive applications.
The unfortunate by-product of this technology has been the increased physical size of
ECC components designed for assisted-return applications. Increased volume capacity
has subsequently increased the NEFSA of the ECC.
Preoperative blood conservation strategies abound, each in its own way attempting to
compensate for the significant hemodilutional impact of traditional ECCs on oxygen-
carrying capacity (anemia), bleeding diathesis (thrombocytopenia), and systemic edema
(hypoproteinemia). The uses of aprotinin, preoperative autologous blood donation, eryth-
ropoietin therapy, preoperative acute normovolemic hemodilution, and platelet-rich
plasmapheresis have both advocates and detractors in the medical literature (35–40). The
absence of compelling and ubiquitous support for many such preoperative strategies
account for their irregular presence in local, regional, or national standards of care.
SETTING FOR PARADIGM SHIFT

The inherent shortcomings of existing ECC technology, and the associated patient
morbidity, has spurred the development of new technologies and techniques. One attempt
to avoid the pathophysiological responses to traditional CPB technology was to eliminate
it completely by reviving and popularizing an old technique, operating on a “beating
heart,” albeit with new and enabling technologies (retractors, immobilizers, apical
positioners) (41–44). Though seemingly an ideal solution to CPB morbidity, persistent
questions about the therapeutic equivalence and technical challenges of beating-heart
revascularization as compared to surgery on the arrested heart have limited the growth
of this technique.
From the recognition that mechanical circulatory and respiratory support of the car-
diac surgical patient is an adjunct to the surgeon’s ability to perform a complete and
effective surgical repair, some different embodiments of minimally invasive cardiopul-
monary bypass systems have been developed. These systems offer the same level of
circulatory and respiratory support as traditional CPB, but without subjecting patients to
the numerous morbidity-producing characteristics of traditional CPB. For example, the
CORx System technology (CardioVention, Inc., Santa Clara, CA) was developed spe-
cifically to address the four most significant and best-documented features of traditional
CPB technology that are causally related to various forms of postoperative patient mor-
bidity. Those four areas are listed in Table 1.
With the creation of these improved, minimally invasive extracorporeal circuits, a
“hybrid” procedure has evolved that enables surgeons and anesthesiologists to provide
to their patients many of the benefits of OPCAB procedures (e.g., no aortic crossclamp,
no global myocardial ischemia) without the hemodynamic instability, volume loading
and hemodilution, extensive vasopressor support, and expensive disposable instrumen-
tation typical of most OPCAB procedures. This hybrid procedure is described as a “pump-
assisted beating-heart,” and is rapidly growing in popularity. Early clinical results support
expectations of improved postoperative patient outcomes. Future and current studies in
the areas of SIRS, hemoglobin levels, plasma protein levels, and coagulation will help
illuminate the mechanisms of these observed improvements.
EXTRACORPOREAL CIRCUIT COMPONENTS

The principal components of “minimally invasive” ECCs vary somewhat among
manufacturers. Some manufacturers have simply interconnected their existing ECC
components with shorter pieces of tubing and renamed the assemblage a minimal or low-
prime ECC. One exception to this approach is the CORx System. This device was newly
designed specifically to address the objectives of a minimally invasive ECC. This new
concept in ECC integrates three primary, and traditionally separate and discrete compo-
nents, of a typical CPB circuit into a single, ultra-low NEFSA and hemodilutional com-
ponent. The three components combined in the integrated oxygenation system (IOS™)
are the venous reservoir (venous air-handling chamber), the blood pump (centrifugal),
and the blood oxygenator (Fig. 2).
ECC Components—Venous Air Removal

A traditional venous reservoir serves two primary purposes: to remove air that inad-
vertently enters the venous drainage line and to store blood in the ECC. Unfortunately,
the perception that a traditional venous reservoir completely removes all air in the venous
Table 1
Traditional CPB Characteristics Causally Related
to Post-operative Patient Morbidity
Large non-endothelial foreign surface area
Large extracorporeal and hemodilutional volume
Large obligatory direct blood-gas interface
Large eluting silicon-oil / silica particle antifoam agent
Fig. 2. Components of the CORx IOS.
blood is incorrect (45). In fact, nearly all instances of air down the venous line are
associated with embolic high-intensity signals in the patient’s cerebral circulation.
Equally unfortunate is the fact that the process of removing some of the air in venous
blood in traditional CPB circuits actually adds a measurably significant dose of lipoidal
and particulate microemboli by a largely unacknowledged, but well-documented, char-
acteristic of the “antifoam” agent that is present in the majority of venous reservoirs
(18,19,21). Most of the minimally invasive ECC circuits available in the United States
and Europe rely on the passive air-removal characteristics of an oxygenator, a traditional
venous reservoir, and/or an arterial line filter. One device on the market in both the United
States and in Europe uses a newly developed, active air-removal system that can serve
as an adjunct to any inherent passive capabilities of an oxygenator or arterial-line filter.
This new active air-removal technology has been integrated into CardioVention’s
CORx™ System. It substitutes the passive and chemical antifoam technologies described
above with an active air-removal system (AirVac™) that utilizes electronics to sense air
in venous blood and trigger its removal before coming into contact with the integrated
centrifugal pump (Fig. 2). This active air-removal mechanism is a significant step in ECC
technology for at least three reasons: (1) it does not rely on the decades-old technology
of chemical defoaming, with its inherent cerebral embolic characteristics; (2) it does not
rely on the passive air-removal characteristics of membrane oxygenators and arterial line
filters; and (3) in contrast to many other “closed” circuits, active venous air removal
evacuates venous air before it contacts the centrifugal blood pump, where large air
bubbles can be atomized into countless gaseous microemboli that cannot be removed by
traditional defoaming technologies.
ECC Components—Venous Reservoir

Many of the improvements in patient outcome measurements promised by the new
concept of minimally invasive CPB are more difficult to demonstrate when traditional
venous reservoirs are used. The process of shortening tubing lengths and repositioning
the traditional, discrete components of the standard perfusion circuit has resulted in some
patient benefits related to a moderate reduction in hemodilution (23). Owing to the
quadruple shortcomings of massive NEFSA, massive direct blood–gas interfacing,
obligatory extracorporeal blood volume (to avoid massive air embolism risk), and ubiq-
uitous use of potentially embolic antifoam agents, the use of these traditional venous
reservoir technologies has been rejected by at least two manufacturers of minimally
invasive ECCs.
Three of these four characteristics of traditional venous reservoirs are attributable
exclusively to their design and function. The remaining characteristic, the “obligatory
extracorporeal blood volume,” is an insidious operational characteristic of traditional
venous reservoirs that often remains unrecognized as a significant contributing factor to
the lack of evolution in the paradigm of traditional cardiopulmonary bypass. The obliga-
tory extracorporeal blood volume of traditional venous reservoirs (i.e., that minimum
level of blood in the reservoir necessary to avoid sending a massive air embolus to the
patient) varies intraoperatively and is directly proportional to the perfusionist’s margin
of safety (i.e., seconds of reaction time available to intervene before the reservoir is
emptied). This obligatory minimum level of blood is necessary even in the instance of
systemic hypovolemia or poor venous return.
In the newest generation of minimally invasive systems, alternatives to traditional
venous reservoirs include approaches such as the “elective venous blood reservoir” (Fig. 3).
The use of closed blood bags as elective venous blood reservoirs is characterized by
the fact that they:
1. Are elective, not obligatory
2. Have no mandatory minimum extracorporeal blood volume requirement for safe use
3. Contain no chemical antifoam agents that can embolize
4. Have no direct blood-gas interface
ECC Components—Centrifugal Pump

Centrifugal pumps are the pumps of choice for the minimally invasive ECC move-
ment. They can be tethered to a location closer to the patient than is possible for a roller
pump, they are both pre- and after-load sensitive, and they are incapable of pumping a
massive air embolism to the patient under catastrophic circumstances. A unique and
functionally significant step in the revolution that describes the departure from traditional
cardiopulmonary bypass is the integration of the centrifugal pump into a single,
multifunctional component. The only integrated centrifugal pump at this time is the
CORx IOS device (Fig. 4). The centrifugal pump in this device, and other closed-loop
Fig. 3. Elective (nonobligatory) venous reservoir.
Fig. 4. Integrated centrifugal pump.
ECC systems, serves two purposes. First, it is the means by which a pressure gradient is
formed between the venous cannula and the ECC to achieve right heart decompression
(i.e., control venous drainage). The same pump also serves as the “arterial” pump, and
propels venous blood through the oxygenator where carbon dioxide and oxygen
transfer occur.
Use of a centrifugal pump in this manner is generically termed augmented venous

drainage or kinetic-assisted venous drainage (KAVD). Some form of augmented venous
drainage is widely practiced in many parts of the world because of its ability to accom-
plish venous drainage in the absence of traditionally large venous cannulae, gravity-
generated hydraulic head pressure to power “siphon” venous drainage, and
large-internal-diameter venous drain lines.
CLINICAL APPLICATION: CLOSED EXTRACORPOREAL CIRCUITS

Closed circuits (i.e., not open to the atmospheric air) have long been recognized as the
most physiologically beneficial and technically safe extracorporeal circuits with which
to perform circulatory and/or respiratory support (46). Avoidance of the continuous
blood-gas interface typical of open circuits is believed to reduce numerous indices of
SIRS, as well as help preserve the coagulation mechanism for better postoperative hemo-
stasis (11). The recent development of active venous air-removal technologies (e.g.,
CORx AirVac) has made the application of a closed circuit, with all its attendant benefits,
a practical option for the routine circulatory and respiratory support needs of cardiac
surgical patients (Fig. 5).
Depending on the manufacturer, the blood flow path through closed ECC systems is
just an abbreviated form of the blood flow path through a traditional CPB circuit with one
or more of the notable qualitative and quantitative differences defined in Table 1. With
few exceptions, blood flow through closed ECC systems follows the following sequence:
venous blood is aspirated from the right heart through the venous cannula, flows down
the venous line (via KAVD) and, ideally, into a venous air-removal device. From there,
venous blood enters a centrifugal pump and is pumped through a membrane oxygenator,
arterial line, and arterial cannula into the patient’s arterial circulation.
SAFETY AND EFFICACY: CLOSED ECCS

The deployment of any new technology raises appropriate questions about patient
safety and efficacy. Challenging the decades-old paradigms of traditional CPB raises
these questions to the forefront. In this new generation of minimally invasive, closed-
circuit ECC systems, several aspects of performance are nearly self-evident (e.g., oxygen
transfer, carbon dioxide transfer, blood flow rate capability, etc.) Less evident are the
design and performance characteristics that argue to improved patient outcome measure-
ments (e.g., hemodilutional volume, NEFSA, venous air-removal efficiency, etc.). Issues
germane to the adequacy with which an extracorporeal circuit satisfies a patient’s meta-
bolic requirement for the delivery of oxygen in a circuit that does not unnecessarily
traumatize the blood include: (1) the ability of the ECC to load molecular oxygen onto
hemoglobin; (2) the concentration of hemoglobin in the patient’s blood; and (3) the blood
flow rate necessary to deliver adequate volumes of oxygenated blood to the patient’s
tissues. New mini-systems (e.g., CORx) that do not create the same magnitude of clini-
cally significant levels of hemodilution (anemia) as traditional CPB systems do (Fig. 6)
are able to transfer at least equivalent volumes of oxygen as those traditional systems, and
under many circumstances, even more. Of added interest is that, secondary to the more
normal levels of hemoglobin in the “minisystem,” the nonphysiologically high pO2 levels
of traditional CPB are not needed to ensure adequate oxygen transfer. Attributable to the
routinely higher concentrations of hemoglobin afforded by minisystems, the resulting
proportionately higher oxygen-carrying capacity of the patient’s blood requires a notably
Fig. 5. CORx “closed” A/V loop circuit.
Fig. 6. Comparison of oxygen transfer and corresponding pO2 of CORx and Medtronic affinity
oxygenators.
lower extracorporeal blood flow rate (Table 2), and thus a greatly reduced number of
times the patient’s blood volume is required to be circulated through the ECC. A reduced
number of passes through the ECC results not only in a reduction of the NEFSA-to-blood
contact ratio, but does so without sacrificing total patient perfusion as reflected by venous
oxyhemoglobin saturation.
Unique among all of these considerations is the most recently recognized requirement
for these very small, closed, minimally invasive ECC systems—venous air removal.
Some of these circuits rely on historical and somewhat ambiguous assumptions about the
reliability of physical characteristics of static circuit components (e.g., oxygenator,
centrifugal pump, etc.) to remove air. Others have developed new technology to
address this need.
Table 2
Early Clinical Comparison of CORx Oxygen Delivery
Capabilities Compared to a Traditional ECC
Conventional ECC CORx
Body surface area/m2 2.1 ± 0.2 2.0 ± 0.1
blood flow/L · min–1 5.31 ± 0.31 3.95 ± 0.55
Hematocrit (%) 26.4 ± 3.0 34.6 ± 4.4
Ven. Saturation (%) 75.9 ± 0.4 78.3 ± 4.5
FiO2 (%) 49.9 ± 9.5 57.3 ± 5.8
Gas flow/L · min–1 3.3 ± 0.4 29. ± 0.7
Art. Temperature (°C) 34.2 ± 2.6 35.1 ± 0.6
Five perfusions each group. Calafiore blood cardioplegia data points
sorted (Flow>2LPM, paO2 195...205 mmHg, SvO2 60...85%).
Fig. 7. Air-removal efficiency of CORx system.
Fundamental to new technological developments in this area is the recognition that

