Cardiopulmonary

Transplantation and
Mechanical Circulatory
Support
ii

Oxford Specialist Handbooks published and forthcoming

General Oxford Specialist Handbooks Oxford Specialist Handbooks in Neurology
A Resuscitation Room Guide Epilepsy
Addiction Medicine 2e Parkinson’s Disease and Other Movement
Applied Medicine and Surgery in Dentistry 3e Disorders, 2e
Cardiopulmonary Transplantation and Mechanical Stroke Medicine, 2e
Circulatory Support Oxford Specialist Handbooks in Obstetrics and
Day Case Surgery Gynaecology
Infection in the Immunocompromised Host Maternal Medicine
Parkinson’s Disease and Other Movement Urogynaecolo gy
Prison Medicine and Health Oxford Specialist Handbooks in Oncology
Disorders 2e Myeloproliferative Neoplasms
Perioperative Medicine, 2e Practical Management of Complex Cancer Pain 2e
Pharmaceutical Medicine Oxford Specialist Handbooks in Paediatrics
Postoperative Complications, 2e Neurodisability and Community Child Health 2e
Renal Transplantation Paediatric Dermatology 2e
Retrieval Medicine Paediatric Endocrinology and Diabetes 2e
Oxford Specialist Handbooks in Anaesthesia Paediatric Gastroenterology, Hepatology, and
Anaesthesia for Medical and Surgical Emergencies Nutrition 2e
Anaesthesia for Emergency Care Paediatric Haematology and Oncology
Cardiac Anaesthesia Paediatric Intensive Care
Global Anaesthesia Paediatric Nephrology, 2e3e
Neuroanaesthesia Paediatric Neurology, 2e
Obstetric Anaesthesia 2e Paediatric Palliative Medicine, 2e
Ophthalmic Anaesthesia Paediatric Radiology
Paediatric Anaesthesia 2e Paediatric Respiratory Medicine 2e
Regional Anaesthesia, Stimulation and Ultrasound Paediatric Rheumatology 2e
Techniques Oxford Specialist Handbooks in Pain Medicine
Thoracic Anaesthesia Spinal Interventions in Pain Management
Vascular anaesthesia Oxford Specialist Handbooks in Psychiatry
Oxford Specialist Handbooks in Cardiology Addiction Medicine, 2e
Adult Congenital Heart Disease 2e Child and Adolescent Psychiatry
Cardiac Catheterization and Coronary Intervention Eating Disorders
2e Forensic Psychiatry 2e
Cardiac Electrophysiology and Catheter Ablation Medical Psychotherapy
Cardiovascular Computed Tomography 2e Old Age Psychiatry
Cardiovascular Imaging Oxford Specialist Handbooks in Radiology
Cardiovascular Magnetic Resonance Head and Neck Imaging
Echocardiography, 3e Interventional Radiology
Fetal Cardiology 2e Musculoskeletal Imaging
Heart Failure, 2e Pulmonary Imaging
Hypertension Thoracic Imaging
Inherited Cardiac Disease 2e Oxford Specialist Handbooks in Surgery
Nuclear Cardiology 2e Burns
Pacemakers and ICDs 2e Cardiothoracic Surgery, 2e
Pulmonary Hypertension Colorectal Surgery 2e
Valvular Heart Disease Current Surgical Guidelines 2e
Oxford Specialist Handbooks in Critical Care Gastric and Oesophageal Surgery
Advanced Respiratory Critical Care Hand Surgery 2e
Cardiothoracic Critical Care Hepatopancreatobiliary Surgery
Oxford Specialist Handbooks in End of Life Care Neurosurgery
End of Life Care in Cardiology Operative Surgery, 2e
End of Life Care in Dementia Oral and Maxillofacial Surgery, 2e3e
End of Life Care in Nephrology Otolaryngology and Head and Neck Surgery
End of Life Care in Respiratory Disease Paediatric Surgery 2e
End of Life in the Intensive Care Unit Plastic and Reconstructive Surgery 2e
Oxford Specialist Handbooks in Infectious Disease Surgical Oncology
Infectious Disease Epidemiology Urological Surgery 2e
Manual of Childhood Infections 4e Vascular Surgery, 2e
Cardiopulmonary
Transplantation
and Mechanical
Circulatory
Support
edited by
Maziar Khorsandi, Steven Tsui,
John Dark, Alan J. Kirk,
Matthew Hartwig,
Mani A. Daneshmand, and
Carmelo Milano
illustrated by
Jay LeVasseur
iv

Great Clarendon Street, Oxford, OX2 6DP,

United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Oxford University Press 2022
The moral rights of the authors have been asserted
First Edition published in 2022
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2022932691
ISBN 978–​0–​19–​286761–​2
DOI: 10.1093/​med/​9780192867612.001.0001
Printed in the UK by
Ashford Colour Press Ltd, Gosport, Hampshire
Oxford University Press makes no representation, express or implied, that the
drug dosages in this book are correct. Readers must therefore always check
the product information and clinical procedures with the most up-​to-​date
published product information and data sheets provided by the manufacturers
and the most recent codes of conduct and safety regulations. The authors and
the publishers do not accept responsibility or legal liability for any errors in the
text or for the misuse or misapplication of material in this work. Except where
otherwise stated, drug dosages and recommendations are for the non-​pregnant
adult who is not breast-​feeding
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
To my parents Shahin and Ali Khorsandi, MD and to the memory
of Edward “Ted” Brackenbury, FRCS
vi
 vii

Foreword
When people speak of vitality, it is often with reference to cardiorespira-
tory function. Indeed, throughout human history people have intuitively
associated heart and lung function with one’s life, virtue, and soul. Thus,
when heart and lung transplantation emerged in the 1960s, it captured the
world’s attention far more than kidney or liver transplantation. After all, we
associate ourselves with every heartbeat, every breath; not so much with
urine and bile. It was as though successful replacement of the heart and
lungs forced us to consider our humanity, and thus seemed more miracu-
lous than technical.
The past 60 years have led the elements of pulse and breath to be char-
acterized so well that they are now regularly exchangeable. The world is
approaching its 100,000th cardiopulmonary transplantation, and mechan-
ical support devices are rapidly becoming commonplace. With such preva-
lence, one would think that cardiopulmonary replacement might have lost
its lustre. Yet every time I think about a heart or lung transplant, or a left
ventricular assist device deployment, it still seems like a miracle to me—​not
because the mechanics and biology are mysterious, but rather because they
are now so well defined. To be sure, it is immensely technical, and it is the
enormity of what we now know about this practice that makes it so im-
pressive. Assembling that knowledge into a usable format is increasingly
important and challenging.
In saying that transplantation is technical, it is not to imply that it is amen-
able to rote education. There is really no transplant procedure that is
exactly like another. Each case has its unique physiology, immunology, and
histocompatibility, the latter of which literally defines individuality. There
are so many reasons that this should not work, yet it does, almost routinely.
I believe this is a tribute to the dedicated people who try to make sense
of each case and derive and curate generalizable insights. Cardiopulmonary
Transplantation and Mechanical Circulatory Support is a collection of those
insights, to be used in the adept navigation of each improbable success.
This is a concise resource to help practitioners quickly refresh the relevant
information needed to recognize problems and deal with them. This is not
a comprehensive treatise, as each chapter could warrant a book in and unto
itself. Rather, this is a means of practical guidance, placing clinical challenges
into the broad contexts of cardiopulmonary physiology and transplant im-
munology. It will be particularly useful to trainees, and to nurses, advanced
practice providers, intensivists, and hospitalists caring for transplant patients
and needing quick reference. This is a resource to provide order to the rap-
idly accumulating practical information needed to routinely transform lives,
one miracle at a time.
Allan D. Kirk, MD, PhD, FACS
David C. Sabiston, Jr. Distinguished Professor of Surgery,
Immunology and Pediatrics
Chairman of the Department of Surgery
Duke University Medical Center
Durham, NC, USA
vii
 ix

Foreword
Cardiopulmonary Transplantation and Mechanical Circulatory Support
will serve as a valuable resource for residents, fellows, and advanced care
practitioners. Thirty-​eight chapters cover a variety of important topics, rep-
resenting several years of effort by an international slate of authors and
editors. While much of what we do in surgery for advanced heart and lung
failure is evidence based, a substantial amount of it is also based on per-
sonal opinion, consensus, and institutional preference. What is presented
here falls into the latter category. However, trainees require a foundational
understanding of key concepts and techniques to begin their education in
the field. This handbook is a practical guide and an excellent starting point
for individuals learning to practice in this field. By no means is this intended
as a definitive treatment on this topic but rather to lay the ground work for
lifelong learning. Managing Editor Maz Khorsandi will be updating this hand-
book to ensure content remains current.
Those who possess a highly-​developed awareness of this field and know-
ledge of related literature may take issue with some of the points made
in the text. However, authors have attempted to simplify the concepts in
order to make them more approachable to the reader. The best use of
this handbook would be for study preparation by trainees with directed
follow-​up discussion with their attending providers, as well as focused lit-
erature searches to enhance and refine their depth of understanding. As
these fields continue in their inevitable evolution, the stewardship of con-
tinued learning will require humility and receptiveness to new concepts and
ideas that challenge existing notions, concepts, or dogma. This handbook is
a starting point that is very readable, approachable, and useful. As such, it
should inspire readers to engage in this complex and rapidly-​evolving field
and serve as a catalyst for advanced learning.
Michael S. Mulligan, MD
Professor and Chief, Division of Cardiothoracic Surgery
Endowed Professor in Lung Transplant Research
Director, Lung Transplant Program
University of Washington
x
 xi

Preface
As I approached the final stretch of my specialty training in cardiothoracic
surgery, I felt that there was a paucity of a quick reference manual for my
subspecialty field of interest, cardiopulmonary transplantation and mech-
anical circulatory support. Most reference guides were large, cumbersome,
and already outdated by the time they were published. Henceforth, I felt
the need to embark on this handbook project. I was further encouraged
and supported by my mentors Dr Carmelo Milano, MD, Duke University
Medical Center, USA, and Mr Edward ‘Ted’ Brackenbury, FRCS C/​Th,
Royal Infirmary of Edinburgh, UK. Furthermore, Duke University Heart
Center generously supported the project financially.
Cardiopulmonary Transplantation and Mechanical Circulatory Support is a
comprehensive yet succinct and readily available pocket guide of the most
up-​to-​date developments in the field of heart and lung transplantation and
mechanical circulatory support. The most prominent and globally respected
experts in Europe and North America were selected as editors and authors
for this handbook and the content, for the most part, reflects their personal
experience, hints, and tips for practitioners on the front line. The handbook
is accompanied by informative text and high-​resolution practical illustrations
making it relevant to all tiers of healthcare professionals such as nurses,
advanced care practitioners, surgical, medical, and critical care residents
and fellows, as well as junior attending/​consultant physicians and surgeons
looking after such critically ill patients. This handbook will be readily avail-
able both as a hard copy and online. I strive to continuously update it with
the most recent evidence-​based practice.
Maziar Khorsandi, FRCS C/​Th
Assistant Professor of Surgery
Attending Cardiothoracic Surgeon
Division of Cardiothoracic Surgery
University of Washington Medical Center
Seattle, WA, USA
xii
 xiii

Acknowledgements
We would like to gratefully acknowledge the generous donation by Duke
University Heart Center in funding the illustrations for this handbook.
xvi
 xv

Contents
Contributors xix
Abbreviations xxvii

Section 1 Heart failure

1 Medical and minimally invasive aspects of heart
failure management 3
Marc D. Samsky and Joseph G. Rogers

2 Myocarditis: diagnosis and treatment 15

Stuart D. Russell

3 Cardiac amyloidosis and other restrictive

myopathies: diagnosis and treatment 27
Rahul Loungani and Chetan Patel

Section 2 Mechanical circulatory support

4 Indications for ventricular assist devices 41
Antonios Kourliouros

5 Surgical techniques for left ventricular assist device

implantation and associated procedures 49
Igor Gosev and Scott Silvestry

6 Clinical trials with durable left ventricular

assist devices 61
Yuting P. Chiang and Mani A. Daneshmand

7 Diagnosis and management of ventricular assist

device complications 69
Maziar Khorsandi, Philip Curry, Sukumaran Nair, and
Nawwar Al-​Attar

8 Extracorporeal membrane oxygenation 79

Jeffrey Javidfar, Maziar Khorsandi, and Mani A. Daneshmand

9 Percutaneous mechanical circulatory support 97

Dominick Megna Jr, Maziar Khorsandi, and Danny Ramzy
 xvi Contents

109 10 The total artificial heart

Dominick Megna Jr and Francisco Arabia

117 11 Post-​cardiac surgery cardiogenic shock

Maziar Khorsandi, Philip Curry, Nawwar Al-​Attar, and
Sukumaran Nair

Section 3 Heart transplantation

12 Patient selection criteria and bridging
131 support for cardiac transplantation
Julie Doberne, Benjamin Bryner, and Chetan Patel

139 13 Cardiac donor selection and management

Ahmed S. Al-​Adhami, Hari Doshi, Nawwar Al-​Attar, and Philip Curry

14 Extended criteria heart donors and

157 perfusion storage
Yaron Barac, Anthony Castleberry, Jacob Schroder, and
Sebastian V. Rojas

15 Donor heart procurement: donation after

brain death and donation after circulatory
171 determined death
Simon Messer, Stephen Large, Marius Berman, and Steven Tsui

185 16 Technical aspects of heart transplantation

Yaron Barac, Christopher White, and Jacob Schroder

17 Critical care management and primary

195 graft dysfunction following cardiac transplantation
Sanjeet A. Singh, Jonathan R. Dalzell, Sukumaran Nair, and
Andrew Sinclair

18 Assessment and management of chronic

209 complications following heart transplantation
Anna K. Barton, Alison I. Smyth, Alan J. Kirk, and Jonathan R. Dalzell

19 Bedside care and rehabilitation with

mechanical circulatory support and
225 thoracic transplantation
Desiree Bonadonna, Lynn McGugan, and Evelyn Watson
 Contents xvii

20 Management of rejection following cardiac

transplantation 233
Sounok Sen and Adam D. DeVore

21 Antimicrobial prophylaxis and treatment of

infection after cardiac transplantation 247
Sana Arif, Sylvia Costa, and Barbara D. Alexander

Section 4 Heart failure in paediatric patients

22.1 Heart transplant for paediatric patients 267
Mohamed Nassar and Asif Hasan

22.2 Paediatric extracorporeal membrane oxygenation 275

Louise Kenny, Ed Peng, and Asif Hasan

22.3 Paediatric heart transplant: specific considerations 287

Ed Peng and Asif Hasan

Section 5 Special considerations in heart

transplantation
23 Adult patients with congenital heart disease 303
Louise Kenny and Asif Hasan

Section 6 Lung transplantation

24 Patient selection criteria and bridging support for
lung transplantation 319
Basil Nasir and John M. Reynolds

25 Lung donor selection and management 337

Stephen C. Clark

26 Lung procurement following donation after brain

death and donation after circulatory death 347
Julie Doberne, Jacob Klapper, and Matthew Hartwig

27 Ex vivo lung perfusion 355

Anders Andreasson, Karen Booth, John Dark, and Abbas Ardehali
xviii Contents

371 28 Technical aspects of lung transplantation

Basil Nasir, Pedro Catarino, and John Haney

29 Critical care management and primary

383 graft dysfunction following lung transplantation
Michael Mulvihill, Nathan Waldron, and Matthew Hartwig

30 Management of rejection following lung

389 transplantation
Lorenzo Zaffiri and Arun Nair

399 31 Late complications following lung transplantation

Yejide Odedina, Jagan Murugachandran, and Jasvir Parmar

32 Antimicrobial prophylaxis and treatment

411 after lung transplantation
Karen Booth and Julie R. Samuel

Section 7 Combined transplantation

423 33 Combined heart–​lung transplantation
Jason Ali, Jasvir Parmar, and Steven Tsui

439 34 Combined thoracoabdominal organ transplantation

Yaron Barac, Jigesh A. Shah, Jacob Schroder, and Andrew S. Barbas

Section 8 Anaesthetic considerations

35 Anaesthetic issues related to
extracorporeal membrane oxygenation
449 and ventricular assist device procedures
Michael W. Manning and Kamrouz Ghadimi

459 36 Anaesthetic issues related to heart transplantation

Philip McCall, Alina Nicoara, and Andrew Sinclair

469 37 Anaesthetic issues related to lung transplantation

Brandi Bottiger, Nazish Hashmi, and Mihai Podgoreanu

Index 481
 xix

Contributors
Ahmed S. Al-​Adhami, Francisco Arabia, MD
FRCS C/​Th Professor of Surgery
Specialty Registrar Division Chief of Cardiothoracic
Department of Cardiothoracic Surgery
Surgery University of Arizona College of
Golden Jubilee National Hospital Medicine
Glasgow, UK AZ, USA
Chapter 13 Chapter 10
Nawwar Al-​Attar, FRCS Abbas Ardehali, MD
C/​Th, PhD Director of UCLA Heart and Lung
Director of Heart Transplantation Transplantation Program
and Consultant Cardiac and Professor of Surgery and William E
Transplant Surgeon Department of Conner Chair of Cardiopulmonary
Cardiothoracic Surgery Transplantation Division of
Golden Jubilee National Hospital Cardiothoracic Surgery
Glasgow, UK Ronald Reagan UCLA Medical Center
Chapters 7,11,13 CA, USA
Chapter 27
Barbara D. Alexander, MD
Professor of Medicine and Sana Arif, MBBS
Pathology Assistant Professor of Medicine
Division of Infectious Diseases Division of Infectious Diseases
Duke University Medical Center Duke University Medical Center
NC, USA NC, USA
Chapter 21 Chapter 21
Jason Ali, MRCS, PhD Yaron Barac, MD, PhD
Speciality Registrar in Associate Professor of Surgery and
Cardiothoracic Surgery Cardiovascular Physiology
Department of Cardiothoracic Director
Surgery Heart and Lung Transplant and
Royal Papworth Hospital Mechanical Circulatory Support
Cambridge, UK Program
Chapter 33 Rabin Medical Center
Sackler Faculty of Medicine
Anders Andreasson, FRCS Tel Aviv, Israel
C/Th, PhD Chapters 14,16,34
Speciality Registrar in
Cardiothoracic Surgery Andrew S. Barbas, MD
Department of Cardiothoracic Assistant Professor of Surgery
Surgery Division of Abdominal Transplant
Freeman Hospital Surgery
Newcastle, UK Duke University Medical Center
Chapter 27 NC, USA
Chapter 34
xx Contributors

Anna K. Barton, MSc, MRCP Anthony Castleberry, MD

Clinical Research Fellow and Assistant Professor of Surgery
Honorary Cardiology Registrar Division of Cardiothoracic Surgery
Centre for Cardiovascular Science University of Nebraska Medical
The University of Edinburgh Center
Edinburgh, UK NE, USA
Chapter 18 Chapter 14
Marius Berman, FRCS C/​Th Pedro Catarino, FRCS C/​Th
Consultant Cardiac and Transplant Director of Heart and Lung
Surgeon Transplantation and Mechanical
Department of Cardiothoracic Circulatory Support and Consultant
Surgery Cardiac and Transplant Surgeon
Royal Papworth Hospital Royal Papworth Hospital
Cambridge, UK Cambridge, UK
Chapter 15 Chapter 28
Desiree Bonadonna, MPS Yuting P. Chiang, MD
Director of Extra-​Corporeal Life Resident in Cardiothoracic Surgery
Support Columbia University Medical Center
Duke University Medical Center NY, USA
NC, USA Chapter 6
Chapters 8,19
Stephen C. Clark, FRCS C/​Th
Karen Booth, FRCS C/​Th Director of Heart and Lung
Consultant Cardiac and Transplant Transplantation and Professor
Surgeon of Cardiothoracic Surgery and
Department of Cardiothoracic Cardiopulmonary Transplantation
Surgery Department of Cardiothoracic
Freeman Hospital Surgery
Newcastle, UK Freeman Hospital and Newcastle
Chapters 27,32 University
Newcastle, UK
Brandi Bottiger, MD Chapter 25
Associate Professor of Anesthesia
Division of Cardiothoracic Sylvia Costa, MD
Anesthesia and Critical Care Adjunct Assistant Professor of
Duke University Medical Center Medicine
NC, USA Division of Infectious Diseases
Chapter 37 Duke University Medical Center
NC, USA
Benjamin Bryner, MD Chapter 21
Co-​Director of Duke ECMO
Program and Assistant Professor Philip Curry, FRCS C/​Th, PhD
of Surgery Clinical Director of Cardiothoracic
Division of Cardiothoracic Surgery Surgery and Consultant Cardiac and
Duke University Medical Center Transplant Surgeon
NC, USA Department of Cardiothoracic
Chapter 12 Surgery
Golden Jubilee National Hospital
Glasgow, UK
Chapters 7,11,13
 Contributors xxi

Jonathan R. Dalzell, FRCP Kamrouz Ghadimi, MD

Consultant Heart Failure Associate Professor of
Cardiologist Anesthesiology
Scottish National Advanced Heart Division of Cardiothoracic
Failure Services Anesthesia and Critical Care
Golden Jubilee National Hospital Duke University Medical Center
Glasgow, UK NC, USA
Chapters 17,18 Chapter 35
Mani A. Daneshmand, MD Igor Gosev, MD, PhD
Director of Heart and Lung Assistant Professor
Transplantation and MCS Programs Division of Cardiothoracic Surgery
and Associate Professor of Surgery University of Rochester Medical
Division of Cardiothoracic Surgery Center
Emory University School of NY, USA
Medicine Chapter 5
GA, USA
Chapter 6 John Haney, MD
Surgical Director of Lung
John Dark, FRCS Transplantation and Assistant
Professor of Cardiothoracic Professor of Surgery
Surgery and Cardiopulmonary Division of Cardiothoracic Surgery
Transplantation Duke University Medical Center
Past President of ISHLT NC, USA
Newcastle University Chapter 28
Newcastle, UK
Chapter 27 Matthew Hartwig, MD
Associate Professor of Surgery
Adam D. DeVore, MD Division of Cardiothoracic Surgery
Assistant Professor of Medicine Duke University Medical Center
Division of Cardiology NC, USA
Duke University Medical Center Chapters 26,29
NC, USA
Chapter 20 Asif Hasan, FRCS C/​Th
Consultant Congenital Cardiac and
Julie Doberne, MD, PhD Transplant Surgeon
Resident in Cardiothoracic Surgery Children’s Heart Unit
Duke University Medical Center Freeman Hospital
NC, USA Newcastle, UK
Chapters 12,26 Chapters 22.1,22.2,23
Hari Doshi, FRCS C/​Th Nazish Hashmi, MD
Consultant Cardiac and Transplant Assistant Professor of Anesthesia
Surgeon Division of Cardiothoracic
Department of Cardiothoracic Anesthesia and Critical Care
Surgery Duke University Medical Center
Golden Jubilee National Hospital NC, USA
Glasgow, UK Chapter 37
Chapter 13
xxii Contributors

Jeffrey Javidfar, MD Stephen Large, FRCS C/​Th

Assistant Professor of Surgery and Consultant Cardiac and Transplant
Assistant Surgical Director of the Surgeon
Lung Transplantation Program Department of Cardiothoracic
Division of Cardiothoracic Surgery Surgery
Emory University School of Medicine Royal Papworth Hospital
GA, USA Cambridge, UK
Chapter 8 Chapter 15
Louise Kenny, FRCS C/​Th Rahul Loungani, MD
Specialist Registrar in Congenital Cardiology Fellow
Cardiac Surgery Division of Cardiology
Children’s Heart Unit Duke University Medical Center
Freeman Hospital NC, USA
Newcastle, UK Chapter 3
Chapters 22.2, 23
Michael W. Manning,
Maziar Khorsandi, FRCS C/​Th MD, PhD
(Managing editor) Associate Professor of Anesthesia
Assistant Professor of Surgery Division of Cardiothoracic
Division of Cardiothoracic Surgery Anesthesia and Critical Care
University of Washington Medical Duke University Medical Center
Center NC, USA
WA, USA Chapter 35
Chapters 7,8,9,11
Philip McCall, FRCA
Alan J. Kirk, FRCS Fellow in Cardiothoracic
Honorary Professor of Surgery Anaesthesia and Critical Care
Consultant Thoracic Surgeon Golden Jubilee National Hospital
Golden Jubilee National Hospital Glasgow, UK
and University of Glasgow Chapter 36
Glasgow, UK
Chapter 18 Lynn McGugan, ACNP
Critical Care Nurse Practitioner
Jacob Klapper, MD Duke University Medical Center
Assistant Professor of Surgery NC, USA
Division of Cardiothoracic Surgery Chapter 19
Duke University Medical Center
NC, USA Dominick Megna Jr, MD
Chapter 26 Assistant Professor of Surgery
Surgical Director of Lung
Antonios Kourliouros, FRCS Transplantation Program
C/​Th, PhD Division of Cardiothoracic Surgery
Consultant Cardiac Surgeon Cedar Sinai Medical Center
Oxford Heart Centre CA, USA
John Radcliffe Hospital Chapters 9,10
Oxford, UK
Simon Messer, MRCS, PhD
Chapter 4
Registrar in Cardiothoracic Surgery
Department of Cardiothoracic
Surgery
Royal Papworth Hospital
Cambridge, UK
Chapter 15
 Contributors xxiii

Carmelo Milano, MD Alina Nicoara, MD

Professor of Surgery Associate Professor of Anesthesia
Chief Section of Adult Cardiac Division of Cardiothoracic
Surgery Anesthesia and Critical Care
Division of Cardiothoracic Surgery Duke University Medical Center
Duke University Medical Center NC, USA
NC, USA Chapter 36
Michael Mulvihill, MD Yejide Odedina, MRCP
Resident in Cardiothoracic Surgery Specialty Registrar in Respiratory
Duke University Medical Center and General Internal Medicine
NC, USA Department of Respiratory and
Chapter 29 Internal Medicine
Royal Papworth Hospital
Jagan Murugachandran, MRCP Cambridge, UK
Registrar in Internal and Respiratory Chapter 31
Medicine
Specialty Registrar in Respiratory Jasvir Parmar, FRCP, PhD
Critical Care and General Internal Consultant in Respiratory Medicine
Medicine Department of Lung
Royal Papworth Hospital Transplantation
Cambridge, UK Royal Papworth Hospital
Chapter 31 Cambridge, UK
Chapters 31,33
Arun Nair, FRCP
Consultant in Respiratory Medicine Chetan Patel, MD
and Lung Transplantation Associate Professor of Medicine
Department of Respiratory Division of Cardiology
Medicine and Lung Transplantation Duke University Medical Center
Freeman Hospital NC, USA
Newcastle, UK Chapters 3,12
Chapter 30
Ed Peng, FRCS C/​Th
Sukumaran Nair, FRCS C/​Th Consultant Cardiac Surgeon, Royal
Consultant Cardiac and Transplant Hospital
Surgeon Children Glasgow and Golden
Golden Jubilee National Hospital Jubilee National Hospital
Glasgow, UK Scotland, UK
Chapters 7,11,17 Chapters 22.2, 22.3
Basil Nasir, FRCSC Mihai Podgoreanu, MD
Assistant Professor of Surgery Chief of Division of Cardiothoracic
Division of Thoracic Surgery Anesthesia and Critical Care and
Centre Hospitalier de l’Université Associate Professor of Anesthesia
de Montréal Duke University Medical Center
Montreal, Canada NC, USA
Chapters 24,28 Chapter 37
Mohamed Nassar, FRCS C/​Th Danny Ramzy, MD, PhD
Consultant Congenital Cardiac and Associate Professor of Surgery
Transplant Surgeon Division of Cardiothoracic Surgery
Freeman Hospital Cedar Sinai Medical Center
Newcastle, UK CA, USA
Chapter 22.1 Chapter 9
xxiv Contributors

John M. Reynolds, MD Jacob Schroder, MD

Medical Director of Lung Director of Heart Transplantation
Transplantation and Associate Program and Assistant Professor
Professor of Medicine of Surgery
Division of Lung Transplantation, Division of Cardiothoracic Surgery
Pulmonary and Critical Care Duke University Medical Center
Duke University Medical Center NC, USA
NC, USA Chapters 14,16,34
Chapter 24
Sounok Sen, MD
Joseph G. Rogers, MD Cardiology Fellow
Chief Medical Officer and Professor Division of Cardiology
of Medicine Duke University Medical Center
Division of Cardiology NC, USA
Duke University Medical Center Chapter 20
NC, USA
Chapter 1 Jigesh A. Shah, DO
Fellow in Abdominal Organ
Sebastian V. Rojas, MD Transplantation
Co-​Director of Cardiac Division of Abdominal
Transplantation Transplantation
Department of Cardiothoracic, Duke University Medical Center
Transplantation and Vascular NC, USA
Surgery Chapter 34
Hannover Medical School
Hannover, Germany Scott Silvestry, MD
Chapter 14 Surgical Director of Thoracic
Transplantation and Associate
Stuart D. Russell, MD Professor of Surgery
Director of Heart Failure and Division of Cardiothoracic Surgery
Professor of Medicine Florida Hospital Transplant Institute
Division of Cardiology FL, USA
Duke University Medical Center Chapter 5
NC, USA
Chapter 2 Andrew Sinclair, FRCA
Director of Mechanical Circulatory
Marc D. Samsky, MD Support and Consultant in
Cardiology Fellow Anaesthesia and Critical Care
Division of Cardiology Golden Jubilee National Hospital
Duke University Medical Center Glasgow, UK
NC, USA Chapters 17,36
Chapter 1
Sanjeet A. Singh, MRCS, PhD
Julie R. Samuel, FRCPath Clinical Fellow in Heart
Clinical Director of the Department Transplantation
of Clinical Microbiology and Department of Cardiothoracic
Consultant Clinical Microbiologist Surgery
Freeman Hospital Golden Jubilee National Hospital
Newcastle, UK Glasgow, UK
Chapter 32 Chapter 17
 Contributors xxv

Alison I. Smyth, MRCP Evelyn Watson, RN

Clinical Fellow in Heart Failure Specialist Charge Nurse
Cardiology Regional Transplant Unit
Scottish National Advanced Heart Freeman Hospital and Newcastle
Failure Services University
Golden Jubilee National Hospital Newcastle, UK
Glasgow, UK Chapter 19
Chapter 18
Christopher White,
Steven Tsui, FRCS C/​Th FRCSC, PhD
Clinical Director of Transplant Heart Failure Surgeon
Services and Director of Mechanical Assistant Professor of Surgery
Circulatory Support Dalhousie University
Consultant Cardiac and Transplant New Brunswick Heart Center
Surgeon NB, Canada
Royal Papworth Hospital Chapter 16
Cambridge, UK
Chapters 15,33 Lorenzo Zaffiri, MD
Assistant Professor of Medicine
Nathan Waldron, MD Division of Lung Transplantation,
Assistant Professor in Anaesthesia Pulmonary and Critical Care
and Critical Care Duke University Medical Center
University of Colorado Health NC, USA
Memorial Hospital Central Chapter 30
CO, USA
Chapter 29
xxvi
 xxvii

Abbreviations
ABOi AABO incompatible CNS central nervous system
ACC American College of COPD chronic obstructive pulmonary
Cardiology disease
ACE angiotensin-​converting enzyme CPB cardiopulmonary bypass
ACEi angiotensin-​converting enzyme CPR cardiopulmonary resuscitation
inhibitor CPX cardiopulmonary exercise
ACHD adult congenital heart disease CRT cardiac resynchronization
ACT activated clotting time therapy
ADHERE Acute Decompensated Heart CSF cerebrospinal fluid
Failure National Registry CVP central venous pressure
AHA American Heart Association CXR chest X-​ray
AL amyloid light chain DBD donation after brain death
AMR antibody-​mediated rejection DC direct current
Ao aorta DCD donation after
aPTT activated partial circulatory death
thromboplastin time DCM dilated cardiomyopathy
ARB angiotensin II receptor DLCO diffusion capacity of the lung
blocker for carbon monoxide
ARDS acute respiratory distress DPC distal perfusion cannula
syndrome
DSA donor-​specific antibody
ASD atrial septal defect
EACTS European Association for
ATG anti-​thymocyte globulin Cardio-​Thoracic Surgery
ATTR-​CM transthyretin amyloid EBV Epstein–​Barr virus
cardiomyopathy
ECG electrocardiogram
AV atrioventricular
ECLS extracorporeal life support
BAL bronchoalveolar lavage
ECMO extracorporeal membrane
BB beta blocker oxygenation
BiVAD biventricular assist device E-​CPR extracorporeal
BMI body mass index cardiopulmonary resuscitation
BNP B-​type natriuretic peptide EF ejection fraction
BOS bronchiolitis obliterans eGFR estimated glomerular
syndrome filtration rate
BSA body surface area ELSO Extracorporeal Life Support
BSD brainstem death Organization
BTT bridge to transplantation ESR erythrocyte sedimentation rate
CAD coronary artery disease EVLP ex vivo lung perfusion
CAV cardiac allograft vasculopathy FDA Food and Drug Administration
CDAD Clostridium difficile-​associated FEV1 forced expiratory volume in
diarrhoea 1 second
CF cystic fibrosis FiO2 fraction of inspired oxygen
CI cardiac index FVC forced vital capacity
CLAD chronic lung allograft GFR glomerular filtration rate
dysfunction GI gastrointestinal
CMV cytomegalovirus GTN glycerol trinitrate
CNI calcineurin inhibitor HB hepatitis B
xxviii Abbreviations

HBV hepatitis B virus LVEDP left ventricular end-​diastolic

HCV hepatitis C virus pressure
HF heart failure LVEF left ventricular ejection fraction
HFpEF heart failure with preserved LVH left ventricular hypertrophy
ejection fraction MAP mean arterial pressure
HFrEF heart failure with reduced MCS mechanical circulatory support
ejection fraction MELD Model for End-​Stage Liver
HFSS Heart Failure Survival Score Disease
HHV herpesvirus MFI mean fluorescent intensity
HIT heparin-​induced MMF mycophenolate mofetil
thrombocytopenia MR mitral regurgitation
HIV human immunodeficiency virus MRA mineralocorticoid receptor
HKTx heart–​kidney transplantation antagonist
HLA human leucocyte antigen MRI magnetic resonance imaging
HLTx heart–​lung transplantation MRSA methicillin-​resistant
HM HeartMate Staphylococcus aureus
HSV herpes simplex virus mTOR mechanistic target of
rapamycin
hTTR hereditary transthyretin
MV mitral valve
HZ herpes zoster
NAAT nucleic acid amplification
IA invasive aspergillosis testing
IABP intra-​aortic balloon pump/​ NICM non-​ischaemic
counterpulsation cardiomyopathy
ICD implantable cardioverter NO nitric oxide
defibrillator
NODAT new-​onset diabetes after
ICU intensive care unit transplantation
Ig immunoglobulin NRAD neutrophilic reversible allograft
IJ internal jugular dysfunction
IL-​2 interleukin-​2 NTM non-​tuberculous mycobacteria
IL2R interleukin-​2 receptor NT-​proBNP N-​terminal pro-​B type
IL2RA interleukin-​2 receptor natriuretic peptide
antagonist NYHA New York Heart Association
ILD interstitial lung disease OHT orthotopic heart
INR international normalized ratio transplantation
INTERMACS Interagency Registry for PA pulmonary artery
Mechanically Assisted PAC pulmonary artery catheter
Circulatory Support PaCO2 partial arterial pressure of
ISHLT International Society for Heart carbon dioxide
and Lung Transplant PaO2 partial arterial pressure
IV intravenous of oxygen
IVIG intravenous immunoglobulin PCCS post-​cardiotomy
LA left atrial/​atrium cardiogenic shock
LAA left atrial appendage PCO2 partial pressure of carbon
LAS lung allocation score dioxide
LKTx lung–​kidney transplantation PCR polymerase chain reaction
LLTx lung–​liver transplantation PCWP pulmonary capillary wedge
pressure
LV left ventricular/​ventricle
PEEP positive end-​expiratory
LVAD left ventricular assist device pressure
LVEDD left ventricular end-​diastolic PFO patent foramen ovale
diameter
PFT pulmonary function testing
PGD primary graft dysfunction
 Abbreviations xxix

PH pulmonary hypertension SVR systemic vascular resistance

PJP Pneumocystis jirovecii TAH total artificial heart
pneumonia TAVR transcatheter aortic valve
PLE protein-losing enteropathy replacement
PO per os (orally) TBBX transbronchial biopsy
PO2 partial pressure of oxygen TEG thromboelastography
PRA panel-​reactive antibody TLC total lung capacity
PT physical therapy TMP–​SMX trimethoprim–​
PTLD post-​transplant sulfamethoxazole
lymphoproliferative disorder TOE transoesophageal
PVR pulmonary vascular resistance echocardiography/​
echocardiogram
RA right atrial/​atrium
TR tricuspid regurgitation
RAA right atrial appendage
TTE transthoracic
RAAS renin–​angiotensin–​ echocardiography
aldosterone system
TTR transthyretin
RAS restrictive allograft syndrome
UFH unfractionated heparin
rATG rabbit anti-​thymocyte globulin
UNOS United Network for Organ
RCM restrictive cardiomyopathy Sharing
rpm revolutions per minute V/​Q ventilation/​perfusion
RV right ventricular/​ventricle VA venoarterial
RVAD right ventricular assist device VAD ventricular assist device
RVEDP right ventricular end-​diastolic VAV venoarteriovenous
pressure
VO2max maximal oxygen
RVEF right ventricular ejection consumption
fraction
VSD ventricular septal defect
SHFM Seattle Heart Failure Model
VV venovenous
SOT solid organ transplant
VZV varicella zoster virus
SVC superior vena cava
WHO World Health Organization
SvO2 mixed venous oxygen
saturation wtTTR wild-​type transthyretin
x
 Section 1

Heart failure

1 Medical and minimally invasive aspects of heart failure

management 3
2 Myocarditis: diagnosis and treatment 15
3 Cardiac amyloidosis and other restrictive myopathies:
diagnosis and treatment 27
2
 Chapter 1 3

Medical and minimally

invasive aspects of heart
failure management
Epidemiology of chronic heart failure 4
Clinical assessment of heart failure 4
Aetiologies of heart failure 4
Evaluation for patients with heart failure with reduced ejection
fraction 7
Pharmacotherapies for heart failure with reduced ejection
fraction 8
Prevention of sudden cardiac death 11
Cardiac resynchronization therapy 12
Transcatheter valvular therapies for heart failure 12
Prognostication for heart failure 13
Referral to advanced heart failure specialists 14
4

4 Chapter 1 Management of chronic heart failure

Epidemiology of chronic heart failure

Heart failure (HF) is a major public health concern affecting >6 million
Americans. The care of patients with HF places a significant burden on the
US healthcare system in terms of morbidity, mortality, and cost. In 2012,
there were nearly 2 million clinic visits, 500,000 emergency room visits, and
>1 million hospital discharges (~25% of discharged patients are readmitted
within 30 days) for HF-​related care with an associated total cost of $30 bil-
lion. Projections estimate $70 billion in total HF-​related spending by 2030,
with approximately 80% of costs coming directly from hospital admissions.

Clinical assessment of heart failure

HF is characterized by functional limitations resulting from sequelae of in-
creased intracardiac filling pressures or low cardiac output. Clinical assess-
ment including a detailed history and physical examination is imperative
to differentiate patients with HF from a variety of alternative diagnoses.
Worsening dyspnoea, orthopnoea, paroxysmal nocturnal dyspnoea, per-
ipheral oedema, and elevated jugular venous pulsations are all common
signs and symptoms of HF patients and are suggestive of increased left-​and
right-​sided filling pressures. Patients with low cardiac output may complain
of fatigue and poor exercise tolerance though cool/​mottled extremities,
resting tachycardia, and a narrow pulse pressure are findings more specific
of a low output state. Pertinent items from the patient history, common
signs and symptoms, and important physical examination findings for HF
are listed in Box 1.1.
After confirmation of a HF diagnosis, further stratification should be
based on disease severity and/​ or functional limitation. The American
College of Cardiology (ACC) and American Heart Association (AHA)
developed a staging system focused on disease progression whereas the
New York Heart Association (NYHA) classification is based upon func-
tional limitation (Table 1.1). Both classification systems provide unique and
complementary schemas that facilitate an understanding of the clinical con-
dition and prognosis of individuals and cohorts of patients.

Aetiologies of heart failure

Development of primary cardiomyopathy can result from impaired struc-
ture/​functioning of the myocardium, cardiac valves, or conduction system.
Cardiomyopathy can also occur secondary to disordered functioning of an-
other organ, such as cor pulmonale, where chronic lung disease leads to
pulmonary hypertension (PH) and subsequent right HF.
Meta-​analyses suggest that underlying coronary artery disease (CAD) is
the most common aetiology of heart failure with reduced ejection frac-
tion (HFrEF), and is responsible for at least 65% of the diagnoses. Chronic
hypertension and diabetes are also important risk factors for development
of HFrEF. In fact, CAD, hypertension, and diabetes act synergistically to
augment the risk of development of HFrEF. As such, an emphasis on HFrEF
prevention has become an important mission of the ACC and AHA.
 Aetiologies of heart failure 5

Box 1.1 Common findings in heart failure

Pertinent items from patient history suggestive of heart failure
• Previous history of HF.
• Other cardiovascular comorbidities (CAD, congenital heart disease,
valvular disease, left bundle branch block).
• Previous myocardial infarction.
• Risk factors for heart failure (diabetes, hypertension, obesity).
• Systemic illnesses with cardiac involvement (amyloidosis, sarcoidosis,
Chagas, rheumatic, glycogen storage disease).
• Chronic human immunodeficiency virus (HIV) infection or acquired
immune deficiency syndrome (AIDS).
• Recent viral-​like illness.
• Toxic exposures.
• Substance abuse (e.g. alcohol).
• Current malignancy or previous exposure to chemotherapy.
Signs and symptoms of heart failure
• Shortness of breath.
• Dyspnoea on exertion.
• Peripheral oedema.
• Orthopnoea/​paroxysmal nocturnal dyspnoea.
• Increasing abdominal girth/​bloating.
• Early satiety.
• Fatigue.
• Altered mental status.
• Dizziness/​presyncope/​syncope.
• Bendopnea (shortness of breath when bending over).
Physical examination findings suggestive of heart failure
• Tachycardia.
• Narrow pulse pressure or thready pulse.
• Tachypnoea.
• Elevated jugular venous pressure.
• Rales or diminished breath sounds.
• Point of maximal cardiac impulse laterally displaced.
• S3 or S4 heart sounds.
• Hepatomegaly with or without pulsatile liver.
• Ascites.
• Peripheral oedema with or without chronic skin changes.
• Cheyne-Stokes breathing (Oscillatory breathing pattern).

Non-​ischaemic cardiomyopathies (NICMs) are also an important group

of diagnoses that should always be considered when evaluating patients
with HF. Valvular heart disease (rheumatic and non-​rheumatic), electrical
conduction system disorders (incessant/​ inappropriate tachycardia or
chronic left bundle branch block chronic right ventricular pacing), toxin and
chemotherapy exposure, excessive alcohol consumption, infiltrative sys-
temic diseases, viral infections, Chagas disease, and other NICMs constitute
approximately 25% of cases of HFrEF. Genetic cardiomyopathies are also
an important aetiology of NICM and have been reported to be responsible
6

6 Chapter 1 Management of chronic heart failure

Table 1.1 Heart failure staging and functional classification

ACC/​AHA staging NYHA functional classification
At high risk of HF but without A None
structural abnormalities or
symptoms
Structural heart disease B I No limitation; ordinary
present but without physical exertion does not
symptoms of HF cause HF symptoms
Structural heart disease C I No limitation; ordinary
present, with previous or physical exertion does not
current symptoms of HF cause HF symptoms
II Slight limitation; no resting
symptoms; ordinary
physical exertion causes HF
symptoms
III Marked limitation; no
resting symptoms; less than
ordinary activity causes HF
symptoms
Refractory HF D IV Significant limitation;
requiring specialized symptoms of HF present
interventions at rest or with minimal
exertion

for up to 50% of idiopathic dilated cardiomyopathies. It is important to

evaluate for these exposures, even in the presence of risk factors for is-
chaemic cardiomyopathy, as many of them can be at least partially treated,
resulting in improvement of native cardiac function. Importantly, a substan-
tial number of patients have an idiopathic cardiomyopathy, which usually

Box 1.2 Aetiologies of heart failure

• CAD (acute myocardial infarction and chronic ischaemia).
• Chronic hypertension.
• Valvular heart disease (stenotic and regurgitant lesions).
• Intra-​or extracardiac shunting.
• Familial/​genetic.
• Toxins.
• Medications (anthracycline chemotherapies, taxane chemotherapies,
tumour necrosis factor alpha inhibitors, neuropsychiatric stimulants, etc.).
• Infiltrative diseases (sarcoidosis, amyloidosis).
• Metabolic disorders (e.g. thyrotoxicosis, beriberi, glycogen storage
disease).
• Myocarditis.
• Chronic tachyarrhythmias.
• Chronic bradyarrhythmias.
– Left bundle branch block
– Chronic RV pacing
• Cor pulmonale.
 MEDICAL EVALUATION FOR HEART FAILURE PATIENTS 7

presents as dilated NICM. Aetiologies of cardiomyopathy and HF are listed

in Box 1.2.

Evaluation for patients with heart failure

with reduced ejection fraction
Transthoracic echocardiography
Non-​invasive cardiac imaging plays a vital role in the assessment of patients
with HF. Transthoracic echocardiography (TTE) can be performed at the
bedside without risk to the patient and without associated radiation ex-
posure. TTE can be used to help determine the aetiology of HF via assess-
ment of structural abnormalities, electromechanical dyssynchrony, valvular
pathology, and pericardial disease. Routine assessment includes quantita-
tively measuring systolic function, wall thickness, chamber sizes and vol-
umes. These measurements can be particularly helpful in considering HF
aetiology, chronicity, and potential for recovery. Furthermore, TTE can be
used to measure intracardiac filling pressures and flow across valves.
TTE is also commonly used to assess right-​sided filling pressures, which
can be helpful in guiding diuretic therapy. Right atrial (RA) pressure can
be estimated by measuring inferior vena cava (IVC) diameter and relative
change in size with inspiration. Normal IVC size with inspiratory collapse of
>50% is associated with normal RA pressure. An enlarged IVC or smaller
inspiratory collapse is associated with increased RA pressures.
Heart catheterization
Determining the presence of significant concomitant CAD in HFrEF has
important therapeutic and prognostic implications. Cardiac catheterization
via invasive coronary arteriography is considered the ‘gold standard’ for as-
sessing the presence and severity of CAD and its contribution to HF. Based
on results from the Surgical Treatment for Ischemic Heart Failure (STICH)
and associated STICH Extension Study (STICHES), patients with CAD and
HF with reduced systolic dysfunction benefit from coronary artery bypass
grafting.
Direct measurement of intracardiac pressures and cardiac output via
right heart catheterization can be useful when there is uncertainty regarding
aetiology of symptoms or when precise measurements are required for
determining candidacy of surgical HF therapies (transplantation or durable
mechanical circulatory support). Furthermore, right heart catheterization
can be helpful when dynamic changes in intracardiac pressure and volume
need to be measured. A common example is an exercise right heart cath-
eterization to determine the severity of mitral regurgitation (MR) or dia-
stolic dysfunction. Right heart catheterization with concomitant pulmonary
vasodilator administration may be performed to determine the aetiology of
PH, candidacy for medications, and response to treatment.
Functional testing
Reduced exercise tolerance is a common complaint in patients with even mild
HF, and can be exacerbated by common comorbidities. The 6-​minute walk
test is a valuable tool that can be safely administered to quantify submaximal
8

8 Chapter 1 Management of chronic heart failure

functional capacity in patients with HF. Furthermore, it can be measured

serially and has been shown to predict outcomes. Cardiopulmonary exercise
(CPX) testing is typically used to more completely quantify and characterize
exercise capacity in patients with HF. CPX testing distinguishes between
aetiologies of functional limitation including poor effort, pulmonary disease,
or lack of cardiovascular reserve. Peak oxygen consumption, commonly re-
ferred to as ‘VO2max’, is measured during CPX testing and provides the
most objective assessment of functional capacity. Peak oxygen consumption
is a foundational component in the determination of patients likely to benefit
from surgical HF therapies including transplantation or durable mechanical
circulatory support.

Pharmacotherapies for heart failure

with reduced ejection fraction
Chronic HF is characterized by a compensatory activation of the sym-
pathetic nervous system and the renin–​angiotensin–​aldosterone system
(RAAS) as a result of decreased cardiac output, reduced stimulation of
arterial baroreceptors, and renal hypoperfusion. Activation of the RAAS
system ultimately results in the conversion of angiotensin I to the biologic-
ally active molecule angiotensin II via angiotensin-​converting enzyme (ACE).
Angiotensin II has several negative effects in chronic HF including vaso-
constriction, sodium retention, stimulation of aldosterone release causing
myocardial fibrosis, and co-​stimulation of the sympathetic nervous system
leading to increased levels of plasma catecholamines. Thus, the pharmaco-
logical foundation for chronic HF is centred around reducing sympathetic
nervous system activation and RAAS signalling.
Beta blockers
Long-​term treatment of chronic HF with beta blockers (BBs) reduces
symptom burden, improves functional status, halts or reverses disease pro-
gression, and reduces hospitalization and mortality. Beta antagonists inhibit
circulating catecholamine stimulation of adrenergic receptors by competi-
tive binding to reduce the compensatory sympathetic overdrive of patients
with chronic HF.
Patients with symptomatic HF and a left ventricular ejection fraction
(LVEF) of ≤40% should be treated with a BB. Three BBs have evidence
supporting their use in patients with chronic HF: bisoprolol, metoprolol suc-
cinate, and carvedilol. Beta antagonists are a complex class of drugs with a
variety of pharmacological effects. BBs may inhibit only the β1 receptor,
both β1 and β2 receptors, or β1, β2, and alpha receptors. In addition, some
of the BBs have intrinsic sympathomimetic properties that mimic the effects
of epinephrine and norepinephrine increasing both heart rate and blood
pressure. Improved outcomes from BB use in patients with chronic HF is
not a class effect. Bucindolol and metoprolol tartrate (short acting) have
not been shown to be as effective, especially in select subgroups of pa-
tients. One of the three guideline-​directed BBs should be prescribed to
all patients with stable HF unless there is a contraindication or there has
been previous intolerance. In general, it is recommended that patients be
 PHARMACOTHERAPIES FOR HEART FAILURE 9

on stable doses of diuretic (if needed) prior to initiation of BB as there is a

well-​documented transient increase in risk of fluid retention with initiation
or increase in BB dose.
Renin–​angiotensin–​aldosterone system antagonists
Inhibition of the RAAS is a key component of treatment for chronic HF.
Increased RAAS signalling as a result of relative renal hypoperfusion in-
creases circulating levels of angiotensin II. Until recently, use of either an
angiotensin converting enzyme inhibitor (ACEi), or angiotensin II receptor
blocker (ARB), in addition to a BB was considered cornerstone therapy for
patients with symptomatic HF and an LVEF of ≤40%. Treatment of patients
with chronic HF with an ACEi or ARB (in patients with ACEi intolerance)
has been repeatedly shown to reduce mortality, reduce HF hospitalization,
and improve functional status.
A new class of drugs that inhibit the angiotensin receptor and neutral
endopeptidase (neprilysin) have been shown to be superior to an ACEi for
reducing mortality and HF hospitalization in patients with symptomatic HF.
The Prospective Comparison of ARNI [angiotensin receptor–​neprilysin in-
hibitor] with ACEI to Determine Impact on Global Mortality and Morbidity
in Heart Failure (PARADIGM-​HF) trial was a randomized double-​blind trial
comparing sacubitril–​valsartan with enalapril in patients with class II–​IV
HFrEF. The trial was stopped prematurely given the benefit of treatment
with sacubitril–​valsartan. Sacubitril–​valsartan was associated with an abso-
lute risk reduction of nearly 5% in the combined endpoint of cardiovas-
cular death or HF hospitalization, and 3% for cardiovascular death alone.
Given the positive data for use of sacubitril–​valsartan, the ACC and AHA
have published updates to the HF management guidelines to assist phys-
icians and other providers with initiation and dosing. Criteria for initiation
include elevated serum natriuretic peptide levels, a systolic blood pressure
of >100 mmHg, and a glomerular filtration rate (GFR) of at least 30 mL/​
min/​1.73 m2. As with all other HF therapies, dose adjustments should be
titrated to systolic blood pressure or evidence of intolerance such as renal
dysfunction or angioedema.
Mineralocorticoid receptor antagonists
Mineralocorticoid receptor antagonists (MRAs) such as spironolactone and
eplerenone also block the effects of RAAS activation and have been shown
to improve outcomes in patients with chronic HF treated with maximally
tolerated doses of BB and ACEi/​ARB. Mechanistically, MRAs inhibit the
deleterious effects of aldosterone, which acts directly on the heart by pro-
moting development of cardiac fibrosis. Mineralocorticoid receptors are
overexpressed in diseased hearts in proportion to the degree of cardiac
dysfunction. Aldosterone binding to the mineralocorticoid receptor in the
diseased heart creates a cycle of production of angiotensin II and thus a
propagating cycle of RAAS activation and worsening left ventricular (LV)
dysfunction and symptomatic HF.
Sodium Glucose Co-transporter 2 Inhibitors
Sodium Glucose Co-transporter 2 Inhibitors (SGLT2i) are a class of medi-
cations that inhibit glucose reabsorption in the proximal convoluted tubule
of the nephron. These medications have been repeatedly shown to reduce
10

10 Chapter 1 Management of chronic heart failure

all-cause mortality, cardiovascular mortality, and HF hospitalizations in pa-

tients with HF regardless of diabetes status. The mechanism of action re-
lated to improving cardiovascular outcomes in patients with HF is not yet
understood.
Ivabradine
Ivabradine is a selective sinus node inhibitor which acts specifically on the
hyperpolarization-​ activated cyclic nucleotide-​ gated channels within the
sinoatrial node. Thus, ivabradine slows depolarization within the sinus node
to lower resting heart rate without affecting myocardial contractility or re-
laxation. Further, ivabradine has no impact on blood pressure. The Systolic
Heart Failure Treatment with the IF Inhibitor Ivabradine Trial (SHIFT) ran-
domized patients already on maximally tolerated HF medications and a
resting sinus heart rate of at least 70 beats per minute to ivabradine or
placebo. Patients randomized to ivabradine experienced an improvement
in the primary combined endpoint of cardiovascular death or hospital ad-
mission for worsening HF. Ivabradine is therefore recommended for pa-
tients with chronic HF with an LVEF of ≤35% and are in sinus rhythm with a
resting heart rate of at least 70 beats per minute despite maximal BB dosage
and RAAS inhibition. Importantly, ivabradine should not be used as a full or
partial substitute for BB.
Hydralazine and isosorbide dinitrate
Fixed-​dose combination hydralazine plus isosorbide dinitrate for treatment
of chronic HF is reserved for selected patient populations. Specifically,
for patients intolerant of RAAS blocking agents due to angioedema, or
for black patients with persistent symptoms of HF despite optimally ti-
trated BB plus ACEi/​ARB and MRA. The African-​American Heart Failure
Trial (A-​HeFT) was terminated early due to significant improvement in
the combined endpoint of death, HF hospitalization, and quality of life in
patients treated with fixed-​dose combination hydralazine plus isosorbide
dinitrate compared to placebo (6.2% vs 10.2%, p =​0.02). While not dir-
ectly assessed in clinical trials, the relative benefits of sacubitril–​valsartan
are sufficiently pronounced such that it should take priority over the use of
hydralazine–​nitrates.
Loop diuretics
Despite a glaring absence of evidence to support their use, loop diuretics
are among the most frequently used pharmacotherapies for HF. Loop diur-
etics including furosemide, torsemide, and bumetanide work on the apical
surface of the thick ascending loop of Henle to directly inhibit transport
of sodium, potassium, and chloride ions. Inhibition of ion reabsorption
in this location effectively leads to the natriuretic effect of loop diuretics.
Administration of loop diuretics intravenously can also have a vasodila-
tory effect via action on ion transporters located in vascular smooth
muscle cells.
Each loop diuretic has a different bioavailability when administered orally.
Furosemide has the greatest variability, with a mean of 50% and ranges from
10% to 90%. Furosemide absorption can be delayed with food intake as well
as in patients with significant gut oedema or low duodenal blood flow. This
in turn leads to reduced peak plasma concentration and can contribute to
diuretic resistance.
 Prevention of sudden cardiac death 11

Torsemide and bumetanide have higher and more predictable oral bio-
availability (>90%) with associated pharmacokinetics not limited by ab-
sorption. Of the loop diuretics, torsemide has the longest half-​life (6
hours vs 2.7 hours for furosemide vs 1.3 hours for bumetanide). There
are limited data comparing loop diuretics in patients with HF. The ongoing
Torsemide Comparison with Furosemide for Management of Heart Failure
(TRANSFORM-​HF) trial will determine if treatment with torsemide im-
proves 12-​month all-​cause mortality for patients with HF.
When dosing loop diuretics, it is important to remember that they all
have steep dose–​ response curves or ‘threshold effect’. Increasing or
escalating doses of medication will not increase the natriuretic effect, but
may act to prolong the time of serum drug concentration above the re-
quired threshold to achieve desired effect. It is important to account for
this concept and relative bioavailability when choosing and dosing diuretics
for patients with HF.
Inotropes
Positive inotropic agents cause an increase in cardiac output via a variety
of pharmacological targets including inhibition of phosphodiesterase-​ 3
(milrinone), beta-​adrenergic receptor agonism (dobutamine), calcium sensi-
tization (levosimendan), and increased intracellular calcium via inhibition of cell
membrane ion exchange (digoxin). Though these medications may tempor-
arily improve haemodynamics via their ability to augment cardiac contractility
and/​or reduce pulmonary and systemic vascular resistance (SVR), the use of
intravenous inotropes for management of acute or chronic HF lacks evidence.
Digoxin can be administered either orally or intravenously and has
been studied prospectively for patients with chronic HF. The Digitalis
Investigation Group (DIG) study randomized patients receiving ACEi and
diuretic to either placebo or oral digoxin. Patients in the treatment arm ex-
perienced a reduced rate of overall hospitalization and HF hospitalization.
There was no difference in all-​cause mortality. The results of the DIG trial,
in combination with use of other evidence-​based guideline-​directed medical
therapy for HF, has dramatically reduced the use of oral digoxin for patients
with chronic HF. Interestingly, findings from the Randomized Assessment of
the Effect of Digoxin on Inhibitors of the Angiotensin Converting Enzyme
(RADIANCE) and Prospective Randomized Study of Ventricular Failure
and the Efficacy of Digoxin (PROVED) trials suggest that discontinuation of
digoxin increases the risk of adverse outcomes in ambulatory patients with
HFrEF. Similar findings have been found in patients who are hospitalized for
HF and have digoxin discontinued during admission.

Prevention of sudden cardiac death

Ventricular arrhythmias are common in patients with chronic HF. Sustained
ventricular arrhythmias and sudden cardiac death are responsible for as
much as two-​thirds of HF-​related mortality. As such, implantation of pri-
mary and secondary prevention cardioverter defibrillators is considered a
class I recommendation with a strong level of evidence. Clinical practice
guidelines recommend placement of an implantable cardioverter defibril-
lator (ICD) in patients with NYHA class II–​III symptoms and an LVEF of
≤35% who are on maximally tolerated guideline-​directed medical therapies
12

12 Chapter 1 Management of chronic heart failure

for HF. For patients with ischaemic heart disease, clinical trials and guidelines
recommend waiting at least 40 days post myocardial infarction and at least
3 months following revascularization for ICD implantation (if applicable).
Placement of a secondary prevention ICD should be done following an
event of sudden cardiac death if obviously reversible causes of ventricular
arrhythmias have been excluded such as acute myocardial ischaemia/​infarc-
tion, electrolyte abnormalities, and toxic exposures including medication
exposures that are proarrhythmic.

Cardiac resynchronization therapy

Observational studies suggest that electrical and mechanical dyssynchrony
induced by a left bundle branch block on electrocardiogram (ECG) can
lead to cardiac remodelling with subsequent cardiomyopathy, chronic
HF with exacerbation of functional MR, and increased mortality. Cardiac
resynchronization therapy (CRT) involves implantation of a permanent
transvenous pacing system that provides concomitant pacing of the left and
right ventricles. This promotes electrical and mechanical ventricular syn-
chrony. Randomized trials have shown benefit in patients with a QRS of at
least 150 milliseconds with a left bundle branch block morphology, symp-
tomatic HF, on maximally tolerated evidence-​based pharmacotherapies,
and an LVEF of ≤35%. Similar to ICD therapy, implantation should be at
least 40 days following an episode of myocardial ischaemia or an acute myo-
cardial infarction. Reversible causes of cardiomyopathy and HF should also
be addressed prior to referral for CRT. Importantly, there is no evidence
of benefit in patients with right bundle branch block or incomplete/​partial
left bundle branch block. For patients with an LVEF of >35% but <50%, if a
high burden of ventricular pacing is needed for another indication, it is also
reasonable to pursue CRT placement.

Transcatheter valvular therapies

for heart failure
Transcatheter mitral repair for secondary
mitral regurgitation
MR can occur as a result of primary pathology of the mitral valve (MV) leaf-
lets or secondary to pathology of the structure and function of the heart.
For secondary MR due to LV dilation and secondary distortion of the MV
annulus, transcatheter MV repair has been recently shown to be a viable
therapy. The MitraClip, based on the Alfieri edge-​to-​edge leaflet repair, is
the only US Food and Drug Administration-​approved device for percutan-
eous repair of MR.
The Cardiovascular Outcomes Assessment of the MitraClip Percutaneous
Therapy for Heart Failure Patients with Functional Mitral Regurgitation
(COAPT) and Percutaneous Repair with the MitraClip Device for Severe
Functional/​Secondary Mitral Regurgitation (MITRA-​FR) trials provide the
only evidence to guide percutaneous repair of secondary MR. The COAPT
trial enrolled >600 patients with moderate to severe or severe secondary
 Prognostication for heart failure 13

MR with associated LVEF of 20–​50% and at least NYHA class II HF symp-

toms despite maximal tolerated doses of HF pharmacotherapies. Patients
were randomized to continue medical therapies plus transcatheter MV re-
pair (MitraClip) or medical therapy alone. Patients treated with MitraClip
had a significant reduction in mortality and HF hospitalization (evident within
30 days of therapy) compared to the medical therapy arm. Furthermore,
patient-​reported outcomes related to functional capacity and quality of life
were improved after receiving transcatheter MV repair compared to medical
therapy alone.
In MITRA-​FR, 304 patients with at least moderate secondary MR, with
associated LVEF of 15–​40% and symptomatic HF, were randomized to
either medical therapy plus transcatheter MV repair or medical therapy
alone. The results of MITRA-​FR were negative for the primary out-
come of all-​cause mortality at 12 months or in the rate of unplanned HF
hospitalization.
The discrepant results of these trials are not entirely obvious but im-
portant differences include a larger study population, longer duration of
follow-​up, and a higher severity of MR in COAPT. Taken in aggregate, the
data from COAPT and MITRA-​FR suggest that in appropriately selected
patients, transcatheter MV repair is a reasonable strategy to improve im-
portant clinical outcomes, functional capacity, and quality of life.
Transcatheter aortic valve replacement for heart failure
Transcatheter aortic valve replacement (TAVR) is an alternative to surgical
valve replacement in patients with severe aortic stenosis, especially those at
increased risk of death from surgery. Aortic stenosis and HF are common
diagnoses in older patients (>75 years old) and likely synergistically lead to
worsened outcomes. Unfortunately, evidence-​based HF medications have
no effect on the valvular component of afterload in aortic stenosis and as-
sociated outcomes. Thus, the Transcatheter Aortic Valve Replacement to
Unload the Left Ventricle in Patients with Advanced Heart Failure (TAVR
UNLOAD) trial (NCT02661451) was designed to test the hypothesis that
TAVR on top of optimal medical therapy improves outcomes in patients
with moderate aortic stenosis and HFrEF.

Prognostication for heart failure

Risk scores
In addition to assessment of ACC/​AHA staging and NYHA functional
classification, routine assessment of the potential for adverse outcome is
an important part of the evaluation and management of patients with HF.
Validated scores for risk stratification in patients with chronic and acute
HF exist. The Seattle Heart Failure Model (SHFM) is the most robust
and commonly used risk score to predict timing and mode of death in
ambulatory patients with chronic HF. The SHFM can be accessed online
(https://​depts.was​hing​ton.edu/​shfm) and incorporates multiple clin-
ical variables including demographics, LVEF, vital signs, medications, and
various laboratory values to accurately predict 1-​, 2-​, and 3-​year survival.
Importantly, the SHFM model has been validated repeatedly in multiple
contexts.
14

14 Chapter 1 Management of chronic heart failure

For patients admitted with acutely decompensated HF, a model devel-

oped from the Acute Decompensated Heart Failure National Registry
(ADHERE) can be used readily at the bedside to predict in-​hospital mor-
tality. Importantly, the ADHERE score for acute HF involves only three
routinely measured clinical variables: systolic blood pressure, blood urea
nitrogen, and serum creatinine.

Referral to advanced heart

failure specialists
Patients with chronic HF who progress to develop persistent and severe
symptoms despite optimal medical therapy should be referred to an ad-
vanced HF team. Further manifestations of worsening HF or development
of end-​stage disease include intolerance of evidence-​based medical ther-
apies due to hypotension, NYHA class IV symptoms, refractory volume
overload requiring frequent hospitalization, cardiac cachexia with associ-
ated unintentional weight loss not otherwise explainable, or requirement of
intravenous inotropes to assist with diuresis or for management of hypo-
tension. Ultimately, earlier referral allows for trained specialists to evaluate
a patient for surgical HF therapies before becoming too unstable to tolerate
life-​sustaining treatments.
 Chapter 2 15

Myocarditis: diagnosis
and treatment
Introduction 16
Definition 16
Epidemiology 16
Aetiology 16
Clinical features (symptoms/​diagnosis) 18
Therapy 22
Therapies for specific forms of myocarditis 23
Outcomes 24
16

16 Chapter 2 Myocarditis: diagnosis and treatment

Introduction
Myocarditis is an inflammatory disease of the myocardium caused by a
number of infectious and non-​infectious problems and the subsequent in-
flammatory response to the initial insult. It can present either acutely or
subacutely and result in a spectrum of clinical manifestations ranging from
mild cardiac damage (normal ejection fraction (EF) and mild symptoms) to
fulminant myocarditis associated with either cardiogenic shock or sudden
death. In this chapter, the aetiologies, clinical presentation, diagnostic mo-
dalities, therapies, and prognosis will be reviewed.

Definition
Myocarditis is defined as inflammation of the heart muscle related to either
a cellular or humoral immune process and is established by either histo-
logical, immunological, or immunohistochemical criteria. It may either occur
in the setting of a systemic immune-​mediated disorder or in isolation to
just the heart. It should also be distinguished from dilated cardiomyopathy
(DCM) which is a clinical diagnosis with dilation and reduced contraction of
the left and/​or both ventricles not related to another cause (CAD, valvular
heart disease, inflammation).

Epidemiology
The true incidence of myocarditis is undefined due to many cases being rela-
tively asymptomatic. A study using the International Classification of Diseases
estimated a prevalence of about 22 cases per 100,000 patients annually.
In young competitive athletes, studies rank myocarditis as the third leading
cause of sudden death. Studies have also shown that between 1% and 5%
of patients testing positive for acute viral infections exhibit a form of myo-
carditis. A large study evaluating causes of cardiomyopathy in 1230 patients
performed endomyocardial biopsies and found that 9% had myocarditis.
Similarly, in the Myocarditis Treatment Trial, evidence of myocarditis was
found in 10% of the patients who underwent endomyocardial biopsy (214/​
2233 patients) for presumed myocarditis based on a history of HF and an
appropriate clinical syndrome. Based on the low prevalence in patients pre-
senting with NICM, one must remember that there are many other reasons
for cardiomyopathy and have an increased suspicion for myocarditis in those
with an acute presentation, a viral prodrome, and elevated biomarkers.

Aetiology
Although this chapter focuses on viral myocarditis, the causes of myocar-
ditis are extensive and include both infectious and non-​infectious agents as
outlined in Box 2.1. The most common cause of myocarditis are viral infec-
tions including adenovirus, cytomegalovirus (CMV), enterovirus including
coxsackie virus A and B, hepatitis C virus, human immunodeficiency virus,
human herpesvirus 6, influenza virus, and parvovirus B19. Although exact
mechanisms are unclear, based on animal data from coxsackie B virus
 Aetiology 17

studies, the currently accepted mechanism is that the virus enters the
cardiomyocyte via specific receptors on the cell surface. An immunological
response consisting of T lymphocytes and macrophages leads to the elim-
ination of the virus. In some patients, this immune response persists for
weeks to months after viral elimination resulting in DCM. Additionally, this
damage may result in the formation of autoantibodies to various cardiac
epitopes resulting in further damage to the myocardium. This response is
more likely to occur in patients with a genetic predisposition, leading to the
hypothesis that there might be a genetic relationship to the development
of myocarditis. The inability to eliminate the virus or turn off the immune
response results in chronic myocarditis.
The European Society of Cardiology Working Group on Myocardial
and Pericardial Disease has developed a model with three phases. In phase
1, direct microbial or toxin damage results in myocyte death, release of
chemokines and cytokines, and activation of the immune system resulting
in active myocarditis. Phase 2 is the elimination of the agent and either reso-
lution of the inflammation (healed myocarditis) or continued infection or
inflammation or both (chronic myocarditis). Phase 3 is split up into four
groups. Chronic microbial myocarditis occurs when the offending agent
is present with ongoing inflammation and destruction. Chronic microbial
and immune myocarditis occurs when both the microbial agent is present
and autoantibodies are present with ongoing inflammation and destruction.
Chronic autoreactive myocarditis occurs when there are autoantibodies but
no microbial agent. Finally, DCM occurs when there is no active inflamma-
tion but either the microbial agent or autoantibodies or both are present
with ongoing destruction.

Box 2.1 Causes of myocarditis

Infectious
• Viruses—​most common.
• Bacteria.
• Fungi.
• Parasites.
• Protozoa.
• Rickettsiae.
• Spirochetes.
• Helminths.
Non-​infectious
• Autoimmune.
• Sarcoidosis.
• Giant cells.
• Systemic disorders
• Hypersensitivity reactions.
• Radiation.
• Antibiotics.
• Psychiatric medications.
• Cardiotoxins.
• Checkpoint inhibitor drugs.
• Illicit drugs.
18

18 Chapter 2 Myocarditis: diagnosis and treatment

In addition to viral and other infectious agents, a similar process can occur
via the non-​infectious mechanisms including drugs, toxins, and many of the
rheumatological disorders. After the initial insult to the heart, the immune
system responds in a similar manner as described above, resulting in an
inflammatory state.
Giant cell myocarditis is an autoimmune process characterized by a mixed
inflammatory process with T lymphocytes and multinucleated giant cells
plus extensive necrosis. Patients tend to present with fulminant myocarditis
with rapidly progressive HF, ventricular arrhythmias, and heart block. It is a
particularly aggressive form of myocarditis with the median survival being
3 months without immunosuppressive therapy. Treatment with aggressive
immunosuppression with two or three different agents similar to early after
heart transplantation has been shown to improve survival with a 5-​year
transplant-​free survival rate of >50%. One of the primary reasons to per-
form a biopsy is looking for giant cell myocarditis since it is responsive to
immunosuppressive therapy.

Clinical features (symptoms/​diagnosis)

The clinical presentation of myocarditis is quite varied and can range from
mild dyspnoea and chest pain with a normal EF to cardiogenic shock. The
‘classic’ presentation is of a patient experiencing a viral illness which is often
fairly benign prior to developing acute HF symptoms including exertional
dyspnoea followed by oedema, orthopnoea, and paroxysmal nocturnal
dyspnoea. Many patients also have conduction disturbances or arrhyth-
mias resulting in complaints of tachycardia and palpitations. Others will
present with chest pain and an acute coronary syndrome-​type presenta-
tion. Unfortunately, some patients present in cardiogenic shock or after
sudden death.
Although classically myocarditis is thought of as a common reason for HF,
many patients will present with a normal EF. An Italian registry of patients
with myocarditis recently reported its findings on clinical presentation. Of
443 patients either under the age of 50 or under the age of 70 with a nega-
tive cardiac catheterization, the diagnosis of myocarditis was confirmed
by either positive endomyocardial biopsy or the combination of a positive
troponin plus peripheral oedema plus late gadolinium enhancement on car-
diac magnetic resonance imaging (MRI). It was found that only 26.6% of the
patients presented with either an EF of <50% or ventricular arrhythmias or
a low cardiac output syndrome. Additionally, there were differences in the
presentation between the two groups with the complicated group having
a higher incidence of dyspnoea (55.7% vs 6.2%), syncope (16.5% vs 2.5%),
and a history of autoimmune disorder (15.4% vs 4.2%). Patients with a
normal EF were more apt to have chest pain (59.1% vs 96.6%) and ST seg-
ment elevation with a normal catheterization as their presenting symptom.
Of note, there was no difference in the number of patients presenting with
a flu-​like syndrome between the two groups with 64.5% having fever, 80.5%
having flu prodromal symptoms, and 36.8% having sore throat.
The diagnosis of myocarditis begins with a high clinical suspicion. These
patients are often young and present with dyspnoea and including cardiac
problems in the differential diagnosis often doesn’t occur. Patients usually
 Clinical features (symptoms/diagnosis) 19

present with either signs and/​or symptoms of HF, chest pain, or arrhyth-
mias. About half of the patients will have a history of a recent upper re-
spiratory illness. The ECG will often demonstrate sinus tachycardia but
might also show signs of myocardial inflammation including repolarization
abnormalities and QRS prolongation. Heart block is often suggestive (but
not required) of either cardiac sarcoidosis or giant cell myocarditis. The
ECG might also only be suggestive of pericarditis with PR segment depres-
sion and non-​specific ST, T-​wave changes instead of elevation.
Laboratory findings will include an elevated B-​ type natriuretic pep-
tide (BNP) reflective of HF and usually elevated troponin I or T. Studies
have shown though that patients with biopsy-​proven myocarditis can have
normal troponin levels so the absence of a troponin elevation does not rule
out myocarditis. An elevated erythrocyte sedimentation rate (ESR) is also
frequently found. A complete blood count with differential should always
be performed to evaluate for peripheral eosinophilia. The European Society
of Cardiology guidelines suggest monitoring of serum troponin, ESR, and
C-​reactive protein. Routine viral serologies have not been proven to be
helpful and are no longer recommended as part of the evaluation.
The chest X-​ray (CXR) can be suggestive of HF with pulmonary vascular
congestion or an enlarged heart. However, the absence of these findings
does not rule out myocarditis and, in general, the CXR should be used to
rule out other aetiologies of dyspnoea or chest pain such as a pneumonia.
The echocardiogram is quite helpful in the diagnosis. Patients will often
present with diffuse hypokinesis and possibly ventricular dilation. An in-
crease in LV wall thickness can be a sign of myocardial oedema and might
correlate with reduced voltage on ECG. A surrounding pericardial effusion
can often be present. If the LV is dilated, that is usually a sign of less acute
presentation and tends to correlate with a reduced incidence of recovery
of LV function. Additionally, the echocardiogram should be used to rule out
other possible causes of HF including valvular heart disease and focal wall
motion abnormalities suggestive of CAD. A normal EF does not rule out
myocarditis as many patients will have a normal EF.
Cardiac MRI is now quite frequently used in the diagnosis of myocarditis.
A consensus conference resulted in the Lake Louise Criteria for the diag-
nosis of myocarditis by MRI which are outlined in Box 2.2.
For patients who present with acute cardiogenic shock or heart block or
arrhythmias, a cardiac MRI often cannot be performed due to safety issues
and one should consider an endomyocardial biopsy. The ACC issued a con-
sensus statement on the indications for biopsy and gave recommendations
for endomyocardial biopsy for the following patients:
• New-​onset HF of <2 weeks and haemodynamic compromise.
• New-​onset HF of 2 weeks to 3 months associated with a dilated LV and
either new ventricular arrhythmias or second-​or third-​degree heart
block or failure to respond to usual medical therapy.
• HF for >3 months with a dilated LV and either new ventricular
arrhythmias or second-​or third-​degree heart block or failure to
respond to usual medical therapy.
• HF and a suspected allergic reaction.
Not included in the guidelines but a new consideration is to perform
endomyocardial biopsy in patients who have recently received checkpoint
inhibitors and have new HF due to the increased incidence of myocarditis.
20

20 Chapter 2 Myocarditis: diagnosis and treatment

Box 2.2 Cardiac MRI criteria for the diagnosis

of myocarditis
Requires the setting of clinically suspected myocarditis. Need at least two
criteria for the diagnosis:
1. Regional or global increased signal intensity of T2-​weighted images
suggesting oedema.
2. Increased global early gadolinium enhancement ratio between
myocardium and skeletal muscle on gadolinium-​enhanced T1-​
weighted images suggestive of capillary leakage.
3. Late gadolinium enhancement defined as at least one focal lesion in
a non-​coronary distribution typically involving the myocardium or
subepicardium.
A repeat study is suggested if none of the above are present but symp-
toms are recent and there is a strong clinical suspicion of myocarditis or
only one of the criteria is present.
The presence of LV dysfunction or pericardial effusion is supportive
evidence for myocarditis.
Reproduced from Friedrich MG, Sechtem U, Schulz-​Menger J, et al; International Consensus
Group on Cardiovascular Magnetic Resonance in Myocarditis. Cardiovascular magnetic reson-
ance in myocarditis: A JACC White Paper. J Am Coll Cardiol. 2009 Apr 28;53(17):1475–​87.
doi: 10.1016/​j.jacc.2009.02.007 with permission from Elsevier.

To improve diagnostic accuracy, one should take at least six pieces of tissue.
In addition to the routine pathological evaluation, it is ideal to perform
immunohistochemistry and polymerase chain reaction (PCR) to look for a
viral genome. One should simultaneously study the peripheral blood to rule
out an acute systemic infection and possible contamination of the myocardial
samples. As outlined in the section on therapy (see p. 22), aggressive immuno-
suppression for patients with ‘virus-​negative’ myocarditis has been shown
to improve outcomes in small trials. The Dallas criteria were developed to
define myocarditis in biopsy specimens (Box 2.3). The definition of active
myocarditis required both an inflammatory infiltrate of the myocardium with
necrosis and/​or degeneration of adjacent myocytes not typical of damage as-
sociated with CAD (Fig. 2.1). Additionally, occasionally the cause of myocar-
ditis can be found for patients with Chagas disease, Lyme carditis, and CMV.
The diagnostic criteria for clinically suspected myocarditis as outlined by the
European Society of Cardiology are outlined in Box 2.4.
Finally, for patients at high suspicion of CAD, cardiac catheterization
should be performed to rule out CAD.

Box 2.3 Dallas criteria for the diagnosis of myocarditis

First biopsy
• Myocarditis with or without fibrosis.
• Borderline myocarditis.
• No myocarditis.
Subsequent biopsy
• Ongoing (persistent) myocarditis with or without fibrosis.
• Resolving (healing) myocarditis with or without fibrosis.
• Resolved (healed) myocarditis with or without fibrosis.
 Clinical features (symptoms/diagnosis) 21

Box 2.4 Diagnostic criteria for clinically

suspected myocarditis
Clinical presentation
• Chest pain.
• New-​onset or worsening dyspnoea (<3 months).
• Subacute/​chronic or worsening dyspnoea (>3 months).
• Palpitations or other arrhythmia symptoms.
• Unexplained cardiogenic shock.
Diagnostic criteria
• ECG: new second-​or third-​degree heart block, bundle branch block,
ST/​T wave changes, ventricular tachycardia or fibrillation, frequent
atrial or ventricular ectopy.
• Elevated troponin.
• Imaging abnormalities (echocardiography/​MRI/​angiography): LV
dysfunction, ± effusion, increased wall thickness, thrombi.
• MRI: tissue oedema of late gadolinium enhancement pattern.
Suspect if >1 clinical and >1 diagnostic criteria are present.
Reproduced from Caforio AL, Pankuweit S, Arbustini E, et al; European Society of Cardiology
Working Group on Myocardial and Pericardial Diseases. Current state of knowledge on aeti-
ology, diagnosis, management, and therapy of myocarditis: a position statement of the European
Society of Cardiology Working Group on Myocardial and Pericardial Diseases. Eur Heart J. 2013
Sep;34(33):2636–​48, 2648a–​2648d. doi: 10.1093/​eurheartj/​eht210 with permission from
Oxford University Press.

Fig. 2.1 Haematoxylin–​eosin staining of an endomyocardial biopsy sample showing

an inflammatory infiltrate with associated cellular necrosis consistent with acute
myocarditis. See plate section.
2

22 Chapter 2 Myocarditis: diagnosis and treatment

Therapy
The initial management of patients with myocarditis and HF should focus on
the usual therapy for patients with HF as outlined in Chapter 1. If patients
are haemodynamically unstable, one should use caution with beta blocker
therapy due to the acute, negative inotropic properties of those agents.
Other agents used to control tachycardia should be used carefully as the
increased heart rate may be compensatory for a low stroke volume. The
treatment of arrhythmias should be as per usual protocol. Because many of
these patients recover function, ICD therapy should be performed as well
as per the usual guidelines.
For patients who present in cardiogenic shock, the initial management
should focus on haemodynamic support of the patient with inotropes and
vasopressors. Additionally, early transfer to a centre equipped with the
ability to perform specialized diagnostic studies (MRI, endomyocardial bi-
opsy) and mechanical support should be consider as patients can quickly
decompensate. There are no randomized studies on the acute management
of shock with myocarditis. In patients with shock after an acute myocar-
dial infarction, norepinephrine has a lower incidence of arrhythmias and
improved survival compared to dopamine. If an endomyocardial biopsy
is performed, consideration should be given to also perform a right heart
catheterization and leave it in place to help with management, although the
use of right heart catheterization has not been associated with improved
outcomes in this group. As outlined later in this chapter, some patients will
be found to have a specific diagnosis on biopsy and in addition to supportive
care, directed therapy towards that diagnosis should be performed. Finally,
since complete recovery from myocarditis occurs so frequently, one should
use aggressive temporary support including percutaneous assist devices or
extracorporeal membrane oxygenation (ECMO) acutely in patients who
are unresponsive to vasopressor therapy.
Left ventricular assist devices (LVADs) should be considered in pa-
tients who require temporary mechanical support and are not improving.
In deciding whether or not to place permanent mechanical support, one
should consider signs of irreversibility as outlined in Box 2.5. Even in that
setting, over a period of a few months many patients will recover and can
potentially have their device removed. In a study of 24 patients with a
ventricular assist device (VAD) placed for fulminant myocarditis, the best
predictor of recovery was an onset of symptoms within 1 week of pres-
entation and female sex. A report from the Intervention in Myocarditis and
Acute Cardiomyopathy 2 (IMAC2) study found that patients who did not
recover after the VAD had a larger LV (left ventricular end-​diastolic diam-
eter (LVEDD) 7.0 cm vs 5.3 cm), less inflammation (0% vs 75%), and more
fibrosis (100% vs 25%).
Finally, in patients who don’t recover after placement of their VAD, con-
sideration should be given to cardiac transplantation if appropriate. Early
case reports of patients transplanted for lymphocytic myocarditis showed
an increased rate of rejection and reduced survival. A larger study from
the United Network for Organ Sharing (UNOS) database showed no dif-
ferences in outcomes between patients with myocarditis versus other aeti-
ologies. Finally, Columbia University reported the results of 32 patients
transplanted with lymphocytic myocarditis and found similar survival and
 Therapies for specific forms of myocarditis 23

Box 2.5 Factors associated with reduced incidence

of recovery in acute myocarditis
• Lower EF (<40%).
• Ventricular arrhythmias.
• Low cardiac output requiring inotropes or MCS.
• Cardiac-​specific autoantibodies.
• Syncope.
• Biventricular dysfunction.
• Persistence of viral genome.
• Presence of fibrosis.
• LVEDD >7.0 cm.

rejection rates to patients transplanted with idiopathic cardiomyopathy. At

this time, one might consider heightened surveillance for rejection in these
patients, but the presence of myocarditis before transplantation should not
be a contraindication for transplantation.

Therapies for specific forms

of myocarditis
Lymphocytic myocarditis is defined as the presence of a cellular infiltrate
and LV dysfunction. A number of clinical trials have been performed to
evaluate the benefits of immunosuppressive therapy in patients with biopsy-​
proven myocarditis. The first large trial was the Myocarditis Treatment Trial
which evaluated the use of ciclosporin and steroids versus placebo. Over
2000 patients were screened to find the 111 with active myocarditis. There
was no significant different in the endpoint of improvement in EF between
the two arms. Based on data showing improvement in children with myo-
carditis, intravenous immunoglobulin (IVIG) has been studied as well. In
two randomized studies, one showed no improvement in EF and the other,
which included the use of plasmapheresis followed by IVIG, showed a small
change in EF of 6% and improved NYHA functional class in 25 patients.
More recently, a number of studies have been performed in patients who
are virus negative but have inflammation on biopsy. The largest trial was a
randomized, placebo-​controlled trial of 85 patients with half receiving ster-
oids and azathioprine versus placebo. Of those receiving immunosuppres-
sive therapy, 88% had an improvement in EF of at least 10% versus none
in the placebo arm. Other retrospective trials have shown similar results.
At this time, if treating a virus-​negative patient with active inflammation,
consideration of immunosuppressive therapy is reasonable but a definitive
approach requires more prospective data.
Randomized trials of antiviral therapy have not been performed in pa-
tients with virus-​positive myocarditis. Small studies have shown improve-
ment in EF with the use of interferon beta for patients with enterovirus or
adenovirus genomes. A follow-​up randomized trial of 143 patients included
those with parvovirus B19 and showed improvement in quality of life and
clearance of virus, but no change in EF when compared to placebo. It should
be noted that these patients had the presence of virus on biopsy, but not
acute, active myocarditis.
24

24 Chapter 2 Myocarditis: diagnosis and treatment

Patients with giant cell myocarditis have been shown to be responsive

to immunosuppressive therapy. In a trial of 11 patients with biopsy-​proven
myocarditis, all received ciclosporin and prednisone and nine of the 11 re-
ceived muromonab-​CD3. Two patients underwent early heart transplant-
ation and one patient died of respiratory complications, but the others
survived past a year which is much improved from historical data. This is in
contrast to an earlier study of 63 patients with giant cell myocarditis who
had a rate of death or cardiac transplant of 89% with a median survival
of 5.5 months. In patients who do progress to require an assist device,
one should strongly consider the placement of biventricular support since
these patients will often progress to biventricular standstill. Finally, there
are reports of recurrence of giant cell myocarditis after transplantation that
has usually been shown to be responsive to enhanced immunosuppressive
therapy.
Eosinophilic myocarditis is characterized by an eosinophilic infiltrate
that may be caused by hypersensitivity to a drug or toxin, parasite, or
hypereosinophilic syndrome. It usually presents as an acute myocarditis as-
sociated with a peripheral eosinophilia and has been shown to be respon-
sive to aggressive steroid therapy. Additionally, one should ensure that the
offending agent is discontinued or removed.

Outcomes
The vast majority of patients with myocarditis have normal LV function and
seem to do well in comparison to patients who are more symptomatic and
have biopsy-​proven myocarditis. The Myocarditis Treatment Trial showed
that all patients improved their EF from a baseline of 25% to 34% by 28
weeks. Twenty per cent of the patients were dead at 1 year and 56% were
dead at 4.3 years. Patients with less severe disease tended to do better.
Slightly better outcomes were reported by a group from Germany in pa-
tients with myocarditis and positive virus with 19.2% mortality at 4.7 years.
Similar data were reported in an Italian registry of 174 patients with myo-
carditis with a rate of 13% at 2 years of death or transplantation. They
found that patients with signs or symptoms of persistent left or right HF had
the largest association with need for transplantation or death.
Patients with fulminant myocarditis defined as LV dysfunction and low
cardiac output requiring inotropes or MCS have had varying outcomes de-
pending on the study. A retrospective analysis of 147 patients with biopsy-​
proven myocarditis reported that the 15 patients with fulminant myocarditis
had a 93% transplant-​free survival rate at 11 years after biopsy compared
to only 45% of those with acute myocarditis. A recent report of 187 con-
secutive patients with clinical myocarditis between 2001 and 2016 provides
more recent data on outcomes of patients with advanced, fulminant myo-
carditis with symptoms for <1 month. One hundred and thirty of the pa-
tients presented with a viral prodrome but endomyocardial biopsies looking
for positive virus were not performed. The authors defined fulminant myo-
carditis as patients with LV dysfunction and low cardiac output requiring
inotropes or MCS. Thirty five of the 55 patients with fulminant myocarditis
required MCS which was usually a balloon pump. Mortality in the fulminant
 Outcomes 25

myocarditis group was 18.2% (10/​55), four had an early heart transplant,
and one received an LVAD. There were no events in the non-​fulminant
group of patients. It should be noted that most of these patients received
immunosuppressive therapy.
Myocarditis is a difficult disease process because of varying clinical pres-
entations, aetiologies, and outcomes. One should have a heightened suspi-
cion for diagnosis, especially in children and young adults who rarely present
with cardiac issues. Aggressive diagnosis and therapy including referral to
institutions equipped with therapeutic modalities is required.
26
 Chapter 3 27

Cardiac amyloidosis
and other restrictive
myopathies: diagnosis
and treatment
Introduction 28
Clinical presentation and diagnosis 28
Amyloid cardiomyopathy 31
Mechanical support 36
Transplantation 37
28

28 Chapter 3 Restrictive cardiomyopathy

Introduction
Restrictive cardiomyopathies (RCMs) are a heterogeneous group of dis-
eases characterized haemodynamically by impaired ventricular filling and ab-
normal diastolic function. These can be characterized pathophysiologically
as infiltrative diseases, metabolic storage diseases (less commonly diagnosed
in the adult population), sarcomeric protein disorders, endomyocardial dis-
eases, or idiopathic. A simplified list of aetiologies is presented in Table 3.1.

Table 3.1 Aetiology of restrictive cardiomyopathy by pathophysiology

Pathophysiological classification of Specific diseases
RCM
Infiltrative Amyloidosis
Sarcoidosis
Metabolic storage disease Fabry disease
Gaucher disease
Haemochromatosis
Glycogen storage disease
Niemann–​Pick disease
Sarcomeric protein disorders Spectrum of hypertrophic
cardiomyopathy
Endomyocardial disease Endomyocardial fibrosis
Radiation induced

Each of these disorders leads to alteration of normal myocardial or endo-

cardial tissue, resulting in progressive ventricular dysfunction in diastole
where efficient cardiac relaxation and filling are critical. Impaired ventricular
filling leads to elevated intracardiac pressures with resultant symptoms of
congestive HF. Ultimately, this progresses to low stroke volume with poor
cardiac output in the advanced stages. Each of these diseases has specific
nuance to its diagnosis and treatment. This chapter will focus on presenta-
tion, management, and indications for referral to advanced therapies for
RCMs. Due to its increasing recognition, amyloid cardiomyopathy will be
discussed in greater detail.

Clinical presentation and diagnosis

RCM should always be suspected when a patient presents with clinical signs
of congestion in the setting of preserved EF. Many types of RCM affect
both ventricles, thus patients can often manifest right HF symptoms (ascites,
anasarca, and lower extremity oedema) in addition to left-​sided HF symp-
toms (dyspnoea, fatigue, exercise intolerance, and pulmonary oedema).
A mainstay of diagnosis is echocardiography. RCM can present with in-
creased wall thickness in infiltrative disorders (amyloidosis) or normal wall
thickness in non-​infiltrative disorders (radiation, endomyocardial fibrosis).
Chronic pressure overload often leads to impressive biatrial enlargement
 Clinical presentation and diagnosis 29

which can predispose to atrial tachyarrhythmias and affiliated embolic com-

plications. Systolic function is often preserved until the later stages of dis-
ease, and echocardiographic signs of diastolic dysfunction are a sine qua
non. These include elevated early diastolic filling velocity (E wave) with
decreased late diastolic filling velocity (A wave), often with an E/​A ratio
of >2; and decreased tissue Doppler velocity of the lateral mitral annulus
(e') with an elevated E/​e', often >14 (Fig. 3.1). Chronic left-​sided pressure
overload eventually leads to postcapillary PH, thus many patients develop
significant tricuspid regurgitation (TR) in later disease stages as a result of
right ventricular (RV) remodelling. Invasive haemodynamics can be used to
support non-​invasive diagnostic studies. Findings on right heart catheter-
ization include elevated right-​and left-​sided filling pressures, PH, and the
classic ‘square-​root sign’ indicating early and rapid ventricular filling with
subsequent plateau phase.

(a) (b)

(c)

Fig. 3.1 Still image of infiltrative RCM highlighting normal LV size, thickened
ventricular walls, enlarged RV, and biatrial enlargement indicative of late-​stage disease
(a). Figure on right reveals low tissue Doppler velocities at medial (b) and lateral
(c) mitral annulus. See plate section.

Echocardiography and haemodynamics can also help distinguish be-

tween RCM and constrictive pericardial disease, which can often have a
similar clinical presentation. As mentioned, RCM presents with low mitral
annular tissue Doppler velocities (e') whereas constriction presents with
normal e' and ‘annulus reversus’ (medial > lateral e'). An invasive haemo-
dynamic study can further help this distinction; RCM is characterized by
ventricular concordance during respiration, elevated pulmonary pressures,
and LV > RV diastolic pressures. Constriction demonstrates ventricular
discordance during respiration, equalization of ventricular diastolic pres-
sures, and often absent PH (Table 3.2). If diagnosis is still uncertain, cardiac
computed tomography (CT) and MRI can enrich diagnostic yield by pro-
viding anatomic features of the pericardium, including thickness, calcifica-
tion, or inflammation.
30

30 Chapter 3 Restrictive cardiomyopathy

Table 3.2 Constellation of findings associated with restrictive cardiomyopathy

Evaluation Findings
History and physical • Right side: lower extremity oedema, elevated JVP,
Kussmaul’s sign, anasarca
• Left side: dyspnoea, fatigue, pulmonary oedema
Echocardiogram • i E/​e' (>14)
• i E/​A (>2)
• d e' velocity (<10 cm/​s)
• Lateral > medial e' velocity
Haemodynamics • Ventricular concordance with respiration
• ‘Square-​root sign’
• i pulmonary artery pressures
• LVEDP > RVEDP
A, late atrial filling velocity; E, early transmitral filling velocity; e', tissue Doppler velocity at mitral
annulus; JVP, jugular venous pressure; LVEDP, left ventricular end-​diastolic pressure; RVEDP,
right ventricular end-​diastolic pressure.

Once RCM is suspected, diagnosis of the specific aetiology often relies

on a constellation of findings including family history, imaging findings, and
endomyocardial biopsy. Histopathology can diagnose amyloidosis, sarcoid-
osis, endomyocardial fibrosis, Fabry disease, Danon disease, and hyper-
trophic cardiomyopathy. A retrospective study from the Johns Hopkins
Hospital revealed that endomyocardial biopsy conducted for unexplained
RCM was diagnostic in 29% of cases, changed patient management in 25%
of cases, and most commonly revealed amyloidosis. Prognosis often de-
pends on the specific condition that is diagnosed, but is generally considered
to be poor if aetiology is felt to be idiopathic.
Management of RCM revolves around treatment of the underlying con-
dition (if such treatment exists) in addition to management of HF. Targeted
therapies are available for specific cardiomyopathies. Fabry disease can be
treated with enzyme replacement therapy with agalsidase beta; haemo-
chromatosis can be treated with iron chelation therapy; endomyocardial
fibrosis can sometimes be surgically managed with endocardiectomy; and
sarcoidosis can be treated with immunosuppression. Treatment for light-​
chain and transthyretin (TTR) cardiac amyloidosis will be discussed further
in the next section (see Amyloid cardiomyopathy, p. 31).
Unlike dilated cardiomyopathies with depressed ventricular function,
there are no specific guideline-​directed medical therapies for RCM. In fact,
as the disease progresses, many therapies used in HF with reduced EF can
actually be detrimental. In advanced RCM, stroke volume becomes mark-
edly reduced, and cardiac output is dependent on heart rate, limiting the
use of beta blockade. Hypotension and renal dysfunction often limit the
use of renin-angiotensin inhibitors. Thus, these therapies are frequently not
tolerated and have resulted in no mortality benefit in RCM. The mainstay of
therapy revolves around management of the patient’s volume status (which
can be tenuous due to the steep LV diastolic pressure–​volume relation-
ship in RCM) and control of atrial arrhythmias (with concomitant thrombo-
embolism risk) which are frequent in this population. There may, however,
 Amyloid cardiomyopathy 31

be a role for mineralocorticoid antagonism to augment diuresis and poten-

tially improve outcomes. Referral for durable mechanical circulatory sup-
port (MCS) or heart transplantation is indicated when patients have severe
functional impairment (NYHA class III or greater HF). Pacemakers may be
needed for progressive conduction disease, and defibrillators have varying
indications dependent on specific RCM and prognosis at the time of disease
identification; indications for cardiac implantable electronic devices are dis-
cussed in detail elsewhere.

Amyloid cardiomyopathy
An increasingly recognized aetiology of RCM is amyloid cardiomyopathy, a
condition characterized by extracellular deposition of misfolded proteins.
While there are many known amyloidogenic proteins, two primarily affect
the heart: immunoglobulin light-​chains or misfolded TTR. Light-​chain amyl-
oidosis (AL amyloid) is a result of abnormal production of immunoglobulin
light chains from a plasma cell dyscrasia.
TTR is a molecule produced primarily by the liver to carry retinol-​binding
protein and thyroxine. It exists normally as a stable tetramer; disassoci-
ation of the tetramer into monomeric intermediates leads to formation of
amyloid fibrils which can subsequently deposit in tissue. Instability of TTR
is either the result of pathogenic mutations inherited in an autosomal dom-
inant fashion with variable penetrance (resulting in hereditary transthyretin
(hTTR) amyloidosis) or age-​related deposition of the wild-​type TTR pro-
tein (wtTTR or senile TTR).
Interstitial deposition of amyloid fibrils results in ventricular wall thick-
ening, decreased ventricular compliance, diastolic dysfunction, and
subsequent RCM. Microvascular deposition can lead to coronary flow ab-
normalities and micro-​ischaemia to the conduction system. Deposition in
atria can lead to decreased compliance, atrial pressure overload, and can
contribute to atrial arrhythmias. hTTR can also present with polyneurop-
athy with symptoms including tendinopathy, carpal tunnel syndrome, and
autonomic dysfunction. The degree of neurological and cardiac involve-
ment is mutation dependent. AL amyloidosis can result in protein depos-
ition anywhere in the body, including the gastrointestinal (GI) tract, kidneys,
and skin.
Prevalence
Recent studies have demonstrated that amyloid cardiomyopathy, in par-
ticular transthyretin amyloid cardiomyopathy (ATTR-​ CM), is far more
prevalent than previously recognized. In a Finnish autopsy study of 256
patients aged >85 years, nearly 25% were found to have myocardial TTR
amyloid deposition via immunohistochemistry, although the clinical signifi-
cance was undefined. In patients with heart failure with preserved ejection
fraction (HFpEF) and left ventricular hypertrophy (LVH), 13% above the age
of 60 years were found to have ATTR-​CM; this number increased to 30%
when the age of 75 years was used as a cut-​off. Finally, in patients under-
going TAVR at one centre, 17% were found to have ATTR-​CM. Mutations
leading to hTTR have geographic variation; the Val122Ile mutation is pre-
sent in 3–​4% of African Americans in the USA, and the Thr60Ala mutation
32

32 Chapter 3 Restrictive cardiomyopathy

is present in 1% of the Northern Ireland population. Overall, there must be

a high clinical suspicion for this disorder when patients present with RCM.
Diagnosis
Diagnosis of amyloid cardiomyopathy involves integration of a constel-
lation of clinical symptoms, imaging characteristics, and laboratory find-
ings. Histopathology can also be used for diagnostic confirmation. Initial,
heightened clinical sensitivity to the various manifestations of amyloid
cardiomyopathy as mentioned previously (autonomic dysfunction, carpal
tunnel syndrome, atrial arrhythmias, conduction disease) is paramount to
diagnosis. ECG may reveal low voltages despite increased ventricular wall
thickness, though this is more common in AL amyloidosis and has poor
sensitivity for ATTR-​CM. Echocardiography may reveal typical features
of RCM highlighted in Table 3.2 (see p. 30). In addition, speckle tracking
echocardiography can be used to define longitudinal myocardial strain.
Characteristic findings in amyloid cardiomyopathy include impaired global
longitudinal strain despite normal EF, but with relative preservation of re-
gional longitudinal strain at the apex, translating to a ‘cherry-​on-​top’ or
‘bullseye’ pattern during strain imaging. Finally, MRI will reveal diffuse late
gadolinium enhancement in a subendocardial pattern and is specific for
diagnosis of cardiac amyloidosis. Imaging findings representative of cardiac
amyloidosis are highlighted in Fig. 3.2.

(a) (b)

(c)

Fig. 3.2 Characteristic imaging findings of cardiac amyloidosis. (a) Typical findings of
longitudinal strain imaging with characteristic ‘bullseye’ pattern. (b) MRI with diffuse
subendocardial late gadolinium enhancement. (c) Pyrophosphate (PYP) scan with
heart/​contralateral (H/​Cl) ratio >1.5. See plate section.

Definitive diagnosis of a subtype of amyloidosis requires further sero-

logical testing, nuclear imaging and, in many cases, biopsy. Our institution’s
algorithm is shown in Fig. 3.3. First, serological testing to exclude AL
amyloid must be performed via measurement of serum free light chains,
serum protein electrophoresis, and urine protein electrophoresis with
immunofixation. If all three of these are negative, this effectively rules out
the presence of AL amyloidosis. Concurrently, nuclear imaging should be
performed. Two molecules, pyrophosphate (PYP) and 3,3-​diphosphono-​
1,2-​propanodiacarboxylic acid (DPD), previously used as bone tracers,
were found to have avidity to TTR deposits in the myocardium. These
 Echocardiography
Clinical features
ECG - Increased wall thickness (often
- Heart failure Cardiac MRI
- Low voltage (despite imaging LVH) biventricular)
- Autonomic dysfunction - Diffuse LGE
- Infarct pattern w/o CAD - Diastolic dysfunction (grade 2/3)
- Bilateral carpal tunnel syndrome - Increased ECV and T1
- Conduction disease - Impaired global longitudinal strain
- Macroglossia - Inability to null myocardium
- Atrial fibrillation (GLS)
- Periorbital ecchymoses
- Apical sparing on GLS

Clinical Suspicion of Cardiac Amyloidosis

Additional labs with prognostic utility:
NTproBNP, Troponin (AL, ATTR)

AL Amyloidosis Evaluation
- Serum IFE ATTR Amyloidosis Evaluation
- Urine IFE - Tc99m-PYP Scan
- Serum free light chains

AL(+) ATTR(+)
ATTR(–) AL(+) AL(–) AL(–)
ATTR(+) or ATTR(–)
Bone marrow biopsy, fat aspirate/fat pad biopsy or EMB Indeterminate TTR Gene Sequencing

Unlikely cardiac amyloid

Amyloid Deposits (+) Or
High rare amyloid (ApoA , etc.)
Amyloid Typing (Mass Spec) Suspicion?
No Cardiac Amyloid
Bone marrow biopsy, fat (–) Wild type TTR Hereditary TTR
AL aspirate/fat pad biopsy
(+) Mass Spec (+) Mass Spec (ATTRwt) (ATTRm)
or EMB

Referral genetic
counseling

Fig. 3.3 Duke University approach to diagnosis of cardiac amyloidosis. CAD, coronary artery disease; ECV, extracellular volume; EMB, endomyocardial
biopsy; IFE, immunofixation; LVH, left ventricular hypertrophy; NTproBNP, N-​terminal pro-​B-​type natriuretic peptide; Tc99m, technetium 99m.
Amyloid cardiomyopathy

Courtesy of Dr Michel Khouri.

33
34

34 Chapter 3 Restrictive cardiomyopathy

are tagged to technetium allowing quantification of TTR deposits via nu-

clear scintigraphy. Absolute counts and mean counts are measured over
the heart and on the contralateral (control) side of the chest. A heart/​
contralateral ratio of mean counts ≥1.5 was found to have a 97% sensi-
tivity and 100% specificity for diagnosing TTR amyloid. Once diagnosed as
ATTR-​CM, further genotyping is required to distinguish wtTTR from hTTR.
If pathways for both AL and ATTR-​CM are negative, cardiac amyloidosis
is unlikely, or a rare type of cardiac amyloidosis is present. If clinical suspi-
cion for cardiac amyloidosis remains high, tissue biopsy is the next step. If
both are positive, then further tissue must be obtained for histopathology.
We favour endomyocardial biopsy, which should be subsequently typed
via mass spectrometry, to definitively diagnose which type of amyloid is
affecting the heart. Fat pad aspirate has low yield for TTR (15–​40%) but has
higher sensitivity for AL (80%). Importantly, it should be noted that many
patients with ATTR-​CM may have concomitant monoclonal gammopathy
of undetermined significance (MGUS), identified in >20% in one series,
which can be a common cause of diagnostic confusion. Consultation of
a haematologist/​oncologist is recommended, as these patients may also
require bone marrow biopsy to assess for the presence of plasma cell
dyscrasia.
Prognosis
Prognosis for this condition is variable. AL amyloidosis typically presents
more acutely, progresses rapidly, and thus has a prognosis that is worse
than ATTR-​CM and is dependent on degree of cardiac involvement. The
Mayo Clinic has developed a staging system utilizing cardiac biomarkers: N-​
terminal pro-​B type natriuretic peptide (NT-​proBNP) ≥1800 ng/​dL, cardiac
troponin T ≥0.025 mcg/​L, and difference in serum light-​chain concentra-
tions ≥18 mg/​dL. Prognosis worsens as more of these biomarkers are ab-
normal. Median survival if none of the biomarkers are above threshold is
55 months, but if all are elevated, median survival is only 5 months. This
improves to 22 months in patients who are able to undergo stem cell
transplantation.
Prognosis for ATTR-​CM is dependent on a variety of factors, including
age, genotype (if hTTR), NYHA functional class, serum biomarkers, and
imaging findings, but overall median survival is in the order of 3–​5 years.
The Mayo Clinic derived a similar biomarker-​based prognostic scoring
system using NTproBNP (>3000 pg/​mL) and cardiac troponin T (>0.05
ng/​mL) and found that 4-​year overall survival if no biomarkers were above
threshold (stage I) was 57%, one biomarker abnormal (stage II) was 42%,
and both biomarkers abnormal (stage III) was 18%. Notably, patients with
stage III had a median survival of only 20 months. NYHA functional class
also plays a role in prognostication (higher NYHA class, worse median sur-
vival) and identifies patients who better respond to treatment (discussed
in following treatment section). Finally, for hTTR amyloid, the Val122Ile
mutation present in many African Americans in the USA confers a worse
prognosis than other genotypes. Fig. 3.4 highlights various prognostic fea-
tures that have been identified in the literature. Prognosis is particularly im-
portant because patients early in the course of disease may derive benefit
from TTR-​targeted therapeutics, whereas those later in the course of dis-
ease may require transplantation or only be eligible for a more palliative
approach to disease management.
 Amyloid cardiomyopathy 35

New Diagnosis of TTR Cardiac Amyloidosis

Multimodality Evaluation

Poor Prognostic Features

Laboratory Genetic (if

Clinical and Electro- Imaging TTRm cardiac
cardiography amyloidosis)
Male NT-proBNP Echocardio- Mutations:
Older age >3000 pg/mL graphy: - Val122lIe
Lower blood cTnT >0.04 - MCF <25% - Thr60Ala
pressure ng/mL - GLS >–14.5 - Glu89Gln
(orthostatic 99mTc-PYP
GFR <45 mL/
syncope) min/1.73 m2 scan:
NYHA Sokolow - H:Cl ≥1.6
class III−IV index <15 CMR:
- Transmural
LGE
- Increased
ECV

Few features present Many features present

Better Prognosis Worse Prognosis
- Prompt referral to - Evaluate to see if
specialist for novel appropriate for heart
therapeutics +/– liver transplantation
- Enrollment in clinical - Consider palliative care
trials referral

Fig. 3.4 Prognostic variables in ATTR-​CM with guidance for future therapy. CMR,
cardiac magnetic resonance; cTnT, cardiac troponin T; ECV, extracellular volume;
GFR, glomerular filtration rate; GLS, global longitudinal strain; H/​Cl, heart/​
contralateral; LGE, late gadolinium enhancement; MCF, myocardial contraction
fraction; NTproBNP, N-​terminal pro B-​type natriuretic peptide; NYHA, New York
Heart Association; TTR, transthyretin; TTRm, mutant TTR.

Treatment
For both AL cardiomyopathy and ATTR-​ CM, therapy is focused on
(1) management of HF, treated similarly as other RCM, and (2) treatment
of the underlying condition. For AL cardiomyopathy, this requires chemo-
therapy directed at the underlying plasma cell dyscrasia and necessitates
consultation of a haematologist/​oncologist. Regimens are generally com-
prised of an alkylating agent (cyclophosphamide), proteasome inhibitor
(bortezomib), and dexamethasone, along with novel CD38 inhibitors
(daratumamab).
36

36 Chapter 3 Restrictive cardiomyopathy

While previously considered untreatable, there are now therapeutic op-

tions for patients with ATTR-​CM. Patisiran, a small-​interfering RNA, and
inotersen, an antisense oligodeoxynucleotide, both function to prevent
translation of TTR mRNA into TTR protein and have been shown to be ef-
ficacious for treatment of familial amyloid polyneuropathy. Although not the
focus of the initial trials, both drugs have demonstrated efficacy in stabilizing
or improving surrogate cardiac endpoints such as LV wall thickness, NT-​
proBNP, and 6-​minute walk distance. Recently, tafamidis, a molecule which
causes stabilization of the TTR tetramer and prevents monomerization,
was approved for use for both hTTR and wtTTR cardiomyopathy based
on reductions in all-​cause mortality and cardiovascular-​related hospitaliza-
tions in the Safety and Efficacy of Tafamidis in Patients with Transthyretin
Cardiomyopathy (ATTR-​ACT) trial. Another molecule, AG-​10, mimics a
tetrameric stabilizing mutation, and has preliminary data demonstrating ef-
ficacy. A clinical trial for AG-​10 is currently underway. Of note, tetrameric
stabilizers do not remove previously deposited amyloid protein, and thus
work best at early stages of disease. Later disease stages require consider-
ation of mechanical support and transplantation.

Mechanical support
As described, RCM leads to progressive elevation in ventricular filling pres-
sures. Most MCS strategies are quite effective at reducing intracardiac
pressures through direct unloading and theoretically should be beneficial in
RCM. However, direct ventricular unloading in these patients can be quite
challenging owing to small ventricular cavity size. Augmentation of unloading
through afterload reduction also cannot be easily achieved given fixed
stroke volume, making strategies like intra-​aortic balloon counterpulsation
(IABP) limited in their capacity to reduce filling pressures. However, IABP
may be used in specific cases to bridge patients to durable support or heart
transplantation, and through augmented coronary perfusion, can limit the
micro-​ischaemia that is a component of many of the infiltrative RCMs.
Surgical implantation of durable MCS faces a similar challenge. This is
particularly true for isolated left-​sided circulatory support devices. Small LV
size can cause mechanical obstruction of a LVAD inflow cannula. Abutment
of the inflow cannula to the myocardium can result in obstruction, risk
suction events, and lead to dangerous ventricular arrhythmias. Restrictive
haemodynamics often lead to chronic RV pressure overload and resultant
RV failure, a major source of morbidity and mortality post LVAD. In patients
with DCM and biventricular dysfunction, post-​implantation RV failure can
occur in 20–​30% of patients. These patients require longer inotropic sup-
port, longer intensive care unit (ICU) stays, and longer hospitalizations. In
many patients with DCM, RV performance eventually improves such that
they can be discharged from the hospital without significant HF symptoms.
However, development of late or recurrent right HF results in a higher 1-​
and 2-​year mortality. In RCM treated with LVAD, this risk of RV dysfunction
is further increased due to the underlying structural and haemodynamic
changes of the RV which are less likely to respond to LV unloading.
The Mayo Clinic has reported their experience of LVAD implantation
in 28 patients with RCM. The most common indications for implant were
 Transplantation 37

amyloidosis, hypertrophic cardiomyopathy, and sarcoidosis, and most pa-

tients had Interagency Registry for Mechanically Assisted Circulatory
Support (INTERMACS) profiles I and II, indicating true end-​stage RCM. As
expected, the vast majority had evidence of RV dysfunction and elevated
right-​sided pressures prior to implantation, and 54% of patients underwent
concurrent tricuspid valve intervention. A LVAD was implanted as a bridge
to transplantation (BTT) in 61% of patients and destination therapy (DT) in
the rest.
Thirty-​nine per cent of patients experienced postoperative RV failure re-
quiring inotropic support, 25% experienced postoperative renal failure, and
in-​hospital mortality was 14%. Overall 1-​year survival in the entire cohort
was 64%, far lower than the 85% 1-​year event-​free survival in patients with
DCM treated with newest generation centrifugal pumps. The authors also
compared LVAD outcomes by aetiology of RCM and found no difference
in mean survival time between amyloid and non-​amyloid aetiologies. Finally,
the authors found that increased LV diameter in systole and diastole were
both significantly associated with greater survival. An LVEDD of ≤46 mm
was significantly associated with mortality.
In registry analysis from 2008 to 2014, only 1% of all LVAD implants were
for RCM and another 0.9% were for hypertrophic cardiomyopathy. The most
common aetiologies of RCM were amyloidosis, sarcoidosis, and idiopathic
disease. Interestingly, those who received LVAD in this cohort had some fea-
tures of ‘burned-​out’ disease and shared a phenotype with DCM: only ap-
proximately 5% had LVEF ≥40% and over half had an LVEDD ≥6.0 cm. In the
context of this mixed phenotype, the authors saw no significant difference in
1-​year survival between RCM (74%) and DCM (81%). However, they also
identified small LVEDD (50 mm in their study) as being significantly associated
with higher overall mortality (1-​year survival estimated at only 30%).
Our group infrequently uses MCS as a long-​term solution for patients
with end-​stage restrictive disease. However, the data previously described
indicate that, in the appropriately selected patient, it may improve short-​
term survival until transplantation can be achieved. Further, there are
also data indicating that MCS may reduce reactive PH, which can be a
contraindication to transplantation. We recommend using LVAD in RCM
only when ventricular geometry is favourable (LVEDD ≥50 mm) and in
patients without significant clinical RV failure. These suggestions are in
keeping with those issued by the International Society for Heart and Lung
Transplant (ISHLT), which state that MCS with LVAD cannot be routinely
recommended, but can be considered in highly select cases at specialized
centres (IIb recommendation). Alternative strategies of MCS in RCM in-
clude biventricular mechanical support (total artificial heart) or ECMO.
Specifically for cardiac amyloidosis, an approach providing biventricular
support via total artificial hearts has demonstrated reasonable long-​term
outcomes in a small cohort of patients.

Transplantation
Referral for orthotopic heart transplantation (OHT) is considered a
class I indication in RCM patients with NYHA class III and IV symptoms,
and is often the only therapy that can improve prognosis in these patients.
38

38 Chapter 3 Restrictive cardiomyopathy

Patients with RCM often develop reactive PH in response to chronically

elevated left-​sided pressures, resulting in a phenotype of combined pre-​and
postcapillary PH. Elevated pulmonary pressures often preclude these pa-
tients from transplantation listing; thus, timely referral is critical.
The number of OHTs performed for RCM is increasing over time, from
1.4% in the period from 1987 to 2010 to 3.4% in the period from 2009 to
2016. In a cohort evaluated from 2000 to 2013, approximately 7.4% of
RCM patients are bridged to transplantation using LVAD support. At our
centre, we often list these patients with inotropic support with or without
haemodynamic monitoring. As mentioned, MCS in these patients can be
technically challenging, and IABP is less helpful. Unfortunately, per the re-
vised UNOS allocation system, these patients often get relegated to status 3
or 4, despite the severity of their illness. Indeed, data indicate that patients
with RCM have a higher risk of death while awaiting OHT as compared to
other forms of cardiomyopathy, though these data were obtained prior to
changes to the UNOS classification and likely reflect the overall poor prog-
nosis of this group of diseases. Outcomes after OHT for RCM appear to be
similar to those for the combined group of ‘non-​RCM’ patients but worse
when compared to DCM directly.
Given the various aetiologies of RCM, there are special treatment consid-
erations that need to be recognized prior to transplantation in this group of
patients. For example, patients with underlying sarcoidosis should be concomi-
tantly treated with disease-​specific immunosuppression; those with haemo-
chromatosis should continue to receive iron-​reduction therapy; and those with
Fabry disease should continue to receive enzyme replacement therapy.
Specifically for amyloidosis, thorough evaluation for extracardiac manifest-
ations should be performed, as these can lead to worsening morbidity and
mortality after transplantation. In particular, evaluation for renal involvement
(proteinuria), neuropathy (peripheral neuropathy, autonomic dysfunction),
and GI involvement (elevated alkaline phosphatase, malabsorption) should
be undertaken. Severe manifestations of the above may serve as contra-
indications to transplantation. For AL amyloidosis in particular, light-​chain
reductive chemotherapy should be given both before and after OHT (as
soon as it is deemed safe).
Wild-​type amyloidosis requires only OHT, but these patients are often
older and have additional comorbidities, which may preclude them from
being optimal candidates. For hTTR, strategies include combined heart-
liver transplant versus isolated heart transplant followed by ongoing TTR-
directed therapies. Recent data suggest outcomes for both isolated OHT
and combined heart–​liver transplantation in cardiac amyloid patients are
approaching those of OHT in other patients.
In conclusion, RCMs represent a minority of HF patients, but with
growing recognition of cardiac amyloidosis, will be increasingly encoun-
tered. Mainstay of therapy involves management of volume status and util-
ization of disease-​specific therapies when they exist. MCS can be technically
challenging, but utilized in the appropriate patients. If patients are safely
bridged to transplantation, they often do well compared to their non-​RCM
counterparts.
 Section 2

Mechanical
circulatory
support

4 Indications for ventricular assist devices 41

5 Surgical techniques for left ventricular assist device
implantation and associated procedures 49
6 Clinical trials with durable left ventricular assist
devices 61
7 Diagnosis and management of ventricular assist
device complications 69
8 Extracorporeal membrane oxygenation 79
9 Percutaneous mechanical circulatory support 97
10 The total artificial heart 109
11 Post-​cardiac surgery cardiogenic shock 117
40
 Chapter 4 41

Indications for ventricular

assist devices
Short-​term VAD therapy 42
Durable LVADs 44
Myocardial recovery following durable LVAD implantation 46
Are durable LVADs alternatives to heart transplantation? 46
42

42 Chapter 4 Indications for ventricular assist devices

Short-​term VAD therapy

Extracorporeal right VAD (RVAD)
Patients requiring short-​term mechanical support of the right heart can re-
ceive a temporary RVAD which is an extracorporeal bypass configuration
of the RV obtaining drainage from the RA or the RV with blood returning
into the main pulmonary artery (PA). This can be done as an open sur-
gical procedure or percutaneously. Illustration of an open RVAD is seen
in Fig. 4.1.
The main indications for extracorporeal RVAD are:
• RV failure post myocardial infarction or severe pulmonary hypertension.
• Post-​cardiotomy isolated RV failure not responsive to inotropic support.
• RV failure post heart transplantation.
• RV failure following durable LVAD implantation.

Fig. 4.1 Open extracorporeal RVAD.

 Short-term VAD therapy 43

Careful assessment of the function of the LV is required so that the in-

creased preload provided by the RVAD can be accommodated by the con-
tractile function of the LV. In those cases, in which the acute RV failure
requiring mechanical support is accompanied by respiratory failure, an oxy-
genator can be attached to the RVAD circuit (oxy-​RVAD).
Extracorporeal LVAD
Selective bypass of the left heart can be accomplished with an open-​chest
drainage of the left atrium (LA) (via the right superior pulmonary vein)
or the apex of the LV and return to the ascending aorta with the use of
an extracorporeal centrifugal pump (Fig. 4.2). Devices implanted percu-
taneously, or via surgical cutdown of a peripheral artery, in the form of
microaxial pumps (Impella, Abiomed, Danvers, MA, USA) are also gaining
popularity. The main indications for temporary LVAD support are:

Fig. 4.2 Extracorporeal LVAD. Ao, aorta; RSPV, right superior pulmonary vein.
4

44 Chapter 4 Indications for ventricular assist devices

• Cardiogenic shock post myocardial infarction.

• Cardiomyopathies affecting mainly the LV.
• Post cardiotomy.
• Acute viral myocarditis.
• Acute decompensation of transplant candidate—​BTT.

Durable LVADs
LVAD systems for long-​term use comprise a pump, which for most modern
devices is centrifugal and intrapericardial, with an inflow component at-
tached to the LV apex and an outflow graft attached to the ascending aorta.
Developments in technology and perioperative management have im-
proved outcomes of patients on durable LVAD support, which has led to an
expansion in their current indications. The principal indications are:
• BTT.
• DT—​lifelong support.
• Bridge to myocardial recovery.
Within the BTT group there are a subgroup of patients who are potentially
transplant candidates but remain satisfied with their quality of life on LVAD
support and may willingly come off the transplant waiting list. Another
subgroup are those who present with transient contraindications to trans-
plantation or are too sick until a donor could be identified, and form the
bridge-​to-​decision cohort, representing almost 40% of LVAD recipients.
Specific physiological criteria for durable LVAD candidacy
Decision-​making for the listing of the heart transplant candidate takes into
consideration the prognostic benefit of heart transplantation against the
natural course of the disease process. Risk prediction models such as the
Heart Failure Survival Score (HFSS) and the SHFM have been utilized for
this purpose. With recent survival outcome data in LVAD patients, these
scores can inform decision-​making for LVAD candidacy as well. In countries
where LVAD therapy is reserved for those who are transplant candidates,
such as in the UK, the ISHLT listing criteria for heart transplantation should
be met:
• CPX testing with peak VO2 ≤14 mL/​kg/​min (if not on BB) or ≤12 mL/​
kg/​min if receiving BB, in conjunction with unfavourable HF prognosis
score (<80% 1-​year survival calculated by SHFM or a HFSS in the high/​
medium risk range).
The following negative criteria suggest unsuitability for transplantation and
can be extrapolated to the LVAD candidates: irreversible renal, hepatic,
or neurological disease remain absolute contraindications. Obesity, renal
failure (with potential reversibility), cancer history, tobacco use, and sig-
nificant pulmonary hypertension, albeit conventional contraindications to
transplantation, should not preclude mechanical support as a bridge to
candidacy.
It is becoming apparent that although transplantation is intended in many
LVAD recipients, only a small percentage will eventually receive an organ.
Hence, the distinction between BTT and DT is less pertinent in the modern
era and the aforementioned guidance for transplant listing can only be in-
formative for the LVAD candidate.
 DURABLE LVADs 45

With a changing landscape in organ availability, heart transplant waiting

lists, evolving technology, and the available evidence from transplant
and VAD studies, it is methodologically challenging to produce stringent
guidelines for durable LVAD therapies. For these reasons, the European
Association for Cardio-​Thoracic Surgery (EACTS) published an expert con-
sensus statement (rather than guidelines) on long-​term MCS highlighting the
main indications for durable LVADs:
• Functional limitations with NYHA class IIIb or IV and
• EF ≤25% and
• At least one of the following criteria:
• INTERMACS levels 2–​4.
• Inotrope dependence.
• Progressive end-​organ dysfunction.
• CPX testing with peak VO2 ≤12 mL/​kg/​min.
• Temporary MCS dependence.

Specific considerations of the LVAD candidate:

• Chronic biventricular failure poses a challenge to durable LVAD support
and temporary extracorporeal or implantable RVAD support should be
considered.
• Unlike transplantation, age is not a contraindication to durable LVADs
as long as age-​associated comorbidities and frailty are carefully assessed.
• Similarly, in patients with poor glycaemic control and end-​organ
complications, durable LVADs may still be considered.
It is essential that before any definitive conclusions are made with regard
to organ dysfunction reversibility, cardiac output and filling status are opti-
mized. That applies to renal but more importantly to liver function, as irre-
versible liver damage poses a contraindication to LVAD therapies.
Other patient groups where durable LVADs are generally
contraindicated
• Patients with intractable ventricular tachycardia are not isolated LVAD
candidates.
• Patients with restrictive cardiomyopathy are not LVAD candidates due
to small LV cavities which would preclude effective LV apical drainage.
• Ventricular septal defect not amenable to repair.
• Infective endocarditis.
• Proven malignancy with expected survival <1 year.
Risk stratification and timing of implantation of
durable LVADs
Risk stratification models have been developed to identify those who would
benefit the most from durable LVADs and help to inform timing for im-
plantation. It is recognized that a deterioration of the general medical con-
dition as evidenced by low albumin, and derangement of liver and kidney
function tests exponentiates the risk associated with LVAD implantation.
The Lietz–​Miller stratification score, which was based on outcomes of
older-​generation devices, yielded strong concordance between those with
high-​risk profile and mortality after LVAD implantation. The Model for
End-​Stage Liver Disease (MELD) was originally a risk stratification tool for
patients with cirrhosis undergoing portosystemic shunts, but it has been
shown to also predict transfusion requirements and mortality following
46

46 Chapter 4 Indications for ventricular assist devices

LVAD implantation. The formula variables of bilirubin, creatinine, and the

international normalized ratio (INR) provide a surrogate of end-​organ func-
tion, that could influence timing of LVAD surgery. The above-​mentioned
risk stratification tools, in addition to the Destination Therapy Risk Score
(DTRS), were predominantly based on pulsatile flow LVADs and their dis-
criminatory ability for continuous flow devices is limited. A more relevant
risk stratification model for the modern era, the Heart Mate II Risk Score,
was superior to both the DTRS and MELD in predicting 90-​day mortality.
Age, albumin, creatinine, INR, and low LVAD volume cardiac surgical units
were the independent variables used to calculate the risk profiles of LVAD
candidates. While these models offer some prognostication, a multidiscip-
linary approach with a dedicated advanced HF team is recommended to
guide the most appropriate treatment strategy.

Myocardial recovery following durable

LVAD implantation
Although reverse remodelling of the failing ventricle supported with a dur-
able LVAD tends to occur within the first 6 months from implantation, only
a small proportion of these patients (~1%) will meet criteria for separation
from mechanical support within the first year. Durable mechanical support
that leads to recovery of the failing LV and to explant is termed bridge to re-
covery and has an incidence of approximately 3% at 3 years. Patient-​related
parameters such as younger age, shorter duration of HF, and absence of
pulmonary hypertension or significant LV dilatation (LVEDD <6.5 cm) are
positive prognostic factors for myocardial recovery. Likewise, myocarditis
and postpartum cardiomyopathy patients are more likely to recover com-
pared to those with an ischaemic cause. Targeted pharmacological inter-
ventions may enhance reverse remodelling, the likelihood of myocardial
recovery, and ultimately LVAD explantation in select patient groups. There
are no universally accepted LVAD explantation protocols, but different
groups have proposed the following parameters as indicators of appro-
priate recovery prior to explant:
At low LVAD speed, ‘off-​pump’ testing:
• LVEDD <60 mm.
• LV end-​systolic diameter <50 mm.
• EF >45%.
• LV end-​diastolic pressure <12 mmHg.
• Resting cardiac index (CI) >2.8 L/​min/​m2.
• VO2 >16 mL/​kg/​min.

Are durable LVADs alternatives to heart

transplantation?
While heart transplantation is the standard of care for those who have fa-
vourable characteristics to qualify and timely receive an organ, there is an
increased number of HF patients who do not meet the criteria for listing.
It is now well established that durable LVADs offer improved survival and
 ARE VADs ALTERNATIVES TO HEART TRANSPLANT 47

quality of life versus optimal medical management. With improved safety

characteristics of the newer devices, the US FDA has given approval for
their use as DT. But for those ambulatory patients who are also eligible for
transplantation, is durable LVAD implantation a suitable option?
For LVAD therapies to be true alternatives to transplantation the fol-
lowing conditions should be met:
• Symptom relief. Optimization of pump speed based on central
venous and pulmonary capillary wedge pressure (PCWP) reading and
echocardiographic indices can improve filling pressures and ventricular
unloading and translate to better symptom control.
• Reduction in adverse events. Pump thrombosis and stroke are the
main complications associated with LVADs; however, modern pump
engineering and blood pressure and anticoagulation protocols have
led to a modest reduction of these events. VAD-​related infections and
sepsis are still prevalent in part due to the need for exteriorization of
the driveline and are associated with high morbidity and mortality.
• Ease of implantation. Modern LVADs allowing for a fully intrapericardial
implantation through median sternotomy or less invasive techniques are
associated with low operative mortality and reproducibility.
In the absence of randomized trials examining mid-​and long-​term survival
and quality of life between LVADs and heart transplantation, we can only
extrapolate data from observational studies, which demonstrated equipoise
at 1 year. However, the shortage of donor organs, which may impact espe-
cially patients with certain group types, size, and antibody profile, mandates
the use of LVADs even if in their current state they are unlikely to match the
excellent long-​term results of a successful heart transplant. Where LVADs
may be a true alternative to transplantation is in the ‘high-​risk’ recipients—​
those who fail to meet standard criteria for transplantation—​receiving mar-
ginal organs, where median survival is approximately 5 years. Finally, for
those who achieve myocardial recovery with their durable LVAD leading to
explant, not receiving a transplant may be advantageous.
48
 Chapter 5 49

Surgical techniques
for left ventricular assist
device implantation and
associated procedures
Operative planning 50
Sternotomy approach 50
Driveline 50
Cardiopulmonary bypass 52
Inflow 52
Outflow 53
De-​airing and weaning from cardiopulmonary bypass 54
Thoracotomy approach 55
Descending aorta and axillary artery outflow 57
Concomitant procedures 57
Exchange 58
50

50 Chapter 5 LVAD implantation technique

Operative planning
A preoperative plan is determined after a full patient evaluation is com-
pleted, reviewing all relevant haemodynamic data, echocardiography, and
radiographic imaging. The need for concomitant procedures is assessed
preoperatively and confirmed in the operating room after review of the
intraoperative haemodynamic status and the transoesophageal echocardi-
ography (TOE). The TOE is reviewed to establish overall function, the in-
tegrity of the interatrial septum, aortic valve competency, and the presence
of other valvular lesions and LV thrombus. An operative time out reviewing
not only patient information but operative plan and conduct is a practice
that high-​performance teams have developed and maintain.

Sternotomy approach
Implanting an LVAD through a median sternotomy utilizing cardiopul-
monary bypass (CPB) is the most common implantation technique. Most
LVADs are recommended to be implanted using this approach in their FDA-​
approved instructions for use. Specific operative steps may relate to size,
outflow location, driveline exit site, and positioning of the implanted LVAD.
The patient is brought into the operating room and placed in a supine
position. General anaesthesia is induced and maintained. Monitoring lines
are placed typically including arterial line, pulmonary artery catheter (PAC),
and a TOE probe. In case of previous cardiac surgery with the associated
risks of sternal re-​entry, vascular access is typically established as per team
preference ranging from arterial and venous wires and/​or lines, to antici-
patory exposure of the femoral or axillary arteries to facilitate emergent
initiation of CPB. A midline incision is made and a full sternotomy is per-
formed. The diaphragm is mobilized and divided to a varying extent to
allow placing the pump body or outflow graft when dictated by pump size
and configuration. The pocket must be large enough to prevent pump dis-
placement or compromise in the final, closed chest pump position. The
pericardium is opened liberally towards the apex. A pericardial cradle is
created with consideration for an asymmetric vertical opening to allow the
pericardium to be re-​approximated at the conclusion of the implantation.
For LVADs that require a true pocket, the diaphragm is divided with a linear
stapler and a preperitoneal pocket is created anterior to the posterior
rectus sheet. Adequate spacing requires a pocket that is wide enough to
allow full sub-​diaphragmatic pump body placement with the pump parallel
to the axial plane. The HeartMate (HM) 3 (Abbott, Abbott Park, IL, USA)
intrapericardial placement as well as extended extrapericardial placement
by augmentation of the pericardium with prosthetic material. Most LVAD
implantation can be completed with most or all of the left pleura intact.

Driveline
After establishing a pump pocket, consideration is given to tunnelling the
driveline prior to giving heparin. The driveline is tunnelled exiting on ei-
ther side of the abdomen in the midclavicular line below the costal margin
 Driveline 51

but above the belt line. The ideal exit site is marked prior to surgery, with
the patient sitting or standing, and wearing typical clothes. The exit site is
selected and a small skin incision is made and dissected to the rectus fascia.
The driveline is tunnelled internally engaging the rectus sheath within the
chest, maintaining the initial depth to avoid compression from the costal
margin. The driveline is ideally placed within the rectus sheath until immedi-
ately below the chosen exit site where it traverses the fascia perpendicular
with minimal subcutaneous distance (Fig. 5.1). All velour should remain
within the fascia with no visible velour at the skin interface to decrease risk
of driveline infection. An anchoring suture is placed approximately 5 cm
from the exit site to stabilize the driveline in the event of sudden driveline
traction until the driveline is healed. Alternative driveline techniques employ
an initial lateral path to a temporary exit on the flank, with a second medial
pass of the trocar to the final exit site to create a longer tunnel in an attempt
to decrease the risk of driveline infection.

Fig.5.1 Driveline tunnelling technique: the trocar is passed inside the chest, engaging
the posterior fascia to a pre-​marked exit site, remaining extraperitoneal, within the
rectus sheath until exiting through the subcutaneous fascia.
52

52 Chapter 5 LVAD implantation technique

Cardiopulmonary bypass
Cannulation is performed according to the surgeon’s preference with con-
sideration to avoid excess tissue dissection in patients with anticipated fu-
ture operations. In patients with pre-​existing MCS, utilizing previously placed
cannulas and/​or other components can be accomplished in many cases.
Before initiating CPB, the intraoperative TOE findings should be reviewed
for significant valvular lesions, atrial septal defect (ASD), or thrombus in LV
or left atrial appendage (LAA). The tissue quality of the LV and apex should
be assessed for apical calcifications with consideration for alternative im-
plantation techniques in the setting of a hostile apex.
Off-​pump implantation
Concern for the body’s pathophysiological response to CPB has led to
some specific pump designs that would facilitate off-​pump procedures and
off-​pump LVAD implantation technique. This has been limited to a few se-
lective centres due to a high risk of significant blood loss and inability to fully
control LV cavity for debris, thrombus, and air.
LVAD implantation off venoarterial (VA) ECMO
A significant subset of patients implanted with an LVAD require temporary
mechanical support for their initial presentation with cardiogenic shock.
When taken to the operating theatre, VA ECMO cannulas can be used for
CPB. Experienced centres will in selected cases performed entire LVAD
implantation off VA ECMO without use of CPB.

Inflow
CPB is initiated, and any additional dissection is completed to adequately
mobilize the heart for apical exposure. The heart is gently positioned using
pericardial traction and/​or operative sponges/​towels. The patient’s apical
anatomy of the heart is examined. Positioning of the sewing cuff and inflow
cannula are planned at the true apex or slightly anterolateral, to avoid the left
anterior descending artery. In order to appropriately orient the inflow can-
nula towards the MV, the cuff is placed on the proposed apical site and the
TOE views are assessed for inflow alignment. The locking mechanism of the
HM 3 to secure the inflow can be accessed at any rotation of the sewing ring.
The cuff can be secured to the LV prior to coring or after (Fig. 5.2).
The ‘sew-​then-​core’ technique of apical cuff attachment usually requires
anchoring sutures followed by partial-​thickness running sutures. Use of
additional sealants is optional and surgeon/​institution specific. HM 3 uses
a barrel-​shaped coring knife to core out the apex. In the ‘core-​then-​sew’
technique, the apex of the LV is cored first. The sewing cuff is then secured
by a series of full-​thickness pledgetted horizontal mattress sutures, starting
on the outer surface of the LV, 1.5 cm from the cored LV. After placing the
sutures through the cuff, the sutures are tied. The LV cavity is inspected
for thrombus or trabeculae that could interfere with the pump function.
The inflow cannula is then inserted and oriented in the desired position
and secured. In the Multicenter Study of MagLev Technology in Patients
Undergoing Mechanical Circulatory Support Therapy with HeartMate 3
(MOMENTUM 3) study, the ‘core-​then-​sew’ technique required longer
CPB duration but the two techniques yielded similar outcomes.
 Outflow 53

(b)

(a)

Fig.5.2 (a) ‘Sew-​then-​core’ technique for apical sewing cuff. (b) ‘Cut-​then-​sew’
technique showing pledgetted mattress sutures placed after the LV apex is cored.

The heart is returned into the pericardium, and the pump position is as-
sessed for stability and orientation on the TOE. Adjustments to the orien-
tation and angulation can be made at this point.

Outflow
At this point, the outflow graft is attached to the pump and clamped. The
HM 3 outflow graft clip is then placed in order to prevent outflow graft
twists and in HM II and HM 3 bend relief is attached to the pump.
The outflow graft is bevelled for an acute angle of 45–​75° and placed on
the lateral aspect of the ascending aorta distal to the sinotubular junction
(Fig. 5.3). A partial occluding clamp is applied to the aorta and the 14 mm
HM 3 graft is anastomosed to the aorta after confirming occlusion and sizing
the aortotomy. The anastomosis is performed with a 4-​0 or 5-​0 continuous
polypropylene suture. Just before tying the anastomotic suture, the outflow
graft clamp is released, the heart is filled, and lung ventilation is initiated to
de-​air the graft. The graft is then clamped and the partial aortic cross-​clamp
is removed.
54

54 Chapter 5 LVAD implantation technique

Ao
60°

Fig.5.3 Outflow graft anastomosis to the ascending aorta (Ao). The graft is cut on a
bevel to create a 40–​60° angle on the lateral aspect of the aorta.

De-​airing and weaning

from cardiopulmonary bypass
Embolism of air into the coronary circulation may contribute to RV failure
whereas air embolism to the cerebral circulation may result in transient
neurological dysfunction or a stroke. Multiple de-​airing strategies can be
used to decrease the risk of significant air embolism. Using CO2 to flood
the surgical field decreases the air entrained in the heart as does continuing
low-​rate and tidal volume ventilation throughout the procedure. At the end
of the procedure, a de-​airing vent can be placed in the ascending aorta distal
to outflow graft anastomosis or directly through the hood of the graft into
the aortic root. Additional manoeuvres include using a small-​bore needle at
the highest level of the outflow graft.
Ventilation is restarted and preparations are made in anticipation of
weaning from CPB. Separation from CPB requires normothermia, a sat-
isfactory cardiac rhythm with appropriate respiratory dynamics, the ab-
sence of surgical bleeding, and mean arterial pressure (MAP) of at least
70 mmHg. TOE is used to assess RV function, residual air, the presence
of an interatrial shunt, septal position, as well as new aortic insufficiency.
Inotropes and inhaled pulmonary vasodilators are used as per institu-
tional protocols primarily to support the RV. Once the left-​sided cardiac
 Thoracotomy approach 55

chambers are sufficiently filled, the LVAD is turned on at its lowest speed
and slowly ramped up while the CPB flow is decreased. The position of the
intraventricular septum, MR and TR, and aortic valve opening are closely
monitored as CPB is weaned. After successful separation from CPB, CPB
cannulas are removed and anticoagulation is reversed. The inflow and out-
flow anastomoses are examined for bleeding. Chest tubes are placed into
the mediastinum and pleural spaces. Using prosthetic material to cover the
LVAD components to enable easier re-​entry is done as per surgeon prac-
tice. The sternum and soft tissue are closed as per surgeon routine.

Thoracotomy approach
Smaller sized LVADs and the availability of instruments designed for minim-
ally invasive cardiac surgery have enabled increased adoption of less invasive
approaches for LVAD implantation as a strategy to improve perioperative
outcomes. Less invasive implants may be accomplished with a small left
anterior thoracotomy in conjunction with an upper hemi-​sternotomy or a
right anterior mini-​thoracotomy (Fig. 5.4).
The patient is positioned in a modified supine position with a slight bump
or inflatable support under the left chest to enhance apical exposure.
Defibrillator pads are placed laterally to avoid planned left-​sided incisions
and/​or right-​sided pericardial sutures. The sterile field is extended to the
side of the bed on the left. This set-​up enhances the ability of two surgeons
working in tandem, one performing the upper hemi-​sternotomy or right
anterior mini-​thoracotomy while the other is focused on the left anterolat-
eral thoracotomy.
To help plan for the left thoracotomy incision, a CT scan or a TTE is used
to locate the LV apex. The specific intercostal space is planned to optimally
access the apex, which is typically the fifth space. A 6–​8 cm skin incision is
made directly over the LV apex and the intercostal muscle is divided lat-
erally to the midaxillary line in order to facilitate exposure. Alternatively, a
portion of the rib overlying the LV apex can be divided to provide the space
to accommodate the device and avoid inadvertently fracturing a rib. A soft
tissue retractor is placed followed by an intercostal retractor. The pericar-
dium is opened anterior to the phrenic nerve and several stay sutures are
placed to improve exposure of the LV apex. To improve exposure, a stay
suture may be placed in the fibrous portion of the diaphragm and brought
out through a stab wound in the lower chest wall for retraction.
The decision to perform an upper hemi-​sternotomy or right anterior
mini-​thoracotomy depends on surgeon preference, prior cardiac surgery,
and the position of the aorta relative to the sternum. The right anterior
thoracotomy is performed using a 4–​5 cm skin incision at the second inter-
costal space. The third rib is disarticulated and a soft tissue retractor in-
serted into the wound. The upper hemi-​sternotomy is performed through
a 4–​6 cm upper midline incision. The sternotomy can be extended into the
right second intercostal space or teed off into both sides with sternal plate
fixation later. A retractor is then placed and the pericardium is opened,
exposing the ascending aorta. Pericardial retraction sutures are placed to
pull the aorta rightward, improving exposure in primary surgery patients.
A slightly larger incision may be considered in reoperative cases.
56

56 Chapter 5 LVAD implantation technique

5 4 3 2 1

Fig.5.4 Less invasive exposure for LVAD implantation. Small bilateral thoracotomy.

Anticoagulation is given as per protocol and the aorta is cannulated dir-

ectly while venous cannulation is performed percutaneously to the fem-
oral vein into the SVC. The sewing cuff can be affixed to the apex prior
to bypass according to surgeon preference, or bypass can be initiated for
enhanced exposure for sewing cuff placement. The patient is placed in the
steep Trendelenburg position for apical coring and the cuff attached in the
sequence as per surgeon preference.
The LVAD pump is connected to the outflow graft and assembled if not
previously completed. The driveline and outflow graft are then tunnelled. In
primary sternotomy patients the outflow graft is tunnelled intrapericardially
while in re-​do patients the outflow graft is tunnelled beneath the sternum
after dissecting a tunnel into the right pleural space. In some reoperative
cases, a small sub-​xiphoid incision is made to assist in outflow graft posi-
tioning and for additional exposure when necessary.
The pump is then inserted in the incision and secured to the apical sewing
ring. The graft is initially de-​aired by allowing the outflow graft to fill before
it is clamped. After tunnelling the outflow graft to the upper incision and
checking the path and length, the outflow graft is cut and bevelled. A partial
occlusion clamp is applied to the ascending aorta and anastomosis is per-
formed. Upon completion, the heart and VAD are de-​aired in antegrade
and retrograde fashion and the patient is weaned off CPB.
Prior to thoracotomy closure, chest tubes are placed, one in the right
chest and two from the left chest, one in the pericardium, and one in
a pleural space. The pericardium is closed around the device with peri-
cardial membrane to minimize adhesions between the pump body and
the lung.
 CONCOMITANT PROCEDURES 57

Descending aorta and axillary

artery outflow
In patients with multiple prior sternotomies or a hostile ascending aorta
due to calcifications or previous surgeries, the descending aorta and axillary
artery may be considered for alternative outflow graft sites.
Access to the intrathoracic descending aorta is achieved by dividing the
left inferior pulmonary ligament to the level of the inferior pulmonary vein.
The diaphragm is retracted with a stitch in the fibrous portion brought out
through the chest wall. A partial clamp is then applied and anastomosis
performed.
For axillary artery outflow, standard exposure of the left axillary artery
is performed with care not to injure the brachial plexus. A right and left
radial arterial line is placed to monitor outflow. The outflow graft is tun-
nelled through the first interspace and the anastomosis performed on the
inferior aspect of the artery. A distal restricting band may be placed around
the axillary artery to limit left arm overflow. With restriction and full LVAD
flow, the arterial pressure in the left radial should be 75% of that in the right
radial arterial line.
Pump speed is adjusted in both alternative outflow positions to allow
aortic valve opening in order to prevent aortic root thrombosis and pos-
sible embolic stroke.

Concomitant procedures
The surgical management of associated valvular disease in patients under-
going LVAD implantation remains controversial. There is evidence (class 1,
level of evidence C) of benefit from repairing or replacing an aortic valve
with more than mild to moderate insufficiency, but the benefit in intervening
on the MV and tricuspid valve is unclear. Closure of a sizeable patent for-
amen ovale (PFO) is recommended in order to prevent hypoxia associated
with significant right-​to-​left shunt after LVAD implantation. There is growing
evidence from retrospective studies that concomitant LAA occlusion may
decrease subsequent thromboembolic events.
Aortic valves with moderate insufficiency can be repaired, replaced with
a bioprosthetic valve, or oversewn. Aortic valve repair using a central stitch
to coapt the leaflets (Park’s stitch) will, if placed appropriately, reduce cen-
tral aortic insufficiency while allowing flow through the aortic valve and
washing of the aortic root. A pledgetted 4-​0 or 5-​0 polypropylene suture is
placed centrally halfway into all three aortic cusps and tied. This manoeuvre
can be performed through the aortotomy for the outflow graft or through
a separate aortotomy. If the aortic cusps are damaged or not coapting,
replacement with a bioprosthetic aortic valve or oversewing the valve can
be performed. These manoeuvres require cross-​clamping the aorta but the
aortic valve can be addressed through a right mini-​thoracotomy or upper
hemi-​sternotomy as well as a full sternotomy.
Concomitant MV repair for severe functional MR remains controversial.
Repair can be performed through full and upper mini-​sternotomy incisions.
The valve is exposed through the interatrial groove after the sewing cuff
58

58 Chapter 5 LVAD implantation technique

is attached and coring has been performed. The patient is placed in the
steep Trendelenburg position to avoid entrapping air. Aortic cross-​clamping
is not necessary to perform mitral repair in this setting. Alternatively, an
Alfieri stitch can be placed through the core of the LV and effectively de-
creases MR.
The tricuspid valve can also be repaired through both sternotomy and
thoracotomy approaches with bicaval cannulation. TR is often secondary
due to left-​sided failure, with a higher afterload in the RV and subsequent
tricuspid annulus dilatation. Consideration should be given for repairing a
tricuspid valve with more than moderate regurgitation and reversal of flow
in the hepatic veins demonstrated on intraoperative echocardiography, in
the absence of an increase in the RV short/​long axis ratio. Repair is per-
formed through the RA without aortic cross-​clamping with an annuloplasty
ring placed prior to VAD implantation. In less invasive approaches, a right
thoracotomy can be performed through the third interspace to allow tri-
cuspid valve exposure in addition to the ascending aorta.
Occlusion of the LAA has been suggested to decrease thromboembolic
events in patients with and without atrial fibrillation with LVAD. Multiple
techniques exist and occlusion can be safely performed through most inci-
sions on CPB with occlusion devices or standard ligation. Prior to occlusion,
the presence of clot is excluded with TOE and post occlusion, the LAA is
assessed to confirm closure.

Exchange
The need for pump exchange due to thrombosis, infection, or driveline/​
pump failure created new surgical challenges. Most commonly, exchanges
performed offer an opportunity to upgrade pump selection to newer de-
vices (e.g. from HM II or HVAD [withdrawn from the market] to HM 3).
Depending on the indication for pump exchange, the option for less inva-
sive and possibly off-​pump exchange of the LVAD body alone is possible in
many circumstances.
HM II to HM II pump body exchange can be performed through either
sternotomy or subcostal access, on or off pump. Subcostal access via an 8–​
10 cm subcostal incision with dissection down to the pocket is performed.
The inflow, outflow elbow, and proximal outflow graft are exposed. The
driveline is controlled from the incision if infection is not present. Once
the pump is explanted, the new pump is implanted and the new driveline is
usually tunnelled to the opposite side from the previous driveline position.
When CPB is used, femoral cannulation is employed. Anticoagulation is ad-
ministered after exposure of the pump and the femoral vessels are cannu-
lated for CPB. The bend relief is disconnected and the pump is stopped.
The outflow graft is clamped and disconnected. In off-​pump exchange, the
inflow may be clamped or occluded with a balloon catheter. De-​airing is
performed with a small-​bore needle, and maintaining LV filling to prevent
air from being entrapped. CO2 is used to flood the field.
Exchange of HM II or HVAD to HM 3 is a more extensive procedure
than a subcostal exchange. The assessment of patient candidacy for this ex-
change includes preoperative TOE and right heart catheterization to assess
 Exchange 59

for valvular abnormalities, an assessment of RV function, and CT of the

chest to exclude outflow graft obstruction/​thrombus.
HM II to HM 3 exchange can be performed through a re-​sternotomy or
through a subcostal incision. In the latter approach, the HM II pump body
and outflow graft elbow is exposed through the main subcostal incision,
while a small left thoracotomy is made to access the apex of the heart
and the prior inflow cannula (Fig. 5.5). In most cases, the HM II sewing
cuff can be trimmed to 12–​15 mm and the HM 3 implanted through this
cuff, securing the pump with cable ties. Alternatively, a HM 3 specific cuff
can be sewn after excising the HM II cuff on CPB. HVAD to HM 3 ex-
change is performed through a median sternotomy or a left thoracotomy.
The thoracotomy exchange is performed on CPB with femoral cannulation.
The HVAD apical cuff is excised and the HM 3-​specific apical cuff in secured
in standard fashion.
Upon removal of the prior HM II, and after securing the HM 3, the new
outflow graft can be tunnelled into the subcostal incision and an end-​to-​end
bevelled anastomosis performed to the original HM II outflow graft.
Due to outflow graft size discrepancies, when exchanging a HVAD
(10 mm outflow graft) to HM 3 (14 mm outflow graft), the outflow graft-​
to-​graft anastomosis needs to be carefully sized and tapered, and bend re-
lief length may need to be trimmed for appropriate visualization. Deairing
is performed by the use of a vent placed in the outflow graft or with a
small-​bore needle.

Old pump
HM 3

Fig. 5.5 HM II to HM 3 exchange through subcostal and thoracotomy approach.

60
 Chapter 6 61

Clinical trials
with durable left
ventricular assist devices
Pre-​trial era 62
Early trials in first-​generation devices 62
Expansion to destination therapy 63
Second-​generation devices 64
Third-​generation devices 65
Conclusion 67
62

62 Chapter 6 Clinical trials related to LVADs

Pre-​trial era
The first human implantation of a total artificial heart as a BTT occurred in
1969. Over the next two decades, several LVADs—​all volume displacement
pumps—​were developed, including the HeartMate implantable pneumatic
left ventricular assist system (IP LVAS), the Thoratec Pierce–​Donachy VAD,
and the Novacor LVAS. However, data about these devices were limited
to case reports and series, and adoption remained limited to a handful
of centres. In the late 1980s, the US FDA granted investigational device
exemption approval to these first-​generation devices, paving the way for
large-​scale trials.

Early trials in first-​generation devices

Issues facing early trials
Double-​blind, randomized, placebo-​controlled trials are the gold standard
of clinical evidence. However, implementation of this design in surgery can
be impractical and even at times unethical. Challenges include the incorpor-
ation of blinding and placebo controls (e.g. sham procedures), the difficul-
ties of standardizing complex surgeries, and the learning curve associated
with novel procedures.
The small population of BTT patients was a key challenge facing early
LVAD trials. The Novacor LVAS trial, which started in 1996, enrolled ‘car-
diac transplant candidates in imminent danger of death’. In that year, heart
transplant waiting list candidates numbered 2,436 in the USA; only a small
fraction would qualify. Eliminating the randomized control arm could reduce
the sample size and shorten recruitment time, potentially leading to more
rapid approval. However, this would be at the cost of study validity. This
trade-​off was accepted by all of the early trials, which used either a single-​
arm design or incorporated non-​randomized controls: historical controls or
contemporaneous controls who either declined the device or for whom the
device was unavailable.
Results of early trials
These early trials convincingly demonstrated improved survival until trans-
plant for deteriorating waiting list patients. The HeartMate IP was im-
planted in 75 patients at 17 centres between August 1985 and September
1993. Compared to 33 non-​randomized controls, HeartMate IP patients
had greater survival until transplantation (71% vs 36%, p =​0.001). The
Novacor LVAS trial, which recruited 129 patients at 22 centres between
March 1996 and June 1998 as well as 35 non-​randomized controls, also
demonstrated improved survival until transplantation: 75% versus 37%
(p <0.001). Similarly, in a single-​arm trial, 63% of 186 BTT patients at 41
centres who received a Pierce–​Donachy VAD (also known as the PVAD)
survived until transplantation. These results led to FDA approval of all three
devices for BTT.
An important secondary finding of these trials was significant end-​organ
function improvement. Additionally, the major complications following
LVAD implantation were characterized. Among HeartMate IP patients,
complications included bleeding (41%), infection (41%), thromboembolism
 Expansion to destination therapy 63

(4%), and right HF (15%) during a mean follow-​up of 76 days. Novacor

LVAS patients experienced similar adverse events (with a notably higher
risk of stroke): bleeding (39%), infection (30% systemic, 12% pocket, and
10% driveline), central nervous system (CNS) thromboembolism (25%),
and RV dysfunction (9%). Thoratec PVAD patients, during a mean support
of 24 days, also experienced bleeding (26%), infection (55%), and thrombo-
embolism (6%). Adverse event data from these trials served as benchmarks
for future trials investigating upgraded devices.
These early trials lacked many features now commonly accepted as
obligatory: protocols were not published, primary endpoints were not
specified, sample size calculations were not reported, and adverse event
definitions were not described. However, they successfully demonstrated
the potential of LVAD therapy. Together, these trials gave LVAD therapy
a solid foothold within the armamentarium of HF therapies and motivated
the development of upgraded, implantable devices that allowed BTT pa-
tients to leave the hospital and paved the way for DT.

Expansion to destination therapy

Bethesda Conference
In 1995, the Bethesda Conference, attended by clinicians, scientists, in-
dustry, and the FDA, met to create guidelines for DT trials, including a gen-
eral definition of the DT population and the endpoints that would define
success in this setting. Additionally, future trials would be expected to de-
termine pre-​specified sample sizes based on statistical power and to use
standardized adverse event definitions.
REMATCH trial
The Randomized Evaluation of Mechanical Assistance for the Treatment of
Congestive Heart Failure (REMATCH) was the first DT trial. This was the
first LVAD trial to publish a detailed protocol and the first to employ a ran-
domized control arm. Between May 1998 and July 2001, 129 patients with
chronic end-​stage HF ineligible for heart transplantation were randomly as-
signed to HeartMate XVE implantation (N =​68) or optimal medical man-
agement (N =​61). Survival at 1 year was 52% in device recipients and 25%
in the controls (p =​0.002). Adverse events remained common, in particular
bleeding and sepsis, and device failure at 24 months was 35%. Based on
these data, the HeartMate XVE became the first LVAD to receive FDA
approval for DT.
INTrEPID trial
Enrolment in the REMATCH trial was slower than expected. This was at-
tributed to a lack of equipoise among patients and clinicians, given the suc-
cessful BTT trials. In view of this concern and the anticipation of high rates
of control arm dropout, the Investigation of Nontransplant-​Eligible Patients
who are Inotrope Dependent (INTrEPID) DT trial with the Novacor LVAS
used non-​randomized controls. INTrEPID enrolled 37 patients between
March 2000 and May 2003 and followed 18 non-​randomized controls.
While Novacor LVAS patients had superior 1-​year survival compared to
controls (27% vs 11%, p =​0.02), these results appeared inferior to the
64

64 Chapter 6 Clinical trials related to LVADs

REMATCH device group. Additionally, Novacor LVAS patients had a high

1-​year stroke risk (62%), a concerning finding consistent with the Novacor
LVAS BTT trial. The investigators hypothesized these differences might have
been due to patient selection, smaller trial sample size resulting in less accu-
mulation of clinical experience, or inherent device differences. Ultimately,
this device did not receive DT approval.

Second-​generation devices
Continuous flow devices
First-​generation volume displacement devices had many disadvantages: their
large size limited application to female and paediatric patients, the multitude
of moving parts limited durability, and the large drivelines promoted infec-
tion. Second-​generation, continuous flow LVADs were much smaller, had a
single moving part (the rotor), and required a slimmer driveline. Examples
include the Thoratec HeartMate II, the MicroMed DeBakey VAD, and the
Jarvik 2000 Heart. However, many feared their minimal pulsatility was in-
compatible with normal physiology. Additionally, aortic root thrombus
was a concern. Definitive demonstration of safety and efficacy required
trial data.
Updated FDA guidance
Second-​ generation devices also raised questions about trial design—​
specifically, whether control arms should include contemporaneous subjects
randomized to first-​ generation devices or whether historical controls
should be used instead. Again, major stakeholders convened to discuss
strategies, and in 2006 the FDA published separate design guidelines for
BTT and DT trials. Given the large DT patient population, DT trials would
include randomized controls. Additionally, given the permanent nature of
DT, high-​quality data demonstrating improved durability and safety were
desired. In contrast, the BTT population remained small. Therefore, these
trials could continue using single-​arm designs, with the following caveats: in
lieu of controls, investigators would use objective performance criteria as
benchmarks for success (e.g. pre-​specifying a threshold proportion of pa-
tients meeting the primary endpoint); additionally, post-​marketing studies
including contemporaneous controls were mandated. This was enabled
by the creation of the INTERMACS registry in 2005, which could facilitate
comparison of first-​and second-​generation devices while controlling for
baseline comorbidities (uniformly collected using standardized definitions).
HeartMate II BTT trial
The HeartMate II BTT trial was a single-​arm, prospective, multicentre
study. The primary endpoint was the proportion of patients who were
transplanted, weaned off LVAD support, or alive on the heart transplant
waiting list at 180 days. To demonstrate success, a pre-​specified objective
performance criteria was defined as the lower end of the 95% confidence
interval for the primary endpoint exceeding 65%. Between March 2005
and May 2006, 133 patients at 26 centres were enrolled. In addition to
similar inclusion and exclusion criteria as prior BTT trials, this trial excluded
patients with mechanical aortic valves and aortic aneurysms. At 180 days,
 Third-generation devices 65

75% of patients achieved the primary endpoint. Additionally, compared to

historical data from the HeartMate XVE BTT trial, reductions in bleeding,
infection, and right HF were demonstrated. After receiving BTT approval
in April 2008, a post-​marketing study using the INTERMACS registry was
performed, corroborating the trial results.
HeartMate II DT trial
The HeartMate II DT trial included a randomized control arm. Given
the focus on reducing device complications in DT patients, the primary
endpoint was a composite endpoint: survival free from disabling stroke
and reoperation at 2 years. Between March 2005 and May 2007, 200 pa-
tients were enrolled and randomly assigned 2:1 to either the HeartMate II
(N =​134) or HeartMate XVE (N =​66). At 2 years, 46% of HeartMate II
patients and 11% of controls achieved the primary endpoint (hazard ratio
0.38; 95% confidence interval 0.27–​0.54; p <0001). Again, significant re-
ductions in device-​related infection and right HF were demonstrated. The
HeartMate II received DT approval in January 2010.
Assessment of second-​generation LVAD trials
The results of the HeartMate II trials and other trials investigating second-​
generation devices represented another leap forward in device perform-
ance as well as trial design. These trials demonstrated how collaboration
between academia, industry, and the FDA could result in creative solutions
to complex challenges that balanced timely approval of life-​saving devices
with patient safety. However, LVAD patients continued to be plagued by
complications such as pump thrombosis and stroke, and further improve-
ments of LVADs were required to offer this therapy to more patients.

Third-​generation devices
The focus on reducing pump thrombosis and thromboembolism motivated
the development of third-​generation devices. The shearing of blood and
heat generation at bearing contact points were thought to promote throm-
bosis. Thus, third-​generation devices, such as the HeartWare Ventricular
Assist Device (HVAD)-​now withdrawn from production and the HeartMate
3 (currently the only FDA approved third generation LVAD on the market)
are characterized by a ‘bearingless’ rotor held in place by magnetic and/​
or hydrodynamic forces. Additionally, these miniaturized devices could be
implanted completely within the thorax, thus eliminating the need for an
abdominal pump pocket. Trial design continued to evolve: innovations in-
cluded use of contemporaneous registry-​based controls, adaptive trial de-
signs, and a paradigm shift in the concept of LVAD indications. Additionally,
while these upgraded devices were anticipated to reduce adverse events,
short-​and long-​term survival were not expected to significantly improve;
trials during this era therefore typically performed non-​inferiority compari-
sons of the primary endpoint. Although the HVAD has been removed from
production, we have made the decision to include some pertinent trials,
since there currently are many patients around the world who are still sup-
ported with this device for both DT and BTT, implanted prior to the device
withdrawal.
6

66 Chapter 6 Clinical trials related to LVADs

ADVANCE trial
The ADVANCE trial, investigating the HeartWare HVAD for BTT, was
the first to include a contemporaneous, registry-​ based, control arm.
Demonstrating that these arms were comparable at baseline was critical. To
accomplish this, the investigators enrolled contemporaneous INTERMACS
registry BTT patients who met identical inclusion and exclusion criteria.
Next, they used baseline data to calculate for each subject a propensity
score predicting likelihood of being in the treatment arm. Subjects were
then stratified into propensity score quartiles, and comparability was as-
sessed by confirming that each arm contained five subjects per quartile.
The primary endpoint was survival until transplantation or recovery or con-
tinued survival on the waiting list at 180 days. The two arms were then com-
pared by calculating the weighted average of quartile-​specific differences in
the proportions of patients achieving this endpoint. Between August 2008
and February 2010, 140 patients were enrolled in the treatment arm; 499
INTERMACS control patients were identified. Overall, 92.0% of HVAD pa-
tients met the primary endpoint compared to 90.1% of controls, and the
HVAD was found to be non-​inferior.
Although the ADVANCE trial successfully resulted in BTT approval for
the HVAD in November 2012, this experience demonstrated the chal-
lenges of using registry-​based controls. Although the pre-​specified com-
parability criteria were satisfied, a significantly larger proportion of control
patients were within INTERMACS profiles 1 and 2. Additionally, differences
in adverse events definitions and reporting precluded direct comparison.
ENDURANCE trial
The HVAD was next evaluated for DT in the ENDURANCE trial, a pro-
spective, randomized, multicentre trial. Subjects were randomly assigned
2:1 to receive the HVAD or HeartMate II. A novel feature of this trial was
the adaptive design for sample size determination, allowing for early trial
termination: after 300 patients reached 2 years of follow-​up, an interim
analysis would determine the proportion of control patients meeting the
primary endpoint (survival free from disabling stroke or reoperation). If this
proportion exceeded 55%, the trial would be terminated (under these con-
ditions the study would still achieve 90% power to establish non-​inferiority);
otherwise, 450 subjects would be enrolled. From August 2010 through May
2012, 446 patients received either the HVAD (N =​297) or HeartMate II
(N =​148). The HVAD was shown to be non-​inferior to the HeartMate II
(55.4% vs 59.1%, p =​0.01). While similar rates of major bleeding and infec-
tions were found, a higher stroke risk was demonstrated in HVAD recipi-
ents (29.7% vs 12.1%, p <0.001). The HVAD received DT approval in 2017.
MOMENTUM 3 trial
During this period, the traditional dichotomy of BTT versus DT began to be
challenged. Rather than seeing these patients as clustering statically within
two distinct populations, some argued that the whole LVAD population
represented a dynamic continuum of transplant eligibility, given that LVAD
support frequently reversed major barriers to eligibility (e.g. renal dysfunc-
tion or PH). According to INTERMACS data, 20% of DT patients ultimately
underwent heart transplantation. Calls emerged instead for a single LVAD
 Conclusion 67

indication and the investigation of short-​term and long-​term endpoints in

a single trial, with the added benefits of reducing device development time
and trial-​related costs.
The MOMENTUM 3 trial, a prospective, multicentre, randomized trial
comparing the HeartMate 3 with the HeartMate II, was the first to imple-
ment a single set of inclusion and exclusion criteria including both BTT and
DT patients. Subjects were organized into a short-​term cohort (consisting
of the first 294 patients, in whom 6-​month outcomes would be analysed)
and a long-​term cohort (consisting of the first 366 patients, in whom 2-​year
outcomes would be analysed). The primary endpoint was survival free from
disabling stroke or pump exchange at both time points. At 6 months, the
primary endpoint occurred in 86.2% of HeartMate 3 patients and 76.8%
of HeartMate II patients (p <0.001 for non-​inferiority). At 2 years, the pri-
mary endpoint occurred in 79.5% of HeartMate 3 patients and 60.2% of
HeartMate 2 patients, meeting significance for not only non-​inferiority but
also superiority. Need for pump exchange and stroke risk were also sig-
nificantly lower in the new device. FDA approvals for BTT and DT were
granted for HeartMate 3 in August 2017 and October 2018 respectively.

Conclusion
Clinical trials have played a central role in the development and adoption of
LVAD devices. The specific challenges raised by investigating these devices
were met by innovations in trial design and aided by cooperation between
clinicians, scientists, industry, and the FDA. Together, these investigators
successfully paved the way for a promising new era in the treatment of
HF patients.
68
 Chapter 7 69

Diagnosis and
management of ventricular
assist device complications
Introduction 70
Surgical VADs 70
Percutaneous and minimally invasive VADs 76
Conclusion 77
70

70 Chapter 7 Clinical trials related to LVADs

Introduction
Over the past two decades, there have been major advancements in the
management of patients with end-​stage HF. Patients with VADs are no
longer bed-​bound and are encouraged to ambulate to mitigate the risks
of physical and psychological deconditioning. Newer generation centrifugal
durable VADs confer superior flow dynamics and significantly improved
event-​free recovery and survival. In this chapter we will focus on the most
common complications pertaining to surgical and percutaneous VADs. In
the surgical VAD category, the term ‘durable VAD’ refers to long-​term,
surgically implantable LVADs (most commonly the HeartMate (HM) 3
(Abbott, Abbott Park, IL, USA) or HVAD (Medtronic, Minneapolis, MN,
USA)-​now withdrawn from production, no approved implantable, durable
RVAD currently exists), and the term ‘extracorporeal VAD’ denotes cen-
tral cannulation for purposes of shorter-​term and temporary LVAD, RVAD,
or biventricular assist device (BiVAD) support, connected extracorporeally
to a centrifugal pump. While these two broad surgical VAD categories are
separate entities, they serve the same purpose. Hence, there are some
complications that are common between the two modes of surgical VAD
support and there are some complications unique to each. We shall also
touch upon the complications of the most commonly utilized percutaneous
VADs. Although the HVAD has been removed from production, we have
made the decision to include some pertinent trials, since there currently are
many patients around the world who are still supported with this device for
both DT and BTT, implanted prior to the device withdrawal.

Surgical VADs
The ‘extracorporeal VADs’ are inserted via direct central cannulation most
commonly through median sternotomy. LVADs are instituted via direct can-
nulation of the LA or the LV apex as an inflow connection to the pump and
a cannula directly inserted into the ascending aorta as an outflow connec-
tion. RVADs are instituted through venous cannulation of the RA as an
inflow connection to the pump and a main PA cannulation as outflow con-
nection. A combination of both would constitute a BiVAD system. A mem-
brane oxygenator can be spliced into either of these two circuits if there is
impaired oxygenation or ventilation (see Chapter 5).
The ‘durable’ LVADs are inserted for the purposes of BTT or for DT.
Eligible patients by definition should have worsening end-​stage LV failure
(INTERMACS level 3 or lower) while being on intravenous (IV) inotropic
support and/​or IABP with sub-​normal cardiac indices and/​or high PVR,
or as an outpatient with worsening symptomatic HF (NYHA class IV) and
EF <25% on maximal medical therapy (see Chapter 4). Over the past two
decades there has been enormous development in LVAD technology from
pulsatile devices (HM XVE) to axial continuous flow (HM II) to centrifugal
pumps (HM 3 and HVAD) with more superior flow dynamics.
Complications of surgical VADs
Mortality rate (durable VAD)
The reported 2-​year ‘event-​free’ (freedom from disabling stroke, need for
device exchange or removal) survival rate according to the MOMENTUM 3
 SURGICAL VADs 71

trial (see Chapter 6) was 77.9% for the centrifugal flow HM 3 device versus
56.4% for the axial flow HM II device, hence there has been a significant
paradigm shift towards this newer technology.
Major mediastinal/​cannulation site bleeding (durable and extracorporeal VAD)
Bleeding is the most commonly encountered complication of surgical VADs
and can occur in as many as 50–​80% of patients. A significant proportion
will require blood product transfusion. As many as 30% of patients would
require re-​exploration in the operating room. Major bleeding occurs be-
cause of coagulopathy, inadequate surgical haemostasis, or a combination
of both, hence checking coagulation screen and thromboelastography
(TEG) and correction of any coagulopathy should take place as a matter of
priority. Frequently the patients may have been on another form of MCS,
for example, IABP and/​or VA ECMO with heparinization for a prolonged
period of time, increasing the risk of bleeding from coagulopathy or platelet
dysfunction (e.g. acquired von Willebrand disorder). Some may have had
previous median sternotomies requiring sternal re-​entry and extensive
adhesiolysis. Hence, there should be meticulous attention to haemostasis
during VAD implantation surgery. Temporary VAD cannulation sites are
particularly prone to bleeding. We recommend that the cannulation sites in
the aorta, the LV apex (in cases of LVAD support), the RA, and the PA (in
cases of RVAD support) should utilize double purse strings with pledgets.
The lines should be tunnelled with care. Wide-​bore chest drains should be
placed in the pericardium and the mediastinum. These typically stay until
after full heparinization is commenced and the desired activated partial
thromboplastin time (aPTT) has been achieved. Unless excessive bleeding
is encountered, the sternum is commonly closed with sternal wires in a rou-
tine fashion. Attention should be paid to wire puncture sites prior to chest
closure as these frequently bleed. Some practices elect to leave the chest
open routinely until all coagulation has been corrected and mediastinal
drainage has subsided in order to mitigate the risk of cardiac tamponade
and bleeding. Anticoagulation with unfractionated heparin is typically held
for the day of operation and recommenced the following day. The aPTT
range for the extracorporeal VADs should be set at 40–​50 seconds and
increased incrementally to 60–​70 seconds with careful attention to chest
drain output. However, for the durable VADs such as the HM 3 with low
risk of thrombosis, aPTT can kept at a lower range (40–​50 seconds). In case
of bleeding, there should be a low threshold for returning to the operating
room for re-​exploration. Human leucocyte antigen (HLA) allosensitization
in BTT patients and transfusion-​related lung injury are two major adverse
effects of blood product transfusion. In cases of excessive bleeding and
TEG prolongation of ‘R time’, coagulation factor transfusion is indicated.
Antifibrinolytics such as tranexamic acid are good adjuncts for perioperative
haemostasis. A low-​dose recombinant factor VII can be administered with
extreme caution in a select number of cases. Desmopressin (e.g. DDAVP)
can be used in patients with suspected acquired von Willebrand disorder.
Gastrointestinal tract bleeding (durable and extracorporeal VADs)
This complication can occur in both durable and extracorporeal VADs.
Durable VAD GI bleeding has an incidence of around 10% per patient
year. The bleeding sites are commonly from the stomach, the duodenum,
or from colonic angiodysplasias. Any GI bleeding should trigger cessation
72

72 Chapter 7 Clinical trials related to LVADs

of anticoagulation, correction of coagulopathy and thrombocytopenia and

administration of antifibrinolytics (e.g. tranexamic acid). Upper and lower
GI endoscopy with or without angiography and embolization are indi-
cated. Emergent surgical intervention should be reserved for massive life-​
threatening haemorrhage. Careful planning and attention to the path of the
durable LVAD driveline as it traverses the anterior abdominal wall should
be made prior to embarking on an exploratory laparotomy as there is a risk
of iatrogenic damage to the driveline.
Driveline infection (durable VADs)
Surgical site infection is a common complication and the rate of driveline
infection can be as high as 30% per patient year. Any evidence of purulent
discharge, erythema, heat, pain, and swelling at the driveline tract or exit
site (with or without systemic manifestations) should alert the surgical team
of the possibility of deep-​seated driveline infection. The most commonly
implicated microorganisms include Staphylococcus aureus, Pseudomonas
species, and enteric Gram-​negative bacteria. Blood cultures should be
drawn and followed by commencement of empirical broad-​spectrum IV
antimicrobial agents and CT scanning. The CT scan should demonstrate
any evidence of abscess formation or subcutaneous gas, warranting surgical
exploration and debridement. A tractotomy is performed along the path of
the driveline and the driveline is relocated from the site of infection. At the
operation, all infected and necrotic tissue should be removed and irrigated
down to the bleeding and healthy tissues. Vacuum-​assisted closure applica-
tion can aid wound healing after debridement; however, in the short term
it can potentially accentuate bleeding from the surgical site especially fol-
lowing recommencement of anticoagulation. If the infection has extended
near the LVAD device, exchange is indicated in a selected group of patients.
The diagnosis of driveline infection prioritizes patients on the heart trans-
plant list in the USA, highlighting the importance of this diagnosis.
Stroke (durable and extracorporeal VADs)
This is one of the most devastating complications of LVAD implantation.
All patients should be anticoagulated following VAD implantation. Stroke in
VAD patients can be haemorrhagic or embolic. Hence, meticulous atten-
tion to coagulation studies should be paid by both clinicians and the patients
themselves. Embolic strokes occur twice as commonly than haemorrhagic
strokes in VAD patients. The rate of stroke in VAD patients can be as high
as 10% for the HM 3 vs 20% for the axial flow HM II device (2-​year out-
comes of MOMENTUM 3 trial, see Chapter 6). The highest risk of stroke
nevertheless appears to be in patients supported with the HVAD even
compared to the axial flow HM II device and careful anticoagulation moni-
toring and blood pressure control of patients receiving this device is advised
(ENDURANCE trial, see Chapter 6). Cerebral CT scanning should be per-
formed urgently, and neurology and neurosurgical opinion should be sought
following a suspected stroke to consider surgical/​interventional therapy if
possible. However, prognosis remains poor in VAD patients sustaining large
strokes, and the BTT patients may be removed from the heart transplant
list with only a minority reconsidered for transplantation upon recovery.
Arrhythmias (durable and extracorporeal VADs)
Atrial or ventricular arrhythmias can be a cause of RV dysfunction, redu-
cing LVAD preload and causing low cardiac output states. In stable patients,
 SURGICAL VADs 73

atrial arrhythmias can be terminated with class III antiarrhythmics (e.g.

amiodarone or sotalol) and with correction of electrolytes (e.g. K+​ and
Mg2+​). Haemodynamically unstable or symptomatic patients should receive
direct current (DC) cardioversion. Ventricular arrhythmias, on the other
hand, can cause more acute deterioration of haemodynamics and should
be treated more promptly with electrical cardioversion or defibrillation.
Fortunately, most patients who receive durable LVADs will have received
a CRT defibrillator or an ICD, which commonly terminate the ventricular
tachycardia/​ventricular fibrillation rhythms as they occur. Beta-​adrenergic
agonists and phosphodiesterase inhibitors should be weaned off; common
antiarrhythmics such as lidocaine or amiodarone should be initiated. BBs
may also be helpful. Less frequently, ventricular arrhythmias are refractory
to pharmacological and electrical measures and approaches such as stellate
ganglion block or catheter ablation may be necessary.
Cannula displacement (extracorporeal VADs)
There is an increased risk of cannula displacement with patient ambulation.
It is imperative that the cannulas are secured with double purse strings and
the snares affixed to the cannulas at multiple levels with heavy silk ties. The
cannulas should be tunnelled laterally along the costal margin so that they
are not in the way of the patient’s lower extremities when they ambulate.
Externally, the cannulas should be affixed to the skin at multiple levels with
heavy braided silk sutures and their tightness should be verified regularly
by the nurses and physicians on daily rounds so as to avoid displacement.
Hypoxaemia (durable and extracorporeal VAD)
Hypoxia can be related to the LVAD or RVAD support. Thrombus in an
RVAD circuit with pulmonary embolization may manifest as hypoxia.
Management consists of increased anticoagulation and replacement of
the circuit (see later section). RVAD flows should be adjusted with cau-
tion and preferably with echocardiographic guidance; over-​setting of RVAD
speeds without careful attention to the function of the left heart can lead
to supraphysiological pressures in the pulmonary circulation and pulmonary
oedema. Where there is intracardiac connection (e.g. PFO), LVADs can
significantly accentuate right-​ to-​
left shunting, which can lead to signifi-
cant hypoxaemia. Hence, any intracardiac shunts should be identified by
preoperative TOE and closed at the time of LVAD placement. Pre-​and
postoperatively, percutaneous device closure (e.g. Amplatzer, Abbott,
Abbott Park, IL, USA) can be considered if the anatomy of the shunt is
suitable. Reverting to VA ECMO support may be a safe salvage strategy if
there is a significant decline in the patient’s oxygenation or ventilation from
any of the above-​mentioned causes of respiratory failure.
Right ventricular failure after LVAD (durable and extracorporeal VADs)
This is one of the most dreaded complications of LVADs. It can occur in 9–​
44% of all durable LVAD implants of which nearly 5% (see Chapter 6 for the
MOMENTUM 3 trial) may require RVAD support. The presentation could
be sudden (i.e. immediately following LVAD implantation and upon weaning
from CPB) or more insidious (i.e. worsening renal and liver function with
high right heart filling pressures in the days and weeks following LVAD im-
plantation). The aetiology of this phenomenon is not entirely clear but it is
postulated that RV overloading due to increase in venous return as a result
74

74 Chapter 7 Clinical trials related to LVADs

of improved cardiac output and leftward displacement of the ventricular

septum can lead to RV dysfunction. Preoperative predictive factors of RV
failure after durable LVAD implantation include the presence of severe TR,
high RA pressures (RA pressure mmHg/​PCWP mmHg ratio >0.6) and low
(<2.0) PA pulsatility index (=​systolic PA pressure –​diastolic PA pressure/​
RA pressure). Severe TR should be repaired at the time of LVAD implant-
ation. TOE and PA catheters are indispensable tools for the diagnosis of
LVAD-​induced RV failure. There should be a high index of suspicion with
increasing RA pressures (>15 mmHg), low SvO2, low CI, low PA pressures,
reduced pulsatility index, and increased suction events. RV function can be
improved through conservative measures (e.g. pacing the RV at higher rates
(100–​110 bpm), inodilator and inhaled pulmonary vasodilator support). If
these measures fail, consideration should be given for reopening the chest
to allow more space for the RV and/​or placement of extracorporeal or
percutaneous RVAD (e.g. ProtekDuo (LivaNova, Pittsburgh, PA, USA)/​
Impella RP (Abiomed, Danvers, MA, USA)—​see below) implantation be-
fore end-​organ failure has set-​in.
Thrombosis and haemolysis (durable and extracorporeal VADs)
Durable VAD thrombosis is an increasingly rare phenomenon in the
context of the newer generation centrifugal HM 3 LVADs as compared
to the axial flow devices (e.g. the HM II has a device thrombosis rate
of ~10%). It nevertheless remains a more common complication in the
context of extracorporeal VADs. The extracorporeal tubing should be
examined on a daily basis for fibrin strands and thrombus formation
especially around the tubing connectors. If there is significant thrombus
formation, the external portion of the tubing should be exchanged to
avoid occlusion and/​or embolization. Heparin bolus of 5000 units IV
should be administered before clamping and dividing any lines. If the
patient has a BiVAD and the LVAD tubing requires exchange, the RVAD
flows should be weaned and decreased before clamping LVAD lines to
avoid pulmonary oedema and haemorrhage. If there is an RVAD that
requires exchange, the team should be prepared for significant reduc-
tion in LV preload and profound hypotension during the process. Hence,
the process of extracorporeal VAD and tubing exchange in any context
should be expedient.
As the HM II axial flow device has been almost completely superseded
by the centrifugal devices, haemolysis and thrombosis have become ex-
ceedingly rare but should be considered as a differential diagnosis with
high power consumption, rising lactate dehydrogenase, and plasma free
haemoglobin and/​or haemoglobinuria. In such instances, the patient should
be admitted to a specialized VAD centre as early as possible. CT angiog-
raphy should be performed to identify large thrombi or kinks in the outflow
graft. Unfortunately, a CT angiogram is frequently unable to definitively
detect pump thrombosis due to metallic artefact caused by the device.
Echocardiography and ‘ramp’ studies could aid diagnose pump thrombosis
as the aetiology of haemolysis if the pump is unable to decompress the
LV with incremental echocardiographically guided speed increases. If the
aforementioned tests confirm pump thrombosis, consideration should be
given for VAD exchange.
 SURGICAL VADs 75

Outflow graft obstruction

Due to the flexibility of the outflow graft, obstruction following insertion
is possible if the graft is not appropriately sized prior to being sewn (see
Chapter 5 for surgical technique). The HM 3 in particular has encoun-
tered problems with torsion of the outflow graft, which has since been
improved upon (additional lock added to prevent torsion). The clinical
manifestation of this complication would depend on the extent of ob-
struction to the flow. Pump power consumption is typically reduced (in
contrast to pump thrombosis, see previous paragraph). The clinical mani-
festation may range from being asymptomatic to having recurrence of
HF or even developing circulatory arrest in extreme cases of obstruc-
tion. Diagnosis can be made with a CT angiogram which would identify
the obstruction. Following this, the patient should return to the operating
room for corrective surgery. To avoid this complication in the first place,
the outflow graft should first be connected to the pump and the cardiac
apex returned to its natural position before pulling and measuring the
graft to reach the right pericardial dimple (pleuro-​pericardial junction) and
from there pulling and measuring the graft again to reach the suprasternal
notch. The excess Dacron should then be removed and the outflow graft
sewn to the ascending aorta. We have found that this technique has been
very effective to avoid graft length redundancy and subsequent kinking.
The lie of the graft should be along the right heart border and away from
the sternal edge. Utilization of percutaneous stents to remedy graft ob-
struction has also been described.
Insertion site infections (extracorporeal VADs)
Such infections are common in extracorporeal VADs as cannulas are fre-
quently in place for protracted periods of time. The dressings around
the cannulas should be changed with care and under aseptic conditions.
Repeated dressing changes can lead to excoriation and irritation and a
silicon, non-​stick dressing is preferred. When infection is encountered,
the most common offending microorganisms include coagulase-​negative
staphylococci, Staphylococcus aureus, and Pseudomonas species. Broad-​
spectrum antimicrobial therapy, followed by sensitivity-​guided, targeted
therapy should be commenced. Any concerns about deep-​seated infec-
tion should prompt CT scanning to delineate pockets of infection and
re-​exploratory surgery to debride or even relocate the cannulation
exit site. Ultimately, the most effective treatment consists of removal
of the infected cannula which can occur with ventricular recovery or
transplantation.
Air entrainment (durable and extracorporeal VADs)
Air can enter the circuit and lead to embolization and rarely air lock. Most
commonly, this occurs when the chest is open and the mechanism consists
of reduced intracardiac volume and creation of vacuum with subsequent
entrainment of air. Dislodgement of the inflow cannula can also create this
problem. As soon as this is identified, the system should be vented to re-
move air. Pump speed should be reduced to eliminate suction and in many
instances the pump may need to be stopped and the outflow clamped. If air
has entered the aorta, then venting of the aorta is required.
76

76 Chapter 7 Clinical trials related to LVADs

Percutaneous and minimally

invasive VADs
Complications such as bleeding, insertion site infection, hypoxaemia, GI
bleeding, thrombosis, and haemolysis and their managements are similar
to extracorporeal VADs (described in previous sections in this chapter).
Percutaneous and minimally invasive VADs can be utilized to support the
LV, the RV, or both. The percutaneous VADs propel blood forward with an
impeller or with centrifugal technology. The most commonly available per-
cutaneous LVAD devices and their characteristics are as follows:
• The Impella 2.5 (9 French (Fr) catheter and 12 Fr pump motor) can
flow up to 2.5 L/​min. It can be placed using a percutaneous approach
and has been used prophylactically in high-​risk percutaneous coronary
intervention (PCI).
• The Impella CP (9 Fr catheter and 14 Fr pump motor) can flow up to
3.5 L/​min. It can also be placed using a percutaneous approach.
• The Impella 5.0 (9 Fr catheter and 21 Fr pump motor) is a ‘minimally
invasive’ VAD requiring surgical cutdown onto the femoral or the
axillary artery for insertion. It can flow up to 5 L/​min. With similar
flow rates to Impella 5.0, the Impella LD (9 Fr catheter and 21 Fr
pump motor) can be utilized to support the left heart. However, it
is inserted directly into the ascending aorta with a Dacron chimney
graft. The Impella 5.0 has been utilized to treat cardiogenic shock for
up to 10 days, whereas Impella CP and 2.5 are primarily suited for
peri-​procedural support during high-​risk PCI and for LV venting and
decompression in the context of peripheral VA ECMO and poor LV
ejection.
• The newer Impella 5.5 (9 Fr catheter and 19 Fr pump motor) is inserted
in the same manner as the 5.0. See page 100 for more information.
Percutaneous RVAD devices include the Impella RP (11 Fr catheter and 22
Fr pump motor) and the ProtekDuo dual-​lumen catheter connected to the
TandemHeart centrifugal pump and are designed as RVAD support. They
are both inserted with fluoroscopic guidance. The Impella RP is inserted
through the femoral vein. The ProtekDuo catheter is most commonly in-
serted through the right internal jugular vein and is floated into the main PA.
Anticoagulation with unfractionated heparin with aPTT range of 60–​70
seconds is the goal. Serial TTE/​TOE assessments should be performed at
the time of insertion, following any repositioning, flow adjustments, and
following weaning and removal of the devices.
Mortality rate
These data are based on reported observational studies; the 30-​day mor-
tality rate while on the Impella device appears to range from 23% to 65%.
The reported mortality rate appears to be slightly higher while on the
Impella 5.0 than with the narrow-​calibre devices.
Major adverse cardiovascular events
The 30-​day risk of stroke has been reported to have ranged from 1.7% to
6.3% with the Impella devices. Impella 5.0 appears to convey a higher risk
of stroke.
 Conclusion 77

Bleeding
The 30-​ day observational rate of bleeding requiring an operation for
haemostasis has been reported to range from 2.6% to 43.8%, with the
Impella 5.0 appearing to have significantly higher bleeding risk as compared
to the Impella 2.5 perhaps due to the need for surgical cutdown to place
this device.
Haemolysis and thrombosis
These complications are most commonly encountered in narrower calibre
percutaneous devices (7.5–​10.3% for Impella 2.5 vs 6.3% for Impella 5.0)
due to the turbulence caused by the impeller. Following VAD implantation,
daily serum lactate dehydrogenase and plasma free haemoglobin should be
measured. Rising levels of such parameters with or without haemoglobin-
uria should alert the HF team of acute haemolysis. If there is thrombosis,
the TTE/​TOE can aid the diagnosis by identifying LV dilation or the pres-
ence of worsening functional MV regurgitation due to the device’s inability
to decompress the LV. If untreated, haemolysis can lead to renal failure
and the need for dialysis. When haemolysis with or without thrombosis
is suspected, the aPTT should be increased to 70–​80 seconds and the
Impella device should be removed emergently and the patient should be
supported with an alternative form of support, such as IABP (if tolerated)/​
VA ECMO/​extracorporeal LVAD.
Catheter migration
The Impella device catheter is positioned with TOE or fluoroscopic guid-
ance, with the inflow placed through the aortic valve and rested in the LV
outflow tract and the outflow orifice is positioned in the ascending aorta.
Catheter migration can lead to significant haemodynamic compromise and
haemolysis. Suspicion of catheter migration can be confirmed with a port-
able plain radiograph or TTE/​TOE. Repositioning also requires TOE or
fluoroscopic guidance.
Vascular injury and limb ischaemia
Incidence of major vascular injury can be as high as 17%. Impella catheters
can cause limb ischaemia by causing obstruction of blood flow distally
leading to limb ischaemia or compartment syndrome. Vascular injury (e.g.
arterial dissection) can also occur at the time of introduction or removal
of such catheters. Therefore, distal flow should be verified by means of
a Doppler ultrasound scan immediately following insertion or removal of
the device and documented. Inability to verify Doppler flow should prompt
emergent vascular intervention for example, relocation of the catheter,
possible angiography, and/​or surgical exploration of the site.

Conclusion
VAD complications are numerous and can be debilitating or fatal.
Anticipation of such complications and early intervention when complica-
tions arise is emphasized and may mitigate deleterious consequences.
78
 Chapter 8 79

Extracorporeal
membrane oxygenation
Clinical applications and indications 80
Maintenance of ECMO and troubleshooting 93
80

80 Chapter 8 Extracorporeal membrane oxygenation

Clinical applications and indications

The basic concept of ECMO is draining the deoxygenated blood from the
venous side, using a centrifugal pump to push the blood across a membrane
oxygenator, and then returning it to the body via a large-​bore cannula. The
oxygenated blood is returned to either an artery or a different venous site.
When evaluating patients for extracorporeal support, it is crucial to pro-
grammatically develop global inclusion and, more importantly, exclusion cri-
teria. Each patient going on ECMO should have an exit strategy (e.g. BTT,
bridge to recovery, etc.) in order to avoid a futile and prolonged ECMO
run (i.e. ‘bridge to nowhere’) which would create an ethical dilemma and
eventual withdrawal of support. Although exceptions can be argued for, it is
recommended that upper limits for age be established. Additional criteria to
consider include prior duration of mechanical ventilation particularly at high
inspiratory pressures, body mass index (BMI), and comorbidities, including
presence and chronicity of renal failure and incidence of multisystem organ
failure. Also, decisions should be made regarding whether or not it should
be offered to patients with advanced cancers.
ECMO can provide temporary mechanical support to facilitate surgery
on the heart, lung, and airway. Patients can be placed on venoarterial (VA)
ECMO to provide haemodynamic support during high-​risk catheter inter-
ventions. Airway surgeries, such as tracheal resection or airway tumour
ablation, can be supported on venovenous (VV) ECMO to obviate the
need for cross-​table ventilation. VV and VA ECMO can be used to provide
intraoperative mechanical support during lung transplantation.
Venovenous ECMO
In terms of respiratory ECMO, inclusion criteria include PaO2/​FiO2 ratio
<60 despite maximum ventilator support. Exclusion criteria can include
mechanical ventilation >7–​10 days, significant known neurological injury, or
profound multiorgan failure.
Before proceeding to ECMO, the critical care team may consider use of
inhaled pulmonary vasodilators, pharmacological paralysis, and ventilation
in prone positioning. Also, consultation from a high-​volume ECMO centre
may provide further expertise in ventilator and patient management that
may obviate the need for VV ECMO. VV ECMO provides an opportunity
to treat patients with refractory hypoxaemic and/​or hypercarbic respira-
tory failure. These are usually single-​system or lung-​specific aetiologies.
Common clinical indications include acute respiratory distress syndrome
(ARDS) resulting from viral or bacterial pneumonia, trauma, pulmonary
contusions, burn/​inhalation injury, reperfusion injury, as well as primary
graft dysfunction (PGD) after lung transplantation. Additionally, carefully
selected patients with significant gas exchange deficit, or profound hypox-
aemia can be supported on VV ECMO as a bridge to lung transplantation.
Configuration
There are three common VV ECMO configurations. Femoro-​femoral (‘fem-​
fem’, Fig. 8.1), femoro-​internal jugular (‘fem-​IJ’, Fig. 8.2), and dual lumen
(see Chapter 9). The least technically challenging approach is the standard
right IJ-​right femoral vein configuration due to the familiarity of most critical
care practitioners in central venous line placement. While image guidance
is strongly recommended throughout, fluoroscopy at the end is helpful to
 Clinical applications and indications 81

ensure that the tips of the cannulas are not in close proximity to mitigate
against ‘recirculation’. In recirculation, the oxygenated blood coming out
of the outflow cannula is immediately drained back into the inflow cannula
and pushed through the ECMO circuit, reducing the volume of oxygenated
blood reaching the patient’s vital organs. Benefits of the fem-​fem approach
include ease of placement and less risk of recirculation, but the IVC can be
crowded and it may be more technically challenging to place a left femoral
cannula with a right-​sided femoral cannula already in place.
The right IJ-​right femoral vein configuration can be modified to include
the left IJ or left subclavian vein site for outflow, and the left femoral vein
for inflow. Each of these represents unique technical challenges and, as
such, an experienced cannulator and image guidance with fluoroscopy is
helpful.