under certain circumstances, all extracorporeal circuits (large, small, open, closed) have
the ability to transform macro-air into micro-air (<100 µm), and, furthermore, have the
ability to pass this micro-air past all currently known levels of safety to the patient’s
arterial system. Traditional ECC design has relied on large extracorporeal, volume-
dependent, discrete devices to provide a sometimes unjustified level of confidence to the
clinician (45,47,48). Venous reservoirs with large-surface-area filter screens treated with
silica-based antifoam agents in conjunction with membrane oxygenators and equally
large-volume arterial line filters comprise the primary line of traditional defense. Of the
minimally invasive ECC systems, some rely on an oxygenator or arterial filter, to the
exclusion of any venous air-removal technology, to protect the patient from gaseous
microemboli. Another ECC (CORx System) augments these more traditional means by
actively removing air from the venous blood before it comes into contact with the cen-
trifugal pump. Because all mechanical systems have design parameter limitations, opera-
tion of these systems within the manufacturer-defined limits is essential. Under those
conditions, the newest venous air-removal technology (CORx AirVac) provides a com-
parable, if not superior, method of ECC air elimination. Figure 7 illustrates the air-
Fig. 8. CORx air entrainment removal model illustrated with two venous cannula, a 29 × 29 fr and
a 29 × 37 fr two-stage cannula at increasing levels of venous outlet resistance.
Fig. 9. Effect of perfusionist intervention in the incidence of air detection in extreme entrainment
model CORx GME output.
removal efficiency of the CORx system compared to two common ECC configurations.
Even at clinically significant air infusion rates of 200 mL/min, the CORx AirVac tech-
nology demonstrates favorable air-handling capabilities.
Although air injection is a commonly used benchmark in device comparison, a more
realistic clinical model is that of air entrainment through a venous aspiration model. This
technique more closely simulates air entrained through a leaking venous cannula purse
string or compromised right atrial integrity. Figure 8 illustrates the ability of this same
device to handle air entrained in this manner.
As the only active venous air-removal technology on the market thus far, a worst-case
experiment challenge creates a baseline performance record to which other future tech-
nologies will be compared. Figure 9 illustrates the ability of the CORx System to remove
air in a simulated, micro-air environment model generated by four 30-gage needles in the
near-venous cannula position under conditions of progressive venous inlet resistance,
and the impact of appropriate physician and perfusionist response to these conditions.
SUMMARY
The adoption of any new technology often involves disruption of old paradigms. The
most immediate outcome benefit of new-generation perfusion systems is the reduced
hemodilutional volume and subsequent effects on allogenic blood use. The use of allo-
genic transfusion adversely affects postperfusion outcomes. Not only is the incidence of
transfusion implicated, the quantity of allogenic transfusion was shown to have a signifi-
cant relationship to the length of stay and rate of postoperative complications (49). In
addition, the Northern New England Cardiovascular Disease Study Group (NNECDSG)
demonstrated that there is a relationship between decreasing hematocrit values and
increased morbidity and mortality. This relationship holds whether or not correction is
achieved through transfusion, which has its own additional level of associated risk and cost.
Quantifiable benefits are also realized through the immediate reduction in direct costs
associated with the procurement and administration of blood products. Table 3 summa-
rizes some of the potential benefits of reduced hemodilution seen by the application of
these technologies.
With a reduced response to the overall impact of the ECC on the homeostatic mecha-
nisms, more attention can be applied to perfecting surgical techniques associated with
both stopped-heart surgeries as well as beating-heart surgeries. A new frontier in mini-
mally invasive cardiopulmonary bypass can now be realized in the search for improved
patient outcomes in numerous applications requiring cardiopulmonary support.
REFERENCES
1. Aouifi A, Piriou V, Blanc P, et al. Effect of cardiopulmonary bypass on serum procalcitonin and
C-reactive protein concentrations. Br J Anaesth 1999;83(4):602–607.
2. Sablotzki A, Friedrich I, Muhling J, et al. The systemic inflammatory response syndrome following
cardiac surgery: different expression of proinflammatory cytokines and procalcitonin in patients with
and without multiorgan dysfunctions. Perfusion 2002;17(2):103–109.
3. Cormack JE, Forest RJ, Groom RC, Morton J. Size makes a difference: use of a low-prime cardiopul-
monary bypass circuit and autologous priming in small adults. Perfusion 2000;15(2):129–135.
4. Aukerman J, Voepel-Lewis T, Riegger LQ, Siewert M, Shayevitz JR, Mosca R. The relationship between
extracorporeal circuit prime, albumin, and postoperative weight gain in children. J Cardiothorac Vasc
Anesth 1998;12(4):408–414.
5. Balachandran S, Cross MH, Karthikeyan S, Mulpur A, Hansbro SD, Hobson P. Retrograde autologous
priming of the cardiopulmonary bypass circuit reduces blood transfusion after coronary artery surgery.
Ann Thorac Surg 2002;73(6):1912–1918.
6. Cromer MJ, Wolk DR. A minimal priming technique that allows for a higher circulating hemoglobin on
cardiopulmonary bypass. Perfusion 1998;13(5):311–313.
7. Brooker RF, Brown WR, Moody DM, et al. Cardiotomy suction: a major source of brain lipid emboli
during cardiopulmonary bypass. Ann Thorac Surg 1998;65(6):1651–1655.
8. Taggart DP, Westaby S. Neurological and cognitive disorders after coronary artery bypass grafting. Curr
Opin Cardiol 2001;16(5):271–276.
9. Spanier T, Tector K, Schwartz G, et al. Endotoxin in pooled pericardial blood contributes to the systemic
inflammatory response during cardiac surgery. Perfusion 2000;15(5):427–431.
10. Heyer EJ, Lee KS, Manspeizer HE, et al. Heparin-bonded cardiopulmonary bypass circuits reduce
cognitive dysfunction. J Cardiothorac Vasc Anesth 2002;16:37–42.
Chapter 29 / Making Cardiopulmonary Bypass Less Invasive
Table 3
Expected Clinical Value of CORx Technology of Reduced ECC Hemodilution
CORx feature Primary benefits Operational benefit Possible clinical benefit Possible cost benefit
Reduction of all Reduction in total
ransfusion-related cost-of-care
morbidity vis-à-vis
additional treatments
Reduced rate of Reduction in total
transfusion-related cost-of-cared
febrile response vis-à-vis additional
time in ICU
Reduced rate of Reduction in total
transfusion reaction cost of care
vis-à-vis additional
423
time in ICU
Higher hemoglobin Better intraoperative Reduced rate of Reduction in total
gas exchange and allogenic cost-of-care
transfusion vis-à-vis
additional treatments
Reduced Hemodilution Higher platelet count None Reduced post-operative Reduction in total
(Minimal to no dilution) blood loss and blood cost-of-care
product transfusion vis-à-vis additional
time in ICU
Higher total protein More stable Less post-op Reduction in total
and colloid osmotic intravascular pulmonary edema cost-of-care vis-à-vis
pressure volume and less ventilator
viscosity time and ICU time
Less post-op Reduction in total
systemic edema cost-of-care
and weight gain vis-à-vis less ICU time
423
11. Nishida H, Aomi S, Tomizawa Y, et al. Comparative study of biocompatibility between the open circuit
and closed circuit in cardiopulmonary bypass. Artif Organs 1999;23(6):547–551.
12. Aldea GS, Soltow LO, Chandler WL, et al. Limitation of thrombin generation, platelet activation, and
inflammation by elimination of cardiotomy suction in patients undergoing coronary artery bypass graft-
ing treated with heparin-bonded circuits. J Thorac Cardiovasc Surg 2002;123(4):742–755.
13. Mueller XM, Tevaearai HT, Horisberger J, Augstburger M, Boone Y, von Segesser LK. Smart suction
device for less blood trauma: a comparison with Cell Saver. Eur J Cardiothorac Surg 2001;19(4):507–511.
14. Anderson RE, Hansson LO, Liska J, Settergren G, Vaage J. The effect of cardiotomy suction on the brain
injury marker S100beta after cardiopulmonary bypass. Ann Thorac Surg 2000;69(3):847–850.
15. Chung JH, Gikakis N, Rao AK, Drake TA, Colman RW, Edmunds LH Jr. Pericardial blood activates the
extrinsic coagulation pathway during clinical cardiopulmonary bypass. Circulation 1996;93(11):2014–2018.
16. Kincaid EH, Jones TJ, Stump DA, et al. Processing scavenged blood with a cell saver reduces cerebral
lipid microembolization. Ann Thorac Surg 2000;70(4):1296–1300.
17. Jones DR, Hill RC, Hollingsed MJ, et al. Use of heparin-coated cardiopulmonary bypass. Ann Thorac
Surg 1993;56(3):566–568.
18. Challa VR, Moody DM, Troost BT. Brain embolic phenomena associated with cardiopulmonary bypass.
J Neurol Sci 1993;117(1-2):224–231.
19. Suchil BL, Hernandez RN, Hurtado J, Orsornio Vargas AR. [Antifoaming agent microembolism in
patients undergoing extracorporeal circulation. Its frequency in post mortem material and its pathogenic
potential in vitro]. Rev Esp Cardiol 1992;45(9):578–583.
20. Orenstein JM, Sato N, Aaron B, Buchholz B, Bloom S. Microemboli observed in deaths following
cardiopulmonary bypass surgery: silicone antifoam agents and polyvinyl chloride tubing as sources of
emboli. Hum Pathol 1982;13(12):1082–1090.
21. Plasse HM, Spencer FC, Mittleman M, Frost JO. Unilateral sudden loss of hearing: an unusual compli-
cation of cardiac operation. J Thorac Cardiovasc Surg 1980;79(6):822–826.
22. Kmiecik SA, Stammers AH, Petterson CM, et al. The effect of volume replacement on serum protein
concentration during cardiopulmonary bypass. J Extra Corpor Technol 2001;33(4):227–232.
23. McCusker K, Vijay V, DeBois W, Helm R, Sisto D. MAST system: a new condensed cardiopulmonary
bypass circuit for adult cardiac surgery. Perfusion 2001;16(6):447–452.
24. Nakanishi K, Shichijo T, Shinkawa Y, et al. Usefulness of vacuum-assisted cardiopulmonary bypass
circuit for pediatric open-heart surgery in reducing homologous blood transfusion. Eur J Cardiothorac
Surg 2001;20(2):233–238.
25. Rousou JA, Engelman RM, Flack JE III, Deaton DW, Garb JL, Owen SG. The “primeless pump”: a novel
technique for intraoperative blood conservation. Cardiovasc Surg 1999;7(2):228–235.
26. Rosengart TK, DeBois W, O’Hara M, et al. Retrograde autologous priming for cardiopulmonary bypass:
a safe and effective means of decreasing hemodilution and transfusion requirements. J Thorac Cardiovasc
Surg 1998;115(2):426–438.
27. Willcox TW. Vacuum-assisted venous drainage: to air or not to air, that is the question. Has the bubble
burst? J Extra Corpor Technol 2002;34(1):24–28.
28. Bevilacqua S, Matteucci S, Ferrarini M, et al. Biochemical evaluation of vacuum-assisted venous
drainage: a randomized, prospective study. Perfusion 2002;17(1):57–61.
29. Mueller XM, Tevaearai HT, Horisberger J, Augstburger M, Burki M, von Segesser LK. Vacuum assisted
venous drainage does not increase trauma to blood cells. ASAIO J 2001;47(6):651–654.
30. Mathews RK, Sistino JJ. In-vitro evaluation of the hemolytic effects of augmented venous drainage.
J Extra Corpor Technol 2001;33(1):15–18.
31. Berryessa R, Wiencek R, Jacobson J, Hollingshead D, Farmer K, Cahill G. Vacuum-assisted venous
return in pediatric cardiopulmonary bypass. Perfusion 2000;15(1):63–67.
32. Ogella DA. Advances in perfusion technology—an overview. J Indian Med Assoc 1999;97(10):
436–437, 441.
33. Munster K, Andersen U, Mikkelsen J, Pettersson G. Vacuum assisted venous drainage (VAVD).
Perfusion 1999;14(6):419–423.
34. Lau CL, Posther KE, Stephenson GR, et al. Mini-circuit cardiopulmonary bypass with vacuum assisted
venous drainage: feasibility of an asanguineous prime in the neonate. Perfusion 1999;14(5):389–396.
35. Groom RC. High or low hematocrits during cardiopulmonary bypass for patients undergoing coronary
artery bypass graft surgery? An evidence-based approach to the question. Perfusion 2002;17(2):99–102.
36. Cross MH. Autotransfusion in cardiac surgery. Perfusion 2001;16(5):391–400.
37. Tempe DK, Banerjee A, Virmani S, et al. Comparison of the effects of a cell saver and low-dose aprotinin
on blood loss and homologous blood usage in patients undergoing valve surgery. J Cardiothorac Vasc
Anesth 2001;15(3):326–330.
38. Ruel MA, Rubens FD. Non-pharmacological strategies for blood conservation in cardiac surgery. Can
J Anaesth 2001;48(4 suppl):S13–S23.
39. Iguchi A, Tanaka S. Preoperative autologous blood donation and plateletpheresis in patients undergoing
elective cardiac operations—factors that influence the need for homologous blood transfusion. Jpn Circ
J 1997;61(3):236–240.
40. Christenson JT, Reuse J, Badel P, Nowicki B, Simonet F, Schmuziger M. Autologous platelet sequestra-
tion in patients undergoing coronary artery bypass grafting. Eur J Cardiothorac Surg 1996;10(12):1083–1089.
41. Do QB, Goyer C, Chavanon O, Couture P, Denault A, Cartier R. Hemodynamic changes during off-
pump CABG surgery. Eur J Cardiothorac Surg 2002;21(3):385–390.
42. Hernandez F, Cohn WE, Baribeau YR, et al. In-hospital outcomes of off-pump versus on-pump coronary
artery bypass procedures: a multicenter experience. Ann Thorac Surg 2001;72(5):1528–1533.
43. Hart JC, Puskas JD, Sabik JF III. Off-pump coronary revascularization: current state of the art. Semin
44. Kim KB, Lim C, Lee C, et al. Off-pump coronary artery bypass may decrease the patency of saphenous
vein grafts. Ann Thorac Surg 2001;72(3):S1033–S1037.
45. Jones TJ, Deal DD, Vernon JC, Blackburn N, Stump DA. How effective are cardiopulmonary bypass
circuits at removing gaseous microemboli? J Extra Corpor Technol 2002;34(1):34–39.
46. Mongero LB, Sistino JJ, Beck J, Smith CR. Current perfusion techniques for repair of giant cerebral
aneurysms using deep hypothermia and circulatory arrest. J Extra Corpor Technol 1994;26:13–17.
47. Kurusz M. Gaseous microemboli: sources, causes, and clinical considerations. Med Instrum
1985;19(2):73–76.
48. Butler BD. Biophysical aspects of gas bubbles in blood. Med Instrum 1985;19(2):59–62.
49. Shapira OM, Aldea GS, Treanor PR, et al. Reduction of allogeneic blood transfusions after open heart
operations by lowering cardiopulmonary bypass prime volume. Ann Thorac Surg 1998;65(3):724–730.
Chapter 30 / The Zeus System 427
VI ROBOTIC SURGERY
30 Robotics and Telemanipulation

The Zeus™ System
Hersh S. Maniar, MD, Sunil M. Prasad, MD,

and Ralph J. Damiano, Jr., MD
CONTENTS
BACKGROUND
ADVANTAGES OF ROBOTIC SYSTEMS
CLINICAL RESULTS
ROBOTIC TRAINING
FUTURE DIRECTIONS
REFERENCES
BACKGROUND
Endoscopic technology has been employed in various surgical specialities to decrease
the morbidity of operations and hasten patient recovery (1–3). These procedures are
performed through small ports, with visualization achieved by using an endoscopic cam-
era. Unfortunately, there has been little success with endoscopic techniques in the field
of cardiac surgery. Despite numerous attempts over the past decade, manual endoscopic
cardiac surgery has proven to be impossible, owing in great part to the limitations of
conventional endoscopic instrumentation.
Traditional endoscopic instruments magnify a surgeon’s tremor. Given their longer
length, even the slightest operator tremor becomes amplified. Involuntary tremor, a
hindrance when performing routine procedures with conventional endoscopy, becomes
prohibitive with the microsurgery required during cardiac procedures. The ports required
for endoscopic instruments also limit the surgeon’s range of motion by producing fixed
pivot points at their insertion sites in the chest wall. The resulting “fulcrum effect”
requires a surgeon to move the handle of an endoscopic instrument in the reverse direction
that he or she intends the instrument tip to travel. These counterintuitive movements

429
430 Part VI / Robotic Surgery
make it difficult to judge accurately the deflection of the instrument tip for prescribed
movements of the instrument handles.
Operative visibility in minimally invasive surgery is also challenging. Endoscopes
provide only a two-dimensional image, instead of the traditional three-dimensional vis-
ibility surgeons have in open procedures. Adapting to the loss of depth perception can be
a difficult transition. In addition, the viewing angle from an endoscope is often mis-
aligned with respect to the instrumentation. This creates a visuomotor incompatibility
and increases the level of procedural difficulty (4,5). Given the combination of impaired
visibility and reduced dexterity, it is not surprising that few endoscopic applications have
been incorporated into microsurgical disciplines.
Robotic surgical systems represent an enabling technology for the development of
endoscopic, minimally invasive cardiac surgery. These systems provide several mecha-
nisms to overcome the inherent difficulties of conventional endoscopy for the perfor-
mance of microsurgery. The two robotic systems currently in use are the Zeus™ system,
developed by Computer Motion (Goleta, CA) and the DaVinci™ system developed by
Intuitive Surgical (Mountain View, CA). The latter is described in the following chapter.
Each system is commercially available and has been used clinically in both general and
cardiac surgery. This chapter focuses on the Zeus robotic system and its applications in
the field of cardiac surgery.
Computer Motion initially introduced a voice-controlled robotic arm (Aesop™) to
position and hold an endoscopic camera in 1994. This robotic arm is mounted on the
operating table and responds to over 20 simple voice commands (Fig. 1). Aesop elimi-
nates the need for a dedicated camera holder, and the surgeon-controlled camera avoids
potential miscommunications between surgeon and assistant. The arm remains motion-
less unless directed, providing a more stable visual field than a hand-held camera. This
feature is particularly important for microsurgery. Kavoussi et al. have shown that a
robotic camera arm more effectively manipulated and controlled the video endoscope
than a human assistant during laparoscopic procedures (6). Overall efficiency, including
repositioning and cleaning of the endoscopic camera, was improved three- to fivefold
when a robotic arm was used. The robotic system has been found to be more stable and
precise than a manually guided camera assistance (7). The Aesop robotic arm has estab-
lished an excellent record of performance and has now been used to assist in over 125,000
clinical procedures, including both coronary and valvular surgery (8).
The Zeus Robotic Microsurgical System was developed and first introduced into
clinical use in 1998 (9). Designed as a telemanipulator, the surgeon’s movements are
digitized and filtered by a signal processor, before being relayed to the robotic arms for
the completion of a given movement. The Zeus system has three primary components:
(1) an interface console, where the surgeon sits and manipulates instrument handles; (2)
a signal processor, where the motions are filtered and scaled; and (3) two effector robotic
arms that are mounted on the operating room table. The arms are designed to hold
instruments similar in length to conventional endoscopic instruments, with custom-made
instrument tips that are freely interchangeable between the robotic arms (Figs. 2 and 3).
Two instrument handles designed to replicate standard microsurgical instruments are
attached to the surgeon console. The instrument handles can be interchanged depending
on the surgeon’s preference. Housed within the console, a 16-in. video monitor displays
the operative field. The surgeon remains seated, with the endoscopic image placed di-
rectly in front, at eye level, and close to the hands. This ergonomic configuration replaces
the often awkward postioning required in laparoscopic procedures. Overall surgical
Fig. 1. The Aesop robotic arm. It can be mounted to the operating table and accommodates most
conventional endoscopes.
performance has been shown to improve with this surgeon–instrument positioning (10).
A second display beneath the video monitor functions as a touch screen to provide control
of instrument type, motion scaling, and the performance characteristics of the instrument
end-effectors.
ADVANTAGES OF ROBOTIC SYSTEMS

The principal advantage of robotic assistance is the incorporation of a signal processor
between the surgeon’s hands and the instrument tips. By converting the surgeon’s move-
ments into a digital signal, his or her motions are processed to help compensate for the
Fig. 2. The Zeus surgical console consists of instrument handles, a computer control system, and
a video monitor.
Fig. 3. The Zeus robotic arms are shown here attached to the operating room table.
Fig. 4. The surgical team is positioned at the operating table and the robotic arms are positioned
to minimize intrusions into the surgical working space.
shortcomings of conventional endoscopy. Tremor, a small-amplitude, high-frequency

signal, can be filtered by the digital processor and effectively eliminated. To further
improve surgical precision, the system provides for motion scaling. This allows gross
movements at the console to be translated into fine movements by the robotic instruments
at the operative site. For example, a 5-mm movement at the console, scaled at a ratio of
5 to 1, yields a 1-mm movement at the instrument tip. The Zeus system permits variable
degrees of motion scaling (from 1:1 to 10:1), depending on surgeon preference, and can
be increased in concert with the magnification. This “scaled telepresence” allows a
surgeon to operate comfortably on extremely small structures. Work in our laboratory has
shown that motion scaling plays a potentially greater role than tremor filtration in enhanc-
ing surgical precision with robotics (11).
The degree of instrument rotation is also enhanced by the computer system. Zeus can
increase instrument pronation and supination as needed, eliminating the rotational con-
straints of conventional endoscopy. Finally, computer-assisted instrumentation can
remove the counterintuitive movements (fulcrum effect) required during manipulation of
traditional endoscopic instruments. The digital processor automatically translates the
surgeon’s movements such that the robotic arms and instrument tips directly follow the
movements of the instrument handles at the surgeon console.
The final components of the Zeus system are the robotic arms, which are mounted to
the operating table. The three arms (two instruments and one camera) are lightweight
(20 kg) and independent, allowing for flexibility in arm and port positions. The surgical
assistant and the remainder of the surgical team are positioned in close proximity to the
robotic arms (Fig. 4), while the surgeon is seated away from the table at the Zeus console
Fig. 5. The operating surgeon is seated at the Zeus surgical interface console.
(Fig. 5). If needed, the arms can be repositioned to accommodate the workspace require-
ments of the operative team.
The endoscopic instruments used in the Zeus system are custom-designed by Scanlan
International. More than 20 different end effectors are offered, including needle drivers,
ring forceps, tissue graspers, and microscissors (Fig. 6). The instruments are between
3 and 5 mm in diameter and are easily inserted through 5-mm ports. These instruments
are smaller than conventional endoscopic instruments, are reusable, and can be conven-
tionally sterilized. They can be interchanged freely within seconds during the operative
case. The time required to set up the Zeus system has routinely been less than 20 min (12).
CLINICAL RESULTS
The Zeus system has now been used in over 400 cases at 48 centers worldwide. This
section describes the clinical results that have been reported with this system.
Coronary Artery Surgery