Fig. 8.1 Illustrating the fem-​fem VV ECMO configuration. Note the higher
position of the outflow cannula (the enlarged image) to minimize ‘recirculation’. TV,
tricuspid valve.
82

82 Chapter 8 Extracorporeal membrane oxygenation

Ao
PA

RA

LV
Outflow
RV

IVC

Air
O2 Blender

Pump

Inflow

Fig. 8.2 Illustrating the fem-​IJ configuration.

Venoarterial ECMO
VA ECMO provides haemodynamic support (cardiac replacement) in
addition to oxygenation and gas exchange. Common indications include
cardiogenic shock, cardiopulmonary arrest, massive pulmonary embolism,
post-​cardiotomy shock, and PGD after heart or lung transplantation.
Patients with end-​stage lung disease and severe right HF can be bridged to
lung transplantation on VA ECMO. As mentioned earlier VA ECMO can
be used to support or rescue patients during high-​risk procedures in the
catheterization laboratory (commonly known as the cath lab). It can also
be used to bridge patients to a more durable mechanical circulatory device
(see Chapter 4).
Configuration
VA ECMO can be configured as central or peripheral (Fig. 8.3). This re-
lates to the location of the cannulas and can offer varying levels of mech-
anical support and unloading of the LV. Central VA ECMO can be deployed
through direct cannulation of the aorta or indirect cannulation via a graft
 Clinical applications and indications 83

sewn onto the innominate or the subclavian artery, with venous drainage
directly from the RA or the vena cava.
Peripheral VA ECMO can be placed percutaneously into the femoral ar-
tery and obviates the need for an incision. It can also be deployed into the
axillary artery or the femoral artery with the cutdown technique. Venous
drainage usually comes from the femoral vein.
A hybrid upper body VA ECMO approach that promotes mobility has
been popularized by certain high-​volume programmes and is dubbed the
‘sport model’. In this approach, the axillary or innominate artery is cannu-
lated and venous drainage is obtained from an upper body site.

IVC

Abdominal
Ao

Inflow

Outflow

Pump
Fig. 8.3 Illustrating the peripheral VA ECMO configuration with a distal perfusion
cannula.
84

84 Chapter 8 Extracorporeal membrane oxygenation

VAV ECMO
In patients who are femorally cannulated for VA ECMO, there is a zone of
mixing in the thoracic aorta between the oxygenated ECMO blood flowing
in retrograde fashion up from the femoral artery and the patient’s natural
and suboptimally oxygenated (due to impaired lung function, e.g. ARDS/​
pulmonary oedema) circulation being ejected from the LV. The blood from
the LV perfuses the coronary arteries, and can also supply the innominate
artery, left carotid artery, and possibly the left subclavian artery depending
on the exact location of mixing. This may initially not be a problem if the
LV isn’t ejecting and the heart is being rested on VA ECMO or if the LV
is ejecting, but the lungs still are functional. However, if the LV starts to
recover in the setting of impaired lung function, then patients will experi-
ence differential hypoxaemia or North–​South syndrome (also known as
Harlequin syndrome). It is important that the upper body end-​organs and
the recovering heart not be subjected to prolonged hypoxaemia as it may
lead to permanent dependence upon MCS.
In severe instances of North–​South syndrome, the patient’s upper body
may appear cyanotic despite being on full femoral VA ECMO and having
normal lower body saturation or a post-​oxygenator PaO2 >300 mmHg.
A simple way to detect suspected differential hypoxaemia would be to
check a right radial arterial blood gas or place a pulse oximeter on the right
hand. If there is a significant difference between that and lower body satur-
ation, or if the saturation is surprisingly low, then the diagnosis of North–​
South syndrome should be considered.
The treatment for differential hypoxaemia includes placement of an upper
body venous return (outflow) cannula in a venoarteriovenous (VAV) configur-
ation (Fig. 8.4). In this configuration the oxygenated blood is split between the
femoral arterial cannula and the new upper body venous return cannula. With
this new venous return cannula, oxygenated blood would reach the right-​sided
circulation and then eventually get pumped across into the ascending aorta.
Usually 1–​2 L of flow down, aided by a Hoffman clamp and a flow meter,
the venous cannula is sufficient. One should avoid excessive outflow from the
venous return cannula especially in the context of right heart dysfunction. An
alternative treatment strategy for the North–​South syndrome would be re-
locating the femoral arterial cannula centrally to the aorta or to an upper body
site such as the axillary, subclavian, or innominate artery. These would all re-
quire surgical cannulation of the new vessel or a graft sewn onto the vessel.
An alternative to mitigating differential hypoxaemia includes jugular
venous drainage alone as inflow with a femoral arterial cannula as out-
flow. With this VA configuration, the oxygen-​poor blood of the RA/​SVC
junction is drained, but the IVC blood, which has been shown to carry an
oxygen saturation of around 80%, enters the right heart. This circuit con-
figuration is designed, firstly, to off-​set concerns that the new upper body
return cannula poses a risk of recirculation. And secondly, that there will be
less driving pressure at the secondary take off for the distal perfusion can-
nula (DPC) and thereby risking limb ischaemia. These risks can be mitigated
by precise placement of the femoral venous drainage line to avoid recircu-
lation and using a flow probe on the distal perfusion catheter in conjunction
with a secondary limb monitoring tool (see Distal perfusion cannula). This
configuration relies on SVC drainage alone and may not be sufficient for
patients who need full ECMO flows.
 Clinical applications and indications 85

Flow meter

ow
Hoffman Outfl
tubing
clamp

Abdominal
Ao

IVC

Outflow

Inflow

Pump
Fig. 8.4 Illustrating the VAV ECMO configuration with an upper body venous return
cannula.

Of note, some patients with an underlying respiratory issue may also

have profound sepsis and vasodilatory shock that may not be sufficiently
management with VV ECMO and pharmacotherapy alone. If a patient is
not responding to vasopressors, then VAV ECMO should be considered
to better support end-​organ perfusion. In these instances, the need for the
haemodynamic support is usually transient and the patient can often be
converted back to VV ECMO as the profound sepsis resolves.
Extracorporeal cardiopulmonary resuscitation (E-​CPR)
Clinical decision-​making
E-​CPR falls under a special subsection of VA ECMO. In this scenario, pa-
tients who suffer cardiorespiratory arrest are emergently placed on per-
ipheral VA ECMO during closed chest CPR or central VA ECMO while
open internal massage is taking place (e.g. post-​cardiotomy cardiac arrest).
86

86 Chapter 8 Extracorporeal membrane oxygenation

In well-​ selected patients, outcomes with E-​ CPR—​ including survival to

discharge—​are statistically better than conventional CPR. For example, a
patient under the age of 50, with a reversible aetiology for a witnessed
cardiac arrest, and <30 minutes of compressions before initiation of cannu-
lation. Patient selection criteria again needs to be programmatically decided
ahead of time.
When emergently evaluating potential patients, providers need to con-
sider age, cause of the arrest, underlying comorbidities, and duration and
quality of compressions (anecdotal tip: ‘age (years) +​duration of CPR
(min) <90’ in favour of E-​CPR). The availability and use of a compression
device can be very helpful in maintaining intrathoracic pressure until ECMO
is initiated.
Technical considerations
In a cardiac arrest situation, high-​quality CPR should be initiated immedi-
ately. If available, an automated compression device, should be utilized.
Appropriate and early use of an automated compression device can extend
the window of time between arrest and cannulation up to 60 minutes at
some high-​volume centres. While chest compressions are taking place, a
member of the cardiac arrest team should rapidly obtain fine-​bore arterial
and venous access.
Simultaneously, the ECMO team decides if the patient is appropriate for
E-​CPR. By running vascular access and clinical decision-​making pathways in
parallel, crucial time can be saved. This also allows for progress to be made
while the ECMO cart or an ECMO response team arrives at the bedside
with the full suite of cannulation supplies.
It should be understood that obtaining correct arterial and venous ac-
cess while chest compressions are in progress is technically challenging.
There is an increased risk of dual venous/​dual arterial access and sub-
sequent inadvertent VV/​arterioarterial ECMO cannulation with the per-
cutaneous technique; as such, we recommend surgical cutdown to the
femoral artery and the vein to more accurately identify these structures
before cannulation. Furthermore, any form of limited image guidance
(echocardiography or fluoroscopy) can be very helpful. If there is a brief
period of return of spontaneous circulation, then TTE can confirm wires
in the IVC and aorta to guide further cannulation. Along those lines, if
time and logistics allow, it may be helpful to place an X-​ray board under
the abdomen. Practically, it will mitigate against flexion at the waist or
hips, which can make placing large femoral drainage cannulas more chal-
lenging. And then it offers the possibility of a quick X-​ray should a tech-
nical difficulty arises. This becomes more important in large or morbidly
obese patients who can sink into hospital beds designed to prevent pres-
sure sores. If an X-​ray plate is not available, a CPR board can prevent hip
flexion.
In cardiac arrest situations, other practical cannulation tips to consider
are using a stiff wire, using smaller cannulas, and less frequent stages of dila-
tion. Due to anatomic considerations, the femoral venous drainage catheter
during E-​CPR should be preferentially placed into the right femoral vein if
it is free. Also, having cannulators with refined endovascular wire skills and
experience with standard ECMO cannulation is very helpful. They may be
better suited to respond quickly to technical difficulties.
 Clinical applications and indications 87

In E-​CPR, the priority should be initiating femoral VA ECMO exped-

itiously. Once the patient has been stabilized on ECMO then an image-​
guided percutaneous DPC or an open surgically placed catheter can be
considered. Some programmes may intentionally employ smaller arterial
cannulas in E-​CPR situations and elect to closely follow the leg for signs of
ischaemia.
Clinical areas for ECMO cannulation
Intensive care unit (ICU)
The ICU, especially one that is already familiar with taking care of ECMO
patients, is an excellent location for initial E-​CPR deployment. For using E-​
CPR in an ICU, one needs to establish E-​CPR protocols and then period-
ically rehearse them with the ICU teams. Practice and simulation are very
important to having a well-​functioning team. ICU providers should know
which patients in code situations broadly meet ECMO inclusion criteria
ahead of time.
Emergency department (ED)
In the ED, there should be a local ECMO chart and a primed ECMO circuit
ready for imminent use. As in ICU settings, the ED teams should practice
and drill simulated E-​CPR codes periodically. There should be an available
designated ECMO specialist who can help the ECMO team cannulate in
the ED. A decision should be made ahead of time regarding which practi-
tioner will undertake the cannulation, for example, designated providers in
the ED teams or the surgical team. Also, programmes should have a policy
regarding cannulating patients who have arrested prior to arriving at the
ED. This is particularly important because quality of compressions in the
field may vary and not every first response team has an automated com-
pression device. However, the combined use of an automated compres-
sion device and an established ED ECMO cannulation may go a long way
towards improving survival for patients who suffer out-​of-​hospital cardiac
arrest. Some institutions have a protocol where such patients are identified
en route to the hospital, akin to a ‘code STEMI’ patient, and they are dir-
ected to the cath lab or a hybrid operating room for emergent cannulation.
Operating room or cardiac cath lab
For patients who arrest in an operating room or in the cath lab suite before,
during, or after a procedure, rapid deployment of ECMO can be lifesaving.
This may be the optimal environment for a successful E-​CPR deployment.
The patient is already on the table; hence, the ergonomics are optimal.
Image guidance is readily available or can be available should the need arise.
In this scenario, open compressions (post cardiotomy) or closed compres-
sions can be started immediately and vascular access may already be es-
tablished and the ECMO supplies may be in close proximity. This logistical
optimization can dramatically decrease time to cannulation.
Stepdown unit
Stepdown unit deployment of ECMO should be undertaken initially in a
limited setting. Once the programme is comfortable deploying ECMO in
the ICU and the ED, then expansion to the regular hospital wards and the
stepdown unit is a reasonable next step. Again, ECMO carts should be lo-
cated at set point throughout the hospital where a team can bring it to a
8

88 Chapter 8 Extracorporeal membrane oxygenation

bedside within 10–​15 minutes of the code being called. In a stepdown unit
situation where patients may not be as closely observed, it is important
to know the period of ‘downtime’ prior to initiation of CPR. A witnessed
arrest has a significantly better prognosis than a patient found down during
periodic nursing rounds. Here patient selection can dramatically improve
outcomes.
Once protocols have been established and this resource is available to
hospital staff, the important thing will be careful patient selection. Strict
criteria need to be established ahead of time and adhered to. Practical chal-
lenges include who fields the ECMO/​E-​CPR consults, who activates the
ECMO team, and if there is an in-​house ECMO specialist available who can
come to the bedside during cannulation.
Technical considerations of ECMO cannulation
Once a determination has been made that a patient needs VV or VA ECMO
cannulation, then the decision has to be made regarding cannula types.
An ECMO cannulation strategy needs to be planned ahead of time. The
target vessels should be evaluated by ultrasound scanning in real time to
determine the vessel calibre as well as the presence of a clot or stenosis.
The size of the cannula should be based on the types of cannulas available,
the patient’s sex and body surface area (BSA), the projected needs of the
cannula, and the patient’s measured vessel size. The cannulator should be
able to easily convert between French size and millimetres (3:1).
Tip: a quick way to calculate a ballpark estimated BSA =​(weight(kg)/​
100) +​1.1.
Tip: a minimum cardiac output (L/​min) needed to be delivered =​BSA ×
2.4 (‘the ideal cardiac index’). This is a rough estimation of ECMO flow. In
most cases there will also be some native heart output, albeit inadequate.
Cannulation strategy
Central ECMO
Central cannulation should be in the remit of cardiac surgeons only as it in-
volves placing cannulas into the heart or sewing grafts onto the great vessels
coming out of the heart. The cannulation strategy will be based on surgeon
preference and likely related to the underlying cause of the need for ECMO.
Peripheral ECMO
Conversely, peripheral cannulation is now being performed by an expanding
field of cannulators all with varying wire and endovascular skills.
ECMO cannula types
Broadly, cannulas can be split into arterial and venous, but select cannulas
can have dual use. A cannula’s performance is determined by its diameter
(French size) and its length. As per Poiseuille’s and Ohm’s laws, a shorter,
wider cannula flows better, and cannula diameter is a more important de-
terminant of flow than its length. Before selecting a particular cannula, the
operator should familiarize themselves with the cannula’s published speci-
fications. This should be done with the understanding that the published
flows and fluid dynamics are in optimal settings, and often the data are based
off of using water instead of the more viscous blood. Usually, a good rule
of thumb is to subtract 0.25–​0.5 L/​min from the published flows to get the
 Clinical applications and indications 89

clinically anticipated flows. ECMO flows are most frequently limited by the
venous drainage cannula. Certain clinical scenarios require high degrees of
ECMO flows and so they may benefit from a second venous drainage line.
Arterial cannula
While they don’t have as much of an impact on flow rates as venous
drainage, an arterial cannula that is too small can lead to increased pressure
drops that will limit flow. That being said, not all VA ECMO patients need full
flow, and partial flow through a smaller cannula may be sufficient to address
the problem. When placing an arterial cannula, operators should ask the
question: what is the smallest cannula required to meet the patient’s needs?
Tip: peripheral arterial cannula flow capacity (L/​min) =​cannula size (Fr)
× 2/​10 (for central cannulation add +​1), for example, size 18 peripheral ar-
terial cannula can deliver 3.6 L/​min and when placed centrally, it can deliver
as much as 4.6 L/​min.
Aortic position
Aortic ECMO cannulas whether they are straight or angled are often CPB
cannulas that have been repurposed for prolonged ECMO utilization. They
have varying performance specifications based on their manufacturer, but
usually run in 1–​22 Fr sizes for adults. A common scenario is with post-​
cardiotomy cardiogenic shock (PCCS) where the patient is unable to come
off of bypass (see Chapter 11). Some surgeons will elect to transition the
patient to a VA ECMO circuit using the existing operative cannula config-
uration. Those patients will arrive in the ICU with an open chest and a
temporary cannulation strategy that will need to be reconfigured to a more
stable platform. Down the line when the patient has stabilized further and
the chest can be closed, the patient may be transitioned to a peripheral
configuration should ECMO still be required.
Axillary, subclavian, or innominate artery positions
These approaches allow for ambulation, physical therapy, and rehabilitation
all of which are needed during the bridging period.
Axillary artery position
Usually a 6 or 8 mm chimney graft (note: French to millimetre conversion
(3:1) to choose appropriately sized cannula for the chimney graft size) is
sewn onto the axillary artery with a bevel. In turn, the graft is cannulated
with a straight arterial cannula. The cannula can be tunnelled through the
soft tissue surrounding the pectoralis muscles or brought through the pri-
mary subclavicular incision. When planning this tunnel, the graft should be
examined with the retractors out and the arm in a natural position to en-
sure that there is no kinking. Extra care should be taken to limit the size of
the arteriotomy and to precisely angle the bevel. The axillary artery ap-
proach is associated with a risk of brachial plexus injury, over-​circulation to
the ipsilateral arm, and bleeding requiring re-​exploration.
Subclavian or innominate artery position
Technical approaches to the subclavian or innominate artery for central
ECMO cannulation have been described. Typically, the grafts are sewn onto
the artery and subsequently cannulated. Although a partial sternotomy may
be required for the initial innominate artery cannulation, decannulation can
be accomplished simply by ligating the graft.
90

90 Chapter 8 Extracorporeal membrane oxygenation

Femoral position
The femoral artery can be cannulated directly percutaneously or through an
open surgical approach. It can also be accessed indirectly through a surgical
chimney graft as described previously. Depending on the manufacturer, the
femoral cannulas often range from 14 to 20 Fr. They are often wire re-
inforced along part of their length to minimize kinking.
Distal perfusion cannula
It is recommended that placement of an antegrade DPC guidewire is
undertaken immediately after placement of the retrograde femoral wire
and before placement of the arterial cannula (with the exception of E-​CPR
scenarios), see Fig. 8.3 (p. 83). This would ensure ultrasonic visualization
and puncture of a filled superficial femoral artery with greater ease. Once
the femoral and venous cannulas are in place and VA ECMO run is initiated,
the DPC cannula is inserted over the previously placed guidewire. Backflow
should be observed through the DPC. The ECMO run should be paused
(maximum 15–​30 seconds), the tubing lines clamped, and the DPC ad-
equately de-​aired and connected to the circuit using the special DPC high-​
pressure tubing and connectors. It is recommended that a wire-​reinforced
DPC cannula is utilized. Kinking is an issue with non-​wire-​reinforced DPC
cannulas. We recommend that the non-​wire-​reinforced DPC is inserted at
the time of cutdown for arterial cannulation. This would allow superficial
femoral artery purse string placement facilitating removal of the DPC can-
nula and would allow the DPC to be advanced into the superficial femoral
artery run off all the way to the hilt of the sheath and buried in the wound
closure leaving the outflow portion out of the wound and connected to the
ECMO circuit. Burying the DPC cannula would eliminate any redundant
length of the DPC cannula between the skin and the vessel entry site, which
would be prone to kinking upon moving the patient’s hip. Pulse oximetry,
distal Dopplers, or near infrared spectroscopy (NIRS) can be used to con-
firm distal flow. If there are concerns after placement of the DPC, an angio-
gram can be performed through the catheter to confirm proper position.
Venous cannula
There are now a host of specially made, wire-​reinforced femoral venous
cannulas for ECMO. They are designed for the femoral position to prevent
kinking and collapse at higher pressures. They come as multistage and single
stage depending on the location of their fenestrations. The multistage can-
nulas have fenestrations at the tip and along the distal portion of the cannula
usually at 5 cm intervals. The single-​stage cannulas have fenestration at the
tip only; as such, they usually have a 2 cm tip that is not wire reinforced.
There are a variety of femoral venous cannulas on the market and each
has differing performance specifications depending on their manufacturer.
These cannulas usually come in two lengths: 38 and 55 cm. Both their length
and diameter should be sized to a patient’s BSA. Usually a 23 or 25 Fr
multistage cannula fits most average sized adults. It is our experience that
multistage cannulas provide better drainage than single-​stage ones. To maxi-
mize drainage, the cannula needs to terminate in the retro-​hepatic IVC. The
liver will help stent open the IVC and prevent it from collapsing around the
drainage cannula if/​when the patient is diuresed while on ECMO. An alter-
native to a second venous drainage cannula is using a larger cannula such
 Clinical applications and indications 91

as a 29 Fr cannula in circumstances where flows >6 L/​min are needed. It

should be noted that these larger cannulas may be more technically difficult
to place in the femoral position especially from the left side of a fem-​fem
VV ECMO configuration. In VV fem-​fem cannulation, once wires have been
placed up both femoral veins, it is recommended that the cannula be placed
first on the left side. This is due to the acute angle that the left common iliac
vein joins the right side for the IVC. The right iliac vein comes in at a new
straight trajectory and if there is already a stiff cannula in place on the right
it may be difficult to place a large multistage cannula up the left side.
Tip: venous cannula siphonage capacity (L/​min) =​cannula size (Fr)/​5,
that is, size 25 multistage venous cannula can drain up to 5 L/​min.
Dual-​lumen cannula
There are now several dual-​lumen cannulas on the market. They all have
a similar design, but slightly different flow dynamics. One lumen drains the
IVC and SVC, and the other returns the oxygenated blood into the RA and
across the tricuspid valve. They can be positioned in the right or left IJ vein,
and for experienced operators in the left subclavian vein. Due to their large
size and the importance of landing the drainage and return ports in the cor-
rect locations, image guidance is necessary. Care should be taken at time
of placement that the cannula is oriented correctly to ensure that the RA
opening is facing the tricuspid valve and not the side wall (see ­chapter 9 for
further description and illustrations).
Cannulation sequence
ECMO cannulation, by nature, is an urgent or emergent procedure and
is performed in patients in extremis. While the procedure has to be per-
formed precisely and expeditiously, there is very little margin for error.
Image guidance with ultrasound, TTE, TOE, or fluoroscopy is imperative.
There are six phases of ECMO cannulation according to the
Extracorporeal Life Support Organization (ELSO): (1) preparation, (2) vas-
cular access, (3) guidewire insertion and confirmation, (4) serial dilation,
(5) cannula insertion, and (6) circuit connection and initiation.
Step 1: preparation
The operator must ensure that the patient has adequate IV access for ad-
ministration of medications or blood products. The operator must ensure
that unfractionated heparin, blood products, and resuscitative fluids are
available, and that there is another clinical provider available to provide
critical care management and heparin administration so that the operator
can focus on the cannulation. All the supplies (the cannulas, wires, clamps,
as well as back-​up supplies must be available on the ECMO cart) must
be prepared. The patient must be positioned appropriately depending on
the intended vascular access site (e.g. shoulder roll placement to better
expose the subclavian region) and access to imaging devices (ultrasound/​
fluoroscopy) and the ECMO circuit. The equipment and the patient’s bed
should be rearranged so as to optimize ergonomics during vascular access
and cannulation. The operator should also plan ahead for potential back-​
up sites if he or she has unexpected difficulties with the primary cannula-
tion sites.
92

92 Chapter 8 Extracorporeal membrane oxygenation

Step 2: vascular access

Careful initial vascular access will facilitate later steps of dilation and can-
nulation. For example, if the vessel is accessed at too steep of an angle or
off-​centre then the wires are at higher risk of kinking or the side wall of
the vessel is more likely to be injured during dilation or cannulation. It is
recommended that the vessel should be accessed in two views using real-​
time ultrasound guidance at a 30° angle. Use of a soft-​tipped micropuncture
needle and a 0.14 microwire should be routine. The needle should enter the
vessels at 12 o’clock.
Step 3: guidewire insertion and confirmation
Ultrasound should confirm that the wire is in the vessel in both the short
and long axis. The guidewire should then be advanced to its target location
under image guidance. If an ultrasound-​only technique is being used, then
it is critical that the wire be seen in real time on echocardiographic (TOE/​
TTE) evaluation of the heart or descending aorta. Fluoroscopy is an ef-
fective alternative for guiding advancement of the guidewire as well as fur-
ther steps. Ideally, all the ECMO arterial, venous, and the DPC guidewires
should be in place before progressing to the next stage.
Step 4: serial dilatation
Once the location of the proximal and distal tip of the guidewire is con-
firmed, systemic heparinization (usually 5000 IU of IV unfractionated
heparin) is accomplished and then the operator can proceed with serial
dilation. It is important to recognize that only the soft tissue between the
skin and the vessel wall needs to be dilated and the dilator does not usu-
ally need to be inserted to its hub. Extra care must be taken to ensure
that the rail system stays straight and that the wire does not kink or form
S-​shapes as it is being manipulated. A taut rail system is the only way to
ensure that a vascular injury does not occur. In and out free movement
of the wire as dilatation is being performed also reassures the operator
that the wire is not kinked (especially in obese patients). For those less
experienced with endovascular skills, real-​time fluoroscopic guidance can
allow for the wire to be visualized along its entire length during cannula-
tion. It is also important to ensure that the wire length is maintained during
various exchanges of the dilators. When working on the artery, it is im-
portant to avoid overdilation as this can lead to vascular injury or cannula
site bleeding.
Tip: in the event of guidewire kinking during serial dilatation, the wire
should be carefully withdrawn to expose the kinked portion. Then the
smallest dilator is inserted past the kinked segment and into the vessel.
The kinked wire is then removed while leaving the dilator in place. A new
guidewire is then passed through the dilator and serial dilatation continued
as above.
Step 5: cannula insertion
After serial dilation, the pre-​selected cannula should be expeditiously
placed over the wire, while maintaining the stiff rail system. When inserting
the cannula, care should be taken to ensure that the inner cannula stylet
is not inadvertently displaced. Specifically, the stylet can get held up on
the skin edge or the soft tissue and lead to kinking of the wire or vascular
injury.
 Maintenance of ECMO and troubleshooting 93

Step 6: circuit connection and initiation

Once the cannula is placed, it should be flushed with sterile saline and
clamped off. The cannula should then be connected to the correct side
of the ECMO circuit using a wet-​to-​wet connection. Care should be taken
to ensure that the circuit tubing is free of air bubbles. Once both limbs of
the circuit are connected, ECMO should be initiated slowly. In parallel, the
ECMO cannulas should secured to the patient at multiple points to mitigate
against dislodgement. To minimize the risk of cannula site bleeding, a purse-​
string suture can be placed.
Arterial and venous decannulation
When cannulating the femoral artery, consideration should be given to plan
for the final decannulation strategy. That is, whether a percutaneous closure
device will be used or if the arteriotomy site will need open repair. If the
femoral artery is surgically cannulated then pursue-​string sutures may be
placed to both help secure the cannula with Rummel tourniquets and later
to close the puncture site. With a femoral artery cannula, care should be
taken to not over-​dilate the vessel and to ensure that the tapered portion
of the cannula is not sliding back and forth. In this scenario, the patient can
develop a large haematoma.
Venous cannula removal is generally more straightforward as this is a
low-​pressure system. Purse strings over the vein can be tied upon cannula
removal if cannulation had been undertaken with an open technique. For
decannulation of a venous cannula which was placed percutaneously (ax-
illary, subclavian, or femoral), a deep figure-​of-​eight or mattress skin stitch
usually suffices in achieving haemostasis.

Maintenance of ECMO and troubleshooting

Role of anticoagulation
Each programme should have a standardized anticoagulation protocol with
set targets for VV and VA ECMO that includes a standardized monitoring
regimen with activated clotting time (ACT), aPTT, or heparin level as well
as a backup anticoagulation drug for patients who have heparin-​induced
thrombocytopenia (HIT), such as direct thrombin inhibitors (e.g. bivalirudin
and chromogenic factor X assay measurements). HIT is a serious limb-​and
life-​threatening complication of heparin and should be suspected in any pa-
tient on heparin with thrombocytopenia. A ‘HIT screen’ would stratify the
likelihood of HIT in patients on heparin. Generally, aPTT could be main-
tained between 60 and 80 seconds. However, this can be down-​titrated
if there is bleeding. If ACT is used, it should be maintained around 150
seconds. It should be noted that if there is a concern for bleeding, holding
anticoagulation in patients on VV and VA ECMO is possible for short
periods of time. The tubing and connectors should be regularly examined
for thrombus formation and tubing exchanged if thrombus is identified.
Longer periods of heparin hold would risk thrombus formation in the cir-
cuit or the gas exchanger. Increased ‘delta P’ (pressure gradient across the
membrane oxygenator) to >60 mmHg (normally <50 mmHg), would raise
the possibility of gas exchanger thrombosis which leads to worse exchange
manifested on the patient’s arterial blood gases.
94

94 Chapter 8 Extracorporeal membrane oxygenation

Blood product transfusion

Blood product transfusion should be minimized to avoid alloimmunization
in BTT patients and to reduce the risk of transfusion-​related lung injury. We
recommend the threshold for red blood cell transfusion be haemoglobin
concentration <7 g/​dL. While on ECMO, platelets, fibrinogen, lactate de-
hydrogenase, and haemoglobin should be routinely checked and patterns
monitored.
Haemodynamic monitoring on ECMO
On VA ECMO support, thermodilution cardiac output determination is
erroneous. Frequent echocardiography is helpful to determine ventricular
functional recovery.
Recirculation in VV ECMO
In recirculation, the cannulas are configured in such a way that the oxygen-
ated return blood is immediately sucked back into the drainage cannula
and pushed back through the ECMO circuit. Very little oxygenated blood
actually makes it to the patient’s left-​sided circulation and, by extension, the
patient’s vital organs. This is diagnosed when the patient’s oxygen saturation
remains low or the systemic PaO2 is unusually lower than what the flows
would predict. On examination, the colour of both the oxygenated and
deoxygenated ECMO tubes is either bright red or the inflow tubing blood
flashes from dark to bright red. This indicates the proximity (<10 cm) of
outflow and the inflow cannula and warrants repositioning.
Impaired gas exchange on ECMO
Impaired gas exchange (hypoxaemia or hypercarbia) on ECMO indicates
one or a combination of the following factors: inadequate drainage or for-
ward flow (inflow or outflow cannula too small for patient size), incorrect
positioning of the inflow or the outflow cannula (e.g. recirculation, see
previous paragraph), gas exchanger problem (raised delta P, see ‘Role of
anticoagulation’ earlier in this topic), high cardiac output on VV ECMO
(blood preferentially ejected by the heart rather than drained into the in-
flow cannula), or differential hypoxaemia on VA ECMO (see p. 82). In any
case, efforts should be made to optimize oxygenation by correcting low
haemoglobin levels. Increasing ECMO flows as much as possible may im-
prove oxygenation and increasing sweep would reduce hypercarbia. If the
ceiling is achieved with these measures, mechanical ventilation of the lungs
can be optimized in order to oxygenate the blood that is ejected through
the pulmonary circulation by the heart (more commonly an issue in VV
ECMO). Finally, additional inflow or outflow cannulas (e.g. VAV ECMO in
fem-​fem VA ECMO, see pp. 82 and 84) can be inserted and connected to
the circuit to improve gas exchange.
Left ventricular distention and venting strategies
LV distention may occur on VA ECMO support and is associated with in-
creased pressurization of the pulmonary circulation and increased alveolar
fluid. LV distention should be addressed quickly to avoid injury to the LV and
to the lungs. LV distension leads to further irreversible myocardial damage
due to LV wall stretching and could lead to pulmonary haemorrhage and/​
or LV thrombus formation. When the LV fails to eject and LV distention is
 Maintenance of ECMO and troubleshooting 95

detected, inotropes should be initiated and with the goal of achieving at

least 20 mmHg of pulse pressure at all times. Should the patient fail to re-
spond, LV venting can be undertaken surgically, that is, via the right superior
pulmonary vein/​LV apex and then connected to the inflow cannula or per-
cutaneously using the Impella CP/​2.5 (Abiomed, Danvers, MA, USA).
Low mixed venous oxygen saturation (SvO2)
Low SvO2 indicates increased oxygen extraction while on ECMO sup-
port. This is commonly due to low-​flow states which can be mitigated by
improving the perfusion pressures by means of inotropes/​vasopressors
and ‘upsizing’ (adding further) inflow and/​or outflow cannulas. Other aeti-
ologies could include low haemoglobin concentration which can be recti-
fied with red blood cell transfusion and sepsis which prompt septic screen
followed by targeted aggressive IV antibiotics. Increased delta P (see ‘Role
of anticoagulation’ earlier in this topic) could indicate membrane gas ex-
changer malfunction.
ECMO pressure alarms
If flows drop off after initially being adequate then it can have one of sev-
eral causes:
• Intravascular fluid depletion. If the circuit, including the patient’s vascular
system, is volume down as would be seen with diuresis then the lines
will chatter or shake, and the pressures in the drainage cannula may
drop below –​100 mmHg (normal would be –​50 to –​80 mmHg). To fix
this, pump speeds can be decreased and volume administered to the
patient.
• Alarms could indicate transmembrane pressure elevations. Any
pressures >60 mmHg should raise the possibility of thrombus
formation in the gas exchanger.
• Alarms could indicate post-​membrane pressure elevation. This pressure
should ideally not exceed 300 mmHg; if so, this possibly indicates
kinking, displacement, thrombosis of the outflow cannula, and/​or
that the cannula is too small a calibre for the patient and the need for
upsizing.
Cannula site bleeding
The best way to avoid cannula site bleeding is to avoid over-​dilation of
the skin and vessel at the time of cannulation. For bleeding at a percutan-
eous venous cannula site, a correctly placed purse-​string suture at the skin
should be able to control the bleeding. For arterial site bleeding, the site
needs to first be assessed for haematoma formation which could become
a nidus of infection. Correct cannula positioning should be confirmed and
partial dislodgment ruled out. Significant ongoing arterial site bleeding may
warrant removal and repair of the site and relocation of the cannula to an
alternative site.
ECMO weaning
A patient can be weaned from VV ECMO when their sweep gas require-
ments have decreased to the point that a formal sweep gas trial can be
conducted. For this, the patient’s ventilator is adjusted from ‘rest’ setting to
lung protective ventilation strategy and the sweep gas is turned off. Most
96

96 Chapter 8 Extracorporeal membrane oxygenation

centres require at least 4 hours off of sweep before they proceed with VV
ECMO decannulation. Once the 4 hours is achieved with acceptable gas
exchange (at least PO2 >60 mmHg and PCO2 <40 mmHg and FiO2 ≤50%)
the patient can be decannulated.
VA ECMO weaning should always be undertaken using echocardio-
graphic guidance. Weaning may also be facilitated by Swan–​Ganz catheter
readings during reduced support (1–​2 L/​min). Prior to weaning, low-​dose
inotropic pharmacological support or even IABP support may be initiated.
After removal of peripheral arterial cannulas, care should be taken to
close the arteriotomy longitudinally to avoid narrowing. If the vessel is dis-
eased or if the arteriotomy is >50% of the vessel diameter, then one may
consider patch closure. With either surgical repair, attention must be paid
to tact down any loose intima. After arterial repair, distal pulses should be
checked with Dopplers in the operating room and postoperatively. If a graft
has been applied to the artery for cannulation, a stump of the graft is left on
the artery and oversewn.
 Chapter 9 97

Percutaneous mechanical
circulatory support
Intra-​aortic balloon pump 98
Intravascular ventricular assist system (iVAS, NuPulseCV) 99
Impella devices (Abiomed) 99
Dual-​lumen VV ECMO cannula 103
TandemLife platform (LivaNova) 104
98

98 Chapter 9 Percutaneous mechanical circulatory support

Intra-​aortic balloon pump

Description
An intra-​aortic balloon pump (IABP) is the simplest form of MCS utilized
commonly in both surgical and non-​surgical HF. The IABP catheter is most
commonly placed through the femoral artery. However, other insertion
sites such as the right and left subclavian arteries, and surgical placement dir-
ectly into the ascending aorta (with a chimney graft or direct) are possible.
Indications for IABP are as follows:
• Support in high-​risk PCI or cardiac surgery.
• Acute management of cardiogenic shock (e.g. post myocardial infarction).
• Management of end-​stage HF (e.g. dilated/​ischaemic cardiomyopathy).
• PCCS (see Chapter 11).
• Bridge to OHT.
• Refractory LV failure.
• Mechanical complications of acute myocardial infarction (ventricular
septal defect (VSD), papillary muscle dysfunction or rupture).
• Unstable angina refractory to medical management.
• Ischaemia-​induced ventricular arrhythmias.
Contraindications
• Aortic dissection.
• Severe aortic insufficiency.
• Extensive peripheral vascular disease.
Methods of insertion
In an elective setting, an IABP should be placed in the cath lab using TOE or
fluoroscopic guidance. However, it can also be inserted in more emergent
settings in the coronary care unit or the cardiac surgical ICU without real-​
time imaging. In such cases, catheter position should be confirmed with a
post hoc plain radiograph.
A femoral IABP is typically placed using the Seldinger technique with
or without its accompanying 8 Fr sheath. The guidewire is fed into the
descending aorta and the catheter tip is positioned distal to the take-​off of
the left subclavian artery.
An axillary IABP would allow for more liberal ambulation but is more
technically challenging and requires the following considerations:
• Fluoroscopic positioning is generally required (C-​arm, cath lab or hybrid
operating room).
• Cutdown is performed onto the right or left axillary with a placement
of a 6 mm side-​arm graft onto this vessel.
• The guidewire is fed through the graft, into the aortic arch and into
the descending aorta. Specialized angulated catheters may be required
to manoeuvre the wire preferentially into the descending aorta. The
IABP is then passed over the guidewire into the descending aorta. More
recently percutaneous technique has become common practice.
Complications
IABP complications are relatively common and include:
• Limb ischaemia (ranges from 0.9% to 31%).
• Catheter migration (ranges from 40% to 60% in the axillary position).
 Impella devices (Abiomed) 99

• Catheter balloon rupture is a rare (<2%) but serious complication

of IABP.
• Insertion site haematoma and infection (ranges from 8.6% to 40%).
• Formation of infective collection in type 1 diabetics (incidence as high as
37%).

Intravascular ventricular assist system

(iVAS, NuPulseCV)
The iVAS (NuPulseCV, Inc., Raleigh, NC, USA) is a form of long-​term
counterpulsation VAD. It would allow the patient to ambulate and even
be discharged from the hospital with the device. The components of the
device include the blood pump, internal driveline tunnelled under the fascia
and connected to a ‘skin interface device’, which in turn is connected to a
‘drive unit’. The insertion technique is very similar to the insertion of the
standard right axillary IABP (see p. 98) with the notable exception that the
iVAS is inserted from the left axillary artery. Hence, a snare from the fem-
oral artery would be required to guide the guidewire into the descending
aorta due to the acute angle of the left subclavian artery take off from the
aortic arch.
The iVAS is currently in the investigational stage and a feasibility clinical
trial is underway.

Impella devices (Abiomed)

Description
The Impella (Abiomed, Danvers, MA, USA) VAD is a microaxial blood
pump. Historically, the most common use of the Impella platform was for
prophylaxis in high-​risk PCI. However, more recently its utility has expanded
to a multitude of uses including the management of cardiogenic shock. It is
placed intravascularly in a retrograde fashion across the aortic valve with its
inflow into the LV and outflow in the ascending aorta (LVAD). In addition, a
right-​sided device (Impella RP) is placed in an antegrade fashion across the
pulmonary valve with its inflow at the RA/​IVC junction and outflow in the
main pulmonary artery (RVAD). The device can be used as a stand-​alone
support for cardiogenic shock or in combination with VA ECMO known as
‘ECPELLA’ where the Impella is utilized as an LV vent. However, it could be
utilized to continue to support the patient’s circulation once they have been
completed weaned off ECMO. Other indications for its use include high-​
risk cardiac surgery and bridging decompensated patients to OHT or LVAD.
For the bridge indication, prior to insertion of the device each patient is
assessed by a multidisciplinary team for candidacy for bridge to OHT or
durable VAD.
The most commonly available percutaneous LVAD devices and their
characteristics are as follows:
• The Impella 2.5 (9 Fr catheter and 12 Fr pump motor) can flow up to
2.5 L/​min.
01

100 Chapter 9 Percutaneous mechanical circulatory support

• The Impella CP (9 Fr catheter and 14 Fr pump motor) can flow up

to 3.5 L/​min. It can also be placed using a percutaneous approach. It
can be placed for high-​risk PCI or LV venting in the context of ECMO
support.
• The Impella 5.0 (9 Fr catheter and 21 Fr pump motor) is a ‘minimal
invasive’ VAD requiring surgical cutdown and sewing a 10 mm chimney
graft onto the femoral or the axillary artery for insertion. It can flow up
to 5 L/​min.
• The Impella LD (9 Fr catheter and 21 Fr pump motor) is not a
percutaneous device. It is centrally inserted through a typically 10 mm
chimney graft into the LV through the ascending aorta.
• Impella 5.5 (9 Fr catheter and 19 Fr pump motor) is a more recently
developed and popularized product with a goal of a longer-​term
temporary support. It is currently FDA licensed for 14 days of support
for cardiogenic shock patients as compared to currently approved 10
days for Impellas 5.0 and LD. However, the FDA is currently considering
a 30 day extension for patients supported on Impella 5.5. Like Impellas
5.0 and LD it is sited surgically and is capable of flowing as much as 6.2
L/​min. It is projected that the Impella 5.5 will supersede Impellas 5.0
and LD as a higher support percutaneous device in the near future as it
is able to be placed both peripherally and centrally. It has the advantage
of significantly lower risk of haemolysis, thrombosis as compared
to other Impella devices. The pump head portion in the 5.5 is more
pliable with a shorter motor segment hence easily inserted than the
Impella 5.0. Furthermore, as there is no pigtail at the tip of the device.
Hence, there is less risk of thrombus dislodgement in cases of apical LV
thrombus.
The Impella RP (11 Fr catheter and 22 Fr pump motor) is a percutaneous
RVAD device. It is inserted under fluoroscopic guidance. The Impella RP
is most commonly inserted through the femoral vein and across the pul-
monary valve.
Contraindications
LV thrombus, intractable ventricular tachycardia, severe aortic regurgita-
tion, severe aortic stenosis, mechanical aortic valve, and small artery ac-
cess size.
Methods of insertion
• Standard right or left axillary artery approach through subclavicular
exposure.
• Ideal size of the artery should be ≥6 mm and otherwise of good quality.
• Systemic heparinization (5000 IU of unfractionated heparin).
• A 10 mm Dacron graft is fashioned and anastomosed in an end-​to-​side
fashion.
• Graft is de-​aired and then tunnelled through a separate inferolateral stab
incision.
• Impella sheath is placed into the Dacron graft and secured (Fig. 9.1).
• A 260 cm 0.035“ guidewire is then passed through the sheath and into
the ascending aorta through the aortic valve and in the LV apex under
fluoroscopic and echocardiographic guidance (inlet should be 4–​4.5cm
from the aortic valve) with a 145° 6 Fr pigtail catheter (Fig. 9.2).
 Impella devices (Abiomed) 101

Subclavian
artery & vein

Fig. 9.1 Cutdown and placement of side-​arm graft on the right axillary/​subclavian
artery.

Output
Intake

Fig. 9.2 Optimal positioning across the aortic valve with the intake area placed
approximately 3.5 cm below the aortic valve and away from the anterior mitral valve
leaflet.
021

102 Chapter 9 Percutaneous mechanical circulatory support

• The pigtail catheter is then exchanged for a 0.018“ guidewire.

• The Impella device is then passed over the 0.018“ guidewire under
fluoroscopic and echocardiographic guidance.
• Care should be exercised to orient the device towards the LV apex and
away from the mitral valve apparatus.
• Performance ‘P’ level is commenced at P9 level, echocardiogram, and
haemodynamics are assessed.
Management
• Standard intensive care haemodynamic monitoring and lines including
arterial line and PAC.
• Inotropes and vasopressors are used as needed and titrated.
• Wean sedation, extubate, and ambulate as soon as possible.
• Aim for aPTT range of 50–​70 seconds.
• Daily monitoring of the axillary cutdown site for potential haematoma.
• Daily complete blood count and comprehensive metabolic panel.
• Daily plasma free haemoglobin and lactate dehydrogenase to assess for
haemolysis (malpositioned inflow portion towards the mitral valve may
lead to haemolysis).
• Daily serum creatinine (especially if haemolysis is suspected).
• SG catheter monitoring is helpful in determining total cardiac output
and effectiveness of unloading of left heart.
• Echocardiographically guided repositioning.
• If the position appears adequate on imaging, the patient’s ‘P’
(performance) level on the device may be turned down if tolerated.
• In the event of device malfunction, urgent replacement is performed by
exchanging the device to the opposite axillary artery site.
• The device can be weaned echocardiographically guided as the patient
haemodynamically improves.
Clinical outcomes
• A meta-​analysis of 163 patients implanted with the Impella 5.0/​LD
use in cardiogenic shock reported an overall survival to discharge was
73.5%. Among the highest group of cardiac recovery and survival were
those with post cardiotomy shock.
• In the authors’ experience of their first 100 axillary Impella 5.0
placements, of those patients who were listed for transplant prior to
insertion, 83.7% were successfully bridged to OHT.
• Among those patients in the bridge to durable device group, 60.9%
went on to receive a durable VAD and 78.6% survived to discharge.
• Survival was 64% overall, and 50%, 48%, and 81% for bridge to
recovery, bridge to durable VAD, and bridge to heart transplantation.
• Of those patients who were rescued with VA ECMO or biventricular
support, many were able to be transitioned to LV support with an
axillary Impella 5.0 with the added benefits including ease of removal
during time of OHT compared to durable VAD and ability to ambulate/​
rehabilitate patients.
Complications
Complications included stroke (10%), need for device exchange (13%), and
clinically significant haemolysis (18.6%).
 Dual-lumen VV ECMO cannula 103

Dual-​lumen VV ECMO cannula

Description
The dual-​ lumen VV ECMO cannula (e.g. the Avalon cannula (Avalon
Laboratories, Rancho Dominguez, CA, USA)) is most commonly inserted
through the right IJ vein. Alternative sites of insertion include the left IJ and
the left subclavian veins. It has two inflow ports draining blood from the
cavae to the oxygenator and one outflow port facing the tricuspid valve to
the RA (Fig. 9.3).
Methods of insertion
The dual-​lumen cannula is inserted with the Seldinger technique under C-​
arm fluoroscopic guidance.
• Commonly, the right IJ vein or the left subclavian vein is accessed with a
‘micropuncture’ needle.

SVC

IVC

Fig. 9.3 Schematic presentation of the dual-​lumen cannula (e.g. Avalon cannula) and
its position (note the position of the outflow orifices facing the tricuspid valve).
041

104 Chapter 9 Percutaneous mechanical circulatory support

• The guidewire is advanced well into the IVC.

• 5000 IU of unfractionated heparin is administered IV.
• Using the dilators (Avalon dilators for the Avalon cannula) the path into
the IJ or the subclavian veins is serially dilated.
• Common Avalon cannula sizes are 31 Fr (suitable for most average
sized males) can achieve up to 6 L/​min and 27 Fr (suitable for most
average sized females) can achieve up to 5 L/​min.
• The cannula is advanced over the wire (Fig. 9.3).
• The cannula position is confirmed on fluoroscopy with the cannula tip in
the IVC at the level of the diaphragm.
• The cannulas are primed and connected to the ECMO circuit and the
pump started.
• Look for differential colour change within cannulas and look for
immediate improvement in SpO2 and PO2 and reduction in PCO2.
• Affix cannulas firmly to prevent dislodgement and in a manner to allow
for ambulation.
Management
• Maintain aPTT at 50–​60 seconds.
• Obtain regular arterial blood gases to titrate pump speed to optimize
PO2 and sweep to optimize PCO2.
• Once pump speed and sweep are optimized, minimize mechanical
ventilator settings to minimize barotrauma and start weaning off
paralytics and sedatives.
• At high pump speeds, ‘mixing’ (recirculation of oxygenated blood into
the inflow cannula) may become an issue which would render the circuit
inefficient.
• Caution should be exercised in RV dysfunction.
• Ambulate with physiotherapy (see Chapter 19).

TandemLife platform (LivaNova)

The TandemLife system (LivaNova, Pittsburgh, PA, USA) is an extracor-
poreal circulatory assist device system which runs on the TandemHeart
pump. The system has both CE marked for 30 days and approved by the US
FDA for up to 6 hours of extracorporeal life support. The TandemHeart
pump is a continuous-​flow centrifugal pump that contains an impeller which
is supported by hydrodynamic bearing. It can be utilized for support in uni-​
or biventricular failure and for respiratory or cardiorespiratory failure. It can
be inserted percutaneously or through cutdown. The system has developed
into four main platforms:
• VA ECMO (TandemLife).
• Percutaneous LVAD (TandemHeart).
• Percutaneous RVAD (ProtekDuo).
• VV ECMO (TandemLung).
TandemLife VA ECMO
Description
TandemLife is a percutaneous VA ECMO system that provides a kit with
all the necessary components to allow a provider to initiate a patient on
 TANDEMLIFE PLATFORM (LivaNova) 105

VA-​ECMO. The kit includes (1) TandemHeart pump, (2) membrane oxy-
genator, (3) sterile priming basin, (4) ProtekSolo arterial cannula (15 or 17
Fr), (5) ProtekSolo venous cannula (24 Fr), and (6) venous dilators (14, 18,
and 22 Fr).
Methods of insertion
The TandemLife VA ECMO can be inserted percutaneously or through a
surgical cutdown. The femoral artery and vein are identified either under
ultrasound guidance or by direct visualization, and wire access is gained.
Heparin bolus (50–​100 units/​kg) is administered. Using the Seldinger tech-
nique, arterial and femoral venous cannulas are inserted. The sterile priming
basin is simultaneously used during this time to prime the TandemHeart
pump and the membrane oxygenator. Once primed, the venous and ar-
terial cannulas are connected to the pump via a wet-​to-​wet connection to
remove all air from the circuit. The pump is then turned on and the patient
is initiated on VA ECMO support with the speed adjusted to provide the
desired level of support.
Management
The management of the TandemLife-​supported patient can be broken up
into two aspects: (1) the system and (2) the patient.
The system
The TandemLife system is managed very similarly to any centrifugal pump
system with one notable difference which is the ‘purge line’. The ‘purge
line’ flows saline at 10 mL/​hour through the motor chamber and com-
pletely surrounds the rotor towards the impeller to provide radial stability.
Heparinized saline (90,000 U/​L) flows around the impeller shaft–​seal inter-
face to flush the area, preventing thrombus formation. The pump rotates
between 3000 and 7500 revolutions per minute (rpm) to provide up to 8
L/​min of flow. When initiating the device, increase the rpm until the desired
flow is achieved. The target ACT goal is 250 seconds which is achieved with
systemic anticoagulation with heparin.
The patient
Management of a patient supported on the TandemLife system is identical
to any patient support on VA ECMO. The patient needs to be adequately
anticoagulated. Pump flows must be adequate for end-​organ function re-
covery. This is achieved with close laboratory blood work and haemo-
dynamic monitoring. Right arterial lines should be placed on all peripherally
cannulated VA ECMO patients to avoid missing cerebral hypoxia as a re-
sult of differential blood perfusion to the right and left carotid. If cerebral
hypoxia is noted, then the patient would require either a placement of
an oxygenated venous return cannula to create a VAV ECMO circuit or
conversion to central cannulation. Pulsatility (pulse pressure ≥20 mmHg)
should be ensured (with inotropes) in peripheral cannulation for VA ECMO
and frequent echocardiography is required to monitor for LV distension.
If there is LV distension, LV venting can be accomplished by either open
surgical LV vent placement or peripheral Impella device placement. When
the LV recovery occurs, the next steps are to ensure adequate ventilatory
function and haemodynamics on minimal inotropic support before weaning
can begin. Weaning can be performed by decreasing the pump speeds until
the flows are decreased by 0.5–​1 L/​min per day until the system flow is 2
061

106 Chapter 9 Percutaneous mechanical circulatory support

L/​min. Prior to pump removal a final echocardiogram is performed at 2

L/​min. When adequate LV function is demonstrated, the patient can be
weaned off fully and decannulated.
TandemHeart LVAD
Description
The TandemHeart system constitutes a temporary LA to aorta LVAD cir-
cuit. It utilizes a transseptal cannula (62 or 72 cm) that draws oxygenated
blood from the LA and delivers blood to the body via the femoral artery
or axillary providing a LA-​to-​femoral/​axillary bypass. The TandemHeart
system improves haemodynamic performance and increases end-​organ
perfusion for patients with LV failure.
Methods of insertion
The TandemHeart system can be inserted percutaneously using TOE and
fluoroscopic guidance. The femoral vein is accessed by the Seldinger tech-
nique and the guidewire is passed into the SVC. Under imaging guidance,
transseptal puncture is then achieved by using the Brockenbrough needle
and a SL1 sheath. Once the transseptal puncture is made, heparin is admin-
istered to achieve a target ACT >250 seconds. Once the sheath is advanced

SVC

IVC
Fig. 9.4 The cannula position in the ProtekDuo.
 TANDEMLIFE PLATFORM (LivaNova) 107

in the LA, a stiff wire (0.035“ Amplatz) is placed into the LA. The sheath is
then removed, and the two-​stage dilator is advanced into the LA. The di-
lator is then removed, and the tip of the 21 Fr cannula is left in the LA. The
cannula tip contains three radiopaque marker discs to aid confirm the pos-
ition of the cannula tip in the LA. TOE is also used to confirm placement.
The femoral artery is then accessed percutaneously using the Seldinger
technique. A 15 or 17 Fr ProtekSolo arterial cannula is inserted. The pump
and the cannulas are de-​aired and are connected via a wet-​to-​wet connec-
tion to remove all air from the circuit. The pump is then turned on and the
pump speed is adjusted according to the patient’s needs.
The TandemHeart system can be placed in the axillary position to facili-
tate patient ambulation. The axillary artery and vein are exposed, and an
8 mm graft side-​arm is sewn to the axillary artery. The arterial cannula is
tunnelled through the skin and then placed within the graft and secured in
place with a heavy tie. The pump inflow cannula is then tunnelled through
the axillary vein and into the LA through direct visualization via a right
atriotomy. The advantage of the axillo-​axillary approach is that the patient
can be ambulatory. This axillo-​axillary approach describes an ambulatory
LVAD strategy. However, if the inflow cannula is placed in the RA and a
TandemLung oxygenator is added to the circuit, then the system consists of
ambulatory VA ECMO.
Management
Adequate anticoagulation should be maintained on the TandemHeart
(aPTT range: 60–​80 seconds). Pump flows must be adequate for end-​organ
function recovery and optimum haemodynamics. Frequent echocardiog-
raphy to monitor for LV recovery is required. Vigilant monitoring of SpO2
is critical since severe life-​threatening hypoxia can develop if the transseptal
cannula is pulled back into the RA. Pump flow is determined by the patient’s
haemodynamic requirements. A PAC and an arterial line are essential tools
of patient management and are extremely useful for weaning. Weaning can
be initiated once the echocardiogram shows good LV function with the pa-
tient maintaining good haemodynamics on little inotropic or pressor sup-
port. The usual method of weaning is by 0.5–​1 L/​min every 12–​24 hours.
TandemLung and ProtekDuo
Description
The TandemLung system comprises the Tandem pump, ProtekDuo can-
nula, and the TandemLung oxygenator. The ProtekDuo VV cannula set is
intended for use as a dual-​lumen single cannula for the RA venous drainage
and PA reinfusion of blood. The combination of the ProtekDuo and the
Tandem pump provides RV support making the ProtekDuo system a tem-
porary RVAD (Fig. 9.4). The TandemLung oxygenator is intended for use
in an extracorporeal circuit making the TandemLung a VV ECMO system.
The ProtekDuo is placed percutaneously via the right IJ vein under fluoro-
scopic guidance and comes in two different sizes: 29 Fr and 31 Fr. The 29
Fr ProtekDuo cannula is a wire-​reinforced dual lumen with a 29 Fr proximal
lumen draining deoxygenated blood from the RA to the pump while the 16
Fr distal lumen returns the deoxygenated blood to the PA to be oxygenated
by the lungs. The 31 Fr ProtekDuo has a 31 Fr proximal drainage lumen and
a 19 Fr distal lumen. Maximum flow across these cannulas are 4.5 L/​min
and 5 L/​min respectively.
081

108 Chapter 9 Percutaneous mechanical circulatory support

The ProtekDuo cannula provides percutaneous ambulatory RVAD

support which makes it unique as compared to the CentriMag which re-
quires an open approach, while the Impella RP despite being a percutan-
eous device is implanted femorally, impeding ambulation. In the VV ECMO
arena the ProtekDuo is the superior platform. Compared to the standard
two-​cannula (‘fem-​IJ’) approach, the ProtekDuo, being a dual-​lumen can-
nula, requires only one neck access which allows the patient to potentially
ambulate.
Methods of insertion
The right IJ vein is accessed using the Seldinger technique and 8 Fr sheath
is placed. Then a 0.035“ compatible balloon-​tipped wedge catheter is ad-
vanced into the main PA and placed in either the left PA or the right PA. The
0.035“ Amplatz super stiff wire is exchanged for the catheter and inserted
into the PA and the sheath is removed. Systemic heparinization is then en-
sued for a target ACT >250 seconds. The path is serially dilated with the
Tandem dilator kit and the ProtekDuo cannula is advanced under fluoro-
scopic guidance into the PA ensuring the distal tip is positioned in the main
PA and the proximal drainage holes in the RA. Then the proximal (inflow-​
RA) and the distal (outflow-​PA) tubes are connected to the Tandem pump.
RVAD support is then initiated and progressively increased (as needed) to
a maximum flow of 4.5 or 5 L/​min depending on the calibre of the cannula
used. If an oxygenator is required, this can be spliced into the circuit.
Management
The TandemLung system is unique as these patients are simultaneously on
VV ECMO and RVAD support (i.e. oxygenated RVAD). The following con-
siderations apply:
• LV function prior and during support must be frequently assessed to
avoid acute LV failure and/​or pulmonary oedema due to overflow
(caution in poor LV function).
• All patients require judicious anticoagulation and ACT monitoring
(range: 160–​180 seconds).
• Arterial blood gas monitoring is crucial and will determine the level of
CO2 sweep, FiO2, and flow needed from the system.
• Weaning can be initiated under echocardiographic guidance once
adequate RV function and oxygenation is achieved.
• Low flows can result in pump thrombosis. If the pump flow is reduced
to 2 L/​min or less, consider higher anticoagulation targets.
 Chapter 10 109

The total artificial heart

Introduction 110
Indications for total artificial heart 110
Assessing for fit (70 cc vs 50 cc) 111
Preoperative and intraoperative management 111
Postoperative management 113
Postoperative complications 114
Outcomes 114
10

110 Chapter 10 The total artificial heart

Introduction
The SynCardia total artificial heart (TAH) (SynCardia Systems, Tucson, AZ,
USA) is a biventricular, implantable, pneumatically driven, pulsatile device
that orthotopically replaces both ventricles and all four valves (Fig. 10.1):
• The polyurethane-​lined ventricles contain a four-​layered diaphragm
creating air and blood separation.
• Two tunnelled drivelines deliver compressed air.
• A maximal stroke volume of 70 mL results in a cardiac output of up to
9 L/​min.

Fig. 10.1 SynCardia System’s 70 cc TAH.

Indications for total artificial heart

The TAH is indicated as BTT for advanced biventricular HF that is worsening
despite maximal medical and surgical therapies.1 More than 75% of patients
in the INTERMACS registry who have had TAHs implanted were profiles 1
or 2. The list of indications for TAH is lengthy:
• Irreversible biventricular failure (central venous pressure (CVP)
>18 cmH2O, RVEF <20%), that is unlikely to be sufficiently supported
with a LVAD alone.
• Severe allograft failure, rejection, or heart transplant vasculopathy with
cardiogenic shock.
• LVAD support with decompensated RV failure.
• Recurrent ventricular tachycardia/​ventricular fibrillation.
• Hypertrophic, restrictive, infiltrative cardiomyopathy (i.e. small, non-​
dilated ventricles).
• Postinfarct VSD which cannot be repaired and ventricular failure.
• End-​stage congenital heart disease.
• Cardiac tumour which requires resection of significant LV and RV.

1 The TAH is currently approved for BTT; clinical trials for use as destination therapy are ongoing.
 Preoperative and intraoperative management 111

• The most common indications in decreasing order of incidence

are idiopathic dilated cardiomyopathy, ischaemic cardiomyopathy,
congenital/​genetic conditions, post-​transplant graft failure, valvular
cardiomyopathy, restrictive cardiomyopathy, and LVAD failure (including
right HF).

Assessing for fit (70 cc vs 50 cc)

Obtain a CT scan of the chest and identify the anteroposterior distance
from sternum to T10 vertebrae. If this is >10 cm, consider implanting the
70 cc TAH.

Preoperative and intraoperative

management
• Continue preoperative management of cardiogenic shock and
multisystem organ failure.
• Monitor routine arterial pressure and CVP; remove PAC.
• These catheters can be exchanged if infection is a concern.
• Perform TOE preoperatively and perioperatively to assess venae cava
compression, pulmonary vein compression, or both.
• Following chest closure, the device may shift and cause compression
of the IVC, LA, or both. Findings of caval obstruction may include
narrowing at the caval atrial junction; turbulent flow past the narrowing,
into the RA, or both; decreased RA size; and flow acceleration across
the stenotic portion.
• LA compression with resultant pulmonary venous obstruction can
occur. Normal pulmonary vein velocities are in the range of 0.4 to 0.7
m/​sec. The resulting pulmonary vein obstruction may manifest as high
velocities of > 1.0 m/​sec in systole or diastole.
Steps to implantation
• Monitor arterial blood pressure, CVP, pulse oximetry, temperature, and
TOE.
• Prior to heparinization, prepare the atrial cuff and arterial grafts and
tunnel drivelines:
• Atrial cuffs:
• Cut the quick connects to the atrial cuffs to 3–​5 mm.
• Arterial grafts:
• Use a surgical sealant to cover the conduits; other techniques
include exposing them to the patient’s unheparinized blood and
allowing to dry over 5 minutes in a stretched position.
• Repeat the process 3 times to preclot.
• Heparinized blood can be used with a combination of protamine
and thrombin to preclot.
• Driveline tunnelling:
• Make an incision in the left midclavicular line 5–​10 cm below the
costal margin and create a tunnel from the subcutaneous layer into
the anterior mediastinum.
12

112 Chapter 10 The total artificial heart

• Place the driveline at the end of a 40 Fr chest tube, and pass it

through the tunnel.
• Tunnel the right-​sided driveline in a similar manner.
• Wrap the LV and RV in antibiotic-​soaked lap sponges and position
laterally away from the wound.
Recipient cardiectomy
Heparinize, and proceed with ascending aortic and bicaval cannulation
technique:
• Place umbilical tapes around the cavae.
• Limit dissection around the aorta and PA for future transplantation.
• Initiate total CPB and snare caval tapes.
• With a goal of preserving the tricuspid and mitral annuli but not the
valves, make an incision on the RV side 1 cm below the atrioventricular
(AV) groove and carry it towards the RV outflow tract proximal to the
pulmonary valve; posteriorly, bring the incision to the intraventricular
septum and across on the LV side of the AV groove, preserving the
mitral annulus.
• Transect the great vessels just proximal to their respective valves.
• Trim excess muscle and chordae, and leave 2 mm of ventricular muscle
around each annulus.
• Oversew the coronary sinus to prevent back bleeding through
transected vessels; look for a PFO and oversew if encountered.
Atrial quick connect and graft attachment
• Invert the quick connect and place it into the LA cuff.
• Using 3-​0 polypropylene (MH), sew the outer, atrial cuff to the quick-​
connect cuff in a running continuous fashion.
• Perform the right quick-​connect anastomosis in a similar fashion.
• Evert both quick connects.
Testing the anastomosis for haemostasis
• Place the plastic testers that are fitted to the quick connects in place
starting on the left side.
• Have the assistant surgeon compress the right and left pulmonary veins
behind the LA as the surgeon injects a saline-​blood admixture with a 60
mL syringe connected to a three-​way stopcock.
• Assess the suture lines for leak and repaired as needed.
Great vessel anastomosis
• Measure the PA and aorta with the prosthetic ventricles brought into
the field to mimic their natural placement.
• Cut the arterial grafts and then anastomose with a 4-​0 polypropylene
suture in a running continuous fashion.
• Take care to preserve as much length of native great vessels as possible
just above the commissures.
• Pressure test the aortic and pulmonic anastomosis while cross-​clamping
the PA.
 Postoperative management 113

Postoperative management
• Monitor arterial blood pressure via arterial line (the waveform will be
pulsatile).
• Precisely position the CVP monitor at the innominate–​SVC junction to
not impinge on the mechanical valve inflow via the RA.
• Monitor pulse oximetry.
• Have a low threshold for temporary chest closure given the propensity
for postoperative mediastinal bleeding.
• At the time of chest closure, perform TOE to assess pulmonary vein
compression, left bronchus IVC compression, or both.
• Provide standard postoperative mechanical ventilatory support and
sedation with early extubation.
Management of preload
Patients may be hypertensive or vasoplegic when weaning from CPB:
• The use of vasodilators (rapid onset and short acting) including
nitroprusside and nicardipine may be necessary.
• Nitroglycerine should be used with caution because of its greater
degree of venodilation and decrease in preload.
• Treat vasoplegia with norepinephrine and vasopressin.
Management of volume status
Assess for signs of hypovolaemia, including decreased systemic blood pres-
sure, decreased CVP, low fill volumes, metabolic acidosis, and elevated
serum lactate levels:
• Caution should be exercised when assessing the patient’s volume status
because changes can be made to the device settings that may alter
cardiac output to normal values.
• Sudden changes to device fills, flows, and output may indicate an acute
driveline or inflow obstruction.
• Atrial tamponade may manifest as decreased flow, fill volumes, or
cardiac output and elevated CVP.
• Pulmonary vein obstruction is also a possible cause manifesting as
pulmonary venous congestion, pulmonary oedema, and hypoxia:
• Promptly evaluate all such cases with TOE.
• Perform expeditious surgical re-​exploration and relief of obstruction
if necessary.
Anticoagulation
Anticoagulation practices are typically institutional specific but, in general,
utilize an antithrombotic approach including antiplatelet and anticoagulant
agents:
• Most commonly initiate with unfractionated heparin at 2–​5 units/​kg per
hour following chest closure when there is minimal chest tube drainage:
• Target an aPTT value of 40–​60 seconds or heparin levels from 0.11
to 0.27 units/​Ml.
• TEG is useful in the early postoperative period:
• The coagulation index should be maintained at <1.2.
• TEG with platelet mapping should also be used to keep the maximum
amplitude (arachidonic acid, adenosine diphosphate) at <50 mm.
14

114 Chapter 10 The total artificial heart

• Begin antiplatelet agents early if the platelet count is >100,000 × 106/​L.

• Prescribe aspirin at 81 mg/​day up to 325 mg/​day with TEG guidance.
• Institute dipyridamole 50 mg every 8 hours up to 400 mg every 6 hours
if necessary, as directed by TEG platelet mapping.
• Begin warfarin once the patient’s condition is haemostatic and end-​
organ recovery is achieved:
• Target a goal INR of 2.0–​3.0.
• If a high degree of haemolysis is evident, pentoxifylline may be used at
200–​500 mg every 8–​12 hours:
• This may be initiated early in the postoperative period.
• HIT is a potential risk in these critically ill patients:
• Use direct thrombin inhibitors such as argatroban or bivalirudin.

Postoperative complications
Renal dysfunction
Renal dysfunction after TAH implantation occurs in up to one-​third of pa-
tients. Proposed mechanisms include the lack of BNP and its compensatory
mechanism due to removal of the ventricles.
• IV infusion of nesiritide was previously utilized to maintain urine output
and renal function but is no longer available.
• Intermittent ultrafiltration is the current standard of practice.
Postoperative anaemia
The aetiology of immediate postoperative anaemia is multifactorial.
Bleeding in the first few days postoperatively requires mediastinal explor-
ation in nearly a quarter of patients. A small proportion of patients have
GI bleeding. Inflammation has been proposed as another cause of anaemia
after TAH, as is evidenced by elevated levels of inflammatory biomarkers.
Another haematological finding is haemolysis, demonstrated in laboratory
findings as decreased haptoglobin, increased lactate dehydrogenase, and in-
creased plasma-​free haemoglobin.
• Target a haemoglobin level as low as 6 g/​dL (a common threshold for
transfusion and tolerated by many patients).
• Avoid transfusion as much as possible given the risk of anti-​HLA
antibody formation in patients being bridged to transplantation.

Outcomes
A recent review of INTERMACS data provides an overview of the TAH
population, factors associated with survival, and rates of adverse events.
Overview
Patients with TAHs were found to be much sicker when compared to pa-
tients receiving an LVAD, with the majority being INTERMACS 1 and 2. RV
failure was present in 82% of patients implanted. Before implantation, 11%
of patients required haemodialysis; an additional 29% required haemodi-
alysis after TAH implantation.
 Outcomes 115

Survival
The SynCardia TAH paired with the Freedom Driver (Fig. 10.2) allows pa-
tients to be discharged from the hospital (new drivers are also being de-
signed); 24% of patients in the INTERMACS registry who were implanted
with the TAH were discharged (91% to home, 4% to a rehabilitation setting,
5% to other). The median length of stay from implant to discharge was
1.6 months.
The most commonly identified causes of death were multisystem organ
failure (36%), neurological injury (18%), and elective withdrawal (12%).
Significant risk factors for early mortality included older age and the need
for dialysis before implantation. Mortality was associated with the patient’s
medical condition (poor nutrition, renal dysfunction), as well as manage-
ment. Risk factors associated with late mortality were high creatinine and
low serum albumin levels.
Centre volume was also associated with mortality rates as well as heart
transplant rates. Centres implanting ten or more TAHs had a 12-​month
survival of 64.8%, compared with 36.7% in those centres implanting fewer
than ten TAHs. Similarly, transplant rates within the 12 months after TAH
implantation were higher in the higher-​volume centres: 58.4% in the centres
implanting ten or more TAHs, versus 43% for low-​volume centres. Overall
survival at experienced centres at 12 months was approximately 73%,
transplanted plus still on device.
Adverse events
The most common adverse events in the early phase after TAH implant-
ation (<3 months) were bleeding and infection; in the later phase (within
6 months), infection and minor device malfunction were most prevalent
(Table 10.1).

Fig. 10.2 SynCardia System’s TAH Freedom Driver, portable driver.

16

116 Chapter 10 The total artificial heart

Table 10.1 Adverse events occurring in TAH patients

Event Earlya Late

Events Rateb Events Rateb p-​valuec
Thromboembolism:
Venous 17 1.7 3 0.2 0.0001
Arterial non-​CNS 20 2.0 2 0.1 <0.0001
Bleeding 414 41.3 96 7.1 <0.0001
Device malfunction:
Major 13 1.3 34 2. 5 0.04
Minor 50 5.0 203 14.9 <0.0001
Pump thrombus 4 0.4 3 0.2 0.4
Hepatic dysfunction 52 5.2 11 0.8 <0.0001
Infection 389 38.8 167 12.3 <0.0001
Neurological 148 14.7 40 2.9 <0.0001
dysfunction
Pericardial drainage 63 6.3 1 0.1 <0.0001
Renal dysfunction 162 16.1 21 1.5 <0.0001
Respiratory failure 219 21.8 38 2.8 <0.0001
a
Early indicates ≤3 months of device implantation. Late indicates >3 months after device
implantation.
b
Rates are reported per 100 patient-​months.
c
The p-​values compare early and late rates.
 Chapter 11 117

Post-​cardiac surgery
cardiogenic shock
Introduction 118
Definitions 118
Management of post-​cardiotomy cardiogenic shock 118
Step-​wise approach to the high-​risk patient 119
Clinical considerations for mechanical circulatory support 122
Outcomes and adverse prognostic indicators 124
18

118 Chapter 11 Post-cardiac surgery cardiogenic shock

Introduction
Post-​cardiotomy cardiogenic shock (PCCS) occurs after about 2–​6% of
all adult cardiac operations. While most patients are salvaged and main-
tained on maximal inotropic support and IABP leading to recovery of the
myocardium, around 1% do not respond to such initial therapies and decline
rapidly with worsening cardiogenic shock (‘refractory PCCS’). In a select
cohort of patients, advanced MCS such as VA ECMO and short-​term VADs
have been utilized with variable success over the years.
In this chapter, our focus will be placed on discussions surrounding man-
agement of ‘refractory PCCS’ in the adult age group with short-​term MCS.
Management of cardiogenic shock following OHT is covered in Chapter 17.

Definitions
The normal range of CI in healthy adults is 2.5–​4.5 L/​m2/​min. PCCS is de-
fined as systolic blood pressure persistently <90 mmHg and MAP 30 mmHg
below baseline in tandem with CI <1.8 L/​m2/​min without haemodynamic
support and <2.0 L/​m2/​min with support while filling pressure should be
adequate or elevated (LVEDP >18 mmHg and RVEDP >10 mmHg).
While all aforementioned values could be measured by means of a PA
catheter, the clinical features of PCCS are manifested by end-​organ dys-
function in the form of oliguria, cool extremities, increasing serum lactate,
and worsening metabolic acidosis. PCCS becomes ‘refractory’ when the
worsening haemodynamic and metabolic parameters do not improve with
high doses of inotropic support and IABP and the patient demonstrates
progressive decline leading to multiorgan failure and demise.

Management of post-​cardiotomy
cardiogenic shock
Early shock
The most common scenario is acute PCCS (within minutes) upon separ-
ation from CPB (‘early PCCS’) with progressive decline in haemodynamic
parameters as mentioned previously and worsening metabolic acidosis.
Prolonged CPB and aortic cross-​clamp times, recent myocardial infarction,
diffuse CAD, poor baseline LV or RV function, and suboptimal myocardial
protection are predictive factors for early PCCS. TOE may demonstrate
regional wall motion abnormality or generalized hypokinesis of the LV, the
RV, or both.
Latent shock
‘Latent PCCS’ represents a more insidious onset of myocardial failure in
the hours or days following cardiac surgery which is usually manifested by
borderline or slowly declining haemodynamic or metabolic parameters
with accompanying end-​organ dysfunction/​failure. Echocardiography and
invasive haemodynamic measurements are indispensable tools to rule out
mechanical causes of HF (i.e. progressive cardiac tamponade).
 Step-wise approach to the high-risk patient 119

Step-​wise approach to the high-​risk

patient
There is no universally agreed protocol in the management of PCCS. The
step-​wise approach detailed below provides a safe and systematic strategy
in the perioperative period for the patients who are high risk for PCCS and
the necessary measures that should be taken in PCCS/​impending PCCS in
order to optimize end-​organ (including myocardial) perfusion.
During all stages of management, TOE by a skilled anaesthesiologist and
appropriately positioned PAC are indispensable tools in real-​time assess-
ment of haemodynamic parameters and myocardial response to treatment.
Step 1: preoperative optimization
• Anticipation of PCCS and appropriate discussion with patient and
family regarding wishes and ‘ceiling of care’ can aid decision-​making
intraoperatively.
• Depending on the PA pressure, mixed venous saturation (SvO2), and
biventricular function, preoperative management of the high-​risk patient
in the ICU with an appropriate inotrope/​inodilator support, diuretic
therapy, and IABP is an important aspect of preoperative optimization,
thereby reducing the risk of PCCS.
Step 2: measures prior to separation from CPB
• Consider and correct any potential reversible aetiology/​ies as a result
of technical errors (e.g. significant para-​prosthetic valvular regurgitation
or occluded coronary graft/​s).
• Adequate intracardiac de-​airing with LV and aortic root vents.
• Rewarm and correction of any metabolic and electrolyte derangements
(artificially correct residual metabolic acidosis to achieve base excess
below –​2 and haemoglobin >100 g/​dL).
• Consider half-​tidal volume ventilation of lungs while on CPB to reduce
the risk of pulmonary atelectasis and high PVR.
• Achieve sinus rhythm, if possible, medically or with DC cardioversion.
• Epicardial pacing to heart rate of 90 bpm with ideally AAI or DDD.
• Adequate filling to maintain euvolemia.
• Initiation of positive inotropes with β1-​adrenergic agonist (epinephrine/​
dobutamine/​dopamine) activity while on CPB support.
• Once the aortic cross-​clamp is released, consider priming the
myocardium with an inodilator/​phosphodiesterase-​3 inhibitor such
as milrinone/​enoximone particularly in patients with preoperative
evidence of RV dysfunction with or without PH.
• Allow adequate reperfusion of the myocardium after release of cross-​
clamp before attempting weaning off CPB.
• Placement of IABP. Timing of inflation should be optimized to the
dicrotic notch.
Step 3: measures during separation from CPB
• Ensure lungs are fully expanded (if suboptimal expansion, check
endotracheal tube, perform bronchoscopy prior to weaning attempt).
• Pace atrially or dual chamber at a rate of 80–​90 bpm ensuring enough
diastolic coronary perfusion time. In certain circumstances, a rate of
201

120 Chapter 11 Post-cardiac surgery cardiogenic shock

100–​110 bpm might be necessary—​in particular, higher pacing rates are

beneficial for RV dysfunction.
• Gentle weaning from CPB, that is, reduction of CPB support by 500
mL/​min increments while monitoring the filling pressures and avoiding
ventricular distention.
Step 4: measures following separation from CPB
• Avoid hypoxaemia or hypercarbia.
• Correct any metabolic/​respiratory acidosis.
• If low, the CVP should carefully be optimized to between 10 and
15 mmHg.
• Titration of vasoactive catecholamine and inodilator infusions to achieve
a CI of 2.5 L/​m2/​min with a SVR of 800–​1200 dynes/​sec/​cm5 or
indexed SVR (SVRI) of 1500–​2000 dynes/​sec/​cm5 per m2.
• Protamine should be administered very cautiously and slowly.
• Reduction of PA pressures utilizing the following:
• Inhaled pulmonary vasodilators (nitric oxide or epoprostenol/​
iloprost) can be used to reduce pulmonary vasoconstriction and
reduce RV afterload especially when there is RV dysfunction.
• Infusion of phosphodiesterase-​3 inhibitors (e.g. milrinone) can aid
improve biventricular contractility and their energy reserves while
reducing the RV and systemic afterload (although these may induce
deleterious systemic hypotension).
Step 5: mechanical support strategies
Short-​term strategy: CPB
If the strategies in previous steps fail, recommencing CPB is the safest and
most expedient, short-​term ‘fall-​back’ strategy while the team conducts
further evaluation and planning. CPB would also allow for further myocar-
dial recovery with further reperfusion and resting of the heart in cases of
myocardial stunning as a result of suboptimal myocardial protection during
the initial surgery. This also allows correction of any electrolyte abnor-
mality before attempting further weaning of CPB. If coronary air embolism
is suspected, mild permissive hypertension while on CPB could improve
coronary perfusion for the next attempt at weaning. Furthermore, CPB
would allow time for ‘bridging to decision’ in such acute settings. If there is
persistent regional LV dysfunction or ST-​segment changes consistent with
ischaemia, coronary grafting should be considered. If there is significant
valvular dysfunction, operative correction should be considered at this time.
Longer-​term strategies: MCS
Depending on which ventricle is failing, the following strategies can be em-
ployed as salvage strategies. More details regarding insertion techniques
have been covered in Chapters 5, 8, and 9. The pros and cons for each
strategy are as follows:
ECMO
VA ECMO (see Chapter 8): can be achieved centrally (transition from cen-
tral CPB l central VA ECMO) or peripherally (typically femoro-​femoral/​
axillo-​femoral).
Pros: expedient, familiar and safe, allows for complete (heart–​lung) sup-
port, femoral or axillary cannulation would allow for chest closure.
 Step-wise approach to the high-risk patient 121

Cons: need for heparin (bleeding risk), limb ischaemia (peripheral

ECMO), North–​South syndrome (in peripheral VA ECMO l need for
VAV ECMO), need for pulsatility (LV vent may be required to prevent LV
stasis). Concomitant use of mild dose of inotrope and/​or IABP is important
to facilitate LV contractility and aortic valve opening in order to avoid LV
stasis and thrombosis). Finally, risk of prosthetic mitral valve (mechanical or
bioprosthetic) thrombosis due to blood stagnation.
Short-​term VADs
(See Chapter 5.)
Surgical short-​term VAD
Can be utilized to selectively support right, left, or both sides of the heart
(RVAD/​LVAD/​BiVAD, respectively) depending on the failing chambers and
with or without an oxygenator depending on the functioning state of the
lungs. LVAD is usually established with inflow through the LV apex and out-
flow into the ascending aorta. Alternatively, a cannula inserted through the
right superior pulmonary vein or roof of LA passed into the LV cavity past
the mitral valve can be used as the inflow port. Alternative outflow ports
to the ascending aorta include the axillary artery or femoral artery. RVAD
is established with inflow through the RA and outflow into the main PA.
Alternatively, inflow can be established using a long two-​stage venous can-
nula inserted through the common femoral vein into the RA past the IVC.
The outflow in the PA could be either by direct cannulation or through a
side-​arm PTFE/​Haemashield graft anastomosed to the main PA and exter-
iorized through the left second/​third intercostal space in the parasternal
region.
Pros: expedient with open chest following termination of CPB, lines can
be tunnelled through the skin allowing for chest closure, direct cardiac
chamber cannulation (no need for venting), allows better ambulation and
physical therapy, and less risk of mitral prosthetic valve thrombosis from
blood stagnation as the LV is well vented.
Cons: can only be placed centrally (typically sternotomy), LV apex inci-
sion (in LVAD), need for heparin (bleeding risk), pulmonary haemorrhage
(RVAD), no oxygenator by default, and thrombosis.
Percutaneous VAD
Percutaneous VADs (see Chapter 9) are available for left and right heart
support (Impella 2.5 and 5.0 (Abiomed, Danvers, MA, USA)). They work
with an ‘impeller/​micro-​axial pump’. In the left heart position, they work by
propelling blood from the LV into the ascending aorta. Placement requires
general anaesthesia and fluoroscopy to ensure that the propeller has appro-
priately crossed the aortic valve and is positioned in the LV outflow tract.
The percutaneous right heart support can be in a form of an impeller
(as for the left heart Impella described above, e.g. the Impella RP) which
can be placed through a 23 Fr peel away sheath into the femoral vein and
fluoroscopically guided across the PA. Alternatively, it can also be in form of
a centrifugal pump e.g. the ProtekDuo (TandemLife, LivaNova, Pittsburgh,
PA, USA) which is typically placed through the right IJ vein and floated to
the PA fluoroscopically such that the inflow orifice sits in the RA and the
outflow in the main PA thereby bypassing the RV.
Pros: less invasive since removal does not require sternal entry/​re-​entry.
21

122 Chapter 11 Post-cardiac surgery cardiogenic shock

Cons: haemolysis, limb ischaemia (left heart support), vessel perfor-

ation, bleeding, cutdown needed for left heart Impella arterial removal,
malposition is common, and post-​insertion ambulation is difficult.

Clinical considerations for mechanical

circulatory support
Given the complexity of the clinical situations when faced with ‘refractory
PCCS’ requiring consideration for MCS, there is lack of set protocol in pa-
tient selection and the subsequent management. There are pre-​and post-​
MCS clinical considerations that can be taken into account as follows:
Pre-​MCS clinical considerations
There is no agreed protocol to aid decision-​making and all cases should
be looked at individually. Given the complexity of such clinical settings and
major implications on the patient and the service at large, we advocate
multidisciplinary (surgical, anaesthesia, perfusion, and critical care) decision-​
making. Consideration of the wishes of the patient and family, at this time,
should be factored into decision-​making.
Some of the considerations to be taken into account are as follows:
• MCS can be resource intensive.
• ‘MCS to where?’, that is, the aim of the MCS should be determined:
• Does the patient qualify for ‘bridge to decision, ‘bridge to transplant’
or ‘bridge to durable LVAD’? Also, what are the chances of ‘bridging
to recovery’?
• The 5-​year survival rate after advanced MCS in the PCCS setting can be
as low as 17% (see Table 11.1, p. 124).
• In contrast, delayed institution of MCS (often end-​organ injury has
begun) is unlikely to rectify the patient’s course. Therefore, early
support is critical to success.
• Complications which are life-​threatening include:
• Major haemorrhage requiring re-​exploration.
• Renal failure requiring renal replacement therapy.
• Thromboembolism.
• Stroke.
• Severe sepsis.
• Limb ischaemia.
• Infection (bacterial and fungal).
• Some of the reported adverse prognostic indicators (see Table 11.1, p.
124) are as follows:
• Advanced age >70 years.
• Female sex.
• Chronic kidney disease.
• Advanced pulmonary disease.
• Diabetes mellitus.
• Obesity.
• High EuroSCORE >20.
• Prolonged ECMO support >48 hours.
• Rising serum lactate levels while on MCS.
 POST-MECHANICAL CIRCULATORY SUPPORT 123

Post-​MCS clinical considerations

Following institution of MCS, the following should be considered:
• Unfractionated heparin can be held to control bleeding and until
correction of coagulopathy. It can be safely started after 24 hours with
aPTT range of 60–​80 seconds. Convert to direct thrombin inhibitors
(e.g. bivalirudin or argatroban) if HIT is suspected.
• Distal limb protection with a DPC should be used with peripheral VA
ECMO and femoral arterial cannulation.
• Attempt to achieve contractility and ejection with positive inotropic
infusions while on VA ECMO of at least 20 mmHg pulse pressure in
order to prevent LV blood stagnation and thrombosis, pulmonary
haemorrhage, and/​or ventricular distention and myocardial damage.
• While on VA ECMO in the absence of LV ejection, LV venting should be
strongly considered. Venting in peripheral VA ECMO can be achieved
through the following routes:
• Right superior pulmonary vein.
• LV apex.
• Percutaneously (e.g. Impella 2.5/​CP)—​short-​term venting due to
haemolysis and limb ischaemia complications.
• Use of IABP and mild inotrope infusion.
• Prevention of differential hypoxaemia (i.e. North–​South syndrome)—​
right radial arterial line is essential for objective PO2 measurement. If
this PO2 is abnormally low, optimize mechanical ventilator support. VAV
ECMO (see Chapter 8) may be required to overcome this complication.
• Any cardiac rhythm disturbance should be treated aggressively with the
goal of the rate of 80–​100 bpm and AV synchronous rhythm.
• Patients supported on VA ECMO or short-​term VADs should be
considered for diagnostic coronary angiography. Furthermore, if there
is a major issue with the coronary circulation, PCI should be considered.
• For VA ECMO weaning, see Fig. 11.1.
• Wean potent vasopressors (e.g. norepinephrine) as much as possible
to maintain systemic vascular resistance at 800–​1200 dyne/​sec/​cm5.
Adding volume expansion may lead to better flow rates, enabling the
weaning of vasopressors.
• Daily TOE to monitor ventricular recovery.
• Weaning of sedation for neurological evaluation. If unarousable
following sedation hold consider head CT scan to assess prognosis for
neurological recovery.
• Patients who show ventricular recovery on echocardiography, should
have VA ECMO/​short-​term VAD flows weaned down to 1–​2 L/​
min. Consider anticoagulation during the weaning process. Continue
echocardiographic assessment. If stable at low flows for 1–​2 hours,
VA ECMO/​short-​term VAD should be removed. After weaning of
VA ECMO/​short-​term VAD, support circulation with IAPB and/​or
moderate inotropic agent (dopamine, dobutamine, or epinephrine).
• If weaning of MCS failed and the patient meets criteria for durable VAD
or transplantation, optimize and proceed accordingly.
• If weaning of MCS fails and the patient does not qualify for destination
MCS or transplantation l discussion with family regarding end of life
care.
241

124 Chapter 11 Post-cardiac surgery cardiogenic shock

Fig. 11.1 Illustrating ‘post-​VA ECMO’ management strategies.

Outcomes and adverse prognostic

indicators
The literature on refractory PCCS is very heterogeneous. The outcomes
data have been derived mainly from retrospective observational studies and
small case series. The reported patient groups range from complex heart
transplant, redo cardiac surgery, to routine first-​time open-​heart surgery
making the data difficult to interpret. Rastan and colleagues reported one of
the largest case series with one of the longest follow-​up periods of 5 years
for refractory PCCS salvaged with MCS. They reported survival to hospital
discharge and 5-​year survival to be 39% and 17% respectively. However, a
meta-​analysis performed by the authors of this chapter (Khorsandi and col-
leagues) found the pooled survival to hospital discharge was lower (~30%).
 Outcomes and adverse prognostic indicators 125

A large body of evidence suggests survival benefit from MCS for refrac-
tory PCCS. Some reports claim post-​transplant MCS results in better sur-
vival rates than when used for PCCS after other forms of adult cardiac
surgery. As PCCS has a universally high mortality without MCS, one should
consider early institution of MCS in these patients, particularly in the ab-
sence of negative predictive factors. Timing is critical; MCS to rectify the
circulation must be initiated before end-​organ injury. Recent studies (Table
11.1) found that survival following ECMO was significantly better in OHT
patients compared to post-​cardiotomy patients. Adverse prognostic indi-
cators common to a few of the studies have been reported as advanced
age (>70 years), elevated serum lactate levels, and multiorgan dysfunction.

Table 11.1 Pertinent studies looking at outcomes of MCS utilization for

refractory PCCS
Author, Study Patient cohort Results Adverse
year, and period prognostic
country indicators
Rubino 2008–​ 101 patients: Weaned from Age >70 years,
et al., 2016 •​ PTE (28.7%) ECMO: 57% persistently
2018, UK In-​hospital elevated lactate,
•​ OHT (21%) multiorgan
survival: 33%
•​ CABG and (OHT: 63%) dysfunction
valve (15%) 1-​year
•​ CABG and survival: 27.7%
other (12%) (OHT: 54%)
•​ Valve (11%) Survival
•​ Other considerably
higher in OHT
Herna 1995–​ Multicentric Operative Urgency of
ndez 2004 retrospective mortality: 45.9% operation,
et al., cohort from STS Reduction in renal failure,
2007, UK database of 5735 mortality and myocardial
patients who had major morbidity infarction,
post-​cardiotomy over time aortic stenosis,
VADs with newer female sex,
cardiac surgical race, peripheral
techniques, vascular
perioperative disease, NYHA
management, class IV, left
and advances main coronary
with devices disease, and
valve procedure
Rastan 1996–​ •​ 516 patients: Weaned from
et al., 2008 •​ CABG (37.4%) ECMO: 63.3%
2010, Discharged
Germany •​ CABG and AVR home: 24.8%
(16.6%)
•​ AVR (14.3%)
•​ Heart/​lungtrans
plant (6.5%)
•​ Other (25%)
continued
261

126 Chapter 11 Post-cardiac surgery cardiogenic shock

Table 11.1   Continued

Author, Study Patient cohort Results Adverse
year, and period prognostic
country indicators
Doll 1997–​ 219 patients: 30-​day
et al., 2002 •​ ​CABG (n =​119) mortality: 76%
2004, (n =​167)
Germany •​ CABG and AVR
(n =​240) Discharged from
hospital after 29
•​ ​MVR (n =​110 ± 24 days: 39%
•​ Other (n =​44) (n =​52)
5-​year follow-​
up: 74% (n =​
37) were alive
Slottosch 2006–​ 77 patients: Weaned from Advanced
et al., 2010 •​ CABG (n =​43) ECMO: 62% age (p =​
2013, 30-​day 0.003); rising
Germany •​ Valve (n =​10) serum lactate,
mortality: 70%
•​ CABG and valve prolonged
(n =​11) ECMO course,
•​ Aortic surgery and ECMO-​GI
(n =​5) complications
were
•​ Heart transplant independent
(n =​2) predictors
•​ Other (n =​6) of 30-​day
mortality
(p <0.05)
Wu et 2003–​ 110 patients: Weaned from Age >60 years,
al., 2010, 2009 •​ CABG (n =​31) ECMO: 61% renal failure,
Taiwan (n =​67) serum bilirubin
•​ Valve (n =​16) >6 mg/​dL,
Survival to
•​ Multiple valves hospital and duration
(n =​26) discharge: 42% of ECMO
•​ Combined valve (n =​46) >110 hours;
and other (n =​19) persistent HF
(EF <60%) was
•​ Aortic surgery a predictor
(n =​8) of mortality
•​ Post-​infarction after hospital
VSD (n =​3) discharge
•​ Pulmonary
endarterectomy
(n =​4)
OHT (n =​3)
Elshar 1995–​ 233 patients: Survival to Advanced
kawy 2005 •​ CABG (n =​86) hospital age, known
et al., discharge: 36% diabetes,
2010, •​ Any valve (n =​69) CABG, longer
USA •​ AVR/​repair CPB time
(n =​42)
•​ MV repair/​MVR
(n =​44)
•​ TV repair/​TVR
(n =​16)
 Outcomes and adverse prognostic indicators 127

Author, Study Patient cohort Results Adverse

year, and period prognostic
country indicators
Li et al., 2011–​ 123 patients: Survival to
2015, 2012 •​ CABG (n =​44) hospital
China discharge: 34.1%
•​ ​CABG and other
(n =​15)
•​ Valve (n =​40)
•​ OHT (n =​11)
•​ Other (n =​13)
AV, aortic valve; AVR, aortic valve replacement; MV, mitral valve; MVR, mitral valve replace-
ment; TV, tricuspid valve; TVR, tricuspid valve replacement.
281
 Section 3

Heart
transplantation

12 Patient selection criteria and bridging support for

cardiac transplantation 131
13 Cardiac donor selection and management 139
14 Extended criteria heart donors and perfusion
storage 157
15 Donor heart procurement: donation after brain death
and donation after circulatory determined death 171
16 Technical aspects of heart transplantation 185
17 Critical care management and primary graft
dysfunction following cardiac transplantation 195
18 Assessment and management of chronic
complications following heart transplantation 209
19 Bedside care and rehabilitation with mechanical
circulatory support and thoracic transplantation 225
20 Management of rejection following cardiac
transplantation 233
21 Antimicrobial prophylaxis and treatment of infection
after cardiac transplantation 247
301
 Chapter 12 131

Patient selection criteria

and bridging support
for cardiac transplantation
Medical assessment of transplant candidates 132
Psychosocial assessment of transplant candidates 135
Bridging support for transplant candidates 137
321

132 Chapter 12 Patient selection for heart transplantation

Medical assessment
of transplant candidates
Cardiac transplantation remains the gold standard treatment for patients
with end-​stage HF refractory to medical therapy. The decision to list for
transplantation is most commonly achieved through multidisciplinary evalu-
ation to ensure the patient meets an indication for transplant and to vet
any alternative therapies that may be less invasive. This assessment also
includes a review of potential contraindications, the collection of clinical
data, the weighing of the risks of the transplant itself against the benefits
provided, and timeframe for ongoing reassessment. The ISHLT first issued
formal guidelines for selecting transplant candidates in 2006 and updated
these recommendations in 2016.
Common causes of HF that may ultimately warrant cardiac transplant-
ation include NICM including RCMs, CAD leading to ischaemic cardio-
myopathy, congenital heart disease, and transplant recipients requiring
re-​transplantation. In an era where long-​term MCS devices are commonly
used to support end-​stage HF, device complications can become a strong
indication even in the absence of refractory symptoms.
Common indications for referral for cardiac transplantation evaluation
are illustrated by the following clinical scenarios or syndromes (Box 12.1):
• Cardiogenic shock requiring continuous IV inotrope support or MCS to
prevent end-​organ malperfusion.
• NYHA class IV HF symptoms refractory to optimal medical
management and limited 1-​year survival. Typical criteria to define this
degree of HF in the ambulatory population include CPX testing with
peak VO2 <14 mL/​kg/​min or <12 mL/​kg/​min on BB therapy, SHFM
suggesting 1-​year survival <80%, or HFSS with medium to high risk:
• The SHFM is a prospectively validated score that predicts survival
and includes the following predictors: age, sex, NYHA class, weight,
EF, systolic blood pressure, medication usage including diuretic usage,
laboratory parameters, and presence of devices.
• The HFSS is a prospectively validated score that predicts survival
without transplant and includes the following predictors: presence of
CAD, resting heart rate, LVEF, mean arterial blood pressure, presence
of conduction delay on ECG, serum sodium, and peak VO2. The three
strata are low (1-​year survival 88%), medium (60%), and high risk (35%).
• RCMs with NYHA class III–​IV symptoms particularly when caused by
infiltrative processes that have limited extracardiac involvement at the
time of evaluation.
• Unstable arrhythmias (typically ventricular in origin) that have become
refractory to antiarrhythmic drug therapy, transcatheter ablations,
defibrillator implantation, or surgical correction.
• Complex congenital heart disease causing NYHA class IV symptoms
refractory to medical management or further surgical palliation including
cyanotic heart disease and failing Fontan physiology after correction of a
single ventricle circuit. This indication is particularly important to identify
before the development of irreversible pulmonary hypertension.
• Cardiac transplant recipients with severe cardiac allograft vasculopathy,
severe and persistent primary graft dysfunction, or severe restrictive
physiology.
 Medical assessment of transplant candidates 133

Other indications for cardiac transplantation include intractable severe

angina not amenable to revascularization, HF with major limitation of
daily activities and VO2 <55% of predicted, and patients with AL or TTR
amyloidosis.