The Zeus system was first used clinically by Dr. Reichenspurner in Munich in Septem-
ber 1998 (9). In his initial report, two patients with isolated left anterior descending
(LAD) coronary artery disease underwent endoscopic harvesting of the left internal
thoracic artery (LITA), followed by a small parasternal incision. The incision allowed for
manual assistance and facilitation of the coronary anastomoses. The time required for
LITA harvesting was 83 ± 110 min, and anastomotic time for both patients was 42 and
40 min. Both patients had uneventful postoperative recoveries, and follow-up revealed
no evidence of recurrent angina.
A total of 25 patients with single- and multivessel disease have since undergone
robotically assisted coronary artery bypass grafting (CABG) by Reichenspurner and his
Fig. 6. The Zeus microsurgical instruments: curved needle driver (external diameter 3.9 mm), 15°
Potts scissors, ring-tipped internal thoracic artery forceps.
group (13). The initial 13 patients underwent a standard median sternotomy and imple-
mentation of cardiopulmonary bypass. Seventeen anastomoses were performed endo-
scopically in this group, including both LITA-to-LAD (n = 13), RITA-to-OM (n = 2), and
saphenous vein graft to diagonal targets (n = 2). The next six patients underwent CABG
(LITA-to-LAD) on the beating heart with a sternotomy approach. Continuing forward in
a stepwise progression, four of the remaining six patients underwent a beating-heart
approach using a limited paramedian incision (2–5 cm) for exposure.
In the entire clinical series, only two cases were converted to a manual anastomosis.
Among the off-pump coronary artery bypass patients, the average anastomotic time was
29 min (range 19–50 min) and total operative time averaged 4.9 h (range 4.0–7.5 h). LITA
preparation time averaged 55 min. One patient developed postoperative pneumonia
requiring prolonged ventilation, but all other patients recovered uneventfully from sur-
gery and were discharged after a median of 10.0 d (range 4–17 d). All patients underwent
control angiography 7–9 d after surgery. All anastomoses sutured using the Zeus system
were patent, without evidence of a stenosis greater than 50%. Twenty-three of the 25
patients have had postoperative follow-up at 3 mo, and all patients were free from angina,
with both mortality and reintervention rates of 0%.
In the United States, the first robotically assisted cardiac surgical procedure was
performed by Damiano and his group at Penn State University in December 1998 (12).
The U.S. Food and Drug Administration (FDA) approved a prospective single-center
Fig. 7. The left and right instrument ports are positioned subxiphoid and in the seventh intercostal
space in the anterior axillary line, respectively. The camera port is positioned in the fifth intercostal
space at the mid-clavicular line.
clinical trial to evaluate the efficacy and safety of robotically assisted endoscopic CABG.
Nineteen patients underwent a robotically assisted endoscopic anastomosis of the LITA
to the LAD. Primary outcome measurements were device-related complications and graft
patency at 6–8 wk.
All anastomoses were performed endoscopically through three instrument ports
(Fig. 7). A 0° endoscope was attached to the Aesop voice-controlled robotic arm. A
continuous end-to-side anastomosis was performed with a specially designed 7-cm,
double-armed, 7-0 Gore-Tex suture (WL Gore & Assoc., Flagstaff, AZ), and graft flows
were assessed using an ultrasonic flow probe and flow meter (Transonic Systems Inc.,
Ithaca, NY). As this study was approved only for a single-vessel bypass, all other grafts
were completed manually prior to the robotic anastomosis.
The robotic system required an average setup time of 16 ± 1 min. There were no
intraoperative complications related to port placement and no mechanical failures of the
robotic system. The time required to perform the endoscopic LITA-to-LAD anastomosis
was 22.5 ± 1.2 min (range 15–31 min), and the last 5 anastomoses were each performed
in less than 20 min. All anastomoses were performed successfully, and no repair stitches
were required. Eighty-nine percent (17/19) of the grafts measured were patent and had
excellent diastolic flow by ultrasound. Average graft flow was 38 ± 5 mL/min. Two of
the grafts had inadequate flow and were reconstructed manually. The average length of
the stay in the ICU was 1.1 ± 1 d, and the average hospital stay was 4.1 ± 0.4 d.
There has been 100% late follow-up with an average follow-up of 17 ± 4.2 mo (14).
At the time, there were no late complications, and all patients were in New York Heart
Association Class I. Eight weeks after surgery, graft patency was assessed by coronary
angiography. This revealed all grafts to be patent, with TIMI 1 flow and no graft with a
stenosis of greater than 50%.
In Canada, Boyd has accumulated a significant experience with endoscopic IMA
harvesting using the Zeus system and has the largest series of totally closed endoscopic
coronary bypass grafting among all Zeus clinical sites. Boyd and associates have also
proceeded in a stepwise fashion toward robotically assisted coronary surgery. Initially,
his group prospectively investigated the use of the Aesop robotic arm during ITA harvest
(15). In 55 consecutive patients, ITAs were harvested endoscopically using three 5-mm
incisions and a 30° endoscope. Anastomoses were initially completed manually through
a limited thoracotomy. The average harvest time was 57.4 ± 22.8 min. Robotic camera
assistance significantly reduced the number of endoscopic cleanings and was felt to
facilitate the more difficult dissections. The Aesop arm responded reliably to greater than
95% of verbal commands, and there was 100% patency in the 14 patients who underwent
postoperative angiography.
Subsequently (16), the Harmonic scalpel (Ethicon Endo-Surgery, Cincinnati, OH)
was adapted to a Zeus robotic arm and 19 patients underwent LITA harvest using a
robotically controlled harmonic scalpel with computer-assisted video control (Aesop). In
three patients, the Zeus system was also additionally used for RITA harvest. The inves-
tigators concluded that the Zeus system could safely be used for bilateral ITA harvesting
even when anterior–posterior working space was limited. The advantages of the robot
controlled endoscope included greater exposure, superior image quality, and a consistent
quality of assistance, improving video dexterity and lessening surgeon fatigue.
Boyd’s group has used the Zeus system for beating-heart coronary anastomoses in 12
patients undergoing single-vessel CABG through a limited thoracotomy. The anasto-
motic times for the ITA-to-LAD were 80 ± 27 min. No repair sutures were required, and
average graft flows were 38 ± 24 mL/min. Follow-up angiography has been performed
in all patients. All anastomoses were patent at an average of 1.4 ± 0.8 d postoperatively,
and 10 of 12 were of Fitzgibbons grade A.
Boyd has since performed a closed-chest, totally endoscopic, beating-heart CABG on
6 patients using the Zeus robotic system (17). The first totally endoscopic beating-heart
case was on September 1999. Using a 0° endoscope, Aesop, and warm carbon dioxide gas
insufflation, sufficient working space and visibility were established within the mediasti-
num. For three patients, a specially designed sternal elevator was required to increase
anteroposterior intrathoracic space. With the patients in a right lateral decubitus position,
trocars were inserted in the third, fifth, and seventh interspaces along the mid to anterior
axillary lines. The Zeus robotic arms were mounted on the operating table (Fig. 8). In
preparation for the beating-heart anastomosis, an articulating endo-stabilizer (Computer
Motion, Goleta, CA) was inserted through a port in the second interspace at the anterior
axillary line for LAD stabilization.
In this clinical series, anastomotic times varied between 40 and 74 min (mean 55.8 min).
All anastomoses had acceptable flows (mean 28.3 mL/min, range 12–46 mL/min), and
no patients required conversion from the robotic technique. Median operative time for the
procedure was 6 h (range 4.5–7.5 h). Patients were intubated for an average of 8.5 h (range
0–23 h) postoperatively. All but one patient was discharged from the ICU on the first
postoperative day. Postoperatively, all patients underwent coronary angiography before
Fig. 8. The patient is positioned in the 30° right lateral decubitus position. The right and left robotic
arms are mounted on the left side of the table opposite the patient’s head and at the mid-thigh level.
The Aesop arm is mounted on the right side of the operating table.
hospital discharge, and five of six grafts were found to be patent. One had a 50% stenosis
in the region of the distal snare site. Hospital average length of stay was 4.0 ± 0.9 d. All
patients were free from angina, had returned to work, and had normal exercise capacity
at a mean follow-up of 145.3 ± 29.6 d.
In summary, the U.S., Canadian, and European experiences have demonstrated the
capabilities of robotic assistance for enabling endoscopic CABG. Currently, a prospec-
tive, randomized multicenter trial is underway. The objectives of this trial are to assess
the safety and effectiveness of the Zeus Robotic Surgical system when used to perform
an anastomosis of the LITA to the LAD. Patients are to be randomized to either robotically
assisted or traditional CABG. All patients will undergo predischarge angiography to assess
LITA-to-LAD patency. This study will be the first randomized trial of robotic cardiac
surgery and represents an important step toward validating the use of computer-assisted
surgery in coronary bypass grafting.
Valve Surgery
The first steps in minimally invasive valve surgery involved the use of smaller inci-
sions than the traditional median sternotomy but remained direct-vision procedures.
Surgeons found that these incisions provided adequate exposure and reported encourag-
ing results, with low mortality and complication rates (18,19). These minimally invasive
procedures were often performed with Heartport Port-Access™ (Redwood, CA) tech-
nology. This endovascular cardiopulmonary bypass system was usually inserted via the
femoral vessels and, as a result, removed the perfusion tubing from the thoracic incision (20).
Chitwood et al. have progressively increased the role of computer assistance for both
mitral valve repair and replacement (21,22). Using a three-dimensional head-mounted
video display and an Aesop-controlled endoscope, this group has performed computer-
assisted surgery for major portions of valvular surgery in 110 patients. In their series,
operative mortality was reported at 0.9%. Although aortic crossclamp times were longer
than for conventional sternotomy patients, there were fewer reported complications. The
surgeon-controlled camera tracking was found to be more intuitive. Technically, the
video assistance was particularly advantageous for providing stable lighting and vibra-
tion-free viewing of the subvalvular apparatus. These benefits have quickly helped tran-
sition this team and others from video-assisted surgery toward video-directed mitral
procedures in which almost all of the procedure is performed under endoscopic vision
(23). Details of this experience are reported in Chapter 19.
Reichenspurner et al. reported a similar series of 50 patients who underwent mitral
valve operations with Port-Access technology and 3-D video assistance, the last 20
patients using the Aesop robotic arm (23). Using a right submammary incision (4–7 cm)
and a three-dimensional endoscopic camera, the surgeon was able to see the operative site
through both the incision and the endoscopic picture displayed inside his helmet without
moving his head. The endoscopic picture was most useful in viewing the subvalvular
apparatus and checking the position of sutures and knots. In this series there was no
mortality, and at 3 mo follow-up, 85% of patients were in NYHA Class I.
Investigators have now begun to incorporate the Zeus system into mitral valve proce-
dures. As part of a FDA-approved Phase I clinical trial, Grossi has recently reported a case
of posterior leaflet prolapse in a 50-yr-old man that was reconstructed with the use of the
Zeus telemanipulator (24). Using a right anterior 6-cm thoracotomy, the valve repair
required 3 h 2 min of cardiopulmonary bypass, and robotic instrumentation was used for
a total of 69 min. Postoperative echocardiography revealed only trace mitral regurgita-
tion, and the patient was discharged on postoperative d 3. Aesop was felt to have afforded
superior visualization, allowing for improved teaching and operating-room integration.
The investigators have also reported successful completion of a mitral valve replacement
using a 3-cm “service entrance” and Port-Access cardiopulmonary bypass in a canine
model (25). The procedure was performed using an interrupted-suture technique. After
the procedure, normal prosthetic valve function was found in all animals.
Aesop and Zeus have also been used in other cardiac procedures, including atrial septal
defect repair and patent ductus arteriosus ligation and pericardiectomy for effusive peri-
cardial disease (26–28). These procedures have principally incorporated the techniques
of minimally invasive valvular surgery, including Port-Access™ cardiopulmonary
bypass and small “service incisions.” These reports are encouraging and indicate the
potential applicability for robotics in various types of cardiac surgery. As a first-genera-
tion robotic system, Zeus has enabled several cardiac procedures. As these systems
improve, their utility in both current and future procedures should continue to expand.
ROBOTIC TRAINING
Before attempting to use a robotic system clinically, it is critical that surgeons undergo
extensive training to gain familiarity with both the endoscopic approach and to learn the
nuances of the robotic system. There is a definite learning curve with robotic systems,
even for simple skill drills. The robotic systems, however, can be easily learned by those
with and without prior endoscopic experience (29). The current recommended training
regimen includes 80–100 h of practice in inanimate models, live animals, and cadavers.
It is recommended that centers without prior robotic experience begin their clinical cases
through a full sternotomy. This provides a safety margin for the patient and allows the
surgical team to become familiar with robotically assisted endoscopic surgery. Endo-
scopic experience should be increased gradually by requiring centers first to perform a

series of robotically assisted endoscopic ITA harvests, after which the surgeon completes
the coronary anastomosis by hand. After the surgical team gains experience and becomes
comfortable with the endoscopic ITA takedown, the surgeon can proceed to robotically
assisted anastomoses. Eventually, the sternotomy and left thoracotomy can be aban-
doned, proceeding to a total endoscopic procedure.
FUTURE DIRECTIONS
Although there has been tremendous progress in the development of robotically
assisted cardiac surgery over the last several years, there still are many challenges ahead
that must be overcome in order to widen the applicability of these techniques and to
demonstrate their clinical value. At present, these operations are often lengthy, techni-
cally difficult, and applicable only to carefully selected patients. However, as with the
introduction of laparoscopy in general surgery, the accumulation of surgical experience
with this sophisticated instrumentation in a closed-chest environment will, over time,
improve operative choreography, resulting in shorter operating times and easier proce-
dures. The development of parallel technologies to facilitate the anastomoses, improve
operative exposure and working space, and assist in port placement will result in signifi-
cant advances in the field.
One of the most significant challenges that face surgeons embarking on endoscopic
procedures is the determination of optimal port placement. Both surgical experience and
likely the use of computer guidance should facilitate this in the future. Using computer-
ized tomography and magnetic resonance imaging, preliminary efforts toward the devel-
opment of a three-dimensional virtual cardiac surgical planning platform has been
initiated for use with totally endoscopic cardiac surgery (30). It also is expected that
improved microsurgical instruments will help facilitate these procedures. Computer
Motion is planning on introducing a Microwrist™. This will add an articulating joint
close to the instrument tip or end-effector (Fig. 9). This extra degree of freedom should
simplify the performance of complex maneuvers in this environment.
The real significance of robotically assisted cardiac surgery lies in the resulting inte-
gration of computers into the operating room. This will affect three principal areas:
surgeon control, intraoperative imaging, and information access. Future improvements
in the digital–manual interface will continue to enhance a surgeon’s technical ability with
these systems. Endoscopic procedures will become more feasible as computers become
more powerful and less expensive. Technological improvements in the robotic systems
should bring us closer to a more ideal microsurgical system over the next decade. This
ideal system would include fully replicated master kinematics, a full range of end-effec-
tors, effective and simple site delivery, tactile feedback, superb three-dimensional optics,
and data fusion capability to allow for computer- and image-guided surgery.
With further enhancements, it is likely that we will be able to preprogram simple
surgical maneuvers to assist in suturing and the performance of an anastomosis. These
systems may eventually “learn” surgical techniques with the use of neural networks. This
will not only allow present procedures to be performed less invasively, it will also enable
surgeons to perform a wide range of procedures that heretofore have been impossible
because of the inherent physical shortcomings of human beings.
Computers will also revolutionize intraoperative imaging. The future will see the
introduction of image-guided cardiac surgery. Surgeons will be able to manipulate images
Fig. 9. The Zeus Microwrist instruments provide “wrist” articulation and an additional degree of
freedom. (Reprinted with permission of Computer Motion; photograph by Bobbi Bennett.)
intraoperatively and view digital echocardiograms, angiograms, CT, and MRI scans
directly on the video monitor. If needed, these images will be superimposed directly on
the operative field. Fused with endoscopic pictures, surgeons may be able to define
intracardiac and extracardiac anatomy precisely without direct visualization, perhaps one
day guiding percutaneous access to intracardiac structures. Manipulation of our digital
visual interface may also make it possible to operate on a beating heart while working in
“virtual stillness.” The movement of the robotic camera and instruments can be synchro-
nized with every heartbeat, effectively canceling cardiac motion and increasing surgical
precision. A prototype already has been developed by industry and may become available
in the near future.
Finally, there will be improvements in information access. In the operating room,
networked video monitors will provide access to the hospital information system and
ancillary services (i.e., radiology, cardiac catheterization laboratory, and echocardiography),
as well as to local area networks, the Internet, and the hospital library. This technology
will allow for experts throughout the world to assist their colleagues via high-speed
video links.
As surgeons, our challenge will be not to let ourselves be defined by the size of our
incisions. We must become cardiac interventionists, performing a wide range of percu-
taneous interventions on various intrathoracic structures. Our understanding of the three-
dimensional anatomy of the chest and surgical training make us ideally suited to perform
these procedures and handle potential complications. It is likely that we will play a key
role in the future of gene therapy, angiogenesis, arrhythmia ablation, and other areas not
traditionally thought to be the domain of the surgeon. The era of computer-assisted
surgery has begun and promises to be one of the dramatic advances in our capability as
cardiac surgeons. Robotics and computer assistance has the potential to transform both
our operating rooms and our specialty as we enter the new millennium.
REFERENCES
1. Rosen M, Ponsky J. Minimally invasive surgery. Endoscopy 2001;33(4):358–366.
2. Vilos GA, Alshimmiri MM. Cost-benefit analysis of laparoscopic versus laparotomy salpingo-oophorec-
tomy for benign tubo-ovarian disease. J Am Assoc Gynecol Laparosc 1995;2(3):299–303.
3. Seifman BD, Wolf JS Jr. Technical advances in laparoscopy: hand assistance retractors, and the
pneumodissector. J Endourol 2000;14(10):921–928.
4. Hanna GB, Cuschieri A. Influence of the optical axis-to-target view angle on endoscopic task perfor-
mance. Surg Endosc 1999;13(4):371–375.
5. Cresswell AB, Macmillan AI, Hanna GB, et al. Methods for improving performance under reverse
alignment conditions during endoscopic surgery. Surg Endosc 1999;13(6):591–594.
6. Kavoussi LR, Moore RG, Adams JB, et al. Comparison of robotic versus human laparoscopic camera
control. J Urol 1995;6:2134–2136.
7. Geis WP, Kim HC, McAfee PC, et al. Synergistic benefits of combined technologies in complex,
minimally invasive surgical procedures: clinical experience and educational processes. Surg Endosc
1996;10:1025–1028.
8. Computer Motion Report. Santa Barbara, CA, June 2000.
9. Reichenspurner H, Damiano R, Mack M, et al. Use of the voice-controlled and computer-assisted
surgical system Zeus™ endoscopic coronary artery bypass grafting. J Thorac Cardiovasc Surg
1999;118:11–16.
10. Hanna GB, Shimi SM, Cuschieri A. Task performance in endoscopic surgery is influenced by location
of the image display. Ann Surg 1998;227(4):481–484.
11. Prasad SM, Maniar HS, Klingensmith ME, et al. Prospective randomized evaluation of robotic versus
laparoscopic surgical training: advantages of computer-assisted surgery. 2002 American College of
Surgeons. San Francisco, CA.
12. Damiano RJ Jr, Ehrman WJ, Ducko CT, et al. Initial United States clinical trial of robotically assisted
endoscopic coronary artery bypass grafting. J Thorac Cardiovasc Surg 2000;119(1):77–82.
13. Boehm DH, Reichenspurner H, Detter C, et al. Clinical use of a computer-enhanced surgical robotic
system for endoscopic coronary artery bypass grafting on the beating heart. Thorac Cardiovasc Surg
2000;48(4):198–202.
14. Prasad SM, Ducko CT, Stephenson ER, et al. Prospective clinical trial of robotically assisted endoscopic
coronary grafting with 1-year follow-up. Ann Surg 2001;233(6):725–732.
15. Boyd WD, Kiaii B, Novick RJ, et al. RAVECAB: improving outcome in off-pump minimal access
surgery with robotic assistance and video enhancement. Can J Surg 2001;44(1):45–50.
16. Kiai B, Boyd WD, Rayman R, et al. Robot-assisted computer enhanced closed-chest coronary surgery:
preliminary experience using a harmonic scalpel® and ZEUS™. Heart Surg Forum 2000;3(3):194–197.
17. Boyd WD, Rayman R, Desai ND, et al. Closed-chest coronary artery bypass grafting on the beating heart
with the use of a computer-enhanced surgical robotic system. J Thorac Cardiovasc Surg
2000;120(4):807–809.
18. Cosgrove DM, Sabik JF, Navia JL. Minimally invasive valve surgery. Ann Thorac Surg 1998;65:1535–1538.
19. Navia JL, Cosgrove DM. Minimally invasive mitral valve operations. Ann Thorac Surg 1996;62:1542–1544.
20. Grossi, EA, La Pietra A, Galloway AC, et al. Videoscopic mitral valve repair and replacement using the
port-access technique. Adv Card Surg 2001;13:77–88.
21. Chitwood WR Jr, Nifong LW. Minimally invasive videoscopic mitral valve surgery: the current role of
surgical robotics. J Card Surg 2000;15(1):61–75.
access mitral valve operation. Ann Thorac Surg 2000;69:1176–1182.
24. Grossi EA, LaPietra A, Applebaum RM, et al. Case report of robotic instrument-enhanced mitral valve
surgery. J Thorac Cardiovasc Surg 2000;120(6):1169–1171.
25. La Pietra A, Grossi EA, Derivaux CC, et al. Robotic-assisted instruments enhance minimally invasive
mitral valve surgery. Ann Thorac Surg 2000;70:835–838.
26. Luison F, Boyd WD. Three-dimensional video-assisted thoracoscopic pericardiectomy. Ann Thorac
Surg 2000;70(6):2137–2138.
27. Torracca L, Ismeno G, Alfieri O. Totally endoscopic atrial septal defect closure using robotic tech-
niques: report of two cases. Ital Heart J 2000;1(10);698–701.
28. Laborde F, Folliguet T, Da Cruz E, et al. Video surgical technique for interruption of patent ductus
arteriosus in children and neonates. Pediatr Pulmonol 1997;suppl 16:177–179.
29. Prasad SM, Maniar HM, Damiano RJ Jr, et al. The effects of robotic assistance on learning curves for
basic laparoscopic drills. Am J Surg 2002;183:702–707.
30. Chiu AM, Dey D, Drangova M, et al. 3-D image guidance for minimally invasive robotic coronary artery
bypass. Heart Surg Forum 2000;3(3):224–231.
Chapter 31 / Robotics and Telemanipulation: The da Vinci System 445
31 Robotics and Telemanipulation

The da Vinci™ System
Volkmar Falk, MD, PHD,

Thomas Walther, MD, PHD,
and Friedrich W. Mohr, MD, PHD
CONTENTS
HISTORICAL BACKGROUND AND BASIC PRINCIPLES
OF TELEMANIPULATION
THE DA VINCI™ SYSTEM: TECHNOLOGY
EXPERIMENTAL AND CLINICAL EXPERIENCE
DISCUSSION
REFERENCES
HISTORICAL BACKGROUND AND BASIC

PRINCIPLES OF TELEMANIPULATION
Telemanipulation systems are a class of robotics that enable the operator to work
remotely by a computerized human–machine interface. The first teleoperated devices
were built in 1890 by Nicola Tesla, an ingenious inventor who also developed the induc-
tion motor and AC power transmission. The concept of teleoperation describes a form of
control in which the human directly guides and causes each increment of motion of the
manipulator. In the 1950s telemanipulators were used mainly to handle toxic waste. In
these early systems, the master and manipulator were kinematically (geometrically)
identical. The joints of the manipulator followed that of the master through the same
trajectory, representing master and manipulator motions as joint coordinates. Since these
systems were purely mechanically linked, the scale between the master and the manipu-
lator was fixed. The first electronically augmented telemanipulator was developed by
Goertz (1). Instead of a pure mechanical linkage, motors and sensors connected the
master and the manipulator unit. With the availability of computers, online calculation

445
Table 1
Limitations of Conventional Endoscopic Instruments—Comparison
to Computer Enhanced Instrumentation Systems
Endoscopic Instruments Telemanipulator
Degress of freedom 4 4–6
Tremor filter No Yes
Motion transmission Fixed, depending on the 1:1 to 5:1
ratio of internal/external
shaft length
Hand–eye alignment Poor Natural
Fulcrum effect Reversed mothio Not an issue
Force ratio (handle/tip) Large/abnormal/not linear Programming/linear
Indexing Not possible Possible
Ergonomics Unfavorable Favorable
of cartesian coordinate transformations became possible. This marked the onset of

telemanipulation systems with different kinematic designs for the master and the
manipulator. The first six-axis manipulator for human bi-directional telemanipulation
was built by Bejczy and Salisbury and included force feedback through motors on the
hand controller (2). Since then, a variety of “computer-enhanced” telemanipulation sys-
tems have been developed for remote handling in hazardous environments or confined
spaces. Most designs follow the anthropomorphic principle, providing a humanlike range
of motion for the manipulator joints. Accordingly, the joints of a serial link manipulator
can be viewed as shoulder, elbow, and wrist joints.
The introduction of endoscopic techniques in the 1980s heralded the birth of mini-
mally invasive surgery. The use of trocars decreased the surgical trauma associated with
large conventional incisions, but at the cost of a substantial loss in dexterity, altered hand–
eye coordination, and a limited range of motion (Table 1) (3,4). Interestingly, the devel-
opment of surgical telemanipulators was not driven by the shortcomings of endoscopic
instruments but rather by the idea of remote care of trauma patients (5). An intuitive
telemanipulator system that would allow distant surgeons to treat injured patients
remotely was thought to potentially improve the outcome from severe injuries. As a
result, a prototype telepresence surgery system (SRI) with bimanual force-reflective
manipulators with four degrees of freedom (DOF), interchangeable surgical instruments,
and stereoscopic video input was developed and used remotely to close arteriotomies and
to repair lacerations of internal organs (6,7). In parallel, the German Research Institute
in Karlsruhe developed the first 6-DOF telemanipulator (Advanced Robot and
Telemanipulator System for Minimally Invasive Surgery, ARTEMIS), which was
designed primarily to improve dexterity during laparoscopic procedures and was used to
perform experimental cholecystectomies (8,9). The potential for combining the comple-
mentary abilities of humans and machines for surgical interventions has been outlined by
Taylor (Table 2) (3,4).
THE DA VINCI™ SYSTEM: TECHNOLOGY

Two telemanipulation systems are currently commercially available, the Zeus™ (Com-
puter Motion, Goleta, CA; see Chapter 30) and the da Vinci™ system (Intuitive Surgical,
Mountain View, CA). The da Vinci technology is based on the early SRI concept. The
Table 2
Complementary Skills of Human Operator and Robotic Devices (Modified from Taylor [3,4])
Strengths Limitations
Humans Strong hand–eye alignment Tremor
Excellent dexterity Fatigue
Flexible Manipulation and dexterity
limited outside natural scale
Adaptive Limited geometric accuracy
Able to use qualitative date Does not use quantitative data
naturally
Easy to Instruct Susceptible to radiation,
infection
Able to learn
Integrates extensive and diverse information
Robots Good geometric accuracy Poor judgment
Untiring and stable Difficult to instruct
Precisely calibrated forces Low bandwidth
Various design options Limited ability to perform
complex control
and hand–eye tasks
Works in hazardous environment
Can incorporate various sensors into control laws
system consists of two major components, a master console and a cart-mounted manipu-
lator. The console houses the display system, the master handles, the user interface, and
the electronic controller (Fig. 1) (10–13). The master handles are serial link manipulators
that act both as high-resolution input devices, reading the position, orientation, and grip
commands from the surgeon, and as haptic displays, transmitting forces and torques to
the surgeon in response to various measured and synthetic force cues (bi-directional). The
masters are gravity-compensated to minimize fatigue of the operator. The image of the
surgical site is transmitted to the surgeon through a high-resolution stereo display (two
separate optical channels with a maximum resolution of 2.0 mrad/line pair). The system
projects the image of the surgical site atop the surgeon’s hands (via mirrored overlay
optics), while the controller transforms the spatial motion of the tools into the camera
frame of reference. In this way the system provides natural hand–eye coordination
(Fig. 2). Motion scaling allows for various ratios for master and manipulator motions. By
activating a foot switch, the operator is able to temporarily uncouple and reposition the
masters in the working field while the instrument tips remain stationary (indexing). A
tremor filter is used to minimize involuntary motions. The camera is also controlled by
the master handles. By activating a foot switch, the masters control motion of the scope
in four axes (rotational, insertion, two translational). A variety of scopes is available (0°,
30°, small, and wide-angled view). By selecting the scope at the console, the masters are
automatically realigned according to the chosen scope angle.
The patient-side cart consists of two instrument manipulators and a central camera
manipulator (Fig. 3). The camera manipulator has 4 DoF and holds the endoscope. The
multiple-use end-effectors attach interchangeably to the two instrument manipulators,
which feature an automated instrument recognition system. The manipulator provides 3
DoF (insertion and translation in two planes), while the end-effectors add another 3 DoF
Fig. 1. Da Vinci surgical telemanipulation system: surgeon’s console.
Fig. 2. Surgeon’s view through viewer at the console (top) and orientation of master
handles (bottom) demonstrating perfect hand–eye alignment.
Fig. 3. Cart-mounted manipulator.
(rotation and two additional pitch and yaw axes at the tip). By means of this endowrist,
a total of 6 DoF are provided, allowing for free motion and orientation of the tip in space.
All three patient-side manipulators use remote-center technology to minimize shear
stress at the site of insertion. The system setup on the patient side takes approx 15 min,
and emergency removal of the manipulator from the patient can be performed in seconds.
The ideal position for the setup joints of the instrument arms is 90° between the primary
and secondary axis (“shoulder”) and 45° between the secondary and tertiary axis
(“elbow”). For the camera arm the net sum of angles should be 0°, resulting in straight
alignment of the scope and the central column. With this setup there should be no neces-
sity to move the setup joints during the procedure. The remote centers should be placed
correctly within the ports to provide the highest precision and lowest friction. The system
has been cleared by the U.S. Food and Drug Administration (FDA) for laparoscopic and
thoracoscopic (ITA take-down) procedures and has received CE mark approval in Europe.
Clinical trials to obtain FDA approval for coronary artery bypass surgery and mitral valve
repair are currently underway.
Table 3
Worldwide Experience with
the da Vinci System as of September 2001
Procedure n
ITAa take-down 1337
TECABb (arrested heart) 121
TECAB (beating heart) 80
Mitral valve repair 135
Atrial septal defect repair 42
aInternal thoracic artery.
bTotal endoscopic coronary artery bypass.
EXPERIMENTAL AND CLINICAL EXPERIENCE

As of September 2002, more than 2000 cardiac procedures including endoscopic
coronary artery bypass grafting, mitral valve repair, and atrial septal defect (ASD) closure
have been performed—in part or completely—using the da Vinci system in some 45 centers
around the world (Table 3).
Coronary Artery Bypass Surgery

Prior to clinical trials, a number of experimental studies were performed that demon-
strated the feasibility of remote suturing of coronary anastomoses (14–16). Other experi-
mental studies demonstrated the superiority of 6- vs 4-DoF telemanipulator design and
highlighted the importance of high-resolution three-dimensional vision to perform
remote endoscopic procedures using a telemanipulation system (17,18). A total endo-
scopic coronary artery bypass grafting procedure (TECAB) was first developed in a
cadaver model through three ports and without assistance. In addition, a transabdominal,
transdiaphragmatic endoscopic approach for double-vessel grafting was developed (19).
In December 1998, the da Vinci system was introduced clinically, with the main focus
on endoscopic coronary surgery (20). In the majority of cases the system was used to
harvest the internal thoracic artery endoscopically. After an initial learning curve that was
similar in most centers using this technology, harvest times for the left internal thoracic
artery (ITA) are now in the range of 30–40 min, and the technique is routinely performed
in a number of centers (21,22). For robotic-assisted ITA harvest, patients are placed in
a supine position with the left chest slightly elevated and the left arm lowered. Single
(right)-lung ventilation is performed. A 30° scope angled up is inserted at the fourth
intercostal space. Continuous CO2 insufflation is applied to enhance exposure by increas-
ing the available space between the heart and the sternum. Although insufflation pres-
sures up to 10 mmHg are usually well tolerated, hemodynamic studies have demonstrated
an increase in right ventricular filling pressures, a decrease of intrathoracic blood volume
index, and right ventricular ejection fraction with increasing insufflation pressures. As
a result, cardiac index and MAP may decrease despite a compensatory increase in heart
rate (23). The instrument ports are created in the third and sixth intercostal spaces.
Depending on the individual’s body habitus, ports are created in a flat triangle (with the
central camera port placed a little bit lower than the two instrument ports) or in an almost
linear fashion following the anterior axillary line. The ITA is usually dissected as a
pedicle from the first rib to the sixth inercostal space using low-energy cautery. Clips are
rarely used. The precision of the instruments allows for a skeletonized take-down tech-
nique. As in conventional harvest, care must be taken to avoid injury of the subclavian
vein and the phrenic nerve while dissecting the proximal part of the ITA. ITA harvesting
is now performed routinely and allows tailoring the thoracotomy incision necessary for
a MIDCAB procedure. As of this writing, 1350 cases of robotic ITA take-downs have
been reported (Table 3). If the ITA is to be used for a TECAB procedure, the vessel is
skeletonized distally, and cut and trimmed for the anastomosis in situ, using the native
tissue for countertraction (24). The pedicle is not detached from the chest wall until the
anastomosis is finally performed, to avoid torsion of the graft and any interference during
pericardiotomy. For bilateral ITA take-down, the right pleural space is opened and the
right ITA is dissected first, which is sometimes facilitated by the use of a 0° scope.
The first successful TECAB procedure on the arrested heart was reported by Loulmet
(25). Following ITA take-down, the pericardial fat is removed and a pericardial window
is created. After the LAD is identified, femorofemoral bypass is initiated using the Port-
Access™ system for closed-chest cardiopulmonary bypass and antegrade cardioplegic
cardiac arrest. The anastomosis is then performed in a running fashion on the arrested
heart through the same ports. More than 100 cases have been reported in the literature,
mostly single-vessel revascularizations of the ITA to the LAD. CPB time and crossclamp
time are in the range of 80–120 and 40–60 min, respectively. The conversion rate to a
sternotomy is now consistently less than 10%, and the reported patency rate for the
TECAB procedure ranges from 95% to 100% prior to discharge and 96% at 3-mo follow-
up angiography (24,26–28). In a few patients, the right ITA was used to graft the RCA,
and successful double-vessel TECAB to the LAD and RCA as well as sequential grafting
of the LAD and a diagonal branch have been reported (29,30). In addition, both ITAs may
be harvested endoscopically, followed by a multivessel arterial revascularization on the
arrested heart through a left parasternal minithoracotomy in the second interspace
(Dresden technique) (22,31).
The application of endoscopic coronary artery bypass grafting on the beating heart
required the development of endoscopic stabilizers and methods for temporary vascular
occlusion. Using a Nitinol-based self-expanding endoscopic stabilizer, complete endo-
scopic bypass grafting was first achieved in a canine model (32,33). More advanced
stabilizers have subsequently been developed, allowing articulation of the pads and
providing easier placement. The latest generation of endoscopic stabilizers also features
vacuum asssistance and an irrigation channel (Fig. 4).
Vascular occlusion can be achieved using vascular clamps or more commonly using
silastic bands that are either locked into the pads of the stabilizer or are used in combi-
nation with a self-locking plate. Complete TECAB procedures on the beating heart have
been reported by us and the Dresden group using the da Vinci system and an endoscopic
stabilizer that was inserted through a subxiphoidal port (Fig. 5) (34,35). After the site for
the anstomosis is identified, occlusion snares are placed around the vessel. Rather than
pushing the needle through, the motion of the heart is used to move the needle passively
through the tissue. Before insertion of the stabilizer, all suture material to be used should
be placed into the chest to avoid CO2 leaks later during the procedure. Once the stabilizer
is placed, the ITA is placed beneath the stabilizer. Alternatively, a first stitch can already
be placed in the ITA while it is still attached to the chest wall. The irrigation is placed from
behind, aiming at the site of the anastomosis. After occlusion (usually proximal and distal
occlusion will be necessary, as even a little bleeding from the anastomotic site is not well
tolerated), the anastomosis is performed in the usual fashion. About 80 closed-chest
Fig. 4. Vacuum-assisted endoscopic stabilizer with irrigating channel.
beating-heart procedures including three double-vessel beating-heart TECAB’s have

been reported in the literature (36,37). Based on an intention-to-treat analysis, the con-
version rate (elective conversion to a MIDCAB procedure) with this approach is cur-
rently in the range of 20–30% in experienced centers (26,36). LAD occlusion times are
in the range of 25–40 min and thus exceed those reported for MIDCAB procedures.
Among the difficulties reported are excessive epicardial fat, left ventricular dilatation,
determination of optimal anastomotic site, target vessel calcification, and back-bleeding
from septal branches. In addition, difficulties with positioning of the stabilizer, incom-
plete immobility of the target site, and the lack of assistance currently limit wider appli-
cation of beating-heart closed-chest bypass grafting.
Fig. 5. Setup for endoscopic beating-heart bypass grafting. The stabilizer is inserted through a
subxiphoidal port.
Mitral Valve Surgery