Box 12.1 Common indications for cardiac transplantation

• Cardiogenic shock requiring IV inotropes or MCS.
• NYHA class IV symptoms refractory to optimal medical management.
• RCM with NYHA class III–​IV symptoms.
• Unstable arrhythmias refractory to other management.
• Complex congenital heart disease with NYHA class IV symptoms
refractory to other management.
• Cardiac transplant recipients with severe graft dysfunction.
• Intractable severe angina not amenable to revascularization.
• Cardiac amyloidosis.
• Candidacy for multiorgan transplantation:
• Heart–​lung: irreversible pulmonary vascular disease, hypoplasia of
vasculature.
• Heart–​kidney: combined HF and end-​stage renal disease.
• Heart–​liver: congenital heart disease with chronic venous congestion
leading to irreversible liver disease.

Patients possibly eligible for multiorgan transplantation should be re-

ferred to experienced transplant centres. Indications for combined heart–​
lung transplantation include irreversible pulmonary vascular obstructive
disease and hypoplasia of pulmonary vasculature. Combined heart–​kidney
transplantation may be considered for patients with end-​stage renal failure
and appropriate indication for kidney transplantation, as well as heart–​liver
transplantation in congenital heart disease patients with chronic venous
congestion leading to cirrhosis. TTR amyloidosis has previously been con-
sidered an indication for heart–​liver transplantation due to the fact that the
TTR protein is produced by the liver, but novel pharmaceutical therapies
may absolve the need for concurrent liver transplantation.
Congenital heart disease represents another separate consideration in
evaluating recipients for heart transplantation. While this indication is dis-
cussed in other chapters, the following considerations are essential:
• Assessing PVR that, if elevated, may lead to rapid RV failure after
transplantation. Calculating PVR is difficult in patients with persistent
shunts and collateral circulation.
• Multiple prior cardiac operations may lead to prohibitively long and
risky explants, specifically in relation to right-​sided structures that may
lie directly behind the sternum.
• Other end-​organ failure (hepatic failure being the most common).
Contraindications to transplantation (Table 12.1) are divided into absolute
and relative; those with reversible contraindications may be appropriate
candidates for bridging MCS to allow the patient to be re-​evaluated in the
future.
341

134 Chapter 12 Patient selection for heart transplantation

Table 12.1 Contraindications for cardiac transplantation

Absolute contraindications Relative contraindications
Irreversible severe PH Advanced age
Life expectancy <2 years secondary Obesity
to unrelated illness
Chronic uncontrollable infection Active smoking
Severe cerebrovascular disease Diabetes mellitus with poor control or
significant end-​organ damage
Prohibitive risk factors identified in Renal dysfunction
psychosocial assessment
Active infection
Anatomic features prohibiting surgery
Prohibitive risk factors identified in
psychosocial assessment

Absolute contraindications to cardiac transplantation are those that have

been adopted at most transplant centres and include the following:
• Irreversible severe PH with PVR >5 Wood units. These patients may
be considered for heart–​lung transplantation but will not tolerate
isolated heart transplantation. Chronic LVAD unloading may result
in improvement of PVR over months. Some centres have reported
successful outcomes when PH can be reduced pharmacologically, but a
useful threshold for PH remains to be clearly defined.
• Life expectancy <2 years secondary to an unrelated severe systemic
illness including malignancy, refractory infection, and other systemic
illness not corrected or modifiable by cardiac transplantation.
• Chronic extracardiac infection that cannot be definitively controlled.
• Cerebrovascular disease that impairs cognitive function and may limit
survival.
• Active substance abuse or medical non-​compliance, identified through
the psychosocial assessment described on p. 135.
Relative contraindications to cardiac transplantation are often more centre
specific or modified slightly based on centre expertise and resources, and
variability among centres has been shown to be significant in programme
surveys. While individually not absolute, the cumulative risk of multiple
relative contraindications can be high and should be considered. They in-
clude the following:
• Advanced age: careful selection of patients over the age of 60–​65 and
extremely careful selection of patients aged >70 may yield candidates
who are suitable for transplantation.
• In general, transplantation in this age group results in reduced but
acceptable post-​transplant survival compared to younger recipients.
Since the cause of death in many of these patients is recipient driven,
many centres focus on the absence of comorbid conditions and frailty
as potential indicators of success rather than chronological age.
• Obesity: while each centre may have a different cut-​off, a BMI of ≥35
kg/​m2 is typically prohibitive as the recipient is at higher risk.
 Psychosocial assessment of transplant candidates 135

• Active smoking.
• Diabetes mellitus with inadequate control or significant end-​organ
damage.
• Renal dysfunction not attributed to cardiac failure such as estimated
glomerular filtration rate (eGFR) <30 mL/​min/​1.73 m2. This
comorbidity may prompt evaluation for combined heart–​kidney
transplantation.
• Active infection or colonization with multidrug-​resistant organisms that
would jeopardize successful transplantation particularly in the face of
immunosuppression. Patients with chronic, controlled infection such
as hepatitis B, hepatitis C, and HIV may be suitable candidates for
transplantation if they have mild forms of disease and no evidence of
end-​organ dysfunction.
• Prior substance abuse and high risk for relapse as identified through the
psychosocial assessment described in the following section (see p. 136).
• Advanced peripheral vascular or cerebrovascular disease.
• Advanced primary lung disease, regardless of pulmonary vascular
resistance can complicate recovery from cardiac transplantation.
• Anatomic features that would make the transplant operation prohibitive.
This may stem from prior operations for acquired heart disease, but is
especially important in congenital heart disease. Chronic cyanosis can
cause significant systemic–​PA collaterals, which can complicate surgery
or persist after surgery and compromise the outcome.
• Other systemic illness which may interfere with postoperative
rehabilitation.
• The presence of extensive anti-​HLA antibodies.

Psychosocial assessment of transplant

candidates
Due to the scarcity of organ donors as well as the complexity and costs of
post-​transplant care, selection of heart transplant recipients must include
a thorough psychosocial evaluation. Recipients must be able to adhere to
treatment recommendations for life; these include frequent clinic visits, la-
boratory assessments, routine catheterizations and biopsies, and potential
hospitalizations for complications. The psychosocial assessment includes
identification of risk factors for poor post-​transplantation outcomes; as-
sessment of factors related to knowledge, understanding, and ability to en-
gage in decision-​making; and an assessment of patients’ personal, social,
and environmental resources and circumstances. These components when
assessed in totality have been demonstrated to play an important role in
clinical outcomes. The ISHLT has made specific recommendations for the
psychosocial evaluation of potential recipients of heart transplantation (or
lung transplantation or MCS).
Key domains for assessment include treatment adherence, mental health
history, substance use history, cognitive status and capacity to give informed
consent, knowledge and understanding of current illness, knowledge and
understanding of current treatment options, coping with illness, social sup-
port, and social history (Box 12.2).
361

136 Chapter 12 Patient selection for heart transplantation

Box 12.2 Psychosocial assessment domains for cardiac

transplant evaluation
Domain
• Treatment adherence and health behaviours.
• Mental health history.
• Substance use history.
• Cognitive status.
• Knowledge and understanding of current illness.
• Knowledge and understanding of treatment options.
• Ability to cope with illness.
• Social support.
• Social history.
Reproduced from Dew MA, DiMartini AF, Dobbels F, et al. The 2018 ISHLT/​APM/​AST/​
ICCAC/​STSW recommendations for the psychosocial evaluation of adult cardiothoracic trans-
plant candidates and candidates for long-​term mechanical circulatory support. J Heart Lung
Transplant. 2018 Jul;37(7):803–​823. doi: 10.1016/​j.healun.2018.03.005.with permission from
Elsevier.

The recommended procedure for psychosocial evaluation of transplant

candidates includes (1) selection of a qualified evaluator, (2) performance
of the psychosocial evaluation and identifying when further evaluation is
needed, (3) communication of findings back to the transplantation team,
and (4) monitoring/​mitigating identified risk factors to optimize transplant
outcomes. The evaluator should be formally trained and licensed in their
healthcare discipline and familiar with local transplant policies and proced-
ures. The psychosocial evaluation should be presented to the patient as
one of many elements in his/​her overall evaluation, ideally with the patient
directly interviewed in a language in which the potential recipient as well as
support system members can converse interactively.
Risk factors for poor post-​transplantation outcomes
Treatment non-​adherence, mental health history, and substance use history
have been identified as risk factors for poor post-​transplantation outcomes.
Assessment of the candidate should include assessment of past medical
adherence, knowledge of the regimen that is required, and willingness to
participate. Poor treatment adherence has been shown to increase the risk
of post-​transplant morbidity and mortality. Psychiatric conditions range
from prohibitive to transplant to relatively common within the transplant
population. Assessment of the psychiatric condition’s interference with
the patient’s ability to adhere to the medical regimen and the degree to
which the condition is controlled is important. Depression and anxiety
are common in the pre-​transplant population. Depression predicts post-​
transplant mortality. Anxiety, bipolar disorder, and personality disorders
have not been shown to directly correlate with increased mortality but
likely have effects on medication compliance, substance abuse, poor social
support, and/​or poor coping skills.
Illegal substances, tobacco, and alcohol abuse have been shown to in-
crease risk for post-​transplant mortality, and active abuse is an absolute
contraindication to transplantation. Assessment should include review of
all substances used in the past, current status of use, treatments received,
 Bridging support for transplant candidates 137

and insight/​willingness to abstain. With the legalization of marijuana use

in some parts of the world without clear evidence for whether it is a risk
factor for postoperative outcomes, assessment should focus on manifest-
ations of substance abuse.
Assessment of factors related to understanding, and ability
to engage in decision-​making
The patient’s cognitive ability to comprehend their condition, the medical
and surgical treatment, and postoperative regimen should be assessed.
Intellectual disability does not necessarily portend worse outcomes, and
can be mitigated by increased social support, although dementia is a contra-
indication to transplantation. However, the complex interaction between
HF and cognitive function is not clear, and defects in cognition need to be
formally assessed prior to transplantation. This is true not only to avoid in-
appropriate transplantation in a patient not able to reap the benefit, but to
provide a baseline since immune suppression may cause long-​term changes
in cognition. One must determine the patient’s level of understanding of
their medical condition and the current treatment options, which is not only
critical to informed consent and shared medical decision-​making but also
can identify gaps in knowledge or support which could help optimize the
patient’s condition. Knowledge deficiencies in one’s medical condition have
been associated with worse self-​care and medical adherence.
Assessment of patients’ personal, social, and environmental
resources and circumstances
A thorough social history elucidating the patient’s education level, financial
resources, living situation, and legal history is important to understanding
the context which will affect the patient’s overall health. Lack of sufficient
social support is a contraindication to transplantation. Lastly, the patient’s
ability to cope with their illness may help predict their post-​transplant
success, as optimism, a greater sense of self-​efficacy, and active problem-​
solving skills have been tied to improved clinical outcomes.

Bridging support for transplant

candidates
Bridging support devices, ranging from femoral catheter IABPs to durable
LVADs to TAHs, can provide a bridge to decision regarding transplant can-
didacy or a support strategy for patients already listed, who are threatening
to die as a result of cardiogenic shock. Longer-​term bridging devices such
as LVADs can allow time for relative contraindications such as obesity, PH,
substance use, or insufficient social support to be reversed or resolved.
Renal, pulmonary, and hepatic function frequently improves during LVAD
support. LVADs can also become destination therapy for those ultimately
deemed ineligible for transplantation either due to irreversible contraindica-
tions or loss of candidacy for new medical/​psychosocial reasons.
Temporary bridging devices such as an IABP, percutaneous VADs
(e.g. Impella (Abiomed, Danvers, MA, USA), TandemHeart LivaNova,
Pittsburgh, PA, USA)), and ECMO are typically used for patients presenting
381

138 Chapter 12 Patient selection for heart transplantation

with haemodynamic collapse. These devices provide short-​term haemo-

dynamic support and can help with preoperative optimization. IABPs are
inserted and removed easily, require less technical support, and are rela-
tively inexpensive. While this strategy is best utilized for patients waiting
days before a suitable organ is identified, modifications in the placement
technique (axillary artery rather than femoral artery) may allow ambulation
and rehabilitation while on support. Furthermore, newer counterpulsation
devices in development, may not only enable ambulation but are also de-
signed for the option of discharge from the hospital setting.
Other percutaneous circulatory assist devices (e.g. Impella, TandemHeart)
are generally considered to provide greater haemodynamic support than
IABPs and are usually placed in the cardiac catheterization laboratory or
in the operating room. The Impella platform is an axial pump that res-
ides in the LV outflow tract and pumps blood from the LV into the aorta.
A number of devices with increasing support (2.5 L/​min, 3.5 L/​min, or 5 L/​
min) are available on this platform which resembles a catheter sitting across
the aortic valve. The TandemHeart relies on a drainage cannula situated in
the LA across the intra-​atrial septum; it provides antegrade flow utilizing
an extracorporeal centrifugal flow pump which delivers blood back via a
femoral arterial cannula. These devices typically provide bridging therapy
from days to 1 week.
VA ECMO, used in cases of cardiogenic shock, is an extracorporeal car-
diopulmonary support system that drains deoxygenated blood from the
venous system, oxygenates it via an artificial membrane lung, and returns it
to the systemic circulation. Cannulas may be placed centrally via sternotomy
or thoracotomy, or peripherally, typically via the femoral vessels. Patients
are typically supported for no longer than 1 week. Extracorporeal VADs
and VA ECMO have a high rate of infection, bleeding complications, and
thromboembolic complications including stroke. Short-​and mid-​term sur-
vival outcomes of patients bridged to transplantation with VA ECMO are
inferior compared to propensity-​matched controls bridged to transplant-
ation via continuous-​flow LVAD.
Durable LVAD implantation has continued to increase since its inception
in the early 1990s and offers patients the ability to be bridged for months
to years. The 2015 ISHLT registry indicated that at least 50% of patients
are bridged to transplantation with MCS, the vast majority utilizing LVAD.
This affords transplant candidates with longer expected waiting list times
the ability to stabilize with reduced risk of clinical deterioration. It also pro-
vides for reversal of some relative contraindications. Single-​centre studies
exist that show improved post-​transplant survival when patients are bridged
with MCS. However, some patients become sensitized related to required
transfusions with LVAD implantation; infection and thromboembolic ad-
verse events may also occur during LVAD support.
 Chapter 13 139

Cardiac donor selection

and management
Pathophysiology of brainstem death 140
Cardiac donor assessment and selection 141
History and examination 142
Blood investigations 142
Electrocardiogram 143
Echocardiogram 143
Pulmonary artery catheterization 145
Coronary angiography 145
Intraoperative assessment 146
Post-procurement assessment 146
Recipient individual requirements and donor–recipient
compatibility 146
Cardiac donor optimization 148
Circulatory and endocrine management 149
Respiratory management 152
General critical care 153
Donor risk scoring systems 154
Mortality risk scores 154
Deficiencies with using risk scores 154
Effect of donor risk factors on outcomes of transplantation in
recipients supported with long-term LVADs 155
401

140 Chapter 13 Cardiac donor selection and management

Pathophysiology of brainstem death

Knowledge of the underlying physiological changes of brainstem death
(BSD) is crucial to understanding the different aspects of donor manage-
ment and optimization. Several physiological changes occur in association
with the rising intracranial pressure that usually precedes BSD and these
consist of cardiovascular, respiratory, endocrine, and metabolic responses
detailed below.
Cardiovascular
• In response to raised intracranial pressure and reduced cerebral
perfusion pressure, a hypothalamic-​induced catecholamine storm
occurs resulting in intense vasoconstriction, raised systemic vascular
resistance, tachycardia, and systemic hypertension. This aims to restore
cerebral blood flow by improving cerebral perfusion pressure.
• Myocardial oxygen demand increases as a result without a simultaneous
increase in oxygen supply due to coronary vasoconstriction. This is
implicated in the impairment of cardiac function seen early after BSD:
• Acute myocardial injury occurs in a quarter of donation after
brain death (DBD) donors and myocardial dysfunction in 40%.
Severity of injury relates to the speed of onset of BSD with rapid
rises in intracranial pressure resulting in exponential increases in
catecholamine release and therefore a higher incidence of myocardial
damage.
• Systemic hypertension stimulates carotid baroreceptors resulting in
slowing of the heart rate often to bradycardic levels.
• Along with respiratory irregularities, which occur secondary to reduced
brainstem perfusion or possible brainstem herniation, hypertension and
bradycardia represent the classic Cushing’s triad.
• As BSD proceeds, sympathetic tone diminishes, leading to a period of
hypotension and global hypoperfusion due to loss of vasomotor tone
and an impaired cardiac output.
Respiratory
• Pulmonary oedema occurs secondary to raised pulmonary hydrostatic
pressure and catecholamine-​induced endothelial injury.
Endocrine and metabolic
• Posterior pituitary function is commonly lost resulting in diabetes
insipidus. Anterior pituitary function may be preserved or partially
affected.
• Hyperglycaemia develops secondary to reduced insulin levels and
increased insulin resistance.
• Temperature control is lost due to hypothalamic failure with initial
hyperpyrexia and subsequent hypothermia.
• A generalized inflammatory response, ischaemia-​reperfusion injury, and
coagulopathy occur secondary to release of tissue thromboplastin from
necrotic neural tissue in the brain-​dead donors.
 Cardiac donor assessment and selection 141

Cardiac donor assessment and selection

Principles of donor selection
Selection of donor hearts is based on two core principles:
• Assessment of donor heart quality including both structure and function.
• Recipient individual requirements and donor–​recipient compatibility.
Donor selection is rarely a black-​ and-​ white process and a degree of
‘balancing’ between the two principles is required.
Assessment of donor quality
Assessment should include a review of the donor’s history, clinical exam-
ination findings, haemodynamic status, and laboratory values and imaging
(echocardiogram, possible cardiac angiography). The desirable and undesir-
able donor characteristics are summarized in Table 13.1.
Table 13.1 Summary of desirable (classical cardiac donor selection criteria) and
undesirable donor characteristics.
Characteristic Desirable (classic Undesirablea
selection criteria)
Age (years) <55 >55
History of chest trauma or Nil Congenital or valvular heart
cardiac disease disease
History of malignancy or Nil Metastatic cancer; uncontrolled
current sepsis sepsis or aspergillosis
History of prolonged Nil History of hypoxaemia (e.g.
hypotension or drowning), hypotension,
hypoxaemia ventricular fibrillation, and CPR
Meets haemodynamic Yes High dose of catecholamines
criteria: MAP >60 mmHg; (dopamine >10 mcg/​kg/​
CVP 8–​12 mmHg; inotropic min, need for epinephrine or
support <10 mcg/​kg/​min norepinephrine)
(dopamine or dobutamine)
ECG Normal Signs of infarction (Q waves),
LVH, and atrial fibrillation
Echocardiogram Normal Regional wall motion
abnormalities, valvular
abnormalities, LVH (>13 mm),
and cavity enlargement
Coronary angiography (if Normal Presence of CAD (or risk
indicated by donor age and factors for CAD if angiogram
history) unavailable)
Serology: hepatitis B Negative HBs-​Ag positive, HIV
surface antigen (HBs-​Ag), positive, or anti-​HCV positive
hepatitis C virus (HCV), History of high-​risk social
and HIV virus behaviour (e.g. incarceration,
substance misuse)
a
Majority of quoted undesirable characteristics are considered relative rather than absolute
contraindications to donor utilization.
Adapted from Sabiston & Spencer Surgery of the Chest, 8th ed. Sellke FW, del Nido PJ,
Swanson SJ, et al. eds. Philadelphia: Saunders Elsevier, 2016.
421

142 Chapter 13 Cardiac donor selection and management

History and examination

Donor age
• Independent predictor of postoperative mortality and the likelihood of
donor heart use.
• Upper age cut-​off has gradually shifted upwards with most centres
accepting donor age <55 years as a cut-​off. Some centres procure
hearts from those aged ≥65 years (see Chapter 14).
• Utilization of hearts form older donors mandates meticulous donor
assessment and assessment of the risk–​benefit ratio for each recipient
balancing the anticipated waiting list mortality against the expected
reduced post-​transplant survival with older donors.
Assess donor comorbidity
• Diabetes mellitus, CAD, or associated risk factors (hypertension,
hyperlipidaemia, peripheral vascular disease, smoking, obesity, family
history, substance misuse), valvular heart disease, and LVH.
Cause of death and associated circumstances
• Deaths secondary to cerebrovascular events have been associated with
reduced transplant survival.
• Causes of death deemed unfavourable for cardiac procurement include
drowning, electrocution, hanging, strangulation, and carbon monoxide
poisoning.
• Preceding cardiac arrest and CPR is associated with reduced post-​
transplant survival but is not an absolute contraindication to donor use.

Blood investigations
Serology
• Test for hepatitis B surface antigen, HCV, and HIV virus.
• Hepatitis B-​or C-​positive donors may be appropriate in selected
higher-​risk recipients.
• For hepatitis C, serology does not specifically differentiate the presence
or absence of viraemia. Positive serology indicates current or past
infection.
Nucleic acid amplification testing (NAAT)
• Directly measures viral RNA using polymerase chain reaction (PCR) or
transcription-​mediated amplification.
• More accurate assessment of transmission risk.
• PCR testing of donor blood for HCV and HIV viral DNA is now
performed routinely in addition to serological testing.
• Differentiates past from current HCV infection:
• HCV-​seropositive donor who is NAAT negative (non-​viraemic)
indicates a spontaneously cleared or successfully treated infection or
a false-​positive antibody result.
• HCV-​seropositive donor who is NAAT positive (viraemic) indicates
active infection with high infection risk.
 ECHOCARDIOGRAM 143

Cardiac enzymes
• Elevated cardiac enzymes in isolation should not be used to justify
donor non-​use.
• Conflicting evidence regarding their association with post-​transplant
survival and outcomes.
• Raised cardiac enzymes have been associated with increased rates of
rejection.
• Higher troponin levels are seen closer to the time of coning and levels
fall subsequently.
• Although troponin levels are inversely correlated with CI and ventricular
function, elevated cardiac enzymes in the presence of normal cardiac
function is commonly encountered.
• Normal troponin levels in the presence of echocardiographic evidence
of dysfunction is helpful and reassuring against recent myocardial injury.

Electrocardiogram
• Need to correlate ECG findings with cardiac enzyme levels and
echocardiographic assessment.
• Look for evidence of ischaemia (ST-​segment changes, Q waves),
conduction abnormalities (e.g. left bundle branch block), rhythm
abnormalities (e.g. atrial fibrillation), LVH, and acute strain patterns.
• ST-​segment, T-​wave changes and prolongation of the corrected QT
interval are described in BSD and are not considered sensitive indicators
of suboptimal donor heart quality. Non-​specific ST changes are also
commonly seen in patients who have undergone CPR, experienced
chest trauma or neurological insult, or are currently being treated with
vasoactive medication.
• Q-​wave formation and left bundle branch block are more suggestive of
underlying disease.

Echocardiogram
• Echocardiography (transthoracic or transoesophageal) is the most
important tool in assessing donor heart function and structure.
• Should be performed in all donors particularly older donors (>40 years)
and those with a history of hypertension, and substance abuse or CAD
risk factors.
• Allows for real-​time assessment of heart function and structure.
• Initial echocardiography should only be performed after initial
haemodynamic resuscitation. Serial echocardiograms may be required
to assess response to treatment (see later in topic).
Functional assessment
• RV function.
• LV function.
• Wall motion abnormalities:
• Global myocardial depression can be seen initially and this in isolation
should not preclude subsequent procurement.
41

144 Chapter 13 Cardiac donor selection and management

• Transient regional wall motion abnormalities occur commonly and

segmental systolic inward motion and thickening abnormalities may
resolve with resuscitation.
Pharmacological stress assessment (PSE)
• Use of PSE in marginal heart donors and those with a high
cardiovascular risk shows potential for extending donor criteria in heart
transplantation:
• Dobutamine or dipyridamole PSE in ‘marginal’ donors defined as age
>50 years or <50 years with concomitant risk factors such as cocaine
use, hypertension, and diabetes.
• Majority of patients with normal resting and PSE echocardiograms
had good outcomes following transplantation.
• Patients with normal resting echocardiograms and abnormal PSE may
have CAD or other cardiac pathology.
Structural assessment
• Chamber size: significant dilation of the RV or LV is a contraindication
to donor organ utilization.
• LVH:
• The observer must ensure adequate filling of the LV when assessing
for hypertrophy as pseudohypertrophy can be encountered in
underfilled subjects and does not represent a contraindication to
transplantation.
• Mild LVH (wall thickness ≤13 mm) is not a contraindication to
transplantation.
• Transplantation is not advisable in presence of both
echocardiographic (>13 mm) and ECG criteria for significant LVH.
• Studies have suggested decreased survival in heart transplant
recipients whose donor heart LV wall thickness exceeded 1.4 mm.
• Combination of LVH and prolonged total ischaemic time may be
associated with higher rates of primary graft dysfunction.
• Congenital or valvular abnormalities:
• Classically. most valvular and congenital cardiac abnormalities
represent a contraindication to procurement.
• ‘Bench’ repairs for mild abnormalities such as secundum ASD and
mild–​moderate MR or TR can be performed on the donor heart to
enable its utilization.
• Normally functioning bicuspid aortic valves are commonly seen and
do not represent a contraindication to procurement.
Limitations of echocardiography and importance of serial
imaging
• Accuracy of TTE interpretation at the donor hospital may be
suboptimal.
• A single echocardiogram may provide initial information that is not
representative of the true function of the donor heart and should be
repeated to assess response to intervention. An important fraction of
hearts with reduced ventricular function will show normalization on
serial echocardiography studies, commonly performed at 4–​8-​hour
intervals.
 CORONARY ANGIOGRAPHY 145

Pulmonary artery catheterization

• Aggressive haemodynamic management with the use of PAC, among
other metabolic and hormonal interventions, yields a 30% increase in
the donor pool and is linked to better transplant outcomes.
• With 42% of donor hearts exhibiting some form of cardiac dysfunction
on echocardiography, the correlation between these abnormalities and
actual underlying pathology is often lacking. PAC should be therefore
considered particularly if the initial evaluation of cardiac function
confirms an EF ≤45%.
• Tailored therapy should be instituted to achieve the following targets:
• PCWP 8–​12 mmHg.
• CI >2.4 L/​min/​m2.
• SVR 800–​1200 dyne/​sec/​cm5.

Coronary angiography
Coronary angiography can be considered if this is available at the donor
hospital to assess for CAD in donors with the following characteristics:
• Male donors aged 35–​45 years and female donors aged 35–​50 years if
there is a history of cocaine use or in the presence of three or more
risk factors (Box 13.1).
• Male >45 years and female >50 years. Angiogram is strongly
recommended particularly in donors >55 years old.
• In the presence of CAD, there are no specific absolute contraindications
to donor use and decisions should be made taking into consideration
the recipient’s status (e.g. urgent or elective, uncontrollable
arrhythmias, haemodynamic deterioration without mechanical support
options), donor cardiac function, and the feasibility of ‘bench’ CABG.
• The combination of increased donor age and CAD correlates
significantly with subsequent coronary allograft vasculopathy.
• Mild CAD should be restricted to high-​risk recipients.
• Bench CABG is associated with a 65% graft patency at 2 years.
• The inability to perform a coronary angiogram at the referring hospital
should not be considered a contraindication to donor use and suitability
should continue to be based on echocardiographic assessment of

Box 13.1 Risk factors for ischaemic heart disease

• Smoking.
• Hypertension.
• Diabetes mellitus.
• Dyslipidaemia.
• BMI >32 kg/​m2.
• Positive family history of premature CAD.
• Previous history of ischaemic heart disease.
• Evidence of ischaemia on ECG.
• Anterolateral regional wall motion abnormalities.
• EF ≤45%.
461

146 Chapter 13 Cardiac donor selection and management

cardiac function, haemodynamic stability, and surgical findings during

procurement (see p. 171 for surgical aspects of organ procurement).
Risk factors for CAD should be factored into the decision to use a
donor heart for a specific recipient.
• To reduce the risk of significant nephrotoxicity in the donor, contrast
left ventriculography should be avoided if echocardiographic images are
deemed adequate.

Intraoperative assessment
Remains an integral part of donor assessment and allows:
• Visual assessment of dynamic cardiac function.
• Palpation for the presence CAD.

Post-​procurement assessment
The donor heart can also be assessed following procurement prior
to implantation, using an extracorporeal circulation, known as ex vivo
heart perfusion (see Chapter 14). Observation of cardiac function and
measurement of serum lactate can be performed during ex vivo heart
perfusion.

Recipient individual requirements and

donor–​recipient compatibility
When matching donors to recipients, it is important to select a donor heart
that meets the needs and characteristics of the recipient. Factors that need
to be considered include the following:
Donor-recipient organ matching Histocompatibility
• ABO compatibility:
• Donor and recipient must be ABO compatible to prevent hyperacute
rejection. Some ABO incompatible transplants have been performed
in neonates with immature immune response.
• HLA compatibility:
• HLA mismatch is not a contraindication.
• HLA-​A1, -​A2, and -​A3 mismatch is associated with an increased risk
of chronic rejection.
• If >10% reactivity on lymphocyte toxicity screen, a prospective
lymphocyte cross-​match between donor lymphocytes and recipient
serum is required by many centres. A positive lymphocyte cross-​
match indicates a high risk of hyperacute rejection and the heart is
rejected for that recipient.
• Other centres rely more heavily on predetermination of specificity
of recipient antibodies and antibody concentrations, enabling a virtual
cross-​match with prospective donors.
 Donor-Recipient compatibility 147

Urgency
• How urgent is the transplantation? For example, ‘super’ urgent, urgent,
or elective.
• What is the waiting list mortality? What is the SHFM or HFSS?
Recipients with a higher waiting list mortality include patients of older
age, or those with congenital heart disease, RCM, or a previous failing
allograft, obesity (BMI >35 kg/​m2), or HIV.
• Extended criteria donor (ECDs) or ‘marginal’ donors should be
considered for sicker recipients without alternatives and poor
physiological reserve in urgent need of transplantation (see
Chapter 14).
• Multiple studies have demonstrated the success of utilization ECD
hearts for heart transplantation (also referred to as extended criteria
cardiac transplantation (ECCT)) in the high-​risk recipient group. This
strategy improved patient survival in this cohort when compared to
optimal medical therapy alone.
• Benefits of ECCT should be weighed against the downside of increased
donor-​related complications such as graft dysfunction, future CAD, and
infection risk with ECD organs.
Expected recipient size and power requirements
(consider current power use)?
• Is the donor heart suitably powerful to support the recipient circulation?
• Greater requirements in males, higher BMI, and greater muscle mass.
• Lesser requirements in female patients, and/​or lower BMI.
• Size and sex mismatch:
• A normal sized adult male (>70 kg) donor is suitable for most
recipients.
• With small donors, size matching with BMI or height is more accurate
than weight.
• Donor heart undersizing must not exceed a 30% mismatch in
predicted heart mass. See Box 13.2.
• Any donor undersizing is undesirable for recipients with known PH.
• Donor heart oversizing must also not exceed a 30% mismatch in
recipients where the pericardial space is likely to be restrictive (e.g.
multiple previous sternotomies, recent large myocardial infarction,
current LVAD in place).
• Sex mismatch downsides are observed mainly in male recipients who
are matched with female donors.

Box 13.2 Predicted heart mass

Predicted LV mass (g) a × height0.54(m) × weight0.61(kg)
where a =​6.82 for women and 8.25 for men
Predicted RV mass (g) a × age–​0.32(years) × height1.135(m) × weight
0.315
(kg)
where a =​10.59 for women and 11.25 for men
481

148 Chapter 13 Cardiac donor selection and management

Anticipated ischaemia times

• Distance from donor to transplant centre must be considered as longer
distances will lead to increased cold and total ischaemia times which are
important predictors for primary graft dysfunction.
• Ischaemia time is an independent risk factor for survival. Total ischaemia
times <4 hours are considered optimal.
• Longer ischaemia times are associated with higher recipient mortality
particularly in combination with other undesirable donor characteristics
(e.g. significant LVH).
Other considerations
Viral-​infected donors can be used in infected recipients and in selected non-​
infected recipients. In a recent trial (DONATE HCV), hearts procured from
HCV-​infected donors showed excellent graft function with undetectable
hepatitis C viral load at 6 months after transplantation (direct acting antiviral
agents were initiated within a few hours of transplantation).

Cardiac donor optimization

What is donor optimization?
Prior to BSD testing, the goal of donor management is directing at maxi-
mizing survival rather than specific optimization for individual organs.
Following BSD confirmation, the emphasis shifts from maintaining patient
survival to specific organ optimization for procurement and subsequent
transplantation. This is termed donor optimization.
Key facts
• Optimization of donor physiology increases the number of
transplantable organs by creating the optimal milieu for organ
preservation and recovery of function.
• Aggressive donor management protocols (PAC-​assisted haemodynamic
management, and hormonal/​metabolic therapy) can increase donor use
by 30%.
Donor optimization set-​up and goals
• Most patients will be in an ICU environment during confirmation of
BSD and, if not, they should be transferred after BSD is confirmed.
• Following initial stabilization, BSD confirmation and death certification,
and confirmation of consent, focus should be directed at specific organ
optimization.
• In the UK and elsewhere, Specialist Nurses for Organ Donation (SN-​
ODs), or similar practitioners, should be present at the donor centre to
assist in the delivery of donor care.
• In the UK specifically, specific cardiothoracic ‘scouting’ teams of trained
personnel are sent from the transplant centre to locoregional (within 2
hours’ travel distance) donor ICUs as soon as a potential heart donor is
identified. The scout team becomes in charge of subsequent donor care
and will perform serial diagnostic and therapeutic procedures, including
TTE/​TOE, PA catheterization, along with careful titration of vasoactive
drugs and fluid resuscitation.
 Circulatory and endocrine management 149

• In the USA, specialists from organ procurement agencies assume care

and manage the donor with input from transplant surgeons.
• Regional organ procurement centres are available in some regions
and assume management after donors are transported from outlying
hospitals; these centres relative to other hospitals are able to focus
entirely on donor management.
• The three primary facets of donor optimization include:
• Circulatory and endocrine management (including hormonal
resuscitation and management of diabetes insipidus).
• Respiratory management and lung recruitment procedures.
• General critical care management including temperature management
and thromboembolic prevention.
• Donor management aims (Box 13.3) to optimize preload (by achieve
an adequate effective circulating volume while avoiding fluid overload),
afterload (by adjusting vasoconstrictors and vasodilators) and cardiac
output (by normalizing cardiac function without relying on high doses
of beta-​adrenergic agonists or other inotropes that increase myocardial
oxygen demand), and to correct acidaemia, anaemia, hyponatremias
and respiratory gas exchange.

Box 13.3 Rule of thumb—​goals of donor optimization

‘Rule of 100’: systolic arterial pressure >100 mmHg, urine output >100
mL/​hour, PaO2 >100 mmHg, haemoglobin concentration >100 g/​L, and
a ‘blood sugar 100% normal’.

Circulatory and endocrine management

Regular clinical monitoring of response to treatment is essential. Fluids,
inotropes, and vasopressors should be adjusted every 15 minutes to achieve
target haemodynamic values listed here with alpha-​adrenergic agonist use
kept to a minimum:
• Sinus rhythm 60–​100 bpm.
• MAP 60–​80 mmHg.
• CVP 4–​12 mmHg.
• PCWP 8–​12 mmHg.
• CI >2.4 L/​min/​m2.
• SVR 800–​1200 dyne/​s1/​cm5.
• SvO2 >60%.
• Urine output 0.5–​2.0 mL/​kg/​hour.
Monitoring and investigations
Adequate monitoring equipment should be installed, and these include:
• Invasive arterial and central venous pressure lines.
• PA catheterization and cardiac output monitoring particularly if
LVEF <45%.
• Continuous ECG monitoring.
• Urinary output measurement.
Investigations
• Blood investigations—​renal function and troponin measurements
particularly if there is a history of a preceding cardiac arrest.
501

150 Chapter 13 Cardiac donor selection and management

• Twelve-​lead ECG (see Cardiac donor assessment and selection,

p. 141).
• CXR following recruitment procedures.
• Serial TTE/​TOE is essential in assessing cardiac function and responses
to treatment, and in excluding any significant structural abnormalities
(see Cardiac donor assessment and selection, p. 141).
Management strategy
See Fig. 13.1.
• Restoring adequate circulating volume should be the primary priority
initially:
• Correct hypovolaemia with fluid boluses (3–​5 mL/​kg of a balanced
crystalloid or colloid as a rapid bolus; repeat as necessary) while
avoiding fluid overload.
• No specific class of fluid resuscitation has been proven to be superior
to the rest but some have expressed concerns about the use of
starch-​based products due to their potential association with graft
dysfunction.
• Restoring vascular tone:
• If hypotensive despite restoration of circulating volume, commence
vasopressin where vasopressin is required and wean catecholamine
vasopressors as able.
• Restoring cardiac contraction:
• If hypotensive despite restoration of circulating volume and vascular
tone, introduce inotropic therapy.
• Myocardial stunning is not uncommonly encountered and reversal can
be achieved with supportive therapy.
• If LVEF <45% then PAC placement and hormonal resuscitation is
strongly recommended (see later in list).
• Dopamine is considered first line when inotropic therapy is required
followed by dobutamine.
• Use of high-​dose catecholamine therapy, especially norepinephrine, is
associated with impaired graft function and a reduced rate of cardiac
retrieval. Catecholamines should therefore be weaned as soon as
possible in favour of dopamine or dobutamine.
• Hormonal resuscitation:
• Inotropic requirements may be reduced with hormonal therapy, and
hormonal resuscitation should be considered in all donors.
• If response to catecholamine therapy is suboptimal, consider a trial
of IV hydrocortisone (50–​100 mg).
Triiodothyronine (T3; 4 mcg bolus at the onset of the infusion
followed by an infusion of 3 mcg/​hour) use has been associated
with increased organs retrieval rates and improved organ function
post-​transplantation. Some advocate its use only in patients with
persistently impaired cardiac performance despite initial measures
of volume resuscitation and vascular tone restoration, while others
advocate its use as first-​line treatment.
• Administer IV methylprednisolone (15 mg/​kg, max 1 g) to reduce the
general proinflammatory response and reduce the accumulation of
extravascular lung water.
 Circulatory and endocrine management 151

• Start insulin infusion to ensure good glycaemic control with the added
anti-​inflammatory benefits including reduced cytokine release:
• Keep blood sugar at 4–​10 mmol/​L (70–​180 mg/​dL) (minimum 1
unit/​hour).
• To avoid hypoglycaemia, add a continuous infusion of 20% or 50%
dextrose at 25 or 10mL/​hour, respectively.

Fig. 13.1 Crystal City algorithm for cardiac donor optimization.

Reproduced from Zaroff JG, Rosengard BR, Armstrong WF, et al. Consensus conference re-
port: maximizing use of organs recovered from the cadaver donor: cardiac recommendations,
March 28–​29, 2001, Crystal City, Va. Circulation. 2002 Aug 13;106(7):836–​41. doi: 10.1161/​
01.cir.0000025587.40373.75 with permission from Wolters Kluwer.
521

152 Chapter 13 Cardiac donor selection and management

• Identification and management of diabetes insipidus:

• Diabetes insipidus should be identified and treated early.
• Dramatic unprovoked rises in urine output should be treated as
suspected diabetes insipidus even before confirmatory urine and
plasma electrolytes levels and osmolalities are obtained.
• Confirmatory tests include a rising serum sodium level, urine specific
gravity <1.005, a urine osmolality >295 mOsm/​kg.
• If >4 mL/​kg/​hour, consider diabetes insipidus and treat promptly
with vasopressin/​terlipressin and/​or desmopressin (DDAVP). Dose
of DDAVP 1–​4 mcg IV titrated to effect.
• Simvastatin therapy:
• Emerging evidence supporting its use in donor optimization.
• A preclinical study revealed that donor simvastatin treatment
enhanced endothelial-​pericyte integrity, and inhibited ischaemia-​
reperfusion injury and the development of allograft vasculopathy after
heart transplantation.
• In a recent double-​blinded randomized controlled trial, donor
nasogastric simvastatin (80 mg) administration within 2 hours of
BSD confirmation reduced the number of treatments with IV pulsed
steroids, plasmapheresis, antithymocyte globulin, or intravenous
immunoglobulins (IVIGs) for haemodynamically compromised
rejection episodes in the first 30 days after heart transplantation.
There was, however, no effect on the incidence, or severity, of
biopsy-​proven acute cellular or antibody-​mediated rejections.

Respiratory management
Regular clinical assessment and monitoring response to treatment is essen-
tial. The following parameters should be monitored and targets achieved:
• SpO2 >95%.
• PaO2 10.0 kPa (75 mmHg) (FiO2 <0.4 as able).
• PaCO2 5–​6.5 kPa (35–​45 mmHg) (or higher as long as pH >7.25).
• Acidosis (pH 7.40–​7.45).
Monitoring and investigations
• Continuous SpO2 monitoring.
• Intermittent blood gas analysis.
• Recent CXR.
• Microscopy and culture of airway secretions.
• Bronchoscopy (diagnostic or therapeutic) if clinically indicated or as part
of lung assessment for lung retrieval (see Chapter 25).
Management strategy
• Recruitment manoeuvres normally employed at each ICU should be
used. These should be repeated after any suctioning procedure.
• Tidal volumes should be 4–​8 mL/​kg of ideal body weight.
• Peak inspiratory pressures should be limited to <30 cmH2O.
• Positive end-​expiratory pressure (PEEP) should be maintained at 5–​
10 cmH2O to reduce airway collapse and improve alveolar recruitment.
 General critical care 153

• FiO2 should be adjusted to maintain PaO2 of >8.0 kPa and maintain

an SpO2 of 92–​95%. The minimum amount of FiO2 should be used to
reduce absorption atelectasis and risk of oxygen toxicity.
• Alter minute ventilation (respiratory rate and tidal volume) to allow
permissive hypercarbia while achieving a pH of >7.25. This minimizes
ventilator-​associated lung injury.
• Measures to improve airway clearance and reduce risk of aspiration
include:
• Monitoring endotracheal tube position and cuff pressures. Remember
cuff leaks lead to lung de-​recruitment and increased risk of aspiration.
High pressures can lead to mucosal injury.
• Raised head-​up position (30–​45°)—​also improves functional residual
capacity.
• Turning, suctioning, and regular physiotherapy.
• If secretions are viscous or purulent, consider regular nebulized
hypertonic saline and/​or bronchoscopy Antibiotics should be
administered if clinically indicated (purulent secretions).

General critical care

• Regular turning to minimize the pressure sores.
• Review all drugs being administered and discontinue those that are not
required.
Nutrition
• Start enteral feeding unless otherwise instructed by the procurement
team:
• Improves electrolyte balance, glycaemic control, and splanchnic
perfusion.
Thromboembolic prevention
• Heparin or low-​molecular-​weight heparin.
• Thromboembolic deterrent (TED) stockings.
• With or without calf compressors.
Haematological management
• Target haematocrit ≥30% and haemoglobin ≥10 g/​dL.
• If haemoglobin levels are <10 g/​dL, blood transfusion can be
considered. There is, however, evidence suggesting that blood
transfusions can adversely affect organ function post transplantation.
• Adequate group and save sample should be present prior to any
procurement procedure.
• Any coagulation derangement should be treated particularly if there
is ongoing bleeding. Fresh frozen plasma and platelets should be used.
Antifibrinolytics may cause microvascular thrombi and should generally
be avoided.
Temperature management
• Aim for a core body temperature between 36°C and 37.5°C.
• Hypothermia is commoner than hyperthermia and can be managed by:
541

154 Chapter 13 Cardiac donor selection and management

• Using active warming blankets.

• Warming administered fluids.
• Heating and humidifying inspired gases.
• Increasing ambient temperatures.

Donor risk scoring systems

• Key determinant of success has more to do with the health of the
recipient than the donor.
• As highlighted previously, there are multiple donor factors that increase
the risk of post-​transplant mortality.
• Multivariate donor risk scores are routinely used in kidney
transplantation (Kidney Donor Profile Index (KDPI) to improve donor
selection, reduce discard rates, and improve outcomes).
• Although in the context of cardiac donor assessment and selection, the
formal use of donor risk scores is not yet routine, their components
highlight the important donor determinants of outcomes and have been
shown to be predictive.

Mortality risk scores

There are multiple mortality risk scoring systems that have been de-
vised. The most detailed is the Eurotransplant Heart Donor Score (HDS)
(Table 13.2):
• Consists of ten variables—​age, cause of death, history of malignancy,
sepsis, drug abuse, meningitis, or positive virology, hypertension, cardiac
arrest, LV function, valvular function, LVH, coronary angiogram, serum
sodium, norepinephrine therapy, and dopamine/​dobutamine therapy.
• Predicts 3-​year mortality.
• Scores above the median were associated with a higher organ non-​use
(35%) versus lower scores (7%).
• Each HDS point above 17 increased the predicted 3-​year
mortality by 4%.
• Only donor age and LVH, as individual donor variables, were statistically
significant predictors of mortality at 3 years after heart transplantation.

Deficiencies with using risk scores

• Not widely used.
• Do not take into account the enlarging patient population with long-​
term LVAD as a BTT.
• They do not combine recipient factors, including expected waiting list
mortality, and donor-​related factors, including risk of primary graft
dysfunction.
 IMPACT OF DONOR FACTORS ON TRANSPLANTATION FROM LVAD 155

Table 13.2 Eurotransplant Heart Donor Score (score ≥17 =​high risk)
Donor characteristics Range of points
Age 1 (age <45 years) to 11 (age ≥60 years)
Cause of death 1 (drugs, sepsis, subarachnoid bleeding,
meningitis, head trauma) to 7 (carbon monoxide
intoxication)
History of malignancy, 1 (no) vs 19 (yes)
sepsis, drug abuse,
meningitis, or positive
virology
Hypertension 1 (no) vs 2 (yes)
Cardiac arrest 1 (no) vs 2 (yes)
Left ventricular function 1 (EF >55%) to 22 (EF <45%)
Valvular function 1 (normal) to 7 (abnormal)
Left ventricular hypertrophy 1 (10 mm) to 4 (>14 mm)
Coronary angiogram 1 (no stenosis) to 70 (>1-​vessel stenosis)
Serum sodium 1 (<130 mmol/​litre) to 3 (≥170 mmol/​litre)
Norepinephrine therapy 1 (<0.1 mcg/​kg/​min) to 5 (>0.8 mcg/​kg/​min)
Dopamine/​dobutamine 1 (<5 mcg/​kg/​min) to 2 (>10 mcg/​kg/​min)
therapy
Reproduced from Kransdorf EP, Stehlik J. Donor evaluation in heart transplantation: The
end of the beginning. J Heart Lung Transplant. 2014 Nov;33(11):1105–​13. doi: 10.1016/​
j.healun.2014.05.002 with permission from Elsevier.

Effect of donor risk factors on outcomes

of transplantation in recipients
supported with long-​term LVADs
• Paucity of data on the impact of donor-​related variables on outcomes
post transplantation in patients bridged with durable LVADs.
• In a prospective cohort study, LVAD recipients, whether transplanted
with high-​or low-​risk donors (on the basis of age >50 years, history
of diabetes, three or more inotropic agents at the time of incision, and
BUN >25 mg/​dL), exhibited worse early (but not late) survival and
higher early complication rates than standard recipients.
• Adverse donor characteristics were less predictive outcomes than in
standard recipients.
561
 Chapter 14 157

Extended criteria
heart donors and
perfusion storage
Introduction 158
Marginal donors: definition and outcomes 158
OCS Heart 160
Use and system set-​up 162
At the procurement site 163
Clinical trials 168
Conclusion 169
581

158 Chapter 14 Extended criteria donors and perfusion

Introduction
As the pool of cardiac donors is limited and the number of patients suf-
fering from HF is increasing, there is an ever-​growing imbalance between
the demand and supply of donor hearts available for transplantation. Thus,
there is mounting pressure to consider donors who are suboptimal. Over
the years, a distinction has emerged between the term ‘high-​risk donors’,
for example, people with HCV or drug users who may be infected with
contagious viral infections such as HIV, and the term ‘marginal donors’ that
reflects either the donor characteristics or the cardiac parameters. As a
consequence, we are constantly revising what should be considered as
‘good enough’ to transplant rather than waiting for the ‘ideal’ organ.
Donor selection is probably one of the most complex decisions that
has to be taken in the transplantation process, reflecting both the recipient
medical condition and urgency, how good or marginal the donor organ is,
and the institutional risk tolerance. Furthermore, a known association exists
between time spent on the waiting list and mortality, thus contributing to
the complexity of matching donors to recipients. Although early experi-
ence of using marginal donors for marginal recipients on an ‘alternate list’
resulted in reduced post-​transplant survival, studies from the last 5–​10 years
examining the use of marginal organs demonstrated results which are non-​
inferior to using ‘ideal organs’. In this chapter, we will review the current
definition of a marginal donor heart and the outcomes of transplanting such
hearts. We shall also discuss the latest technology available to try and miti-
gate some of the risks associated with the use of these organs.

Marginal donors: definition

and outcomes
Throughout the 1980s when heart transplantation was increasingly
adopted, there was consensus on the required quality of donor hearts for
optimal outcomes. Over the subsequent decade, the demand on heart
transplantation continued to rise due to the increasing number of HF pa-
tients, forcing HF teams to consider what was termed ‘marginal donor’
hearts, for example:
• LVH.
• Reduced LV function.
• Prolonged CPR.
• Advanced age.
Recent reports on studies analysing the UNOS database were able to miti-
gate some of the original concerns about these donors. For example, re-
cipients of donor hearts without LVH (<1.1 cm), with mild LVH (1.1–​1.3
cm), and moderate–​severe LVH (≥1.4 cm) had equivalent 30-​day to 3-​year
survival. However, subgroup analyses showed an increased risk of death
in recipients of allografts with LVH and donor age >55 years or ischaemic
time ≥4 hours, attesting in favour of the ‘one hit’ theory and pointing out
the need for something else in order to mitigate the risk completely (e.g.
perfusion systems).
 Marginal donors: definition and outcomes 159

Sibona and colleagues examined the impact of LV dysfunction using the

UNOS database between January 2000 and March 2016. Among 31,712
heart transplant recipients, no significant differences were found in post-​
transplant survival up to 15 years of follow-​up between those receiving
donor hearts with or without LV dysfunction. When using Cox regression
analysis with adjustment for propensity variation, EF was not found to have
any significant impact on mortality when analysed as a categoric or con-
tinuous variable and LVEF normalized 1-​year post transplantation. These
findings support the notion of the neurogenic stunned myocardium, trig-
gered by the high catecholamine release that is typical after brain death,
hence causing reversible LV dysfunction. For recipients receiving donor
hearts with a history of CPR (even exceeding 30 minutes), post-​transplant
outcomes were not found to be inferior.
Older donor age used to be a common reason for concern in the past.
However, a recent study reported that there is no survival disadvantage
in recipients of older donor hearts (age ≥50 years), albeit with a higher
incidence of cardiac allograft vasculopathy at 5 years post transplantation.
In order to facilitate the use of marginal hearts and reduce the related
risk, perfusion systems may have a role in their retrieval. This will be re-
viewed later in the chapter.
Ex vivo organ perfusion
Ex vivo perfusion of isolated hearts has been under investigation for
>150 years. Following the pioneering experimental works of Ludwig and
Cyon with frog hearts in 1866, Langendorff and Martin independently de-
veloped methods to perfuse mammalian hearts at the end of the nineteenth
century. However, it wasn’t until the middle of the last century that Morgan
and Neely converted Langendorff ’s model into a working heart model.
From a transplant perspective, ex vivo heart perfusion with oxygenated
and nutrient enriched blood has been developed during the past 20 years
with the aim to reduce the cold ischaemic time and to evaluate metabolism
and function of preserved grafts prior to transplantation. In times of donor
organ shortage, the possibility of organ procurements at geographically dis-
tant areas and assessment of hearts from extended criteria donors and do-
nation after circulatory determined death (DCD) are essential to increase
organ availability for patients on the heart transplant waiting list. A major
difference compared to standard cold storage is that ex vivo perfusion is
performed using normothermic donor blood and thus minimizes the cold
ischaemic time and avoids another ‘hit’ on the retrieved heart. This may
be of potential benefit when marginal donor hearts are being considered.
Currently, the only system commercially available with the largest clin-
ical experience is the Organ Care System (OCS) Heart (TransMedics,
Andover, MA, USA) (see following section). It is hypothesized that using
such a system for marginal donor hearts may reduce the risk associated
by abiding to the ‘one-​hit theory’ and taking the cold ischaemic time out
of the equation.
601

160 Chapter 14 Extended criteria donors and perfusion

OCS Heart
The OCS Heart is a portable ex vivo perfusion platform for retrieved donor
hearts in which the heart is beating and metabolically active. While there
have been two reports of out-​of-​the-​body times exceeding 10 hours, the
median duration for OCS Heart perfusion is 4–​5 hours. The system consists
of a durable console with a wireless monitor and a disposable perfusion
module (Fig. 14.1). The latter includes an organ chamber, a perfusion circuit,
a gas exchanger, a heater, a blood reservoir, and a pulsatile pump (Fig. 14.2).

Fig. 14.1 Portable OCS Heart console with the disposable module.

 OCS Heart 161

Warmer

O2

SvO2
probe
Gas
exchanger

Reservoir
To chiller

Pulsatile pump
Fig. 14.2 Schematic of the OCS Heart module components.

The OCS Heart system basically represents a modified non-​working

heart Langendorff perfusion model. In other words, the heart is beating
but the LV is not volume loaded. Blood from the OCS reservoir is pumped
through an oxygenator and heater in a pulsatile manner into the ascending
aorta of the retrieved donor heart via an aortic cannula. The competent
aortic valve remains closed and the oxygenated blood is directed into the
coronary arteries to perfuse the myocardium. Venous effluent from the
coronary sinus collecting in the RA (IVC and SVC are temporarily closed
for OCS perfusion), fills the RV during diastole and is ejected across the
pulmonary valve into the PA. A cannula in the pulmonary trunk directs the
venous blood back to the reservoir.
There is a line that connects the aortic cannula with the reservoir for
de-​airing and a port for administration of cardioplegia administration just
prior to removing the heart from the OCS Heart. A vent is placed across
the MV into the LV to ensure that it remains unloaded. The vent is allowed
to drain passively into the organ chamber of the OCS Heart and then back
to the blood reservoir.
621

162 Chapter 14 Extended criteria donors and perfusion

Use and system set-​up

While evaluating the donor, a decision should be made if the OCS Heart
is to be used. The OCS Heart system may be considered in high-​risk heart
transplants, for example:
• An expected longed ischaemic time:
• Long travel time.
• Surgical preparation of recipient is expected to be long, usually due to
previous operation(s).
• Extended donor criteria:
• Presence of ventricular hypertrophy.
• Prolonged down-​time.
• Reduced EF.
• Older age.
• Special procurement conditions—​DCD.
Nevertheless, in some circumstances the OCS system cannot be used, for
example:
• Moderate aortic valve regurgitation.
• Myocardial contusion (often associated with thoracic trauma of the
donor).
• Unrepaired interatrial or interventricular septal defects including PFO.
System set-​up
Before setting off for the donor hospital, the procurement team is advised
to discuss with the coordinator at the donor hospital the following issues:
• Transfuse packed red blood cell to the donor if the haematocrit
is < 25%.
• Verify the latest inotropic doses and donor arterial lactate.
Since three different proprietary solutions are required to perfuse a
donor heart with the OCS Heart and other additives have to be prepared
and added to the fluids, it is advisable for the procurement team to use the
checklist that is provided by TransMedics before leaving the base hospital:
1. The priming solution has to be supplemented with 20 mEq of sodium
bicarbonate.
2. The maintenance solution containing adenosine requires the addition
of 50 IU of regular insulin.
3. The epinephrine solution is made up of 0.25 mg epinephrine and 30 IU
of regular insulin in 500 mL of 5% dextrose.
Other additions to the reservoir include:
• Ciprofloxacin: 100 mg.
• Cefazolin: 1 g.
• Methylprednisolone: 250 mg.
• Heparin: 10,000 units (required for the blood collection bag).
• Multivitamin: 1 unit.
• 25% albumin: 100 mL (especially recommended for longer OCS Heart
runs).
Two lots of cardioplegia solution are also required, one for flushing of the
donor heart before explant and one before removing the donor heart from
the OCS Heart. The type of cardioplegia that is used differs regionally.
Whereas the US experience was initially based on a variety of different
 At the procurement site 163

solutions, during the past 5 years most centres have been using the crys-
talloid portion of del Nido’s cardioplegia. Most European centres rely ei-
ther on histidine–​tryptophan–​ketoglutarate solution (HTK, Custodiol) or
St. Thomas formulations (Sterile Concentrate for Cardioplegia Infusion,
Martindale Pharma, UK).
The following checklists should be reviewed before setting off for the
donor hospital:
• Equipment checklist:
• The batteries are charged,
• The gas cylinder is sufficiently full.
• The data card is inserted properly.
• The defibrillator is operational.
• Handheld lactate analyser and cartridge available.
• External pacing box and pacing lead available.
• Medication checklist:
• Sodium bicarbonate (2 × 20 mEq).
• Methylprednisolone (250 mg).
• Adult multivitamin.
• Regular insulin (80 IU).
• Epinephrine (0.25 mg).
• 5% dextrose in water (500 mL).
• Heparin (10,000 IU).
• Calcium gluconate.
• 50% dextrose.
• Potassium chloride.
• Sterile water for injection.
• Disposable checklist:
• BGA syringes.
• Needles.
• Alcohol wipes.
• Gloves.
• Disposal bag.
• 9 V batteries.
• LV vent cannula.
• Organize proper transportation vehicles according to the material
(running bag, OCS module box, OCS console).
It is advisable to prepare the priming, the maintenance, and the epineph-
rine solutions before setting off for the donor hospital. However, the OCS
module should be kept sterile until after donor heart inspection and the
final decision has been made to retrieve the heart for OCS perfusion and
transplantation.

At the procurement site

The donor heart should be assessed as described in Chapter 16. Briefly, this
should include the four Cs:
• Conduction.
• Contraction.
• Contusion.
• Coronaries.
641

164 Chapter 14 Extended criteria donors and perfusion

Once the decision has been made to accept the donor heart for OCS per-
fusion, preparation is made for donor blood collection. It is crucial to refrain
from intensified vasopressor therapy immediately before and during blood
collection because this could lead to graft failure during OCS perfusion.
When all the procurement teams are ready to proceed with organ pres-
ervation, a two-​staged venous cannula is placed into the RA. Alternatively,
a 24 Fr cannula is placed into the proximal aortic arch. The canula is con-
nected to the blood collection bag placed at floor level for passive blood
drainage. All retrieval teams are reminded not to commence perfusion of
their respective organs until donor blood collection is complete. A min-
imum of 1.2 L of donor blood is required and this process can take up to
90 seconds. In some cases, the Trendelenburg position is required to collect
the required amount of donor blood. The arterial blood pressure will inev-
itably decrease during blood collection and it is paramount that the donor
is not given boluses of vasoconstrictors during this time. Once blood col-
lection is complete, the vascular cannula is removed, the ascending aorta is
clamped, and cardioplegia infusion is started as usual (0.5–​1 L for the heart).
Priming of the OCS module should be performed simultaneously with
donor blood collection.
Preparing the heart for the OCS Heart chamber
After the donor heart has been explanted, it is placed in a bowl of cold
cardioplegia solution for back table preparation. The SVC is tied and the
IVC (and PFO if present) is oversewn. This step can also be performed after
OCS perfusion is initiated in order to reduce ischaemia time.
The aorta is transected just below the cardioplegia cannula site and four
double pledgetted sutures are inserted along the cut edge. There are four
different sizes of aortic cannulas. After inserting the optimal sized cannula,
a cable tie is tightened below the pledgetted sutures and the excess sutures
are trimmed (Fig. 14.3 and Fig. 14.4).
Donor heart perfusion in the OCS Heart
1. The donor heart is positioned in the perfusion chamber with the back
of the LA facing the surgeon. The aorta is positioned at 12 o’clock, the
SVC at 1 o’clock, the IVC at 5 o’clock, and the PA at 11 o’clock.
2. As the aortic cannula is approximated with the perfusion port of the
heart chamber, the OCS Heart is set to an aortic flow rate of 0.25 L/​
min, taking care to de-​air the aorta before completing the connection.
The winged nut is tightened to secure this connection and the aortic
flow can be increased back to the target flow rate of 1 L/​min.
3. At this point, the donor heart is perfused and the perfusion clock has
to be started on the OCS monitor.
4. A vent cannula is placed across the MV into the LV and suture secure
along the LA edge.
5. To prevent distension of the perfused donor heart, it is necessary to
perform manual massage before the heart starts to beat.
6. The electrode pads in the organ chamber are positioned behind the
RA and LV. The external defibrillator is connected and when the ECG
is detected, DC cardioversion can be performed if required starting
with 10 J.
 At the procurement site 165

(a) (b)

Ao

PA
Ao

(c)

(d)

Fig. 14.3 Aortic cannulation for OCS perfusion. There is only one size for the
pulmonary cannula. A purse string and umbilical tape secure the cannula position
(Fig. 14.4).

7. The donor heart can be paced with the electrode pads or separately
inserted epicardial pacing wires if the native heart rate is <80 bpm.
8. After the heart starts beating, the maintenance solution should be
delivered in ‘AUTO AOP’ (aortic pressure) mode with a target AOP
of 75–​80 mmHg.
9. The PA cannula is connected to a port in the lower right corner of the
OCS Heart chamber; the LV vent is secured on the left-​hand border
of the perfusion chamber.
10. The OCS temperature is set up to 34°C.
11. The positions of the electrode pads are adjusted to ensure good
tissue contact in case further DC shocks are required during
transportation. These pads can be sutured to the foam backing of the
organ chamber if required.
61

166 Chapter 14 Extended criteria donors and perfusion

(a)

(b)

Fig. 14.4 PA cannulation.

Initial stabilization is normally accomplished by the following configuration:

• Pump flow: 1 L/​min.
• Coronary flow: 700–​800 mL/​min.
• Mean AOP: 75–​80 mmHg in AUTO AOP mode.
In order to ensure adequate OCS perfusion in DBD donor hearts, paired
arterial and venous blood samples should be taken at repeated intervals for
lactate evaluation at the following time points:
• Baseline is the first OCS arterial lactate after priming with donor blood
but before insertion of the donor heart into the perfusion module.
• Ten minutes after perfusion initiation.
• Every 30–​60 minutes of OCS perfusion or 10 minutes after any manual
changes to the OCS setting.
For the myocardium, a variety of substrates can provide energy including
free fatty acids, glucose, and lactate. As fatty acids are not provided during
OCS perfusion, the donor heart would take up and metabolize lactate in-
stead. Thus, if perfusion is adequate, there will be a fall between the arterial
and venous lactates (i.e. arterial > venous) (Fig. 14.5). This phenomenon
 At the procurement site 167

is called ‘absorbing lactate’ and the lactate level would trend downwards
over time. On the other hand, if myocardial perfusion is inadequate, the
venous lactate would become higher than the arterial lactate and the heart
is said to ‘secrete lactate’. If left uncorrected, the lactate level would trend
upwards over time.

Art. Lactate
Ven. Lactate
Minimum of 1 lactate sample
prior to blood collection Glucose & electrolytes
Donor Management

Stabilization
OCS

Donor Blood OCS 0 30 60 90 120 150 180

Collection Priming
OCS Perfusion Time (min)
Fig. 14.5 Lactate evaluation scheme in the context of OCS heart perfusion and
‘trending down’ over time.

It is expected that the first lactate measurement after initiation of OCS

perfusion will be higher than the donor arterial lactate just before blood
collection. However, if heart perfusion is adequate, the lactate level should
trend down over time. For prolonged OCS runs, lactate levels may slightly
increase but without relevance for graft performance.
Pearls of OCS Heart perfusion
• Lactate values should always be analysed in the context of the latest
donor lactates. In DCD donors, interpretation of lactate levels (high
levels) is different from DBD donors.
• Lower venous lactates, compared to arterial samples, as well as stable
or decreasing lactate trend are indicative of adequate heart perfusion.
• The OCS Heart aortic flow and the adenosine-​rich maintenance solution
infusion rate are the main regulators for AOP and coronary flow.
• If AUTO AOP mode is not achieving targeted AOP, the mode
should be switched to manual. This can become necessary when the
maintenance solution infusion rate of >30 mL/​hour is required.
• Venous and arterial lactates should be assessed after no more than 10
minutes following any perfusion adjustments.
• A rapid decrease in the lactate level can be indicative of ‘over
perfusion’. The aortic flow should be reduced in order to avoid
myocardial oedema but maintaining perfusion pressure.
• By applying the ECG synchronization mode, adequate coronary flow
can be achieved with lower mean AOP. The circulatory pump is
synchronized with the ECG R wave. This mode is recommended for
longer OCS Heart runs.
• The decision to begin the patient explant operation should be based on
communication between the organ retrieval team and the implanting
surgeon regarding the perfusion status of the donor heart on the OCS.
681

168 Chapter 14 Extended criteria donors and perfusion

Table 14.1 Summarizers correlation between pump and maintenances solution

flow rates with aortic pressure and coronary flows

Aortic pressure (AOP) Coronary flow

Pump flow d/​i d/​i d/​i
Maintenance solution rate d/​i i/​d d/​i

Pre-​implantation procedure
The pre-​implantation procedure should be initiated when the recipient
surgeon is ready to transplant the donor heart. At this point, the OCS
Heart should be connected to a standard cooler. The cooler temperature
should be set to 34°C before being connected to the OCS Heart. The
aortic access port is de-​aired and the cardioplegic infusion line is connected.
The wireless monitor is undocked from the console, the lid of the organ
chamber is opened, and sterile drapes are attached. The cooling process
can then begin.
The OCS Heart temperature is set to OFF, the temperature on the
cooler is reduced by 10°C, and the pump flow reduced by 100 mL/​min.
This step has to be repeated once the monitor temperature is 2°C higher
than the cooler temperature. As the temperature decreases, heart action
will decline. Final arrest is induced at 14°C. The aortic vent has to be closed,
the aortic line is clamped, the OCS pump is stopped and the cardioplegia
infusion is started. The PA cannula is then disconnected to allow the right
heart to vent.

Clinical trials
The clinical trials that were performed thus far have established safety
of the OCS Heart. The Ex-​Vivo Perfusion of Donor Hearts for Human
Heart Transplantation (PROCEED II) study was a multicentre randomized
study comparing patients who were transplanted with donor hearts pre-
served on ice and those who were transplanted with donor hearts using
the OCS Heart. This showed equivalent outcomes regarding patient and
graft survival. However, significant differences in the OCS Heart group
included longer total preservation time and shorter cold ischaemic time,
demonstrating that the OCS Heart system enabled longer out-​of-​the-​body
time without any impact on cardiac function and survival. PROCEED II
demonstrated non-​inferiority of the OCS Heart compared with cold static
storage for standard donor hearts.
The International EXPAND Heart Pivotal Trial was a single-​arm study de-
signed to evaluate the effectiveness of the OCS Heart to recruit, preserve,
and assess donor hearts that may not meet current standard donor heart
acceptance criteria for transplantation to potentially improve donor heart
utilization for transplantation. Of the 93 donor hearts placed on the OCS
Heart, 75 (81%) were transplanted with 95% recipient survival at 30 days
after transplantation. At 6 months, survival was not statistically different
when compared to UNOS national average (87.9% vs 92.8%). The inci-
dence of ISHLT-​defined moderate or severe PGD was low (14.7%) for this
extended criteria donor population.
 Conclusion 169

Conclusion
The ever-​growing HF population has driven heart transplant specialists
to embrace marginal donor organs and gain experience to achieve op-
timal post-​transplant outcome. Restrictions on age, ‘down’ time, LVH,
comorbidities, as well as reversible LV dysfunction have all been relaxed.
Several recent registry database analyses have consistently shown non-​
inferiority for most of the extended criteria which is reassuring. Although,
this chapter does not cover ‘high-​risk’ donors, similar non-​inferior out-
comes have been reported.
Going forward, more experience will be gained using novel perfusion
systems for retrieved donor hearts and subgroups of marginal organs that
can truly benefit from these systems will be defined. Other subgroups of
marginal donor hearts that do not require the use of these systems will also
be explored. One should remember that as these systems are new, the limi-
tations attached to their use (e.g. organ oedema) are still under evaluation
and longer follow-​up will shed light on their efficacy and cost-​effectiveness.
701
 Chapter 15 171

Donor heart
procurement: donation
after brain death and
donation after circulatory
determined death
Introduction 172
Donation after brain death heart procurement with cold
storage 172
Normothermic machine perfusion 177
Donation after circulatory death heart procurement 180
721

172 Chapter 15 Donor heart procurement: DBD and DCD

Introduction
The world’s first successful heart transplant performed by Christiaan
Barnard at the Groote Schuur Hospital, Cape Town, South Africa, in 1967
was from a DCD donor as brain death was not legally recognized at the
time. Even though criteria for defining brain death were proposed as early
as 1968, it was not until the late 1970s that the definition became widely ac-
cepted. In 1967, Barnard placed the asystolic donor on CPB following dec-
laration of death and cooled the heart down to 16°C before transplanting
it into the recipient in an adjoining operating theatre. Accordingly, the is-
chaemic times were minimal in comparison to today’s practice. Although
there was an initial flurry of heart transplant activity worldwide following
Barnard’s success, the outcomes were poor with a 1-​year survival rate of
15%. It was not until Stanford trialled the use of ciclosporin in 1980 that
1-​year survival following heart transplantation became acceptable. By the
arrival of ciclosporin, brain death criteria had become universally accepted
and controlled DBD heart donation with cold static storage became widely
practised with excellent results. Cold static storage with its simplicity and
relative reliability remains the universal technique practised today, with
>100,000 DBD hearts globally procured and transported this way.
Over the last two decades, as the demand for donor hearts increased, the
acceptance criteria for DBD hearts have been relaxed and there has been
a resurgence of using DCD hearts for transplantation. With these donor
hearts, the limitations of cold static storage have become apparent. In order
to overcome the challenges that these particular organs pose and transplant
them safely, strategies involving in situ and ex situ donor heart perfusion
have been embraced with the aim of minimizing ischaemic time and recon-
ditioning donor organs in order that successful outcomes can be achieved.

Donation after brain death heart

procurement with cold storage
Donor assessment
The donor demographics, history, consent, blood group, virology status,
brain death confirmation, ECG, CXR, and echocardiogram must be re-
viewed and confirmed prior to starting the procurement.
Ideally, a TOE should be undertaken to assess valvular pathology, LV and
RV function, and presence of congenital heart disease.
In some countries, a coronary angiogram will be undertaken in donors
>40 years old or those with risk factors.
If the donor has not undergone formal right heart catheterization, a PAC
can be inserted to assess RA pressure, PA pressures, PCWP, and cardiac
output.
Inotropes and filling status should be adjusted to achieve the following
parameters while minimizing the vasoconstrictor requirement:
• Mean systemic arterial blood pressure >60 mmHg.
• RA pressure <12 mmHg.
• PCWP <12 mmHg.
• CI >2.5 L/​min/​m2.
• EF >50%.
 DBD HEART PROCUREMENT WITH COLD STORAGE 173

Team brief and communication

Introductions and a team brief with members of staff from the donor
hospital, donor coordinator, and abdominal procurement team should be
undertaken. If separate cardiac and pulmonary teams are involved, a clear
strategy should be discussed. If machine perfusion technology is involved,
for either heart or lungs, measures should be taken in order to ensure length
of PA, and/​or the need of reconstruction with thoracic aorta. The cardiac
retrieval team should be aware of potential specific considerations related
to donor, that is, redo surgery which will impact potential cross-​clamp time,
and need for long vascular structures (SVC, aorta, PA). The WHO surgical
checklist should be followed, giving particular importance to preoperative
broad-​spectrum antibiotics, methyl prednisolone administration, and avail-
ability of blood products should they be required. Whether the abdominal
procurement team requires the thoracic team to place a cross-​clamp on
the descending thoracic aorta during abdominal organ perfusion should be
established in advance, and a clear strategy for venting of abdominal organ
preservation fluid should be agreed upon.
Surgical preparation
• A median sternotomy is undertaken with meticulous attention to
haemostasis as it may be 4–​6 hours before the donor aorta is cross-​
clamped to allow for dissection time in complicated recipients.
• The thymus is either removed or divided in the midline. The innominate
vein is identified and divided between ligatures to allow the sternum to
be retracted fully without stretch on the vessel.
• The pericardium is opened with an inverted T incision and the
pericardial edges hitched up.
• An intraoperative assessment of the heart is made. During assessment
the heart can become irritable. Manipulation should be undertaken
carefully with minimal disturbance to the heart. Internal defibrillation
paddles should be available.
Visual inspection
• Overall heart size is assessed to determine the presence of
cardiomegaly or LVH.
• Size of LA and RA are assessed determining filling status and evidence
of poor cardiac function.
• The LV and RV are assessed for corkscrew-​like contraction. A gentle lift
of the apex with the palm of the right hand or swab enables this.
• Look for evidence of myocardial contusion or prior infarct.
• Rule out congenital abnormalities, such as left-​sided SVC, partial
anomalous left-​sided drainage, etc.
• Assess the size of the ascending aorta.
Palpation
• CAD: palpation of the entire coronary artery tree should be made in
order to assess CAD paying particular attention to the left main stem.
This is good practice, even if coronary angiography was performed.
• Aorta: the ascending aorta and aortic root should be palpated to
exclude calcification and to detect for palpable thrills which should alert
the surgeon to the possibility of aortic stenosis.
741

174 Chapter 15 Donor heart procurement: DBD and DCD

• If it has not been possible to insert a PAC, a transduced needle can be

placed directly to measure LA and RA filling pressures as well as the PA
pressure.
• If right side chambers are unusual enlarged and there is a suspicion of
right-​to-​left shunt, a direct blood gas sample should be taken from the SVC
and PA, looking for step-​up oxygenation. Ensure ventilation of FiO2 <50%.
Once the assessment has been made, findings should be relayed to the ac-
cepting transplant centre and a provisional cross-​clamp time agreed upon.
Preparation for cross clamp
• The distal ascending aorta is dissected from the PA and encircled with a
nylon tape.
• Traction of the nylon tape towards the left facilitates access to the SVC
and azygos vein.
• The pericardium over the anteromedial aspect of the SVC is incised to
the level of the divided innominate vein.
• The SVC is dissected free circumferentially, separating it from the right
PA which is directly posterior.
• The azygos vein can then be found draining into the posterior aspect
of the SVC cranial to the right PA—​this can be divided between heavy
ligatures although others may wish to divide it immediately prior to
cross-​clamping as it can tear easily and be difficult to control.
• The SVC is encircled with a silk tie at the level of the right subclavian
vein to be snared down upon later.
• The IVC should be dissected out and encircled with a nylon tape. The
diaphragmatic pericardial reflection should be carefully dissected from
the IVC freeing up additional length.
• If the heart–​lung block needs to be split, Sondergaard’s groove should
be incised to extend the length of the right-​sided pulmonary vein cuff.
Cross-​clamp
Permission should be sought to proceed to cross-​clamp from the accepting
recipient transplant centre, and ensure that the abdominal team and donor
coordinator are aware and happy to proceed.
• Administer 300 IU/​kg of heparin IV and allowed to circulate for
5 minutes. The Swan–​Ganz catheter should be removed and
the sheath and the central venous line pulled back out with the
intrapericardial SVC.
• An infusion cannula should be placed into the ascending aorta as
cranially as possible and secured (some surgeons advocate division
of the innominate artery to facilitate more length on the ascending
aorta). The cardiac preservation fluid should be primed, de-​aired, and
connected up to the cannula.
• Once the abdominal team have cannulated and agreed to proceed to cross-​
clamp, the SVC should be occluded by snugging down on the SVC silk tie.
• The intrapericardial IVC is clamped and almost completely transected
on the heart side of the clamp, enabling right-​sided venting of the IVC.
Alternatively, the IVC can be divided and a sucker placed in the right
pleura to vent the abdominal flush and also vent the right side of the
heart. Whichever method is adopted, care must be taken to ensure that
enough IVC is left to provide adequate margin from the coronary sinus.
 DBD HEART PROCUREMENT WITH COLD STORAGE 175

• After 5–​10 beats to facilitate emptying of the heart, a clamp is applied

across the distal ascending aorta and the cardiac preservation fluid is
infused into the isolated aortic root (Fig. 15.1).
• The heart is lifted out of the pericardium and the left inferior pulmonary
veins are divided (if heart retrieval alone) or the LA appendage tip is
divided (if retrieving the heart–​lung block) to vent the left heart. It must
be stressed that the venting site needs to be large enough so that the
heart does not distend when flushing with heart (and lung) preservation
solution(s).

Fig. 15.1 Cannulation of the ascending aorta and venting sites of the DBD
donor heart.

• Vigilance must be maintained to ensure that the aortic root is

pressurized while being alert to avoiding ventricular distention. The
retrieval surgeon may choose to increase preservation fluid volume
pending heart size and predicted prolonged ischaemic time. Cold saline
is applied for topical cooling of the heart.
761

176 Chapter 15 Donor heart procurement: DBD and DCD

• Following completion of the cardiac preservation fluid infusion, the IVC

is transected fully. For heart-​only retrieval, the four pulmonary veins and
the two PAs are individually divided at the level of the pericardial edges.
• If the heart and lung block is to be split, the heart is lifted from the
pericardial well following division of the IVC and an incision made in the LA
10 mm medial to the confluence of the left superior and inferior pulmonary
veins. An identical incision is then made in the LA 10 mm medial to the
confluence of the right superior and inferior pulmonary veins. These two
incisions are the joined in a circular fashion giving rise to the LA cuff to the
heart while paying particular attention to allow enough pulmonary vein to
go with the lungs. Ensure the dissection plane is visualized and clear, looking
both inside and outside the LA during dissection. Care should be taken
when pulling the heart up, as the dissection plane might jeopardize the right
pulmonary veins, in particular the inferior.
• The pulmonary trunk is then divided 2 cm distal to the pulmonary valve
(Fig. 15.2).

Fig. 15.2 Division of the major vascular structures.

 Normothermic machine perfusion 177

• The aortic clamp is removed and the proximal aortic arch is divided
(some surgeons will divide the arch vessels individually and take the
aortic arch with the donor heart).
• Finally, the SVC snugger is removed, and the vessel is palpated to ensure
it is free from catheters before it is divided as cranially as possible and
beyond the divided azygos vein.
Inspection and packing for cold storage
• The heart is placed in a bowl of cold saline solution on the back table
and examined for abnormalities or damage. The IVC cuff is assessed to
ensure that there is no de-​roofing of the coronary sinus and to exclude
a left-​sided SVC. The SVC is inspected to confirm that there is sufficient
length while the fossa ovalis is probed to look for a PFO. The aortic
valve is inspected through the ascending aorta to ensure it is trileaflet
and free of calcification and vegetation.
• The heart is then triple bagged with 2 L of sterile cold saline or
cardiac preservation fluid in each bag being placed in the ice box. The
documentation is then signed, and lymph nodes and spleen obtained
before the heart is sent.

Normothermic machine perfusion

The Organ Care System (OCS) Heart (TransMedics, Andover, MA, USA)
is currently the only device commercially available for isolated donor heart
perfusion, receiving its CE mark in 2006 (Fig. 15.3).
The OCS Heart has been assessed in four clinical trials to date, with the
ongoing US EXPAND trial for extended criteria donors. The most notable
study was the PROCEED II trial where standard DBD hearts were ran-
domized to either the OCS Heart device or cold static storage. Although
there was no proven advantage in clinical outcomes, it revealed that the
preservation time could be extended by an additional 2 hours compared to
cold storage alone. This significant advantage has been reinforced by case
reports of successful DBD heart transplantation when the heart has been
perfused on the device for >10 hours. The early EXPAND trial results re-
veal that in a cohort of high-​risk donors, OCS Heart-​perfused hearts re-
sulted in a rate of primary graft dysfunction which was reduced relative to
a predetermined performance standard.
In the current configuration, the OCS Heart perfuses the retrieved
donor heart with oxygenated donor blood enriched with nutrients at
normothermia. Although the donor heart is beating, the LV is not volume
loaded and, therefore, it is not ejecting. This is an important limitation as it
is not possible to assess donor heart contractility (Fig. 15.4).
Technique of normothermic machine perfusion
• Following systemic heparinization, 1.2–​1.5 L of donor blood is rapidly
drained from the donor via a cannula in the RA or aorta in order to
prime the device.
• Care is taken to avoid administration of vasopressors to donors during
this time period (as this may subsequently affect perfusion of the graft).
781

178 Chapter 15 Donor heart procurement: DBD and DCD

Fig. 15.3 TransMedics OCS Heart device.

• This blood is passed through a leucocyte filter and added to the OCS
reservoir pre-​primed with 500 mL of the proprietary priming solution.
This perfusate is then circulated, oxygenated, and warmed with the
addition of the TransMedics maintenance solution containing amino
acids with added adenosine.
• Following procurement of the donor heart, the aorta is cannulated
with an interfacing aortic connector and secured by a cable tie. This
connector is then attached to the organ chamber while carefully de-​
airing the aortic root.
 Normothermic machine perfusion 179

Fig. 15.4 Cannulation of the donor heart on the OCS Heart device.

• After perfusion is restored to the heart and cardiac rhythm is restored,

the IVC and SVC are sutured closed to channel the coronary sinus
blood through the main PA.
• A vent is placed in the LV to prevent air from being ejected into the
aortic root.
• The PA is cannulated and connected to the organ chamber where
coronary flow, haemoglobin, haematocrit, and oxygen saturations are
measured.
• If the heart fails to restore sinus rhythm, the integrated ECG pad can be
used to defibrillate the heart. Pacing wires can be placed in the event of
bradycardia.
• Heart perfusion is regulated by ensuring that the aortic root pressure is
maintained between 65 and 90 mmHg and coronary flow between 650
and 850 mL/​min. This can be controlled by altering the infusion of the
vasodilatory maintenance solution as well as varying the pump speed.
An automatic perfusion mode is now available to reduce operator
interaction and a synchronization mode aimed at reducing oedema by
timing pulsatile perfusion with the ECG.
801

180 Chapter 15 Donor heart procurement: DBD and DCD

• Donor heart viability and suitability for transplantation are assessed by

measuring arterial and venous lactate levels. It is recommended that the
perfusate lactate levels do not trend upwards and rise above 5 mmol/​L.
It is also recommended that the heart should only be used if the heart is
consuming lactate with the venous lactate level lower than the arterial.
• At the recipient centre, when the transplant team is ready to implant
the donor heart, the SVC and IVC sutures are removed, the heart
is cooled to 18°C, the pump is stopped, and 1 L of cold crystalloid
cardioplegia is used to arrest the donor heart through the aortic root
while the vents remain in the heart.

Donation after circulatory death heart

procurement
Livers, lungs, kidneys, and pancreases have all been routinely transplanted
from DCD donors for over a decade. At an international consensus
meeting in 1994 (Maastricht, Netherlands), four types of non-​beating heart
donors were recognized in a cascade of shorter warm ischaemic times
(Table 15.1). The terms uncontrolled (I, II) versus controlled (III, IV) dona-
tion were later added.

Table 15.1 DCD donor Maastricht criteria

Category Description
I Declared dead outside hospital
II Witnessed cardiac arrest and unsuccessful resuscitation within
hospital
III Expected death following withdrawal of life supportive therapy
IV Brain-​dead donors who have undergone cardiac arrest

One of the main challenges facing DCD heart transplantation is the wide-
spread variation that exists across the globe relating to what is acceptable
ethical practice, such as antemortem drug administration, interventions,
and the duration of the observation period from mechanical asystole to
declaration of death, which can vary from 5 to 20 minutes across different
European countries. Although all other solid organs from category III DCD
donors are routinely retrieved using cold static storage, there had been
considerable reservations to transplant DCD hearts due to the associated
prolonged obligatory warm ischaemic time which is not well tolerated by
the heart. In order to overcome these challenges, pioneers in DCD heart
transplantation have adopted techniques aimed at minimizing the warm and
cold ischaemic periods, while also allowing a quality check of the heart prior
to transplantation, all obstacles to previous successful DCD heart trans-
plant attempts.
 Donation after circulatory death heart procurement 181

There are currently four techniques or combination of techniques that

have been used globally for DCD heart procurement leading to successful
clinical transplantation in the modern era.
1. Direct procurement after flushing with cold cardiac preservation fluid
and cold static storage (DP-​CS).
2. Direct procurement and normothermic machine perfusion (DP-​MP).
3. Normothermic regional perfusion followed by machine perfusion
(NRP-​MP).
4. Normothermic regional perfusion followed by cold static storage
(NRP-​CS).
1. DP-​CS
This technique, reported in 2008 by the Denver (CO, USA) group, used cold
static storage to preserve three paediatric hearts from DCD donors with
a mean donor age of 3.7 days. In order to be successful, they co-​located
the donor and recipient, employed antemortem cannulation of the femoral
vessels, antemortem heparinization, and reduced the observation period
from 3 minutes down to 75 seconds to minimize ischaemia. Following dec-
laration of death, an intra-​aortic endo-​balloon was inflated and cold cardiac
preservation fluid was administered into the aortic root before the heart
was removed and cold stored. The average total cold ischaemic time was
106 minutes. One recipient required ECMO post transplantation but all
were discharged home on average at 20 days post transplantation and all
were alive 6 months later.
2. DP-​MP
In 2015, the Sydney (Australia) group reported the first three successful
adult DCD heart transplants using the DP-​MP technique for distantly pro-
cured DCD hearts using the OCS Heart device. The protocol is based on
donors <40 years old and a donation withdrawal ischaemic period of <30
minutes (time from withdrawal of life-​sustaining therapy to starting the in-
fusion of the cardiac preservation fluid). The observation period from asys-
tole to the declaration of death varied from 2 to 5 minutes depending on
which Australian state the donor was located in. Following declaration of
death, a rapid sternotomy was performed and the donor was exsanguin-
ated. The collected donor blood was leucocyte depleted and added to the
pre-​primed OCS Heart device. One litre of cold St. Thomas cardioplegia
supplemented with erythropoietin and glycerol trinitrate (GTN) was given
to the asystolic heart to minimize reperfusion injury. The heart was then
explanted and instrumented for OCS perfusion before being transported
to the recipient centre. Of the three DCD heart recipients, all survived to
discharge with one requiring ECMO support post transplantation.
A recent update on the Sydney experience reported 23 DCD heart
transplants using this technique with a 96% overall survival, a 35% reliance
on ECMO post transplantation, and an overall increase of heart transplant
activity of 15%.
3. NRP-​MP
DCD heart transplantation using the NRP technique was first reported by
Papworth Hospital (Cambridge, UK) in 2016 following nine successful clin-
ical adult DCD heart transplants. The technique evolved to minimize the
risk of primary graft failure in the recipient as previous attempts at DCD
821

182 Chapter 15 Donor heart procurement: DBD and DCD

heart transplantation had never involved a functional assessment of the

DCD donor heart prior to transplantation with primary graft failure re-
quiring ECMO reported at 30%. No antemortem interventions or drugs are
permitted in the UK and the observation period is set at 5 minutes. The
protocol consisted of donor age <50 years old, a donor warm ischemic time
of <4 hours, and a functional warm ischaemic time (systolic <50 mmHg
to onset of in situ blood reperfusion) <30 minutes. Following declaration
of death, the donor is transferred to the operating room where a rapid
sternotomy is undertaken. The pericardium is opened and 30,000 IU of hep-
arin is injected into the RA and 20,000 IU into the main PA. The cerebral
circulation is excluded by clamps or staple ligation of the three aortic arch
branches. The donor then undergoes arterial and venous cannulation before
reperfusing the thoraco-​abdominal compartments (Fig. 15.5). Recently the
technique has been refined to open the distal stumps of the arch vessels to
allow active drainage, preventing any pressure build-​up through collateral
blood flow throughout the NRP period, which can last up to 2 hours.

SVC
Ao
PA

Oxygenator
Right atrial
cannulation

Reservoir

Pump

Fig. 15.5 Normothermic regional perfusion circuit configuration.

 Donation after circulatory death heart procurement 183

Following reperfusion, a dopamine infusion is started and the donor is re-​

intubated and ventilated before weaning the reanimated donor heart from
circulatory support. A full functional assessment is undertaken using TOE
and thermodilution cardiac outputs by inserting a PAC. Following satisfac-
tory assessment, for distantly procured DCD hearts the TransMedics OCS
Heart device is primed with donor blood. The DCD heart is re-​arrested
with 500 mL of St. Thomas No. 2 cardioplegic solution supplemented with
erythropoietin and GTN, explanted from the donor and instrumented for
the OCS Heart device. To date, of the 18 DCD heart transplants under-
taken at Papworth using NRP-​MP, there remains a 100% survival with a 5%
reliance on ECMO post transplantation. The main drawback of the practice
is that in some countries, restoring heart function in a donor declared dead
from circulatory arrest has raised ethical objections.
4. NRP-​CS
If the DCD NRP heart donor and the heart transplant recipient happen to
be co-​located in the same or adjoining hospitals, the recipient operation be-
gins following the satisfactory functional assessment of the DCD heart after
NRP has been weaned off. When the recipient surgeon is ready to implant
the DCD heart, it can be retrieved in a similar way to a standard DBD heart
retrieval with the exception that the DCD heart is arrested with 1 L of cold
crystalloid cardioplegia supplemented with erythropoietin and GTN. The
heart is then explanted, packaged, and expedited to the awaiting recipient
surgeon. The Papworth Hospital team have undertaken three DCD heart
transplants using this technique with all surviving to discharge with one re-
quiring ECMO for 12 hours in the early post-​transplant period (who was on
ECMO as a bridge to heart transplantation). There have been subsequent
reports of two successful DCD heart transplants using this technique from
Liege in Belgium. The benefit of this technique is a full functional assessment
while saving on the financial expense of an ex situ perfusion device.
841
 Chapter 16 185

Technical aspects
of heart transplantation
Introduction 186
Recipient operation 186
Cardiac explanation 187
Special situations 189
Allograft implantation 190
Conclusion 193
861

186 Chapter 16 Technical aspects of heart transplantation

Introduction
The first successful cardiac transplantation was performed by Christiaan
Barnard in 1967. However, this was preceded by years of methodological
development conducted by Norman Shumway. Heart transplantation has
since become the treatment of choice for eligible patients with advanced
HF, and >112,000 heart transplants have been reported to the International
Heart and Lung Transplant Registry thus far. Despite adverse changes in the
donor and recipient populations, post-​transplant outcomes continue to im-
prove with a median survival of >12 years.
The most common method for cardiac transplantation is implantation in
the orthotopic position using the bicaval technique. In the following chapter
we will describe this approach: OHT using hearts from brain-​dead donors.

Recipient operation
Preparation and timeline
Recipient ‘go-​call’ is usually made approximately 2.5 hours before the an-
ticipated start time, in order to provide adequate time for the recipient to
be transported to the operating room and anaesthesia induction prior to
final acceptance of the organ in the donor hospital. Recipient skin incision
usually takes place 30 minutes after the designated donor operation begins,
once the procuring surgeon has evaluated the donor heart and given the
final approval.
Occasionally, there are considerable concerns regarding the viability of
the donor heart. In this scenario, the recipient will not be anaesthetized
(though some peripheral lines may be placed) until the procuring surgeon
has accepted the organ.
The recipient operation is performed under general anaesthetic with
orotracheal intubation. Large-​bore peripheral and central venous lines and
an arterial line are placed. A PAC with continuous cardiac output monitoring
and TOE are utilized in all patients. If the INR is elevated, vitamin K and/​
or prothrombin complex concentrate may be administered preoperatively.
Methylprednisolone (1000 mg) and appropriate antibiotic prophylaxis are
administered after the donor heart has been accepted. If the recipient has
an ICD, it should be deactivated and external defibrillator pads should be
placed on the patient.
Any foreign bodies that will be removed following completion of the
transplant (ICD, IABP, LVAD driveline), both groins, and axillary regions
should be prepped into the surgical field. Foreign bodies should be covered
with a surgical towel and exposed only at the end of the case to minimize
the chance of mediastinal contamination.
The timing of the donor cross-​clamp is set following communication be-
tween the procuring and implanting surgeons, considering anticipated de-
lays in both the donor and recipient operating rooms. If the recipient is a
primary sternotomy, skin incision is made 60–​90 minutes before the antici-
pated organ arrival time in the operating room, and CPB is commenced
30–​45 minutes before the anticipated arrival time. After confirmation has
been received that the air transportation has touched down, the recipient
aortic cross-​clamp is placed and the cardiectomy is performed. The goal is
 Cardiac explanation 187

to coordinate donor heart arrival in the recipient operating room with com-
pletion of the recipient cardiectomy, in order to minimize ischaemic time
for the donor heart while minimizing the necessary CPB time to complete
the cardiectomy. The aim is to limit the total ischaemic time to <4 hours in
adult recipients.