In 1998, a prototype of the da Vinci system was first used to perform part of a mitral
valve repair procedure (20). Thus far, access to the mitral valve is through a right ante-
rolateral minithoracotomy similar to that used for most current techniques for minimally
invasive mitral valve repair (38–40). The actual mitral valve repair is then performed
using the telemanipulation system (41–43).
Patients with a primary indication for mitral valve repair are suitable candidates for a
telemanipulator-assisted repair. Prior right-sided chest surgery, a highly calcified mitral
valve annulus, severe peripheral atherosclerosis, or concomitant cardiac disease are rela-
tive contraindications
Standard anesthetic monitoring is applied, including a central venous line inserted in
the left internal jugular vein and transesophageal echocardiography to assess mitral valve
pathology. External defibrillator pads are placed. A single-lumen endotracheal tube may
be used, since cardiopulmonary bypass is initiated before entering the chest. Use of a
pulmonary artery catheter is optional and should be used according to individual patient
requirements. A 17 fr venous cannula is inserted percutaneously in the right internal
jugular vein for drainage of the superior vena cava.
The patient is placed with the right chest elevated by 20–30°, allowing the shoulder to
drop as low as possible. The right arm is lowered beyond the level of the table, thus
providing access to the mid-axillary line.
The right femoral vessels are exposed through a 3-cm incision in the right groin. The
femoral vessels are cannulated using Seldinger technique. For venous cannulation, a
23 Fr. venous return cannula is used and placed at the level of the right atrial–inferior vena
cava junction under TEE guidance (a PFO or small ASD should always be excluded by
TEE prior to cannula insertion). Since some adjustments of the cannula may become
neccessary during the procedure, the cannula is only loosely fixed. For arterial cannula-
tion a 17 or 19 fr cannula is inserted in the femoral artery via a guide wire and after
dilatation. Both venous cannulae are connected to the CPB using a Y-connector. Venous
drainage is enhanced using vacuum assistance. Once CPB is initiated, cooling is started
and both lungs are deflated. The incision is usually made in the inframammary crease in
the mid-clavicular line extending laterally. After division of the pectoralis muscle, the
chest is entered in the fourth intercostal space. In case of excess fat tissue, a soft-tissue
retractor may be placed to provide improved exposure. A rib retractor is placed tempo-
rarily. A percutaneous stay suture is placed on the diaphragm and used to pull the
diaphragm caudally. The pericardium is opened anterior to the phrenic nerve, and addi-
tional percutaneous stay sutures are placed along the posterior pericardial edge to
approximate the heart toward the incision. The interatrial groove is dissected using
endoscopic instruments under direct vision.
A straight cardioplegia needle is inserted percutaneously in the second intercostal
space and inserted in the ascending aorta. At our institution, cold crystalloid cardioplegia
is preferred, but blood cardioplegia as well as retrograde cardioplegia (provided a coro-
nary sinus catheter has been placed via the internal jugular vein) may be used. The
transthoracic aortic clamp is placed through a stab incision in the second intercostal space
in the mid-axillary line. Placement of the clamp is ideally performed under video guid-
ance and according to the guidelines described by Chitwood (44). Alternatively, the port-
access technique may be used.
The atriotomy is performed in the usual fashion beginning at the level of the right
superior pulmonary vein and extending inferiorly just below the inferior vena cava. A
pulmonary vent is inserted through the incision. Exposure of the mitral valve is achieved
using a left atrial roof retractor inserted in the third intercostal space in close proximity
to the sternal border. Once exposure of the mitral valve is achieved, the rib retractor is
removed, since its blade may potentially interfere with the instruments. For deairing
purposes, CO2 insufflation is begun and maintained throughout the case until the left
atrium is closed.
The manipulator is placed from the patient’s left with the central camera arm aligned
with the thoracotomy incision. The scope is inserted through the incision. A port is
created in the same intercostal space below the incision and used for the right instrument.
Although this position does not provide a perfect triangle, it avoids unintentional enlarge-
ment of the atrial incision, which has been described for a more rightward port (41). A
second port is created in the third intercostal space in the anterior axillary or mid-clav-
icular line above the transthoracic clamp. Potential problems can be caused by internal
(scope and right instrument) or external (transthoracic clamp and left instrument) colli-
sions and require careful attention of the table-side surgeon. A 30° scope angled up is
preferred, because it allows easier access for the assisting surgeon. The table-side sur-
geon and the surgeon at the console should then check for potential collisions and
nonintuitive behavior at all potential positions the instruments may encounter during the case.
The actual repair is performed remotely from the surgeon’s console, applying standard
repair techniques. Resection of leaflet segments, chordae shortening, sliding plasty, and
annular decalcification can all be accomplished using the system. Suture material is
provided by the table-side surgeon, who is also in charge of retrieving excised material
or used needles using an endoscopic grasper or a magnet on a stick. Complete or partial
rings may be implanted as indicated. Annuloplasty rings with a soft cuff are preferred.
Knot tying is best performed using the sliding-square-knot technique. Motion scaling is
set to 3:1 or 1:1 according to the surgeon’s preference. After repair of the valve is
accomplished, the instruments are withdrawn, the left atrium is closed manually, and the
aortic clamp is released after de-airing through the aortic root needle. At our institution,
22 of 25 valves have been successfully repaired endoscopically using the da Vinci sys-
tem, including complex repairs requiring chordal transfer, and complete ring insertion
(41,42). In 3 patients, manual repair became necessary. Cardiopulmonary bypass and
crossclamp times exceed those for manual endoscopic techniques (38).
Few cases of total closed-chest mitral valve repair have been reported [Lange R,
personal communication]. After CPB is initiated, the instruments and camera ports are
placed as described above. The atriotomy is performed with the system. Exposure of the
valve is achieved by attaching the roof of the left atrium to the chest using patch enforced
stay sutures. Until an exposure device (internal atrial expander) becomes available, this
procedure is not recommended, however, as it is very time-consuming and requires long
CPB and crossclamp times.
ASD Repair
For total endoscopic ASD repair, placement of the patient, system setup, and the
technique for CPB are essentially the same as that described for mitral valve repair. After
insertion of the ports in the third, fourth, and fifth interspaces, the pericardium is opened
above the phrenic nerve. Both venae cavae are dissected and encircled with snare tapes,
which can be occluded either inside (self-locking) or outside the chest (external snaring).
A suction line is inserted percutaneously. The right atrium is then opened and the defect
closed with a running suture or with a patch as required (45,46). Details of this procedure
are delineated in Chapter 32.
DISCUSSION
From the experience presented here, it can be concluded that the use of the da Vinci
system is safe and allows for true endoscopic cardiac surgery. The use of the system is
currently restricted to a few indications (single-vessel bypass grafting of the LAD, occa-
sionally double-vessel grafting, and mitral valve repair, as well as some ASD closures),
but it is conceivable that it may be used for additional procedures in the near future. An
ergonomic human–machine interface and multilevel servo controlling allow for precise
tissue handling despite the lack of fine tactile feedback. The three-dimensional vision
technology provides enough optical resolution to visualize and manipulate small struc-
tures such as coronary arteries.
Importance of Training
As with all new technologies, a substantial learning curve has to be overcome and a
structured training is considered essential for procedural success. This includes a prin-
cipal understanding of telemanipulator technology and its components, the way it may
expand our capabilities but also the limitations that are associated with its use. Mastering
of the human–machine interface should be attained before any patient contact. This
includes all maneuvers performed at the surgeon’s console. Forces reflected at the master
side as well as changes in dexterity need to be interpreted correctly, as they may indicate
collisions or singularities (collapsing of 2 DoF into 1, reflecting joint limits at extreme
articulations). Because fine force reflection is limited, visual cues become more impor-
tant and need to be trained. Working with no assistance using natural tissue as a source
for countertraction is a helpful concept that is best achieved in an animal model. Equally
important is training of the patient-side surgeon. It is crucial that the table-side surgeon
understand the basic mechanisms of joint motion of the manipulators in order to provide
a setup that provides unrestricted range of motion during the case. With the use of the da
Vinci system, another complex technical system is introduced into the surgical suite.
Working in an endoscopic environment presents additional surgical challenges, includ-
ing new anatomical orientation and no direct access to cardiac structures. A stepwise
approach is therefore suggested. Initially, take-down of the ITA should be accomplished
routinely, before aiming at a TECAB procedure. Several steps occur between ITA take-
down and creation of the anastomosis. While these are trivial and automatic in an open
scenario (removal of pericardial fat, identifying the target vessel, temporary vessel
occlusion, delivering material inside the chest and others), they require sophisticated
choreography to be mastered endoscopically. A low threshold for conversion is manda-
tory, to avoid any risk to the patient. Elective conversion is safe and should by no means
be considered a failure. Importantly, algorithms for reacting to emergency situations
should be familiar to everyone in the team.
FUTURE DEVELOPMENTS
The focus for developments in endoscopic coronary artery bypass grafting is twofold:
(1) multivessel revascularization on the arrested heart using a percutaneous cardiopulmo-
nary bypass system; and (2) improvements in endoscopic beating-heart bypass grafting.
The latter will require the development of better endoscopic stabilizers as well as new
methods for coronary occlusion and blood drainage.
For multivessel revascularization, endoscopic devices for exposure of the back wall
of the heart need to be developed. Alternatively, different access routes (transabdominal,
right chest) may be explored. To help identifying coronary pathology and to define the
ideal location for an anastomosis in the absence of tactile feedback, endoscopic ultra-
sound probes may be useful (47).
With refinements in telemanipulator technology and the development of adjunct

devices to enhance exposure, the technique of computer-enhanced endoscopic cardiac
surgery will evolve further and may prove beneficial for selected patients. Smaller and
more flexible modular robotic arms will be developed, and new control algorithms will
eventually allow one operator to control multiple arms. Three-dimension HDTV systems
will provide even better optical resolution in the near future. The application of
multimodal three-dimensional imaging and computational modeling of the range of
motion of the robotic arms in an individual patient data set may optimize preoperative
planning of the procedure (48). The use of preoperative imaging may also help to better
identify suitable candidates for an endoscopic approach. Multidetector CT scanning may
help to identify intramyocardial LADs preoperatively and thus decrease the risk of con-
version or erroneous grafting to a diagonal branch (49). New devices for facilitated
anastomosis (see Chapter 28), such as the Ventrica magnetic coupling device, may facili-
tate endoscopic coronary artery bypass grafting in the future (50).
Endoscopic cardiac surgical simulation programs are currently being developed and
may help to overcome the initial learning curve for beginners. One such system is based
on the Karlsruhe Endoscopic Surgery Trainer, a virtual-reality training system for mini-
mally invasive surgery. It provides several surgical interaction modules for deformable
objects such as grasping, application of clips, cutting, coagulation, injection, and sutur-
ing. Additionally, it is possible to perform irrigation and suction in the operation area.
Special attention is paid to elastodynamically deformable tissue models of the heart and
vessels and geometric modeling techniques for graphical real-time performance. The
system is currently programmed to run with the original da Vinci console as the user
interface and thus allows a realistic simulation of beating-heart scenarios. Coupling of
two consoles may serve as an educational tool, as it enables the introduction of the
“driving school concept” into cardiac surgical training.
REFERENCES
1. Goertz RC. Fundamentals of general-purpose remote manipulators. Nucleonics 1952;10:36–45
2. Bejczy AK, Salisbury JK. Kinesthetic coupling for remote manipulators. Comput Med Eng 1983;2:48–62.
3. Taylor RH, Lavallee S, Burdea GC, Mösges R. Computer-Integrated Surgery. Technology and Clinical
Applications. Cambridge, MA: MIT Press, 1996.
4. Taylor RH, Jensen M, Whitcomb L, et al. A steady-hand robotic system for microsurgical augmentation.
Rob Res 1999;12:1201–1210.
5. Green PS, Hill JW, Jensen JF, Shah A. Telepresence surgery. Proc IEEE Eng Med Biol 1995:324–329.
6. Bowersox JC, Shah A, Jensen J, Hill J, Cordts PR, Green PS. Vascular applications of telepresence
surgery: initial feasibility studies in swine. J Vasc Surg 1996;23:281–287.
7. Bowersox JC, Cordts PR, LaPorta AJ. Use of an intuitive telemanipulator system for remote trauma
surgery: an experimental study. J Am Coll Surg 1998;186:615–621.
8. Schurr MO, Breitwieser H, Melzer A, et al. Experimental telemanipulation in endoscopic surgery. Surg
Laparosc Endosc 1996;6:167–175.
9. Schurr MO, Arezzo A, Buess GF. Robotics and systems technology for advanced endoscopic proce-
dures: experiences in general surgery. Eur J Cardiothorac Surg 1999;16(suppl II):S97–S105.
10. Guthart GS, Salisbury JK: The Intuitive™ telesurgery system: overview and application. Proc IEEE
ICRA 2002; in press.
11. Falk V. Robotic surgery. In: Yim AP, Hazelrigg SR, Izzat MB, Landrenaeau RJ, Mack MJ, Naunheim
KS, eds. Minimal Access Cardiothoracic Surgery. Philadelphia: Saunders, 1999:623–629.
12. Falk V, Diegeler A, Walther T, Autschbach R, Mohr FW. Developments in robotic cardiac surgery. Curr
Opp Cardiol 2000;15:378–387.
13. Falk V, Swarup N. Robotic Cardiac surgery —technology and procedure development. Cardiovasc Eng
2001;6:39–43.
14. Falk V, Gummert J, Walther T, Hayesi M, Berry GJ, Mohr FW. Quality of computer enhanced endo-
scopic coronary artery bypass graft anastomosis—comparison to conventional technique. Eur J
15. Shennib H, Bastawisy A, Mack MJ, Moll FH. Computer-assisted telemanipulation: an enabling technol-
ogy for endoscopic coronary artery bypass. Ann Thorac Surg 1998;66:1060–1063.
16. Shennib H, Bastawisy A, McLoughlin J, Moll F. Robotic computer-assisted telemanipulation enhances
coronary artery bypass. J Thorac Cardiovasc Surg 1999;117:310–313.
17. Falk V, Mc Loughlin J, Guthart G, et al. Dexterity enhancement in endoscopic surgery by a computer
controlled mechanical wrist. Min Inv Ther Allied Technol 1999;8:235–242.
18. Falk V, Mintz D, Grünenfelder J, Fann JI, Burdon TA. Influence of 3D vision on surgical telemanipulator
performance. Surg Endosc 2001;15:1282–1288.
19. Falk V, Moll F, Rosa D, et al. Transabdominal endoscopic computer enhanced coronary artery bypass
grafting. Ann Thorac Surg 1999;68:1555–1557.
20. Mohr FW, Falk V, Diegeler A, Autschbach R. Computer enhanced coronary artery bypass surgery.
21. Kappert U, Cichon R, Schneider J, et al. Robotic coronary artery surgery—the evolution of a new
minimally invasive approach in coronary artery surgery. Thorac Cardiovasc Surg 2000;48:193–197.
22. Cichon R, Kappert U, Schneider J, et al. Robotically enhanced “Dresden technique” with bilateral
internal mammary artery grafting. Thorac Cardiovasc Surg 2000;48:189–192.
23. Raumanns J, Diegeler A, Falk V, Ender J, Petry A. Hemodynamic effects of CO2 insufflation under one
lung ventilation for robot-guided surgery. Anesth Analg 2000;90:SCA55.
24. Falk V, Diegeler A, Walther T, et al. Total endoscopic coronary artery bypass grafting. Eur J Cardiothorac
Surg 2000;17:38–45.
25. Loulmet D, Carpentier A, d’Attellis N, et al. First endoscopic coronary artery bypass grafting using
computer assisted instruments. J Thorac Cardiovasc Surg 1999;118:4–10.
26. Mohr FW, Falk V, Diegeler A, et al. Computer-enhanced robotic cardiac surgery—experience in 148
patients. J Thorac Cardiovasc Surg 2001;121:842–853.
27. Kappert U, Schneider J, Cichon R, et al. Development of robotic enhanced endoscopic surgery for the
treatment of coronary artery disease. Circulation 2001;104(suppl I):I102–I107.
28. Dogan S, Aybek T, Andressen E, et al. Totally endoscopic coronary artery bypass grafting on cardiop-
ulmonary bypass with robotically enhanced telemanipulation: report of forty-five cases. J Thorac
29. Aybek T, Dogan S, Andressen E, et al. Robotically enhanced totally endoscopic right internal thoracic
coronary artery bypass to the right coronary artery. Heart Surg Forum 2000;3:322–324.
30. Dogan S, Aybek T, Westphal K, Mierdel S, Moritz A, Wimmer-Greinecker G. Computer enhanced
totally endoscopic sequential arterial coronary artery bypass. Ann Thorac Surg 2001;72:610–611.
31. Kappert U, Cichon R, Schneider J Schramm I, Guliemos V, Schueler S. Closed chest bilateral mammary
artery grafting in double vessel coronary artery disease. Ann Thorac Surg 2000;70:1699–1701.
32. Falk V, Diegeler A, Walther T, et al. Endoscopic coronary artery bypass grafting on the beating heart
using a computer enhanced telemanipulation system. Heart Surg Forum 1999;2:199–205.
33. Falk V, Grünenfelder J, Fann JI, Daunt D, Burdon TA. Total endoscopic computer enhanced beating
heart coronary artery bypass grafting. Ann Thorac Surg 2000;70:2029–2033.
34. Falk V, Diegeler A, Walther T, Jacobs S, Raumans J, Mohr FW. Total endoscopic off-pump coronary
artery bypass grafting. Heart Surg Forum 2000;3:29–31.
35. Kappert U, Cichon R, Schneider J, et al. Closed chest coronary artery bypass surgery on the beating heart
with the use of a robotic system. J Thorac Cardiovasc Surg 2000;120:809–811.
36. Kappert U, Cichon R, Schneider J, et al. Technique of closed chest coronary artery surgery on the beating
heart. Eur J Cardiothorac Surg 2001:20:765–769.
37. Dogan S, Aybek T, Khan MF, et al. Computer enhanced telemanipulation enables a variety of totally
endoscopic cardiac procedures. Thorac Cardiovasc Surg 2002;50:281–286.
38. Mohr FW, Onnasch JF, Falk V, et al. The evolution of minimally invasive mitral valve surgery—two
years experience. Eur J Cardiothorac Surg 1999;13:233–239.
39. Falk V, Walther T, Autschbach R, Diegeler A, Battellini R, Mohr FW. Robot assisted minimally invasive
solo mitral valve operation. J Thorac Cardiovasc Surg 1998;115:470–471.
40. Chitwood WR. State of the art review: videoscopic minimally invasive mitral valve surgery. Trekking
to a totally endoscopic operation. Heart Surg Forum 1998;1:13–16.
41. Falk V, Autschbach R, Walther T, Diegeler A, Chitwood WR, Mohr FW. Computer enhanced mitral
valve surgery—towards a total endoscopic procedure. Semin Thorac Surg 1999;11:244–249.
42. Autschbach R, Onnasch JF, Falk V, et al. The Leipzig experience with robotic mitral valve surgery.
J Cardiac Surg 2000;15:82–87.
43. Chitwood WR, Nifong LW, Elbeery JE, et al. Robotic mitral valve repair: trapezoidal resection and
prosthetic annuloplasty with the da Vinci™ surgical system. J Thorac Cardiovasc Surg 2000;120:1171–1172.
44. Chitwood WR. Video-assisted and robotic mitral valve surgery: toward an endoscopic surgery. Semin
45. Torracca L, Ismeno G, Alfieri O. Totally endoscopic atrial septal closure using robotic techniques: report
of two cases. Ital Heart J 2000;1:698–701.
46. Dogan S, Wimmer-Greinecker G, Andressen E, Mierdl S, Westphal K, Moritz A. Totally endoscopic
coronary artery bypass (TECAB) grafting and closure of an atrial septal defect using the da Vinci™
system. Thorac Cardiovasc Surg 1999;48(suppl. I):21.
47. Falk V, Fann JI, Grünenfelder J, Burdon TA: Endoscopic Doppler for detecting vessels in closed chest
bypass grafting. Heart Surg Forum 2000;3:331–333.
48. Chiu AM, Dey DD, Drangova M, Boyd WD, Peters TM. 3-D Image guidance for minimally invasive
robotic coronary artery surgery. Heart Surg Forum 2000;3:224–231.
49. Vogl TJ, Abolmaali ND, Diebold T, et al. Techniques for the detection of coronary atherosclerosis:
multidetector row CT coronary angiography. Radiology 2002;223:212–220.
50. Falk V, Walther T, Stein H, et al. Facilitated endoscopic beating heart coronary bypass grafting using
a magnetic coupling device. J Thorac Cardiovasc Surg, in press.
Chapter 32 / Totally Endoscopic ASD Repair 461
32 Totally Endoscopic Atrial Septal Defect