Cardiac explanation
See Fig. 16.1.
Cardiac explanation is performed in the following sequence:
1. Median sternotomy and IV heparin.
2. Separate the aorta and PA.
3. Aorto-​bicaval cannulation and place umbilical tapes around the IVC
and SVC.
4. Place a purse-​string in the right superior pulmonary vein.
5. Initiate CPB and cool to 28°C.
6. Place the aortic cross-​clamp.
7. Snare the IVC and SVC, and open the free wall of the RA from
the right atrial appendage (RAA) parallel with the tricuspid annulus
inferiorly into the coronary sinus.
8. If pacemaker or defibrillator leads are present, they are pulled down
and out of the SVC as much as possible and then amputated flush
with the SVC.
9. With a No. 11 blade, make an incision in the medial aspect of the
foramen ovale medially and superiorly towards the dome of the LA
and aortic valve.
10. Divide the aorta at the aortic valve (you may place a 2-​0 silk suture
on the anterior aspect of the remaining aorta for retraction and
orientation).
11. Divide the PA at the pulmonic valve (you may place a 2-​0 silk suture
on the anterior aspect of the remaining PA for retraction and
orientation).
12. Extend to the left the incision previously made onto the foramen
ovale and the dome of the LA, along the posterior MV annulus and
connect it to the previous incision made in the fossa ovalis from the
right side, and extract the heart.
13. Create the IVC cuff (2–​3 cm away from the cannula) by separating
it from the remaining LA tissue, being careful not enter the LA and
keep the plane of dissection inferior to the right inferior pulmonary
vein (you may place a 2-​0 silk suture in the anterior aspect of the
remaining atrial tissue for retraction and orientation).
14. Create the SVC cuff by transecting it cranial to the cavoatrial junction
with scissors and then separating it from the right PA posteriorly (you
may place a 2-​0 silk suture in the anterior aspect of the remaining
atrial tissue for retraction and orientation).
15. Tailor the remaining LA tissue to create a symmetrical cuff. Be sure to
excise any remaining coronary sinus and the LAA, but leave adequate
LA tissue to ensure a generous cuff.
81

188 Chapter 16 Technical aspects of heart transplantation

(a) (b)

(c) (d)

(e) (f)

(g) (h)

Fig. 16.1 Recipient cardiotomy. (a) Right atriotomy is made extending from the
RAA caudally. (b) The right atriotomy is extended parallel to the AV groove from the
RAA into the coronary sinus. (c) The medial aspect of the fossa ovalis (FO) is incised
and extended superiorly onto the roof of the LA between the SVC and the aorta.
(d) The aorta and PA are divided at the level of the semilunar valves. (e) The left
atriotomy is extended across the roof of the LA towards the posterior margin of the
LAA and the left atriotomy is then extended along the posterior mitral valve annulus.
(f ) The remnant of the RA is divided and the cavae are mobilized. (g) The SVC
and IVC cuffs are completed by separating the posterior aspect of the caval vessels
from the anterior aspect of the LA. The redundant tissue between the two marked
incisions can be excised. (h) Completed recipient cardiotomy.
 Special situations 189

16. Place a 3-​0 SH polypropylene suture in the LA cuff that corresponds

to a point under the middle of the PA (where the LAA was
previously). This suture will be used for the LA anastomosis.
17. Quadrangulate the remaining LA cuff with three 4-​0 polypropylene
sutures to optimize exposure during the LA anastomosis.
18. Mobilize the posterior aspect of the aorta, PA, and SVC.
19. Obtain haemostasis of the posterior pericardial contents.
20. Place a LV vent through the previously placed purse string in the right
superior pulmonary vein.

Special situations
Redo sternotomy
The additional time required to complete dissection of the recipient heart
and cannulate for CPB in the redo scenario may necessitate delaying of
the donor cross-​clamp time. When the recipient operation has progressed
to a point such that the estimated surgical time remaining until recipient
cardiectomy will be completed is equivalent with the estimated organ
transport time from the donor hospital, it is safe to proceed with donor
cross-​clamp.
A number of additional precautions are taken prior to initiating the re-
cipient operation in a patient with prior sternotomies:
1. The CT scan is reviewed to identify the proximity of vital structures to
the posterior sternal table (outflow graft, aorta, innominate vein, RV,
driveline).
2. External defibrillator pads are placed on the patient.
3. Micropuncture catheters (5 Fr) are placed in the femoral artery and
vein in case the need for emergency cannulation arises.
4. Peripheral CPB cannulas and a minimum of 4 units of red blood cells
should be readily available prior to sternotomy.
5. If the redo sternotomy is deemed to be particularly high risk based
on preoperative imaging, right axillary artery and femoral venous
cannulation are performed and the patient is placed on CPB prior to
sternotomy.
LVAD explanation
Preoperative precautions are taken as described above for patients with
previous sternotomies.
1. Following redo sternotomy, appropriate sites for aortic and venous
cannulation are dissected out. The remainder of the heart and LVAD
are mobilized, paying particular attention to not damage the outflow
graft or LVAD driveline. As much of this dissection as possible is
performed prior to initiation of CPB; however, complete dissection
typically requires initiation of bypass.
2. Following initiation of CPB, the speed on the LVAD should be reduced
such that the LV is adequately decompressed but not completely
collapsed.
3. More time may be required to complete the recipient cardiotomy and
obtain haemostasis following placement of the aortic cross-​clamp in
901

190 Chapter 16 Technical aspects of heart transplantation

the redo scenario. Consequently, the aortic cross-​clamp may need to

be placed prior to touchdown of air transportation in some scenarios
in order to minimize donor organ ischaemic time.
4. Following placement of the aortic cross-​clamp, the outflow graft is
ligated and divided, and the LVAD is turned off (the driveline can be
divided later).
5. Meticulous haemostasis of the posterior pericardium should be
obtained following recipient cardiectomy.
6. If the dissection is particularly difficult and the heart has arrived in the
recipient operating room before the cardiectomy has been completed,
the LV apex can be amputated to facilitate cardiectomy and minimize
ischaemic time. The LVAD can be removed after the cardiectomy has
been completed or even after the heart has been implanted.
Heart transplantation following a congenital operation
1. The procuring surgeon should discuss with the implanting surgeon what
additional tissue should be procured. This may include the entire aortic
arch and a portion of the descending thoracic aorta, the PAs out into
the lung hilum, and the entire SVC and innominate vein complex.
2. The procuring team should ensure an appropriately sized cardioplegia
cannula and sternal retractor are available.
3. Appropriate recipient imaging should be obtained to determine the
patency of upper and lower extremity vasculature, and to investigate
the proximity of vital structures to the posterior sternal table.
4. The implanting surgeon should:
a. Take appropriate precautions as for any redo sternotomy.
b. Be prepared to address the significant aortopulmonary collaterals
that are common in this patient population.
c. Be prepared for the need for complex reconstruction of the
IVC and SVC, the innominate vein, the PAs, and the aortic arch,
depending on the patient’s HF aetiology and the nature of the
previous surgical procedures. As much of the reconstruction as
possible should be performed prior to donor heart arrival.

Allograft implantation
Back-​table preparation
1. Place the donor heart in a basin of cold saline.
2. Verify the integrity of all cardiac valves and the coronary sinus.
3. Close the foramen ovale if patent.
4. Create a common LA cuff by connecting the pulmonary veins. Tailor
the LA cuff by resecting excess tissue.
5. Separate the aorta and the PA (exercise caution so as to not injure the
left main coronary artery as it courses posterior to the main PA).
Heart transplantation
See Fig. 16.2.
 Allograft implantation 191

(a) (b)

(c) (d)

Fig. 16.2 Allograft implantation. (a) Pericardial well prepared for the new
heart. A right superior pulmonary vein (RSPV) vent placed in the LA well. (b) LA
anastomosis. RSPV vent will be placed through the MV into the LV before tie-​down.
(c) IVC, aorta, and PA back-​wall anastomosis. (d) Completed aorta anastomosis.
(e) IVC and PA anterior wall anastomosis as well as SVC anastomosis.

1. Deliver a dose of cold blood cardioplegia to the donor heart following

completion of the back-​table dissection (300 mL).
2. Place the heart on the recipient left chest wall, such that the LA cuff is
facing the implanting surgeon. Using the 3-​0 SH polypropylene suture
previously placed under the recipient PA, begin the LA anastomosis
starting at the donor LAA. The heart is parachuted down on top of an
icy sponge in the left lateral pericardial space and covered in ice.
3. The leftward and then inferior portions of the anastomosis are
performed in an everting and running fashion. The IVC marks the half-​
way point. The right superior pulmonary vein vent should be placed
across the mitral valve. The completed portion of the anastomosis
should be examined and 4-​0 polypropylene reinforcing sutures can be
placed on the back wall if needed.
921

192 Chapter 16 Technical aspects of heart transplantation

4. The superior and rightward portion of the anastomosis is then

completed using the other end of the 3-​0 SH polypropylene suture.
5. Deliver a dose of cold blood cardioplegia (250 mL) and IV
liothyronine (dosed according to patient weight).
6. Perform the posterior half of the IVC anastomosis using a 4-​0 SH
polypropylene suture.
7. Tailor the donor PA to length and orientation. Perform the posterior
half of the anastomosis using a 4-​0 polypropylene suture.
8. Tailor the donor aorta to length and place a 2-​0 silk suture at the
anterior aspect for retraction and orientation. Perform the posterior
half of the anastomosis using a 4-​0 polypropylene suture. The
completed portion of the anastomosis should be examined and 4-​
0 polypropylene reinforcing sutures can be placed. The LV vent is
turned off and the aortic anastomosis is then completed.
9. An antegrade cardioplegia needle is placed in the anterior aspect of
the recipient aorta. The heart is then de-​aired and the aortic cross-​
clamp is removed. The LV and root vents are resumed. CPB flows are
maintained at half-​flow for 3 minutes following removal of the cross-​
clamp and then increased back to full flow. Rewarming is commenced
after the donor heart has been reperfused for a period of 15 minutes.
10. The anterior half of the PA anastomosis is completed.
11. A suction catheter is placed through the SVC into the coronary sinus,
and then the IVC anastomosis is completed.
12. The posterior half of the SVC anastomosis is completed using a 5-​
0 polypropylene suture. Consideration should be given to locking
portions of this anastomosis to prevent narrowing. The PAC is
then placed through the SVC into the RV. The anastomosis is then
completed.
13. The caval snares are removed and the SVC cannula is turned 180°
into the RA to optimize drainage.
14. An epinephrine infusion is started at 0.04 mcg/​kg/​min after the
patient is rewarmed.
15. All anastomoses are inspected for bleeding sites that require
additional sutures. Atrial and ventricular pacing wires are placed and
the heart is paced sequentially at 90 bpm.
16. The patient is ventilated with an inhaled pulmonary vasodilator (e.g.
nitric oxide (NO)) and any necessary corrections to the arterial blood
gas are made.
17. Wean from CPB and begin collecting continuous cardiac output data.
18. Assess cardiac function using TOE.
19. Optimize haemoglobin concentration, ventilation, oxygenation, heart
rate and rhythm, preload, afterload, and inotropic support.
20. Evaluate the need for delayed chest closure, or mechanical support.
21. If clinical stability is achieved and there is no need to return to CPB,
basiliximab (20 mg) is administered.
22. After the sternal incision has been closed and dressed, remove the
ICD and the remainder of the driveline (if present) and pack the
wound with gauze.
 Conclusion 193

Primary graft dysfunction (PGD)

A successful cardiac transplantation requires thoughtful donor–​recipient
matching, minimized donor ischaemic time, diligent myocardial protection
during implantation, and meticulous surgical technique. Weaning from CPB
is facilitated through optimization of haemoglobin concentration, oxygen-
ation, and ventilation with a pulmonary vasodilator. The heart is paced at a
rate of 90–​100 bpm (AAI or DDD), the CVP is maintained at 8–​12 mmHg,
and the MAP at 80–​90 mmHg. Inotropic support is optimized (e.g. epineph-
rine, norepinephrine, and dopamine for the LV and milrinone and NO for
the RV, if needed) and adequacy of the systemic circulation is evaluated.
However, regardless of the above-​mentioned points, PGD may de-
velop in patients undergoing heart transplantation, and is associated with
30% mortality at 30 days. PGD may result from LV failure, RV failure, or
biventricular failure and is graded as mild, moderate, or severe LV PGD and
as RV PGD as described by Kobashigawa.
If high-​dose inotropic support is required to maintain an adequate CI,
there should be a low threshold to insert an intra-​aortic balloon and/​or
more advanced MCS devices. In the situation of isolated RV failure, central
RVAD support can be considered. In this situation, it is advisable to tunnel
the cannulas through the epigastrium to allow closure of the sternotomy
incision. A 34–​36 Fr angled cannula is placed in the RA, and a 19–​21 Fr
percutaneous femoral cannula (the additional length of this cannula can be
helpful) is placed in the PA. A 2-​0 Ethibond suture is used to place one
purse string in the RA and two purse strings in the PA. Multiple intervening
pledgets are placed on each purse string. The tip of the PA cannula can be
trimmed back, and care is taken when inserting this cannula so that only
2 cm of cannula is in the PA past the last infusion port of the cannula. This
ensures that the tip of the cannula is proximal to the PA bifurcation and
prevents streaming of the RVAD outflow into one PA. The cannulas are
de-​aired and connected to the temporary pump (e.g. RotaFlow, CentriMag,
etc.). Flow is increased slowly and typically in order to adequately fill the LV
and facilitate weaning of inotropic support.
If there is any concern regarding LV or pulmonary function, central VA
ECMO should be instituted. If the duration of support is expected to be
short (24–​48 hours), then the CPB cannulas can be used and brought out
through an open chest. However, tunnelling new cannulas through the epi-
gastrium is preferred as this would facilitate chest closure, early extubation,
and mobilization until myocardial recovery occurs. Cannulation (34–​36 Fr
angled cannula in the RA and 19–​21 Fr percutaneous femoral cannula in the
ascending aorta) proceeds in the same manner as described for insertion
of the RVAD. If myocardial recovery does not occur, the patient should be
considered for relisting for retransplantation. That being said, one should
take into consideration that acute heart retransplantation has very poor
outcomes and there is much ethical debate about its appropriateness.

Conclusion
Since the first heart transplantation performed in 1967, improvements in
surgical techniques, preservation methods, and immunosuppression have
enabled outstanding 1-​year survival and graft half-​life of >12 years.
941

194 Chapter 16 Technical aspects of heart transplantation

The current era of heart transplantation is evolving around preservation

machines, which will enable substantial increases in transport time, min-
imizing the cold ischaemic time and enabling the use of marginal/DCD
donors, thus increasing the donor pool and as a consequence increasing the
number of transplants being performed all around the world substantially.
Furthermore, it may enable organ sharing between countries that previ-
ously were not able to do so owing to transport time.
Future research efforts should be directed to minimize cardiac allograft
vasculopathy (10%) and malignancy (20%) that account for most cardiac
recipient deaths. These numbers highlight that regardless of the progress
that has been accomplished over the years in all aspects of cardiac trans-
plantation, the consequences of chronic administration of immunosuppres-
sion on long-​term survival is pronounced. Thus, the need to develop new
drugs or tolerance strategies that achieve better long-​term graft survival is
genuine.
Hence, standardizing the surgical technique as explained in this chapter is
a crucial step, but merely one of many steps needed.
 Chapter 17 195

Critical care management

and primary graft
dysfunction following
cardiac transplantation
Monitoring and investigations 196
Cardiovascular management 196
Fluid management 198
Respiratory management 198
Immunosuppression 199
Infection management 199
Nutrition 200
Psychosocial support 200
Multidisciplinary team working 201
Early post-​transplant complications 201
Acute kidney injury 203
Right ventricular dysfunction 203
Primary graft dysfunction 204
Vasoplegic syndrome after heart transplantation 206
961

196 Chapter 17 Immediate postoperative management

Monitoring and investigations

The patient should arrive from theatre with full monitoring in place. The
ISHLT recommends the following:
1. Peripheral oxygen saturation.
2. ECG (telemetry).
3. Invasive arterial blood pressure.
4. CVP/​RA pressure.
5. PA pressure.
6. PCWP/​LA pressure.
7. Cardiac output.
8. Urinary output.
Baseline investigations should be performed on arrival and thereafter as
clinically indicated.
Arterial blood gases provide an early assessment of gas exchange, oxygen
delivery, acid–​base status, and important serum electrolyte levels including
potassium and ionized calcium. Bedside tests of coagulation including
heparinase ACT plus thromboelastometry may be repeated following
blood products or if there is ongoing blood loss.
Haematology tests including full blood count and coagulation screen are
used to guide transfusion of blood products. Biochemistry tests including
urea and electrolytes allow diagnosis of common electrolyte imbalances
such as hypokalaemia, hypomagnesaemia, and hypocalcaemia.
CXR should be performed to confirm correct positioning of endo-
tracheal tube, invasive monitoring lines, and mediastinal/​intercostal drains.
Lung ultrasonography in the ICU setting has increased in the past
decade and can be invaluable for the assessment of lung function in the
postoperative period. A consensus document by the International Liaison
Committee on Lung Ultrasound provides recommendations for the iden-
tification and follow-​up of common complications such as pulmonary oe-
dema, consolidation, and pleural effusions.
Echocardiography plays a crucial role in the postoperative period, guiding
cardiovascular management and identifying cardiac complications such as
tamponade. Filling status, ventricular function, and pericardial collections
can all be assessed and monitored at the bedside.

Cardiovascular management
Haemodynamic management starts in theatre and continues into the
postoperative period in the ICU. Common haemodynamic goals include an
adequate CI and systemic blood pressure to maintain organ perfusion while
avoiding excessive myocardial oxygen demand and promoting recovery of
graft function.
Chronotropic medications and/​or AV pacing continue to achieve heart
rate of 90–​110 bpm with AV synchrony. Fluid or diuretics are titrated ac-
cording to filling pressures and appearances on echocardiography noting
that diastolic dysfunction may be a prominent feature of the first 12–​24
hours. A CVP <10 mmHg is desirable to promote RV recovery. Pulmonary
vasodilators continue in the context of elevated pulmonary pressures and
RV failure to promote left-​sided filling. Inotropes are titrated according to
 Cardiovascular management 197

ventricular function on echocardiography to achieve a CI of 2–​2.5 L/​min/​

m2. IABPs implanted in theatre are generally continued for 48–​72 hours
with careful attention to lower limb perfusion. Vasopressors are titrated to
achieve a MAP of 65–​70 mmHg. Packed red blood cells are transfused to
maintain haemoglobin >7.5 g/​dL. Higher levels (e.g. haemoglobin 10 g/​dL)
may be targeted if SvO2 is <60%.
Commonly used inotropes/​vasopressors are summarized in Table 17.1.

Table 17.1 Commonly used inotropes/​vasopressors

Agent Dosage Half-​life Peripheral Cardiac Peripheral Chrono

vasocon­ con­ vasodi­ tropic
striction tractility lation effect

Dopamine 1–​10 mcg/​ 1 minute +​ +​ +​ +​ +​ +​ +​

kg/​min

Dobutamine 1–​10 mcg/​ 4 0 +​ +​ +​ +​ +​ +​

kg/​min minutes

Epinephrine 0.01–​0.2 2–​3 +​ +​ +​ +​ +​ +​ +​+​

mcg/​kg/​min minutes +​ +​

Milrinone 0.1–​0.5 mcg/​ 2–​4 0 +​ +​ +​ +​ +​ +​

kg/​min hours

Norepinephrine 0.01–​0.2 2–​2.5 +​ +​ +​ +​ +​ +​ +​ 0 +​

mcg/​kg/​min minutes

Vaso pressin 0.018–​0.12 10–​35 +​ +​ +​ +​ 0 0 0

IU/​kg/​min minutes

Levosimendan Loading 70–​80 0 +​+​+​ +​+​ +​

dose: 6–​24 hours
mcg/​kg over
10 minutes
Infusion:
0.05–​0.2
mcg/​kg/​
min

Mechanical circulatory support

MCS is generally reserved for patients with persisting graft dysfunction
and compromised organ perfusion despite the above-​discussed measures.
Ideally, MCS should be instituted in theatre but may also be required in the
postoperative period in the context of escalating pharmacological support.
Commonly used devices include VA ECMO for biventricular failure and
short-​term RVAD for isolated RV failure. These devices support cardiac
output and maintain organ perfusion while allowing weaning of inotropes and
recovery of the injured ventricle. Ongoing management requires regular input
from the multidisciplinary team including surgery, ICU, nursing, and perfusion
to avoid device-​related complications for the duration of the support.
981

198 Chapter 17 Immediate postoperative management

Weaning support
Pharmacological support commenced in theatre should be continued for
48–​72 hours postoperatively. Following a period of haemodynamic sta-
bility, support is weaned as cardiac function recovers and vascular tone
returns. Inhaled pulmonary vasodilators are weaned first to facilitate
extubation. Nebulized iloprost or nasogastric sildenafil may be used to
avoid rebound PH. Inotropes are weaned slowly in the context of ad-
equate CI and ventricular function on echocardiography. Vasopressors
are weaned in the context of an adequate MAP for organ perfusion (usu-
ally 65–​70 mmHg). IABP may be explanted once inotropes are weaned
to moderate levels (e.g. dobutamine 5–​ 10 mcg/​ kg/​min) facilitating
mobilization.
Hypertension may become a prominent feature following myocardial re-
covery reflecting the patient’s vasculature adapted to a low cardiac output.
Commonly used antihypertensive agents include GTN, hydralazine, and
calcium channel blockers. BBs and ACEis are best avoided in the immediate
postoperative period.

Fluid management
The patient may be subject to large fluid shifts intraoperatively due to
increased capillary permeability secondary to CPB, on-​pump ultrafiltra-
tion, bleeding (especially in re-​sternotomies), and blood product transfu-
sion. The goals of fluid management in ICU are to maintain preload while
avoiding RV distension and ultimately achieving euvolaemia. Invasive moni-
toring including CVP/​PCWP and echocardiography should be used to
guide management including cautious fluid resuscitation or IV diuretics by
bolus or infusion (e.g. furosemide 10–​40 mg/​hour) as required. A posi-
tive fluid balance may be necessary for 24–​48 hours postoperatively be-
fore diuresis can be used to achieve a daily negative fluid balance targeting
euvolaemia. Continuous renal replacement therapy should be considered
early in the context of fluid overload and inadequate urine output despite
diuretic infusion/​acute kidney injury.

Respiratory management
The goals of respiratory management after transplantation are to avoid
physiological causes of increased PVR such as hypoxia and hypercapnia
while minimizing the deleterious effects of positive pressure ventilation on
RV function. While the patient remains sedated and intubated, protective
ventilation is employed with low tidal volumes and minimum PEEP required
to avoid de-​recruitment. FiO2 and respiratory rate are titrated to achieve
PO2 >10 kPa, and CO2 <5.5 kPa. Following a period of haemodynamic
stability, sedation is weaned with a view to spontaneous ventilation and
early extubation. In scenarios where extubation is felt likely to be delayed,
percutaneous tracheostomy should be considered to facilitate spontaneous
ventilation and weaning from the ventilator.
 INFECTION MANAGEMENT 199

Immunosuppression
Immunosuppression protocols should be well established at each centre
and should be easily accessible to all staff. Timing of induction, tailoring of
agents, and timing of sampling (CD3 count/​tacrolimus/​ciclosporin levels)
should be clearly outlined, alongside management of the early adverse
effects of immunosuppression (renal impairment, thrombocytopaenia,
marrow suppression, etc.). A multimodal approach involving the surgeons,
transplant physicians, intensivists, and nursing staff is needed with precise
handovers to ensure continuity of care. Further details regarding immuno-
suppression are discussed in Chapter 20.

Infection management
Perioperative immunosuppression is associated with an increased risk of
infection during the intensive care period and beyond. Post-​transplant in-
fection remains one of the leading causes of death in the first-​year after
transplantation, accounting for 25–​32% of all deaths. The risk of death is
highest between 1 month to 1 year post transplantation.
Risk factors for post-​transplant infections include age, use of MCS,
reoperation, previous bacterial infections, and relapsing viral infections
(Table 17.2).

Table 17.2 A list of infections that predominate in the first month post
transplantation
Infection type Early infection (<1 month)
Viral Herpes simplex virus
Bacterial Nocardia
Clostridium difficile
Pseudomonas
Vancomycin-​resistant enterococci
Methicillin-​resistant Staphylococcus aureus
Fungal Candida
Aspergillus

Ideally the post-​transplant patient should be cared for in an isolated,

positive pressure room with restricted visitor numbers. People entering the
room should wear a protective gown and perform hand hygiene prior to
interacting with the patient’s environment.
Antibiotic prophylaxis commenced in theatre as per local policy and is
continued in the ICU for 24–​48 hours. Commonly used agents include beta-​
lactams, cephalosporins, and glycopeptides often requiring level monitoring
and/​or dose reduction in the context of renal dysfunction.
Early extubation, patient mobilization, and removal of indwelling cath-
eters are important steps in reducing the risk of nosocomial infections.
20

200 Chapter 17 Immediate postoperative management

Candida species are common fungal pathogens. Oral nystatin is rou-

tinely used to prevent mucocutaneous candidiasis. Pneumocystis pneu-
monia is a common opportunistic infection in immunosuppressed patients.
Prophylaxis with antiprotozoals is commenced in the early postoperative
period and continued for up to 12 months. CMV-​positive donor or re-
cipient are considered at high risk of CMV infection and prophylaxis with
ganciclovir/​valganciclovir as recommended by the ISHLT is started within
24–​48 hours.
Special indications, such as the use of organs from HCV-​infected donors,
may require administration of antivirals, guided by locally agreed protocols
involving discussions with hepatologists and virologists.
The systemic response to sepsis may also be suppressed in the immuno-
compromised patient, necessitating a high degree of clinical suspicion for
diagnosis. Clinicians should have a low threshold for performing cultures
and commencing or changing antibiotics based on new positive results and
sensitivities. Daily input from a specialist microbiology team is vital. Vigilance
is also needed to promptly identify wound or line infections in the immuno-
suppressed patient. Severe infections often require dose adjustments of im-
munosuppression, increasing the risk of allograft rejection.

Nutrition
Weight loss may be a feature of chronic HF and may be accelerated pre-
operatively with low cardiac output states particularly in patients supported
with short-​term MCS. Early enteral nutrition is the goal in the postoperative
period preferably via the oral route. Nasogastric feeding should be com-
menced early if delayed extubation appears likely. Attention should be paid
to feeding interruptions, such as for invasive procedures, and feeding rates
adjusted accordingly to achieve overall nutritional requirements. Glycaemic
control may be particularly challenging in the context of steroids and cat-
echolamine infusions and a variable rate insulin infusion is commonly
required.

Psychosocial support
Cardiac transplant recipients are often young patients who may have been
affected by long-​standing illness with multiple hospital admissions and po-
tentially prolonged inpatient care. Patients and their families may be a long
way from home, bringing additional challenges around normal life com-
mitments. Perceptions of intensive care can vary. Ideally, the patient and
their family should visit the ICU prior to surgery but most will require on-
going education and reorientation in the ICU environment. Emotions in the
postoperative period will also vary from patient to patient and can impact
their recovery. Patients may benefit from psychology support when dealing
with emotions of fear, low mood, and guilt. Ongoing communication with
the patient and family is also essential to update them on progress. Donor
anonymity should be maintained at all times including during communication
within the clinical team and with the patient and family members.
 EARLY POST-​T RANSPLANT COMPLICATIONS 201

Multidisciplinary team working

Postoperative care for the cardiac transplant patient is delivered by a multi-
disciplinary team including the various different medical specialties, specialist
nursing groups, and allied health professionals. Management in intensive care
is directed by the intensive care consultant with regular input from other
members of the team on ward rounds and during regular multidisciplinary
team meetings. Each unit should have local cardiac transplant guidelines
including protocols on aspects of perioperative care. Daily management
plans should be clearly documented in the medical notes with deviations
from protocol noted and communicated to the multidisciplinary team.

Early post-​transplant complications

Hyperacute rejection
Hyperacute rejection is a relatively rare event in organ transplant recipients.
It is an irreversible process due to:
• The previous formation of antibodies to the HLA system present in the
transplanted graft, above all antibodies to class I antigens (HLA-​A, -​B,
or -​C).
• The existence of preformed ABO antibodies in the graft recipient at the
time of transplantation.
• Xenotransplantation between phylogenetically distant species.
It occurs from the time of transplant to 48 hours postoperatively
and result in rapid ischaemic and necrotic changes to the graft with
polymorphonucleocyte infiltration. It may occur upon allograft reperfusion
when the preformed antibodies against the donor are first in contact with
the graft. It was relatively more common in the earlier days of transplant-
ation, before HLA-​matching techniques were well developed. It results in
profound haemodynamic instability, for which inotropic and vasopressor
support is invariably required alongside MCS. High-​dose immunosuppres-
sion (corticosteroids, cytolytic induction therapy, antimetabolites) are often
administered and plasmapheresis may have a role in removing circulatory
antibodies from the blood while on the bypass circuit. IVIG can be adminis-
tered to inactivate antibodies. Endomyocardial biopsy can be performed to
confirm the diagnosis. Retransplantation may be considered, but the mor-
tality remains high.
Bleeding
Risk factors for postoperative bleeding include patient factors such as pre-
operative anticoagulation and operative factors such as reoperation and
prolonged CPB time. Clinical signs of ongoing blood loss include hypo-
tension with low cardiac output and low filling pressures. Invasive moni-
toring may show low blood pressure and low cardiac output with low CVP
and/​or pulmonary diastolic pressure/​PCWP. Chest and mediastinal drain
monitoring may identify ongoing drain losses. CXR and/​or lung US may
demonstrate pleural/​mediastinal collections. Echocardiography may show
underfilled LV and RV with or without pericardial collection.
20

202 Chapter 17 Immediate postoperative management

Hypothermia and hypocalcaemia should be reversed. Heparin ex-

cess should be identified with a heparinase ACT and treated with add-
itional protamine. In the context of ongoing bleeding, transfusion of
blood products should be guided by point-​ of-​
care testing (e.g. TEG,
thromboelastometry) along with laboratory tests including full blood
count and coagulation screen. Blood products for transfusion should be
irradiated for all transplant patients. CMV-​negative recipients should only
receive CMV-​negative blood.
Importantly, coagulopathic bleeding should be medically managed.
Bleeding which continues despite correction of coagulation abnormalities
should prompt surgical re-​exploration to identify a bleeding source.
Tamponade
Diagnosing tamponade in the post-​transplant setting may be challenging.
Tachycardia is often masked and the triad of hypotension, low cardiac
output, and raised CVP may be attributed to RV failure, another common
complication following cardiac transplantation. In the presence of clinical
suspicion, the team should have a low threshold for further investigation.
CXR may demonstrate a widened mediastinum and echocardiography
may reveal a local or generalized pericardial effusion with or without
chamber compression. Variation in transmitral velocities may not be seen
in the ventilated patient. Clinical diagnosis should prompt management
including chest drain clot removal and emergent surgical re-​exploration.
If the patient has been extubated, induction of anaesthesia should be
managed with extreme caution given the high vascular tone required to
maintain cardiac filling pressures and the potential for additional cardiac
dysfunction.
Arrythmias
Conduction abnormalities in the immediate post-​transplant setting are
multifactorial. Electrolyte disturbance including abnormal potassium, mag-
nesium, and phosphate levels are common following prolonged CPB and
predispose to arrhythmias. The denervated heart has a higher resting
heart rate due to the loss of vagal stimulation and does not exhibit the
expected response to many cardiac medications. The biatrial surgical im-
plantation technique involves the anastomosis of the donor and recipient
RA and is associated with a higher rate of sinoatrial node injury, increasing
the risk of atrial arrhythmias including bradycardias. It is also associated with
a higher incidence of permanent pacemaker implantations compared with
the bicaval technique. Rhythm control is desirable in the early postoperative
period with stroke volume being particularly dependent on ventricular
filling. Temporary epicardial pacing wires are left in situ until rhythm has
been stable for 48 hours off chronotropic drugs and/​or after the first
endomyocardial biopsy.
Atrial arrhythmias
Atrial fibrillation, atrial flutter, and supraventricular tachycardia are the
most common arrhythmias post transplantation though the frequency of
atrial arrhythmias in general is significantly lower than following general car-
diac surgery, possibly due to the exclusion of the pulmonary veins. New
or sustained atrial tachyarrhythmias should prompt a full workup to rule
 RIGHT VENTRICULAR DYSFUNCTION 203

out acute rejection. Treatment is similar to the general post-​cardiac surgical

patient including electrolyte replacement, antiarrhythmic drugs, and DC
cardioversion if required. Amiodarone is associated with significant drug
interaction with ciclosporin or tacrolimus, requiring close monitoring of im-
munosuppressant levels.
Ventricular arrhythmias
Non-​sustained ventricular tachycardia is relatively common in the early
postoperative period. It is often benign but may also be associated
with acute rejection or early graft failure. Sustained ventricular tachy-
cardia is uncommon and should prompt consideration of hyperacute
rejection.

Acute kidney injury

Acute kidney injury in the immediate postoperative period occurs in up to
70% of heart transplant patients. Perioperative renal failure may significantly
complicate the management of intravascular volume and ventricular filling
pressures.
The incidence of renal replacement therapy in the immediate
postoperative period varies (6–​25%) which may indicate the heterogeneity
of patients and their comorbidities. Previous cardiac surgery, high levels of
preoperative creatinine, prolonged CPB time, HF medical therapy, diabetes
mellitus, advancing age, low cardiac output, and low serum albumin concen-
tration have all been reported as independent preoperative risk factors for
acute kidney injury requiring renal replacement therapy.
Management includes optimization of cardiac output and MAP to maintain
renal perfusion with careful attention to fluid balance. Nephrotoxic drugs
are discontinued, avoided, or delayed until renal function has recovered
including immunosuppressant drugs such as the calcineurin inhibitors.
Renal replacement therapy should be considered early in the context of
fluid overload refractory to diuretics or electrolyte disturbance with the
increased risk of cardiac arrhythmias. If renal replacement therapy is re-
quired, continuous renal replacement is preferred to intermittent haemodi-
alysis aiming to avoid cardiovascular instability and large variations in filling
pressures. Care should be taken to monitor drug levels where possible,
including antimicrobials and immunosuppression drugs.

Right ventricular dysfunction

RV dysfunction is common following cardiac transplantation and is a
cause of major morbidity and mortality. It results from a combination
of donor and recipient factors including the ischaemic injury sustained
during organ procurement and increased PVR in the recipient. Diagnosis
requires clinical and echocardiographic assessments and management in-
cludes optimization of filling status, titration of vasoactive medications,
and early consideration of MCS. RV dysfunction is covered in detail in
Chapter 36.
204

204 Chapter 17 Immediate postoperative management

Primary graft dysfunction

PGD is defined as:
• Severe ventricular dysfunction of the donor graft which fails to meet
the circulatory requirements of the recipient despite adequate filling
pressures.
• In the immediate post-​transplant period.
• Manifest as either single or biventricular dysfunction.
It is differentiated from secondary graft dysfunction, which is when a dis-
cernible cause for allograft dysfunction is identified (e.g. hyperacute rejec-
tion, PH, or surgical complications).
PGD can be classified into two categories:
• PGD-​LV: includes LV dysfunction or biventricular dysfunction.
• PGD-​RV: includes RV dysfunction alone.
Both occur within 24 hours of transplantation.
Classification
Classification follows the ISHLT 2014 ‘Consensus Statement Definition
of Severity Scale for Primary Graft Dysfunction (PGD)’ (Table 17.3 and
Table 17.4).
Incidence
The incidence of PGD has a significant variation in incidence reporting
(2.3–​32.4%) prior to the ISHLT consensus statement. There is an increasing
trend towards utilization of extended criteria donor organs which may have
lowered the threshold for initiating MCS in some patients during the initial
phase of reperfusion, potentially causing an overestimation of the true in-
cidence of PGD.
Mortality
The 30-​day mortality ranges widely from 17% to 80% due to variations in
earlier definitions. Early mortality has improved with increased timely initi-
ation of MCS for PGD. To date, treatment for PGD is primarily supportive.
Retransplantation may be of benefit in selected individuals; however, mor-
tality from early retransplantation remains high.
Risk factors
Risk factors for PGD can be classified into donor, recipient, and procedural
factors (Table 17.5). The RADIAL score (Table 17.6) and PREDICTA score
(Table 17.7) are also useful tools for PGD. They are simple additive scores
used to predict recipient and donor combinations that may be at a higher
risk of PGD. PGD incidence increased significantly as the PREDICTA or
RADIAL score increased. Other risk factors described in the literature are
as follows:
Management
The management of PGD is primarily supportive, initially with inotropic
support and adjuncts such as vasopressors and/​or inhaled pulmonary vaso-
dilators. Early escalation to an IABP or temporary MCS (percutaneous/​
surgical) is warranted for patients who remain haemodynamically unstable.
In some instances, retransplantation may be considered following failure to
wean from MCS.
 Primary graft dysfunction 205

Table 17.3 PGD-​LV severity is separated into mild, moderate, and severe PGD
based on either echocardiographic/​invasive cardiac monitoring and inotrope
score or intervention criteria
PGD-​LV LVEF by echo Invasive cardiac Inotrope Intervention criteria
severity cardiography monitoring scorea
Mild ≤40% 1. RA pressure ≤10 n/​a
>15 mmHg
2. CI <2.0 L/​
min/​m2
3. PCWP
>20 mmHg
4. MAP
<70 mmHg
(Lasting >1 hour)
Moderate ≤40% 1. RA pressure >10 Newly placed
>15 mmHg, IABP (regardless of
2. CI <2.0 L/​ inotrope score)
min/​m2
3. PCWP
>20 mmHg
4. MAP
<70 mmHg
(Lasting >1 hour)
Severe n/​a n/​a n/​a Dependence on
left/​biventricular
MCS (ECMO,
BiVAD, LVAD, or
percutaneous LVAD)
Excludes IABP
a
Inotrope score =​dopamine (×1) +​dobutamine (×1) +​amrinone (×1) +​milrinone (×15) +​
epinephrine/​adrenaline (×100) +​norepinephrine/​noradrenaline (×100) with each drug dosed
in mcg/​kg/​min.

Table 17.4 PGD-​RV is diagnosed by either invasive cardiac monitoring or

intervention criteria
Invasive cardiac monitoring Intervention criteria
1. RA pressure >15 mmHg Need for RVAD
2. PCWP <15 mmHg
3. CI <2.0 L/​min/​m2
And
1. Transpulmonary pressure gradient
<15 mmHg and/​or
2. PA systolic pressure <50 mmHg
206

206 Chapter 17 Immediate postoperative management

Table 17.5 Risk factors for PGD

Donor factors Recipient factors Procedural factors
Advancing age Advancing age Prolonged ischaemic time
Donor cause of death Weight Donor–​recipient sex
mismatch
Trauma MCS Weight mismatch
Inotropic support Congenital heart Experience of procurement
disease as aetiology team and centre volume
for HF
Comorbidities Multiple reoperations Increased blood transfusion
(diabetes, requirement
hypertension)
Prolonged downtime LVAD explant Elective vs emergency
following cardiac arrest transplant
LVH Comorbidities: renal Prolonged CPB time
dysfunction, liver
dysfunction, diabetes
mellitus
Drug abuse Infection
CAD/​wall motion Amiodarone therapy
abnormalities on TTE

Table 17.6 RADIAL score

Variable Points
Right atrial pressure ≥10 mm Hg 1
Age (recipient) ≥60 years 1
Diabetes mellitus 1
Inotrope dependence 1
Age (donor) ≥30 years 1
Length of ischaemic time ≥240 minutes 1
Total score _​_​/​6
Source data from Segovia et al., 2011.

Vasoplegic syndrome
after heart transplantation
Vasoplegic syndrome is characterized by persistent hypotension, reduced
SVR, and normal or increased cardiac output. It occurs in up to one-​third
of patients following cardiac transplantation (depending on definition) and
has been shown to be associated with increased morbidity and mortality.
While the exact cause is unknown, several risk factors have been iden-
tified following cardiac surgery. These include the use of medications
 Vasoplegic syndrome after heart transplantation 207

Table 17.7 PREDICTA score

Variable Points
Preoperative MCS (ST-​VADs and ECMO) 3
Diabetes mellitus (recipient) 3
Cardiopulmonary bypass time >180 minutes 2
Implant Time
≤45 minutes 0
46–​60 minutes 1
61–​90 minutes 2
>90 minutes 3
Donor Age
<21 years 0
21–​40 years 1
41–​50 years 2
>50 years 3
Total _​_​/​14
Source data from Avtaar Singh et al., 2019.

targeted at the renin–​angiotensin system (ACEi, ARBs neprilysin/​angio-

tensin receptor inhibitors), obesity, male sex, higher serum thyroxine level,
preoperative IV heparin usage, aspirin, preoperative vasodilators (e.g.
milrinone), prolonged bypass time, prior cardiac surgery, continuous-​flow
LVADs, and prolonged ischaemic time.
The pathogenesis of vasoplegic syndrome may have several mechanistic
pathways which combine and contribute to the evolution of the shock state.
Current management strategies include titration of norepinephrine and
vasopressin, with methylene blue, a direct NO synthase inhibitor, used as
required for refractory hypotension. Additional treatment options including
high-​dose hydroxycobalamin and N-​acetylcysteine have also been described
though their role in management is less clear. In cases of refractory organ
failure despite maximum dose vasoconstrictors, circulatory support should
be considered.
208
 Chapter 18 209

Assessment and
management of chronic
complications following
heart transplantation
Cardiac allograft vasculopathy 210
Common general side effects and complications of
immunosuppression 212
Calcineurin inhibitors 212
Antiproliferative agents 213
Corticosteroids 213
Sirolimus 214
Malignancy 214
Infection 215
Diabetes mellitus 218
Bone mineral density loss and osteoporosis 219
Chronic kidney disease 220
Hypertension 221
Reproductive health and pregnancy 221
Psychological disorders 222
Recurrence of heart failure and retransplantation 223
021

210 Chapter 18 Chronic complications of heart transplantation

Cardiac allograft vasculopathy

Definition
• Cardiac allograft vasculopathy (CAV) is a form of aggressive CAD that
occurs following cardiac transplantation.
• Accelerated fibroproliferative disease resulting in pathological smooth
muscle proliferation within the intima of the coronary vessels, with
accumulation of inflammatory cells and lipid deposition.
• Circumferential intimal thickening—​diffuse affecting epicardial and
intramural coronary vessels (note contrast to coronary artery
atherosclerosis).
• Almost one in three patients develop some form of CAV at 5 years and
the condition is said to account for one in eight deaths beyond a year.
Pathophysiology
Involves a combination of immunological and non-​immunological factors
contributing to vascular inflammation, endothelial injury, and a cascade of
vascular cell proliferation, fibrosis, and adverse remodelling (Table 18.1).

Table 18.1 Cardiac allograft vasculopathy pathophysiology

Immunological factors Non-​immunological factors
Histocompatibility mismatch Age
Presence of HLA antibodies Sex
Rejection—​both cellular/​antibody Obesity
mediated
Dyslipidaemia
Hypertension
Smoking
CMV infection
Brain death
Organ preservation
Ischaemia-​reperfusion injury
Donor native atherosclerosis

Diagnosis
The recognition of early CAV as an adverse prognostic marker has led to
awareness of the need for surveillance and early detection. Diagnosis can
be challenging:
• Due to the cardiac denervation at the time of transplant, most patients
do not experience anginal chest pain.
• Presentation can involve graft dysfunction noted on imaging, clinical HF,
dysrhythmia, or even sudden cardiac death.
• Coronary angiography can be an insensitive tool due to the diffuse and
concentric nature of CAV—​it cannot visualize beyond the arterial lumen
or beyond the larger epicardial vessels. Often luminal obstruction is a
late phenomenon in CAV.
 Cardiac allograft vasculopathy 211

• Intravascular ultrasound can be employed in addition to routine

coronary angiography, providing cross-​sectional imaging of the lumen
and vessel wall.
• The use of optical coherence tomography can provide high-​resolution
intravascular imaging that is suited to assess the vessel intimal and
consequent plague morphology. However, it is limited by cost, contrast
load, and lower tissue penetration.
• Imaging modalities such as dobutamine stress echocardiography,
myocardial perfusion imaging, cardiac MRI, and CT coronary
angiography may all have a role in detection of occult CAV.
Treatment and prevention
• Clinically, the main focus in treatment of cardiac allograft vasculopathy
remains preventative. Primary prevention of CAV in heart transplant
recipients should involve strict control of cardiovascular risk factors
(hypertension/​diabetes/​hyperlipidaemia/​smoking/​obesity) as well as
strategies to prevent CMV infection (ISHLT class I, level of evidence C).
• Aspirin: tends to be used empirically in routine post-​transplant care,
although little evidence to support its use (ISHLT class I, level of
evidence C).
• Statin therapy: unfortunately, hyperlipidaemia is often the result of
treatment with immunosuppressive regimens including steroids,
ciclosporin, and tacrolimus. Statin therapy should be considered for all
transplant recipients (ISHLT class I, level of evidence A).
• Mycophenolic acid: reduces progression of intimal thickening as
compared to azathioprine and is the most preferred antimetabolite.
• In cardiac transplant recipients with established coronary allograft
vasculopathy, the substitution of mycophenolate mofetil (MMF)/​
azathioprine with sirolimus or everolimus can be considered (ISHLT
class IIa, level C).
• Annual or biannual coronary angiography should be considered to
assess the development of CAV (ISHLT class I, level C).
• Treatment options such as coronary angioplasty, PCI, and redo
transplantation offer symptom control and palliative solutions once
cardiac allograft vasculopathy has been established:
• PCI may be limited due to the diffuse nature of the disease, distal
vessel pruning, presence of diabetes/​dyslipidaemia/​renal dysfunction,
and high rates of in-​stent restenosis.
• Drug-​eluting stents are recommended (ISHLT class IIa, level of
evidence C) and follow-​up coronary angiography is recommended
6 months after a percutaneous intervention (ISHLT class I, level of
evidence C).
• Surgical revascularization is an option in highly selected patients but is
associated with a high operative mortality.
• The option of repeat cardiac transplantation exists, although this is
rare in clinical practice and is associated with increased mortality and
increased rates of subsequent CAV.
21

212 Chapter 18 Chronic complications of heart transplantation

Common general side effects and

complications of immunosuppression
Immunosuppression is essential in all heart recipients to prevent allograft
rejection. As with all pharmacotherapy, these can be associated with side
effects and interactions.
• Calcineurin inhibitors (CNIs) (tacrolimus/​ciclosporin).
• Antiproliferative agents (mycophenolate mofetil (MMF)/​azathioprine).
• Corticosteroids.
• Mechanistic target of rapamycin (mTOR) inhibitors (sirolimus/​
everolimus).
A typical immunosuppression regimen following cardiac transplantation in-
cludes an antiproliferative agent (e.g. MMF or azathioprine), a CNI (e.g.
tacrolimus or ciclosporin), and, initially, a corticosteroid.

Calcineurin inhibitors
CNIs antagonize interleukin-​2 (IL-​2). The most commonly encountered
agents are tacrolimus and ciclosporin. These each have different branded
preparations. As different brands have potential diversity in bioavailability, a
patient should ideally not be switched between brands or from branded to
generic preparations without caution.
Tacrolimus
Complications/​side effects
• Nephrotoxicity—​care with co-​administration of other nephrotoxic
agents (e.g. gentamicin/​vancomycin, NSAIDs).
• Diabetes mellitus.
• Neurotoxicity (most seriously—​posterior reversible encephalopathy
syndrome (PRES). Also, peripheral neuropathy, tremor, headache,
paraesthesia).
• Electrolyte disturbance: hyperkalaemia (care with co-​administration
of other drug causing hyperkalaemia, e.g. spironolactone/​ACEi, etc.),
hypomagnesaemia.
• Nausea.
• Diarrhoea.
• Alopecia.
Ciclosporin
Side effects
• Hypertension.
• Nephrotoxicity—​as per tacrolimus.
• Neurotoxicity (headaches, tremors, convulsions).
• Hyperkalaemia and hypomagnesaemia.
• Gingival hyperplasia.
• Hirsutism.
• Hepatic dysfunction.
 Corticosteroids 213

Antiproliferative agents
Antiproliferative agents are generally required lifelong. The most commonly
encountered agents are MMF and azathioprine.
Common side effects
• GI: nausea, anorexia, diarrhoea, oesophagitis, and abdominal pain most
common.
• Minimized by taking the dose with food, or a dose reduction may be
necessary.
• Changing from twice-​daily to four times daily dosing may also help.
• Bone marrow suppression: MMF is less myelosuppressive than
azathioprine.
• Adverse effects are usually reversed on discontinuation of MMF.
• Stop or reduce dose if white cell count <4 × 109/​L, or
thrombocytopenia, cholestatic jaundice, or pancreatitis.
• Teratogenic: counsel young patients (male and female) accordingly as
per Medicines and Healthcare products Regulatory Agency (MHRA)
guidance.
Azathioprine
Usually given second line when MMF is not tolerated—​typically, due to GI
side effects.
Caution
Consider withholding in situations where bone marrow suppression is a
concern, such as leuco/​neutropenia, thrombocytopenia, or if there is evi-
dence of jaundice.
Side effects
• Haematological abnormalities:
• Bone marrow suppression.
• Macrocytic anaemia.
• Hepatic abnormalities:
• Jaundice.
• Transaminitis.

Corticosteroids
Duration of treatment and dosing
• Corticosteroids are commenced immediately post transplantation,
and are converted to an oral steroid tapering regimen (typically
prednisolone).
• Corticosteroids are generally commenced at high dose during or
immediately post transplantation. Weaning and withdrawal may be
considered 3–​6 months after transplant in low-​risk patients (e.g. no
circulating anti-​HLA antibodies, non-​multiparous women, no history of
rejection, older age).
Consider weaning if corticosteroids cause significant side effects and there
have been no recent (i.e. within 6 months) episodes of rejection.
421

214 Chapter 18 Chronic complications of heart transplantation

Sirolimus
Sirolimus is a mTOR inhibitor. It is not first line but used in certain
circumstances:
• Renal-​sparing chronic immunosuppressive regimen (>3 months from
transplant) in place of CNI.
• Maintenance immunosuppression in place of CNI in post-​transplant
lymphoproliferative disorder (PTLD).
• Maintenance immunosuppression in addition to CNI in significant CAV.
• Rejection refractory to tacrolimus and MMF.
Side effects
• Lipid disturbances: hyperlipidaemia, hypercholesterolaemia.
• Bone marrow suppression: thrombocytopenia, leucopenia, anaemia.
• Pharyngitis (usually transient).
• GI—​diarrhoea, oral aphthous ulceration.
• Acneiform and other skin rashes.
• Significant proteinuria/​nephrotic syndrome.
• Impaired wound healing—​therefore caution needed in peri-​transplant
period.
• Drug-​induced pneumonitis—​rare but potentially serious.

Malignancy
More than 10% of heart transplant recipients developed a de novo malig-
nancy 1–​5 years post transplantation.
Skin cancers
Compared to the incidence of skin cancer in the general population, in
heart transplant recipients, the incidence is approximately 65–​250 times
more frequent. More than 90% of cases are due to squamous and basal
cell carcinomas. Squamous cell carcinoma tends to be more aggressive in
the post-​transplant patient in comparison to the general population, with
greater metastatic potential and more local recurrence noted.
Factors that contribute to the development of skin malignancies:
• Duration of immunosuppression.
• Intensity of the immunosuppressive regimen used.
• Increasing age.
• Male sex.
• Caucasian.
• Smoking.
• Sun exposure.
• The presence of Epstein–​Barr virus (EBV), human papillomavirus,
herpes simplex virus (HSV), herpes zoster virus, and polyoma.
In clinical practice, yearly review at a specialist dermatology clinic is recom-
mended and patients are counselled regarding reducing sun exposure.
Post-​transplant lymphoproliferative disorder
PTLD encompasses disorders ranging from infectious mononucleosis to
malignant lymphoma (T-​cell/​B-​cell lymphoma, Hodgkin’s lymphoma, and
Hodgkin’s like PTLD).
 Infection 215

• Results from uncontrolled lymphoid growth in the immunosuppressed

patient.
• PTLD has been reported in 6.4% of cardiac transplants.
• Over a 10-​year period, the risk of a post-​transplant patient developing
lymphoma is 11.8 times higher than the general population.
• The pathogenesis of PTLD is a combination of immunosuppressive
agents, infection by oncogenic viruses such as EBV, and possible genetic
and epigenetic mutations.
Presentation
Often non-​specific—​fever, lymphadenopathy, weight loss, abdominal pain,
splenomegaly, etc. Often presents at extranodal sites (e.g. GI tract).
Diagnosis
Excisional biopsies are preferred and fine needle aspiration should be
avoided where possible. Staining for EBV is mandatory.
Treatment
• Modification and reduction of immunosuppression regimen wherever
possible.
• Surgery for localized disease.
• Localized radiotherapy.
• Targeted chemotherapy (most commonly used regimen CHOP/​
rituximab).
• More recently, adoptive T-​cell immunotherapy (via infusion of EBV-​
specific cytotoxic T lymphocytes in patients with EBV-​positive PTLD)
has shown some promise but presently is far from established in clinical
practice.
Solid organ malignancies
There is no specific evidence to recommend reduction in immunosup-
pression in patients with solid tumours unrelated to the lymphoid system.
A multidisciplinary team approach involving the transplant team and onco-
logists to tailor strategies on a case-​by-​case basis is advised.
Heart transplant recipients should undergo routine screening for breast,
colon, and prostate cancer as per the general population recommendations.

Infection
Infection in transplant patients can range from mild to fatal.
General principles
• Infection is one of the leading causes of death in transplant recipients.
• The need for effective lifelong immunosuppression places patients at an
increased risk of developing infection.
• Pathogens may be bacterial, viral, fungal, or protozoal.
• The pathophysiological response to infection is altered which can make
timely diagnosis challenging.
• Responsible microorganisms can differ from those affecting the general
population.
• Common microorganisms include CMV, Candida species, Gram-​
negative bacilli, staphylococci, and Pneumocystis jirovecii.
621

216 Chapter 18 Chronic complications of heart transplantation

• Common sites of infection include the respiratory tract, mouth, skin,

and urinary tract.
• Less common sites such as CNS infection should always be considered.
• Consider temporary suspension of antiproliferative agents in severe
infection or leucopenia/​neutropenia.
• In all cases, the cardiac transplant recipient with infection should be
managed in close discussion with the transplant medicine, infectious
diseases, and microbiology team.
Physiological response to infection in
the immunosuppressed
• Immunosuppressive medication (particularly antiproliferative agents)
attenuate the physiological ability to generate pyrexia and leucocytosis
in response to infection.
• Corticosteroids do not appear to have these effects, but may mask
clinical signs of infection, allowing the condition to worsen and present
late.
CMV infection
• Symptoms can be initially be constitutional and generalized if CMV
viraemia with no end-​organ involvement.
• Sometimes there may be focal organ involvement and specific related
symptoms (e.g. retinitis, pneumonitis, colitis, hepatitis).
• Treatment guided by monitoring blood for CMV by PCR.
• Antimicrobial therapy generally involves ganciclovir first line.
Pneumocystis jirovecii infection
See Fig. 18.1.
• Consider in the patient with shortness of breath or other respiratory
tract symptoms.
• Pulse oximetry on exertion should be performed. Classically, this would
show an exertional drop in saturations.
• There should be a low threshold for bronchoalveolar lavage if there is a
high index of suspicion.
• Management is with respiratory support and aggressive antimicrobial
therapy—​usually co-​trimoxazole (adjusted for renal function).
• Alternatives (e.g. in G6PD deficiency, sulfa allergies) include a
combination of dapsone ± trimethoprim or pyrimethamine,
atovaquone, or clindamycin and pyrimethamine.
• Consider concurrent steroids and folinic acid.
• Oral prophylaxis should continue indefinitely following recovery.
Infective endocarditis prophylaxis
• Limited evidence to support any particular approach.
• Transplant recipients at increased risk of valvular disease.
• Outcomes in infective endocarditis are extremely poor.
• Reasonable to give antibiotic prophylaxis prior to dental procedures in
transplant recipients with valvular heart disease.
 Infection 217

(a)

(b)

Fig. 18.1 Anteroposterior CXRs demonstrating Pneumocystis jirovecii infection. Sternotomy

wires from cardiac transplantation, VV ECMO cannula in situ (a), and left intercostal chest
drain for associated left-​sided pneumothorax (b) are also seen. See plate section.
821

218 Chapter 18 Chronic complications of heart transplantation

Diabetes mellitus
• Deranged glucose metabolism and development of diabetes mellitus is
common following cardiac transplantation.
• The chances of developing diabetes post transplantation increases with
time and may occur as late as 15 years post transplantation.
• Some series report rates up to 32% at 5 years following heart
transplantation.
• The International Consensus Guidelines for new-​onset diabetes after
transplantation (NODAT) were published in 2003, and although based
on renal transplant patients, they have been adopted in the setting of
other solid organ transplants.
Risk factors
• Risk factors for developing diabetes following heart transplantation
are not well defined, but higher preoperative blood glucose levels,
corticosteroid dose post-​transplantation, and family history have been
predictive in some populations.
• In liver and kidney transplantation, risk factors are more extensively
defined and include:
• Black/​Hispanic ethnicity.
• Family history of diabetes.
• Age >40 years.
• Pre-​transplant glucose intolerance.
• Immunosuppressive agents.
• Metabolic syndrome.
• Hepatitis C infection.
• Cadaveric kidney.
• Obesity.
• In many cases, diabetes may develop as a result of immunosuppressive
medication (e.g. corticosteroids, tacrolimus) which cause
hyperglycaemia by increasing insulin resistance.
• Ciclosporin confers less of a risk than tacrolimus.
Consequences of developing diabetes following heart
transplantation
Though data are limited to relatively small populations, developing diabetes
following heart transplantation appears to be associated with worse out-
comes, notably, increased risk of serious infection, reduced long-​term sur-
vival, increased risk of ischaemic heart disease, cardiovascular mortality, and
risk of developing CAV.
Diagnosis
The NODAT guidelines recommend that diabetes in the post-​transplant
population should be diagnosed by the same criteria as used for the general
population, namely:
• Symptoms of diabetes, and:
• Random glucose ≥11.1 mmol/​L or 200 mg/​dL or
• Fasting glucose ≥7 mmol/​L or 126 mg/​dL or
• Plasma glucose ≥11.1 mmol/​L or 200 mg/​dL 2 hours following a 75 g
glucose load.
 Bone mineral density loss and osteoporosis 219

• HbA1c has been adopted as a diagnostic test for diabetes since the
publication of the NODAT guidelines in 2003. A threshold of ≥48
mmol/​mol or 7.8% is diagnostic of diabetes mellitus in the presence of
symptoms.
• In asymptomatic patients, HbA1c should be repeated in 6 months, or
sooner if risk of developing diabetes is high.
Management
• Finger-​prick glucose should be monitored weekly for the first 4 weeks
following transplantation, then at 3 and 6 months, then annually.
• If a patient develops diabetes, assess other risk factors such as serum
lipid profile, blood pressure, and evidence of end-​organ complications.
• Stepwise management of diabetes will include:
• Conservative measures: dietary and lifestyle advice.
• Oral anti-​h yperglycaemics.
• Insulin.
• Combination of oral anti-​h yperglycaemics and insulin.
• Transplant-​specific considerations: modification of
immunosuppression (reducing corticosteroid dose, ciclosporin as an
alternative to tacrolimus).
• A specialist in diabetes should be involved in the care of the post-​
transplant patient with new diabetes.
• Disordered lipid profile and hypertension should be treated aggressively
to reduce the risk of diabetic complications.
Oral anti-​hyperglycaemic agents
• The choice of oral agent should be tailored to the patient based
on factors such as weight, renal function, and occupation (i.e. if
hypoglycaemic episodes would be dangerous).
• Thiazolidinediones can cause fluid retention and HF. Conversely,
sodium-glucose co-transporter 2 (SGLT2) inhibitors reduce adverse
events in HF with reduced ejection fraction. No robust data exist
examining their use in the post-​heart transplant recipient.

Bone mineral density loss and

osteoporosis
• Cardiac transplant recipients require prolonged courses of prednisolone
or high doses of IV methylprednisolone (in cases of rejection) and are
therefore at high risk of developing corticosteroid-​related bone disease.
• Prevention strategies should be considered early in a patient’s
transplantation journey, particularly as rapid bone loss and fractures
generally occur during the first 3–​12 months post transplantation.
Pre transplantation
All adult heart transplant patients should be screened for pre-​existing bone
disease prior to placement on the transplant waiting list wherever feas-
ible. Baseline bone mineral density is obtained with a dual-​energy X-​ray
20

220 Chapter 18 Chronic complications of heart transplantation

absorptiometry scan of lumbar spine and neck of femur. A bone mineral

density score is calculated (T score and Z score).
The presence of pre-​existing low bone mineral density should prompt
further evaluation and treatment of any correctable causes. Bisphosphonate
therapy is recommended where indicated.
All patients are advised on the recommended daily allowance of cal-
cium (1000–​1500 mg) and vitamin D (400–​1000 IU) to maintain serum 25-​
hydroxyvitamin D levels >30 ng/​mL.
Post transplantation
Patients are encouraged regarding regular weight-​ bearing and muscle-​
strengthening exercise.
All patients should begin regular bisphosphonate therapy immediately
after transplantation and continue for at least the first post-​transplant year
(or longer if necessary). IV therapy is an option for patients who are unable
to safely swallow tablets in the immediate post-​transplant phase.
Bone mineral density is usually reassessed at the 1-​ year post-​
transplant mark. It is reasonable to stop the bisphosphonate if cortico-
steroids have been discontinued and bone mineral density is preserved
(T score >1.5).

Chronic kidney disease

Causes
Renal dysfunction can result from a variety of causes including hypertension,
immunosuppressive therapy (in particular, CNIs), and recurrence of HF.
Screening
Check an eGFR at least annually, more frequently if <60 mL/​min/​1.73 m2
and/​or decline by >4 mL/​min/​1.73 m2 per year.
Management
• Consider referral to specialist renal services if eGFR decreases to <30
mL/​min/​1.73 m2.
• Measures that have been showed to slow progression of chronic kidney
disease in the general population should be applied to heart transplant
recipients:
• Aggressively control abnormalities in blood pressure and blood
glucose.
• Consider introducing an ACEi/​ARB.
• Check haemoglobin at least annually.
• If low, check iron status and commence erythropoiesis-​stimulating
agents where appropriate, targeting haemoglobin 110–​130 mg/​L or
11–​13 g/​dL.
 Reproductive health and pregnancy 221

• End-​stage renal failure should be preferentially managed with kidney

transplantation in all heart transplant recipients for whom this is
appropriate.
Immunosuppression adjustment
• Renal-​sparing immunosuppression strategies can be considered in select
cases. These can involve either replacement of a CNI with a different
agent (e.g. sirolimus) or stopping CNI with subsequent maintenance on
a corticosteroid/​antiproliferative agent dual-​agent regimen. Caution is
advised in patients with a history of rejection or those with circulating
donor specific antibodies.
• If CNIs are continued, lower the target level to the minimum acceptable
to prevent rejection.

Hypertension
• Although there is limited specific evidence pertaining to hypertension in the
post-​cardiac transplant population, it is generally recognized that the same
blood pressure targets should be sought as for the general population.
• Lifestyle factors such as increased exercise, weight loss, and reducing
dietary salt and fat should be advocated.
• Medical management should follow established guidance, and will
generally involve calcium channel blockers or ACEis/​ARBs as first-​line
therapy.
• Calcium channel blockers are often preferred due to their renal-​sparing
properties.
• Both dihydropyridine and non-​dihydropyridine calcium channel blockers
can interfere with metabolism of CNIs, resulting in CNI toxicity.
• As such, close monitoring of therapeutic CNI levels should be
considered when commencing a calcium channel blocker.
• Presence of other comorbidities should be taken into account as this
may guide choice of an initial agent (e.g. ACEi/​ARB in patients with
diabetes mellitus).
• Combination of two agents (e.g. ACEi/​ARB and calcium channel
blocker) may be required to achieve adequate control.
• Screening for additional risk factors such as high cholesterol or diabetes
mellitus should be undertaken.
• End-​organ damage should be assessed for and treated, such as urine dip
for protein/​formal urinary protein:creatinine to assess for nephropathy,
and fundoscopy to assess for hypertensive retinopathy.

Reproductive health and pregnancy

Pregnancy
• Having children is very important to many patients following cardiac
transplantation.
• Clinicians should take a patient-​centred approach to discussions
surrounding pregnancy and should support patients and their families.
2

222 Chapter 18 Chronic complications of heart transplantation

• Ideally, pregnancies should be planned and monitored with involvement

from the multidisciplinary team. This would typically include specialists
such as an obstetrician specializing in maternal and fetal medicine, a
neonatologist, psychologist, clinical geneticist, transplant cardiologist, an
anaesthetist, and the patient’s primary care doctor and midwife.
Female patients
Pre-​conception counselling
• The approach to managing a pregnant transplant recipient should be
individualized according to patient factors such as immunosuppressive
regimen, graft function, time since transplantation, and the patient’s
own wishes and preferences.
• Ideally, pregnancy would be delayed until at least 1 year following
cardiac transplantation.
• The likely risks of complications such as infection, rejection, cardiac
graft vasculopathy, and fetal abnormality, should be considered and
communicated to the patient.
• If, despite the best efforts to modify them, the risk of any of the above
factors is high enough to preclude a successful pregnancy and birth, the
patient should be advised against pregnancy.
Immunosuppression
• Immunosuppressive regimen should be reviewed and modified by
reducing/​switching any agents likely to be teratogenic or toxic to fetal
growth, and graft function assessed on the modified regimen to ensure
stability.
• MMF should be discontinued.
• CNIs and corticosteroids can be continued.
• CNI levels may fluctuate due to physiological changes in plasma volume
and renal function in pregnancy and should be monitored closely.
Male patients
• Patients hoping to achieve pregnancy with a female partner should have
their immunosuppressive regimens reviewed and modified.
• MMF should be discontinued.
Hormonal contraception
As these agents inhibit cytochrome P450, monitor CNI drug levels carefully.
Barrier methods
Barrier methods such as the male or female condom should be advocated
as an adjunct to other methods of contraception and to reduce risk of sexu-
ally transmitted infection.

Psychological disorders
General principles
• Robust psychological assessment is mandatory prior to cardiac
transplantation to ensure candidacy for the procedure.
• A nurse specialist or psychologist should be accessible for all patients
following cardiac transplantation for any problems that may arise.
 Recurrence of heart failure and retransplantation 223

Adherence to therapy
• Review the patient’s adherence to therapy at each outpatient visit, being
sure to identify and address any barriers to adherence using a patient-​
centred solution-​focused approach.
Depression
• Screen for evidence of depression routinely during follow-​up.
• Any patients found to have depression should have input from a
psychologist and the wider multidisciplinary team.
• Pharmacotherapy:
• Selective serotonin reuptake inhibitors do not impact haemodynamics
or cardiac conduction so represent a good choice.
• Avoid cytochrome P450 agents as these may interact with tacrolimus
metabolism and alter levels.

Recurrence of heart failure and

retransplantation
• May arise from a range of pathologies—​valvular failure (e.g. TR arising
from damage from right heart procedures such as endomyocardial
biopsy), chronic rejection, or ischaemic heart disease from graft
vasculopathy.
• Limited evidence base for recurrent cardiac failure following cardiac
transplantation, but generally managed as for failure in a native heart
with disease-​modifying therapies—​Angiotensin receptor-neprilysin
inhibitor (ARNI)/ACEi/ARB, BB, MRA, SGLT2 inhibitor.
• Retransplantation can be considered in the following circumstances:
• CAV not amenable to medical therapy, or revascularization either
percutaneously or surgically, where there is clinical evidence of HF or
myocardial ischaemia.
• CAV not amenable to medical therapy or revascularization, with
asymptomatic moderate to severe LV systolic dysfunction.
• Allograft dysfunction and clinical manifestations of HF without
evidence of acute rejection.
• The decision to list for retransplantation should be made by the wider
transplant multidisciplinary team, taking into consideration changes in
the patient’s comorbidities and biopsychosocial circumstances, as these
may affect eligibility.
• The patient’s views and wishes should be taken into account in any
decision to pursue retransplantation.
24
 Chapter 19 225

Bedside care and

rehabilitation
with mechanical
circulatory support and
thoracic transplantation
Introduction 226
Rehabilitation of patients with IABP and MCS 226
Rehabilitation post heart/​lung transplantation 229
26

226 Chapter 19 Bedside care and rehabilitation

Introduction
Myopathy and polyneuropathy of critical illness is a well-​recognized major
complication in ICU patients. Hence, there is a move towards early mobil-
ization and physical therapy (PT) to optimize postoperative recovery and to
reduce the rate of physical and psychological deconditioning.
In this chapter, we present strategies for ambulation and rehabilitation in
patients with long-​term IABP, MCS, and those who have undergone cardio-
pulmonary transplantation.

Rehabilitation of patients with IABP

and MCS
There is ample evidence that early mobilization after implantation of IABP
and MCS confers great benefits to patients both physically and psycho-
logically while they await organ transplantation. However, there are some
inherent risks with ambulation, which should be weighed against its bene-
fits. Furthermore, ambulation of HF patients on IABP or MCS is very time
and resource consuming. There are a few barriers to ambulation, some
of which include haemodynamic instability, presence of multiple lines and
cannulas, muscle weakness, altered mental status, and shortage of staff or
equipment required to ambulate the patient safely.
MCS patients may have up to four additional cannulas in the case of
extracorporeal biventricular support as compared to the routine adult car-
diac surgery patients. On transport, ECG, pulse oximetry, and PA and ar-
terial lines pressures should be monitored closely. However, interpretation
of haemodynamic parameters for MCS patients may be more difficult. The
pump speed and flow parameters should also be taken into consideration
prior to ambulation.
Prior to ambulation of the IABP or MCS patient, anticoagulation status
should be noted; suboptimal anticoagulation and presence of thrombus in
the oxygenator/​external tubing may preclude ambulation due to risk of
embolization. Overzealous anticoagulation predisposes patients to bleeding
(trauma from falls, stress on cannulation sites and wounds).
IABP ambulation strategy
IABPs are traditionally inserted via the femoral artery, which limits the pa-
tient in terms of PT, as ambulation is often institutionally prohibited with the
device in the femoral artery position. Not only is insertion site bleeding a
possibility, but the IABP may become kinked with hip flexion leading to mal-
function or rupture during ambulation. Patients with femoral IABPs should
be turned at least 30° from side to side every 2 hours, have active and
passive range of motion, and have the bed placed in reverse Trendelenburg
position.
Alternative sites which would allow for more liberal ambulation are
where IABP catheters are sited in the axillary or subclavian positions (see
Chapter 9). These alternative positions would allow for mobilization out of
bed, sitting on the chair or the commode, and walking. However, axillary or
subclavian positions preclude arm raising above the head level on the side
of the IABP to avoid the risk of catheter displacement. An elbow immobil-
izer may be used.
 Rehabilitation of patients with IABP and MCS 227

Ambulation should be performed in stages to avoid orthostatic hypo-

tension. The patient and the medical team should be mindful of the in-
creased risk of catheter migration with ambulation, which may require IABP
repositioning. Patients with an axillary IABP should have the radial pulse
on the ipsilateral side verified and the function and sensation of the upper
extremity assessed before and after ambulation. If there is a suspicion of
IABP displacement, a plain CXR would be indicated and repositioning can
be undertaken at the bedside using fluoroscopy.
MCS ambulation strategy
ECMO and LVAD may be instituted centrally or peripherally. Common per-
ipheral sites include the IJ veins, subclavian veins, the subclavian or axillary
arteries, and femoral arteries and veins. Ambulation with an open sternum
is contraindicated. Femoral cannulation may prove ambulation more chal-
lenging but is not a contraindication. While all cannulas are affixed securely
at the time of cannulation, these should be examined before and after am-
bulation regimentally as there remains an inherent risk of cannula displace-
ment leading to flow interruptions, limb ischaemia, and vascular injury.
The procedure for mobilizing patients with all forms of MCS is similar,
although patients requiring ECMO are typically more acutely ill and require
more support.
The following checklist should be completed prior to ambulation of MCS
patients:
• All team members have discussed pertinent information and safety
concerns.
• The patient can follow verbal directions and communicate answers to
questions.
• For patients unable to verbalize, ‘yes’ and ‘no’ signals are clearly
demonstrated.
• Body strength assessment is adequate.
• The patient is doing well (no significant bleeding, excessive fluid volume
requirements, or electrolyte abnormalities).
• The patient has maintained adequate baseline oxygenation and
haemodynamic stability in the sitting and standing positions.
If the patient does not meet all of these criteria or if adequate resources
are not available to ambulate the patient safely, ambulation session should
be postponed. Ideally, the patient should receive daily ambulation with PT
which should be detailed in their occupational therapy plan of care.
Recommended ambulation procedure checklist
• Prior to each ambulation session, the aforementioned checklist is
validated.
• Roles are clearly assigned and defined among the ambulating team.
• The route along the ward is created and decluttered.
• Two personnel jointly assess that the cannula sites are secure for
ambulation (e.g. an arm sling for upper chest cannulation immobilizes
that arm). A headband with self-​adhering wrap or indwelling catheter
securing devices secures neck cannulas to the head. Binders are worn
for tunnelled cannulas.
28

228 Chapter 19 Bedside care and rehabilitation

• The MCS device console battery is checked prior to each ambulation

and the gas tanks are verified to be adequately filled. Emergency back-​
up equipment or hand cranks are available as are tubing clamps and
hospital standard MCS emergency supplies.
• The patient should stand only with the assistance of a fully trained
physical therapist.
• Prior to each ambulation, the patient will perform weight shifting and
marching on the spot. If stable with these, the patient will then take a
few test steps prior to leaving the room. If a concern is raised by the
patient or any member of the team regarding ambulation at this time,
the effort will be paused, and issues rectified.
• The primary nurse will watch the monitor at all times and alert the team
of any changes.
• A staff member will be in direct physical contact with the patient at
all times.
• One clinical provider should carry a chair behind the patient. Should
ambulation need to stop prior to returning to the room, the patient will
sit on a chair and be wheeled back.
• Upon returning to the room, the care providers should reinstitute
monitoring in the patient’s room as per routine, remove stabilization
equipment, and secure cannulation sites.
• Two staff members will re-​examine cannulation sites.
Durable LVAD ambulation strategy
Durable (long-​term) LVAD patients should be educated on how to in-
corporate the controller and batteries into their day-​to-​day activities,
and physically adjust to the additional weight of the batteries and con-
troller. The changes in gait, posture, and balance from the additional
weight of the device and its batteries may result in early fatigue and can
lead to falls.
PT in durable LVAD patients should be commenced in the immediate
post-​LVAD implantation period in the ICU, once the chest is closed, the
patient is extubated, and is haemodynamically stable. A physiotherapist fa-
miliar with LVADs should evaluate the patient, set goals, and create a treat-
ment plan. PT usually starts with lower extremity exercises in bed, followed
by sitting on the edge of the bed, standing, and then gait training with a
walker. Adverse effects of PT with durable LVAD patients are few and de-
creased pump flows are common. Keeping LVAD flows at >3 L/​min during
PT is recommended.
The driveline should be well secured to the skin and immobilized with
a binder since excessive movement and jarring can result in driveline in-
fections. Movements should be slow and steady, as rapid standing can
cause orthostatic hypotension. Kinking or twisting of the driveline should
be avoided. The PT session should be discontinued if there are significant
fluctuations in LVAD flows, hypotension, chest pains, extreme fatigue, or
by patient request.
Outpatient cardiac rehabilitation in durable VAD recipients has been as-
sociated with lower re-​hospitalization rates and mortality risk.
 Rehabilitation post heart/lung transplantation 229

Nutrition
Commencing feeds within 24 hours is recommended in the LVAD patient to
prevent weight loss and ileus; feeding in VA ECMO patients within 2 days of
its initiation has been associated with decreased mortality.
In many cases, enteric feeding tubes are placed as soon as the patient
is haemodynamically stable and coagulation screen is appropriate. If the
nasoenteric feeding tube can be placed before systemic anticoagulation
is started, the risk of epistaxis is lower; epistaxis can be severe in the
anticoagulated patient and can lead to aspiration.
Lung transplant patients with gastro-​oesophageal reflux disease have a
higher incidence of PGD, and should ideally be fed via a post-​pyloric en-
teric tube (initiated at 10–​15 cc/​hour, then advanced to meet caloric goals).
Swallowing assessment should be performed prior to commencing oral nu-
trition to avoid aspiration.
The ICU cardiopulmonary transplant and MCS patients are routinely
suboptimally fed, for example, due to haemodynamic instability, GI in-
tolerance, or if they are fasted pending further procedures or operations.
Signs such as abdominal distention/​pain, vomiting, residual gastric volumes
>200 mL (if feeding in the stomach), and diarrhoea can indicate intoler-
ance of enteral nutrition, which may be helped by prokinetic agents. Use of
muscle relaxants, sedation, or prone positioning are not contraindications
to nasoenteric feeding.
Daily protein requirements for patients on ECMO have been recom-
mended at 1.2 g/​kg ideal body weight, with energy requirements deter-
mined by using the Schofield equation and multiplying by 1.1–​1.2; sodium
administration should be kept to <2000 mg/​day. If the patient is obese
(BMI >30 kg/​m2), consider using the patient’s ideal body weight to deter-
mine protein and caloric needs.

Rehabilitation post heart/​

lung transplantation
Following heart/​lung transplantation, close observation and monitoring is
of paramount importance. Some of these observations are as follows:
• Respiration rate, oxygen saturations, oxygen therapy, and arterial
blood gases.
• Blood pressure, heart rate, and temperature.
• CVP.
• Temporary pacing.
• Telemetry.
• Chest drain output.
• Fluid balance (avoid fluid overload).
• Daily weight (echocardiogram and diuretics if weight increasing).
• Blood tests (complete blood count, renal function, and coagulation
screen).
• CXR (looking for fluid overload, effusion, pneumothorax, etc.).
• Immunosuppressant level assay.
• Nutritional assessment.
• Serum glucose measurement.
• Evidence of infection (wounds, respiratory, lines).
230

230 Chapter 19 Bedside care and rehabilitation

Pain management
Optimal pain management is key to patient comfort, and speedy recovery
in the postoperative period. Effective pain management improves the ability
to take deep breaths, cough and expectorate, sleep, mobilize, and per-
form self-​care. The World Health Organization (WHO) pain ladder should
be adhered to. In cases of inadequate pain control, consultation with the
‘chronic pain team’ would ensure sufficient follow-​up and continuity of care
regarding pain management.
Wound management
Sternotomy, thoracotomy, and clamshell wounds can take up to 4 months
to fully heal in routine heart–​lung surgery. However, the impact of major
surgery, combined with long-​term chronic illness, deconditioned state,
comorbidities, and immunosuppressive drug regimen, means that the trans-
plant recipients often experience a longer period of wound healing and
carry a higher risk of wound complications such as superficial (involving
skin/​subcutaneous tissue) or deep wound infections (involving the medi-
astinum/​sternum).
Psychological rehabilitation
Emotional health affects physical health. The most common presentations,
following the acute physiological impact of cardiopulmonary transplant-
ation include anxiety, depression, post-​traumatic stress, and ‘survival guilt’.
Occasionally, recipients may develop steroid psychosis, delirium, delusions,
or hallucinations. It is important, therefore, to consider the role of medica-
tions in any psychological presentation. Patient concerns regarding physical
challenges of lifestyle modification following cardiopulmonary transplant-
ation, fear of life-​or limb-​threatening complications, and mortality, as well
as non-​physical challenges (e.g. financial and social), lead to psychological
burden which would require support and intervention. Psychological/​
psychiatric support should be sought readily in such patients. The use of
antidepressants, antipsychotics, or anxiolytics can be helpful in overcoming
some of the sequalae of acute psychiatric decompensation. However, the
medical and social services team should, whenever possible, strive to re-
solve the underlying aetiology of the stressors, for example, by providing
social support or fund raising for financial support to ensure the best pos-
sible prognosis.
Physical rehabilitation
Rehabilitation is influenced by the length of time in ICU. Skeletal muscle
mass loss can be as much as 5–​20% after 7–​10 days in the ICU.
A programme of physical exercise post transplantation would aim to
improve muscle strength and exercise tolerance. It is also beneficial in re-
ducing risks of hypertension, obesity, and osteoporosis and psychological
dysfunction.
A combination of aerobic, resistance, and endurance exercise and physical
rehabilitation is beneficial to both heart and lung recipients. Physiotherapy
is implemented immediately post transplantation to optimize recovery and
continues to be expanded throughout the rehabilitation programme.
A programme is determined by the multidisciplinary team of clinicians,
physiotherapists, and occupational therapists and depends on various fac-
tors including:
 Rehabilitation post heart/lung transplantation 231

• Degree of frailty prior to transplantation.

• Organ implanted (heart or lung).
• Physiological alterations post transplantation.
• Complications post surgery.
• Comorbidities.
Physiological alterations post cardiac transplantation which influence exer-
cise capacity may include surgical denervation of the transplanted heart, de-
crease in chronotropic competence, cardiac function, vascular endothelium,
and pulmonary diffusion.
In the absence of clear evidence to support a ‘one-​size-​fits-​all’ exercise
programme, an individually tailored plan is required. Currently, there is
no evidence to categorically indicate a recommendation for one rehabili-
tation programme over another. Exercise regimens are currently based
on developing the frequency, intensity, timing, and type of exercise and
adopting an integrative approach to building individual programmes, which
incorporate both cardiac and respiratory rehabilitation.
Patient education programme
Successful transplantation requires the recipient’s full cooperation as well as
adequate social support. It is imperative that the organ recipients are made
aware of the risks and what is required of them.
Patient education before transplantation provides the opportunity for re-
cipients to gain information on the following topics:
• Transplant-​specific information.
• Surgical procedure and complications.
• In-​hospital course.
• PGD and rejection.
• Importance of compliance with treatment.
• Implications for employment.
• Outcomes.
Leaflets and patient information sheets should be provided in the pre-​trans-
plantation interval to the patient and family members.
Self-​medication programme
The success of the transplant patient is dependent on adherence to the
self-​medication programme. This is a lifetime commitment. Recipients are
educated in the importance of adherence to a strict regimen. They should
be educated in the role of each medication, the time and method of ad-
ministration of each, maintaining a meticulous medication diary, as well as
any changes in their weight and temperature. Such information may prove
invaluable for the transplant team in subsequent clinic visits.
Infection
Transplant recipients are prone to opportunistic infections which could be
bacterial, viral, or fungal, due to the high levels of immunosuppressive medi-
cations (see Chapters 21 and 30).
Rejection
Rejection is unpredictable in its frequency and severity. The recipient is edu-
cated in the signs of rejection (see Chapters 20 and 28).
23

232 Chapter 19 Bedside care and rehabilitation

Follow-​up clinic
In the first 6 months, frequent visits are scheduled and the intervals are
variable depending on unit protocols. In the interval period between clinic
visits, the organ recipient will have an assigned contact in the transplant unit
whom they can get in touch with for any queries or concerns. Education
continues in follow-​up clinics and recipients are made aware of the import-
ance of continued dose monitoring, identification of complications, infec-
tion, rejection, and prompt intervention.
 Chapter 20 233

Management
of rejection following
cardiac transplantation
Induction of immunosuppression 234
Maintenance immunosuppression 236
Rejection surveillance 240
Cellular rejection 242
Antibody-​mediated rejection 243
Management of sensitized patients 245
Conclusion 246
234

234 Chapter 20 Rejection after heart transplantation

Induction of immunosuppression
Acute rejection and graft failure continue to be common causes of death in
the first 30 days after cardiac transplantation. Induction immunosuppression
regimens are therefore administered immediately after cardiac transplant-
ation to decrease the risk of acute rejection while maintenance immuno-
suppressive regimens are initiated and titrated. The potential benefits of
induction therapy include (1) decreased early rejection, (2) the ability to
delay initiation of nephrotoxic maintenance CNIs, and (3) the opportunity
to decrease corticosteroid dosing. There are potential risks including early
infections and increased risk of malignancy hence the decision to utilize
induction immunosuppression is often individualized, taking into account
individual recipient factors (e.g. baseline renal function, pre-​sensitization,
younger age, female sex, and African American race) that may confer higher
rejection risks.
Induction immunosuppression regimens typically include one of the fol-
lowing: interleukin-​2 receptor antagonist (IL2RA; basiliximab), polyclonal
anti-​thymocyte antibodies (ATG), or historically and least commonly a
murine monoclonal antibody that binds to the T-​cell receptor–​CD3 com-
plex (OKT3) (Table 20.1). These agents, when given, are administered
along with high-​dose corticosteroids.
International Heart and Lung Transplant Registry data suggest more than
half of transplant recipients receive some form of induction immunosup-
pression with approximately 30% of recipients receiving IL2RA and approxi-
mately 20% receiving polyclonal ATG. However, the recommendation for
induction therapy is based on limited prospective randomized controlled
clinical trial data. Data from recent meta-​analyses do not suggest any differ-
ence in all-​cause mortality, infection, or malignancy with the use of induc-
tion immunosuppression with either IL2RA or ATG. The data are mixed in
terms of impact on rejection, with some analyses suggesting a reduction in
acute rejection with induction therapy with either IL2RA or ATG compared
to no induction, and others reporting no difference in moderate or severe
rejection with induction therapy. In non-​randomized studies, there are also
data to suggest that ATG-​based induction therapy may be associated with
less acute rejection compared with IL2RA.
Anti-​thymocyte globulin
ATGs are polyclonal immunoglobulin (Ig)-​G antibodies derived from im-
munization of rabbits (rATG) or horses (eATG) with human thymocytes.
As a result, ATG is comprised of cytotoxic antibodies directed against a
wide array of epitopes found on antigens expressed by human T lympho-
cytes. The antibody–​antigen interactions have been found to affect a spec-
trum of immune processes both in the peripheral blood and lymphoid
tissue. Antibody binding has been implicated in the following: (1) depletion
of T and B cells by both cytolysis and apoptosis, (2) modulation of cell ad-
hesion, (3) interruption of cell signalling necessary for immune activation,
(4) depletion of both immature and mature dendritic cells responsible for
tolerance and T-​cell activation, and (5) expansion of regulatory T cells ne-
cessary for immune tolerance.
ATG is highly immunosuppressive in the timespan of weeks to months
after administration and causes a dose-​dependent depletion of T cells.
 Induction of immunosuppression 235

Table 20.1 Induction immunosuppression

Agent Mechanism of Typical Considerations
action dosage
Basiliximab IL2RA 2D mg Common indications:
IV post abnormal baseline renal
operative function, diabetes, elevated
day 0 and 4 panel-​reactive antibodies
(PRAs), transplant from LVAD,
young female recipients
ATG Polyclonal, 1.5 mg/​ Prophylaxis against infusion
cytotoxic anti-​ kg/​day for reactions is common.
thymocyte 3–​5 days Monitor for leucopenia and
antibodies thrombocytopenia
Alemtuzumab Monoclonal 3D mg IV Can cause serious infusion
antibody once intra reaction, bone marrow
against CD52 operatively suppression, and lymphopenia
receptor
Muromonab Monoclonal 5 mg/​day Can cause fatal cytokine
antibody release syndrome usually
against T-​cell between the first and second
receptor–​ doses
CD3 complex

Given that these antibodies are rabbit or horse derived, they are contra-
indicated in individuals who have allergy or anaphylaxis to rabbit or horse
proteins. Prophylaxis against infusion reactions is commonly administered
including diphenhydramine, acetaminophen, and corticosteroids (which are
given in high doses during induction). The most common adverse effects
include leucopenia and thrombocytopenia which may require dose reduc-
tion by 50% for white blood cell count of 2–​3 × 109/​L or platelets 50–​75 ×
109/​L or discontinuation if white blood cell count <2 × 109/​L or platelets
<50 × 109/​L. Prophylaxis against CMV is necessary in moderate to higher
risk recipients during treatment with IV ganciclovir with a transition to oral
valganciclovir after treatment.
Basiliximab
Basiliximab is a chimeric mouse–​human monoclonal antibody with spe-
cificity to CD25, the alpha chain of the interleukin-​2 receptor (IL2R) on
T cells. Antibody–​antigen binding disrupts the subsequent clonal prolifer-
ation of activated T cells. Basiliximab is administered on postoperative day
0 and postoperative day 4 for induction therapy and saturates the IL2R
for 4–​6 weeks. Basiliximab has not been associated with a cytokine release
syndrome.
Alemtuzumab
Alemtuzumab is a humanized rat monoclonal antibody with specificity to
the CD52 receptor present on not only B and T cells but also other im-
mune cells including neutrophils, natural killer cells, and eosinophils. Binding
236

236 Chapter 20 Rejection after heart transplantation

of alemtuzumab leads to initial activation but subsequent complement and

antibody-​mediated cytolysis of B and T cells. It is used less frequently in
cardiac transplantation and is associated with infusion reactions and bone
marrow suppression.
Muromonab
Muromonab (OKT3) is a murine-​derived monoclonal antibody with speci-
ficity for the CD3 coreceptor on T cells. Antibody–​antigen binding to the
CD3–​T-​cell receptor complex leads to initial activation but subsequent
rapid depletion in CD3+​T cells via apoptosis. The initial activation leads to
a cytokine release syndrome with release of interferon gamma and tumour
necrosis factor which can be life-​threatening. Given the adverse effect pro-
file, OKT3 is rarely used. Similar to ATG, premedication with diphenhydra-
mine and acetaminophen is necessary.

Maintenance immunosuppression
The basis of maintenance immunosuppression regimens consists of three
pharmacological classes of medications including CNIs such as tacrolimus
or ciclosporin, antiproliferative agents such as MMF or azathioprine, and
corticosteroids such as prednisone. In addition, inhibitors of the mTOR can
be used in lieu of or in addition to CNIs when CAV is detected or instead of
CNIs when there is intolerance or toxicity (i.e. renal toxicity).
The goal of triple drug therapy is to target separate biochemical pathways
for immune-​modulation while minimizing dose-​dependent drug toxicities
and oversuppression of the immune system that can increase infectious
and malignancy risks. Since the risk of acute rejection is highest in the first
year after transplantation, maintenance regimens are designed to begin with
higher levels of drug therapy that tapers during the first year with more
frequent surveillance until a low-​dose two-​or three-​drug regimen can be
achieved without significant graft rejection.
Calcineurin inhibitors
The CNIs ciclosporin and tacrolimus form the foundation of immunosup-
pressive regimens in cardiac transplant. Both drugs ultimately inhibit the
transcription of IL-​2 among several other proinflammatory cytokines but
do so with slightly different mechanisms.
Ciclosporin is a cyclic peptide consisting of 11 amino acids derived from
fungus and binds to cyclophilin in the cell cytoplasm; this complex inhibits
calcineurin. Under normal conditions, calcineurin is a phosphatase which
dephosphorylates the transcription factor NF-​AT which allows for tran-
scription of IL-​2 and other related cytokines such as tumour necrosis factor
alpha, granulocyte colony-​stimulating factor, interferon gamma, IL-​4, and
CD40L. Calcineurin inhibition impairs the transcription of these cytokines,
particularly in helper T cells and decreases effector T-​cell activation and
proliferation.
Tacrolimus is a macrolide also derived from fungus which inhibits
calcineurin by binding to a different cytoplasmic protein called FKBP
with similar downstream reduction in transcription of IL-​2 and related
cytokines.
 Maintenance immunosuppression 237

There are several important toxicities to CNIs that require careful moni-
toring and these include hypertension, dyslipidaemia, new-​onset diabetes,
renal insufficiency, hypomagnesaemia, and hyperkalaemia for tacrolimus
versus hypokalaemia for ciclosporin, and neurological side effect such as
seizures, headache, tremors, and posterior reversible encephalopathy syn-
drome (i.e. PRES). Ciclosporin has a few unique adverse effects including
hirsutism and gingival hyperplasia.
Several randomized controlled trials compared tacrolimus to ciclosporin-​
based regimens and, overall, the survival is similar but tacrolimus has a better
side effect profile and lower rejection rates. Given the variable toxicities and
risk for both infection and malignancy with CNIs, serum drug level moni-
toring is imperative. While target trough concentrations are individualized
based on balancing history of rejection, infection, and other toxicities, gen-
eral guidelines are tabulated for reference (Fig. 20.1 and Table 20.2).