Repair with Robotic Assistance
Michael Argenziano, MD, Mehmet C. Oz, MD,

and Craig R. Smith, Jr., MD
CONTENTS
INTRODUCTION
METHODS
RESULTS
CONCLUSIONS
REFERENCES
INTRODUCTION
In the past decade, the face of cardiac surgery has been changed by a number of
advances, most notably the development of minimally invasive techniques, including
minimally invasive direct coronary artery bypass (MIDCAB), off-pump coronary artery
bypass (OPCAB), and minimal-access valve surgery. Initial attempts to perform cardiac
operations through small incisions were hindered by the lack of appropriate accessory
technology, such as visualization systems, retractors, stabilizers, and alternate methods
of vascular cannulation and cardiopulmonary bypass. With the development of these
enabling technologies, surgeons have been increasingly able to perform complex cardiac
procedures, including coronary artery bypass, mitral and aortic valve replacement, and
atrial septal defect (ASD) closure, through smaller-than-traditional incisions. Nonethe-
less, in many cases, the extent to which incision size has been reduced by these minimally
invasive approaches has been matched by a corresponding increase in technical difficulty
and operative time—and a potentially decreased safety margin—due to the constraints
imposed by limited or incomplete cardiac exposure.
Computer (robotic) enhancement has emerged as a potential facilitator of minimally
invasive surgical procedures. Initially, this technology was utilized to maximize visual-

461
ization of intracardiac structures by providing enhanced (including voice-activated)

endoscopic camera control (1). More recently, robotic surgical systems have permitted
the manipulation of surgical instruments through limited thoracic incisions (2). This
chapter describes the next step in this progression: the performance of atrial septal defect
repair entirely through thoracoscopic port incisions. Torracca and colleagues have
recently reported a small series of patients undergoing this operation in Europe (3). Our
series of robotic ASD repairs supplements this experience and represents the first U.S.
application of robotic technology for totally endoscopic open-heart surgery.
METHODS
Robotic Surgical System
The da Vinci™ surgical system (Intuitive Surgical, Inc., Mountainview, CA) consists
of two primary components: the surgeon’s viewing and control console and the surgical
arm unit that positions and maneuvers detachable surgical EndoWrist instruments
(Fig. 1). These pencil-sized instruments, which possess small mechanical wrists with
seven degrees of motion, are designed to provide the dexterity of the surgeon’s forearm
and wrist at the operative site through entry ports less than 1 cm in size. One port allows
access for the endoscope, and the other two ports provide access for surgical instruments.
The wrists of the surgical instruments mimic the motions made by the operating surgeon,
who sits at a console away from the operating table. The surgeon peers through an
eyepiece that provides high-definition, full-color, magnified, three-dimensional images
of the surgical site provided by the endoscope.
All patients in our series were enrolled in a U.S. Food and Drug Administration (FDA)-
sanctioned trial, entitled “Atrial Septal Defect Closure Using Intuitive Surgical Inc.’s da
Vinci™ Surgical System” and gave informed consent. This study was approved by the
Columbia University Institutional Review Board and received an FDA Investigational
Device Exemption (IDE # G010156).
Patient Selection
Inclusion criteria for the trial included age between 18 and 80 yr, and the presence of
a secundum-type atrial septal defect with Qp:Qs ratio > 1.5, or patent foramen ovale with
a documented neurological event. Patients were excluded if they could not tolerate single-
lung ventilation or peripheral cardiopulmonary bypass, or otherwise were considered
poor candidates for a thoracoscopic approach.
Surgical Technique
After induction of general anesthesia, a left-sided double-lumen endotracheal tube is
positioned to allow single-lung ventilation. A transesophageal echocardiography probe
and bilateral arterial pressure monitoring lines are inserted to facilitate monitoring of
endoaortic balloon positioning later in the procedure. A 15 or 17 fr arterial cannula
(Medtronic Bio-Medicus, Eden Prarie, MN) is placed percutaneously into the right inter-
nal jugular vein and passed into the superior vena cava with echocardiographic guidance.
This cannula is heparinized before and after insertion to avoid thrombus formation. The
patient is placed in a modified left lateral decubitus position, with the right arm either
suspended above the head (n = 7) or tacked at the side (n = 4) and the pelvis relatively flat
to facilitate femoral cannulation. After sterile preparation and draping, the right femoral
vessels are accessed through a 2-cm oblique incision along the inguinal crease. After
Fig. 1. The da Vinci surgical cart, with a central camera arm and two lateral instrument arms.
systemic heparinization, the right common femoral artery is cannulated with a 17 or 21

fr Remote Access Perfusion cannula with endoaortic balloon (ESTECH, Inc., Danville,
CA) (Fig. 2). The distal tip of the arterial cannula is passed under echocardiographic
guidance into the ascending aorta, approx 3 cm above the aortic valve. The bypass circuit
is completed by inserting a 19 or 21 fr venous cannula (Medtronic Bio-Medicus, Eden
Prarie, MN) into the right common femoral vein and passing it into the inferior vena cava,
with its tip just inferior to the IVC-RA junction.
After establishment of selective left-lung ventilation, a port incision is made in the
fourth intercostal space, in the midclavicular line, and a 12-mm endoscopic trocar
(Ethicon, Inc., Somerville, NJ) is placed into the pleural space. The endoscopic camera
is inserted and after pleural adhesions are ruled out, the pleural space is insufflated with
carbon dioxide to a maximum pressure of 8 mmHg. Two additional 8-mm port incisions
are made in the third and sixth intercostal spaces, in the anterior axillary line (Fig. 3). The
da Vinci surgical cart is positioned at the operating table, and the left and right robotic
arms are inserted into the pleural space. A fourth port incision (15 mm) is made in the fifth
intercostal space, in the posterior axillary line, for use as a service entrance.
Fig. 2. The ESTECH arterial perfusion cannula, with endoaortic balloon and distal cardioplegia
lumen.
Fig. 3. Port sites for totally endoscopic, robotically assisted ASD repair.
Fig. 4. The da Vinci surgeon’s console, from which the robotic camera and instrument arms are
controlled.
Next, the operating surgeon moves from the operating table to the surgeon’s console,
and begins the intrathoracic portion of the operation by controlling the robotic camera and
surgical instrument arms (Fig. 4). A pericardiotomy is made using the “long tip forceps”
and “cautery” attachments, and pericardial stay sutures placed, with traction provided by
passing the sutures out of the chest through the service port (posteriorly) or by sewing the
suture to the inside of the chest wall (anteriorly). Caval snares are placed using the “long
tip forceps” and passed out of the service entrance port. Cardiopulmonary bypass with
moderate hypothermia is initiated with kinetically assisted bicaval venous drainage.
Atrial stay sutures are placed. The perfusion pressure is reduced, and the endoaortic
balloon is inflated to a pressure of 250–300 mmHg. Antegrade cold blood cardioplegia
(4:1) is administered through the distal cannula port, and a satisfactory cardiac arrest
is confirmed.
Using the “round tip scissors” attachment, a right atriotomy is created, and the right
atrium is explored. Cardiotomy suction is provided by a specially modified instrument
(the “Flora sucker”), which is passed through the service entrance port by the patient-side
surgeon. In each case, anatomic landmarks, including the fossa ovalis, coronary sinus
ostium, and eustachian valve, are identified. Whether the pathology involves a large ASD
or PFO with or without a septal aneurysm, we close the entire fossa with a double-layer
primary suture technique. Before tying the atrial septopexy suture, the left atrium is
de-aired by inflating the left lung. After the suture is tied, the endoaortic balloon is
deflated, and the patient is rewarmed. The atriotomy is closed with two layers of running
4-0 polypropylene, and the patient is weaned from cardiopulmonary bypass. Integrity of
the septal closure is confirmed by transesophageal echocardiography, and protamine is
administered. After adequate hemostasis is achieved, the robotic arms are removed from
the chest, and a small flexible drainage tube is placed in the pericardium and one or two
conventional chest tubes are placed in the right pleural space, all through existing port
incisions. The femoral vessels are decannulated, and the percutaneous catheter removed
from the internal jugular vein. All incisions are closed in layers with absorbable
suture material.
Postoperative Regimen
The patients are transported to the intensive care unit, where they remain until extu-
bated. Upon extubation, patients are transferred to the general postoperative unit. Chest
drains are removed when drainage subsides. Upon discharge, patients are given an
appointment for a clinical visit and a transthoracic echocardiogram. Anticoagulants are
administered only for patients with specific postoperative indications such as atrial fibrillation.
RESULTS
Demographics
Over 7 mo, 13 patients were enrolled in the study and provided informed consent. Of
these, two were excluded intraoperatively prior to initiation of the procedure. In one case,
transesophageal echocardiography revealed a hypoplastic descending aorta, which was
considered a contraindication to passage of the endoaortic balloon cannula. This patient
underwent ASD repair via a minithoracotomy, with central vessel cannulation. In another
case, the patient had an anaphylactic reaction upon induction of anesthesia, with pro-
found vasodilation, urticaria, and bronchospasm. After stabilization, this patient was
awakened and the surgery deferred until appropriate allergic evaluation and testing were
undertaken. The other 11 patients underwent the procedure robotically. The group of
operated patients included 1 man and 10 women, with mean age 46.1 ± 14.2 (mean ± SD).
Eight patients had secundum ASD, while 3 had PFO with septal aneurysm and a history
of cerebrovascular accident not attributable to another cause.
Operative Results
All operations were completed robotically through four port incisions, with no intra-
operative conversions to alternate techniques. All patients underwent primary suture
repair of the ASD or PFO, and transesophageal echocardiography postoperatively
confirmed successful closure in all cases. Median cardiopulmonary bypass time was
124 min, and median crossclamp time was 38 min. In the last five cases of our series,
median values for cardiopulmonary bypass and crossclamp times were 121 and 28 min,
respectively. Postoperatively all patients were extubated on the night of surgery, and
median ICU length of stay was 20 h. One patient remained in the ICU for 3 d for treatment
Fig. 5. A robotic ASD repair patient, 30 d after surgery, with four well-healed port incisions.
of a postoperative pneumonia that was presumed to be related to aspiration, and recovered