• Tacrolimus • Mycophenolate
• Ciclosporin mofetil
• Mycophenolate
Sodium
• Azathioprine

Calcineurin Antiproliferative
inhibitors agents

mTOR Corticosteroids

• Sirolimus • Prednisone
• Everolimus • Methylprednisolone

Fig. 20.1 Maintenance Immunosuppression

Antiproliferative agents
Antiproliferative agents include MMF and azathioprine and are used as ad-
juncts to CNIs and corticosteroids. Both drugs impair nucleic acid synthesis
and thus inhibit proliferation of both B and T cells.
Azathioprine is a prodrug that is hydrolysed within tissues to 6-​
mercaptopurine (6-​MP) and undergoes a series of further biochemical
modifications to become a purine analogue that is incorporated into DNA
and interrupts cell replication. The purine analogue also impairs purine bio-
synthesis in actively replicating B and T cells.
MMF is a newer class of antiproliferative agent that is also a prodrug
which is hydrolysed to mycophenolic acid which inhibits the enzyme inosine
238

238 Chapter 20 Rejection after heart transplantation

monophosphate dehydrogenase involved in the de novo pathway of guanine

nucleotide synthesis. B and T cells rely on the de novo pathway for purine
synthesis and thus are selectively inhibited.
Important toxicities for antiproliferative agents include myelosuppression
and notably leucopenia, and increased risk for both bacterial and viral infec-
tions such as CMV which requires surveillance. MMF can also cause notable
GI side effects with nausea, vomiting, and diarrhoea which often require
either dose adjustment or substitution with mycophenolate sodium which
is an enterically coated delayed-​release formulation.
Azathioprine is associated with more myelosuppression than MMF
but less diarrhoea and can cause pancreatitis and hepatic dysfunction.
Importantly, patients with impaired function or expression of a particular
enzyme, thiopurine methyltransferase (TPMT), are at increased risk for
myelosuppression with azathioprine and testing for TPMT activity is recom-
mended prior to initiating therapy with azathioprine.
Currently, MMF is often preferred over azathioprine as an initial agent given
a more favourable side effect profile with randomized controlled trial data
showing some signal towards decreased mortality and decreased rejection.
Corticosteroids
Corticosteroids are non-​specific immunosuppressive agents that are used
in high doses early after cardiac transplantation with a subsequent taper
over the first year with the goal of maintaining patients on either very low
doses or no corticosteroids at all after the first year. Corticosteroids re-
main a mainstay of treatment of acute cellular and antibody-​mediated rejec-
tion (AMR) in addition to their role as maintenance immunosuppressants.
Long-​term corticosteroid use has many adverse effects including diabetes,
osteoporosis, gastritis, and increased risk of infection. Prophylaxis against
Pneumocystis jirovecii and GI candidiasis is recommended when on higher-​
dose steroids especially during the first year.
mTOR inhibitors
mTOR inhibitors are often used either as adjuncts to standard immuno-
suppressive regimens or as substitutes for either CNIs or antiproliferative
agents such as MMF and azathioprine. These agents (i.e. sirolimus and
everolimus) function by binding to the FK-​BP and cause downstream cell
cycle arrest, leading to decreased T and B cell proliferation. Although mTOR
inhibitors bind to the same cytoplasmic protein as tacrolimus, their mech-
anism of action signalling is independent of the calcineurin pathway. The
benefits of these agents are severalfold: (1) allow for reduction in dose or
substitution of CNIs when there is concern for nephrotoxicity, (2) reduce
the progression of CAV, and (3) reduce malignancy risk associated with
CNIs and antiproliferative agents. The toxicities of mTOR inhibitors include
hyperlipidaemia, myelosuppression, oral ulcers, and proteinuria. Delayed
wound healing is an important adverse effect and thus mTOR inhibitors are
not used immediately after transplantation and careful consideration must
be made prior to other surgical procedures. There are no official guideline
recommendations for therapeutic drug troughs for these agents but often
CNI dosing is reduced when used as an adjunct. Drug levels depend on each
clinical scenario and are usually programme and assay dependent. Typical
drug trough levels for sirolimus and everolimus are listed in Table 20.2.
 Maintenance immunosuppression 239

Table 20.2 Maintenance immunosuppression

Pharmacological Agent Dose Therapeutic
class trough levelsa
CNI Ciclosporin 4–​8 mg/​kg/​day at 275–​375 ng/​mL;
twice a day dosing 0–​6 weeks
200–​350 ng/​mL;
6–​12 weeks
150–​300 ng/​mL;
3–​6 months
150–​250 ng/​mL;
6–​12 months
~200 ng/​mL;
12–​24 months
~150 ng/​mL:
>24 months
Tacrolimus 0.075 mg/​kg/​day at 10–​15 ng/​mL;
twice a day dosing 0–​6 months
8–​12 ng/​mL;
6–​12 months
5–​10 ng/​mL;
>12 months
Anti MMF 2000–​3000 mg/​day Mycophenolic
proliferative at twice a day dosing; acid 2–​5 mcg/​mL
agent maintain white blood
cell count >3.5 ×
109/​L
Azathioprine 2 mg/​kg/​day; —​
maintain white blood
cell count >3.5 ×
109/​L
Mycophenolate 720–​2160 mg/​day at —​
sodium twice a day dosing
mTOR Sirolimus 1–​5 mg/​day 4–​12 ng/​mL
inhibitor
Everolimus 1.5 mg/​day at twice a 3–​8 ng/​mL
day dosing

continued
240

240 Chapter 20 Rejection after heart transplantation

Table 20.2   Continued

Pharmacological Agent Dose Therapeutic
class trough levelsa
Corticosteroid Methyl 1000 mg IV on day 1 —​
prednisolone 125 mg IV every 8
hours on day 2
Prednisone 0.5mg/​kg twice daily —​
and decrease by 5
mg twice daily every
2 days until 10 mg
twice daily
15mg PO four times
daily; month 1–​2
12.5mg PO
four times daily;
month 2–​3
10mg PO four times
daily; month 3–​4
7.5mg PO four times
daily; month 4–​5
5mg PO four times
daily; month 5–​6
2.5mg PO four times
daily; month 6–​8
a
Dosing and therapeutic trough levels intended to be representative and general guidelines; each im-
munosuppressive regimen must be tailored based on balancing rejection history with adverse effects

Rejection surveillance
Down-​titration of triple drug immunosuppression during the first year of
transplantation necessitates frequent monitoring and surveillance for cel-
lular mediated rejection and AMR as acute rejection is inversely related to
time from cardiac transplantation. The most recent International Heart and
Lung Transplant Registry data suggest a 12.6% incidence of treated rejection
within the first year of transplant. Endomyocardial biopsies are the standard
of care with frequent biopsies in the first 6 months following transplant-
ation. While the exact schedule for biopsies is dependent on individual
patient level factors such as previous history of rejection, baseline immuno-
logical response of the recipient, clinical signs and symptoms of rejection,
concurrent infection, or rapid changes in immunosuppression regimens, bi-
opsies at our institutions are performed on the following schedule: a weekly
basis during the first month; every 2 weeks through post-​transplant month
2; monthly through post-​transplant month 6; and finally every other month
until post-​transplant month 12 (Table 20.3).
After the first postoperative year, surveillance with endomyocardial bi-
opsies can be further spaced out to every 4–​6 months in higher-​risk re-
cipients and often is not utilized in those who have not had any significant
rejection history and are otherwise clinically stable. Under these circum-
stances non-​invasive testing with echocardiograms or gene expression
profiling assays (AlloMap (CareDx, Brisbane, CA, USA)) can be utilized as
a surrogate. AlloMap testing may be a useful alternative in patients who
have reached the nadir of their immunosuppression regimen and can help
 Rejection surveillance 241

reduce the number of endomyocardial biopsies. For example, for low-​risk

transplant recipients who are >6 months from transplantation, if AlloMap
scores are ≤34 then surveillance with this method can be continued. If
there is an abnormal AlloMap score >34 then further characterization with
an endomyocardial biopsy may be warranted for a pathological diagnosis.
Among these lower-​risk individuals without significant prior rejection or
graft dysfunction, AlloMap profiling has been associated with a high negative
predictive value for acute cellular rejection (ACR) surveillance. Additionally,
low serial test variability which is measured as the standard deviation of
4 AlloMap scores (≤0.6) has also been shown to have prognostic value
related to a high negative predictive value. These strategies are intended
for surveillance and if there is any concern for rejection by history, physical
examination, or laboratory findings then non-​invasive modalities should not
supplant histopathological determination with biopsy.
CAV manifests as a chronic pan-​ arterial concentric intimal hyper-
plasia in the epicardial arteries and concentric hyperplasia of the media
of the microvasculature of the cardiac allograft. Factors associated with
CAV include immunological factors such as prior cellular mediated rejec-
tion and AMR, presence of donor-​specific anti-​HLA antibodies, and non-​
immunological factors such as CMV infection, diabetes, hypertension,
obesity, smoking, and hypercholesterolaemia. Progression of CAV can lead
to allograft dysfunction from myocardial ischaemia leading to symptoms of
dyspnoea on exertion and chest pain, progressive conduction disease leading
to symptoms of palpitations, presyncope, or syncope. Monitoring and sur-
veillance of CAV with angiography with or without intravascular imaging
such as intravascular ultrasound or stress imaging are important as changes
in immunosuppression, namely introduction of mTOR inhibitors, can curb
progression. CT coronary angiography can be used as well when technically
feasible. Statin therapy has also been shown to reduce CAV independent of
lipid levels. PCI with drug-​eluting stents is recommended for angiographically
amenable vessels. Retransplantation is required in severe cases.

Table 20.3 Rejection surveillance schedule

Months Example biopsy Example CAV Gene expression
schedule screening profiling assays
(AlloMap)
0–​1 Every week —​
1–​2 Every 2 weeks Baseline angiogram
± intravascular
ultrasound
2–​6 Every 4 weeks —​
6–​12 Every 8 weeks Repeat angiogram Begin AlloMap
at ~1 year ± at 6–​10 months
intravascular
ultrasound
12–​24 Every 4–​6 months in Stress imaging Annually
higher-​risk individuals alternating with
angiogram pending
>24 As needed
renal function
24

242 Chapter 20 Rejection after heart transplantation

Cellular rejection
ACR is diagnosed with endomyocardial biopsy. Histological examination
demonstrates inflammatory infiltrate comprising lymphocytes, macro-
phages, and less commonly eosinophils. Neutrophils are not character-
istic of cellular rejection and suggestive of alternative processes such as
healing ischaemic injury, AMR, or infection. The presence of plasma cells
may suggest an alternative process such as healing ischaemic injury, nodular
endocardial infiltrates (i.e. Quilty effect), or lymphoproliferative disorders.
In addition to inflammatory infiltrate, myocyte damage is an additional fea-
ture used to grade severity of rejection. Myocyte damage is histologically
characterized by myocytolysis with encroachment of mononuclear cells at
the periphery of the myocytes resulting in scalloped myocyte borders and
distortion of cellular architecture.
The incidence of ACR has decreased with the development of more
robust immunosuppressive drugs. The risk is inversely proportional to
time, with a higher risk immediately after transplant. The ISHLT revised its
grading system in 2004 to address inconsistencies in classification among
pathologists and variability in treatment related to the previous classifica-
tion scheme.
Treatment options for ACR consist of oral or IV corticosteroids with the
option to add ATG or less commonly OKT3. The intensity of immunosup-
pressive treatment depends on histological severity of the rejection epi-
sode in combination with the degree of haemodynamic compromise. Graft
dysfunction from rejection can be ascertained by invasive haemodynamic
study with a right heart catheterization well as by echocardiography
where wall thickening from myogenic oedema and systolic dysfunction are
characteristic.
Grade 1R typically does not require treatment unless there is haemo-
dynamic compromise, in which case a short pulse of oral or IV steroids
is administered. If there is severe graft dysfunction, cytolytic therapy can
be given but this is rare. Grade 2R rejection without haemodynamic com-
promise is often treated with oral or IV steroids with resumption of the
original steroid taper. Background immunosuppression therapy is adjusted
if serum levels are subtherapeutic. Grade 2R with haemodynamic com-
promise or grade 3R rejection or recurrent lower grade rejection refractory
to steroid treatment is treated with cytolytic therapy, most commonly ATG
on the background of pulse IV steroids. Addition of ATG in severe ACR
requires premedication with steroids, antipyretics, and antihistamine to
reduce the risk of infusion-​related reactions. Antiproliferative agents such
as MMF or azathioprine can be held to reduce the effect of leucopenia.
Antiviral prophylaxis can be escalated with IV ganciclovir during treatment
and then transitioned to oral valganciclovir among patients who are me-
dium to high risk for CMV. Histological evaluation of response to treatment
should be performed with an endomyocardial biopsy within 1–​2 weeks of
treatment for severe symptomatic rejection and can be within 2–​4 weeks
for mild to moderate asymptomatic rejection. Serial echocardiograms are
recommended to assess functional improvement after treatment.
Importantly, severe ACR with haemodynamic compromise is a global
biventricular inflammatory process and, as such, circulatory support with
inotropic or vasopressor therapy is frequently needed while anti-​rejection
 Antibody-mediated rejection 243

treatment is administered to protect end-​organ function. In the most se-

vere cases, MCS may need to be implemented spanning from percutaneous
mechanical circulatory support (e.g. IABP) to full biventricular mechanical
circulatory assistance (e.g. VA-​ECMO) (Table 20.4).

Table 20.4 Endomyocardial biopsy grading criteria

ISHLT standardized cardiac biopsy grading: ACR (revised 2004)

Grade Histological Treatment Dosage range
features considerations
0 No rejection No treatment —​
1R Interstitial and/​ Treat with Prednisone 1–​3 mg/​kg/​
(mild) or perivascular steroids if day (3–​5 days)
infiltrate with haemodynamic Methylprednisolone
up to 1 focus of compromise 250–​1000 mg/​day
myocyte damage (3 days)
2R Two or more Treat with Methylprednisolone
(moderate) foci of infiltrate steroid 1000 mg/​day (3 days)
with associated pulse if no Thymoglobulin 0.75–​
myocyte damage haemodynamic 1.5 mg/​kg/​day for
compromise. 5–​14 days
With Atgam 10 mg/​kg/​day
haemodynamic for 5–​14 days
compromise add
cytolytic therapy ATG-​Fresenius 3 mg/​
with ATG kg/​day for 5–​14 days

3R Diffuse infiltrate High-​dose IV Methylprednisolone

(severe) with multifocal steroids with 1000 mg/​day for 3 days
myocyte damage cytolytic therapy ATG (as above)
± oedema ±
haemorrhage ±
vasculitis

Antibody-​mediated rejection
AMR is thought to occur when recipient antibodies interact with spe-
cific HLA antigens on the endothelial cells of the cardiac allograft leading
to complement activation and tissue injury. The diagnosis is made with
endomyocardial biopsy. Histologically, AMR is characterized by myocardial
capillary injury with endothelial cell swelling, intravascular accumulation of
macrophages, cytokine-​mediated inflammatory changes, and microvascular
thrombosis. Risk factors for AMR include pre-​sensitized recipients such
as those with a history of prior transplantation, multiple blood product
transfusions, female sex, pregnancy, CMV seropositivity, or LVAD implant-
ation. Importantly, pre-​existing and de novo anti-​HLA antibodies have been
associated with cellular rejection, the development of CAV, and decreased
survival.
In addition to standard histological evaluation, immunohistochemical
and immunofluorescent assays intended to detect complement depos-
ition are utilized to make the diagnosis of AMR. C4d, which covalently
24

244 Chapter 20 Rejection after heart transplantation

binds to the endothelium and thus has a longer half-​life than other com-
plement cascade intermediates, can be detected in the capillaries by
immunohistochemistry and immunofluorescence. The presence of C4d
along with histological features is suggestive of AMR. Detection of C3d
in capillaries, which is another complement split product, is also useful
in conjunction with C4d staining and the combination of C4d and C3d
increases the likelihood of AMR. HLA-​DR staining can also be utilized to
assess the integrity of the capillary endothelium which can be disrupted
with AMR-​mediated endothelial damage. Since AMR can be concurrent
with cellular rejection, delineation of the two processes can be aided by
macrophage-​specific staining such as CD68 which is more specific for
AMR. In addition, CD34 and CD31 which are endothelial cell markers
can be stained to understand intravascular localization of macrophages
which occurs in AMR.
Other diagnostic information that can aid in the diagnosis is detec-
tion of circulating donor-​specific antibodies (DSAs) along with clinical
allograft dysfunction. In addition to anti-​HLA antibodies, non-​HLA anti-
bodies may also contribute to endothelial cell injury via apoptosis and
can be implicated in AMR when there is pathological evidence of AMR
in the absence of detectable circulating DSAs. These non-​HLA anti-
bodies bind to autoantigens such as endothelial cytoskeletal proteins,
polymorphic minor antigens, and polymorphic non-​HLA antigens such
as major histocompatibility complex class I chain-​related antigens A and
B (MICA, MICB).
Similar to cellular rejection, AMR (Table 20.5) can be subclinical and
asymptomatic but can also lead to allograft dysfunction and HF symptoms
and in severe cases can lead to cardiogenic shock. Treatment guidelines for
AMR are based on expert consensus and typically involve a combination of
high-​dose IV corticosteroids, removal of circulating antibodies with plasma-
pheresis, inhibition of circulating antibodies with IVIGs, anti-​lymphocyte
antibodies (ATG), B-​cell depleting therapy (rituximab), plasma cell depleting
therapy (bortezomib), complement inhibition (eculizumab), and B-​and T-​
cell depletion (alemtuzumab).

Table 20.5 AMR classification

Category Histological and immunopathological features
pAMR 0 Negative for pathological AMR
pAMR 1 (H+​) Histological findings present and immunopathological findings
negative
pAMR 1 (I+​) Histological findings negative and immunopathological
findings positive
pAMR2 Both histological and immunopathological findings present
pAMR 3 Severe AMR with histological findings of interstitial
haemorrhage, capillary fragmentation, mixed inflammatory
infiltrates, oedema, endothelial cell pyknosis, and/​or
karyorrhexis
 Management of sensitized patients 245

Management of sensitized patients

Sensitization occurs when the transplant recipient has pre-​existing anti-​HLA
antibodies which are associated with increased risk of rejection and CAV.
Moreover, pre-​sensitization has been associated with longer wait times to
transplantation and impacts post-​transplant survival. HLA cell surface pro-
teins are encoded by genes within the major histocompatibility complex and
there are two subgroups of HLA antibodies; namely, class I (A, B, C) and
class II (DR, DQ, DP).
Risk factors for sensitization include prior exposure to foreign HLA
antigens through blood transfusion, prior transplantation, pregnancy, and
implanted tissue allografts. Other risk factors include African American
race, prior viral infection, MCS, and LVAD implantation.
The degree of sensitization is determined by detection of circulating
antibodies using PRA testing. Contemporary methods utilize solid phase
immunoassays where purified or recombinant HLAs are attached to poly-
styrene beads that can be distinguished by differences in internal fluores-
cence when interrogated by flow cytometry or a commercial Luminex
fluoroanalyser. The polystyrene beads contain distinct concentrations of
two fluorescent dyes and can be segregated into 100 different bead popu-
lations with each population conjugated to either a known pool of HLA
antigens or single HLA antigen. Recipient serum is introduced to these
beads and recipient antibody and bead antigen binding is detected with a
reporter dye that binds to the recipient antibodies. The combination of the
individual bead fluorescent signal along with the reporter signal indicates
specific binding which is reported as a mean fluorescent intensity (MFI).
Importantly, MFI is a marker of binding or bead saturation and not of actual
antibody titre. Serum dilution is performed instead to quantitate the highest
dilution at which antibody–​antigen binding is detected. Often antibody–​
antigen binding can become undetectable at low serial dilutions and suggest
lower antibody concentrations independent of the MFI.
The bead composition ranges in terms of specificity. For example, basic
screening for sensitization can be determined with a pooled antigen panel
where beads are coated with either class I or class II proteins. This method
is mostly qualitative and indicates whether there is positive or negative
signal. Phenotype panels include bead populations that have either class I or
class II proteins from 98 individuals and offer more specificity in tracking
changes in the level of class I or class II antibody binding. The most specific
assays utilize single antigen beads where each bead population is coated
with a unique HLA antigen and thus individual antibody–​antigen inter-
actions can be identified. The percentage of positive reactions is quanti-
fied as a percentage PRA and a percentage PRA>10% indicates significant
allosensitization. Determination of unacceptable antigens is determined
using these assays for donor–​recipient matching.
A separate entity known as a calculated PRA (cPRA) is also routinely de-
termined. A cPRA represents the percentage of actual organ donors from
an ethnically weighted database who express more than one of the un-
acceptable HLAs. That is, a high cPRA would correspond to more sensitiza-
tion and less compatibility. A cPRA >50–​80% has been cited as significant
sensitization warranting consideration for desensitization therapies.
246

246 Chapter 20 Rejection after heart transplantation

When sensitization is quantified and significant, desensitization strat-

egies can be used to clear circulating antibodies and thus increase donor
compatibility and improve post-​transplant outcomes associated with pre-​
sensitization. There are limited data evaluating and comparing various
desensitization protocols and thus treatment choices are often based
on programmatic experience. In general, treatment strategies often in-
volve some combination of mechanical removal of antibodies with either
plasmapheresis or immunoadsorption, addition of immunosuppressive
medications to reduce antibody production with rituximab, bortezomib,
alemtuzumab, eculizumab, ATG, cyclophosphamide, and treatment with
IVIG. The optimal regimen, treatment goal, and timing of desensitization is
variable and programme specific.
Cross-​matching is performed to determine recipient antibody to donor
antigen interaction and can be done directly with donor cells in a pro-
spective fashion; this is frequently done when feasible geographically and
when there is concern for recent sensitization in the recipient. Alternatively,
virtual cross-​matching is performed which compares recipient HLA anti-
bodies detected with single antigen bead assays with donor HLA antigens
using prespecified thresholds based on either MFI or serial dilution to deter-
mine which antigens are unacceptable. Physical cross-​match is performed
regardless immediately after transplant. Transplanting across a positive
cross-​match can be done with added risk of rejection and graft dysfunction
but can be considered on a case-​by-​case basis with addition of desensi-
tization therapies. A positive physical cross-​match is generally treated with
induction therapy, plasmapheresis, and IVIG. Addition of more long-​lasting
immunosuppression outside of conventional maintenance therapy including
rituximab, cyclophosphamide, alemtuzumab, and bortezomib can be con-
sidered if there is ongoing rejection, graft dysfunction, or presence of DSAs.
The detection of DSAs and particularly de novo DSAs is associated with
AMR, decreased graft survival, and CAV. While the presence of DSAs is
not synonymous with AMR, it can be an important clue to the immune
environment of the cardiac allograft. Monitoring of DSAs is recommended
at months 1, 3, 6, and 12 post transplantation and more frequently among
sensitized recipients. Furthermore, DSAs are often checked when there are
concerns for graft dysfunction, CAV, or signs of RCM. Whether to treat
detected DSAs is a multifactorial decision with several factors that need to
be considered case by case including whether background immunosuppres-
sion is at optimal levels, associated biopsy pathology, and whether there is
clinical or haemodynamic evidence of graft dysfunction.

Conclusion
Advancements in immunomodulatory agents have been critical in improving
survival among heart transplant recipients. However, there are still formid-
able challenges with the longer-​term adverse effects of immunosuppres-
sion. As such, there is a strong need to have an improved mechanistic
understanding of the alloimmune response that will inform newer thera-
peutics and prolong the longevity of donor organs. With newer biological-​
based therapies and a deeper understanding of allograft tolerance, there
is a compelling opportunity for a more personalized approach towards
immunosuppression.
 Chapter 21 247

Antimicrobial prophylaxis
and treatment of
infection after cardiac
transplantation
Pre-​transplant evaluation 248
Prophylaxis of infections 249
Infection diagnosis and management 250
Immunizations post transplantation 261
248

248 Chapter 21 Infections after heart transplant

Pre-​transplant evaluation
Prior to listing for transplantation, recipients need to undergo a thorough
infectious diseases evaluation. The aim of this is to:
• Identify any latent or active infections that need to be addressed.
• Determine the patient’s immune status against vaccine-​preventable
infections and other potential donor-​derived infections.
• Identify colonization with multidrug-​resistant organisms that require
modification of standard post-​transplant antimicrobial prophylaxis or
infection control procedures or that could preclude transplantation.
As part of the evaluation, a comprehensive lifelong infection history should
be obtained, including review of specific pertinent microbiological records.
Important areas to review include:
• Social history; examples of key risk factors to address are listed in
Table 21.1.
• Current medications including use of immunosuppressive agents that
could predispose to atypical infections earlier than otherwise expected
in the post-​transplant course.
• Complete immunization record.
• Drug allergies including exact reactions and severity.

Table 21.1 Risk factors associated with specific infections

Risk factor Examples of associated infections
Travel or residence in Mycobacterium tuberculosis, Trypanosoma spp.,
areas with known endemic Histoplasma, Coccidioides, Strongyloides, etc.
infections
Travel to areas with recent Zika virus, Ebola virus, West Nile virus, Candida
outbreaks auris, etc.
Close contact with certain Cats (Toxoplasma), birds (Cryptococcus), horses
animals (Rhodococcus), bats (rabies), etc.
Occupational exposures Soil (Nocardia and moulds), water (non-​
tuberculous Mycobacteria, Legionella), etc.
Unpasteurized dairy Listeria monocytogenes, Brucella spp.
products
Well water Cryptosporidium
Time spent in homeless Mycobacterium tuberculosis
shelter or prison
IV drug use HIV, hepatitis B, hepatitis C

Routine testing is performed for certain pathogens pre transplantation (see

Box 21.1); however, unique epidemiological exposures may require add-
itional testing in some cases (e.g. need for Strongyloides or Coccidioides anti-
body screening in patients who have spent time in endemic regions).
 Prophylaxis of infections 249

Box 21.1 Recommended pre-​transplant testing

Serologies
Cytomegalovirus (CMV) antibody
Epstein–​Barr virus (EBV) antibody
Varicella zoster virus (VZV) antibody
Herpes simplex virus (HSV) antibody
Hepatitis B (HB) virus: HB surface antigen and antibody, HB core
antibody
Hepatitis C virus (HCV) antibody
HIV antibody/​antigen
Mumps, measles, rubella antibodies
Toxoplasma gondii antibody
Treponema pallidum screening (rapid plasma regain (RPR) or fluores-
cent treponema antibody absorption (FTA-​ABS))
Other tests
Tuberculin skin test (purified protein derivative (PPD)) or interferon
gamma release assay for Mycobacterium tuberculosis
Tests individualized per patient’s unique exposures (e.g. Coccidioides
antibody, Strongyloides antibody, etc.)

Prophylaxis of infections
Various strategies have been implemented to aid in prevention of infection
after transplantation. Based on timing relative to the transplant procedure,
these include:
• Pre transplantation: update vaccines (see Table 21.5, p. 262).
• Peri transplantation: administer antimicrobials at the time of surgery
to prevent nosocomial infections (see following discussion on peri-​
transplant prophylaxis).
• Post transplantation:
• Prescribe ‘universal prophylaxis’—​administer an antimicrobial agent
to all recipients for a defined period of time post transplantation to
prevent infection.
• ‘Pre-​emptively monitor’ for infection—​serially screen the recipient
with specific tests at defined intervals to monitor for early
reactivation or acquisition of an infection.
Table 21.2 provides some commonly used preventative regimens in heart
transplant recipients.
Peri-​transplant antimicrobial prophylaxis
Surgical site infections, which occur in 4–​19% of heart transplant recipi-
ents, are associated with significant morbidity and mortality. Risk factors
for surgical site infections include a BMI >30 kg/​m2, prior cardiac surgery,
LVAD as a BTT, prolonged mechanical ventilation, and use of sirolimus in
the immunosuppressive regimen. Bacterial infections tend to be the most
common, though Candida species and other fungi can be encountered as
250

250 Chapter 21 Infections after heart transplant

well. Antimicrobial prophylaxis is recommended at the time of surgery to de-

crease the frequency of surgical site infections. Because there are no formal
guidelines on choice of drug or duration, transplant centres formulate their
own protocols. Antibiotic regimens should have good anti-​staphylococcal
activity and be individualized based on patient-​specific factors such as
colonization with multidrug resistant organisms (e.g. methicillin-​resistant
Staphylococcus aureus (MRSA)). Additionally, if the patient has a history of
an infected LVAD or other cardiac hardware, susceptibility data from those
infections should be used to guide antibiotic selection. Modification to the
antibiotic regimen may also be required based on positive donor cultures.
Routine antifungal prophylaxis in the heart transplant population is contro-
versial and not routinely recommended unless the recipient has multiple
and specific risks for invasive candidiasis (e.g. acute kidney injury, prolonged
central venous catheter, etc.) or there is an outbreak of mould infections
within the institution.

Table 21.2 Post-​transplant prophylactic regimens

Pathogen Prophylactic regimen
Pneumocystis jirovecii TMP–​SMX SS daily or TMP–​SMX DS 3×
weekly or daily for 6–​12 monthsa
Toxoplasmosis gondii TMP–​SMX DS 3× weekly for 6–​12 months
CMV D+​/​R–​ serostatus (high Valganciclovir or IV ganciclovir for 3–​6
risk) or months
CMV R +​and anti-​
lymphocyte antibody
therapy
CMV R+​serostatus Valganciclovir or pre-​emptive monitoring for
(intermediate risk) 3–​6 months
CMV D–​/​R–​ serostatus Valaciclovir or aciclovir for 3 months (for
(low risk) prevention of HSV/​VZV)
EBV Pre-​emptive monitoring and reduction of
immunosuppression if detected (antivirals not
effective)b
D, donor; DS, double strength; R, recipient; SS, single strength; TMP–​SMX,
trimethoprim–​sulfamethoxazole.
a
Second-​line agents include dapsone, atovaquone, and aerosolized pentamidine.
b
Reduction in immunosuppressive needs to be weighed against risk of rejection. There is no con-
sensus as to monitoring algorithms or viral load threshold for intervention. There is variability be-
tween quantitative NAATs and controversy regarding best sample type (whole blood vs plasma)
for testing; the same NAAT and sample type should be used for the individual patient.

Infection diagnosis and management

A single-​centre study of 600 OHT recipients over a 16-​year time period re-
ported infectious episodes as 44% bacterial, 42% viral, 10% fungal excluding
Pneumocystis jirovecii, 4% P. jirovecii, and 0.6% parasitic. The timing and type
of infection after transplantation is influenced by prophylactic regimens,
enhanced immunosuppression, and environmental exposures. Because of
 Infection diagnosis and management 251

their immunosuppressed status and altered anatomy, transplant recipients

may not mount a robust inflammatory response and may have minimal
or atypical signs and symptoms of infection. Accordingly, a high index of
suspicion for infection has to be maintained. Concurrently, non-​infectious
syndromes such as rejection and drug reactions (e.g. sirolimus-​induced
pneumonitis, dapsone-​induced methemoglobinemia/​hypoxia, etc.) need to
be kept in mind in the heart transplant recipient presenting with signs and
symptoms concerning for infection. The diagnostic work-​up for infection
has to be individualized in accordance with the patient’s presentation and
history. Cultures relative to the infected body site should be obtained to aid
in reaching a specific diagnosis. Conventional radiographs frequently lack
sensitivity and the patient may need a CT scan or an MRI scan to delineate
the presence and characteristics of specific radiological findings.
Donor-​derived infections
Donor-​derived infections are characterized by UNOS as expected or unex-
pected. Expected transmissions are those in which the donor was known
to be infected with the pathogen prior to donation and for which routine
prophylaxis or monitoring is employed in the recipient post transplantation
(e.g. CMV, EBV, etc.). Unexpected transmissions are those wherein a donor
was not known to harbour a pathogen but transmitted it to a recipient.
Unexpected donor-​derived infections, while rare, are often associated with
high mortality as their diagnosis is frequently delayed in the recipient and
treatment options may be limited. Clearly, maintaining a high index of sus-
picion for donor-​derived infections is necessary. Most donor-​derived in-
fections present in the recipient within the first 3 months after transplant.
Thus, if a recipient presents with unusual or unexplained symptoms within
the first 3 months of transplant, a donor-​derived infection should be con-
sidered and epidemiological clues for potential donor exposures should
be investigated. In the US, when donor-​derived infection is suspected, the
Potential Donor-​Derived Disease Transmission Event (PDTE) should be re-
ported using the Secure Enterprise UNOS Improving Patient Safety portal
(instructions available at: https://​unos.org/​wp-​cont​ent/​uplo​ads/​unos/​
Gui​danc​e_​DT​AC_​P​DDTE​_​06-​2011.pdf ).
HBV/​ HCV/​HIV
US Public Health Service (PHS) guidelines outline criteria that could indicate
if a recipient would be at risk for acquiring HBV, HCV, and HIV. As part of
donor screening, hepatitis B (HB) surface antigen, HB core antibody, HIV
antigen and antibody, HCV antibody, and NAAT testing for HBV, HCV and
HIV are performed. NAAT testing supplements antibody tests which can
be falsely negative as a consequence of haemodilution/​large volume resus-
citation of the donor and during the ‘window’ period after acute infection
if adequate viral replication has not taken place by the time the test is per-
formed. While the window periods of new-​generation NAAT testing are
very small (<5 days for HIV and HCV, ~20 days for HBV), it is still possible
for disease transmission to occur. Therefore, repeat testing in all recipients
is recommended within 4-8 weeks of transplantation.
The advent of safe and effective HCV therapy in recent years has allowed
for use of HCV-​positive organs in HCV-​negative recipients. Preliminary
data on heart transplantation from HCV viraemic donors look promising
25

252 Chapter 21 Infections after heart transplant

though long-​term outcomes are not yet available. These recipients are
typically treated with direct-​acting HCV therapy post transplantation and
require close monitoring of therapy by either a hepatologist or infectious
disease physician.
Unlike HCV, current therapeutic options for HBV only suppress the
virus and are not curative. HBV serologies have to be carefully inter-
preted to determine patient status (Table 21.3). In the US, organs from
donors who are HBV surface antigen positive (i.e. have active infection)
are not considered for transplantation apart from urgent circumstances.
However, in countries where HBV is highly endemic, organs from donors
with active infection may also be accepted for transplantation, with plans
for treatment with HB immunoglobulin and antivirals post transplantation.
HB core antibody positive donors can be accepted for transplantation as
cases of transmission have been rare and outcomes have been similar
to recipients who received a HBV-​negative organ. Based on individual
recipient factors, some patients may require revaccination and antiviral
therapy as prophylaxis, therefore it is important to involve your local in-
fectious diseases physician or hepatologist in the management of these
patients.
Other donor-​derived pathogens

Table 21.3 Interpretation of hepatitis B serology

Interpretation HBV surface HBV surface HBV core
antigen antibody antibody
Active infection +​ –​ +​
Immunity due to natural –​ +​ +​
infection
Immunity due to vaccination –​ +​ –​
Possible interpretations: –​ –​ +​
chronic infection, false-​
positive test, resolving
acute infection

There is a wide range of other pathogens which can be transmitted via

donor organs. For example, rabies, lymphocytic choriomeningitis virus,
West Nile virus, Mycobacterium tuberculosis, Cryptococcus, Coccidioides
immitis, Aspergillus, and free-​living amoebae (Balamuthia, Acanthamoeba)
have all been transmitted from donors with meningitis or encephalitis of
unknown cause. Numerous cases of unexpected tuberculosis transmission
from donors to recipients have been reported including via heart, liver,
kidney, and lung allografts. Table 21.4 provides a partial list of some of
the more common ‘unexpected’ potential donor-​derived pathogens and
the most common clinical presentation in the recipient. Bottom line, if
donor-​derived infection is a concern, infectious diseases should be imme-
diately consulted to aid in the diagnostic work-​up and management of the
recipient.
 Infection diagnosis and management 253

Table 21.4 Possible donor-​derived infections

Pathogen Donor exposure risks Recipient presentation
Lymphocytic Rodent exposure, including Encephalitis
choriomeningitis pets (hamsters, etc.)
virus
Rabies Unreported bat bites Encephalitis
Toxoplasmosis Travel to endemic area Diffuse pneumonia
(Latin America, parts of myocarditis
Europe); ingestion of cat Retinitis
faeces or undercooked Encephalitis
meat
West Nile virus Mosquito exposure Meningitis
Encephalitis
Poliomyelitis-​like flaccid
paralysis
Chagas’ disease Travel to endemic area Fever
(Trypanosoma cruzi) (Latin America) Myocarditis
Acanthamoeba Warm, fresh water Skin lesion
exposure (drowning/​near Encephalitis
drowning in a lake/​pond,
tap water nasal irrigation)
Strongyloides Travel to endemic area GI complaints;
(Southeast Asia, Central hyperinfection
and South America, Africa) syndrome (polymicrobial
bloodstream infection,
pneumonitis, meningitis)
Malaria Travel to endemic area Fever
(Southeast Asia, Central
and South America, Africa)

Sepsis
Transplant recipients are at a higher risk for developing sepsis than the gen-
eral population and are more likely to have an atypical pathogen. Specific
studies looking at sepsis mortality in heart transplant recipients are limited;
one retrospective review of a national US database showed the mortality
rate to be 15.7%. Empiric broad-​spectrum antibiotics should be started
promptly, as delay in initiation is associated with increased mortality. The
antibiotic regimen should be tailored according to the suspected site of
infection, susceptibility of prior infections, and local antimicrobial resistance
patterns. It should typically include an antibiotic with MRSA coverage such as
vancomycin or linezolid and a Gram-​negative agent with anti-​pseudomonal
activity such as piperacillin–​tazobactam or a carbapenem. An echinocandin
for Candida coverage may be considered, depending on the risk. In cases
where there is concern for a mould infection such as Aspergillus, therapy
with a mould active azole may be required. We recommend prompt con-
sultation with your local infectious diseases physician to help in the manage-
ment of these complex cases.
254

254 Chapter 21 Infections after heart transplant

Other specific pathogens

Listeria
L. monocytogenes is a Gram-​positive rod usually acquired by eating food con-
taminated with the bacterium, most commonly unpasteurized dairy prod-
ucts, deli meats, smoked salmon, and produce (recent outbreaks have been
traced to celery, sprouts, and cantaloupe). Solid organ transplant (SOT)
recipients, owing to their suppressed cell-​mediated immunity, can present
with febrile gastroenteritis associated with sepsis and meningoencephalitis.
In one case–​control study, the median duration of symptoms prior to hos-
pitalization in 30 SOT recipients with listeriosis was 3 days; at presentation,
87% of patients had bacteraemia, 33% had meningoencephalitis, and 23%
had septic shock. Mortality at 30 days was high (27%). Blood and cerebro-
spinal fluid (CSF) should be submitted for culture; stool can also be cultured
but requires special media for recovery. Thus, communication directly with
the microbiology laboratory is recommended before sending stool for cul-
ture. Listeria should be considered when ‘diphtheroids’ are reported to be
growing from blood or CSF cultures.
High-​dose ampicillin (2 g IV every 4 hours renally adjusted for adults) is
the drug of choice for listeriosis; trimethoprim–​sulfamethoxazole (TMP–​
SMX) may be used for patients allergic to penicillin. Duration of therapy
depends on site of infection, but typically 3–​6 weeks in SOT recipients with
bacteraemia and longer in those with CNS disease. Of particular note,
Listeria is intrinsically resistant to cephalosporins and these agents should
not be used for empiric treatment for Listeria. TMP–​SMX used as prophy-
laxis after SOT has protective activity against Listeria.
Nocardia
At least 33 Nocardia species can cause disease in humans. Nocardiosis is
most often acquired via inhalation of spores, and classically presents as a
respiratory infection characterized by pulmonary nodules often mistaken
radiographically for invasive pulmonary aspergillosis. It also frequently dis-
seminates and based on its neurotropism, recovery of Nocardia from any
body site should prompt imaging of the CNS to rule out an asymptom-
atic brain abscess. In one retrospective review, the median time to diag-
nosis among 37 SOT recipients was 370 days post transplantation. Blood,
CSF, and tissue biopsies may be submitted for routine bacterial and fungal
culture; Nocardia is frequently recovered from fungal culture as the media
used supports its growth and fungal cultures are typically held longer than
bacterial cultures. Because it grows more slowly than other bacteria, the
microbiology laboratory should be notified when this pathogen is sus-
pected. Nocardia appear as delicate, filamentous, branching Gram-​positive
rods that can be confirmed as Nocardia based on partial acid fastness on
modified acid-​fast stains.
High-​dose TMP–​SMX is considered the drug of choice for Nocardia;
in cases of intolerance, treatment should be based on susceptibility
results and site of infection. The duration of therapy ranges from 3
months up to 1 year or longer depending on disease severity, location,
and response. Of note, while TMP–​SMX is the treatment of choice for
Nocardia, doses used for Pneumocystis prophylaxis are not protective
against Nocardia.
 Infection diagnosis and management 255

Clostridium difficile-​associated diarrhoea (CDAD)

C. difficile is an anaerobic Gram-​positive rod that can be part of the normal
stool flora in up to 5% of healthy adults and up to 30% of healthy neonates.
Colonization with C. difficile alone does not result in CDAD. Only strains
that are able to produce C. difficile enterotoxin (TcdA) and/​or cytotoxin
(TcdB) are capable of causing CDAD. In addition to being colonized with a
TcdA/​TcdB-​producing C. difficile strain, other risks (e.g. antibiotic exposure,
proton pump inhibitor use, etc.) combine to influence toxin production re-
sulting in CDAD. C. difficile is highly transmissible via the faecal–​oral route
which is facilitated by the production of spores that can persist on envir-
onmental surfaces for months and which are resistant to commonly used
cleaning agents and alcohol-​based hand gels (hence the need for washing
hands with soap and water after contact with infected patients). Diagnosis
of CDAD should be pursued only in patients with acute diarrhoea (≥3
loose stools in 24 hours) and with no other obvious explanation for the
diarrhoea. Diagnosis typically includes testing a liquid stool sample with an
enzyme immunoassay (EIA) for C. difficile toxins A and B. Use of the newer
NAAT, which only detects the presence of toxin-​producing C. difficile but
not the actual toxin itself, has been associated with false-​positive tests.
Heart transplant recipients have a CDAD incidence ranging from 1% to
8%. Patients with CDAD should be assessed for disease severity. Criteria
proposed for severe CDAD include white blood cell count >15,000 cells/​
mL or serum creatinine ≥1.5 mg/​dL. Complications may include pseudo-
membranous colitis, toxic megacolon, perforation of the colon, and sepsis.
Oral vancomycin or oral fidaxomicin are typically first-​line therapy for
CDAD. Fidaxomicin is more expensive than vancomycin but is associated
with a lower recurrence rate. Alternatively, oral metronidazole, may be
used for non-​severe CDAD. Treatment duration is typically 10 days. Test
of cure is not indicated as tests may remain positive during or after clinical
recovery.
CMV
The most common presentation of CMV is CMV viral syndrome, a
mononucleosis-​like syndrome which consists of the presence of CMV in
blood and fever and one or more of the following: malaise, leucopenia,
atypical lymphocytes, thrombocytopenia, or elevated hepatic enzymes.
CMV may also cause tissue-​invasive disease with end-​organ damage, with
predilection to involve the allograft and presenting as myocarditis in heart
transplant recipients. As CMV is a known immunomodulatory virus, it also
plays an indirect role in driving other post-​transplant infections, such as in-
vasive fungal infections, P. jirovecii pneumonia, EBV-​related PTLD, as well as
allograft rejection. The risk for CMV disease is based on the SOT recipient’s
immune status prior to transplantation and the type of organ transplanted.
Heart is considered an intermediate-​risk organ. Quantitative NAAT is the
method of choice for rapid diagnosis of CMV in blood.
IV ganciclovir is recommended for severe or life-​threatening disease.
Antiviral treatment should be administered for 2–​3 weeks, and continued
until viremia clears or reaches a negative threshold, and there is resolution
of clinical symptoms. Oral valganciclovir may be used in mild to mod-
erate disease. Drug resistance should be suspected if there is prolonged
ganciclovir exposure (>6 weeks) and no reduction in viral load or lack of
256

256 Chapter 21 Infections after heart transplant

clinical improvement after 14 days of treatment in which case genotypic

testing for specific resistance mutations (UL97 and UL54) should be per-
formed. Based on the specific mutation present, therapeutic options include
high-​dose IV ganciclovir, foscarnet, maribavir or cidofovir.
EBV
EBV, the aetiological agent of mononucleosis, is also involved in the un-
controlled proliferation of EBV-​infected lymphocytes resulting in PTLD.
Clinical manifestations vary from a non-​specific febrile mononucleosis-​like
illness with lymphadenopathy to solid tumours. Being an EBV seronegative
recipient is a well-​known risk for PTLD which is a significant complication
after paediatric heart transplantation in particular. In a retrospective co-
hort, 80% of heart transplant recipients presented with late-​onset PTLD
(>1 year post transplantation) and the GI tract, with its high content of
lymphoid tissue was the site most frequently involved, followed by hilar
lymph nodes, then lung, cervical, axillary, para-​aortic, and inguinal lymph
nodes, and finally skin. Involvement of the GI tract may present as abdom-
inal discomfort, nausea, or diarrhoea and requires a high index of suspicion.
Overall 5-​year mortality was 29%, mainly related to disease progression or
treatment-​related mortality. Tissue biopsy and histopathology is the gold
standard for the diagnosis of PTLD.
The mainstay of management is reduction in immunosuppression.
Treatment regimens that include rituximab, an anti-​ CD20 monoclonal
antibody, are also used. Though antivirals have been used in this setting,
they lack activity on latently infected lymphocytes. Factors which may af-
fect overall survival after PTLD include bone marrow involvement and
hypoalbuminemia, both poor prognostic factors which suggest advanced
disease at presentation. Complete response to therapy is associated with
improved survival.
VZV
VZV is the aetiological agent for varicella, or chicken pox, and for herpes
zoster (HZ). Most primary infections occur in childhood, though immunity
may also be acquired by vaccination with a live attenuated VZV vaccine.
After infection, VZV establishes lifelong latency. Primary varicella infection
in SOT may present with fever, constitutional symptoms, and a vesicular
disseminated rash. SOT patients are at increased risk for complications,
such as pneumonia. The incidence of HZ reactivation, or shingles, in
heart transplant recipient ranges from 5.5% to 16.8%. While most cases
of HZ occur in the first year post transplantation, in a multicentre cohort
of mixed SOTs, HZ occurred a median of 2.6 years after transplantation.
Complications such as post-​herpetic neuralgia was reported in 26–​45% of
patients and HZ ophthalmicus in 14%. Risk factors for HZ include MMF
use, being African American, and older age at time of transplant. Severe,
disseminated or CNS infection with VZV may require additional testing.
Laboratory tests include VZV NAATs, direct fluorescent antibody assay,
and viral culture, though the latter is used principally for epidemiological
purposes.
SOT patients who develop primary varicella should be treated with
IV aciclovir. For limited dermatomal HZ, oral agents such as aciclovir,
valaciclovir, or famciclovir, may be used. Severe HZ necessitates IV aciclovir,
and, for HZ ophthalmicus, an ophthalmology consult.
 Infection diagnosis and management 257

Influenza
Influenza virus usually causes a self-​limited illness characterized by fever,
rhinorrhoea, and cough. In SOT recipients, however, fever may be absent,
though upper respiratory symptoms, with or without myalgia, headache,
and dyspnoea during influenza season are highly suspicious. In this patient
population, approximately 22% of patients present with pneumonia, and
may have co-​infection with other viruses or bacteria. Influenza infection
may occur at any time post transplantation and early infections, <3 months
post transplantation, may be associated with severe illness. Risk factors for
severe influenza in SOT recipients include age, use of MMF, and early infec-
tion (<3 months) after transplantation. NAAT is the preferred diagnostic
method. Since it is not possible to clinically differentiate disease caused by
respiratory viruses, multiplex PCR assays that test for influenza and other
respiratory viruses are recommended. Preferred clinical specimens in early
illness include nasopharyngeal swab, wash, or aspirate. If negative or if
diagnostic uncertainty, or if symptoms and signs of lower tract disease are
present, a lower respiratory tract specimen obtained via bronchoalveolar
lavage (BAL) may have higher diagnostic yield.
Neuraminidase inhibitors, oseltamivir and zanamivir, are associated with
improved outcomes in SOT. Early initiation of antiviral therapy is associated
with lower risk of progression to pneumonia, as is influenza vaccination in
the same season.
Other respiratory viruses
Other respiratory viral pathogens causing infection in SOT recipients in-
clude parainfluenza (PIV), respiratory syncytial virus (RSV), human
metapneumovirus (hMPV), rhinovirus (hRV), severe acute respiratory
syndrome coronavirus 2 (SARS-CoV-2) and adenovirus (ADV). Most
are community acquired and cause upper respiratory tract infections. In
SOT, however, these may be nosocomially acquired and have increased
risk of progression to lower respiratory tract infections (pneumonia) as
well as prolonged duration of illness and viral shedding. Some viruses
have year-​round prevalence while others circulate with a seasonal pat-
tern. Transmission of the different respiratory viruses may be via direct
or indirect contact through droplet or aerosol. For ADV, there is also the
faecal–​oral route. In hospital settings, appropriate infection control meas-
ures including consistent hand washing, standard and contact precautions,
droplet precautions, and, depending on the virus, airborne precautions
should be instituted. Molecular diagnostic tools, such as multiplex PCR is
the diagnostic method of choice. In cases with tissue invasive disease, histo-
pathology with immunohistochemistry is used to evaluate for select viruses
and to distinguish between infection and rejection.
Supportive treatment is the mainstay of therapy, with reduction in im-
munosuppression when possible. Antiviral agents with effect against
some of these viruses include ribavirin, cidofovir, remdesivir and possibly
palivizumab. IVIG in conjunction with antiviral therapy is occasionally used
in select patients. Management should be individualized with consultation by
an infectious disease specialist.
Pneumocystis jirovecii
P. jirovecii is an opportunistic fungal pathogen that causes pneumonia in
SOT not receiving effective prophylaxis. Primary infection is acquired via
258

258 Chapter 21 Infections after heart transplant

the respiratory route usually in childhood, which then may enter a latent
phase. Reactivation or de novo exposure can lead to P. jirovecii pneumonia
(PJP), presenting with fever, non-​productive cough, and dyspnoea with hyp-
oxaemia. Increased risk for PJP is seen with lack of effective prophylaxis,
increased immunosuppression (corticosteroids or antilymphocyte therapy),
CMV reactivation, allograft rejection, older age, and low absolute lympho-
cyte count post transplantation. Nosocomial infections via interhuman
transmission in outbreak situations have been reported in heart transplant
recipients. Without prophylaxis, the risk of PJP is high in the first 6 months
following SOT and during periods of increased immunosuppression. The
use of prophylaxis may lead to delayed-​onset PJP. Diagnosis of PJP is made
through a combination of clinical, radiological, and laboratory features.
The preferred specimens for diagnosis include induced sputum, BAL, or
biopsy. Immunofluorescence testing of specimens, or validated NAATs will
detect P. jirovecii. Calcofluor stains of respiratory specimens, and lung tissue
histopathology stained with polychrome or silver, will also demonstrate the
organisms. Serum testing of lactate dehydrogenase and (1l3) β-​D-​glucan
may be useful adjuncts. Abnormal CXR is present in most patients, with
patterns ranging from bilateral interstitial infiltrates to alveolar infiltrates,
nodules, or consolidation.
TMP–​SMX is the drug of choice for treatment of PJP. Clindamycin with
primaquine is a second line regimen. IV pentamidine use is limited by numerous
adverse effects. Atovaquone at higher doses has also been used. Adjunctive
steroids are controversial but may have a role in severe hypoxaemia.
Aspergillus
Aspergillus is the most frequent cause of invasive mould infection following
heart transplant and A. fumigatus is the most common species causing
invasive aspergillosis (IA). Incidence of IA in heart transplant recipients
ranges between 3.5% and 7.1%. Though the time of IA onset varies in
heart transplant recipients, most occur in the first year post transplant-
ation. Risk factors for IA in heart transplant recipients include Aspergillus
colonization, reoperation, CMV disease, post-​transplant haemodialysis,
and an episode of IA in the heart transplant programme within 2 months
prior to or post transplantation. Older age has been associated with late
IA. Infection is acquired most commonly by the inhalation of Aspergillus
spores and accordingly, the site most frequently involved is the lower re-
spiratory tract, followed by dissemination, with or without CNS infection.
Mediastinitis and surgical site infection have also been reported. Mortality
due to IA in a Swiss cohort was 26.7% for heart transplant recipients and
in a Spanish cohort, IA mortality ranged from 26% for early IA to 63% for
late IA. Clinical signs and symptoms, together with laboratory results and
radiographic findings on CXR or chest CT, are used to make the diagnosis
of IA. Isolation of A. fumigatus from a respiratory site in a heart trans-
plant recipient has a high positive predictive value and is highly suggestive
of IA. BAL specimens should be sent for culture and a galactomannan
test and tissue should be sent for histopathology and culture. Culture al-
lows for identification of the Aspergillus species and susceptibility testing.
Radiographic findings suggestive of IA include nodules or cavitation,
though more than half may present with atypical finding such as ground-​
glass opacities or pleural effusion.
 Infection diagnosis and management 259

Voriconazole is the drug of choice for the treatment of IA. Isavuconazole

or lipid amphotericin are considered alternative agents, and posaconazole
has been used in refractory IA. Some centres add an echinocandin as part
of salvage therapy.
Non-​Aspergillus fungal infections
Infections due to infrequently encountered fungi have increasingly been
reported in immunocompromised patients. Some invasive moulds include
non-​Aspergillus hyalohyphomycetes, phaeohyphomycetes, or dematiaceous
(black) moulds, and Mucorales (also referred to as Zygomycetes). The most
commonly reported moulds in each category are as follows:
• Hyalohyphomycetes: Scedosporium, Lomentospora, and Fusarium.
• Phaeohyphomycetes: Alternaria, Bipolaris, Exophiala, Verruconis, and
Cladophialophora.
• Mucormycetes: Rhizopus, Mucor, and Cunninghamella.
Portals of entry include the skin or the respiratory tract. If infection ensues,
clinical signs and symptoms may be indistinguishable from those caused by
Aspergillus or Nocardia. Clinical manifestations include cutaneous or sub-
cutaneous infection, pulmonary disease, and disseminated infection, with or
without CNS involvement. Other specific locations may be associated with
underlying medical conditions, secondary to focal trauma, or be seeded
during dissemination. The diagnostic approach includes a comprehensive
history (with exposure if known), clinical findings, specimens for culture and
histopathology, and radiographic studies. Response to treatment is predi-
cated not only on the type of fungal pathogen and the minimum inhibitory
concentration to the antifungal agent being considered, but also site of in-
fection, underlying host factors, degree of immunosuppression, therapeutic
drugs levels, and need for adjunctive surgical excision or debridement.
Expert consultation with infectious disease physicians should be obtained
when managing these infections.
Cryptococcus
Cryptococcus is an encapsulated yeast and the third most common inva-
sive fungal infection in SOT after invasive candidiasis and IA. Acquisition
may occur through inhalation or through the skin. Cryptococcosis is usu-
ally a subacute, late occurring infection in SOT due to reactivation of la-
tent or quiescent infection. Very early-​onset cryptococcosis may be due
to early reactivation, unrecognized cryptococcosis pre transplantation, or
in some cases, donor derived. The most common clinical presentation is
pulmonary disease, with cough, dyspnoea, and sputum production. Fever
may or may not be present. Extrapulmonary disease may include CNS dis-
ease/​meningitis, skin/​soft tissue infection, or fungemia. Clinical manifest-
ations of cryptococcal meningitis may be subtle and include headache, gait
disorder, hearing loss, and altered mental status. Necrotizing fasciitis and
infection of the prostate have been reported in heart transplant recipients.
It is important to determine site and extent of infection. Blood, sputum
or BAL, and tissues should be sent for culture and tissues for histopath-
ology (Cryptococcus appears as round yeasts with narrow budding and the
mucicarmine stain reveals the organism’s capsule). Serum cryptococcal
antigen should also be submitted. In patients with a positive serum crypto-
coccal antigen or diagnosed with pulmonary disease, a lumbar puncture
260

260 Chapter 21 Infections after heart transplant

should be performed to evaluate for cryptococcal meningitis (send CSF for

fungal culture and cryptococcal antigen); documenting the opening pressure
is critical as this guides therapeutic taps as outlined below. Radiographic
findings include pulmonary nodules which may be solitary or multiple and
indistinguishable from other infections. Occasionally, findings include diffuse
interstitial infiltrates or effusion. Some centres also conduct CNS imaging
with CT prior to lumbar puncture.
For CNS disease, disseminated disease, or moderate to severe pul-
monary disease, the primary induction regimen consists of a lipid formula-
tion of amphotericin B plus 5-​fluocytosine for ≥14 days or until CNS culture
sterilization, followed by long-​term fluconazole therapy for the maintenance
phase. If the patient has elevated intracranial pressure (CSF opening pres-
sure >25 cm), CSF drainage by lumbar puncture to reduce opening pres-
sure by 50% or to 20 cm of CSF is recommended. Patients may require
serial lumbar punctures or placement of a lumbar percutaneous drain for
pressure control. For individuals with mild to moderate disease limited to
the lung, azole therapy may be used as the primary regimen.
Toxoplasma
Toxoplasma gondii is a protozoan with seroprevalence varying by geographic
region. In the US, the most recent seroprevalence was 12.4% in those ≥6
years, and 9.1% in women of child-​bearing age. Information regarding the
frequency of toxoplasmosis in SOTs is scarce; one multicentre study re-
ported the overall frequency of toxoplasmosis to be 0.14%, and highest
in heart transplant recipients at 0.6%. SOT recipients may acquire primary
toxoplasmosis from ingesting food or water contaminated with oocysts
from faces of infected felines or by eating undercooked meat containing
cysts. SOT recipients are also at risk for endogenous reactivation of latent
disease, and for acquisition from a cyst-​containing donor organ. SOTs who
have negative serostatus prior to transplantation, especially if discordant
to the donor status (D+​/​R–​), are at highest risk for primary infection in
the first 6 months post transplantation. Reactivation toxoplasmosis in R+​
recipients may be asymptomatic. However, in those who were seronega-
tive prior to transplant and acquire primary infection, clinical manifestation
may be severe, including pulmonary toxoplasmosis, cerebral toxoplasmosis,
and disseminated disease. Fever is present in approximately two-​thirds of
patients. Lung infections present with cough and dyspnoea, with diffuse
infiltrates on imaging. Myocarditis can present as HF resembling allograft
rejection. CNS infections, including chorioretinitis, meningitis, encephalitis,
brain abscess, and disseminated infection may present with headache, con-
fusion, focal neurological signs, and visual abnormalities. Since infection may
be asymptomatic or initially present with non-​specific signs and symptoms,
diagnosis depends on laboratory tests. PCR is the method of choice, es-
pecially in patients with discordant serostatus and those unable to tolerate
TMP–​SMX prophylaxis. PCR can be performed on BAL, blood, CSF, and
other body fluids. Histopathology may also reveal the pathogen.
Based on studies in HIV patients, the preferred regimen for induction
therapy is pyrimethamine plus sulfadiazine plus leucovorin, for at least
6 weeks, followed by lifelong chronic suppressive therapy with the same
agents at a reduced dose. Alternative regimens are available but should be
managed by an infectious disease consultant.
 Immunizations post transplantation 261

Strongyloides
S. stercoralis is a soil-​transmitted nematode found worldwide, with a preva-
lence up to 60%-​≥ 71% in resource-​poor countries. Prevalence of S.
stercoralis in the US ranged between 0% and 3.8% in areas of Appalachia
and the southern US. Infective filariform larvae in the soil directly pene-
trate human host skin, migrate to the small intestine, then reach the lung
either directly or via the bloodstream. They are then coughed and ingested,
reaching the small intestine where they mature to adult form. Reproduction
within the human host results in endogenous autoinfection. In the immuno-
competent host, uncomplicated intestinal infection may present with non-​
specific GI symptoms, such as diarrhoea and abdominal pain, but is typically
asymptomatic. However, in SOT recipients, hyperinfection syndrome and
disseminated disease resulting from reactivation of chronic latent infection,
or primary infection from an infected donor may occur. Hyperinfection syn-
drome is characterized by a massive production of larvae in the GI system
and invasion of the lungs by the parasite, often combined with polymicrobial
bloodstream infection. Disseminated disease involves larval migration to
other organs. Meningitis by enteric Gram-​negative rods may also occur.
Heart transplant patients with Strongyloides hyperinfection syndrome may
present with nausea, vomiting or diarrhoea, cough, dyspnoea or respiratory
failure, fever, malaise, headache, altered mental status, and cutaneous rash.
A combination of diagnostic methods is often need for the diagnosis of
Strongyloides in the SOT recipient. Stool should be collected for ova and
parasite exam and PCR testing. For hyperinfection or disseminated dis-
ease, stool studies along with lower respiratory specimens (BAL) or other
affected body fluid or tissue should be evaluated. Larvae can be seen in
respiratory secretions, BAL, CSF, peritoneal fluid, urine, pleural effusion,
blood, and in other tissue specimens. Blood and CSF cultures should also
be collected to evaluate for Gram-​negative rods. Imaging of the chest often
demonstrates diffuse bilateral alveolar and interstitial pulmonary infiltrates,
reflecting underlying diffuse alveolar haemorrhage or acute respiratory dis-
tress syndrome. Brain imaging may reveal areas of new infarcts.
The drug of choice for Strongyloides is ivermectin by mouth then repeated
2 weeks later. Albendazole is considered a second-​line agent. Strongyloides
hyperinfection syndrome and disseminated disease require longer duration
of treatment. Management also involves a combination of broad-​spectrum
antimicrobials for systemic bacterial sepsis and reduction of immunosup-
pression, if feasible.

Immunizations post transplantation

Whenever possible, the recommended series of vaccines should be ad-
ministered before transplantation, allowing for completion at least 2 weeks
prior to transplantation (Table 21.5). Live-​attenuated vaccines should be
administered at least 4 weeks prior to transplantation to ensure resolution
of viral replication. Vaccines noted to be safe for administration after trans-
plantation may not be sufficiently immunogenic post transplantation and
should be delayed at least 3–​6 months following transplant. Live vaccines
are not recommended after SOT. For patients who are unvaccinated or in-
completely vaccinated prior to transplant, infectious disease consultation
should be obtained.
26

262 Chapter 21 Infections after heart transplant

Table 21.5 Vaccinations in the post-​transplant period

Vaccine Inactivated Can be Comments
(I)/​live given after
attenuated transplant
(LA)
Anthrax I No
BCG LA No
Cholera LA No Travel vaccine
Cholera and I Yes Travel vaccine; oral inactivated
traveller’s vaccine against cholera and
diarrhoea enterotoxigenic Escherichia coli
provides short-​term protection.
(Not available in the US)
Haemophilus I Yes If residential college student,
influenzae B receiving eculizumab, asplenic or
functional asplenia (e.g. sickle cell
disease)
Hepatitis A I Yes If hepatitis A antibody negative
(and vaccination not previously
administered)
Hepatitis B I Yes Hepatitis B series should be offered
prior to transplantation. If HB
surface antibody and surface antigen
are negative, may administer
Human I Yes If age ≤26 years and not previously
papillomavirus administered (patients up to age 45
can receive the vaccine at patient/​
provider discretion)
Influenza, I Yes High-​dose or booster during same
inactivated season has higher immunogenicity
than single dose
Influenza, LA NO
intranasal
Japanese I Yes Travel vaccine
encephalitis
Measles/​ LA No
mumps/​
rubella
Neisseria I Yes If residential college student,
meningitides receiving eculizumab, asplenic or
ACYW and N. functional asplenia (e.g. sickle cell
meningitides B disease)
Pertussis I Yes
(Tdap)
Polio, I Yes
inactivated
Polio, oral LA No
 Immunizations post transplantation 263

Vaccine Inactivated Can be Comments

(I)/​live given after
attenuated transplant
(LA)
Rabies I Yes Not routinely administered.
Recommend for exposure
or potential exposure due to
occupation
Rotavirus LA No Usually administered prior to
transplantation in paediatric patients
Salmonella LA No Travel vaccine
typhi, oral
S. I Yes Travel vaccine
typhi,
intramuscular
Streptococcus I Yes PCV20 conjugate, PPSV23
pneumoniae polysaccharide
conjugate
or poly
saccharide
Smallpox LA No Transplant recipients with face-​to-​
face contact with smallpox should be
vaccinated. Vaccinia immune globulin
may be administered concurrently if
available. With less intimate contact,
vaccine should not be administered
Tetanus I Yes
Varicella, I Yes Varicella IgG positive and age
subunit ≥50 years (and not previously
administered)
Varicella, live LA No
Yellow fever LA No Travel vaccine. Severely
immunosuppressed travellers are
encouraged to avoid travel to areas
where there is risk for yellow fever
264
 Section 4

Heart failure
in paediatric
patients

22.1 Heart transplant for paediatric patients 267

22.2 Paediatric extracorporeal membrane
oxygenation 275
22.3 Paediatric Heart transplant: specific
considerations 287
26
 Chapter 22.1 267

Heart transplant
for paediatric patients
Selection criteria and evaluation for donor and recipient 268
Technical aspects of heart transplantation for congenital
conditions 270
268

268 Chapter 22.1 Heart transplant for paediatric patients

Selection criteria and evaluation

for donor and recipient
Paediatric heart transplant listing criteria are not rigid. The scarcity of donor
organs means that each patient needs to be carefully assessed in order to
identify patients with the greatest need and most favourable outcome post
transplantation.
Recipient selection criteria
The following are used as guidance criteria for listing in the UK.
Indications for listing
• End-​stage HF despite maximal medical therapy.
• Complex congenital heart disease with no further medical or surgical
options.
• Progressive PH secondary to systemic ventricular failure that would
preclude transplantation at a later date.
• Failing Fontan circulation with no medical or surgical option.
• Refractory malignant arrhythmia causing haemodynamic compromise.
• Unresectable cardiac tumours causing haemodynamic consequences.
• Unacceptable poor quality of life secondary to advanced HF.
Contraindications
• Active malignancy or history of previous treatment with a high
recurrence risk.
• Irreversible end-​organ failure:
• Decompensated liver cirrhosis is a contraindication to heart-​only
transplantation.
• Severe renal abnormality/​dysfunction not likely to be reversible
and not suitable for continuing dialysis or renal transplantation at a
later stage.
• Progressive or refractory multisystem organ failure unlikely to be
resolved by heart-​only transplantation.
• Other significant organ dysfunction (including neurodevelopmental
delay) likely to lead to an inability to adhere to post-​transplant
treatment regimens or poor quality of life.
• Progressive systemic disease with early mortality (genetic, metabolic,
syndromic, or neurological disease).
• Anatomical concerns regarding appropriate anastomosis (e.g. severely
hypoplastic pulmonary arteries or veins).
• Severe, fixed elevation of PVR is a contraindication to heart-​only
transplantation.
• History of recurrent medical non-​compliance despite an appropriate
multidisciplinary and safeguarding team involvement.
• Psychological or psychiatric conditions not amenable to treatment
leading to non-​compliance with treatment regimens after transplantation
in spite of appropriate multidisciplinary team involvement.
• A lack of appropriate social support to enable compliance with
treatment regimens after transplantation.
 DONOR AND RECIPIENT SELECTION CRITERIA 269

Relative contraindications
• Malignancy in remission, patients should be assessed on an individual
basis.
• Severe renal impairment not likely to be reversible.
• Active sepsis from bacterial, fungal, or severe viral infection.
• Prior sensitization with high levels of HLA antibody.
• Diabetes mellitus with evidence of severe end-​organ damage.
• Morbid obesity.
• Limited venous access.
Indication for retransplantation
• Significant CAV.
• Graft dysfunction without evidence of acute graft rejection.
Criteria for urgent listing
• Short-​term MCS for acute haemodynamic instability (e.g. ECMO, Berlin
heart VAD).
• Need for more than one IV inotrope.
• Need for IV inotrope and mechanical ventilatory support.
Donor evaluation
Essential donor information at first call includes
• Blood group.
• Weight and height.
• Cause of death.
• Duration of CPR (if any).
• Amount of inotropes required to maintain haemodynamic stability.
• Echocardiography.
The paediatric heart has great potential for recovery after acute events. A
short run of ECMO support following transplant is not an uncommon prac-
tice. This has made the use of grafts with borderline function more accept-
able than in adult population with no difference in outcome.
Donor–​recipient size matching
Donor weight is the most commonly used size-​matching criteria. Donor–​
recipient size mismatch is common in paediatric transplants. Oversizing up
to three times is usually feasible. Delayed sternal closure should be per-
formed following diuresis and resolution of allograft myocardial oedema.
The disadvantage of oversizing are hypertension and possible left bronchial
compression.
Smaller donors with >20% mismatch are usually avoided as this is associ-
ated with poor outcome.
Procurement
• Harvesting team assessment is crucial. Visual assessment, TOE (if
available), and Swan–​Ganz readings (in older donors) can confirm the
quality of the graft.
• Proper communication between the implant and the harvesting teams
is essential. Extra donor tissue (e.g. the entire length of the SVC, the
left innominate vein, the aortic arch, the branch PAs, and even the
pericardium) may be required for complex reconstruction.
270

270 Chapter 22.1 Heart transplant for paediatric patients

Technical aspects of heart

transplantation for congenital conditions
Heart transplantation in congenital populations is more technically challen-
ging than in cardiomyopathy patients. Challenges commonly faced are:
• Multiple previous sternotomies.
• Proximity of the cardiac structures to the sternum.
• Anomalous systemic or pulmonary venous connections.
• Previous reconstruction of the great vessels.
• Previous catheter intervention and placed stents/​devices.
• Questionable patency of peripheral vessels.
• Significant collateral load in cyanotic patients.
Conduct of bypass
Preoperative preparation
• Cross-​sectional imaging is vital to plan entry strategy, CPB cannulation,
and need for any technical modifications.
• Imaging of peripheral arterial and venous access.
• Planning monitoring and invasive lines with anaesthetic team.
• Coordination between the implant and the harvesting teams is essential
to avoid any unnecessary long cold ischaemia time, especially in patients
with VAD support, those with anticipated dense adhesions, or those
who need reconstruction of the arterial or the venous pathway.
Operative steps
• Bicaval OHT is currently the most used technique.
• Set up is similar to any open-​heart surgery with stress on redo
precautions.
• Groin or neck vessels included in the surgical field and exposed or
cannulated as planned.
• Standard median sternotomy is performed.
• Aortic and bicaval cannulation performed as distal as possible with caval
tapes placed.
• CPB flow adjustment for collateral run-​off and extra suction lines for
significant pulmonary return or bleeding.
• A LA vent is placed to reduce the distension of the donor heart.
• Any required reconstruction of the arterial or venous pathway is done.
• In HLA and ABO incompatibility, plasma exchange is undertaken prior
to commencing CPB using a short period of circulatory arrest.
Recipient cardiectomy
Usually performed just before the donor heart arrives in theatre:
• The great vessels are cut just distal to their respective valve.
• The SVC is cut at the SVC–​RA junction and IVC to be cut with a
generous cuff of RA.
• Caution is required when cutting the LA to leave sufficient cuff around
the pulmonary veins.
• Cardiectomy may be done earlier if there is a need for extensive
reconstruction of arterial or venous pathway.
• Haemostasis is addressed before implanting the donor heart as it is
usually difficult to reach afterwards.
 TECHNICAL ASPECTS OF HEART TRANSPLANT 271

Preparation of the donor heart

• The aorta and PA are separated.
• LA cuff is created by connecting incisions between four pulmonary veins
(if the lungs were not used).
• The fossa ovalis is inspected (some units prefer to keep PFO patent
or create a small hole in the floor of the fossa ovalis that helps
decompression of the left side if ECMO is required in the postoperative
period).
• The SVC is trimmed to have as wide an anastomosis as possible.
Bicaval anastomosis
Usually carried out in the following order:
• LA anastomosis starting adjacent to the site of LAA.
• A vent is left in place to avoid distension of the donor heart and help
de-​airing.
• PA anastomosis avoiding a long cuff that may produce kinking.
• Anastomosis of the aorta. A dose of corticosteroids is usually given at
this point before release of the aortic cross-​clamp.
• Some centres prefer to inject warm cardioplegia and/​or warm shot
before release of THE clamp.
• The anastomosis of the IVC and finally SVC are to be performed on
beating heart.
• Attention is required for SVC anastomosis. Interrupted or semi-​
continuous running sutures are used to avoid A purse-​string effect.
Preparation for weaning off CPB
Once anastomosis is done and patient temperature is acceptable, the fol-
lowing has to be in place before weaning off CPB:
• Temporary pacing wires (multisite pacing is favourable to achieve the
best biventricular synchronization possible).
• LA line is placed through the LA vent site to monitor LA pressure.
• Appropriate inotropic support is in place.
• NO is routinely introduced to lower the PVR which is commonly high in
those patients.
• Proper de-​airing is confirmed on TOE or epicardial echocardiography.
After weaning off CPB, drains are inserted and the chest closed in the usual
fashion. In case of significant bleeding or borderline graft function, the chest
can be left open and sealed with membrane until chest closure is possible.
If the graft function is not good enough, ECMO is used to support the
circulation for few days until the graft recovers. ECMO cannulation could be
central (RA–​aorta) using CPB cannulas or peripheral, which would facilitate
chest closure and reduce the risk of infection.
Technical considerations
Bicaval versus biatrial
• Bicaval transplant is the commonly used technique nowadays:
• Lower risk of tricuspid insufficiency.
• Less arrhythmogenic.
• Avoids need for complex intra-​atrial adjustments (e.g. baffle
redirection as in Senning or Mustard procedure).
27

272 Chapter 22.1 Heart transplant for paediatric patients

• Biatrial can be helpful in rare situations as in left SVC with azygos

continuation of the interrupted IVC. It avoids the risk of SVC
anastomosis commonly seen with the bicaval technique. Also, it can
be used in neonatal heart transplantation to avoid stenosis at the caval
anastomosis site.
Dextrocardia
• The pericardial cavity is usually big enough to accommodate the donor
heart.
• You might need to cut the left side of the pericardium and open the left
pleura (avoid phrenic nerve).
• There is usually a dead pericardial space that was previously occupied
by the cardiac apex. The pericardium needs to be plicated and attached
to the chest wall to obliterate this space and avoid pericardial collection
or rightward rotation of the apex, which may distort the tricuspid valve
and the interventricular septum.
Venous pathway
Left SVC
• Presence of additional left superior caval vein with a bridging
(innominate) vein; it can be ligated and divided.
• Presence of a single left SVC or additional left superior caval vein
without a bridging vein:
• Anastomose to donor innominate.
• Use a prosthetic conduit and place in pre-​or retro-​aortic position
(Fig. 22.1.1).
Anomalous pulmonary veins
• Those are usually dealt with in a previous stage. The principle is to have
wide confluence and enough cuff of tissue around to anastomose to the
donor heart.
Great vessels
Aorta
• Arch hypoplasia or coarctation are usually dealt with at early stage.
• In presence of distortion, re-​coarctation, or calcified patches: repair is
carried out before the donor heart is brought into the chest. That might
require a period of circulatory arrest or selective cerebral perfusion.
• In patients with ventriculoatrial discordance (transposition of the great
arteries), restoring the normal relation between the aorta and the PA
may require:
• Reversing LeCompte manoeuvre with or without shortening of the
ascending aorta.
• Repositioning of the pulmonary confluence to the proximal left PA
(Fig. 22.1.2).
Pulmonary arteries
• Diminutive branch PA is usually a contraindication to heart-​alone
transplantation.
• Reconstruction of any focal narrowing of branch PA (preferable with
donor tissue).
• Resection/​division of previously inserted stents may be required.
• Intraoperative hybrid stenting of branch PA may be needed to avoid
kinking or compression.
 TECHNICAL ASPECTS OF HEART TRANSPLANT 273

RSVC

LSVC
Ao

uit
nd
Co

PA

Patch

Fig. 22.1.1 In case of a single left SVC (LSVC), a conduit can be used to connect it
to the RAA in a pre-​or retro-​aortic position. RSVC, right superior vena cava.

Transplantation in single-​ventricle patients

• Usually shows higher complexity and combination of previously
mentioned challenges.
• Patients are usually in poor general condition as result of Fontan failure
(liver cirrhosis, PLE, chylothorax, plastic bronchitis, etc.).
• Higher collateral load (commonly assessed on MRI or aortic angiogram)
requires higher CPB flows, avoidance of donor heart distension, and
meticulous haemostasis.
Outcome
The outcome of transplantation in patients with congenital heart disease
is relatively good and has improved over the years. However, considering
all the previously mentioned challenges, early survival remains lower than
in children with cardiomyopathy (3-​year survival of 79% vs 88% respect-
ively) this gap tends to get smaller with time (10-​year survival 68% vs 70%).
Although results have improved, patients with a single ventricle still have
10-​year survival of 53%.
274

274 Chapter 22.1 Heart transplant for paediatric patients

(a)

(b)

Ao
LPA

PA

Fig. 22.1.2 (a) Typical aortopulmonary arrangement following LeCompte

manoeuvre with the PA anterior to the aorta. (b) The donor main PA is anastomosed
to the proximal left PA (LPA) while the distal cut end of the recipient’s main PA (at
previous confluence) is now closed.
 Chapter 22.2 275

Paediatric extracorporeal
membrane oxygenation
Introduction 276
Patient selection in paediatric ECMO 276
Technical aspects of ECMO 277
Cannulation strategies 279
Monitoring of ECMO 282
Anticoagulation 283
Complications of ECMO 285
Weaning from ECMO 285
Outcomes of ECMO 286
276

276 Chapter 22.2 Paediatric ECMO

Introduction
Extracorporeal membrane oxygenation (ECMO), also known as extracor-
poreal life support (ECLS), is well established as therapy for cardiorespira-
tory failure in the paediatric population.
The care of children on ECMO is a complex task, requiring the expertise
of specialized intensivists, surgeons, haematologists, cardiologists, and re-
spiratory physicians. ECMO is deployed in increasingly complex situations
to provide the time required to tackle the underlying pathophysiology and
allow intrinsic recovery of lungs and heart or bridging to decision.

Patient selection in paediatric ECMO

Patients who are potential ECMO candidates should be discussed at the
earliest opportunity with the appropriate paediatric ICU/​surgical team.
Early initiation of ECLS can reduce end-​organ malperfusion and failure.
The following inclusion and exclusion criteria are suggestive. Each case
merits individual discussion with an ECMO centre.
Inclusion criteria
• >34 weeks of gestation.
• >2.5 kg body weight. Neonates with lower birth weight may be
candidates in certain centres depending on assessment of vessel size for
cannulation. Successful ECMO has been reported in preterm neonates
of <32 weeks of gestation and <1.5 kg.
Exclusion criteria
• Major intracranial haemorrhage.
• Severe neurological injury.
• Lethal congenital malformation, chromosomal anomalies.
Indications
Cardiac
• Post cardiotomy:
• Failure to wean from CPB.
• Postoperative instability or cardiac arrest.
• Post transplantation: donor heart dysfunction, RV failure secondary to
PH.
• HF:
• Cardiomyopathy (idiopathic, viral, nutritional), structural heart
lesions (anomalous left coronary artery from the pulmonary artery
(ALCAPA), obstructive cardiac lesions, borderline ventricle).
• As a bridge to diagnosis, recovery, VAD support, or transplantation.
• Intractable arrythmia.
Pulmonary
• Pulmonary infection. Viral pneumonia is the most common cause of
respiratory failure in children beyond the neonatal period.
• Acute respiratory distress syndrome/​acute lung injury.
• Refractory asthma.
• Aspiration.
 Technical aspects of ECMO 277

• Neonatal lung disease:

• Primary PH.
• Congenital diaphragmatic hernia.
• Meconium aspiration.
• Hyaline membrane disease (infant respiratory distress syndrome).

Others
• Septic shock.
• Drowning.
• Hypothermic arrest.
• Arrest of unknown cause: bridge to diagnosis and resuscitation.
• ECMO support in cardiac arrest (E-​CPR) can be offered for
witnessed cardiac arrest with ongoing resuscitation attempt where
conventional measures have failed. Caution should be exercised in
unwitnessed arrest situations.

Technical aspects of ECMO

Various configurations exist to provide pulmonary-​ alone or cardiopul-
monary support. Regardless of configuration, the basic principle of ECMO
combines a centrifugal pump, oxygenator, sweep gas exchanger, and
heater–​cooler and optional haemofiltration device in complete circuit with
the patient.
Venoarterial support
(Also see Chapter 8.) VA ECMO supports both the cardiac and pulmonary
systems and is warranted where both gas exchange and cardiac output are
inadequate. Blood is extracted from the venous system, oxygenated, and
returned to the arterial system via a centrifugal pump.
VA ECMO should be considered in patients with inadequate cardiac
output to maintain tissue perfusion despite maximal inotropic therapy.
Venovenous support
(Also see Chapter 8.) VV ECMO can be employed to support the pul-
monary system alone in paediatric patients with potentially reversible
causes of respiratory failure. Blood is extracted from the venous system,
oxygenated, and returned to the venous system via a centrifugal pump.
Concomitant cardiac dysfunction warrants consideration of VA ECMO as
cardiac output is not supported in a VV configuration. Older children with
irreversible pathology such as cystic fibrosis may also be candidates as a
bridge to lung transplantation.
VV ECMO should be considered in patients with:
• Oxygenation index >35–​40 in neonates and >25 in older children.
• Severe hypoxic respiratory failure demonstrated by PaO2/​FiO2 ratios
<60–​80.
• Severe hypercapnic respiratory failure demonstrated by sustained
respiratory acidosis.
Patients will have reached maximal ventilation therapy including appropriate
application of high-​frequency oscillatory ventilation, prone positioning, and
inhaled NO therapy.
278

278 Chapter 22.2 Paediatric ECMO

Other configurations
VA ECMO with left heart drainage
LA venting is recommended in cardiac ECMO for poor ventricular func-
tion to optimize recovery. As ECMO flow is established, afterload in-
creases which may negate any ejection from the poorly functioning LV. A
poorly contractile left heart is unable to empty collateral return effectively.
Consequently, left heart distension occurs resulting in pulmonary oedema
and limited myocardial recovery due to increased wall tension, comprom-
ised coronary flow, and subendocardial ischaemia.
Options for management
• Ideally, a left heart vent should be routinely placed and incorporated
into the venous return limb:
• Central cannulation: direct placement via LAA or right superior
pulmonary vein, LV apex is another option; in the context of previous
cardiac surgery, LV apex can be approached via left thoracotomy.
• Peripheral cannulation: percutaneous insertion via femoral vein across
PFO/​ASD under fluoroscopy/​TOE guidance.
• A balloon/​blade atrial septostomy with placement of a stent is
performed. This decompresses the left heart into the negatively
pressurized RA.
• Inotropic support allows some ejection to empty to left heart. This
must be balanced with resting of the myocardium to allow recovery.
VAV ECMO
(Also see Chapter 8.) VAV ECMO may be required in patients with both
cardiac and pulmonary failure who are being supported by femoral VA
ECMO. In VA ECMO with minimal cardiac ejection, the myocardial and
cerebral flow is achieved in a retrograde fashion from the femoral ECMO
cannula to the head and neck vessels. As cardiac function recovers, the
heart will eject poorly oxygenated blood, due to the concomitant pres-
ence of pulmonary failure, preferentially to the upper body, while the lower
body is supplied with oxygenated blood via the ECMO circuit. The mixing
zone is dependent upon the competing cardiac ejection and can result in
differential hypoxia. This phenomenon can also be observed in congenital
heart patients with large amounts of collateral returns into the left heart
bypassing the lungs.
Patients on femoral VA ECMO should have continuous monitoring of
saturations in both upper and lower limbs and cerebral/​flank oximetry.
Options for management
• Increase ECMO flow to drive mixing zone retrograde to ascending
aorta. This may be limited by cannula size and may be detrimental to
cardiac recovery.
• Optimize ventilation to improve oxygenation of pulmonary venous
return. This may be limited by barotrauma and oxygen toxicity concerns.
• VAV configuration (see Chapter 8): additional arterial inflow cannula
via right IJ vein to deliver oxygenated blood returning to the RA
and thereby improving saturations of ejected blood from left heart.
Alternatively, hybrid combination of FA inflow and Avalon cannula can
be used.
• Conversion to central cannulation.
 Cannulation strategies 279

Conversion of ECMO configuration

A changing clinical picture may require a reconfiguration or conversion of
ECMO. Patients on VV ECMO may develop hypotension and inadequate
perfusion requiring a conversion to VA ECMO. Patients on VA ECMO may
experience resolution of cardiac dysfunction but persist with hypoxia re-
lated to lung injury. The patients should be considered for conversion from
VA to VV ECMO to mitigate complications associated with arterial cannula-
tion and to provide longer support on VV ECMO until lungs recover.

Cannulation strategies
Suitable cannula size, site, and safe insertion is critical to achieve adequate
support and achieve successful patient outcomes. Strategy depends on
the desired mode of ECMO, age/​weight of patient, vessel availability,
underlying condition, and experience of surgeon.
Choice of cannula
Imaging of vessels pre ECMO can be useful to guide cannula sizing. It is im-
portant to note that adequate flow is dependent on both inlet and outlet
cannula calibre. The size of cannula is determined based on patient weight
(Table 22.2.1). While the larger cannula offers higher potential flow, these
are more likely to cause vessel damage, lead to venous obstruction, inter-
mittent cannula obstruction due to vessel wall proximity, and distal flow
compromise. The cannula size/​weight table is used as a guidance only to
achieve the estimated full ECMO flow for the patient size. Smaller cannula
should always be made available and used when native vessels are small or
fragile and in difficult cannulation to avoid vessel damage or dissection. The
modern design of ECMO cannulas is capable of achieving optimal flow at
a smaller calibre.

Table 22.2.1 Choice of cannula

Patient size (kg) Arterial cannula (Fr) Venous cannula (Fr)
2 8 8–​10
3–​6 10 10–​12
6–​8 12 14
8–​16 14 17
16–​30 17 19
30–​40 17 21
>40 21 25

VV ECMO
• The classic configuration of VV ECMO includes two venous cannulas:
bilateral common femoral or femoral–​right IJ. Direct open cannulation
or Seldinger technique are both possible. Guidance can be with TOE
or fluoroscopy. The separation distance between the tips of these two
cannulas is important to minimize recirculation of the oxygenated blood.
280

280 Chapter 22.2 Paediatric ECMO

• A bicaval double lumen cannula can be inserted into the right IJ

permitting venous outflow to the oxygenator from the RA and IVC and
oxygenated outflow into the RA through a single site. The oxygenated
return is directed toward the tricuspid valve orifice to maximize forward
flow and avoid recirculation. VV ECMO cannulation can be achieved
with a fully percutaneous method under ultrasound guidance. While the
benefits of a single IJ cannula include facilitating ambulation of patients,
a wired reinforced bicaval design cannula is associated with a higher risk
of caval and cardiac perforation.
VA ECMO
Central configuration
Central cannulation is accessed via median sternotomy. Cannulas are placed
into the RA and ascending aorta. Central cannulation will be the technique
most familiar and reproducible to surgeons due to the routine of placing
patients onto CPB.
• Central cannulation may be most appropriate in post-​cardiotomy
patients requiring ECMO. Cannulas placed during CPB can be used
if cardiac dysfunction necessitating ECMO is present at separation
from bypass. In patients requiring ECMO due to instability in the
postoperative period, re-​entry through the sternotomy may be the
easiest route.
• Central cannulation can also be achieved rapidly via innominate artery
when there is a Gore-​Tex shunt in situ, such as following reconstruction
of the hypoplastic arch or after the Norwood procedure.
• Central cannulation is the preferred route for patients requiring ECLS
for sepsis as higher flows can be achieved secondary to larger vessels
and cannula calibre.
• Central cannulation is not feasible in children who require emergent
ECMO and who have had prior cardiac surgery.
• A disadvantage of central cannulation is the requirement to keep the
chest open, exposing the patient to infection, bleeding, sedation, and
paralysis.
Peripheral configuration
Neck cannulation
Paediatric patients can be cannulated in the neck for VA and VV ECMO. See
Fig. 22.2.1 and Fig. 22.2.2 for technical aspects.
• Cervical cannulation is the primary modality of peripheral ECMO in
neonates and infants.
• Higher risk of cerebral events is reported in older children >2–​3 years
old. Central cannulation should be considered instead, or femoral
cannulation in bigger children.
• Patency of bilateral carotid arteries and jugular veins can be established
quickly by bedside neck ultrasound scanning.
• If the clinical situation allows, pre-​ECMO cross-​section cerebral imaging
for baseline and assessment of the circle of Willis can be helpful. In
neonates, cranial ultrasound should suffice. Bilateral near-​infrared
spectroscopy can suggest cerebral flow compromise and indicate a
change in strategy is required.
 Cannulation strategies 281

• Benefits of neck cannulation include avoidance of sternotomy and its

risks of bleeding and infection. In urgent situations, neck cannulation can
be rapid and CPR can be continued during cannulation.
• TTE may be used to guide and check cannula position.
• CXR post cannulation to confirm good cannula position (Fig. 22.2.2).
• Upon decannulation, vessels can usually be repaired to re-​establish
ipsilateral antegrade cerebral flow.

(a) (b)

Fig. 22.2.1 (a) The patient is positioned, and anatomy identified for dissection.
(b) The vessels are isolated. Caution is required regarding the vagus nerve running
between the vessels.

(a) (b)

Fig. 22.2.2 (a) The dotted lines indicate the direction of the arteriotomy.
(b) Arterial cannulation is performed first, approximately 2.5 cm of advancement
into the carotid artery. The cannula is secured, and the process repeated in the
same manner for venous cannulation. Venous cannulation can be more challenging.
Advancement of 6–​7 cm is correct for most neonates. CXR and echocardiography
should be used to confirm position and status of LA distension.

Femoral cannulation
In older children weighing >15–​20 kg, VA ECMO can be achieved via fem-
oral vessels. While vessel calibre may be amenable to cannulation with ad-
equate size cannulation to achieve flows, caution must be taken to avoid
limb ischaemia due to occlusive cannula size.
28

282 Chapter 22.2 Paediatric ECMO

Distal limb perfusion can be achieved by various methods:

• Avoid direct cannulation of the vessel by a side graft anastomosis to the
femoral artery allowing bidirectional flow to the limb and body. Control
of flow into the limb can be difficult to monitor and control, resulting in
hyperperfusion. Leak via graft wall could be problematic—​this could be
mitigated by advancing the arterial cannula to the end of the prosthetic
graft, just above the femoral arteriotomy.
• A limb perfusion cannula can be placed into the distal femoral artery,
superficial femoral artery, or posterior tibial artery. Accurate perfusion is
possible with this technique. Distal limb perfusion cannula is usually feasible
only in larger patients, that is, in teenagers close to young adult size.

Monitoring of ECMO
Patient
While invasive and quantitative monitoring is essential during an ECMO
run, the importance of bedside clinical serial assessment must not be for-
gotten. Patients on ECMO require close observation and management of
haemodynamics as well as vigilance for complications related to bleeding
and thrombosis (see Complications of ECMO, p. 285).
• Tissue perfusion: cap refill, urine output, serial lactate measures.
Monitor lower limb perfusion with pulse oximetry and Doppler
ultrasonography where peripheral cannulas are sited.
• Invasive blood pressure monitoring is required: in cardiac dysfunction
with loss of ejection, arterial tracing may be non-​pulsatile.
• Pump inlet pressures are helpful in guiding management of
hypovolaemia. Shuddering of the lines may occur in a low-​volume state.
• Mixed venous saturations:
• Oxygen delivery.
• Recirculation in VV.
• Bloods:
• Coagulation profile (see Anticoagulation, p. 283).
• Platelet consumption: indicative of circuit thrombus.
• Echocardiography:
• Useful for confirmation of cannula placement in peripheral configurations.
• Guide to fluid status.
• In non-​pulsatile flow, attention must be paid to left heart distension.
• Establishing any underlying diagnosis and temporal changes.
• Detection of intracranial complications during ECMO therapy:
• Near-​infrared spectroscopy.
• Transcranial ultrasonography.

ECMO circuit
• Flow (L/​min): maintain adequate cardiac output. Flow can be adjusted
by changing the revolutions per minute (rpm).
• Inlet and outlet pressures: helpful in guiding fluid status.
• Pre-​and post-​membrane pressures: elevated transmembrane gradient
may indicate oxygenator compromise.
• Emboli count: where air emboli are noted, check all access ports and
taps of circuit and patient (e.g. central line).
 Anticoagulation 283

Flow rates
VA ECMO
• Individual patients require clinical evaluation to determine the ideal rate
based on perfusion parameters. The following calculations provide a
baseline guide for commencing ECMO:
• Patients <10 kg: 100–​150 mL/​kg/​min.
• Patients >10 kg: 2.5–​3.5 L/​m2/​min.
Consider higher flows in patients with septic shock, parallel circulation
(single-​ventricle physiology), and extracardiac shunts. A systemic–​PA shunt
such as modified Blalock–​Taussig shunt is not usually clipped or ligated.
VV ECMO
• Patients <10 kg: 70–​150 mL/​kg/​min.
• Patients >10 kg: 1.8–​3.5 L/​m2/​min.

Anticoagulation
Haemorrhagic and thromboembolic complications remain the major cause
of mortality and morbidity in ECLS. Blood contact with artificial surfaces of
the circuit, pump, and oxygenator initiate the clotting and inflammatory cas-
cades in an unpredictable manner. Plasma proteins become denatured during
ECMO, increasing plasma viscosity and haemolysis occurs secondary to
shear stress, cavitation, turbulence, and osmotic forces. Patients frequently
have multifactorial coagulopathies related to underlying pathophysiology.
Important factors of pro-​haemorrhagic status:
• Excessive heparin use.
• Consumption of coagulation factors.
• Low fibrinogen levels.
• Thrombocytopenia.
• Platelet dysfunction.
• Hyperfibrinolysis.
Important factors of pro-​thrombotic status:
• Inadequate use of heparin.
• Acquired thrombin deficiency.
• Consumption of protein C and protein S.
• Consumption of tissue factor pathway inhibitor.
• Endothelial dysfunction.
• HIT.
• Blood stasis in chambers and venous system.
Unfractionated heparin (UFH) is the most commonly used anticoagulant
used in paediatric ECMO, although UFH has an unpredictable response in
newborns and children. UFH has a short half-​life and is reversible with ad-
ministration of protamine sulphate. UFH is given as a bolus (50–​100 units/​
kg of body weight) at ECLS cannulation and is then infused continuously
during ECMO support under tight control of anticoagulation monitoring.
Alternative anticoagulation includes direct thrombin inhibition (e.g.
bivalirudin), factor Xa inhibitors (rivaroxaban, argatroban), and nafamostat
mesilate. No large-​scale studies on these alternatives are available in the
paediatric population. Their use may be indicated in children with HIT and
heparin resistance.
284

284 Chapter 22.2 Paediatric ECMO

Monitoring of anticoagulation
Individual ECMO centres will have protocol-​ driven monitoring of
anticoagulation depending on availability of testing panels. Quantitative
tests should be utilized in combination with the clinical picture. Most clot-
ting laboratories assays are chromatography based: elevated plasma bili-
rubin, plasma free haemoglobin, and triglycerides will affect the accuracy
of results.
The following tests are applicable in ECMO but not universally available:
• ACT: this is a readily available quick bedside test. It is the most
commonly used parameter in CPB. ACT is affected by many factors
such as platelet function, fibrinogen and factor levels, temperature, and
haemodilution status. An ACT of 180–​220 is typically targeted in the
non-​bleeding ECMO patient.
• Anti-​Xa activity: plasma anti-​Xa indirectly measures the activity of
heparins. It is commonly used as the primary monitoring tool in ECMO,
although controversy exists due to the poorly defined therapeutic
range. Anti-​Xa below therapeutic range may be due to inadequate
heparin or patient antithrombin deficiency.
• Antithrombin (AT) level: heparin effect is due to binding of heparin to
AT which results in a many-​fold increase in the anticoagulant activity of
AT. Adequate AT is required for effect of heparin. A relative deficiency,
and lower activity of AT is seen in the neonate and infant population. AT
concentrate replacement is indicated when reduced effect of UFH is evident.
• aPTT: aPTT is reportedly less reliable in the neonatal and paediatric
population when evaluating UFH status.
• aPTT plasma: measures aPTT after neutralization of heparin.
• Platelet count: should be maintained at a minimum of 80,000 cells/​
mm3. The decision to transfuse platelets is based on the count alongside
clinical assessment of bleeding status.
• Fibrinogen level: maintain at a minimal level of 100 g/​dL. Fibrinogen
concentrate or cryoprecipitates can be administered at subtherapeutic
levels.
• Viscoelastic tests (TEG): evaluate whole-​blood coagulation and
fibrinolysis. These are relatively new tests and their application is not
yet fully established. Time to fibrin formation, kinetics, clot strength and
stability, and function of platelets are evaluated.
Principles of clinical activity
• Perform an initial screening following initiation of ECMO to assess
baseline coagulation status: consider multifactorial coagulopathy (liver
compromise, factor deficiency, sepsis).
• Priming of the ECMO circuit will depend on urgency. In children, a
combination of banked blood and colloids is preferable with UFH
additive at 50–​100 units per unit of packed red cells. Heparin–​albumin
coated priming tubing is used in certain units to reduce blood–​plastic
interface.
• Blood products such as platelets, FFP, or cryoprecipitate can be
administered if significant coagulopathy exists prior to ECMO initiation.
During an ECMO run, the decision to transfuse packed red cells,
platelets, fresh frozen plasma, fibrinogen concentrate, or cryoprecipitate
is based on laboratory data with accurate correlation of clinical
assessment of coagulation status.
 Weaning from ECMO 285

• Close observation of all components of the circuit (tubing, connectors,

oxygenator, pump head) are required to monitor for clot or fibrin build-​
up. Monitor transmembrane pressure gradients which may indicate clot
formation not visible to the eye.
• Rigorous monitoring of patients for clinical indicators of haemorrhagic
or thrombotic sequelae.

Complications of ECMO
• Haemorrhagic events (cerebral, GI, pulmonary, surgical site).
• Thrombotic events.
• Infection.
• Limb ischaemia: in femoral ECMO due to occlusion of the femoral
artery—​this is mitigated with the use of graft anastomosis to allow
bidirectional perfusion or distal limb cannula flow.
• Stroke:
• Embolic or haemorrhagic.
• In neck cannulation: consider imaging of the circle of Willis to confirm
patency. Avoid neck cannulation in children >2–​3 years.
• Massive air embolus.
• Disseminated intravascular coagulation.