without incident. Hospital length of stay ranged from 2.5 d (our first patient) to 10 d
(a 68-yr-old patient who developed postoperative atrial fibrillation and required several
days to establish effective rate control), with a median of 4 d. Cosmetic results were
excellent in all cases (Fig. 5).
Complications
There were no deaths in our series. There were no cerebrovascular accidents, wound
infections, reoperations for bleeding, or femoral vascular complications. There were two
cases of perioperative atrial fibrillation, one case of pneumonia, and one patient with
temporary arm discomfort due to intraoperative arm positioning. In one case, trans-
thoracic echocardiography revealed a recurrent interatrial shunt on postoperative d 5, in
a 58-yr-old patient who had had a large secundum ASD. When this was confirmed by
transesophageal echocardiography, the patient underwent reoperation via right
minithoracotomy. At operation, the primary suture line was found to be intact, but a tear
was noted in the septum medial to the repair. This was presumably induced by excessive
tension on the septum postoperatively, and might have been prevented by a patch tech-
nique. A pericardial patch was used for this repair, and the patient was discharged 4 d later
without incident. Since the robotically placed sutures and the line of closure were intact,
it is our conclusion that the failure of this repair was not related to the robotic procedure
per se, but rather to an error in judgment regarding the use of a primary repair technique.
In the other 10 patients in the series, 30-d postoperative echocardiography confirmed
successful repair.
CONCLUSIONS
In the past several years, technical advances in peripheral cardiopulmonary bypass
access and endoaortic balloon technology have allowed a number of intracardiac proce-
dures to be performed through smaller than usual (but not necessarily small) incisions.
The development of these procedures has required the adaptation of surgical instruments
and techniques to the challenge of operating “in a deep hole,” with less than optimal
visualization. The least invasive of these procedures have required small thoracotomy or
partial sternotomy incisions (4,5). Although these approaches employ smaller than tra-
ditional incisions, they are still associated with significant perioperative pain, due largely
to the division or retraction of intercostal muscles, ribs, and/or sternal bone. For these and
other technical reasons, these procedures have been performed predominantly at selected
centers, and have not gained widespread popularity.
The minimally invasive cardiac surgical movement has more recently been propelled
by the introduction of a new category of technological achievement: the computerized
telemicromanipulator. Utilizing this device, also known as the surgical robot, surgeons
can manipulate small instruments, which are inserted through small chest incisions, in
tight spaces, achieving many of the technical maneuvers previously possible only with
traditional open exposure. In 1997, the first intracardiac procedures—mitral valve
repairs—were performed using a prototype of the current da Vinci system (6,7). These
operations were performed through small thoracotomy incisions, since the “micro-wrists”
allowed the surgeons to complete complex maneuvers without placing their hands within
the chest. In December 2000, Chitwood and associates reported the first such mitral valve
operation performed in the United States, using an inframammary mini-thoracotomy (2).
To date, over 100 such mitral valve repairs have subsequently been performed by
Chitwood and others, under the auspices of an FDA-sanctioned multicenter trial. Details
of this technique are outlined in Chapter 19. Our center is a participant in this trial, and
we have performed 17 such repairs to date.
The success of robotically assisted operations through small thoracotomy incisions
was a critical step in the evolution of minimally invasive open-heart surgery, because it
proved that complex intracardiac procedures could be performed with robotically con-
trolled instruments. The next step—a totally endoscopic intracardiac procedure—would
wait until 2001, when closed-chest, robotically assisted atrial septal defect repairs would
be performed, as described in this chapter. In addition to the 6 cases reported by Torracca
and the 11 procedures performed by our group, an additional 5 procedures have been
performed in Frankfurt, Germany (G. Wimmer-Greinecker, personal communication).
Our series of robotically assisted ASD repairs constitutes the first U.S. report of a
totally endoscopic, robotically assisted open-heart procedure. By utilizing femoral can-
nulation and port incisions in the right chest, the da Vinci surgical system is utilized to
perform every step of the repair. By avoiding thoracotomy incisions and rib spreading,
this procedure results in minimal pain and postoperative recovery time. Despite the
impressive results in our small series, and the potential advantages of this approach with
regard to cosmesis and patient acceptance, it is currently unclear whether these benefits
can be expected in all patients. For this reason, additional patients are being enrolled in
our FDA-sanctioned clinical trial, in order to determine whether this technology will be
of reproducible value in the management of patients with intracardiac disease on a
larger scale. Furthermore, it is clear that the continued evolution of totally endoscopic
cardiac surgery depends on the development of new accessory technology, such as retrac-
tion systems, perfusion catheters, noninvasive monitoring techniques, and sutureless
anastomotic devices.
Thus, although the surgical robot allows unprecedented closed-chest access to the
heart, it is only one of several tools that must be used in concert to achieve the technical
goals of a particular operation. In this respect, the immediate future of closed-chest
cardiac surgery depends on the ability of physicians and industry to work together to fill
in the many technological gaps that are present in our current armamentarium of mini-
mally invasive tools. With such advances, and continued success from pioneering centers,
completely endoscopic cardiac surgery—literally an inconceivable notion only a few
years ago—may become a widely available clinical reality in the near future.
REFERENCES
1. LaPietra A, Grossi EA, Derivaux CC, et al. Robotic-assisted instruments enhance minimally invasive
mitral valve surgery. Ann Thorac Surg 2000;70(3):835–838.
2. Chitwood WR Jr, Nifong LW, Elbeery JE, et al. Robotic mitral valve repair: trapezoidal resection and
prosthetic annuloplasty with the da vinci surgical system. J Thorac Cardiovasc Surg 2000;120(6):1171–1172.
3. Torracca L, Ismeno G, Alfieri O. Totally endoscopic computer-enhanced atrial septal defect closure in
six patients. Ann Thor Surg 2001;72:1354–1357.
access mitral valve operation. Ann Thorac Surg 2000;69(4):1176–1181.
5. Grossi EA, LaPietra A, Ribakove GH, et al. Minimally invasive versus sternotomy approaches for mitral
reconstruction: comparison of intermediate-term results. J Thorac Cardiovasc Surg 2001;121(4):708–713.
6. Carpentier A, Loulmet D, Aupecle B, et al. Computer assisted open heart surgery. First case operated
on with success. C R Acad Sci III 1998;321(5):437–442.
7. Falk V, Walther T, Autschbach R, Diegeler A, Battellini R, Mohr FW. Robot-assisted minimally inva-
sive solo mitral valve operation. J Thorac Cardiovasc Surg 1998;115(2):470–471.
Index 471
INDEX
A preoperative preparation, 294

Aesop 3000 system, Aprotinin, attenuation of material-dependent
congenital heart surgery, 320 activation in cardiopulmonary
micro-mitral operation, 265 bypass, 11, 12
mounting, 430 ASD, see Atrial septal defect
popularity, 430 Atrial fibrillation (AF),
voice commands, 430 atrial transection, 344
AF, see Atrial fibrillation epidemiology, 343
Anastomosis, left atrial isolation, 344
adhesives, Maze procedure,
BioGlue, 406 ablation techniques,
gelatin–resorcin–formaldehyde glue, chemical ablation, 352
406 cryoablation, 346
Guidant OPCAB System, 111 factors affecting, 345, 346
laser welding, 406, 407 hyperthermic ablation, 346
mechanical systems, laser ablation, 350, 351
advantages, 394 lesion characteristics, 345, 355
arcuate-legged clips, 396 microwave ablation, 348, 350
Automatic Anastomotic Device, 400, 401 radiofrequency ablation, 346–348
GraftConnector, 402, 403 ultrasound ablation, 352
Heartflo device, 401, 402 lesion sets, 353–355
Magnetic Vascular Positioner, 403, 404 prospects, 356
Nitinol U-Clip device, 404, 406 pulmonary vein isolation, 353
One Shot System, 401 success rate, 344, 355
prospects, 408 medical therapy, 343
requirements, 394 stroke, 343
staplers, 394–396 treatment goals, 344
Symmetry Bypass System, 398 Atrial septal defect (ASD),
vascular connectors, 396, 397 da Vinci system assistance,
minimally invasive direct coronary artery complications, 467–469
bypass, 133 outcomes, 455, 466, 467
robotics, 408 patient selection, 462, 466
Aortic valve surgery, minimally invasive, postoperative regimen, 466
economic analysis, 361, 362 surgical technique, 462, 463, 465, 466
historical perspective, 293, 294 hybrid closure of septal defects, 321
incisions, Automatic Anastomotic Device, features,
mini-sternotomy, 297–299 400, 401
right anterolateral thoracotomy, 296, 297 Axius Stabilizer, see Guidant OPCAB System
right parasternal incision, 294–296 B
transverse sternotomy, 297
BioGlue, anastomosis, 406
operative techniques, 300–303
Bow-tie repair, see Percutaneous mitral
outcomes, 303–306
valve repair
471
472 Index
C Carotid artery disease, prevalence in cardiac

CABG, see Coronary artery bypass grafting patients, 218
Cardiopulmonary bypass (CPB), Cognition, see Neurocognition
adhesion molecule response and effects, 8, 9 Complement,
attenuation of material-dependent activation in cardiopulmonary bypass, 4, 5
activation, neutrophil binding effects, 5
aprotinin, 11, 12 terminal complement complex recruitment,
corticosteroids, 12, 13 4, 5
surface coating of extracorporeal circuits, Congenital heart surgery, see also Patent
10, 11 ductus arteriosus; Video-assisted
blood conservation strategies, 413 thoracoscopic surgery,
complement activation, 4, 5 hybrid closure of septal defects, 321
cytokine response and effects, 6–8 incisions,
elastase release and pulmonary dysfunction, lower sternotomy, 318
6 partial sternotomy, 316, 317
endothelial injury, 29, 30, 32, 33 right anterolateral thoracotomy, 318, 319
extracorporeal circuit, thoracotomy, 315, 316
contact system activation, 4 upper sternotomy, 317, 318
milestones in development, 411, 412 operative imaging and documentation, 315
traditional circuit features, 412 robotics, 320
tolerance for morbidity, 412, 413 thoracotomy avoidance for palliative
minimally invasive circuit design, shunts, 314
centrifugal pumps, 416–418 Coronary artery bypass grafting (CABG),
closed circuits, 418–423 endothelial effects, 28, 30, 36
CORx system, 414, 418, 420, 421, 423 fusing interventional cardiology and cardiac
prospects, 422 surgery to reduce trauma, 320, 321
venous air removal, 414–416 minimally invasive surgery, see also Dresden
venous reservoir, 416 technique; Minimally invasive
improvement prospects, 16–18 direct coronary artery bypass; Off-
material-dependent activation, 4 pump coronary artery bypass
material-independent activation, grafting; Partial sternotomy
cardiotomy suction, 13 coronary artery bypass grafting; Port
hemodilution, 14, 15, 413 access coronary artery bypass graft-
heparin response, 16 ing; Snaring tourniquet; Total
hypothermia, 15, 16 endoscopic coronary artery by-
ischemia-reperfusion injury, 13, 14 pass,
neurocognition effects, access limitations, 49
assessment, 219, 220, 223 circumflex artery revascularization, 50
mechanisms, 218 definition, 53, 129, 130
stroke, 218 historical perspective, 45–47, 129
nitric oxide response, 9, 10 learning curve, 54
platelet defects, 32, 33 popularity, 53
postperfusion syndrome, 4 stabilizers, see Genzyme-OPCAB Elite
stepwise logistic regression analysis of system; Guidant OPCAB Sys-
risk factors for untoward clinical tem; Octopus
outcomes, 142, 143 percutaneous transluminal coronary
terminal complement complex recruitment, angioplasty outcome comparison,
4, 5 147, 148
whole-body inflammatory reaction, 3, 411 prevalence, 75
Index 473
prospects for minimally invasive surgery, demographics, 241

321–323 off-pump coronary artery bypass grafting,
reoperation rates, 157 mortality, 243, 247, 248, 249
Corticosteroids, attenuation of material- multivessel grafting, 248, 249
dependent activation in operative technique, 242, 243
cardiopulmonary bypass, 12, 13 outcomes compared with conventional
CORx system, features, 414, 418, 420, 421, 423 bypass grafting, 47
CPB, see Cardiopulmonary bypass popularity, 246, 247
quality of life outcomes, 243, 249
D
stroke outcomes, 248, 249
da Vinci system, study design, 242–245
atrial septal defect repair, Endoscopic vein harvesting, see Saphenous
complications, 467–469 vein harvesting
outcomes, 455, 466, 467 Endothelium,
patient selection, 462, 466 acute injury,
postoperative regimen, 466 ischemia-reperfusion injury, 34, 35
surgical technique, 462, 463, 465, 466 mechanical trauma, 34
components, 447, 449 cardiopulmonary bypass activation, 28–30,
congenital heart surgery, 320 32, 33
conventional endoscopy limitations, 446 cell–cell interactions, 33, 34
historical perspective of telemanipulation, chronic injury,
445, 446 atherosclerosis, 37
learning curve, 456 intimal hyperplasia, 36
micro-mitral operation, 264, 265, 267, 468 functions,
mitral valve surgery outcomes, 453–455 coagulation regulation, 30–32
prospects, 456, 457, 469 vasomotor function, 28, 29
strengths and limitations compared with injury response, 28, 29
conventional surgery, 447 intravascular homeostasis maintenance, 27
total endoscopic coronary artery bypass minimally invasive bypass grafting effects,
outcomes, 450, 451 28, 30, 36
Diaphragmatic pacing, video-assisted Extracorporeal circuit, see Cardiopulmonary
thoracoscopic surgery, 338–340 bypass
Direct limited-access vein harvesting, see
Saphenous vein harvesting F
Doppler ultrasound, intraoperative graft Fulcrum effect, endoscopic surgery, 429, 433
patency assessment, 182, 183
G
Dresden technique,
advantages, 62 Gelatin–resorcin–formaldehyde (GRF) glue,
disadvantages, 62, 63 anastomosis, 406
indications, 62 Genzyme-OPCAB Elite system,
adverse hemodynamic effect mechanisms,
E coronary insufficiency, 97
Economic analysis, left ventricle, 96, 97
minimally invasive direct coronary by- right ventricle, 97
pass grafting, 362, 363 exposure of inferior and lateral walls for
minimally invasive valve surgery, 361, 362 multivessel grafting, 95, 96
off-pump coronary artery bypass grafting, exposure techniques,
362–366 apical suction devices, 102
Elderly patients, deep pericardial sutures, 97, 98
cardiovascular mortality, 241 right hemi-sternal elevation, 99–102
474 Index
right pleurotomy and pericardiotomy, Interleukin-10 (IL-10), cardiopulmonary

98, 99 bypass induction and effects, 7,
Heart Manipulation Device, 102 8
immobilizer platform,
L
coronary capture, 91, 92
development, 90 Laser welding, anastomosis, 406, 407
intracoronary shunt placement, 95 LAST, see Left anterior small thoracotomy
positioning, 93, 94 Left anterior small thoracotomy (LAST), see
prototypes, 90 Minimally invasive direct coronary
GraftConnector, automatic anastomosis, 402, artery bypass
403 M
Graft harvesting, see Saphenous vein harvesting
Magnetic resonance imaging (MRI),
GRF glue, see Gelatin–resorcin–formaldehyde
glue neurocognition assessment in
Guidant OPCAB System, heart patients, 220, 224
Magnetic Vascular Positioner (MVP), automatic
Lenox Hill Hospital experience, 111, 112
technical considerations, anastomosis, 403, 404
cardiac anesthesia, 106 Maze procedure, see Atrial fibrillation
Micro-mitral operation (MMO),
coronary artery anatomy, 106
heart displacement, 107, 110 advantages, 269, 270
proximal anastomosis, 111 Aesop 3000 system, 265
annuloplasty rings and sutures, 262–264
sternotomy and pericardiotomy, 106, 107
vessel stabilization, 110, 111 camera placement, 260
cardioplegia, 260, 261
H closure, 264
Heartflo device, automatic anastomosis, 401, da Vinci robotic assistance, 264, 265, 267
402 incision, 258–260
Heartport, see Port-access mitral valve surgery insufflation, 261
Heparin, learning curve, 270
material-independent activation in outcomes, 265, 267, 268, 269, 272
cardiopulmonary bypass, 16 preoperative preparation, 258
surface coating of extracorporeal circuits, principles, 257, 258
10, 11 MIDCAB, see Minimally invasive direct
coronary artery bypass
I Minimally invasive direct coronary artery
ICAM, see Intercellular adhesion molecule, bypass (MIDCAB),
IL-1, see Interleukin-1 access, 57, 58
IL-6, see Interleukin-6 advantages, 58, 59, 141
IL-8, see Interleukin-8 contraindications, 59–61, 114, 115, 131
IL-10, see Interleukin-10 definition, 113
Integrins, cardiopulmonary bypass induction economic analysis, 362, 363
and effects, 8, 9 endoscopic coronary artery bypass procedure,
Intercellular adhesion molecule (ICAM), 58
cardiopulmonary bypass induction historical perspective, 113, 114
and effects, 8, 9 hybrid revascularization,
Interleukin-1 (IL-1), cardiopulmonary by- case scenarios, 210–212
pass induction and effects, 6, 7 indications, 208, 213, 214
Interleukin-6 (IL-6), cardiopulmonary by- operative procedure, 208, 209
pass induction and effects, 7, 8 outcomes, 212–215
Interleukin-8 (IL-8), cardiopulmonary by- overview, 207, 208
pass induction and effects, 7, 8 percutaneous intervention, 209, 210
Index 475
sequence of procedures, 214 da Vinci system assistance, 453–455

indications, 57, 58–61, 64, 114, 115, 130, definition, 255
131 direct vision, 256
limitations, 49, 59, 142 economic analysis, 361, 362
neurocognition outcomes, 223 edge-to-edge technique, 384–386
open conversion rate, 136 evolution, 255, 256
outcomes, 125, 126, 134–136 micro-incision, 257
patency rates, 59, 148 micro-or port-incision and video-directed
popularity, 57 robotic operation, 257
postoperative course, 133, 134 mini-incision, 256
quality-of-life outcomes, 374–380 video-assisted vision, 257
reoperation, Zeus system assistance, 438, 439
anterior MIDCAB, MMO, see Micro-mitral operation
left anterior descending artery MRI, see Magnetic resonance imaging
exposure, 163, 164 MVP, see Magnetic Vascular Positioner
left internal mammary artery
N
preparation, 62, 163
subclavian MIDCAB, 164–170 Neurocognition,
subxiphoid MIDCAB, 64, 170 cardiopulmonary bypass effects,
transabdominal MIDCAB, 170, 172, assessment, 219, 220, 223
173 mechanisms, 218
subxiphoid procedure, stroke, 218
advantages, 154, 155 factors affecting, 219
approach, 58, 148, 149 minimally invasive direct coronary by-
mortality, 154 pass grafting outcomes, 223
open conversion rate, 154 neuropsychological test batteries, 224, 225
reoperation, 64, 170 off-pump coronary artery bypass grafting
technique, 119, 120, 123, 148, 149 outcomes, 47, 48, 217–225
technique, presurgical baseline and change definition,
anastomosis, 133 221, 222
anesthesia, 131 risk factors for deficits after heart surgery,
endoscopic MIDCAB, 125 224
graft harvesting, 131, 142 Neuron-specific enolase (NSE), neurocognition
heparin reversal, 133 marker, 223
H Graft, 119 Nitinol U-Clip device, automatic anastomosis,
intubation, 131 404, 406
left anterior thoracotomy, 115–118 Nitric oxide (NO),
left anterior thoracotomy with axillary atherosclerosis dysregulation, 37
artery inflow, 118, 119 cardiopulmonary bypass induction and
left posterior lateral thoracotomy, 123 effects, 9, 10
stenting, 141 endothelial injury prevention, 33, 34
xiphoid approach, 119, 120, 123, 148, 149 functions, 29
Mitral regurgitation, NO, see Nitric oxide
prognostic factors, 384 NSE, see Neuron-specific enolase
timing of surgery, 384 O
Mitral valve surgery, minimally invasive,
see also Micro-mitral operation; Octogenarians, see Elderly patients
Octopus,
Percutaneous mitral valve repair;
Port-access mitral valve surgery, comparison of prototypes, 81
Carpentier–Loulmet classification of de- development, 76, 78, 79
hemodynamic changes, 80, 82–84
grees of surgical invasiveness, 256
476 Index
histological changes, 82, 85, 88 limitations, 54, 55