Weaning from ECMO

On resolution of underlying physiology, the child will need to be safely
trialled and weaned from ECMO.
Prior to weaning ensure:
• The processes and pathology leading to ECMO have been addressed
adequately.
• Pulmonary function and cardiac function have been adequately
assessed with ECHO, parameters of perfusion, CXR, and check of lung
compliance.
• Inotropy is running to an adequate level to support weaning process.
• Patient is on inhaled NO at 20 parts per million.
• Fluid/​blood boluses/​emergency drugs are available.
• Full ventilatory support and titration of sweep gas accordingly.
• Arterial blood gases should be checked every 5–​10 minutes during
wean. Echocardiography may be useful to assess cardiac function when
separated from ECMO.
Two techniques are reported:
• Once the ECMO is weaned to minimal flow 200 mL/​min, a bridge
is placed into the circuit between the inflow and outflow pipes. The
cannulas are left in situ and clamped, resulting in separation from ECMO.
The pump continues to circulate volume through the bridge to prevent
stasis. Intermittently, the cannulas are unclamped to flush through to
prevent clots. In case of a failed wean, the clamps are simply removed
and full ECMO circulation reinitiated. After a successful trial period of
separation, the child can be decannulated.
286

286 Chapter 22.2 Paediatric ECMO

• Retrograde flow trial. The rpm speed is reduced to the point where
native cardiac ejection overwhelms the pump vacuum. The child begins
to eject into the aortic cannula creating retrograde flow. If the child
is unable to generate flow >70 mL the trial should be abandoned. In
case of a failed wean, the rpm speed is turned up to achieve adequate
antegrade flow to re-​establish ECMO. After a successful trial period, the
child can be decannulated.
Coming off ECMO urgently
In rare situations, one may need to come off ECMO emergently, such as
major haemorrhage from accidental decannulation, circuit rupture or leak,
or air in the circuit between oxygenator and patient.
• Call for help (theatre team). Do not panic. Continuation of ECMO
could be detrimental and lead to exsanguination, Control haemorrhage,
cease ECMO, and resuscitate, before re-​establishing ECMO.
• Clamp the arterial cannula then the venous cannula, and open the
bridge if present.
• Local compression to control bleeding.
• Hand ventilate or optimize ventilator setting.
• Resuscitate with CPR, inotropes, and blood transfusion as necessary.

Outcomes of ECMO
• The use of ECMO has significantly increased over recent years: 68,693
neonatal and paediatric ECLS events have been reported to the
Extracorporeal Life Support Organization international registry.
• Short-​term survival rates of ECMO are dependent on underlying
diagnosis, severity of illness, type of support, duration of support,
comorbidities, and complications.
• Survival to discharge is lower for cardiac support with VA ECMO at
43% for neonates and 52% for paediatric compared to 73% and 59% for
pulmonary VV support.
• E-​CPR carries a 42% survival to discharge for both neonatal and
paediatric populations.
Cardiac ECMO and outcomes
Factors reportedly associated with mortality among ECMO recipients in the
cardiac population include E-​CPR, single-​ventricle physiology, lower body
weight, prematurity, pre-​ECMO acidosis, increased time to diagnosis of re-
sidual lesions, and longer time between intubation and initiation of ECMO.
The majority of survivors separated from ECMO at <8 days. Early initiation
of VA ECMO in cardiac patients and early address of residual lesions may
improve outcomes.
The single-​ventricle population
ECMO support of this group of patients remains uncommon but is
increasing. The unique anatomy and physiology of this population demands
thought in regard to cannulation, circuit flow, systemic thrombotic risk, and
resting ventilation. Previous opinion was of poor outcomes in this group of
patients, but recent data from the Extracorporeal Life Support Organization
registry suggest ECMO is a viable option in the single-​ventricle population.
 Chapter 22.3 287

Paediatric heart
transplant: specific
considerations
Common pathologies 290
Paediatric waiting list 291
Special considerations 292
28

288 Chapter 22.3 Paediatric heart transplant

Heart transplantation remains the most effective therapy in paediatric

patients with end-​stage heart disease not amenable to conventional sur-
gery. Besides many challenges that can be overcome in terms of unique
anatomy, physiology, and technical demands, the shortage of paediatric
donors remains the key limiting factor. Despite all these challenges, heart
transplantation offers the most durable solution in paediatric patients,
after exhausting all available repair options. Currently, with improved sur-
gical outcome, immunotherapy, and long-​term surveillance, median post-​
transplant survival surpasses 15 years in children and 22 years in infants
(data from ISHLT).
The first successful human-​ to-​
human heart transplant operation (by
Christian Barnard on 3 December 1967) was preceded by many pion-
eers, notably James Hardy (first human heart transplant, chimpanzee
donor, 1964); Norman Shumway and Richard Lower (Stanford); Vladimir
Demikhov (Moscow); and Charles Guthrie and Alexis Carrel (Nobel Prize
in Physiology or Medicine, 1912).
The first paediatric heart transplant operation followed shortly after.
Pioneers in paediatric heart transplantation
• Adrian Kantrowitz (1967, New York: neonatal Ebstein anomaly,
survived 6.5 hours).
• Denton Cooley (1968, Houston: 5-​year-​old, endocardial fibroelastosis,
survived 8 days; 2-​month-​old, complete AV septal defect, survived 14
hours).
• Magdi Yacoub (1984, London: 10-​day-​old neonate, survived 18 days).
• Eric Rose (1984, New York: 4-​year-​old, single ventricle, survived late).
• Leonard Bailey (Loma Linda, 1985: neonate, survived late).
Early attempts
Although ‘baby Fae’ captured the headlines in 1984 by receiving a baboon
heart, the first paediatric human heart transplantation took place 3 days
after Barnard’s success. After many years of intensive research, Adrian
Kantrowitz carried out the first paediatric heart transplant in a 20-​day old
neonate with severe Ebstein anomaly (6 December 1967). Unfortunately,
the baby died some 6 hours later, and the procedure was declared a ‘failure’.
Prior to the discovery of cyclosporine A (1976), there was little progress
due to acute allograft rejection.
Successful transplantation
On 9 June 1984, the first successful paediatric heart transplant procedure
was performed in Columbia University, New York. The recipient was a
4-​year-​old boy with a single ventricle, who lived until adulthood after a
second heart transplant. Sadly, he died in his sleep during the first week of
medical school. In Loma Linda, after baby Fae (survived 20 days), Leonard
Bailey performed the first successful newborn-​to-​newborn transplant on
20 November 1985 in a patient with hypoplastic left heart syndrome. This
marked the inception of a paediatric heart transplant programme. The first
neonatal transplant recipient was still alive 33 years post transplantation in
a late report (2019).
  Paediatric Heart Transplant: specific considerations 289

Unique characteristic of paediatric heart transplantation

Paediatric heart transplantation is unique from that for acquired heart dis-
ease in adults. Congenital heart disease affects 1 in 120 (0.8%) live births,
and constitutes approximately 40% of heart transplantation in paediatric
population. More than half of infant recipients have congenital heart lesions.
• Size of recipient is diverse in the paediatric population, ranging from a
teenager with young adult size to a tiny newborn.
• Unique anatomy that is rarely seen in acquired heart disease and include
higher proportion of left SVC, situs inversus, dextrocardia, heterotaxy,
and arch anomalies.
• Pulmonary vascular disease, malnourishment, and sensitization are all
more common in paediatric patients. Therefore, heart transplantation
presents with specific challenges in the paediatric population.
• On the other hand, paediatric patients benefit from their immature
immune system, and, using this advantage, the ABO-​incompatible
(ABOi) heart transplantation was introduced.
Another important distinction from the adult population, who primarily
presented with cardiac (pump) failure, is a paediatric cohort with preserved
cardiac function. These patients have circulatory failure as their primary
reason for transplantation following Glenn/​Fontan failure or multifocal ob-
structive disease. They form an ‘unconventional’ group who need a heart
transplant.
Specific challenges in paediatric heart transplantation
Besides anatomical and technical demands, the availability of a suitable
donor is probably the greatest challenge in paediatric heart transplantation.
In end-​stage congenital heart disease, patients usually present with a gradual
decline rather than acute cardiac decompensation. Therefore, the urgency
on the waiting list is under-​prioritized, further adding to the waiting time.
Excess waiting time increases waiting list mortality. Specific challenges in
paediatric patients include:
• Shortage of donors especially in small recipients (particularly <10 kg).
• Lack of clinical prioritization: absence of overt HF as in failing Fontan
circulation.
• Sensitized patients: presence of HLA antibodies due to previous blood
transfusion or homograft materials.
• Elevated PVR secondary to left heart disease, diastolic dysfunction, or
borderline left heart structure as in Shone’s complex.
• Multiple significant collateral load in cyanotic patients.
• Anatomical challenges: bilateral SVC, dextrocardia, transposed great
vessels, heterotaxy.
• Technical challenges include multiple previous sternotomies, extensive
collaterals in cyanotic patients, occluded peripheral vessels for bypass,
previous stent or shunts, and additional reconstruction required for arch
of pulmonary arteries
Primary transplantation (or conventional surgery)
Although heart transplantation was offered as a primary treatment option
for complex congenital heart lesions in the early era, this is rare nowadays.
Transplantation is usually reserved for after palliative or repair procedures
290

290 Chapter 22.3 Paediatric heart transplant

which have failed to improve the patients, and hence is termed a ‘rescue
therapy’. Nevertheless, late survival appears to be more encouraging
in patients who have had a primary transplant procedure for complex
univentricular heart.
The first successful neonatal transplantation recipient who had
hypoplastic left heart syndrome was still alive 33 years later with cur-
rent ongoing follow-​up. With reported long-​term follow-​up of neonatal
heart transplantation that exceeded 64% survival at 20 years and 55%
at 25 years, late outcome surpassed that of conventional palliation for
hypoplastic left heart syndrome. The shortage in the donor pool for
the paediatric population precludes heart transplantation as a primary
therapeutic option, which is reserved for when conventional surgery is
unfavourable.

Common pathologies
The two most common groups of pathologies referred for consideration or
evaluation for transplant are cardiomyopathy and congenital heart disease. In
infants, congenital heart disease is the more common indication for trans-
plantation than cardiomyopathy. This differs in children over 1 year old
where cardiomyopathy is the most common indication. Cardiac tumours
and retransplantation form very small groups.
Cardiomyopathy
• Dilated cardiomyopathy: idiopathic (66%); acute or post myocarditis
(16%); neuromuscular disorder (e.g. Duchenne and Becker dystrophy)
(9%); Barth syndrome; familial cardiomyopathy (5%); inborn error of
metabolism (1%); autoimmune disease; toxins such as chemotherapy;
nutritional cause: vitamin D deficiency, carnitine (amino acid) deficiency;
thyrotoxicosis.
• Hypertrophic cardiomyopathy: idiopathic (most common), familial,
inborn errors of metabolism (Pompe disease, mitochondrial disease),
Noonan syndrome.
• Restrictive cardiomyopathy.
• Tachycardia /​arrhythmia induced cardiomyopathy.
The incidence of cardiomyopathy is significantly higher in the infant
population. DCM is the most common form (approximately 50–​75% of
all cardiomyopathies) and presents earlier in childhood. Forty per cent
of paediatric DCM cases either die or need a transplant within 5 years of
diagnosis.
Hypertrophic cardiomyopathy is the second largest group (40% of car-
diomyopathies). Although presenting later than DCM, hypertrophic cardio-
myopathy has a similar long-​term outcome with a mortality rate 2 years
after diagnosis of 12.7% versus 13.6% for DCM.
Other forms of cardiomyopathy, that is, restrictive and arrhythmia-​
induced cardiomyopathy, are rare (<5%) in the paediatric population.
Restrictive cardiomyopathy is particularly problematic due to early devel-
opment of PVR and its consequences related to early graft failure.
 Paediatric waiting list 291

Congenital heart lesions

Rescue therapy following conventional surgery:
• Biventricular repair: Shone’s complex, borderline left heart, endocardial
fibroelastosis.
• Univentricular palliation: failing single ventricle following
arteriopulmonary shunts, Glenn or Fontan procedures; protein-​losing
enteropathy (PLE), more rarely for plastic bronchitis.
Heart transplantation as primary therapy can be considered in the following:
• Hypoplastic left heart syndrome with (1) tricuspid, that is, systemic
atrioventricular valve that is severely dysplastic and regurgitant or
(2) poor ventricular function.
• Other single ventricle disease with (1) poor ventricular function or
(2) severely stenotic or regurgitant atrioventricular valve.
• Pulmonary atresia with intact septum and RV-​dependent coronary
circulation with aorto-​coronary atresia or significant proximal coronary
stenosis.
• Pulmonary atresia with intact septum, sinusoids in RV myocardium, with
reduced LV function and/​or sign of myocardial ischaemia but patent
proximal coronaries.
• Pulmonary atresia with intact septum with severe tricuspid regurgitation
and massive (wall-​to-​wall) RV dilatation.
• Single ventricle with heterotaxy and total anomalous pulmonary venous
drainage.
• Most severe form of Ebstein anomaly in neonates.
Cardiac tumours
• Benign: multifocal and non-​resectable intracardiac tumour.
• Primary cardiac malignancy (e.g. sarcoma) that is confined to the heart.

Paediatric waiting list

Waiting list mortality
The size of the donor pool remains at best stagnant if not reduced, while
more candidates are added onto the waiting list. MCS in paediatric pa-
tients has improved and allowed more children to wait longer. Due to the
shortage of donor hearts, time and mortality on waiting list are higher; the
issue is more pronounced for the paediatric population. The mortality rate
on the waiting list among paediatric candidate under 5 years old is 15–​20%
by 12 months after listing. Congenital heart disease patients are also more
likely than those with cardiomyopathy to die on the waiting list (nearly one-​
third) prior to transplantation.
The law
There is a regional discrepancy of donor availability that could be due to
culture, faith, belief, and public awareness. In order to increase organ do-
nation, the law has changed in the UK. From May 2020, organ donation
in England has changed to an ‘opt out’ system after Parliament approved
‘Max and Keira’s law’: All adults are now presumed organ donors in England
29

292 Chapter 22.3 Paediatric heart transplant

unless they’ve registered to opt out or informed their family. The law does
not apply to (1) those under 18 years old; (2) those who lack mental cap-
acity, and therefore cannot opt out; and (3) visitors or those with less than
12 months residency in England before death. The new law, however, may
not improve the paediatric waiting list, as it applies only to adults.
Strategies to increase the paediatric donor pool
In order to broaden the donor pool for children, oversized donors are
already used for paediatric patients, as well as ABOi donors. In order
to further broaden the donor pool, size matching using criteria other
than weight has been evaluated (see Special considerations, p. xxx).
Awareness among public and intensive care units continues to play a piv-
otal role to increase donors for paediatric recipient. The rate of organ
donation among infants is ten times lower than older age groups. A lower
organ donor recruitment rate could be due to unfamiliarity among paedi-
atric practitioners or difficulty in the diagnosis of brainstem death in
young patients.

Special considerations
Oversize donor
In the paediatric population, the use of an oversized donor has been shown
to be safe: the strategy will widen the donor pool and reduce time on the
waiting list. The 10-​year survival has been shown to be comparable be-
tween oversize and conventional size matching.
Donor versus recipient size
Current practice on the paediatric waiting list is to match recipient and
donor based on weight (donor weight 0.6–​3.0 times that of the recipient
has not been shown to be associated with adverse outcomes). In general,
an oversized donor with weight up to three times that of the recipient can
be accepted, although some centres reported safe use of ‘extreme’ mis-
matched heart that is over three times the recipient weight. Older children
(>10 years old or >20 kg) usually can accept a heart from the adult donor
pool. Newer assessment methods to match include using height versus es-
timated cardiac dimension (by echocardiography) to assess for a suitable
donor otherwise not matched by conventional weight method.
Potential issues with oversizing
The use of an oversized donor heart is associated with a higher rate of
delayed sternal closure.
Other issues that have been linked to oversized donor hearts include
(1) hypertension in recipient, (2) early overperfusion and potential cerebral
injury, (3) lobar collapse, and (4) RV compression/​failure following sternal
closure. Many of these issues are rare in practice including ‘big heart syn-
drome’ (neurological complication manifesting as post-​transplant seizure
due to overperfusion and reflex cerebral vasoconstriction).
Technical considerations
To fit in a bigger heart, technical considerations include (1) opening the
pleural space; (2) left pericardiectomy above phrenic nerve, rarely bilateral;
 Special considerations 293

(3) reduction of donor atrial size; and (4) relaxing incisions on diaphragmatic
surface
Cautions in oversizing
Careful considerations are required in the following settings: (1) very small
neonate who has tiny intrathoracic space, and (2) history of multiple re-​
sternotomies or known to have severe adhesions. In patients with densely
cemented adhesions, the pericardial space is tight and pleural spaces may be
densely adherent and could not be opened to fit in the donor heart.
When there is concern that a donor heart may not fit, further assess-
ments could aid in the decision of the implanting surgeon whether to accept
the offer: (1) cardiothoracic ratio on chest radiographs, (2) echocardio-
graphic assessment (intercaval distance, between SVC/​RA and IVC/​RA
junctions and LV end-​diastolic dimension), and (3) feedback from the pro-
curing surgeon after measuring the width and length of the donor heart.
Marginal donor
A significant proportion of paediatric patients awaiting heart transplantation
die on the waiting list. The mortality is higher in younger recipients where
donors are scarce. On the other hand, offers are turned down in a sig-
nificant proportion of potential donors (one-​third of all donors). Potential
donors with impaired ventricular function (EF <50%), on high-​dose or mul-
tiple inotropes, with functional valvar regurgitation are often turned down.
These marginal hearts could be considered as donors:
• Depressed donor heart function could be reversible, such as myocardial
dysfunction associated with brain death or in the setting of sepsis.
• Use of donor hearts with mildly impaired (LVEF 45–​54%) and
moderate-​severely depressed (<45%) and/​or segmental wall motion
abnormalities resulted in comparable post-​transplant survival outcomes
with donor hearts with normal LVEF (≥55%).
• History of CPR should not prelude a donor.
• Use of vasopressors and inotropes are not contraindications.
• Elective post-​bypass ECMO following implantation may be considered
to support cardiac recovery.
ISHLT guidelines recommend refusal of donor heart with EF less
than 40%.
ABO incompatibility (ABOi transplantation)
ABOi heart transplantation was first performed in infants (late 1990s,
Canada). ABOi transplantation is usually contraindicated due to hyperacute
rejection: anti-​A or anti-​B antibodies attack the donor’s antigen on vascular
endothelial cells, activate the complements, and cause vascular thrombosis.
ABO compatible only donor pool would result in excessive wait time and
mortality especially for children in O-​group.
Immunological basis
The immune system in infants is still immature. In the first few months
of life, infants have only maternal antibodies. Anti-​A or -​B antibodies
(isohaemagglutinins) to blood group antigens develop gradually from 6 to
24 months of age due to gastrointestinal colonization of microorganisms.
The complement attack system is also not fully mature yet. When an ABOi
294

294 Chapter 22.3 Paediatric heart transplant

donor heart is implanted, hyperacute rejection does not occur due to ab-
sent or lack of anti-​A or -​B antibodies.
Interestingly, antibody levels remain absent or very low late after ABOi
transplantation in most patients, suggestive of development of long-​term
tolerance towards the donor’s ABH antigen. This immune tolerance (B-​
cell tolerance) allows excellent long-​term graft survival and freedom from
antibody-​mediated rejection.
ABOi listing is institution based. Currently, UNOS recommends primary
ABOi listing for those under 2 years old and low isohaemagglutinin level
1:4 or less.
• Check for detectable anti-​A or -​B levels in plasma.
• If positive, quantify the level of antibodies by haemodilution (called
‘isohaemagglutinin titre’). The plasma is diluted until no antibody is
detected (1:16 titre has a higher level of antibodies than 1:4).
• Repeat titre periodically. Immediate repeat test on donor heart offer. If
titre is high, discuss with surgeon/​cardiologist.
• Intraoperative complete whole body exchange transfusion is
commenced until titre is less than 1:4.
• Repeat titre regularly prior to release of cross-​clamp and reperfuse
heart only after titre level fall. Repeat titre at regular interval after donor
heart reperfusion.
Older children can be considered based on urgency (oldest ABOi recipient
was 8.8 years old) and higher titre of 1:16 or greater is possible and does not
exclude ABOi transplantation (up to 1:256 has been reported successfully).
ABOi listing and transplantation outcome
ABOi transplantation shortens the waiting time (by 20–​25%) and reduces
mortality (from >50% to <10% in infants). Although ABOi listed patients
are younger, with higher urgency status and risks (more mechanical ventila-
tion, congenital heart disease), post-​transplant survival, graft rejection, and
vasculopathy rates are not inferior in ABOi recipients.
Blood transfusion during and after ABOi transplantation
(See Table 22.3.1.)
Blood group A has A-​antigen on erythrocytes (and platelets) and anti-​B
antibodies in plasma. Group B has B-​antigen and anti-​A.
Group O has no antigen, but both anti-​A and anti-​B in plasma, therefore
avoid giving O plasma or platelet (as platelet preparation from O blood
comes with antibodies).
Blood group AB has both antigens but no anti-​A or anti-​B antibodies in
plasma. Therefore, use AB plasma product or platelet only for A, B, and
AB recipients. In an O recipient, product from the AB group or the donor
heart’s blood group can be used.
For red cell transfusion, use O red cells or recipient’s blood group.
Sensitization in paediatric recipients
Allosensitization in paediatric recipients is an important issue that limits the
availability of a matching donor and has an important impact on graft lon-
gevity. This refers to immunological sensitization due to preformed anti-
bodies against HLA and occasionally non-​HLA.
 Special considerations 295

Table 22.3.1 Compatible typing of blood products for transfusion during and
after ABOi heart transplantation

ABOi Donor heart Blood transfusion

recipient
Plasma product/​ Red cell concentrate
platelet
O AB AB O
A AB or A
B AB or B
A AB AB O or A
B
B AB AB O or B
A
AB AB, A, B, AB AB, A, B, or O
or O

Human leucocyte antigen

HLA is a cell surface protein that differentiates between self and non-​self.
• MHC (major histocompatibility complex) is the region on the genome
that encodes HLA. In human, the MHC (genes and proteins) is also
called HLA and is located on chromosome 6 (>200 genes).
• Most studied HLA genes are A, B, C and D (DP, DQ, DR) and divided
into: Class I MHC (A, B, C genes) is expressed on all nucleated cells
(HLA-​A, HLA-​B, and HLA-​C); Class II MHC (D genes) is usually
expressed by antigen-​presenting cells (B-​cells, macrophages, dendritic
cells, etc.).
Sensitized candidates (defined as PRA >10%)
These have preformed antibodies against specific donor HLAs (DSAs):
• Candidate’s serum is added in wells that contain lymphocyte cells from
random population samples: anti-​HLA antibodies cause cell lysis. PRA
represents the proportion of wells with cell lysis (lymphocytotoxicity
assay).
• The initial PRA test screens for sensitized candidates and how often
their sera will cross-​react with the population. It does not specify which
type of anti-​HLA.
Sensitized paediatric patients formed about one-​quarter of the waiting list,
with higher-​risk morbidity and mortality both while on the waiting list and
post transplantation due to longer waiting times, higher rate of acute allo-
graft rejection and vasculopathy, and reduced survival post transplantation.
However, HLA sensitization does not preclude transplant. If PRA is greater
than 10%, the next steps involve:
• More specific newer assays to assess the type of DSA and its strength
of binding (MFI).
• Deciding the presence of which HLA(s) on the donor will be
unacceptable (certain DSAs are more detrimental than others, e.g. anti
HLA-​DQ).
296

296 Chapter 22.3 Paediatric heart transplant

• Determining the calculated PRA: an estimate of the percentage of

organ donors with unacceptable HLAs. A calculated PRA of 70% implies
only 30% of donors will be compatible. Calculated PRA is computed
based on national data of HLA frequencies.
• Interval retesting: antibody levels can fluctuate over time and after
sensitizing events (e.g. transfusion). Also retest if negative PRA if risk
for sensitization (e.g. on VAD).
• Specific antibodies assay allows virtual cross-​matching with donor HLA
typing (versus prospective cross-​matching).
• Consider desensitization therapy (plasmapheresis, IVIG, rituximab) on
waiting list for high calculated PRA (threshold is institutional based, e.g.
>50%); consider lower threshold for more urgent patients.
• ‘Wait’ or ‘Take’ of a donor heart offer for a sensitized candidate is
weighed against clinical urgency: (1) in non-​urgent patients, await
negative virtual cross-​match; and (2) in urgent patients, consider ‘lower-​
risk’ positive cross-​matching (based on type of DSA and its MFI).
• At the time of donor offer for sensitized patients, discuss with HLA lab
expert (whether virtual or prospective cross-​match). If cross-​match is
positive and organ is accepted: perform plasmapheresis and/​or whole
blood exchange on CPB and intraoperative immunosuppressive agents
(methyl-​prednisolone pre/​post bypass and eculizumab post clamp
removal).
Paediatric candidates are at higher risk of sensitization. Humoral sensitiza-
tion occurs after exposure to non-​self antigen such as:
• Blood or platelet transfusion.
• Allograft material (i.e. homograft valve, patch, conduit).
• Previous organ transplant.
Risk groups include Fontan patients, older children, congenital heart disease
and multiple prior surgery, and those with VADs.
New DSAs can also form de novo post transplantation. Persistent de novo
DSA can contribute to chronic rejection and allograft vasculopathy.
Pulmonary hypertension
Evaluation of PVR is important in paediatric populations. PVR is a challen-
ging issue in paediatric population due to (1) higher risk of PH in paedi-
atric candidates, (2) challenges in assessment of PVR index (e.g. in Fontan
patients), and (3) complex interaction between complex congenital heart
disease and pulmonary artery pressure.
The paediatric candidates, especially in a congenital heart disease popu-
lation, are at higher risk of developing PH due to a number of risk factors:
• Unprotected pulmonary vascular bed: single ventricle with unrestrictive
pulmonary blood flow; large left-​to-​right shunt, such as in complete AV
septal defect; pulmonary blood flow at systemic pressure (e.g. common
arterial trunk/​hemitruncus).
• Chronic left atrial hypertension: left-​sided obstructive lesion such as
congenital aortic or mitral valve disease, multilevel left heart obstruction
(e.g. Shone’s complex), borderline left heart structure with endocardial
fibroelastosis.
• Additional source of high pulmonary blood flow: major aortopulmonary
collaterals.
 Special considerations 297

• Failing Fontan patients.

• Other comorbidities that contribute to PH: prematurity, chronic lung
disease, chromosomal anomalies (e.g. trisomy 21).
Irreversible PH is unusual in the paediatric age group as there will have been
some form of intervention; nevertheless, children may present late without
prior diagnosis. Inappropriate pursuit of a biventricular strategy in border-
line left heart may also contribute to development of PH. Eisenmenger syn-
drome implies fixed PH with reversal of a large intracardiac shunt usually
late in adulthood and development of central cyanosis—​heart transplant-
ation is precluded and heart–​lung transplantation is required.
Cardiac catheterization is required prior to transplantation to assess
PVR. However, paediatric PVR index assessment is more challenging due
to limited peripheral vascular access with occluded vessels, more difficult
assessment with central/​BT shunt, or cavopulmonary shunt.
The assessment of PVR index and TPG is particularly difficult and un-
reliable in failing Fontan due to low pulmonary flow and loss of hydro-
dynamic energy. There may be associated intrapulmonary arteriovenous
malformations or microthrombi which further complicate true PVR index
assessment.
Assessment of PH incudes (1) PVR: (MPAP –​PCWP)/​ Qp
(Qp =​calculated pulmonary blood flow) and (2) transpulmonary gradient
(TPG =​MPAP –​PCWP).
Criteria for heart transplantation
• PVR index less than 6 Wood units (WU).m2
• A 50% or greater decrease in PVR index after pulmonary vasodilator.
• MPAP less than 25 mmHg.
• TPG less than 15 mmHg (mPAP –​PCWP).
High-​risk transplantation with PVR index above the traditional criteria may be
offered on institutional basis. A PVR index less than 9 WU.m2 was not as-
sociated with increased post-​transplant mortality in recent paediatric study.
Failing univentricular and Fontan physiology
Congenital heart disease with univentricular physiology represents a di-
verse group of patients with heterogeneous morphology and physiology,
at different stages of palliation. Their physiology can differ in terms of pul-
monary blood flow, atrial pressure, and volume loading which can lead to
chronic elevation of pulmonary arterial or venous pressure and ventricular
dysfunction. Primary transplantation as a therapeutic option for congenital
heart disease is uncommon and most commonly, they present after failed
surgical palliation. Univentricle is the most common diagnosis among con-
genital heart disease patients who need transplantation.
Waiting list mortality is the highest after first-​ stage palliation with
systemic–​PA shunt; consideration should be given to progress them to
second-​stage palliation with Glenn shunt. The Fontan procedure com-
pletes the cavopulmonary connection, creating a total non-​pulsatile sys-
temic venous flow return to the pulmonary vascular bed. The original
atriopulmonary connection has undergone a number of evolutions in
298

298 Chapter 22.3 Paediatric heart transplant

the last four decades, and currently the most common modification is an
extracardiac conduit followed by lateral atrial tunnel.
Mechanisms of Fontan failure
Fontan patients represents a unique group of patients needing transplant,
as patients can present with (1) pump failure, and/​or (2) failure of ‘Fontan
physiology’. Fontan circulation is non-​physiological, as there is decoupling
between RV and pulmonary vascular bed with loss of pulsatile phasic flow
which results in chronic elevation of systemic venous pressure, pulmonary
endothelial dysfunction, ventricular unloading, and low cardiac output.
Unfavourable haemodynamics pre Fontan or trigger factors post Fontan
such as arrhythmia, further aggravate and accelerate the decline in Fontan
circulation.
Therefore, systemic ventricular dysfunction and progressive AV valve re-
gurgitation are suboptimal high-​risk substrate to construct a Fontan circuit.
With higher risk of early failure after Fontan completion and worse post-​
transplant survival in Fontan patients (versus Glenn), serious consideration
should be given for transplantation prior to stage two palliation.
Early Fontan failure
The contemporary surgical outcome has vastly surpassed that of
atriopulmonary connection; as such, rescue cardiac transplant in the set-
ting of early Fontan failure is uncommon. Early mortality (<30%) is low
(approximately 1.0–​1.6%); nevertheless, early Fontan failure that includes
mortality, takedown, revision/​re-​intervention, or ECMO/​VAD can occur
in the region of 3–​6%.
Failure to wean ventilation and inotropic support after Fontan operation
with impaired systemic ventricular function or recurrent arrhythmia des-
pite optimal medical therapy are primary reasons for referral to transplant-
ation. Another rescue option is Fontan takedown, if ventricular function is
permissible—​the early mortality of takedown is high (approximately 25%).
A timely rescue cardiac transplant in the setting of early Fontan failure is a
more superior long-​term option.
Late Fontan failure
Attrition continues with time in all Fontan patients. Those with
atriopulmonary connection and hypoplastic left heart syndrome are at the
highest risk. Late failure can present as declining functional capacity (NYHA
III/​IV), HF, PLE, cyanosis, and plastic bronchitis. Freedom from late Fontan
failure at 20 years is 70%. Some 5% of paediatric Fontan patients (<18 years
old) require heart transplantation. Mean age at transplantation ranges from
8 to 15 years. Patients may remain stable for a long time before acute de-
compensation, often triggered by loss of sinus rhythm.
Cardiac catheterization
This is an essential pretransplant assessment but precise evaluation is dif-
ficult in the Fontan group. This is due to significant dysfunction despite
normal EDP due to inadequate volume loading; underestimation of any
gradient along the Fontan pathway due to low cardiac output; and unreli-
able measurement of pulmonary blood flow and PVR index (non-​pulsatile,
non-​phasic flow, systemic venous obstruction, presence of other sources of
pulmonary blood flow including minor and major collaterals, arteriovenous
malformations, chronic microemboli).
 Special considerations 299

Specific challenges in transplantation for Fontan patients

Only some 4–​7% of Fontan patients ever receive a transplant within
20 years of surgery; undoubtedly, they are among the most challenging and
highest-​risk transplantation candidates. The following is a list of factors as-
sociated with the difficulties in undertaking transplantation in this group:
• Under-​recognized PH due to difficulty in assessing PVR index in pre-​
transplant catheter.
• High transplant mortality risk due to comorbidities: severe
malnourishment associated with PLE, liver disease, coagulopathy, renal
dysfunction—​among transplants for congenital heart disease, Fontan
patients had an eightfold increase in mortality.
• Technical complexities: multiple prior operations, collaterals and
bleeding, extensive PA reconstruction, longer bypass and donor
ischaemia time.
• High risk of sensitization, need of cross-​matching, limited donor pool;
higher risk of acute rejection.
• Under-​prioritization on waiting list.
• Associated liver cirrhosis requires careful hepatic assessment and heart
transplantation alone may not be adequate.
Due to these challenges, early listing should be considered for Fontan pa-
tients. Prompt listing should be considered in the setting of early Fontan
failure.
Protein-​losing enteropathy and plastic bronchitis
PLE is an indication in 20–​40% of Fontan patients receiving heart trans-
plantation, hence it needs special mention. Chronic systemic venous and
lymphatic congestion in Fontan patients leads to lymphatic channels rupture
into the GI lumen and leak of serum protein. However, the mechanism of
PLE and why some 5–​15% of Fontan patients develop PLE while others do
not remain to be understood, and a link between level of systemic venous
pressure and the development of PLE has not been demonstrated. PLE can
develop as early as 2 months post Fontan but usually occurs years later.
Severe hypoalbuminemia is the main issue, and diagnosis is made by low
serum albumin (<3.5 g/​dL or total protein <5.5 g/​dL), and elevated alpha-​
1-​antitrypsin in stool samples. PLE is suspected when Fontan patients pre-
sent with diarrhoea, oedema leading to weight gain, abdominal distension,
and occasionally GI bleeding. Pleural effusion can be present resulting in
breathlessness.
Severe hypoproteinaemia leads to severe malnutrition and immune com-
promise prior to transplantation with low immunoglobulin levels (IgG, IgA,
IgM) and CD4 counts. PLE, without any intervention, carries a risk of in-
fection with mortality as high as 49% after 5 years of diagnosis. This is also
reflected by high waiting list mortality, which is almost twice as high with PLE
compared to Fontan patients without PLE (21 versus 12%).
However, post-​ transplant survival is comparable to Fontan patients
without PLE. Transplantation is considered for PLE refractory to standard
medical therapy (including diet with high protein, medium-​chain triglycer-
ides, diuretics, heparin, steroid, bowel rest, total parenteral nutrition) and
fenestration of Fontan pathway. Enteric-​coated steroid improves PLE and
30

300 Chapter 22.3 Paediatric heart transplant

outcome. Heart transplantation remains the most effective treatment for

PLE, which provides a 100% long-​term cure in all transplantation survivors.
Plastic bronchitis
This is much rarer than PLE. The hallmark of the disease is the forma-
tion of thick, mucofibrinous casts in the tracheobronchial tree, which is
expectorated on coughing or visualized on bronchoscopy. The develop-
ment of plastic bronchitis is inadequately understood; like PLE, it reflects
a sequela of chronic elevation of systemic venous pressure and low car-
diac output in Fontan physiology, with contribution from inflammatory
process. Symptoms range from mild cough and breathlessness to severe
respiratory distress when the airway is severely obstructed. Plastic bron-
chitis has a high mortality of 30% and life-​threatening airway obstruction
requires emergency clearance with rigid bronchoscopy. Medical therapies
include high-​molecular-​weight heparin, bronchodilator, tissue plasminogen
activator, sildenafil, and steroid. Catheter or surgical intervention including
atrial pacing, thoracic duct ligation, fenestration of Fontan, and conversion
of Fontan has resulted in variable degrees of resolution of plastic bron-
chitis. Unlike PLE, plastic bronchitis is a rare indication for transplantation in
Fontan, perhaps reflecting its much lower incidence. Transplantation mor-
tality is higher, but a new heart results in early and durable resolution of
plastic bronchitis in all transplantation survivors, with comparable long-​term
survival with non-​Fontan patients.
Ex vivo cardiac perfusion
Ex vivo cardiac perfusion is an organ preservation system which keeps the
donor heart perfused in a warm beating state after procurement (Organ
Care System (OCS) Heart, TransMedics, Andover, MA, USA). Currently,
the system is suitable to support bigger donor hearts from older teenagers
onwards due to instrumentation size. Normothermic donor heart perfu-
sion has been shown to be non-​inferior to standard cold storage and allows
(1) extension of cold ischaemia time, which is relevant in the context of
transplantation for failing Fontan and other complex congenital heart dis-
ease; and (2) assessment of marginal donor heart and DCD heart.
Donation after circulatory death
The first successful DCD heart transplantation was carried out in paediatric
population in 2004 (DN Campbell, Denver). Despite a successful series of
three paediatric patients in Denver, DCD transplantation did not take off
widely until in recent years in adult heart transplantation. Far fewer DCD
procurements were reported in paediatric registries. The long-​term post-​
transplant survival with DCD hearts remains to be defined. The use of
DCD hearts could help increase the donor pool for paediatric candidates.
 Section 5

Special considerations
in heart
transplantation

23 Adult patients with congenital heart disease 303

302
 Chapter 23 303

Adult patients
with congenital
heart disease
Overview 304
Outcomes of transplantation 304
Waiting list criteria and outcomes 305
Common aetiologies 306
Indications for transplant assessment 307
Indications for transplantation 308
Contraindications and special considerations 308
Pulmonary artery reconstruction 312
Transplant assessment 312
304

304 Chapter 23 Adult patients with congenital heart disease

Overview
The prevalence of all congenital heart disease (CHD) is 0.5–​0.8% of live
births. With drastic improvements in management over the past four dec-
ades, particularly palliative procedures of the more complex congenital
anomalies, these patients are now surviving into adulthood and presenting
with late complications, including HF. A recent population study suggested
approximately 1000 adults are living in the UK (population 66 million) with
single-​ventricle physiology. Many of these patients will require advanced
failure management, including transplantation.
HF in adult congenital heart disease (ACHD) is frequently due to struc-
tural abnormalities and as such is less amenable than acquired pathologies
to pharmacological preservation of the ventricle. An increasing number of
young adult patients are presenting to the advanced HF and transplant team
and CHD now accounts for approximately 3% of adult cardiac transplant-
ation, with the number of ACHD recipients having increased by 40% over
the past two decades.
This expanding group brings unique challenges in the shape of highly
complex surgical histories, varied physiology, unusual anatomy, and limited
mechanical support options. The worldwide literature reports that cardiac
transplantation (and, at times, multiorgan transplantation) is an effective
strategy to improve survival and quality of life in a carefully selected group
of ACHD patients. These patients carry a higher 1-​year postoperative
mortality than other indications for transplantation. The success of trans-
plantation for the ACHD group requires multidisciplinary involvement of
transplant and congenital heart surgeons, advanced HF and ACHD cardi-
ologists, hepatologists, intensivists, and allied professionals in the transplant
assessment and perioperative clinical care.
Considering the rarity of such cases and anatomical challenges, early re-
ferral to an advanced HF team for assessment, and ultimately heart trans-
plantation must be undertaken in a centre with an adequate volume of
ACHD practice.

Outcomes of transplantation
ACHD recipients carry a high mortality risk in the first year following
transplantation. A diagnosis of ACHD confers additional risk with post-​
transplant 1-​year mortality being 16–​19% in ACHD recipients versus 6–​9%
in adults with other diagnoses. Higher incidence of primary graft failure,
renal failure, multiorgan failure, and stroke in ACHD patients contributes
to early mortality. In-​hospital mortality is even higher for those with single-​
ventricle physiology (8% in biventricular vs 23% in single-​ventricle ACHD
patients).
Prognosis for those who survive the first year is favourable, with sig-
nificantly better long-​term survival than all other pathologies. The median
survival conditioned to 1-​year survival is better for ACHD than any other
pretransplant diagnoses.
 Waiting list criteria and outcomes 305

Factors that may contribute to poor early results in ACHD patients include:
• Increased risk of acute rejection due to sensitization from previous
blood product transfusion and use of homograft materials in palliative
procedures.
• Longer graft ischaemia and CPB times. This is related to factors
associated with multiple previous sternotomies, structural anomalies
requiring reconstruction, and challenges of haemostasis due to
coagulopathy and surgical bleeding.
• Concomitant organ failure—​commonly hepatic involvement in the
Fontan patient which contributes to turbulent postoperative course in
single organ transplant or may require heart–​liver transplantation.
• Increased risk of infection due to protein-​losing enteropathy.
• PH—​this can be challenging to diagnose preoperatively, resulting in RV
compromise in the donor heart or the need for higher-​risk heart–​lung
transplantation. PVR >4 Wood units is an independent risk factor for
death post transplantation in ACHD recipients.
• Centre volume is associated with post-​transplant survival in patients
undergoing ACHD transplantation. Low-​volume centres perform
poorly compared to high-​volume centres.

Waiting list criteria and outcomes

Waiting list criteria
ACHD patients may not fit the routine criteria for transplantation despite
their deteriorating condition.
Further consideration must be given to additional criteria specific to the
failing ACHD patient:
• Refractory arrhythmia (more than one hospital admission over last
3 months with haemodynamic instability or associated with kidney or
liver dysfunction).
• ACHD patients with no option for conventional escalation of therapy
who are unsuitable for inotropes and/​or VAD with one of the following:
• Bilirubin and transaminases greater than twice normal levels.
• Deteriorating renal function (eGFR <50 mL/​min/​1.73 m2, or 20%
reduction from baseline).
• Requirement for dialysis/​continuous venovenous haemofiltration for
fluid or electrolyte management.
• Recurrent admissions (greater than three in preceding 3 months) with
episodes of right HF or protein-​losing enteropathy requiring ascites
drainage.
Waiting list mortality
Waiting list mortality is similar for ACHD and non-​ACHD, although death
from HF and sudden cardiac death is more likely in ACHD. Despite this,
there is a lower use of ICDs and MCS. While on the waiting list, close clin-
ical observation should highlight patients who could benefit from an ICD
given the prevalence of sudden death.
The use of MCS has previously been reported as a risk factor for in-
creased waiting list mortality in ACHD patients. This, along with the tech-
nical challenges of device implantation in the congenital heart has deterred
306

306 Chapter 23 Adult patients with congenital heart disease

the use of MCS as a bridge to transplant. A more contemporary evaluation

of the INTERMACS database of 126 ACHD patients supported with mech-
anical circulatory devices on the waiting list reports a positive outcome for
71% at 6 months (21% successfully bridged to transplantation and 50% re-
main alive on VAD). Use of MCS is increasing in ACHD and consideration
should be given to device employment for waiting list patients anticipated
to have an extended wait or clinical deterioration. Recent reports describe
individual ACHD cases in which VA ECMO, axillary IABP, and durable and
temporary VADs have been successfully employed for bridging.
Time on the waiting list
ACHD patients wait longer on the list than non-​ CHD HF patients.
Suggested reasons for this include:
• High prevalence of elevated panel reactive antibodies, limiting the
donor pool and requiring a negative prospective cross-​match.
• The perceived need to find an ideal donor with adequate up-​sizing to
tolerate occult PH.
• The desire to reduce ischaemic time by limiting distance of donor
organ transport. This is overcome in some UK centres with the use
of the Organ Care System (OCS) Heart (Transmedics, Andover, MA,
USA) which allows transport with a beating heart and can limit warm
ischaemic time.
• A higher proportion of ACHD patients will require heart–​lung or
heart–​liver transplantation.
• Waiting list status criteria may be more appropriate and designed for
non-​ACHD patients.

Common aetiologies
Underlying pathologies in ACHD vary greatly on the spectrum of com-
plexity and diagnostics are rarely straightforward requiring description of
situs, cardiac orientation, atrial arrangement, intracardiac anomalies, ven-
tricular and valve arrangements, and any shunt defects.
Two distinct groups exist: those with single-​ventricle physiology (i.e.
Fontan circulation) and those with two ventricles.
For simplicity, common aetiologies found in ACHD patients undergoing
heart transplantation have been divided into three broad groups according
to the number of functional ventricles.
Biventricular
• Dextro-​transposition of great arteries (d-​TGA) patients with corrective
surgery undertaken before the era of arterial switch operation (before
the 1990s). These patients had atrial inversion procedures (Senning or
Mustard) in childhood. This configuration results in a sub-​systemic right
morphological ventricle, which is less robust than the morphological LV
resulting in early sub-​systemic HF.
• Congenitally corrected transposition of great arteries (ccTGA). This
configuration results in a sub-​systemic ventricle of right morphology,
with a tricuspid valve as the sub-​systemic AV valve but with normal
sequential pulmonary and systemic systems. The morphology of
 Indications for transplant assessment 307

the RV and valve in the high pressure left-​sided system leads to early
failure. These patients may not have undergone surgery, or may have
undergone extensive corrective surgery or palliation including the
Fontan operation.
• Eisenmenger’s syndrome resulting from any uncorrected shunt
defect (ASD, VSD, or rarely patent ductus arteriosus). Rarely seen
in the current era of transplantation due to improved detection and
management, and require combined heart–​lung transplant.
• Ebstein’s anomaly of the tricuspid valve.
Univentricular
In the adult population, these patients will have undergone palliation staging
to achieve a Fontan circulation. Previous surgery is ubiquitous and extensive
(see pp. 296–297).
• Hypoplastic left heart syndrome.
• Double outlet RV.
• Double inlet LV.
• Tricuspid or mitral atresia.
Can be either univentricular or biventricular
• Complete atrioventricular septal defect (CAVSD). Number of ventricles
and subsequent management depends on whether balanced or
unbalanced.

Indications for transplant assessment

Consider earlier discussion with transplant services to allow time for
serial assessment. These patients are highly unique in anatomy and treat-
ment history and do not make for a straightforward assessment or listing
decision.
Transplant assessment for heart failure
Recognizing HF in the ACHD patient is imperative to permit early re-
ferral before complications occur which elevate risk, or contraindicate
transplantation. Patients may have lived with the classic symptoms of HF
for many years. Baseline expectations, younger age, and lifelong adap-
tations may mask deterioration of normal parameters and symptom-
atology. Profound circulatory dysfunction may exist before patients seek
help. With this in mind, clinicians must interrogate closely any ‘hidden’
functional changes which may reflect underlying declining function. To
avoid late detection, it may be prudent to undertake frequent and regular
haemodynamic assessment via catheterization or CPX in asymptomatic
patients.
Transplant assessment for other reasons
Unlike most acquired HF, cardiac transplantation may be considered
in ACHD with preserved ventricular function due to the idiosyncratic
presentations related to Fontan circulation; patients with debilitating
protein-​losing enteropathy or plastic bronchitis should be referred for
discussion.
308

308 Chapter 23 Adult patients with congenital heart disease

Transplantation or further intervention

Atriopulmonary connection Fontan patients with refractory arrhythmia
may benefit from Fontan conversion. Similarly, those with sub-​systemic AV
valve regurgitation may benefit from valve replacement. When such po-
tential interventions exist, the decision between further corrective surgery
or listing for transplantation should be undertaken as part of a transplant
assessment and multidisciplinary team discussion between congenital and
transplant services.

Indications for transplantation

Indications with impaired ventricular function
• Patients with stage D HF refractory to therapy who will not benefit
from surgical, interventional, or electrophysiological intervention.
• Patients with stage C HF associated with reactive PH and potential risk
of developing fixed irreversible elevation of PVR that could preclude
OHT in the future.
• Patients with stage C HF and severe limitation of exercise and activity.
Indications with preserved ventricular function
• Patients with associated near-​sudden death or life-​threatening
arrhythmias refractory to all therapeutic modalities.
• Patients with symptomatic cyanosis that is not amenable to surgical
correction.
• Patients with protein-​losing enteropathy/​plastic bronchitis despite
optimal medical and surgical therapy.

Contraindications and special

considerations
Traditional contraindications to cardiac transplantation apply to ACHD
patients. Additionally, there are ACHD-​specific considerations which may
be a single absolute contraindication or, more commonly, multiple relative
contraindications amass to collectively elevate risk and render a patient
untransplantable.
Pulmonary hypertension
• PH is common in this group and is aetiologically different from non-​
CHD patients. Elevated left heart pressure may be causative, but
pulmonary venous irregularities and other left-​sided mechanical
obstruction, cyanosis, pulmonary over-​circulation via collateral flow,
shunting, and intrinsic lung disease are frequently present. Patients with
established PH require heart and lung transplantation.
• ACHD patients require cautious evaluation and interpretation of
pulmonary haemodynamic data including transpulmonary gradient.
The quantitative diagnosis is not straightforward due to presence of
shunts and collateral flow. RV pressures may be high due to outflow
obstruction mimicking a PH picture.
 Contraindications and special considerations 309

• In addition, the low cardiac output underestimates PVR. Occult PH

in these patients can be unmasked following transplantation, with
subsequent RV failure of the implanted heart.
• Given the clear challenges in diagnostics, attention should be paid to
history including pattern of disease, duration of unprotected pulmonary
circulation in infancy, and any extended periods of cyanosis. Qualitative
and quantitative data in combination will permit a better understanding
of individual situations.
• Pharmacological reversibility of PH is an incremental risk factor but does
not preclude cardiac transplantation. Preoperative risk management with
pulmonary vasodilators and anticipatory measures of donor RV support,
including inhaled NO, milrinone, and RVAD, should be considered.
• In some instances, mechanical unloading of the systemic ventricle may
reduce PVR and should be considered.
High levels of sensitization
• ACHD patients presenting for transplant assessment have commonly
undergone multiple surgeries with exposure to blood products and
homograft material resulting in a high prevalence of sensitization.
Strategies to minimize exposure and subsequent sensitization are limited:
• High PRA level >10% is associated with poorer outcomes post
transplantation.
• Anti-​HLA antibodies are associated with worse long-​term graft
survival.
• Presence of DSAs is associated with cell-​mediated and antibody-​
mediated rejection and graft vasculopathy.
• Desensitization therapy can be considered in patients with calculated
PRA levels >50% although absolute values for initiating desensitization
prior to transplantation are unit dependent. Desensitization strategies
aim to mitigate risk by perioperative reduction of circulating antibodies
and cessation of production. Such deletion of immunological
memory is a multi-​therapy approach with variable reported success:
plasmapheresis, IVIG, anti-​B-​cell therapies (rituximab), and proteasome
inhibitors (bortezomib).
The single-​ventricle patient
The single-​ventricle patient presents an idiosyncratic set of challenges in
peri-​transplant management. Patients will likely have undergone pallia-
tive staging to reach partial or complete Fontan physiology (Fig. 23.1 and
Fig. 23.2).
The Fontan circulation results in non-​pulsatile venous return to the PAs
via cavopulmonary connections with subsequent adverse physiological
consequences:
• High venous pressures leading to liver cirrhosis, protein-​losing
enteropathy, and plastic bronchitis. These idiosyncratic pathologies can
persist long after transplantation and significantly complicate recovery.
• The reduction of hepatic flow to one lung or both lungs in
cavopulmonary arrangements results in pulmonary arteriovenous
malformations and subsequent chronic cyanosis which can persist
postoperatively necessitating angiographic coiling.
031

310 Chapter 23 Adult patients with congenital heart disease

• Chronic cyanosis leads to aortopulmonary collaterals.

• The assessment of pulmonary pressures can be hampered by non-​
pulsatility, collateral flow, and arteriovenous malformations. RV
dysfunction in the donor heart is commonly encountered in Fontan
patients as high pulmonary pressures are unmasked.
Poorer outcomes are reported in Fontan patients undergoing transplant-
ation, likely due to the additive consequences of prolonged and challenging
surgery, longer ischaemic times, and extensive comorbid disease: liver dis-
ease, coagulopathy, occult PH, immune dysfunction, renal impairment, and
hypoalbuminemia.
Fontan patients presenting for transplantation may be in the unusual
position of preserved ventricular function, but be classified as Fontan
failure due to protein-​losing enteropathy, plastic bronchitis, cyanosis,
or intractable arrhythmia. In the presence of poor ventricular func-
tion, transplantation may be the only option, but further repair is an at-
tractive option for those with relatively preserved ventricular function in
whom failure can be attributed to residual structural defect such as valve
regurgitation.
Consideration should be given to Fontan conversion for those patients
with poor function and arrhythmia in the setting of the now outdated
atriopulmonary connection as good results are reported as an alternative
to transplantation.

BDG
anastomosis

SVC Ao/
neoaorta LPA

LA
d o
blo

Mixing

Fig. 23.1 Partial Fontan circulation. The SVC is anastomosed to the right PA as a
bidirectional Glenn shunt (BDG) and provides pulmonary flow. Deoxygenated IVC
return and oxygenated pulmonary venous return mix in the common atrium and
provide systemic flow via the single ventricle.
 Contraindications and special considerations 311

Fontan
anastomosis

SVC
PA

LA

Fig. 23.2 Complete Fontan circulation. In addition to the bidirectional Glenn

shunt demonstrated in Fig. 23.1, the IVC is anastomosed to the PA as shown.
A fenestration remains between the Fontan conduit and the RA to allow offloading.
All deoxygenated venous return is now directed to pulmonary flow in a non-​pulsatile
passive system.

Surgical risk
• ACHD patients frequently have an extensive surgical history with
multiple sternotomies. Re-​entry to the chest requires planning
to allow safe access. Peripheral CPB may be needed prior to
sternotomy, or indeed during, in the event of unmanageable injury
to a great vessel or ventricle. The extra time for re-​entry must be
considered to minimize ischaemic time. Ex vivo perfusion of the
donor organ may be an important strategy to limit total ischaemic
time (see Chapter 14).
• Reconstruction, particularly of the PA, is commonly required to
permit implantation (see Pulmonary artery reconstruction, p. 314).
Reconstruction of the PA at the time of transplantation may be
associated with higher mortality in Fontan patients. Patients with
heterotaxy or situs inversus may need baffling of left-​sided systemic
venous return to the donor RA necessitating longer IVC/​SVC
harvesting. With removal of baffles and caval connections, limited
native tissue remains and donor tissue or xenograft pericardium may
be required. This adds time to the preimplantation stage. The retrieval
team need to be aware if extra tissue is required.
• Aortopulmonary collaterals occur secondary to chronic cyanosis.
When damaged, these cause extensive haemorrhage. The increased
venous return to the LA can render the surgical field challenging when
commencing implantation and maintaining adequate systemic pressures
231

312 Chapter 23 Adult patients with congenital heart disease

on bypass in the presence of large collaterals can be challenging.

Consideration should be given to coil occlusion pre transplantation or in
the postoperative period if high-​output cardiac failure persists.
All of the above factors result in prolongation of ischaemic and operative
times. Acute graft failure remains the commonest cause of early mortality
in these patients.

Pulmonary artery reconstruction

Single-​ventricle patients commonly require great vessel reconstruction at
the time of implantation, most frequently the PAs. Extent of reconstruction
is dependent upon whether there is complete discontinuation of the PAs or
partial absence of the vessel wall following takedown of the cavopulmonary
connections and removal of stents.
Choice of material includes on-​the-​shelf materials such as Gore-​Tex,
polytetrafluoroethylene, or bovine pericardium or jugular vein, or donor
tissue: PAs, aorta, and pericardium.
• The benefit of on-​the-​shelf materials is that reconstruction can occur
prior to organ arrival to limit ischaemic time.
• The benefit of donor materials is better haemostasis which may be
preferable in these patients with high risk for coagulopathy and bleeding.
The surgeon must be aware of the anatomic variety, reconstruction mater-
ials available, and technique options.
Partial PA reconstruction
Limited reconstruction is required following take down of the
cavopulmonary anastomoses. A single-​or double-​patch repair can be per-
formed to complete PA continuity (Fig. 23.3).
Complete PA reconstruction
Hilum-​to-​hilum reconstruction is required. This can occur due to PA stents
in situ requiring removal of the majority of vessel (Fig. 23.4a), or less fre-
quently with previous intentional surgical discontinuation due to classical
and antiquated Glenn and Fontan arrangements.
Hilum-​to-​hilum reconstruction is possible utilizing Gore-​Tex conduits,
donor PAs, or donor descending aorta. The conduit is fashioned to anasto-
mose to both PAs in an interpositional end-​to-​end technique and an aper-
ture created to allow anastomosis with the graft main PA (Fig. 23.4b–​d).

Transplant assessment
ACHD patients required the follow additional considerations to the routine
transplant assessment.
History and examination
• Cardiac diagnosis: close attention to anatomy and physiology.
• Symptoms and signs of associated liver, kidney, and pulmonary
disease.
 Transplant assessment 313

(a) (b)

(c)

Fig. 23.3 Series demonstrates an extracardiac Fontan and bidirectional Glenn

configuration (a), which is frequently encountered in single-​ventricle patients
presenting for transplantation. The cavopulmonary anastomoses are taken down, and
a two-​patch technique is used to reconstruct the PAs (b). Cannulas are seen in the
IVC and SVC to allow bicaval venous drainage for CPB. An aperture is then opened
into the patch to allow the donor heart to be implanted (c). LPA, left pulmonary
artery; RPA, right pulmonary artery.

• Current status: dyspnoea, palpitations, exercise tolerance, ascites,

changes in bowel habit (protein-​losing enteropathy), weight loss or
gain, peripheral oedema. Temporal change in symptoms and exercise
tolerance since last review. Pay close attention to masked symptoms
(see Indications for transplantation, p. 268).
• Surgical history: number of sternotomies, intra-​and extracardiac
catheter and surgical interventions, baffle stenting, previous use of groin
vessels, and history of MCS. Increasing number of sternotomies has
been associated with higher risk of mortality.
• Social history: establish support network.
431

314 Chapter 23 Adult patients with congenital heart disease

(a) (b)

(c)
(d)

Fig. 23.4 Series demonstrates management of PA discontinuation due to the

presence of a hilum-​to-​hilum stent within the PA in the context of complete Fontan
circulation (a). Stent removal at time of transplantation results in PA disruption
leaving only partial back wall of the vessel (b). Cannulas are seen in the IVC and
SVC to allow bicaval venous drainage for CPB. Following removal, hilum-​to-​hilum
reconstruction is performed in this case using a Gore-​Tex conduit (c). An aperture
is then created in the conduit to allow the donor heart to be implanted (d). Native
aorta is shown with cross-​clamp in situ. Note the donor PA is trimmed in a bevelled
fashion in order to avoid kinking or distortion.

Investigations
To establish criteria fulfilment, planning, and feasibility for transplantation. In
addition, investigations should evaluate the presence of residual lesions or
repairable structural anomalies common in ACHD which are contributing
to the HF picture. Management of such lesions can alter the clinical course
and potentially extend time to transplantation.
 Transplant assessment 315

• Cardiopulmonary catheter data.

• CT: cardiac diagnosis and anatomy. Important to confirm adequacy
of pulmonary venous drainage which can be addressed at time of
transplantation. Examine proximity of vessels and heart to posterior
sternum in planning for re-​entry of the chest.
• Echocardiography/​MRI: assess structural pathologies which may be
amenable to repair or replacement. Confirm valve regurgitation, conduit
and baffle patency, and leak. Intervention on valves/​stenoses/​leaks may
be performed transcatheter and can improve symptoms and prolong
time to transplantation.
• ECG: consider ambulatory 24-​hour ECG if history of palpitations.
Electrophysiological studies may be appropriate to determine if
arrhythmia intervention is possible to extend time to transplantation.
• Venogram/​Doppler studies: patency of access points for CPB and
anastomoses.
• CPX: formal exercise testing.
• Liver assessment: liver disease is common due to high venous pressures
and significantly increases morbidity and post-​transplant mortality. A
MELD-​XI score should be calculated and discussion with hepatologists
as part of the multidisciplinary team. Some centres may consider liver
disease as a contraindication for heart transplantation, while others will
consider heart–​liver transplantation.
• Transient elastography (Fibroscan) for cirrhosis.
• CT for monitoring of lesions (hepatocellular carcinoma) with or
without biopsy.
• Liver biochemistry and synthetic function is a late indicator of liver
compromise.
• Hepatitis C: ACHD patients have a higher prevalence of hepatitis C
due to transfusion prior to routine screening in 1992.
• Hepatic venous wedge pressure.

Perioperative considerations
• Surgical planning is critical. Perusal of investigations and previous
operative notes is essential to avoid unexpected anatomy. Close liaison
with the donor team is critical to reduce ischaemic period.
• A surgical team including both congenital and adult surgeons may be
optimal.
• Ex vivo donor heart perfusion can mitigate the total ischaemic time
and has been used in cases requiring extensive reconstruction prior
to implant. One perfusion system is the OCS Heart. This system
provides normothermic blood perfusion at near physiological perfusion
parameters.
• Redo sternotomy precautions:
• External defibrillator pads.
• Ice on the head for cooling.
• Blood and heparin in the operating room with a primed circuit.
• CT imaging displayed in the operating room may be helpful.
• Carbon dioxide suffusion of operative field to reduce impact of air
embolism.
• Team aware of CPB plan in event of inadvertent thoracic injury.
631

316 Chapter 23 Adult patients with congenital heart disease

• Blood loss can be significant before CPB is established due to redo

surgery, aortopulmonary collaterals, and concomitant hepatic
dysfunction:
• Consider where access is possible due to previous surgery and which
vessels will be required for CPB.
• Concentrated recombinant factor replacement should be available.
• Unmasking of elevated pulmonary pressures can occur post
implantation.
• Routine use of pulmonary vasodilators in this group of
patients: inhaled NO/​isoprenaline/​milrinone.
• Early deployment of RV support if required.
• Hypotension and low systemic resistance are common during CPB,
especially in Fontan circulation due to vasoplegia. Use vasopressin
for low SVR. This can persist into the postoperative period requiring
vasoconstrictors including vasopressin, methylene blue, and vitamin B12
infusions.
 Section 6

Lung
transplantation

24 Patient selection criteria and bridging support for

lung transplantation 319
25 Lung donor selection and management 337
26 Lung procurement following donation after brain
death and donation after circulatory death 347
27 Ex vivo lung perfusion 355
28 Technical aspects of lung transplantation 371
29 Critical care management and primary graft
dysfunction following lung transplantation 383
30 Management of rejection following lung
transplantation 389
31 Late complications following lung transplantation 399
32 Antimicrobial prophylaxis and treatment after lung
transplantation 411
831
 Chapter 24 319

Patient selection criteria

and bridging support for
lung transplantation
Introduction 320
Indications for referral to a transplant centre 320
Indications for transplantation/​activation on the waiting list 326
Pre-​transplant evaluation 328
Clinical management after listing 331
Special considerations 332
320

320 Chapter 24 Lung transplant selection criteria

Introduction
Lung transplantation has been shown to prolong patient survival and im-
prove quality of life but is dependent upon appropriate timing of referral
and activation on the transplant waiting list to minimize pre-​transplant mor-
tality. In 2017, 2478 lung transplants were performed (1860 of these were
bilateral) in the USA across 26 centres contributing to 14,000 living lung
transplant patients in the US at that time. Conversely, the ISHLT registry
reported 4554 adult lung transplants performed in 2016 worldwide.
Recipients with restrictive lung disease (group D) comprised the highest
volume at 1407 (57%), followed by obstructive lung disease (group A,
27.1%), suppurative lung disease (group C, 10.7%), and PH (group B, 5.1%).
The timing of referral for transplantation, surgical plan, and when to ac-
tivate patients on the waiting list vary for these lung diseases as different
parameters within these disease groups are indicators of the appropriate
time to move forward with listing for transplant.

Indications for referral to a

transplant centre
General indications
It is important to understand the distinction between indications for re-
ferral for transplant and indications for transplant (i.e. to activate on the
waiting list and proceed with transplant surgery as soon as a suitable donor
is identified). Guidelines for lung transplant candidacy (>50% death within 2
years without a transplant, >80% likelihood of survival at least 90 days after
transplant, and >80% likelihood of survival for at least 5 years concerning
patients’ non-​pulmonary medical conditions) generally focus on listing and
not indications for referral. Patients with end-​stage lung disease generally
have a progressive disease with some degree of clinical deterioration over
time. Ideally, patients should be listed as soon as their lung disease becomes
life-​threatening and substantially affects their quality of life. However, as
their disease progresses, patients may become too deconditioned or too
sick to survive the lung transplantation process. In these instances, patients
may miss the opportunity to undergo a lung transplant. This period be-
tween the development of an indication for listing and progression to a
point where it may be too late to perform lung transplantation is referred
to as the ‘transplant window.’
Over time, early referral for transplant has become the recommended
approach to patients with end-​stage lung disease. Many patients who meet
criteria for referral will be too early for listing. However, it is essential to rec-
ognize that referral does not necessarily lead to a full evaluation or listing.
Instead, early referral gives individuals with end-​stage lung disease and
their care team access to transplant-​specific expertise of a lung-​transplant
centre. This allows for management and optimization of medical conditions,
patient education, improved physical conditioning (preferably in a formal
pulmonary rehabilitation setting), and solidification of psychosocial factors
within the variable rates of symptom progression across these disease
groups. Patients are typically surprised by the complexity and variety of the
 Indications for referral to a transplant centre 321

issues (both medical and non-​medical) involved in lung transplantation, and

early referral provides the opportunity for patients and their caregivers to
acclimate and learn what is required throughout their transplant journey.
When a patient is referred for transplant, a composite of testing (pul-
monary function testing (PFT), ambulatory and exercise oxygen require-
ments, arterial blood gas) is necessary to determine the severity of disease
as resting oxygen saturation and flow rate do not always adequately capture
the degree of pulmonary disease. Some patients may be more advanced
than originally appreciated while other patients initially felt to be in their
transplant window can prolong their time to transplant with tailored ther-
apies. The patient’s recent trajectory can also be an important predictor
of clinical demise. Early referral can save patient lives for those with rapid
deterioration, but even for those with a slower disease progression, early
referral has advantages to improve survival outcomes by allowing the trans-
plant team time to address the host of issues involved. Please refer to Box
24.1 for a list of contraindications to lung transplantation.

Box 24.1 Absolute contraindications to lung

transplantation
• Active or recent history of malignancy. Programmes vary as to an
acceptable duration of disease-​free interval and their willingness to
proceed with lung transplantation but the disease-​free interval typically
ranges between 2 and 5 years based on programmes’ risk assessment
and experience. Early-​stage active malignancies may be amenable to
successful lung transplantation (e.g. lung cancer radiographic stage 1,
breast cancer stage 1, prostate cancer Gleason score ≤6).
• Major uncorrectable non-​pulmonary organ dysfunction (cardiac,
kidney, liver) unless combined multivisceral transplantation can be
considered.
• Significant multivessel CAD not amenable to revascularization with
percutaneous pre-​transplant or bypass during transplant surgery.
• Severe chest wall or spinal deformity.
• Acute severe sepsis.
• Untreatable pulmonary infection or intolerance to requisite
antimicrobial therapy to treat pulmonary infection.
• Active Mycobacterium tuberculosis infection.
• Uncorrectable bleeding disorder.
• Significant obesity (BMI >35 kg/​m2); weight distribution primarily
across torso can also be problematic at lower BMIs as well.
• Substance abuse or dependence (e.g. ethanol, tobacco, marijuana,
cocaine, methamphetamine, or other illicit substances).
• Medical non-​compliance.
• Lack of suitable caregiver plan.

Specific aspects of different disease processes

Obstructive lung disease (group A)
See Table 24.1.
Prior to the lung allocation score (LAS) in the USA, transplant for severe
chronic obstructive pulmonary disease (COPD) was associated with an im-
provement in the quality of life more than improved survival. Even under
32

322 Chapter 24 Lung transplant selection criteria

the LAS or composite allocation score (CAS) systems, the timing of referral
and transplant to maximize the lifespan of patients with severe obstruction
can be challenging. One needs to balance the overall indolent course of the
disease with the potential complications of transplantation.
Utilization of the BODE (incorporating BMI, airway obstruction, dys-
pnoea, and exercise capacity) index in the decision-​making can be helpful as
a part of the mortality risk assessment. A BODE score of 7–​10 has been as-
sociated with a 4-​year mortality risk of 80% whereas a BODE score of 5–​6
has been associated with 4-​year mortality risk of 50–​60%, suggesting that
stratifying COPD patients by BODE score is a helpful adjunct as part of the
decision process of when to refer and list for transplantation. Lahzami et
al. evaluated COPD patients by BODE index and found improved survival
outcome for patients with a BODE score of 7–​10. Patients with a BODE
score of 5–​6 did not achieve a survival benefit; however, they did achieve
an improved quality of life following transplantation. A subset of COPD pa-
tients may be functionally limited by severe hyperinflation, which is not fully
captured in the BODE scoring model. Although the assessment of patients
by the BODE index is helpful in most cases, it should be incorporated into
a composite approach to evaluate patients’ candidacy for transplantation.
Assuming optimization of medical therapy and functional status, other
indications that should prompt referral to a transplant centre include an
increased frequency of pulmonary exacerbations, patients who are not can-
didates for lung volume reduction surgery, spirometry revealing a forced
expiratory volume in 1 second (FEV1) <30%, and those who manifest evi-
dence of hypercarbia on blood gas sampling.

Table 24.1 Indications for referral and listing for patients with obstructive lung
disease
Indications for referral Indications for activation on the
transplant waiting list
Progressive disease despite maximal BODE index ≥7
medical therapy, including pulmonary FEV1 <15–​20% of predicted
rehabilitation and domiciliary oxygen Three or more exacerbations in the
Patient is not a candidate for lung volume last 12 months
reduction One very severe exacerbation
BODE index 5–​6 with acute hypercarbic respiratory
PaCO2 >50 mmHg or 6.6 kPa and/​or failure
PaO2 <60 mmHg or 8 kPa Moderate to severe PH
FEV1 <25–​30% of predicted

Pulmonary vascular disease (group B)

See Table 24.2.
Timing of referral of patients with PH is also challenging. The increased
number of medical options for management of PH has improved the po-
tential for response to treatment depending on the underlying pathophysi-
ology. It is therefore recommended that patients undergo a trial of medical
therapy (phosphodiesterase inhibitor, endothelin receptor antagonist, or
prostacyclin analogue) before committing to lung transplantation particu-
larly in patients with idiopathic pulmonary arterial hypertension.
 Indications for referral to a transplant centre 323

The aetiology of primary pulmonary vascular disease is often unknown,

and this contributes to uncertainty regarding response to therapy and the
rate of symptom progression. These patients should be followed closely
and those manifesting NYHA class III or IV symptoms, rapidly progressive
disease, presyncopal/​syncopal symptoms, need for IV prostacyclin therapy,
or suspected pulmonary veno-​ occlusive disease/​ pulmonary capillary
haemangiomatosis should be referred to a transplant centre for evaluation
and monitoring to minimize mortality. Since this patient population can
manifest sudden, expected deterioration of symptoms following extended
periods of stability, it is particularly beneficial to refer these patients for
transplant consultation earlier on in their disease rather than later to avoid
missing an opportunity for transplant.

Table 24.2 Indications for referral and listing for patients with pulmonary
vascular disease
Indications for referral Indications for activation on the transplant
waiting list
NYHA class III or IV symptoms NYHA class III or IV symptoms after a 3-​
despite maximal medical therapy, month trial of maximal medical therapy
including prostanoids, endothelin CI <2 L/​min/​m2
receptor antagonists, and Mean RA pressure >15 mmHg
phosphodiesterase inhibitors
Six-​minute walk test <350 m
The need for IV agents
Haemoptysis
Known or suspected veno-​
occlusive disease or capillary Pericardial effusion
haemangiomatosis Development of signs of right HF (renal
insufficiency, increased serum bilirubin,
recurrent ascites)

Cystic fibrosis (CF) (group C)

See Table 24.3.
Recent initiatives within the CF community have drastically altered prior
guidelines for referral of CF patients. Historically, the timing of referral was
recommended when patients were manifesting signs of very advanced lung
disease based on severely reduced spirometry, increased number of ex-
acerbations, declining clinical status, pneumothorax, and life-​threatening
haemoptysis. Now, these indications are more associated with timing for
listing for transplant as the referral for consideration of lung transplantation
is strongly encouraged to occur much sooner.
The CF Foundation now contends that the above guidelines are not ag-
gressive enough to effectively manage the scope of issues faced by CF pa-
tients and instead recommends the following:
1. Routine clinician-​led discussion of disease trajectory and treatment
options. This may include a discussion of lung transplantation.
2. As soon as spirometry reveals a FEV1 <50% of predicted volume,
team-​based discussion regarding lung transplantation should occur.
Additional heightened concern should be maintained for patients with
rapidly progressive disease or CF-​related complications. Early referral
may be considered in such patients, which include:
324

324 Chapter 24 Lung transplant selection criteria

a. Relative FEV1 decline of >20% in 12 months.

b. CF exacerbation requiring the use of non-​invasive positive
pressure ventilation.
c. BMI <18 kg/​m2 (or BMI <5th percentile for children).
d. Increased frequency of pulmonary exacerbations (more than two
exacerbations per year requiring IV antibiotics or one exacerbation
requiring positive pressure ventilation).
e. Massive haemoptysis requiring ICU admission or bronchial artery
embolization.
f. Pneumothorax.
3. For patients with FEV1 <40%, referral should be initiated for the
following reasons:
a. All patients <18 years of age.
b. Six-​minute walk test <400 m.
c. Room air hypoxia (SpO2 <88% or PaO2 <55 mmHg, at rest or
with exertion; at sea level). Or if supplemental oxygen is required.
d. Hypercapnia (PaCO2 >50 mmHg, confirmed on arterial
blood gas).
e. PH (PA systolic pressure > 50 mmHg on echocardiogram
or evidence of RV dysfunction in the absence of a tricuspid
regurgitant jet).
f. BMI <18 kg/​m2.
4. Referral for all patients with an FEV1 <30% of predicted.
5. Use of up-​to-​date CF-​specific transplant resources.
6. Development of partnerships between CF and transplant centres.
7. Identify psychosocial and physical concerns regarding lung
transplantation and address these as part of the transition process to
transplantation.
8. Pre-​emptive optimization of modifiable barriers to transplantation
(nutrition, functional status, diabetes, medical compliance, mental
health, substance use, psychosocial concerns).
9. Special consideration for early referral of female patients (due to sex
disparate waiting list mortality) and those with short stature defined as
<162 cm or 5'3'' (due to difficulty in finding a suitably matched donor).
10. Special consultation with transplant centres for patients with micro-​
organisms that pose a risk for transplantation (non-​tuberculous
mycobacteria, Burkholderia cepacia complex, Scedosporium).
For a more comprehensive list of the CF Foundation recommendations
on the timing of referral for lung transplantation, please refer to their 2019
publication in the Journal of Cystic Fibrosis. Also note that newer therapies
for CF are being introduced and will likely significantly curtail the need for
lung transplantation in patients with the appropriate underlying genetic mu-
tations. Specifically, elexacaftor–​tezacaftor–​ivacaftor is a newly approved
triple-​combination cystic fibrosis transmembrane conductance regulator
(CFTR) modulating therapy that contains two correctors and a potentiator
of the CFTR channel. This drug was found to be efficacious in patients with
CF with Phe508del-​minimal function genotypes, in whom previous CFTR
modulator regimens were ineffective. CFTR modulators will likely have
utility for CF patients both prior to lung transplantation (in postponing or
potentially eliminating need for transplant) and after transplantation (to
manage significant sinus and/​or GI symptoms).
 Indications for referral to a transplant centre 325

Table 24.3 Indications for referral and listing for patients with CF
Indications for referral Indications for activation on the
transplant waiting list
FEV1 <30% of predicted With hypoxia (PaO2 <8 kPa or
FEV1 <40% with markers of increased <60 mmHg)
disease severity: With hypercapnia (PaCO2 >6.6
1. Age <18 years kPa or >50 mmHg)
2. Room air hypoxia Long-​term non-​invasive
3. Hypercapnia ventilation therapy
4. Six-​minute walk test <400 m PH
5. PH Frequent hospitalization
6. BMI <18 kg/​m2 Rapid lung function decline
FEV1 <50% or decline in FEV1 >20% in WHO functional class IV
12 months
FEV1 <50% and disease-​related
complication:
1. Increased frequency or severity of
exacerbation
2. Haemoptysis
3. Pneumothorax
4. Need for positive pressure ventilation

Restrictive lung disease/​interstitial lung disease (ILD) (group D)

See Table 24.4.
Patients with restrictive lung disease have the worst 5-​year survival of
lung groups at 50–​55%. Restrictive lung disease patients include idiopathic
pulmonary fibrosis, usual interstitial pneumonitis, non-​specific interstitial
pneumonitis, hypersensitivity pneumonitis, rheumatoid lung disease and
connective tissue diseases (scleroderma, lupus, polymyositis, dermatomyo-
sitis, mixed connective disorders). Given the potential for accelerated de-
cline in these diseases, early referral is emphasized based on symptoms,
oxygen requirements, PFT, and biopsy results if obtained.
Patients with ILD should be referred for lung transplantation following
the development of shortness of breath or limitation in their activity that
is a result of their lung disease. Any patient with ILD who requires sup-
plemental oxygen should be referred for transplantation as well as those
with abnormal PFT defined as a reduction in the forced vital capacity (FVC)
<80% of predicted or diffusion capacity of the lung for carbon monoxide
(DLCO) to <40% of predicted.
A referral is also recommended whenever a biopsy reveals histopatho-
logical evidence of usual interstitial pneumonitis or non-​specific interstitial
pneumonitis irrespective of oxygen requirements and PFT. Recently, the
advent of antifibrotic agents, nintedanib and pirfenidone, has somewhat al-
tered the current treatment of patients with usual interstitial pneumonitis.
However, it is unknown what effect, if any, these agents have on the indica-
tion for referral and activation on the waiting list.
Those patients with inflammatory ILD are appropriate to consider for
medical therapy and may experience stabilization or improvement in their
symptoms, oxygen requirements, or PFTs. Those who do not respond or
decline after an initial response should then be referred for transplantation.
326

326 Chapter 24 Lung transplant selection criteria

Table 24.4 Indications for referral and listing for patients with restrictive lung
disease
Indications for referral Indications for activation on the
transplant waiting list
Histopathological or radiographic Decline in FVC ≥5–​10% in 6 months
evidence of usual interstitial Decline in DLCO ≥15% in 6 months
pneumonitis or fibrosing non-​specific Desaturation to <88% or distance
interstitial pneumonitis, regardless of <250 m on 6-​minute-​walk test or >50
lung function m decline in 6-​minute-​walk distance
FVC <80% predicted over a 6-​month period.
DLCO <40% predicted PH on right heart catheterization or
Dyspnoea or functional limitation two-​dimensional echocardiography.
attributable to lung disease Hospitalization because of respiratory
Any oxygen requirement, even if decline, pneumothorax, or acute
only during exertion exacerbation
For inflammatory ILD, failure
to improve dyspnoea, oxygen
requirement, and/​or lung function
after a clinically indicated trial of
medical therapy

Indications for transplantation/​activation

on the waiting list
Obstructive lung disease (group A)
Patients with severe obstructive lung disease should be considered for
listing for transplantation if they experience an increase in the frequency
of exacerbations to greater than three in 1 year or if they have an admis-
sion for management of acute hypercarbic respiratory failure. Patients with
three or more exacerbations in 1 year had a hazard ratio of 4.13 for death
at 5-​year follow-​up, compared to patients with fewer number of exacer-
bations. As stated previously, the BODE index can be a useful guide for
referral and listing with a BODE index ≥7 conferring a survival benefit to
patients with severe COPD.
Spirometry in isolation is not a useful parameter for listing in patients with
COPD but a decline in FEV1 to <20% is commonly associated with signifi-
cant disease progression to warrant consideration of listing. Patients with
severe COPD who require higher rates of oxygen supplementation often
have accompanying secondary PH. In those COPD patients with moderate
to severe PH, listing should occur regardless of their FEV1. Hypercarbia
also predicts mortality among COPD patients and should be carefully
scrutinized when present.
Pulmonary vascular disease (group B)
Patients with pulmonary vascular disease should be listed for transplant-
ation if they have not responded to combined pulmonary vasodilator
therapy, including a prostanoid for at least 3 months and are symptomatic
with NYHA class III or IV symptoms.
 INDICATION FOR LISTING FOR LUNG TRANSPLANT 327

Patients with early signs of right HF (CI <2.0 or RA pressure >15 mmHg
on right heart catheterization) should be listed for transplantation before
they experience further disease progression. Those with evidence of ad-
vanced failure (development of ascites, elevated bilirubin, elevated BNP, or
decreased forward flow with elevated creatinine) should be urgently listed
if otherwise functional and felt to be good candidates. However, careful
consideration of whether to proceed should be performed as these pa-
tients are at high risk for perioperative mortality and overall poor outcome.
Cystic fibrosis (group C)
Despite multiple attempts to develop predictive models of mortality risk for
CF patients based on a variety of patient characteristics, ‘change in lung func-
tion over time’ continues to be the single best predictor to guide timing to
list for transplantation. Reliance on FEV1 alone, however, can be deleterious
and attention must be paid to other aspects in the CF population as patients
with FEV1 <30% may experience increased mortality if spirometry alone is
followed. The risk of death among female patients with FEV1 values between
20% and 30% is reported as high as 56%, as compared with 38% among male
patients. Development of increased severity and frequency of CF exacerba-
tions, haemoptysis, pneumothorax, or non-​tuberculous mycobacterial pul-
monary infection may all be indicators of a rapid decline, and these patients
need to be closely monitored. Those with hypercarbia (arterial PCO2 >50
mmHg), need for non-​invasive positive pressure ventilation, or secondary
PH should be strongly considered for immediate activation on the transplant
waiting list regardless of their FEV1. Female CF patients should be monitored
closely, given their increased mortality risk. A lower threshold for listing
should be considered if they begin to manifest early signs of chronic respira-
tory failure or increased difficulties with respiratory infections.
Resistant micro-​organisms may be considered as relative contraindications
to transplant. Burkholderia cenocepacia is typically avoided given its asso-
ciation with increased mortality following transplant. Other infections such
as Mycobacterium abscessus or pan-​resistant organisms (e.g. Pseudomonas,
Achromobacter) can present significant challenges for successful transplant
surgery. Patients with multidrug or pan-​ resistant pulmonary infections
must be thoroughly evaluated with a vigorous multidisciplinary approach
involving infectious disease, transplant surgery, and pulmonology teams
to review past cultures and susceptibilities, chest imaging, and proposed
surgical plan. A patient’s condition, presence of parenchymal cavities, or
significant pleural disease may determine the willingness to proceed with
transplantation in the face of these resistant organisms. A patient bridged
on ECLS with large cystic cavities adjacent to the pleura is at high risk for
bleeding and soiling of the pleural space during explantation of the lungs,
and these should be considered as to whether the patient is appropriate for
transplantation. Patients with resistant organisms and significant pulmonary
architectural abnormalities are generally best served by a referral to a large
transplant centre to manage these aspects of their disease.
Interstitial lung disease (group D)
Given the high mortality rate of patients with ILD, early referral for trans-
plantation and frequent monitoring is recommended. Signs of disease
progression manifested by an increase in oxygen requirement or decline
328

328 Chapter 24 Lung transplant selection criteria

in PFT (FVC decline >10% or diffusion capacity decline >15%) should

prompt consideration of listing for transplantation. Development of sec-
ondary PH or hospitalization for management of acute exacerbation of ILD
should also trigger activation on the waiting list. In general, the potential
for rapid disease progression should lead to frequent, close follow-​up of
ILD patients with anticipation for listing when manifesting any of the above-​
mentioned signs.
Maintenance of adequate nutritional status and conditioning are essential
for all transplant candidates, but these are particularly crucial for ILD pa-
tients. Frailty is common in ILD patients (due to the age of disease onset,
pulmonary cachexia, and accumulation of other medical comorbidities) and
correlates with poor outcomes following transplant surgery. It is imperative
that patients with ILD be referred early for transplantation as well as to
pulmonary rehabilitation at the time of their diagnosis to either maintain
or improve their functional status in anticipation of transplantation. ILD pa-
tients are at risk for pulmonary cachexia, given the catabolic nature of their
disease, which can be exacerbated by an exercise programme or antifibrotic
drug therapies. An early and aggressive comprehensive treatment pro-
gramme to address both nutrition and conditioning may be the key factors
of ILD treatment to bridge these patients to transplantation successfully.
Patients who are robust physically can be safely transitioned to transplant-
ation despite rapid progression of their disease whereas those with frailty
and malnutrition assume a much higher and potentially prohibitive risk.

Pre-​transplant evaluation
Pulmonary evaluation
The pulmonary evaluation serves three purposes. The first is to determine
whether the pulmonary disease has progressed to the point that justifies
transplantation. The second objective is to identify complications of the dis-
ease that may or may not be amenable to tailored therapies that may ameli-
orate the lung disease. For example, the CT scan and ventilation/​perfusion
(V/​Q) scan may identify COPD patients who may be candidates for lung
volume reduction surgery. Finally, the pre-​transplant evaluation is useful for
operative planning.
• PFT: PFTs are used to assess disease severity. FEV1 is an important
determinant of indication for waiting list activation in patients with CF
and COPD. FVC and DLCO serve a similar purpose in patients with
restrictive lung disease. Additionally, FVC is used in the calculation
of the LAS. Finally, the total lung capacity (TLC) is important in
determining the size match for potential donor lungs.
• Arterial blood gas: the degree of hypoxaemia or hypercapnia
is determined. These are used to titrate supplemental oxygen
requirements as well as the utility of non-​invasive positive pressure
ventilation. Hypoxaemia and hypercapnia are used to determine
indications for activation on the waiting list.
• VQ: the main purpose of the V/​Q scan is to determine the laterality in
patients who are listed for single-​lung transplantation. Additionally, in
programmes that perform bilateral sequential lung transplantation off
pump, the perfusion scan guides the operative team by identifying which
side to transplant first.
 Pre-transplant evaluation 329

• Six-​minute walk test: like blood gases and PFTs, the 6-​minute walk
test is an important determinant of disease severity and is used as an
indication for activation in certain disease processed (e.g. distance <250
m in patients with restrictive lung disease). It can also serve as a crude
assessment of functional status and deconditioning.
• CT imaging: in patients with CF, the CT scan is important in identifying
the degree of infection and cavitary lung lesions, which may aid in
operative planning. In patients with obstructive and restrictive lung, who
are predisposed to lung malignancy, a CT scan is important to rule out
lung cancer.
Cardiac evaluation
The cardiac evaluation is integral to the pretransplant evaluation. It is
very important to assess the cardiac function of the transplant candidate
and their ability to undergo the immense stress of lung transplantation.
Furthermore, it is important to identify modifiable heart disease, such as
CAD, which needs addressing before lung transplantation. Non-​modifiable
heart disease is important, and a determination needs to be made as to
whether this would constitute a contraindication for lung transplantation.
Finally, it is important to determine the presence of PH and the condition
of the right heart. Irreversible severe right heart dysfunction is a contraindi-
cation for lung transplantation, and the patient may be better served with
combined heart–​lung transplantation instead.
• Left heart catheterization: the main purpose of the left heart
catheterization is to rule out significant coronary disease. Identification
of important CAD will prompt a discussion of how to address it.
Options include medical management and PCI. Combined coronary
artery grafting at the time of transplantation is used as well in selected
patients. In some institutions, imaging with CT coronary angiography
is used. If an important lesion is identified, then traditional coronary
angiography is indicated for intervention.
• Right heart catheterization: this is used to determine right-​sided heart
pressure, including RA pressure, PCWP, and PA pressure. These values
are used to compute the LAS or CAS. They are also important in
stratifying the severity of disease in patients with secondary PH. This
test is mandatory in patients with pulmonary vascular disease.
• Echocardiogram: determination of cardiac function is important in
determining the ability of a patient to undergo lung transplantation,
especially in patients with PH. Mild left or right heart dysfunction is
acceptable but raises the possibility of intraoperative cardiopulmonary
support in programmes that use it selectively. In patients with severe
heart right heart dysfunction, it is important to determine whether the
lung transplantation may reverse this, or whether the patient should
undergo a heart–​lung transplant instead.
• Cardiac MRI in selected patients (sarcoidosis; stress MRI in certain cases
with CAD).
• ECG.
Gastrointestinal evaluation
The GI evaluation focuses on the assessment of upper GI function and the
likelihood of aspiration. Patients with oesophageal dysmotility are at increased
risk for post-​transplant complications from gastro-​oesophageal reflux and
30

330 Chapter 24 Lung transplant selection criteria

aspiration. Those with systemic sclerosis or mixed connective tissue disease

may have significant gastro-​oesophageal dysfunction that can present across
a wide spectrum of symptoms and clinical findings. Given the role of chronic
aspiration and gastro-​oesophageal reflux in the development of chronic lung
allograft dysfunction (CLAD), it may be advantageous to identify these prob-
lems preoperatively so that they can be addressed. Early fundoplication may
reduce the incidence of CLAD in this population. In patients with CF, GI ab-
normalities and delayed gastric emptying are common. Identification of these
issues helps delineate a nutrition strategy that may or may not incorporate
direct enteral access, whether by gastrostomy or jejunostomy. Finally, patients
with scleroderma and other connective tissue disorders are at risk for signifi-
cant oesophageal motility problems. Upper GI evaluation is very important
to identify patients who would need modification and tailored therapy to
address the oesophageal dysmotility in the post-​transplant period. Each trans-
plant centre’s experience with scleroderma and scleroderma-​like patients will
dictate their willingness to accept these patients for transplantation. Some
centres have achieved very acceptable outcomes with this patient subset, but
symptomatic patients with air-​fluid levels in a patulous, immotile oesophagus
are at high risk to do poorly outside of experienced centres.
• Barium oesophagram.
• Oesophageal manometry and 24-​hour pH monitoring.
• Speech therapy assessment for oropharyngeal dysphagia (consider
fibreoptic endoscopic evaluation of swallowing).
• Solid gastric emptying in patients with symptoms or history of
gastroparesis.
Nutritional evaluation
This evaluation is preferably undertaken by a dietician with experience in
managing lung transplant patients. Malnutrition is common in patients with
CF and restrictive lung disease. Malnutrition should be addressed with a
diet plan, oral dietary supplements, and enteric nutrition in some cases.
Occasionally, activation on the waiting list is delayed until the malnutrition is
addressed. If possible, it is best to avoid transplantation of patients who are
extremely underweight (BMI <18.5 kg/​m2) or extremely overweight (BMI
>35 kg/​m2). Outcomes in these patients are significantly worse, and it is
preferable to address the weight before listing.
Hepatic evaluation
• Ultrasound of the liver in patients <50 years of age.
• CT imaging to assess liver, vasculature, rule out malignancy >50 years
of age.
• Consider Fibroscan of liver or transjugular liver biopsy with pressure
measurements in a subset of patients with increased risk for liver
disease (e.g. CF, hepatitis C, etc.).
Renal evaluation
• Serum creatinine and eGFR generally adequate (creatinine clearance
>50 mL/​min is considered acceptable for lung transplantation).
• Consider measured GFR in patients with chronic kidney disease to
establish creatinine clearance >50 mL/​min.
• Screen for significant proteinuria with urinalysis (may need 24-​hour
urine collection/​nephrology referral if significant).
 Clinical management after listing 331

Infectious evaluation
• Hepatitis serologies.
• HSV, VZV, CMV, EBV serologies.
• HIV screen.
• Toxoplasmosis.
• Syphilis testing (rapid plasma reagin/​Treponema screen).
• Sputum for bacteria, acid-​fast bacilli, fungal stain, and culture.
Immunological evaluation
• HLA screen to evaluate for preformed PRAs.
Functional status/​frailty
• Varies by centre (regular participation in pulmonary rehabilitation
strongly recommended).
Cancer screening
• Colonoscopy for patients >50 years of age (endoscopic preferred;
virtual colonoscopy may be considered in certain conditions).
• Mammogram, pelvic exam.
• Prostatic serum antigen.

Clinical management after listing

General guidelines for patients on the list
Once patients have completed their evaluation and have met their trans-
plant centre’s criteria, they can be activated on the waiting list. Time on
the waiting list until undergoing transplantation varies by centre and may be
impacted by blood type, HLA sensitization, extremes of chest cavity size,
and stature. To optimize patient outcome following transplantation, it is im-
perative patients achieve and maintain adequate nutrition and conditioning.
All patients should be active in a regular exercise programme up until the
time of transplantation. In centres where patients have relocated to partici-
pate in their pulmonary rehabilitation, this approach has the advantage of
having trained therapists to actively assess patients’ clinical status while on
the waiting list to avoid surprises at the transplant surgery involving patients
who have significantly deteriorated since they were last seen.
Patients should generally be evaluated every 4–​12 weeks while on the
waiting list by the transplant team to ensure stability and determine if there
are any clinical status changes (increased oxygen requirements, decreased
walk distance, worsening hypercarbia) that would warrant an adjustment of
the allocation score or sequence on the waiting list. If patients are on the
waiting list for 6 months, certain tests including blood work and right heart
catheterization may need updating to avoid a default entry that could nega-
tively impact the allocation score.
The deteriorating patient
For patients who are deteriorating on the waiting list, every attempt should
be made to identify any aetiology for the decline that may respond to
therapy. Their sequence on the waiting list may need to be modified. For ex-
ample, their allocation score should be recalculated to see if they should be
32

332 Chapter 24 Lung transplant selection criteria

prioritized on the waiting list. If unable to remain in the outpatient setting,

hospitalization can provide a setting to stabilize and assess frequently to de-
termine suitability for transplantation. Occasionally, oxygen requirements
exceed flow rates that can be delivered outside the hospital and an admis-
sion to the hospital provides an opportunity to remain functional, improve
their clinical status, and still safely undergo transplant surgery. These pa-
tients should be carefully assessed as it has been shown that those with rap-
idly increasing allocation scores immediately prior to transplantation have
worsened survival following transplantation. In these instances, the large
change in clinical status or allocation score may be a surrogate marker that
the mortality risk outweighs survival benefit which warrants a thoughtful
assessment of the decision whether to move forward with transplantation
as much as is possible.
ICU management
For those inpatients who develop worsened hypercarbic and/​ or
hypoxaemic respiratory failure that necessitates management in the ICU
and are felt to remain suitable transplant candidates, an early discussion by
the team to determine the extent of interventions that will be offered is
prudent for effective management. Some patients may require increased
oxygen delivery (via high-​flow device), intubation with mechanical ventila-
tion, tracheotomy placement, or ECMO cannulation. All of these inter-
ventions increase the level of risk for transplantation and potential for
worsened outcome. Although ECMO (most commonly VV ECMO but
also VA ECMO in certain patients with right HF) has become an acceptable
and effective strategy to bridge unstable and declining patients, it is pref-
erable that the patient be ambulatory or at a minimum off sedation and
neurologically intact. Sedation and ECMO support can hide a multitude
of unstable conditions for successful transplantation. Each centre should
establish criteria in these instances for which they feel they can achieve a
good outcome for these deteriorating patients to not only guide medical
care and clinical decision-​making but also to provide clear communication
to patients and their families and appropriately manage their expectations.