off-pump coronary artery bypass grafting, multivessel grafting, 95, 96, 142
clinical experience, 86–88 neurocognitive outcomes, 47, 48, 217–225
Octopus III, 70, 71 Octopus stabilizer, 70, 71, 86–88
Off-pump coronary artery bypass grafting operative procedure, 68, 69, 136, 138
(OPCAB), perfusion-assisted direct coronary artery
advantages, 50, 54, 67, 73 bypass, 72, 73
cardiac displacement and coronary target popularity, 229, 230
presentation, 70, 71 postoperative course, 138
contraindications, 50, 55, 56, 58 preoperative evaluation, 68
coronary stabilization and grafting, 71, reoperation,
72, 75, 76 anesthesia, 159
economic analysis, 362–366 indications, 158
elderly patients, intraoperative monitoring, 159
mortality, 243, 247, 248, 249 Lenox Hill Hospital outcomes, 173–177
multivessel grafting, 248, 249 pain control, 159, 160
operative technique, 242, 243 preoperative planning, 158
outcomes compared with conventional resternotomy,
bypass grafting, 47 exposure of lateral and inferior
popularity, 246, 247 wall, 160–162
quality of life outcomes, 243, 249 median sternotomy, 160
stroke outcomes, 248, 249 mediastinal dissection, 160
study design, 242–245 surgical technique, 159, 160
exposure techniques, severe left ventricular dysfunction patients,
apical suction devices, 102 operative technique, 231, 232
deep pericardial sutures, 97, 98 outcomes of multivessel revascularization,
median sternotomy, 136 232–237
right hemi-sternal elevation, 99–102 patient selection, 230–232
right pleurotomy and pericardiotomy, quality of life outcomes, 235, 237
98, 99 study design, 230, 231, 237
goals, 106 training, 238
graft patency, stabilizers, see Genzyme-OPCAB Elite
angiography, 183, 184 system; Guidant OPCAB System;
Doppler ultrasound, 182, 183 Octopus
fluorescein angiography, 184 temporary intraluminal shunt, 49
intraoperative assessment, verticalization of heart, 136, 138
postoperative assessment, One Shot System, automatic anastomosis, 401
early angiography, 185, 187 OPCAB, see Off-pump coronary artery
intermediate-term angiography, bypass grafting
187, 188
P
late angiography, 188
thermal coronary angiography, 184 Pacemaker, implantation with video-assisted
transit time flow measurement, 180–182 thoracoscopic surgery, 338
prospects for postoperative evaluation, PADCAB, see Perfusion-assisted direct
188, 189 coronary artery bypass
rates, 55, 139, 140, 142 Partial sternotomy coronary artery bypass
grafting sequence, 69, 70 grafting,
historical perspective, 46–48, 89 access, 60, 61
indications, 55, 56–58, 67, 68 advantages, 61
inflammatory response, 48, 49 contraindications, 61, 62
Index 477
disadvantages, 61 closure, 285

indications, 61, 62 endoscopic instruments, 284
Patent ductus arteriosus (PDA), video-assisted femoral artery cannulation, 277
thoracoscopic surgery, 312, 313 Heartport system overview, 273
PDA, see Patent ductus arteriosus historical perspective, 273, 274
Percutaneous mitral valve repair, incisions, 275–277
bow-tie repair, 384–386 intraoperative difficulties, 286
coil-fastening device, 388, 389 learning curve, 289
difficulty, 383–385 left atriotomy, 283
imaging, 391 mitral valve repair, 284, 289
indications, 389, 390 modifications, 285, 286
prospects, 391 operative technique, 274, 275
Percutaneous transluminal coronary angioplasty outcomes,
(PTCA), hospital outcomes, 286, 288
coronary artery bypass grafting outcome intraoperative factors, 286
comparison, 147, 148 quality of life, 288, 289
hybrid revascularization with minimally pain, 289
invasive direct coronary artery patient selection, 274
bypass, 207–215 postoperative care, 285
Perfusion-assisted direct coronary artery prospects, 290
bypass (PADCAB), off-pump rib repair, 285
coronary artery bypass grafting, robotics, 290
72, 73 Prostacyclin, cardiopulmonary bypass induction
Pericardial disease, and effects, 10
anatomy and physiology, 328, 329 PTCA, see Percutaneous transluminal coronary
etiology of pericarditis and effusion, 331, 333 angioplasty
historical perspective, 328 Pulmonary artery stenosis hybrid therapy,
medical therapy, 327, 328 321
video-assisted thoracoscopic surgery,
Q
advantages, 340
anesthesia, 329, 330 QoL, see Quality-of-life
congenital defects, 335, 336 Quality-of-life (QoL),
patient selection and preparation, 329 Centro Cardiologico Monzino IRCCS
pericardial constriction, 334 experience with minimally invasive
pericardial cysts, 336, 338 cardiac surgery,
pericardial effusion, 331–334 early results, 376
pericardial tamponade, 331–334 late results, 376
pericardial tumors, 335 measures of quality-of-life, 376–379
surgical technique, 330, 331 study design, 375, 376
Port access coronary artery bypass grafting, cosmetic issues, 374, 375
advantages, 62 definition, 369, 370
disadvantages, 62, 64 importance in overall health outcomes,
indications, 62 370, 371
Port-access mitral valve surgery, minimally invasive cardiac surgery study
advantages, 289 abundance, 370
aortic decannulation, 285 off-pump coronary artery bypass grafting
blood transfusion, 288 outcomes, 235, 237, 243, 249
cardioplegia, 281, 282 pain, 373, 374
central aortic cannulation, 279, 281 port-access mitral valve surgery outcomes,
chest tube, 285 288, 289
478 Index
prospects for study, 379–381 Total endoscopic coronary artery bypass

tools, (TECAB),
Nottingham Health Profile, 371, 373, access, 63
375, 379 advantages, 63
physical activity score, 371, 376 contraindications, 63
Psychological General Well-Being da Vinci system assistance, 450, 451
Index, 373, 376, 378 disadvantages, 63
table, 372 learning curve, 63
Transit time flow measurement (TTFM),
R
intraoperative graft patency
Radial artery harvesting, minimally invasive assessment, 180–182
harvesting, 205 TTFM, see Transit time flow measurement
Robotics, see Aesop 3000 system; da Vinci Tumor necrosis factor-α (TNF-α),
system; Zeus system cardiopulmonary bypass induction
S and effects, 6–8
TXB2, see Thromboxane B2
S-100, neurocognition marker, 223
Saphenous vein harvesting, V
angioscopy for intraluminal lesion Vascular cell adhesion molecule (VCAM),
monitoring, 204 cardiopulmonary bypass induction
complications, 193, 194 and effects, 8, 9
direct limited-access vein harvesting, VATS, see Video-assisted thoracoscopic
closure, 196 surgery
Genzyme retractor system, 195, 196 VCAM, see Vascular cell adhesion molecule
mid-calf incision, 196 Verapamil, controlled hypotension in minimally
popularity of grafting, 193 invasive surgery, 46
traditional technique, 193 Video-assisted endoscopic vein harvesting,
video-assisted endoscopic vein harvesting, see Saphenous vein harvesting
complications, 201, 202, 204 Video-assisted thoracoscopic surgery
disadvantages, 203, 204 (VATS),
dissection, 199–201 diaphragmatic pacing, 338–340
Guidant Vasoview Uniport Plus System, epicardial pacemaker implantation, 338
198 extracardiac congenital lesion repair, 314
insufflation, 199 patent ductus arteriosus repair, 312, 313
learning curve, 197, 198 pericardial disease,
vein quality, 203 advantages, 340
Selectins, cardiopulmonary bypass induction anesthesia, 329, 330
and effects, 8, 9 congenital defects, 335, 336
Snaring tourniquet, coronary artery trauma, patient selection and preparation, 329
47 pericardial constriction, 334
Stabilizers, see Genzyme-OPCAB Elite system; pericardial cysts, 336, 338
Guidant OPCAB System; Octopus pericardial effusion, 331–334
Staplers, anastomosis, 394–396 pericardial tamponade, 331–334
Symmetry Bypass System, automatic pericardial tumors, 335
anastomosis, 398 surgical technique, 330, 331
T vascular ring division, 313, 314
TECAB, see Total endoscopic coronary X
artery bypass Xpose, see Guidant OPCAB System
Thromboxane B2 (TXB2), cardiopulmonary
bypass induction and effects, 10 Z
TNF-α, see Tumor necrosis factor-α Zeus system,
Index 479
advantages of robotic systems, 431, 433 instrument rotation, 433

components, 430 learning curve, 439, 440
computer integration, 440, 441 mitral valve surgery outcomes, 438, 439
coronary artery surgery outcomes, 434–438 prospects for minimally invasive cardiac
development, 430 surgery, 440–442
endoscopic instruments, 434, 440 robotic arms, 433, 434
instrument handles, 430, 431
Contemporary Cardiology ™
Christopher P. Cannon, MD, Series Editor
Minimally Invasive Cardiac Surgery

SECOND EDITION
Edited by
Daniel J. Goldstein, MD Mehmet C. Oz, MD
Newark Beth Israel Medical Center, Newark, NJ New York Presbyterian Hospital, New York, NY
From Reviews of the First Edition:

“…succeeds in providing a timely review of an important new arena in cardiac surgery…” —TEXAS HEART INSTITUTE JOURNAL
“…a superb contribution to the literature and…perhaps the most comprehensive of the field of MICS to date…must reading for
surgeons in this field.” —DOODY’S HEALTH SCIENCES BOOK REVIEW JOURNAL
“…important reading for anyone contemplating this form of surgery, as well as for all cardiac surgeons needing to keep abreast of
developments in their specialty.” —JOURNAL OF THE ROYAL COLLEGE OF SURGEONS OF EDINBURGH
“…[offers] some very stimulating ideas about what the future of cardiac surgery may be.” —BRITISH JOURNAL OF ANAESTHESIA
In this updated and expanded edition of Goldstein and Oz’s much-praised classic survey, world experts
comprehensively review the many innovations and changes that have occurred in the practice of minimally
invasive cardiac surgery (MICS). The authors summarize the latest findings and results regarding such novel
techniques as robotic surgery, hybrid revascularization, the use of off-pump coronary artery bypass grafting
(OPCABG) in high-risk populations, and new less invasive surgical approaches to the treatment of atrial fibrilla-
tion. They also provide a medium-term update on the current status of MICS in dealing with valvular and con-
genital heart disease and offer a more comprehensive understanding of procedures like mid-CABGs with longer-
term followup and randomized data to help tease out real benefits and limitations. In addition, the book addresses
the economic impact of MICS and quality-of-life issues, and shows how surgeons are not only performing the
old operations with new tools, but are also creating new procedures precisely adapted to the new tools.
With its state-of-the-art descriptions of each of the relevant procedures, Minimally Invasive Cardiac Sur-
gery, Second Edition should be in the libraries of all cardiac surgery residency program directors, as well as in the
pockets of all their training and graduating surgery residents.
Features
䉴 Full descriptions of the latest minimally invasive 䉴 Comprehensive understanding of minimally
coronary surgical techniques invasive cardiac surgery procedures
䉴 Coverage of robotic surgery, hybrid 䉴 New less invasive surgical approaches to the
revascularization, and the use of OPCAB treatment of atrial fibrillation
䉴 Long-term follow-up and randomized data to tease
out real benefits and limitations
Contents
Part I Physiology of Coronary Bypass Grafting With and Without Car- Surgery. Port-Access Mitral Valve Surgery. Minimally Invasive Aortic Valve
diopulmonary Bypass. Pathophysiology of Cardiopulmonary Bypass. Surgery. Part IV Minimally Invasive Congenital, Pericardial, and Ar-
Endothelial Injury During Minimally Invasive Bypass Grafting. Part II rhythmia Surgery. Strategies for Reducing Trauma in Congenital Heart
Minimally Invasive Coronary Bypass Grafting. Minimally Invasive Surgery. Thoracoscopic Pericardial Surgery. Less Invasive Surgical Treat-
Bypass Grafting: A Historical Perspective. Tailoring Minimal Invasive ment of Atrial Fibrillation. Part V Miscellaneous Aspects of Minimally
Coronary Bypass to the Patient. OPCAB: A Primer on Technique. Mechani- Invasive Cardiac Surgery. The Economic Impact of Minimally Invasive
cal Stabilization: The Medtronic Octopus System. Mechanical Stabiliza- Cardiac Surgery. Minimally Invasive Cardiac Surgery: Quality-of-Life
tion Systems: The Genzyme-OPCAB Elite System. Mechanical Stabiliza- Issues. Experimental Percutaneous Mitral Valve Repair. Alternative
tion Systems: The Guidant OPCAB System. The MIDCAB Operation. Mini- Anastomotic Techniques. Making Cardiopulmonary Bypass Less Invasive.
mally Invasive Coronary Artery Bypass Grafting on the Beating Heart: Part VI Robotic Surgery. Robotics and Telemanipulation: The Zeus™
The European Experience. Minimally Invasive Coronary Artery Bypass System. Robotics and Telemanipulation: The da Vinci™ System. Totally
Grafting: The South American Experience. Reoperative Off-Pump CABG. Endoscopic Atrial Septal Defect Repair with Robotic Assistance. Index.
Perioperative Evaluation of Graft Patency in OPCAB. Minimally Invasive
Conduit Harvesting. Hybrid Revascularization. Neurocognitive Issues in
Off-Pump CABG. Multivessel Off-Pump Revascularization in High-Risk
Patients: Severe Left Ventricular Dysfunction. Multivessel Off-Pump
Revascularization in High-Risk Populations: Octogenarians. Part III Mini- 90000
mally Invasive Valvular Surgery. Minimally Invasive Mitral Valve
Contemporary Cardiology™
Minimally Invasive Cardiac Surgery, Second Edition
ISBN: 1-58829-170-7 E-ISBN: 1-59259-416-6
humanapress.com 9 781588 291707

Minimally Invasive Cardiac Surgery

Uploaded by

Copyright:

Available Formats

Minimally Invasive Cardiac Surgery

Uploaded by

Document Information

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

Minimally Invasive Cardiac Surgery

Uploaded by

Copyright:

Available Formats

What surgical techniques are discussed in the document?

What surgical techniques are discussed in the document?

What cardiovascular conditions are addressed through minimally invasive techniques?

What cardiovascular conditions are addressed through minimally invasive techniques?

Minimally Invasive

Coronary Disease in Women: Evidence- Essentials of Bedside Cardiology: With a

All rights reserved.

Cover design by Patricia F. Cleary.

This publication is printed on acid-free paper. ∞

Photocopy Authorization Policy:

Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1

Library of Congress Cataloging-in-Publication Data

Part I Physiology of Coronary Bypass Grafting With and

Part II Minimally Invasive Coronary Bypass Grafting

11 Minimally Invasive Coronary Artery Bypass Grafting:

Part III Minimally Invasive Valvular Surgery

Part IV Minimally Invasive Congenital, Pericardial,

23 Thoracoscopic Pericardial Surgery .............................................. 327

Part V Miscellaneous Aspects of Minimally Invasive

Part VI Robotic Surgery

Index ........................................................................................................................... 471

KEVIN D. ACCOLA, MD, Florida Heart Institute, Florida Hospital, Orlando, FL

BRENDA DICKEY, RN, Florida Heart Institute, Florida Hospital, Orlando, FL

P. MICHAEL MCFADDEN, MD, Surgical Director Lung Transplantation, Ochsner Clinic

I PHYSIOLOGY OF CORONARY BYPASS

Ron G. H. Speekenbrink, MD, PHD,

TNF-α Monocytes, macrophages TNF receptors on Hypotension, fever,

macrophages, T-lymphocytes hepatocytes APP synthesis/release ↑,

(41,42). Interleukin-6 (IL-6) is produced by activated monocytes, endothelial cells, fibro-

METHODS TO ATTENUATE MATERIAL-DEPENDENT ACTIVATION

reduced complement activation, the inflammatory responses of leukocytes, platelets, and

elucidated. Possibly, aprotinin inhibits the activity or production of trace amounts of

Ischemia and Reperfusion

include oxygen-radical scavengers (150), maintenance of physiological oxygen concen-

dure of cardiopulmonary bypass can only be obtained when a multifactorial approach is

Direct monitoring of the microcirculation is possible with the technique of orthogonal

2 Endothelial Injury During Minimally

Robert J. Dabal, MD, Craig R. Hampton, MD,

From: Contemporary Cardiology: Minimally Invasive Cardiac Surgery, Second Edition

OVERVIEW OF ENDOTHELIAL FUNCTION

Fig. 1. Endothelial cell responses to injury.

prostacyclin, and adenosine—all of which have been shown to be potent antiadhesive

ACUTE ENDOTHELIAL INJURY

20 min, myocardial necrosis is inevitable, with concomitant endothelial cell damage.

CHRONIC ENDOTHELIAL INJURY

difference is likely secondary to differences in endothelial function. Mammary grafts are

cells. Cardiovascular surgery, however, results in damage of the endothelium that

II MINIMALLY INVASIVE CORONARY

3 Minimally Invasive Bypass Grafting

Enio Buffolo, MD, PHD

After some isolated cases of direct myocardial revascularization performed by Goetz

From: Contemporary Cardiology: Minimally Invasive Cardiac Surgery, Second Edition

patients after the techniques of myocardial revascularization with extracorporeal circu-

cerebral microembolic signals were analyzed by transcranial Doppler in a randomized

Thirty patients were studied; 15 patients received adjunctive methylprednisolone and 15

4 Tailoring Minimal Invasive Coronary

Michael A. Borger, MD, PHD

From: Contemporary Cardiology: Minimally Invasive Cardiac Surgery, Second Edition

Another disadvantage is that beating-heart surgery is technically more demanding and

Cardiac Indications and Contraindications for OPCAB