Special considerations
Single-​versus double-​lung transplantation
In patients with suppurative lung disease or severe PH, a bilateral lung
transplant is mandatory to ameliorate the disease effectively. However,
in patients with restrictive lung disease and COPD, single orthotopic
lung transplantation (SOLT) is considered an appropriate transplantation
approach.
Improved survival in patients receiving bilateral orthotopic lung trans-
plantation (BOLT) has been demonstrated in multiple retrospective studies.
Studies employing the registry of the ISHLT have likewise supported this
finding. Being retrospective in nature, these findings may be affected by
selection bias. A study by Thabut et al. showed that in patients with pul-
monary fibrosis, this survival advantage might be related to baseline differ-
ences between the groups, and correcting for these differences may nullify
the mortality benefit in favour of BOLT. Moreover, they concluded that
 Special considerations 333

while single-​lung transplantation confers short-​term survival benefit but

long-​term harm, and bilateral transplantation confers short-​term harm but
long-​term survival benefit.
The balance between short-​term risk versus harm and the desire for the
optimal good long-​term survival is the underlying tenet that drives the de-
cision between SOLT and BOLT. Multiple criteria have been proposed as a
means to define the group of patients where overall short-​term risk pro-
file favours SOLT, such as advanced age. Based on these studies, it would
seem that with increasing donor age and frailty, the lower perioperative
mortality of a single-​lung transplant may outweigh the advantage of better
long-​term survival. Gulack et al. found that bilateral lung transplantation
was associated with improved long-​term survival in older patients with idio-
pathic pulmonary fibrosis, except in patients with reduced functional status.
Villavicencio et al. found that bilateral lung transplantation does not confer a
survival advantage in patients >70 years old. However, they recommended
against the use of single-​lung transplantation in patients with PA pressure
>30 mmHg and LAS >45.
Recognizing that there are no definitive data on the subject, the deci-
sion to offer single-​lung transplantation has varied from centre to centre.
At our institutions, we offer bilateral lung transplantation to all patients with
CF, suppurative lung disease, and pulmonary vascular disease. Bilateral lung
transplantation is our preferred strategy for patients with restrictive and
obstructive lung diseases. However, we will offer single-​lung transplantation
for some elderly patients >65 years of age, especially if they have reduced
functional capacity, significant comorbidities, CAD, or previous chest sur-
gery. We would generally avoid single-​lung transplantation in patients with
significant secondary PH (mean PA pressure >40 mmHg), since registry
data suggest that BOLT may be a preferred option in these patients.
Highly sensitized patients
Patients with class I or class II PRAs are considered sensitized. The typical
approach is to avoid donors with the same antigen–​antibody profile as the
recipient given the risk for graft dysfunction. Several considerations of PRAs
are important in determining their significance, including their MFI and pres-
ence on 1:16 dilution, which correlates with strength. Whether they are
class I (present on all cells) or class II (only present on immune cells) is also
important in determining their significance. Most often, patients either do
not have these antibodies or have reasonably low levels such that avoiding
donors with the corresponding antigens will not result in a prolonged time
on the waiting list. Some patients, however, do have very high PRA levels
(>80–​90%) and when combined with their blood type and stature, can re-
sult in very long wait times.
Attempts at ‘desensitization’ (a reduction in the antibody burden) prior
to transplantation by plasmapheresis (PLEX) and medications have not
shown consistent results thus far. Examples of medication used for de-
sensitization include proteasome inhibitors (e.g. carfilzomib, bortezomib)
to impact the plasma cells which produce antibodies and rituximab to af-
fect B lymphocytes. Positive results have been seen in the administration
of PLEX, proteasome inhibitors, and rituximab in highly sensitized pa-
tients following transplantation in patients with a negative or even positive
cross-​match.
34

334 Chapter 24 Lung transplant selection criteria

Retransplantation
Patients who develop CLAD may be considered for retransplantation.
According to the 2017 Organ Procurement and Transplantation Network
Annual Data Report, retransplantation comprised 3.2% of all lung trans-
plants. Those with obstructive CLAD tend to have a slower rate of disease
progression and overall better prognosis, including for retransplantation
than patients with restrictive CLAD.
Each transplant centre has its own guidelines regarding retransplantation
(some will avoid retransplantation altogether and others perform multiple
operations in the same patient). Centres that perform retransplantation ad-
here to their basic criteria of candidacy for the initial transplantation. In
general, obstructive CLAD, long duration following prior transplant surgery,
absence of antibody sensitization, and absence of significant chronic kidney
disease all confer a better prognosis with retransplantation. Even with these
more favourable characteristics, programmes and patients should expect a
retransplantation to potentially be more challenging than their prior trans-
plant surgery due to the likelihood of pleural adhesions and bleeding as well
as the risk of worsened kidney function.
Multiorgan transplant
Heart–​lung transplantation
The majority of patients with end-​stage lung disease will benefit from iso-
lated lung transplantation and preservation of the native heart. Even the ma-
jority of patients with some degree of cardiac dysfunction will not require
cardiac transplantation, and cardiac dysfunction is likely to improve after
lung transplantation. However, there are rare situations where combined
heart–​lung transplantation is unavoidable. These include the following:
• Patients with PH secondary to CHD and irreparable congenital defects.
• Patients with acquired heart disease (CAD, valvular heart disease),
who are being considered for cardiac transplantation, and who have
irreversible PH:
• In these patients, assessment with a heart catheterization should
be performed. Pulmonary arterial hypertension and elevated
PVR (defined as a PVR >5 Wood units, a PVR index >6, or a
transpulmonary pressure gradient 16–​20 mmHg) should be considered
contraindications for isolated cardiac transplantation. Vasodilator
therapy may be used to reduce PVR. However, this is typically limited
by the precipitation of systemic hypertension. In these cases, heart–​
lung transplantation should be considered, as the incidence of RV
failure after isolated cardiac transplantation is very high.
• Patients with end-​stage lung disease, complicated by PH and irreversible
RV or biventricular dysfunction:
• The majority of these patients with RV dysfunction will benefit from
lung transplantation only. However, there is a subset of patients
where RV dysfunction is irreversible and may not improve with
lung transplantation alone. This is a very difficult group to define. In
some programmes, lung transplantation is offered for all patients
with pulmonary arterial hypertension, and RV dysfunction, with
heart–​lung transplantation, offered only to patients with significant LV
dysfunction (defined as EF <40%, CI <2.2 L/​min/​m2, with or without
PCWP <15 mmHg).
 Special considerations 335

• In patients with PH, chronic pressure or volume overload imposed

on the RV initially promotes compensatory mechanisms. This is
characterized by elevated PA pressure, elevated PVR, and elevated
RA pressure, and a preserved ventricular function. With time, and
with prolonged exposure to overload, the ventricle transitions from
a compensated to a decompensated phase characterized by myocyte
loss and scarring. In the decompensating phase, as PVR and RA
pressure remain elevated, RV function declines and forward flow of
blood into the PA is decreased. Cardiac output subsequently declines,
followed by a reduction in PA pressure. Declining PA pressure in
the setting of high PVR is an ominous clinical finding, heart–​lung
transplantation should at least be considered.
Combined kidney and lung transplantation
Deterioration in kidney function is common in SOT recipients. The aeti-
ology is multifactorial, but the unavoidable use of CNIs is most likely to
blame. It has been shown that kidney function declines rapidly in the first
month after transplantation, declines slightly in the first year after that, and
stabilizes up to 10 years. GFR is used to stratify kidney function, with a GFR
of <50 mL/​min/​1.73 m2 used as a threshold for determining the risk of
adverse outcomes. However, a study of the UNOS database shows that
some of these adverse outcomes can be ameliorated by receiving a kidney
transplantation. In this study by Osho et al. of the lung recipients listed for
kidney transplantation after receiving a lung allograft, 56% were not on dia-
lysis. Time to listing was approximately 80 months. Patients who received
a kidney transplant had better survival than those who did not. Given the
high morbidity and mortality following lung transplantation in patients who
develop end-stage renal disease in the first year following their transplant
surgery, an increasing attractive option is utilizing living donor kidney trans-
plantation. Those candidates noted to have marginal renal function prior to
their lung transplant may benefit from identifying a living donor kidney to
provide an option to proceed with isolated living donor kidney transplant-
ation following their lung transplant if end-stage renal dysfunction develops
and chronic hemodialysis is required.
Combined liver and lung transplantation
Certain diseases, such as CF and alpha-​1-​antitrypsin deficiency, will have
combined lung and liver dysfunction. In these patients, lung dysfunction
is typically more severe. Combined liver and lung transplants have been
reported mainly in UNOS database publication and small case series.
Patients with combined liver and lung dysfunction should meet disease-​
specific criteria for activation for both lung and liver transplantation.
However, caution should be exercised in patients with very advanced
liver disease (albumin <2.0 g/​dL, INR >1.8, or the presence of severe
ascites or encephalopathy). However, many patients with less advanced
liver disease may be considered for combined lung and liver trans-
plantation. The decision to proceed with concomitant transplantation
during the same operation is influenced by many aspects. These include
concern about the progression of organ dysfunction after single-​organ
transplantation, the rate of decline of lung or liver dysfunction, concern
for coagulopathy due to the liver disease, and time on the waiting list
36

336 Chapter 24 Lung transplant selection criteria

for patients with unfavourable characteristics (high PRA levels, short

stature).
De-​listing
Removal from the waiting list may be one of the most difficult decisions
undertaken by a lung transplant team. However, it is inevitable that some
patients on the list will progress or develop new problems that render them
unsuitable for lung transplantation. Patients on the waiting list should be
followed regularly to screen for the development of issues, progression of
disease, and contraindications for lung transplantation. The development
of any of the contraindications from Box 24.1 (see p. 321) should prompt
removal from the waiting list. Perhaps one of the most difficult decisions
to make is to determine the level of deconditioning that should prompt
removal of the waiting list. The decision should be made by taking into con-
sideration input from all members of the team, including physicians, physio-
therapists, nurses, and dietician.
 Chapter 25 337

Lung donor selection

and management
Introduction 338
Donor care 338
Criteria for lung decline 338
Surgical assessment at the donor hospital 339
Lung donor acceptance criteria 339
Donation after circulatory death 344
Conclusion 344
38

338 Chapter 25 Lung donor selection and management

Introduction
As lung acceptance rates from donors stand at <20%, there is a stark con-
trast with the demand for transplantation. As a result, much work has been
done on widening donor lung selection and acceptance criteria to safely use
less than ideal (marginal) lungs.
Current criteria for ideal lung donors include a donor age <55 years,
PaO2 >300 mmHg (40 kPa), no smoking history, and a clear bronchoscopy
and CXR. Finding a donor that meets all of these criteria is rare in the
modern era but accepting lungs with extended criteria can lead to excellent
and equivalent outcomes.

Donor care
Donor lung functional criteria can be improved through careful and active
donor management by the team caring for the patient. This can optimize
donation criteria, increase conversion rates, and support better long-​term
outcomes after transplantation.
As brain death and the catecholamine surge lead to haemodynamic in-
stability, the activation of inflammatory pathways, and endocrine dysfunc-
tion, the quality and function of donated lungs can be affected significantly.
The sympathetic storm leads to increased pulmonary capillary pressures
and disruption of the alveolar–​capillary membrane. The loss of protein-​rich
fluid into the alveoli can cause hypoxaemia and prolong this phase of high-​
permeability pulmonary oedema.
To mitigate against these effects, several strategies can be employed in
intensive care to maximize the yield of lungs from suitable donors.
The beneficial effects of employing lung protective ventilation strategies
(tidal volume 6–​8 mL/​kg predicted body weight, PEEP 8–​10 cmH2O) and
careful fluid balance to avoid the neurogenic pulmonary oedema associated
with increased capillary permeability are important. Ventilation should aim
to maintain a PaO2 ≥10 kPa and a pH >7.25.
Optimal intensive care management of the lung donor ahead of retrieval
should also include:
• Administration of methylprednisolone (15 mg/​kg).
• IV crystalloid maintenance fluid to maintain Na+​ <150 mmol/​L and
urine output between 0.5 and 2.0 mL/​kg/​hour.
• Vasopressin (0.5–​4 units/​hour) to treat diabetes insipidus.
• An insulin infusion to keep blood sugar at 4–​10 mmol/​L.
• The use of antiembolic stockings and low-​molecular-​weight heparin
to prevent thromboembolic issues. Subclinical microemboli are quite
commonly encountered in donor lungs at retrieval and can be flushed
out of the PA during retrograde perfusion of the lungs through the
pulmonary veins.

Criteria for lung decline

Transplant surgeons and physicians must maintain broad acceptance criteria
to avoid discounting potentially transplantable organs that could benefit
those on the waiting list. The usual exclusion criteria for donation of any
 Lung donor acceptance criteria 339

organ always apply but certain donors are clearly unsuitable for lung pro-
curement and can be declined.
These include:
• Donors with a PaO2 <25 kPa (187 mmHg) on an FiO2 of 1.0 and
PEEP of 8 cmH2O after rigorous attempts have been made to recruit
atelectatic segments, clear the airways of secretions, and endotracheal
tube malposition has been excluded.
• Where a clear and irreversible cause for hypoxaemia has been
established such as bilateral significant pulmonary contusion or other
trauma, documented aspiration with gross airway contamination which
cannot be cleared, or CXR evidence of major pulmonary consolidation.
In the presence of a PaO2 <25 kPa on an FiO2 of 1.0 and PEEP of 8 cmH2O
and unilateral CXR changes only, the possibility of single-​lung transplant-
ation should be considered (pulmonary venous sampling during organ
retrieval is recommended as this may demonstrate good function in the
unaffected lung making it suitable for donation despite concerning systemic
blood gases).
Ex vivo lung perfusion (EVLP) where available should always be con-
sidered before declining organs where potential exists to recondition the
lungs to transplantable status. Situations where EVLP would be beneficial
will include those with a PO2 <40 kPa (FiO2 100% PEEP 8 cmH2O). Lungs
showing pulmonary oedema clinically or radiologically, those with persistent
atelectasis despite optimal recruitment manoeuvres, or excessive purulent
secretions are especially suitable. Trauma lungs (e.g. multiple transfusions,
lung contusion, ‘hanging’ cause of death) and logistical reasons are also very
reasonable indications for considering the use of this technology.

Surgical assessment at the

donor hospital
After reviewing the age, previous medical history, behavioural history, and
travel history, further tests are required: standard clinical blood tests, blood
group determination, infectious disease screening, and review of CXRs or
CT scans. A standardized blood gas analysis will reveal, determined while
ventilating on an FiO2 of 100% and a minimum PEEP of 8 cmH2O, whether
the gas exchange is adequate (PO2 >35 mmHg). Appropriate recruitment
manoeuvres and bronchoscopy should be undertaken before evaluating
function. Donor and recipient can then be matched on blood group, height,
sex, total lung volume, and antibody status. The latter can be a direct cross-​
match or the more often used virtual cross-​match.

Lung donor acceptance criteria

Inherent donor characteristics may influence the quality of the lungs and
potentially impact recipient outcome. In general, but not exclusively, ventila-
tion of the donor should be for <10 days. The presence of a tracheostomy
does not preclude lung donation.
There are a number of characteristics that should be considered by the
retrieval and implanting surgeons:
340

340 Chapter 25 Lung donor selection and management

Cause of death
Donor death by asphyxiation or drowning should not automatically ex-
clude transplantation provided airway contamination with marine flora and
algae or parenchymal injury have been excluded. However, aspiration of
seawater can induce a direct lung injury and so one should be cautious in
this scenario.
Donors arise from suicidal hanging where occlusion of the airway is com-
bined with carotid artery obstruction. The former will cause pulmonary
oedema which can be neurogenic or due to high negative intrathoracic
pressures. Inspiration against the occluded upper airway causes a fall in
intrathoracic pressure, increased venous return, and increased pulmonary
capillary pressure, with a simultaneous fall in pulmonary interstitial pres-
sure resulting in pulmonary oedema. Hypoxia also increases pulmonary
capillary vascular resistance increasing hydrostatic pressure and pulmonary
capillary permeability. Air trapping may cause parenchymal barotrauma
and lung surface bullae and haemorrhage may be evident at inspection on
retrieval.
However, studies have shown no effect on the rate of airway dehiscence,
rejection, or long-​term survival from drowning or asphyxiation donor lungs
after transplantation. Ten-​year survival curves in >18,000 lung transplant
recipients were not influenced by donor cause of death, either when ana-
lysed individually or when asphyxiation or drowning was compared with all
other causes.
However, alterations in surfactant have been described and so care must
be taken in the evaluation of the lungs at retrieval.
Carbon monoxide poisoning is a rare cause of organ donation but
should not necessarily lead to rejection of the lungs. In a small series from
Cambridge, UK, satisfactory outcomes were achieved. Concomitant smoke
inhalation, however, suggests that caution should be taken and bronchos-
copy may show airway burns which preclude use for transplantation.
In patients who have sustained severe chest trauma, the lungs may still be
utilized for transplantation. Mild to moderate lung contusions are acceptable
and do not contraindicate use. CT scanning may be helpful if available to
specifically evaluate this. The presence of rib fractures or pneumothorax is
not a problem in isolation nor is the presence of a chest drain. Inspection of
the lung at retrieval, however, is critical in the final decision-​making pathway
as parenchymal injury that cannot be easily repaired may be evident.
Age
The mean age of donors in Europe has been increasing. Registry data
suggest that donors with ages <18 and >64 years are related to a higher
mortality at 1 and 3 years. Donor age up to 70 years for brainstem dead
donors (DBD) and 65 years for donors after circulatory death (DCD) are,
however, acceptable with good outcomes demonstrated in several series.
This can be extended further to 75 years especially if the donor had not
smoked for 10 years or was a non-​smoker. Donor age criteria can be safely
extended, and caution need only be applied if there are compounding risk
factors in addition, such as a prolonged ischaemic time.
The interaction between donor and recipient age alone is not an inde-
pendent factor in determining survival.
 Lung donor acceptance criteria 341

Sex and race

Donor and recipient sex and racial combinations have been analysed with
diverse results.
Fessart et al. failed to discern a difference in recipient survival after ana-
lysis of all gender combinations but other retrospective single-​institution
studies have demonstrated an increase in survival and decrease in bron-
chiolitis obliterans syndrome (BOS) for donor–​recipient male into female
and female into male sex mismatches. Male donor to male recipient com-
binations had a significant decrease in survival. ISHLT registry review has
shown decreased survival in female donors to male recipients. Female
donor to female recipient matching though showed a short-​and long-​term
survival benefit.
Race-​matched donor/​recipients have improved outcomes and African
American donors are associated with worse survival irrespective of the
race of the recipient. Overall, though, recipient race is not associated with
survival variability. No changes in 1-​year rejection rates have been associ-
ated with race matching.
As the nature of gender interactions between donor and recipient have
yet to be fully elucidated, it should not play a major role in donor selection
or acceptance.
Donor size
TLC (both predicted and actual) and recipient pathology (e.g. fibrosis or
emphysema) are important in considering an appropriate donor-​to-​re-
cipient match. For double lung transplants, patients with emphysema should
be matched to a donor with a TLC of 60–​100% of the recipient’s TLC. For
pulmonary hypertension and CF patients, the predicted TLC of the donor
should be 100–​120% of the recipient’s actual TLC.
Due to the reduced TLC seen in recipients with pulmonary fibrosis, the
donor’s TLC should be approximately halfway between the recipient’s ac-
tual TLC and predicted TLC plus 20%.
Generally, it is preferable to slightly oversize rather than undersize.
Experienced surgeons may consider graft volume reduction using tech-
niques such as right middle lobectomy and multiple wedge resections using
pericardial strip reinforced staplers.
Very good outcomes comparable to conventional transplants have been
described using this technique.
Blood gases and ventilation
On an FiO2 of 100% and PEEP of 8 cmH2O, a PaO2 ≥35 kPa (or a PaO2
≥14 kPa with an FiO2 of 40%) are acceptable for transplantation. The add-
ition of 8 cmH2O of PEEP is recommended to prevent atelectasis and
help minimize neurogenic pulmonary oedema. Initial PaO2/​FiO2 ratios of
<300 mmHg (40 kPa) should not exclude donor organ usage and recruit-
ment techniques should be implemented with intent to transplant alongside
the use of bronchoscopy to clear secretions.
Normal ventilatory parameters and normal lung compliance are essential.
Oxygenation (although considered by many to be the most important
element in lung donor acceptance) is probably the weakest physiological
measure to define and determine the suitability of the lung for transplant-
ation but is frequently used as a critical decision-​making parameter.
342

342 Chapter 25 Lung donor selection and management

In a recent Australian study of 93 brainstem dead lung donors, an arterial

PO2/​FiO2 ratio of <300 mmHg (40 kPa) was largely caused by a low ratio in
the lower lobes. There were no differences between the recipients receiving
donor lungs where the ICU PO2/​FiO2 ratio was <300 mmHg compared
with those ≥300 mmHg in time to extubation, severity of primary graft dys-
function, pulmonary function at 6 and 12 months, and survival at 1 year.
Had a donor PO2/​FiO2 threshold of 300 mmHg been adhered to, then up
to a third of the donor lungs available would have been declined. A ratio of
300 mmHg therefore can be considered very conservative, wastes accept-
able donor lungs, and leads to the unnecessary and expensive use of EVLP.
The blood gases should always be evaluated in the context of other
donor characteristics, but lungs should not be discarded out of hand in
the face of initially poor blood gases. These can very often be improved
in situ in the donor, may mask a usable single lung, or can be reconditioned
with EVLP.
Smoking history
Smoking history is not a concern if lung function and assessment at retrieval
is otherwise good.
A smoking history of up to 30 pack-​years is perfectly acceptable but if in
excess of this then other relevant factors and donor characteristics should
be considered in combination to ensure that hazards are not compounded
(such as donor age, ischaemic time) as decline might then be appropriate
for particular recipients. Donor lungs with a >20 pack-​year history are not
associated with greater BOS or reduced survival.
Smoking history is associated with a reduced 3-​year recipient survival
(by 10% compared to non-smoker donors) when compared to non-​
smoker donor lungs. Patients receiving lungs from donors with a positive
smoking history had a lower hazard of death after registration than those
who remained on the waiting list. Thus, the transplantation of recipients
with smoker lungs remains advantageous as mortality on the waiting list is
greater than the added postoperative risk.
However, in the UK experience, recipients whose donors had a smoking
history >20 cigarettes/​day have a 6-​year survival of 20% less than that ob-
served with non-​smoker donor lungs (and 10% less than those who had
smoked <20 cigarettes/​day).
The collapse test at retrieval is important in determining whether signifi-
cant air trapping is present suggesting chronic obstructive airways disease.
If the lungs do not rapidly deflate, they should be declined. EVLP is not
appropriate in such situations as this is not a reversible pathology where
reconditioning will help.
Bronchoscopy and microbiology
Post-​transplantation pneumonia and sepsis are serious concerns to the
transplant surgeon. It is critical to gain the expert advice of a transplant
microbiologist in most cases to advise on donor culture results where they
have been obtained.
Purulent secretions do not automatically preclude lung donation unless
there is evidence of inflamed mucosa and contact bleeding on bronchos-
copy. Gram stain evaluation of the airways in a single-​centre retrospective
study found that 12% of donors with a positive Gram stain subsequently
 Lung donor acceptance criteria 343

developed recipient pneumonia compared to 20% in those with negative

Gram stains! This refutes the association of donor Gram stain with recipient
pneumonia. In this study, however, donor lungs were not accepted if there
was evidence of frank aspiration on bronchoscopy.
Multiple organisms on Gram stain may indicate normal flora and are
unlikely to lead to post-​transplant infection despite the patient being im-
munosuppressed. Positive Gram stains in tracheal aspirates usually do not
indicate ongoing pneumonia but simply purulent secretions harboured in
the upper airways. These are readily removed at bronchoscopy. No specific
Gram stain findings are associated with or reflect a risk of poor outcome in
recipients. Heavy fungal contamination of the bronchial tree may, however,
exclude donation.
Importantly, though, positive cultures from donor BAL specimens may
indicate ongoing and clinically significant infection and lead to longer ICU
stay and prolonged ventilation in recipients when compared to donors with
negative BAL cultures.
Lavage specimens should be taken for culture from all donors to aggres-
sively adapt and refine antibiotic therapy postoperatively. This practice has
significantly reduced the incidence of recipient pneumonia.
Prospective analysis of donor airway cultures and bronchial tissue cultures
revealed a <1.5% transmission rate of donor organ contamination. With
appropriate antibiotic prophylaxis to cover Pseudomonas and Staphylococcus
aureus, the risk of transmission of donor-​associated infections is negligible.
CMV mismatches are acceptable unless specified in high-​risk recipients.
Radiological imaging
CXRs are a poor indicator of lung donor function or long-​term post-​
implant recipient survival. No studies have ever correlated CXR findings to
infections in recipients.
As such, CXR should not be used alone to decline organs. Radiological
abnormalities may be transient and signs such as infiltrates do not influence
outcome and may often resolve. Borderline blood gases combined with a
unilateral abnormality on CXR may mask a usable contralateral lung.
CT scans may have been obtained and are increasingly made available to
transplant teams. This can be useful in better resolving lung abnormalities
such as nodules, but also in assessing the severity of consolidation as op-
posed to atelectasis.
Lung contusions from trauma can be evaluated accurately on CT in terms
of number and severity. If modest, lungs may still be acceptable for use but
may lead to lung decline if severe.
Pleural effusions are usually reactive or inflammatory but can be associ-
ated with chest trauma. The presence of an effusion alone is not a reason
to decline the lungs, only if the pleural fluid is infected or there is radiological
evidence of empyema should microbiological advice on the usability of the
lungs be sought.
Ischaemic time
Ischaemic times have traditionally been kept to <6 hours, but more recent
studies have suggested that the lung is very resistant to ischaemia and with
better organ preservation techniques, ischaemic times can safely be more
generous. An ischaemic time of up to 8 hours is acceptable.
34

344 Chapter 25 Lung donor selection and management

Ischaemic time alone therefore should not limit donor retrieval distances
or transport times.
Several single-​centre studies have demonstrated no effect on survival
with preservation >6 hours, but surgeons should be cognisant of the addi-
tive effects of adverse donor characteristics where factors such as age or
heavy smoking history may combine with a long ischaemic time to increase
risk in certain recipients.
A review of >5000 lung transplant recipients showed greater early mor-
tality with cold ischaemic times of 7–​8 hours and donors >45 years of age.
Ischaemic times of ≥10 hours have been successfully reported with op-
timal donors.

Donation after circulatory death

When, after careful assessment of a patient in ICU, it appears that fur-
ther medical treatment is considered futile, withdrawal of treatment is ap-
propriate. The considerations outlined previously in this chapter for DBD
donors apply similarly in this category of donor although the retrieval pro-
cess is very different.
The donor can be extubated with the intent of allowing death to occur.
Respiratory and haemodynamic parameters are observed: blood pressure,
pulse frequency, respiration rate, and oxygen saturation are documented
at regular intervals. For lung retrievals, the only important parameter is the
systolic blood pressure. A systolic blood pressure <70 mmHg cannot be
permitted for >1 hour. After this period of warm ischaemia, the lungs are
considered non-​transplantable and should be declined. After 30 minutes of
warm ischaemia the use of EVLP can be considered if available.
Once the donor has been declared dead, a legally defined 10-​minute
period has to be respected before the actual retrieval procedure can start.
Unfortunately, most DCD procedures do not progress to organ retrieval
as the donor remains haemodynamically stable beyond the prescribed time
limits.

Conclusion
Given the low acceptance rates of donor lungs for transplantation, every
effort should be made to increase acceptance, improve function, and there-
fore utilization. This begins with optimal ICU care of the donor with ap-
propriate management of the catecholamine storm, careful fluid balance to
mitigate against neurogenic pulmonary oedema, and lung protective venti-
lation strategies.
Ideal donors are now a rarity. Surgeons must be prepared to thoroughly
evaluate and accept marginal lungs as excellent outcomes in recipients can
be obtained.
Donor age can be extended to >65 years and even to 75 years. Donor
smoking history is associated with good outcomes after transplantation and
is certainly better than the chances of survival on the waiting list.
CXR abnormalities should never be used alone as a reason for decline.
Abnormalities may be transient or reversible.
 Conclusion 345

Poor blood gases can be misleading and should not be used to decline
lungs out of hand. Excellent outcomes can be obtained from lungs where
the donor initially demonstrated hypoxaemia and setting the bar too high
can result in the decline of perfectly suitable lungs for transplantation.
Lung recruitment manoeuvres and bronchoscopy are very important in
the optimization process. Upper airway secretions are easily cleared and
lavage specimens should always be obtained to guide postoperative anti-
biotic strategies.
Ischaemic times can safely be prolonged well beyond the traditional 6-​
hour limit with good results.
In lungs that cannot meet acceptance criteria for transplantation despite
careful evaluation and optimization, the use of EVLP shows great promise
in reconditioning lungs where there is a reversible pathology.
346
 Chapter 26 347

Lung procurement
following donation
after brain death
and donation
after circulatory death
Donation after brain death donor lung procurement 348
Donation after circulatory death donor lung procurement 352
348

348 Chapter 26 Lung procurement

Donation after brain death donor

lung procurement
Upon arrival at the procurement site, the team should begin by assessing
available imaging, recent blood gases, and measuring the length and width
of the prospective donor’s chest. Any CT imaging should be examined for
any emphysema or nodules that may not have been mentioned previously.
Once the candidate donor is deemed suitable, flexible bronchoscopy is per-
formed to evaluate the extent of secretions, to assess for any anatomical
variations that may complicate implantation, and to thoroughly suction the
airways. Secretions may be copious, but attention is paid to the rate of re-​
accumulation and whether the lungs contain enteric material.
The lungs are procured via a median sternotomy incision. The pericardial
and bilateral pleural spaces are entered and the lungs are visually inspected.
Visual inspection should include not only looking for blebs but also areas of
diffuse emphysema. During the inspection phase prior to organ harvest, it is
important to specify that the settings on the ventilator are appropriate with
a PEEP of 8 cmH2O and an FiO2 of 100%. A sustained recruitment man-
oeuvre of 30 cmH2O for several seconds is made to recruit atelectasis and,
upon deflation, both lungs are manually palpated to detect for pathology.
Lung deflation should be done by disconnecting the ventilator, which has the
effect of allowing the surgeon to accurately assess the compliance of the
potential donor lungs.
A central gas should be obtained no sooner than 5 minutes after this
‘recruitment manoeuvre’. If performed too quickly, the challenge gas may
be artificially low due to donor hypotension during lung manipulation and
recruitment. Individual pulmonary vein gases may also be sampled to as-
sess the PaO2 in the individual lobes. A central arterial blood gas with a
PaO2 of >300 mmHg is generally deemed suitable. At this point, commu-
nication with the implanting team should be made to confirm or abort the
transplantation.
Once assessment of the lungs deems them suitable for transplantation,
it is generally advisable, if the donor is stable, to proceed with additional
dissection. This serves the purpose of reducing the chances for iatrogenic
injury at the time of explant. Encircling the aorta and freeing it from the
underlying right PA helps to ensure that it is not injured at the time of aortic
transection. The aortopulmonary window is dissected, separating the main
and right PAs from the aorta, in order to prepare for the aortic cross-​clamp.
The SVC is dissected circumferentially and a tourniquet loosely passed
around it cephalad to the azygous vein. The azygous vein is tied off approxi-
mately 0.5–​1 cm off of the SVC to prevent stenosis. Dissecting the SVC
away from the underlying right PA as it courses under this vessel also en-
sures that it is not injured later during explant. The trachea is then mobilized
by retracting the aorta and PA, incising the superior posterior pericardium
along the PA and using a finger to bluntly dissect in the plane around the
trachea. If a longer length of trachea is required as is often desired with
subsequent EVLP use, the innominate artery can be dissected out and en-
circled and retracted which will provide ample exposure to the trachea at
a more proximal level. Our institutional preference is to avoid dissection in
the interatrial groove of the heart (Waterston’s or Sondergaard’s groove)
 DBD lung procurement 349

until right before cross-​clamp as this has a tendency to provoke potentially

unstable heart arrhythmia.
Once dissection is complete, and cardiac and abdominal harvest teams
ready, the patient is systemically heparinized and the main PA cannulated,
typically with a 6.5 mm high-​flow cannula. If the heart is not being procured
then separate aortic cannulation is not necessary. Cannulation of the PA
generally occurs a short length distal to the pulmonary valve; cannulating
too far distally on the main PA can lead to an uneven flush as the tip of
the catheter is directed down one of the two main PAs. Prostaglandin E1
(500 mcg) is directly injected into the main PA using a fine bore needle
and should be done prior to cross-​clamp as it can induce hypotension.
Alternatively, prostaglandin can be administered via the preservation solu-
tion by injecting it into the perfusate to ensure its delivery.
Immediately prior to cross-​clamping, the SVC tourniquet is then tightened
and preparations are made for venting the heart. If the wish is to vent via
the LA, the interatrial groove (Waterston’s or Sondergaard’s groove) is de-
veloped and a stab incision made for the purposes of venting. Alternatively,
the venting of the LV can occur through the LAA. The cardioplegia and
pulmonary perfusate are prepared for infusion, flushed, and de-​aired. The
right heart is vented via transection of the IVC at the level of the diaphragm.
With the heart adequately vented, the aorta is then cross-​clamped and
the cardiac and pulmonary preservation solutions are initiated. Typically,
preservation solution such as Perfadex (XVIVO Perfusion AB, Göteborg,
Sweden) is used. Note that cross-​clamping is not necessary in instances
when the heart is not being procured.
The preservation solution should be administered via gravity. During
antegrade perfusion, it is important to first confirm that the flush is running
efficiently by communicating with the preservationist. Once that has been
established, one must then proceed with placing ice slush into each pleural
space to offset the warm blood that tends to collect in this area and impede
uniform cooling of the lungs. Focus is then turned back to assessment of
the flush which is best done by paying close attention to the rate and colour
change of the flush as it exits the LA. Initially bloody, it should quickly dilute
and become clear. A full antegrade flush is at least 4 L. The antegrade flush
should not be administered with a pressure bag.
With completion of the perfusate, the heart is explanted. This can be
performed a number of different ways, but if done in the absence of a heart
team, our team generally transects the great vessels, beginning with the PA,
then the aorta, and finally both the SVC and IVC. In the presence of a heart
team, the successful negotiation of the pulmonary vein cuff is critical. One
of the subtleties of this step is realizing that with the heart elevated and on
stretch, the pulmonary vein cuff can appear satisfactory, but that it always
retracts upon cutting. One must also pay particularly close attention as the
heart surgeon cuts along the right pulmonary veins, an area in particular
where the cuff can be inadvertently shortened. Should this happen, and
it often does, it can be corrected by adding a strip of pericardium to the
edge of the vein cuff at the time of back-​table preparation. It is therefore
critical to procure excess donor pericardium as well for possible vascular
reconstruction.
Division of the LA usually begins with the LV elevated out of the chest
and a stab incision made halfway between the left inferior pulmonary vein
350

350 Chapter 26 Lung procurement

and the AV groove towards the base of the LAA (Fig. 26.1). It is then con-
tinued from the inside of the LA to include all four pulmonary vein orifices.
The ideal LA cuff includes a rim of LA muscle encircling all four pulmonary
vein orifices.
After removal of the cardiac structures, retrograde pulmonary perfusion
is carried out by delivering 2 L of perfusate into each of the pulmonary vein
orifices. The flow of perfusate out of the PA cuff is important to note. The
PAs are also inspected at this time to look for clot that might impede an
effective back flush.

IVC
LIPV
LA
RIPV

Fig. 26.1 Division of left atrium (LA). LIPV, left inferior pulmonary vein; RIPV, right
inferior pulmonary vein
 DBD lung procurement 351

The lungs are then explanted using the following steps. The pericardium
is cut sharply down to the hilum bilaterally, removed, and saved. The PA
is transected at the level of the pulmonary perfusion cannulation site. The
left lung is eviscerated, and inferior pulmonary ligament divided, identifying
the thoracic aorta. The right lung is then eviscerated, the inferior ligament
divided, and the anterior surface of the oesophagus bluntly dissected
(Fig. 26.2). This can be carried up under the lung bloc to free it from pos-
terior mediastinal attachments. With the lung bloc freed posteriorly, the
trachea is encircled with instructions to the anaesthesiologist to slowly re-
move the endotracheal tube. When an adequate transection site on the
trachea is identified, the anaesthesiologist is instructed to give a final Valsalva
breath of sustained 30 cmH2O pressure for several seconds with the endo-
tracheal tube completely retracted, and the trachea is stapled twice and
cut sharply between the two fires. Traction is placed on the bloc inferiorly
and the surgeon sharply divides additional soft tissue attachments in the su-
perior mediastinum. The lung bloc is removed from the chest (Fig. 26.3) and
placed in a sterile plastic transport bag filled with cold preservation solution
(no ice). The lungs are examined briefly prior to bag closure for surgical
damage, unexpected pathology, or inadequate margins. The lungs are triple
bagged with the outer bags containing ice.
Injuries to the PA can and do occur in general during the time of explant,
usually because the right PA, for instance, has not been adequately dis-
sected out during initial dissection. Proximal injuries are not a concern as
much of the right PA is sacrificed on the back table but more distal injuries
can require technical reconstructions that are readily avoidable if the SVC is
separated from the underlying artery. A short atrial cuff is also a common
problem, but again readily overcome by simply adding a lip of pericardium
to the abbreviated atrial cuff.

Posterior surface,
right lung

Trachea
Ao

Oesoph

Fig. 26.2 Dissection of posterior aspect of trachea. Oesoph, oesophagus.

352

352 Chapter 26 Lung procurement

PA

LA

Fig. 26.3 Explanted lungs.

Donation after circulatory death donor

lung procurement
The lung procurement operation in the DCD setting differs from the DBD
setting in that the dissection prior to explant is generally more limited. In
most hospitals, systemic heparin can be administered prior to extubation
and cardiac arrest, or alternatively in the perfusate solution. After death is
pronounced, the patient is transported to the operating room if not there
for withdrawal and reintubated. Sternotomy is performed, the heart ex-
posed, the PA cannulated prior to the bifurcation, and the heart vented
by direct cannulation of the RV/​IVC and aorta/​LA. The descending aorta
or abdominal aorta may be clamped to prevent perfusion of the bronchial
arteries with renal preservation solution. The pleurae are opened bilaterally
and the lungs inspected, atelectatic areas re-​inflated with recruitment man-
oeuvres, and topically cooled. Pulmonary perfusate is administered during
the initiation of dissection. The azygous is divided, and ascending aorta sep-
arated from right and the main PA. The interatrial groove is developed.
The pericardium is excised to the level of the pulmonary hila. The SVC and
IVC are transected and the posterior plane of the lungs are dissected from
the oesophagus and descending thoracic aorta. The aorta is transected as
distally as possible, leaving the ligamentum arteriosum as a cuff on the PA
to avoid PA injury. The separation of the heart from the lungs occurs by
dividing the LA as described previously (see p. 349). The trachea is bluntly
dissected, doubly stapled, and transected and any remaining posterior me-
diastinal attachments divided with scissors. Retrograde perfusion via each
pulmonary vein is performed. The lungs are removed and sterilely packed
for transport.
 DBD lung procurement 353

The use of DCD lung donors has increased in recent years due to dem-
onstrated comparable outcomes with DBD donors, although still remains
much less common than DBD. It has been theorized that the lungs may tol-
erate longer ischaemic times because the lungs have a relatively low meta-
bolic demand, and cells and alveoli have a high oxygen concentration, which
permits continued aerobic cellular metabolism. The lungs are resilient and
tolerate substantial cold storage time, which allows for thorough investi-
gation of the graft on the back table prior to packaging for delivery to the
recipient hospital.
The evaluation of DCD lung quality has been assisted by the develop-
ment of EVLP, described elsewhere in this text (see Chapter 27). Results
from abdominal organ transplantation suggest that allograft viability is
optimized if perfused with cold flush within 30 minutes of withdrawal of
life-​sustaining therapy; however, this interval has been demonstrated with
successful implantation up to 120 minutes of withdrawal of life-​sustaining
therapy. Typically, most centres accept 60–​90 minutes of warm ischaemic
time for lung DCD allografts. Whereas DBD donor criteria are gener-
ally widely accepted, several factors add complexity to selection of DCD
donors, such as warm ischaemic time, time of extubation, onset of low
blood pressure or oxygenation, the administration of heparin pretreatment,
and use of EVLP. A meta-​analysis pooling five studies reporting on DCD
lung transplantation (n =​271) compared with DBD (n =​2369) found no
difference in 1-​year mortality, either within the studies or in the pooled
analysis. There was also no difference in primary graft dysfunction or acute
rejection in the pooled analysis.
354
 Chapter 27 355

Ex vivo lung perfusion

Introduction 356
History of ex vivo lung perfusion 356
Criteria for EVLP assessment and reconditioning 362
Decision to transplant EVLP lungs 363
Surgical technique 363
Ventilation strategy 364
Cellular or acellular perfusate? 365
EVLP centres or EVLP in every centre? 365
Static or portable EVLP? 366
Platform for translational research 367
Conclusion 369
356

356 Chapter 27 Ex vivo lung perfusion

Introduction
Lung transplantation is now a well-​established treatment for end-​stage lung
disease with a median survival of >10 years in specific patient groups. There
is, however, a critical shortage of suitable donor lungs and a significant attri-
tion among those on the lung transplant waiting list.
The high susceptibility of donor lungs to injury and dysfunction during
the time of the donors’ demise and before procurement mean that <20%
of lungs from multiorgan DBD and as few as 6% of those from DCD are
at present deemed suitable for transplantation in the USA and the UK.
Despite a year-​on-​year increase in absolute transplant numbers, an annual
waiting list mortality of up to 20% persists.
Ex vivo lung perfusion (EVLP), whether it is mobile or static, has emerged
as a promising new technique for evaluating and reconditioning donor lungs
that would previously have been regarded as unusable. By increasing the
number of lung transplant procedures in individual centres by 15–​30%,
EVLP is already showing potential to substantially increase the availability of
suitable donor lungs. The clinical experience has rapidly evolved over the
past decade and with the introduction of portable EVLP, we might now face
a new era of lung preservation where EVLP may also improve outcomes
from donor lungs already deemed acceptable.

History of ex vivo lung perfusion

With the development of the first CPB machines in the 1950s grew an
interest in its potential in organ transplantation and various groups came
to investigate the effects that perfusion, storage, and preservation had on
tissue viability and function. In the case of hearts and lungs, the preserved
organs were found to be particularly delicate yet had to be capable of per-
forming at almost full normal function immediately following transplant-
ation. Early attempts at perfusing the lung were almost invariably frustrated
by oedema formation and deteriorating function.
Contribution of Steen
Originally designed to enable an improved assessment of lungs from DCD
donors, Stig Steen and colleagues in Lund, Sweden, developed the first well-​
functioning EVLP circuit for clinical use. A fundamental step towards success
was the development of Steen Solution, a buffered perfusate solution with
two major components. Firstly, a high albumin concentration to create an
‘ideal’ colloid osmotic pressure (about 30 mmHg) allowing physiological
perfusion pressures and flow to be maintained without the development
of either tissue oedema or dehydration. Secondly, a high dextran content,
which is designed to coat the endothelium and protect it from excessive
leucocyte interaction.
The preclinical work was translated by Steen and colleagues in 2000 when
they performed the first human lung transplantation using a DCD lung as-
sessed by EVLP and the first successful lung transplantation of an initially un-
acceptable donor lung reconditioned ex vivo in 2005. The Lund group EVLP
circuit provided the blueprint for all circuits currently used in clinical centres
worldwide. A pump generates a perfusate flow through a leucocyte filter
 HISTORY OF EX VIVO LUNG PERFUSION 357

and an oxygenator connected to a heater–​cooler unit and gas exchange

membrane before entering the lung through a cannula in the PA. Pulmonary
venous return is collected in a sterile reservoir either through an open LA
or through a closed LA cannula and is then recirculated. A ventilator is con-
nected to the trachea allowing ‘protective’ ventilation to be carefully started
only after the lungs have been rewarmed to 32°C and steadily increased
while approaching normothermia (Fig. 27.1).
In 2019, around 550 human donor lungs are expected to be transplanted
after EVLP reconditioning worldwide. Three main groups (Lund, Toronto
(Canada), and Hannover (Germany)) have contributed the research into
the physiology and technology to facilitate it and the theory behind their
creations differ slightly. The Lund and Toronto groups were the pioneers
inventing and spreading the technique of static EVLP around the world
Ventilator

Temperature
Tracheal probe &
cannula perfusate
sampling line

Filter
PA

LA

Flow probe

Pressure
measurement

Gas e Warmer
xchan
ger

Reservoir

Pulsatile pump
Fig. 27.1 Line diagram of an EVLP circuit. HCU, heater–​cooler unit.
358

358 Chapter 27 Ex vivo lung perfusion

and the Hannover group/​Organ Care System (OCS) Lung (TransMedics,

Andover, MA, USA) have since facilitated a portable adaptation of the EVLP
circuit (Table 27.1).
The Lund EVLP protocol is designed to mimic the physiological con-
ditions a transplanted lung will face after reperfusion in the recipient. It
serves primarily as a thorough assessment of borderline lungs not meeting
standard acceptance criteria to reclaim those demonstrating suitability for
clinical transplantation on the ex vivo circuit.

Table 27.1 EVLP protocols in clinical lung transplantation

Lund Toronto OCS Lung
Circuit Static Static Portable
Perfusion
Target flow 100% of cardiac 40% of cardiac 2.5 L/​min
output (70 mL/​ output
kg/​min)
PA pressure <20 <15 <20
(mmHg)
LA pressure 0 (open LA) 3–​5 0 (open LA)
(mmHg)
Pump Roller Centrifugal Pulsatile
Perfusate 2. Steen Solution 2. Steen 1.5. ‘OCS Lung
with red cell Solution solution’a with red-​
concentrates cell concentrates
(Haematocrit (haematocrit
10–​15%) 15–​25%)
Ventilation
Mode Volume Volume Volume controlled
controlled controlled
Tidal volume 6–​8 7 6
(mL/​kg)
Frequency 10–​15 7 10
(bpm)
PEEP (cmH2O) 5 5 5
FiO2 (%) 50 21 21
Temperature
(°C)
Start of 32 32 32
ventilation
Start of 15 15 32
perfusion
Start of 37 37 37
evaluation
a
The ‘OCS Lung solution’ is similar in composition to low-​potassium dextran with added
glucose, and does not contain any albumin (personal communication, Dr Warnecke, Study
Director of the OCS Lung INSPIRE Trial).
 HISTORY OF EX VIVO LUNG PERFUSION 359

Toronto: developing and validating EVLP

Shaf Keshavjee and his colleagues in Toronto have with their extensive con-
tributions changed the landscape of EVLP into a technique to significantly
improve transplantation rates. The focus of Keshavjee and Cypel’s studies
have not been to just evaluate if a graft is usable or not, but to prolong the
perfusion times to be able to potentially treat and better recondition injured
lungs before transplantation. They have most notably revised the Lund
protocol to possibly increase the option of longer-​term perfusion with:
• An acellular perfusate to avoid potential detrimental haemolysis.
• A low-​flow strategy with only 40% of estimated cardiac output to
reduce pulmonary vascular sheer stress and oedema formation.
• A closed circuit with both the PA and LA cannulated creating a positive
LA pressure.
In 2011, they presented their landmark results from the HELP trial in The
New England Journal of Medicine where they showed non-​inferiority in se-
vere PGD (grade 3) and 30-​day mortality between 20 transplants of initially
declined lungs reconditioned with EVLP compared to 116 standard lung
transplants performed in their centre. Table 27.2 shows the most pivotal
clinical EVLP trials to date.
Portable EVLP
In a pilot study published in The Lancet in 2012 followed by a multicentre
randomized controlled trial (INSPIRE) in The Lancet Respiratory Medicine
in 2018, Gregor Warnecke et al. investigated the effect of normothermic
preservation and transportation of standard criteria human donor lungs
on a portable EVLP system. Standard donor lungs were, instead of being
brought to their centres by means of cold preservation on ice, preserved
by normothermic perfusion and ventilation on the transportable OCS
Lung. These are the first reports of a portable EVLP system used in clin-
ical transplantation, with short-​term outcomes non-​inferior to controls.
Of the randomly assigned 370 patients in the INSPIRE trial, >80% under-
went transplantation, roughly divided between perfused lungs and controls.
Portable EVLP preserved lungs showed significantly lower rates of severe
PGD compared to standard cold preserved lungs after transplantation. A
feasibility study (the EXPAND trial) has since shown that portable EVLP can
result in a high utilization rate for also extended criteria donor lungs with
short and long-term post-transplant clinical outcomes comparable to what
we see in standard lung transplants preserved on ice.
The OCS protocol is a hybrid of the Lund and Toronto EVLP proto-
cols. A cellular perfusate based on OCS Lung Solution supplemented with
erythrocytes and an open LA is combined with a perfusate flow limited to
2.5 L/​min resembling the protective approach developed by the Toronto
group. Fig. 27.2 shows the semi-​automated EVLP circuits used in the listed
clinical trials in Table 27.2.
 360

360
Table 27.2 Pivotal trials investigating EVLP in clinical lung transplantation
Trial title Design Comparison Location EVLP EVLP Start Primary endpoint Outcome
protocol system date
HELP trial Prospective EVLP Toronto Toronto In-​house 2008 Primary graft No significant difference in PGD
Chapter 27

non-​ reconditioned XVIVO dysfunction rates or 30-​day mortality between

randomized donor lungs vs circuit 72 hours after EVLP reconditioned (n =​20) and
controlled trial standard criteria transplantation standard criteria transplant lungs
donor lungs (n =​116)
NOVEL Lung Prospective EVLP USA Toronto XPS 2011 Survival rates and 1-​year reported results shows
trial non-​ reconditioned rates of grade 3 no difference in PGD rates or
randomized donor lungs vs PGD at 72 hours early mortality between EVLP
controlled standard criteria with success reconditioned (n =​42) and
multicentre donor lungs if and only if standard criteria transplant lungs
trial the secondary (n =​42)
endpoint of
Ex vivo lung perfusion

survival at
3 years is met
OCS Lung Prospective OCS preserved Germany, OCS OCS 2011 Absence of Significantly lower incidence of
INSPIRE trial randomized standard criteria Belgium, Lung PGD3 within the severe PGD rates in the OCS
controlled donor lungs vs France, first 72 hours preserved lungs (n =​141) compared
multicentre standard cold Italy, after transplant to standard cold storage preserved
trial static organ Spain, and 30-​day lungs (n =​165). However,
preservation UK, USA, survival significantly higher 30-​day mortality
Canada, in the OCS preserved lung group
Australia compared to the control group.
Mortality was similar between
groups at 1 year
DEVELOP-​ Prospective EVLP UK Lund/​ Vivoline 2012 1-​year survival No significant difference in 1-​
UK trial non-​ reconditioned Toronto LS1 year survival between EVLP
randomized donor lungs vs reconditioned (n =​18) and
controlled standard criteria standard criteria transplant lungs
multicentre donor lungs (n =​184). Higher rates of severe
trial PGD and more ECMO required in
the EVLP group, study stopped
VIENNA trial Prospective EVLP Vienna Toronto In-​house 2013 Safety of EVLP No significant difference in PGD
randomized reconditioned XVIVO in standard rates or 30-​day mortality between
controlled trial standard donor circuit criteria lung EVLP reconditioned standard lungs
lungs vs standard donors (n =​35) and standard cold storage
cold static organ preserved lungs (n =​41)
preservation
International Prospective OCS USA, OCS OCS 2013 Composite of 30-​day survival in a cohort of 79
EXPAND non-​ reconditioned Belgium, Lung patient 30-​day transplanted recipients was 98.7%.
Lung Pivotal randomized extended criteria Germany survival and The utilization rate (number of
trial non-​controlled donor lungs, no PGD grade donor lungs transplanted/​number
multicentre controls 3 in the first of donor lungs placed on OCS)
trial 72 hours post was 87%
transplantation
HISTORY OF EX VIVO LUNG PERFUSION
361
362

362 Chapter 27 Ex vivo lung perfusion

(a) (b) (c)

Fig. 27.2 Semi-​automated EVLP circuits used in pivotal multicentre trials

investigating EVLP in clinical lung transplantation. (a) Vivoline LS1 (Vivoline Medical
AB, Lund, Sweden) static EVLP with the Lund EVLP protocol. (b) XPS (XVIVO
Perfusion AB, Göteborg, Sweden), static EVLP with the Toronto EVLP protocol.
(c) OCS Lung (TransMedics, Andover, MA, USA) portable EVLP with the OCS EVLP
protocol.

Criteria for EVLP assessment

and reconditioning
As recently described by Cypel and colleagues, the criteria for EVLP assess-
ment and reconditioning can be grouped into four different categories: ex-
tended criteria DBD, routine DCD, extended criteria DCD, and logistics
(extension of preservation time).
• Extended criteria DBD: this accounts for the majority of EVLP
indications. Donor lungs with marginal oxygenation, pulmonary oedema,
CXR infiltrates, unacceptable bronchoscopy, pulmonary emboli,
pneumonia, or aspiration can meet these criteria. EVLP can be used
simply to reassess organ quality or to recondition the lungs by means
of antibiotics, thrombolysis, or amelioration of pulmonary oedema by
optimal V/​Q strategies.
• Routine DCD: these are controlled DCD (Maastricht category 3–​5)
that present no concerning donor findings prior to or after withdrawal
of life-​sustaining therapies. Normally the time from withdrawal of
treatment to cardiac arrest in this group will be <30 minutes. There is
an ongoing discussion in the lung transplant community whether EVLP
is essential in this category. While some studies have demonstrated that
routine use of EVLP provides improved outcomes by ruling out some
organs with an unrecognized lung injury that ultimately would lead to
significant PGD, many centres have demonstrated excellent results with
routine DCD donors without the use of EVLP. Therefore, EVLP does
not appear to be a requirement for DCD lung transplantation, but may
aid in these more difficult assessments.
• Extended criteria DCD: these are uncontrolled DCD or controlled
DCD with concerning clinical features similar to high-​risk DBD or
donors with a delayed time to arrest (>60 minutes) after withdrawal
of life-​sustaining therapies. EVLP has potential to be very helpful in
expanding DCD transplants by confirming stable function of these
 Surgical technique 363

lungs where very little history or time for assessment is available.

If uncontrolled DCD donation would become a safely established
procedure it has the potential to completely alter donor graft
availability.
• Logistics: EVLP can give an indispensable logistical advantage by
extending preservation times from the traditional 6–​8 hours to >20
hours of total ‘ischaemic time’. Experimental and clinical studies have
demonstrated the safety of this approach which has the potential to
not only extend the donor pool, but to also bring lung transplantation
procedures to a semi-​elective operation with multiple benefits for
patients and clinicians. Common logistical reasons that could extend
donor lung ischaemic time >10–​12 hours and prevent standard donor
organ use are:
• Awaited viral studies.
• HLA compatibility studies.
• Pathology assessment of indeterminate mass in any donor.
• Awaited recipient admission.

Decision to transplant EVLP lungs

Most centres use the following combination of acceptance criteria after re-
conditioning to decide whether to use the lungs or not. All of the following:
• Any DBD or DCD donor lungs meeting criteria for standard transplant.
• PA pressure <20 mmHg, while achieving target perfusate flow.
• Oxygen capacity shown by deltaPaO2 of >300 mmHg (perfusate LA
PaO2 –​perfusate PA PaO2)/​FiO2.
• Selective pulmonary vein gas >225 mmHg on 100% FiO2 and
5 cmH2O PEEP.
• Stable or improving lung compliance and stable or falling lung resistance.
• No pulmonary oedema build-​up in the endotracheal tube.
• Satisfactory assessment on inspection and palpation including
bronchoscopic evaluation and deflation test (as a marker of
compliance).

Surgical technique
Technically, the procurement of lungs from an EVLP donor follows the
standard protocol, apart from the necessity to have a longer section of tra-
chea and PA for swift cannulation ex vivo. If a long enough main PA cannot
be achieved, due to simultaneous heart procurement, we usually resect a
segment of descending aorta which is anastomosed end-​to-​end to the main
PA to facilitate secure cannulation. After anterograde (through the PA)
pneumoplegia and retrograde (through pulmonary veins) Perfadex (XVIVO
Perfusion AB) flush and graft retrieval, the lungs are either connected to
the portable circuit or cool stored as new cold storage devices are now
available (LungGuard for example) and transported to the recipient site
where EVLP is ready to start in case of static EVLP. The lungs are connected
to the machine perfusion by cannulation of the PA and, if using a ‘closed
atrium technique’, by an end-​to-​end anastomosis between the LA cuff and
364

364 Chapter 27 Ex vivo lung perfusion

a venting cannula connected to the reservoir chamber. If an ‘open atrium

technique’ is used, the perfusate from the LA is drained freely and collected
by gravity in the reservoir chamber. A temperature probe is placed in the
drainage area of the LA cuff as a surrogate for core lung temperature. The
flow of perfusion is incremented slowly with a maximum perfusion pressure
of 15–​20 mmHg and the ventilation started at 32°C once the desired soft-
ening of the tissues has been achieved. During EVLP multiple examinations
are carried out, even selectively from each pulmonary vein, to assess the
function of lungs and lobes separately.
This type of selective evaluation, along with a visual and bronchoscopic
assessment, helps guide the decision of whether to accept the lungs for
transplantation, to further assess and recondition, or to discard the lungs
from clinical use and direct them to a strict research pathway.

Ventilation strategy
The re-​warming EVLP lung, having endured the donor’s demise, lung pro-
curement, cooling to 4°C, and then slow rewarming towards normothermia,
is vulnerable and highly sensitive to hyperventilation. The EVLP ventilation
strategy is therefore adapted from that of ARDS lungs.
In brief, mechanical volume-​controlled ventilation is initiated when tem-
perature in the perfusate outflow reaches 32°C:
Reconditioning phase
• Intermittent positive pressure ventilation (IPPV) setting.
• Pmax 20 to keep peak airway pressure <20 cmH2O.
• PEEP 5 cmH2O.
• FiO2 to 0.21–​0.5.
• Set I:E ratio 1:2 and inspiratory pause at 10%.
• Lund advocates an incremental approach to ventilation, beginning
with a minute volume of 1 L/​min at 32°C (tidal volume 3 mL/​kg) and
increasing by 1 L/​min for each degree Celsius, finally reaching 6 L/​min
at 37°C (tidal volume 5–​7 mL/​kg). Breaths are 50% oxygen given at
5–​20 breaths/​min with a PEEP of 5 cmH2O.
• Toronto recommends an immediate tidal volume of 7 mL/​kg at a rate
of 7 breaths/​min, 21% oxygen, and with a PEEP of 5 cmH2O.
• If persistent atelectasis, perform a recruitment manoeuvre either by
careful hand ventilation or by transiently increasing PEEP from 5 cmH2O
in increments up to 10 cmH2O for a few breaths while always keeping
airway pressure (Paw) <25 cmH2O.
Evaluation phase
Once rewarming is complete and target perfusion established, the function
of the donor lungs undergoing EVLP can be assessed. Once the perfusate
is deoxygenated and confirmed on blood gas analysis, perform recruitment
manoeuvres as previously described and set the ventilator for evaluation
as shown here:
• Increase FiO2 via the ventilator from 21–​50% to 100%.
• PEEP can be increased to a maximum of 8 cmH2O for a short period.
• Keep peak airway pressure (Paw) <25 cmH2O.
 EVLP centres or EVLP in every centre? 365

• Perform blood gas analyses 15 minutes after FiO2 is increased to 100%.

• A lung deflation test is performed by disconnecting the tracheal tube
at the end of inspiration. Remember to first reduce perfusate flow to
maximum of 1.5 L/​min to avoid tracheal oedema. Recoil of the lungs
is evaluated subjectively; global collapse of the lungs is defined as
normal.
• If transplant suitability criteria have not been achieved, return to the
reconditioning phase.
• If transplant suitability criteria have been achieved, move immediately to
cooling phase for organ preservation.
• Before discontinuing ventilation, perform a bronchoscopy for final
assessment and toilet. Stop ventilation at 32°C, clamp trachea
with lungs partially inflated with 50% FiO2, and carry on with the
transplantation.

Cellular or acellular perfusate?

One of the fundamental questions in EVLP is whether or not to use blood
in the circulating perfusate. The Lund and OCS protocols advocate a cel-
lular perfusate, whereas the Toronto technique uses an acellular solution.
Comparative animal studies have shown no significant difference in lung
performance between the two.
The EVLP logistics are simplified, costs lowered, and ethical conflicts
arising from the use of limited blood products in studies of organs that may
not be used clinically are prevented by avoiding erythrocyte concentrate.
The oxygen supply to the lung cells during EVLP also appears to be suffi-
ciently provided by the ventilator alone without the need of oxygen carriers
for parenchymal preservation. A potential benefit of a cellular perfusate is
the oxygen binding capacity that erythrocytes provide and its possible ad-
vantage during lung evaluation. It has been shown that blood gas analyses
during EVLP with an acellular perfusate are unreliable and could disguise
oxygenation deficits during assessment, especially in the case of a V/​Q mis-
match. It is also argued that the presence of red blood cells provides a
more physiologically relevant assessment of flow through the pulmonary
microvasculature.

EVLP centres or EVLP in every centre?

There is an ongoing debate on how the EVLP service should best be de-
livered in the future. The Toronto group has shown the feasibility of the
‘EVLP centre’ approach in a case report with the team in Chicago, USA. To
manoeuvre around the restriction on EVLP in the USA outside the NOVEL
Lung trial, a pair of unacceptable donor lungs was transported on ice to
Toronto for reconditioning and then flown back to Chicago for successful
implantation. Thus, concentrating the volume to fewer centres will presum-
ably increase the experience of the EVLP team. The most obvious draw-
backs of establishing such specialized EVLP centres are transportation and
environmental costs, more complex logistics, and the inevitable prolonged
cold ischaemia inflicted on an already injured organ.
36

366 Chapter 27 Ex vivo lung perfusion

By either approach there is witness of less quantifiable benefits with the

mere availability of the EVLP technique. In Milan, Italy, they stress how the
implementation of EVLP in their low-​volume centre has facilitated the safe
use of extended criteria donor lungs in recent years and in the UK we have
seen a clear increase in the numbers of donors assessed by our retrieval
teams since EVLP was put into practice in 2009. Currently, the retrieving
surgeon frequently manages to optimize intended EVLP donors, who would
previously not have been approached, to reach standard transplant criteria.
Initially rejected lungs are therefore now being brought immediately to im-
plantation as often as taken back to our centre for reconditioning, an im-
portant ‘side effect’ of EVLP provision.

Static or portable EVLP?

The optimal method for lung preservation remains unclear. In the process
of lung transportation, cold storage at 4–​8°C has routinely been used to
decrease cellular metabolic activity and preserve lung function. This can,
however, compound lung injury due to ischaemia-​reperfusion and ATP
depletion and increase the risk of PGD. A large registry review in 1999
of 5052 lung transplants, reported higher 30-​day mortality with cold is-
chaemic times exceeding 8 hours. Since the introduction of improved
extracellular lung-​preservation solutions, ischaemic times of up to 10–​
12 hours have been safely reported. Cypel and colleagues showed that
normothermic EVLP can interrupt hypothermic ischaemic lung injury after
12 hours of cold storage and ameliorate immediate transplant outcomes
in a porcine model. What effect shorter periods of cold ischaemia have
on graft function and patient survival in lung transplantation is still to be
elucidated.
The INSPIRE lung trial showed that in comparison to static cold
storage, portable EVLP reduced the amount of PGD. There is still much
uncertainty about whether or not the benefits of this approach will
balance its considerable costs and logistical hurdles. If it does however,
the portable EVLP technique has the potential of leading the way of a
new era in lung assessment and organ preservation. This could enable
transport of organs over longer distances, opening up many new pos-
sibilities for organ allocation and therapeutic intervention. An exciting
avenue is its implication for combined transplants where there are now
reports of lung–​liver transplantation being done liver first with the lung
kept on EVLP. Liver first has several technical and physiological advan-
tages, but is only feasible with EVLP. With distance no longer a factor in
donor–​recipient matching, closer matches could also be achieved. This
could decrease rejection and the need for immunosuppression and ex-
tend graft lifespan, while increasing access to transplantation for disad-
vantaged patient populations. Organs could also more easily be routed
through specialized facilities, which have been suggested by several
groups as a way to make technically challenging assessment, repair, func-
tional augmentation, or banking procedures a clinical reality. Thus, ap-
proaches that today would not be seriously considered could become
practical and fruitful areas of innovation.
 Platform for translational research 367

Platform for translational research

The recent success of EVLP in clinical lung transplantation is also contrib-
uting significantly to advances in translational research in the lungs. The
EVLP circuit has allowed many types of research projects, including pharma-
cotherapy, stem cell therapy, gene therapy, and organ/​tissue engineering.
Results obtained using lungs on EVLP have potential to be highly translat-
able to clinical practice. In theory, EVLP should allow the administration
of the most effective medications based on the specific causes of lung in-
juries while avoiding systemic toxicity. Some notable successes have been
reported using this approach.
Antimicrobial treatment
The hazard of respiratory infections in the early post-​transplant period is
notorious and donor-​to-​host transmission of bacterial and fungal infections
is a known cause of morbidity in the immunosuppressed lung transplant
recipient.
With a localized closed circuit, EVLP is ideal for high-​dose antimicrobial
treatment with no risk of side effects to other organs. Extended criteria
donor lungs subjected to the assessment form by nature a subpopulation
likely to have a higher microbial load than standard lungs. All clinical EVLP
protocols therefore use prophylactic broad-​spectrum antibiotics in their
perfusate solution. Several studies have shown that EVLP with high-​dose
antimicrobials in the perfusate is associated with an effective reduction of
both the bacterial and fungal burden of the donor lung. We empirically use
meropenem and amphotericin B, and recommend the addition of a broad-​
spectrum fungicide to the perfusate. Yeast is a well-​known cause of infec-
tions in the immunosuppressed post-​transplant patient and is frequently
cultured in the BAL fluid from these borderline donor lungs.
Thrombolysis for pulmonary embolism
Fibrinolytic treatment may be of highest importance in DCD donors when
timely heparin use is not possible. Urokinase administration during EVLP
has been shown to reduce PVR and improve oxygenation in a preclinical
DCD model by Inci et al. The same group later reported a clinical case
where normothermic EVLP was used as a platform to deliver therapeutic
thrombolysis (urokinase) in lungs with known massive pulmonary emboli
followed by complete clot fragmentation and successful transplantation of
the treated lungs.
HCV-​infected donors
A growing interest in the use of HCV-​infected donors followed from the
current epidemic of overdose deaths in the North American donor popu-
lation. Some geographic areas in the USA report up to 20% of all organ
donors being NAAT positive for HCV. Underuse of these organs is par-
ticularly relevant given that they are often young with less comorbidity
than other donors. However, organs from donors with contagious viral in-
fections are traditionally not offered for transplantation due to a high risk
of transmission. In a recent report, using normothermic EVLP as a treat-
ment platform, the Toronto group presented a method for treatment of
HCV-​infected human donor lungs with physical clearance and light-​based
368

368 Chapter 27 Ex vivo lung perfusion

therapies that efficiently prevented HCV transmission without deleterious

effects on the lung or perfusate solution. This strategy of treating viral in-
fections in a donor organ during preservation could significantly increase the
availability of organs for transplantation and encourages further clinical de-
velopment, especially in an era where available rescue therapies using novel
direct-​acting antivirals have been shown to be very effective.
Acute lung injury
Interleukin (IL)-​8 and other proinflammatory proteins have been suggested
by several groups as prognostic markers in lung transplantation indicating
severity of donor lung injury. We could in a study of 42 human donor
lungs clinically investigated for transplantation demonstrate the potential
role of interleukin-​1β as a biomarker of EVLP reconditioning and more
importantly post-​transplant survival. Interestingly, IL-​1β levels in the EVLP
perfusate correlated with extent of neutrophil adhesion to conditioned pul-
monary endothelial cells (R2 =​0.33, p <0.001), upregulation of ICAM-​1 in
donor lung vasculature (R2 =​0.68, p <0.001), and upregulation of ICAM-​1
(R2 =​0.30, p =​0.001) and E-​selectin (R2 =​0.29, p =​0.001) on condi-
tioned pulmonary endothelial cells, and importantly neutralization of IL-​1β
in perfusate strongly inhibited neutrophil adhesion to conditioned endothe-
lium (91% reduction, p =​0.002). Therapeutic targeting of IL-​1β or other
proinflammatory components therefore provides one of many exciting
opportunities to decrease endothelial activation and potentially reduce
the incidence of early graft injury post transplantation. If successful, these
principles may expand to other organs experiencing ischaemia-​reperfusion
injury.
Blood type conversion
In a recent Science publication Wang and colleagues showed that deple-
tion of donor lung A-Ag can be achieved with EVLP treatment. By adding
antigen cleaving enzymes to the perfusate they managed to effectively con-
vert group A (ABO-A) donor lungs to group O (ABO-O) universal blood
type lungs within 4hr of EVLP. This strategy has great potential to expand
ABO-incompatible lung transplantation and lead to improvements in fair-
ness of organ allocation.
Stem cells
Because of a number of evolving technologies, there have been rapid ad-
vances in the field of stem cell therapy over the past few years and stem
cell therapy has become a potential future for treatment of chronic lung dis-
ease. The beneficial effects of mesenchymal stem cells have been confirmed
in lungs on EVLP with various types of acute lung injury. In Escherichia coli-​
injured human lungs, for example, tracheal instillation of mesenchymal stem
cells during EVLP restored alveolar fluid clearance, reduced inflammation,
and exerted antimicrobial activity. It has also been reported that intravas-
cular administration of mesenchymal stem cells in pig lungs on EVLP leads
to decreased circulating IL-​8 levels, suggesting potential therapeutic effects
against ischaemia-​reperfusion injury and PGD. Multipotent adult progenitor
cells are another promising resource for stem cell therapy with notable
anti-​inflammatory effects in lungs on EVLP. Mesenchymal stem cells and
multipotent adult progenitor cells are able to regenerate damaged tissue
 Conclusion 369

with new cells. However, the primary role of stem cell therapy in recovery
from acute lung injury appears to be attributed to paracrine signalling
from the cells by regulating epithelial and endothelial permeability, thereby
enhancing alveolar fluid clearance and lessening the immune response in
injured lungs.
Gene therapy
Similar to drug therapy, gene therapy administered during EVLP is also a
feasible targeted approach for reconditioning. A pilot study of IL-​10 gene
transfection using adenovirus infection during EVLP of porcine lungs was
reported by the Toronto group with notable success and improved post-​
transplantation outcomes. This was an encouraging study that foreshadows
the potential major impact of gene therapy during EVLP. We must be
mindful of viral toxicity, however, as adenoviral vectors can induce inflam-
mation. Non-​viral gene delivery methods during EVLP may be valuable
alternatives.

Conclusion
As organ perfusion has become one of the ‘hot topics’ in all fields of trans-
plantation in recent years, EVLP continues to lead its progression. Its appli-
cation in clinical lung transplantation is spreading rapidly across the globe at
the same time as exciting research into new ways to treat and recondition
the perfused lung continues. Mapping and targeting the compounded inflam-
matory insults caused by brain death, aspiration, infection, and reperfusion
and limiting oedema formation by haemoconcentration of the perfusate are
some of the more promising advancements in recent years. EVLP offers a
unique platform tailored to this clinical and investigative work, aiming to find
novel ways to increase the donor pool, and to make lung transplantation a
reality for more waiting list patients with life-​threatening lung disease.
370
 Chapter 28 371

Technical aspects
of lung transplantation
Lung transplantation operative technique 372
Special situations 380
372

372 Chapter 28 Technical aspects of lung transplantation

Lung transplantation
operative technique
There are many methods and techniques for performing a lung transplant
operation. Each variation in technique has advantages and disadvantages.
The prototypical lung transplant operation is bilateral sequential lung trans-
plantation performed through a bilateral anterior thoracosternotomy. This
is the most common way that transplantations are performed in North
America and Europe. However, it is by no means the only method or best
method. Differences in technique arise because of clinical situations, institu-
tional resources, and surgeon preferences.
In this chapter, we will describe the technical aspects of performing bi-
lateral orthotopic lung transplantation (BOLT), sequentially via a bilateral
anterior thoracosternotomy (clamshell incision). During each step, we will
discuss variations in technique, and situations where perhaps such variation
may be reasonable, or even preferred alternatives to the standard method.
Anaesthetic considerations
After identification of a suitable donor, and confirmation that the donor
lungs are of good quality, then the recipient operation is started. The re-
cipient procedure begins with the induction of anaesthesia, obtaining
IV access, placement of adequate monitoring, and positioning on the
operating table.
Specifics of anaesthesia are discussed in Chapter 37.
Pulmonary hypertension with right heart dysfunction
Patients with severe PH and right heart dysfunction are a particularly risky
group during induction. These patients are susceptible to haemodynamic
instability during induction that may be catastrophic. If standard anaes-
thetic induction is undertaken, there is a high probability that the agents
will compromise any pre-​existing right heart dysfunction to the point of
inducing systemic hypotension and even cardiac arrest. The dysfunctional
RV is largely dependent on an adequate arterial blood pressure for myo-
cardial perfusion. Even minimal periods of hypotension can induce signifi-
cant RV ischaemia, further compounding the right heart dysfunction. The
result is further dysfunction and the creation of a vicious cycle resulting in
cardiac arrest which is extremely difficult to recover and will necessitate
‘crashing’ on cardiopulmonary bypass or ECLS. This situation is potentially
avoidable.
In such patients, a safer strategy is to place adequate central access and
monitoring lines prior to induction. Radial arterial line and central access
via the IJ vein or femoral vein are placed in the awake patient under local
anaesthesia. Care is taken to avoid hypoxia. Placement of the PAC is not
necessary at this point and may induce arrhythmias. These arrhythmias are
usually inconsequential, but in a patient with a tenuous right heart, they
may be detrimental. A vascular sheath may be inserted but the PAC is only
floated after the chest is opened. Once adequate vascular access is ensured,
anaesthesia induction is performed under invasive haemodynamic moni-
toring to avoid cardiac depression. When performed in this manner, any
instability can be detected early and treated promptly to avoid a spiralling
situation and catastrophe at induction.
 Lung transplantation operative technique 373

In cases of extremely severe right heart dysfunction and an unstable pa-

tient, it may be prudent to have the femoral areas exposed and ready for
initiation of emergent femoral–​femoral ECLS. Placement of femoral arterial
and venous catheters will expedite the process in case of the need to ‘crash
on’. Taking it one step further, the patient may be placed on ECMO prior
to induction. After placement of adequate venous access and monitoring
lines, and prior to induction, the femoral artery and vein are exposed under
local anaesthesia. They are cannulated and low-​flow ECMO is initiated. This
technique has been described with good results. Alternatively, cannulation
can be performed percutaneously using a Seldinger technique. Preclosing of
the femoral artery using a percutaneous vascular closure device facilitates
decannulation without open repair.
Choice of incision
Bilateral anterior thoracosternotomy
This incision, also known as the ‘clamshell’ incision, is the most commonly
used incision for performing a BOLT. The incision provides excellent ex-
posure to the pulmonary hilum, visualization of the phrenic nerves, and
access to the heart for central cannulation. Access to the chest wall is
good for managing pleural adhesions. However, adhesions in the posterior
costophrenic angles may be challenging to manage. The major disadvantage
of this incision is the transverse sternal transection and the potential for
sternal wound complications.
The skin incision is made starting in the midline at a point two-​thirds
of the way between the sternum and the xiphoid process. The incision is
extended laterally to the point that is one fingerbreadth below the nipple
on each side, and then carried slightly superiorly in a curvilinear fashion
(Fig. 28.1). In women, the incision is carried along the inframammary folds
to avoid incising the breast tissue. The subcutaneous tissue and the pectoral
muscles are transected. The thoracic cavity is entered in the fourth or fifth
interspace, depending on the size of the cavity. Both internal mammary
arteries are ligated and transected. A self-​retaining retractor is placed on
each side.
Sternal-​sparing bilateral/​unilateral anterior thoracotomy
For single orthotopic lung transplantation (SOLT) performed without car-
diopulmonary support, there is no need to transect the sternum. The op-
eration is performed via an anterolateral thoracotomy or a posterolateral
thoracotomy. For the anterolateral thoracotomy, the rib may or may not
need to be transected.
For BOLT, a sternal-​sparing approach has been described. The operation
is performed using bilateral anterolateral thoracotomy incisions; however,
the sternum and mammary arteries are not transected. The landmarks are
otherwise similar to the ‘clamshell’ approach.
Sternotomy
Median sternotomy is an option for performing lung transplantation. The
incision can be performed rapidly, making it the incision of choice in cases
where cardiac arrest happens on induction. It is also easier to close when
compared to an anterolateral thoracotomy. The incision is particularly ad-
vantageous for women with pendulous breasts. Large breasts may result in
significant issues with wound healing for the anterolateral incisions. There
374

374 Chapter 28 Technical aspects of lung transplantation

Fig. 28.1 Illustrates the bilateral anterior thoracosternotomy (‘clamshell’) incision

and positioning of the retractor for optimal exposure.

are two main disadvantages. The first is that access to the pulmonary hilum
may be difficult, especially on the left, without using intraoperative ECLS.
Second, exposure to the lateral and posterior chest wall is limited, and cases
with significant intrapleural adhesions may result in a more difficult pneu-
monectomy and increased bleeding.
Posterolateral thoracotomy
This is the least commonly used incision; however, it is an option for the
performance of BOLT or SOLT. Exposure to the pulmonary hilum is excel-
lent through this incision. Its primary disadvantage during BOLT is the need
to reposition patients for the second lung. Also, access for central cannula-
tion is more difficult when compared to the anterior approaches. In cases
where ECLS is anticipated, especially for a left posterolateral thoracotomy,
the femoral area needs to be prepared and draped into the field.
Intraoperative cardiopulmonary support
Regarding intraoperative cardiopulmonary support, there are two schools
of thought. In one type of practice, intraoperative cardiopulmonary sup-
port is used routinely for all transplants. The second school of thought,
which is also the more prevalent, is to use intraoperative cardiopulmonary
support only when needed. Each has its own attributes. The former has
the advantage of a smoother intraoperative haemodynamic course with
 Lung transplantation operative technique 375

minimal hypotension. It also allows controlled low-​pressure reperfusion of

the lung grafts after implantation. Exposure is also easier, especially for the
performance of the left implantation. The advantage of the latter is a reduc-
tion of the complications associated with CPB or ECLS, such as bleeding.
Both approaches yield acceptable results, and the decision to proceed with
one over the other is surgeon or centre dependent.
Technique of intraoperative cardiopulmonary support
Traditionally, intraoperative support was always initiated with full CPB,
usually via central cannulation. However, with improvements in technique,
ECMO has become an option. With the elimination of a reservoir and an
air–​blood interface, the intensity of anticoagulation required for ECMO is
much less than that required for CPB. Studies comparing intraoperative
ECMO to CPB showed a reduction in intraoperative bleeding, transfusion
requirements, and ICU length of stay with no difference in 90-​day mor-
tality. As a result, ECMO has largely replaced CPB as the default means to
provide intraoperative support. However, there remain situations whereby
CPB may still be preferred. Therefore, the means of intraoperative support
should be selected based on the operative scenario encountered.
In centres where intraoperative support is used selectively, a situation
commonly arises where the cardiac function is satisfactory, but lung func-
tion is not enough to sustain the patient without some sort of support.
This may occur as a result of severe lung disease, or a patient who was
bridged with pre-​transplant ECMO. Occasionally, this situation arises after
implantation of the first lung as a result of PGD affecting the newly trans-
planted lung. During such situations, single-​lung ventilation will result in sig-
nificant desaturation and substantial hypoxia. Traditionally, these patients
were managed by the initiation of CPB. However, if haemodynamic support
is not required, and only respiratory support is needed, then there may
not be a need for CPB. In such a situation, it may be possible to provide
adequate support with VV ECMO. In this situation, femoral–​femoral VV
ECMO can be initiated to provide intraoperative pulmonary support. Once
the lungs are implanted and reperfused, then an attempt at weaning can be
performed before closure, or the patient can be transferred to the ICU
where weaning can be undertaken there.
If additional cardiac support is needed, then VV ECMO will not suffice.
Situations such as known CAD, pre-​existing cardiac dysfunction, or inadequate
exposure are better managed with CPB or VA ECMO. As mentioned above,
VA ECMO is preferred over CPB given the lesser degree of anticoagulation,
reduced transfusion requirements, and reduced complications.
There are rare situations where CPB may be the optimal support method
in cases where concomitant cardiac surgery is required, especially if it is
intracardiac surgery (e.g. lung transplantation in a patient with PH and a
repairable congenital cardiac defect). An additional situation where CPB
is advantageous is when significant intraoperative bleeding is expected.
Certainly, it would seem counterproductive to use CPB over ECMO, and by
consequence, a higher level of anticoagulation, in a situation where bleeding
is encountered. However, massive blood loss is much better handled by
using pump suckers and recirculating the blood through the CPB circuit,
rather than putting it through a cell saver and transfusions to maintain an
adequate intravascular volume.
376

376 Chapter 28 Technical aspects of lung transplantation

Recipient pneumonectomy
• Determining laterality:
• For transplants planned without the use of extracorporeal support,
single-​lung ventilation is required, and the contralateral lung will
support the patient during the explant pneumonectomy and
implantation. The preoperative V/​Q scan can be used to determine
which lung should be transplanted first. Typically, the lung with the
least amount of perfusion is transplanted first, so the functionally
better lung can support the patient during single-​lung ventilation. For
procedures performed with ECLS, the sequence of pneumonectomy
is unimportant.
• The phrenic nerve is identified early and protected.
• Dissection and division of the superior pulmonary vein with a vascular
stapler.
• Dissection and division of the PA:
• The PA is dissected and encircled. Care is taken to ensure that the
PAC is withdrawn to avoid including it in the staple line. A test clamp
of the PA is performed to ensure that the patient tolerates single-​lung
perfusion. If test clamping is not tolerated, then ECLS is required.
This will need to be initiated before completing the pneumonectomy.
• The truncus anterior branch is isolated and stapled, followed by the
remaining artery going to the lower and middle lobes. Transecting the
artery at the level of its branches gives more length to perform the
anastomosis. Excess length can always be trimmed when preparing
the hilum.
• Once the PA is transected, the ventilation to the ipsilateral lung is
ceased. Single-​lung ventilation before clamping the main PA causes
significant desaturation and should be avoided.
• Incision of the inferior pulmonary ligament.
• Dissection and division of inferior pulmonary vein.
• Dissection and division of the bronchus:
• The bronchus is transected just proximal to the take-​off of the
upper lobe bronchus, and the anastomosis is performed at this level.
One could transect the lobar bronchi and trim the excess length off
subsequently. Bronchial arterial branches and peribronchial lymph
nodes will need to be controlled with judicial use of electrocautery,
clips, or suture ligation.
• Completion of pneumonectomy:
• Once the bronchus is transected, the pneumonectomy is completed.
Occasionally, adhesions to the chest wall, diaphragm, or mediastinum
are encountered. These can be easily incised, and the specimen is
sent off to pathology.
• Preparation of the hilum (Fig. 28.2):
• The next step is to prepare the recipient’s hilum for the implantation.
If not already performed, the bronchus is trimmed to the point
that is just proximal to the take-​off of the upper lobe bronchus.
Overzealous dissection of the peribronchial tissues should be avoided
to maintain the blood supply of the recipient bronchus, and it is not
necessary to dissect more than one or two rings for the performance
of the anastomosis. During dissection of the posterior mediastinum,
bleeding can be encountered from the paraesophageal and subcarinal
 Lung transplantation operative technique 377

Fig. 28.2 Illustrates preparation of the hilar structures and mobilization of the right
main bronchus in preparation for anastomosis. Note the position of the phrenic
nerve during mobilization of hilar structures.

lymph nodes. When obtaining haemostasis, care should be taken not

to injure the oesophagus or the vagus nerves, which are particularly at
risk during this part of the procedure.
• The PA is dissected circumferentially and as far centrally as possible
to provide adequate length for the anastomosis. The pulmonary
vein stumps are gently retracted, and the pericardium is opened
circumferentially to allow space for placement of a vascular clamp.
Donor lung preparation
• The donor lung/​s is/​are unpacked:
• The lung block is immersed in cold saline and covered in cold gauze
or towels to keep the lungs cool during the preparation.
• Inspection of the lung block for quality or iatrogenic injuries.
• The posterior pericardium is split.
• The LA cuff:
• The LA is split exactly down the middle. The atrial cuffs are trimmed
to size. Occasionally, especially if the heart was procured for
another recipient, the LA cuff may be left too short. This situation is
encountered occasionally but is easily corrected by sewing a patch
on the atrial cuff. This could be a patch of pericardium from the
donor block, or autologous pericardium from the recipient, or bovine
pericardium.
• The pulmonary trunk is divided at its bifurcation.
• The tracheobronchial tree is divided.
• The is usually achieved by double stapling the left mainstem bronchus
just distal to the carina. The airway is transected distal to the staple
line, and the excess length is trimmed. Care should be taken not to
denude the bronchus of the peribronchial tissue and damage its blood
supply. Finally, the bronchus is trimmed leaving no more than one
cartilaginous ring from the origin of the upper lobe bronchus. This
avoids having a long ischaemic donor bronchus and helps to minimize
the incidence of anastomotic complications.
378

378 Chapter 28 Technical aspects of lung transplantation

Implantation
• The donor lung is placed in the pleural cavity in the correct orientation.
• Bronchial anastomosis:
• The first step of the implantation is the bronchial anastomosis (Fig.
28.3). This is typically performed with an absorbable suture with a long
absorption time. Our preference is to use 4-​0 polydioxanone (PDS) on
a tapered needle. The anastomosis can be performed as a single running
suture technique. Alternatively, the membranous wall of the bronchus
can be approximated using a single running suture, and the cartilaginous
portion can be approximated with several interrupted sutures.
The technique for bronchial anastomosis is dependent on surgeon
preference, and both techniques yield similar results. Testing of the
anastomosis is not always necessary but is easily achieved by submerging
the anastomosis in water and ventilating that lung to a sustained pressure
of 25 cmH2O. The appearance of air bubbles signifies a defect in the
anastomosis which may need reinforcement with additional sutures.
• Pulmonary arterial anastomosis (Fig. 28.4):
• A vascular clamp is placed on the recipient’s proximal PA and the
vascular staple line is excised to lay open the stump. The arterial
anastomosis is performed with a single continuous suture of 4-​0 or
5-​0 polypropylene suture. The vascular clamp is left in place until all
anastomoses are complete and the lung is ready to be perfused.
• LA anastomosis (Fig. 28.5):
• The atrial anastomosis is next. A curved vascular clamp is placed on the
recipient LA. The vascular staple lines of the two pulmonary veins are
excised and the junction incised to provide a LA cuff in preparation for
the anastomosis. The anastomosis is performed with a single continuous
suture of 4-​0 polypropylene. In order to avoid LA thrombosis, every
attempt should be made to restore endo-​atrial continuity by everting and
excluding the atrial cut edges and epicardial fatty tissue. Once the suturing
is completed, the sutures are left long and not tied to allow for de-​airing.
• Reperfusion and de-​airing:
• Once the anastomoses are completed and verified, the lung can be
reperfused and de-​aired. The patient is placed in steep Trendelenburg
position. The lung is ventilated gently. After a few puffs, the arterial
vascular clamp is opened partially, and the lung is reperfused. The atrial
clamp is left in place, the atrial suture line is left loose, and the de-​airing is
performed through the atrial suture line (Fig. 28.5). Once the graft is de-​
aired, the atrial clamp is removed, and the suture is tied. The pulmonary
arterial clamp is then released slowly over 10 minutes. Reperfusion is
performed in a controlled fashion to avoid injuring of the graft.
• Alternatively, retrograde de-​airing can be performed to wash the lung
preservation solution. Using this technique, the LA clamp is opened
first, allowing the blood to flow retrograde and out through the
pulmonary arterial suture line, washing out the air and preservation
solution. Once adequate de-​airing is performed, then antegrade flow
can be re-​established.
• After completion and reperfusion of the first lung, attention is turned
to the contralateral side. For cases performed on ECLS, attention
must be given to ensure reperfusion of the newly transplanted lung.
Typically, the flow rate is reduced to ensure pulsatility, creating partial
bypass support. PA pulsatility can be monitored by floating a PAC.
 Lung transplantation operative technique 379

Donor left
atrial cuff

Do
no
rr
igh
t lu
ng

Fig. 28.3 Illustrates the technique of bronchial anastomosis.

Fig. 28.4 Illustrates the technique for PA anastomosis.

380

380 Chapter 28 Technical aspects of lung transplantation

Fig. 28.5 Illustrates the technique for pulmonary vein–​LA anastomosis. Note the
clamp is kept on the PA and sutures untied in preparation for de-​airing following
completion of the pulmonary vein anastomosis.

Closure
Once haemostasis is assured and the patient is separated from ECLS, then
the chest can be closed. It is standard to leave at least one or two chest
tubes in each pleural space.

Special situations
Extensive adhesions
Pleural adhesions are sometimes encountered in patients requiring
lung transplantation. These are typically not extensive, and intrapleural
adhesiolysis can be performed without much added difficulty. However, in
patients with a history of extensive intrapleural surgery, septic lung condi-
tions such as CF or bronchiectasis, or certain disease processes, such as
pleuroparenchymal fibroelastosis and coal worker’s pneumoconiosis, the
adhesions may be significant.
Dealing with such adhesions can be challenging and may increase mor-
bidity. First, it is crucial to gain safe access to the pleural cavity. Identification
of the phrenic nerve is important early in the dissection for its protection.
In patients who previously underwent lung volume reduction surgery, the
staple line may be adherent to the anterior mediastinum where the phrenic
nerve is located.
If the hilum is spared from adhesions, the hilar dissection can proceed in a
standard fashion. However, if extensive adhesions affect the hilum and dis-
section is treacherous, then an option is to commence the dissection within
 Special situations 381

the pericardium. The planes within the pericardium are typically unaffected
in these cases. The pericardium is opened in the midline, and the vessels are
identified within the pericardium. They are dissected from the inside of the
pericardium, but it is important to transect the vessels outside the pericar-
dium to obtain adequate lengths to perform the anastomoses.
Retransplantation
Patients who undergo retransplantation can be managed similarly to pa-
tients with extensive pleural adhesions. The major difference is that hilar
adhesions are more extensive. Furthermore, if the pericardium was ac-
cessed during the initial transplantation, then intrapericardial adhesions will
make intrapericardial dissection more difficult. It is imperative to have the
ability to initiate CPB or ECLS rapidly in case of vascular injury or severe
haemodynamic embarrassment. Once the incision is made, the first step
is to commence intrapericardial dissection and ensure that there is access
to the aorta and RA. This manoeuvre allows for rapid cannulation and ini-
tiation of CPB. If intrapericardial adhesions are too extensive, then access
to the femoral vessels should be considered. In certain situations, it may be
beneficial to initiate ECLS or CPB before commencing the hilar dissection.
Once the recipient pneumonectomy is performed, and the hilum is pre-
pared, then the implantation can proceed in a similar fashion to a first-​time
lung transplantation.
Postoperative ECLS
Occasionally, postoperative ECLS is needed for respiratory or haemo-
dynamic support. This is most commonly required in patients who de-
velop PGD. Increased FiO2 may increase the production of free oxygen
radicals and, therefore, prolonged exposure to high FiO2 may exacerbate
ischaemia-​reperfusion injury and PGD. As such, it is advisable to avoid high
FiO2 and have a low threshold of commencing VV ECMO when the FiO2
requirements reach 80% or higher. In such situations, VV ECMO is initiated
peripherally with bifemoral cannulation, or dual cannulation of the right IJ
vein and femoral vein.
On rare occasions, postoperative VA ECMO may be required. This is usu-
ally the case in patients with severe PH and RV dysfunction. Postoperative
VA ECMO is typically established via dual femoral cannulation. However, if
there is any element of PGD, the VAV ECMO should be initiated to avoid
upper body and myocardial hypoxia due to Harlequin syndrome.
382
 Chapter 29 383

Critical care management

and primary graft
dysfunction following
lung transplantation
Introduction 384
Physiology of the lung transplant recipient 384
Postoperative considerations 385
Postoperative complications and management 387
Conclusion 388
384

384 Chapter 29 Postoperative care after lung transplantation

Introduction
Lung transplantation is the gold standard therapy for appropriately selected
individuals with end-​stage lung disease. Lung transplantation is a durable
therapy that offers improvements in overall survival and quality of life for
recipients. As a resource-​intensive therapy that is in part limited by the
availability of suitable donor allografts, outcomes are appropriately highly
scrutinized. Excellent multidisciplinary care is a prerequisite to successful
outcomes following lung transplantation.
Successful management of the lung transplant recipient in the periopera-
tive period is the culmination of not just postoperative decision-​making and
care, but rather is the composite outcome of candidate management in the
pre-​transplant and intraoperative period, donor selection, and intraoperative
technique. As such, outcomes after lung transplantation are heavily impacted
by the decisions made with respect to the candidate (now recipient), donor
lung allograft, and intraoperative technique and management.
As is the case across many disciplines within cardiothoracic surgery, cur-
rent literature notes that complications occur at a predictable rate, and that
centres with excellent outcomes demonstrate a superior ability to both
diagnose and manage these early complications. Postoperative care of the
lung transplant recipient requires constant vigilance for the early detection
and rescue from postoperative complications.

Physiology of the lung

transplant recipient
While the approach to the care of the lung transplant recipient must be
individually tailored to the recipient’s history and indication for transplant-
ation, donor history and associated considerations, and technical factors
from the procedure itself, several overarching principles should be con-
sidered in the early postoperative period. This early postoperative man-
agement centres on fluid balance and ventilation. Adequate tissue perfusion
and gas exchange must be preserved despite the need to minimize IV fluid
administration, cardiac work, and barotrauma. Selection of sedation, an-
algesia, and strategies for haemodynamic support are all informed by the
need to optimize fluid balance and ventilatory performance in the imme-
diate postoperative setting.
Two physiological considerations are unique to all lung transplant recipi-
ents and warrant consideration in the postoperative period. First, the trans-
plant allografts are by convention denervated following implantation, and as
such lack the cough reflex that otherwise protects in part the functioning
lung from pulmonary infections. This highlights the need for aggressive pul-
monary toilet, as well as careful assessment of the swallowing reflex and
for the presence of reflux. Second, the mandatory period of ischaemia fol-
lowed by reperfusion—​along with the disruption of the lymphatic drainage
of the lung allograft—​results in a predisposition to increased vascular per-
meability and susceptibility to oedema. The management of fluid balance
as such requires attention to the minimization of the development of pul-
monary oedema despite this increased vascular permeability.
 Postoperative considerations 385

Medical histories of both the donor and recipient bear consideration in

individualizing the approach to postoperative management. Candidates—​
now recipients—​differ significantly in their underlying aetiology of lung dis-
ease, which may necessitate changes to their postoperative management.
Lung transplant recipients with CF may require ongoing antibiotic therapy
tailored to the microbiome of their (explanted) lungs, while those with vas-
cular indications may require more aggressive management of PA pressures
in the early postoperative phase. Donors, likewise, increasingly are ac-
cepted despite histories of pneumonia or potentially transmissible disease.
Antibiotic therapy for the recipient can similarly be tailored to culture data
from the donor taken at the time of procurement. As donors with a history
of hepatitis C—​usually absent evidence of viraemia—​are increasingly con-
sidered for use in lung transplantation, institution-​specific protocols for the
use of direct-​acting antivirals should be employed to decrease the risk of
disease transmission to the recipient.
Like other cardiothoracic procedures, mechanical support may be indi-
vidually tailored to the recipient depending on the degree of haemodynamic
support required to tolerate the procedure and the need for concomitant
cardiac procedures. Lung transplantation may be performed with inotropic
support without mechanical support, with an IABP, on VV or VA ECMO,
or on full CPB. Each strategy carries advantages and disadvantages with
respect to the degree of haemodynamic (or ventilatory) support provided,
the propensity towards bleeding and coagulopathy, and the degree of ex-
pected systemic inflammatory response. The appropriate selection of these
strategies to perform lung transplantation safely is described in Chapter 28.
The recipient’s postoperative physiology may therefore likewise be im-
pacted by the choice of MCS to perform the procedure.

Postoperative considerations
Sedation and analgesia
Lung transplantation is a necessarily invasive procedure. The degree
of postoperative pain may be impacted by the incision selected to at-
tain adequate exposure to the thorax. Bilateral lung transplantation
may be safely performed via median sternotomy or by bilateral an-
terior thoracosternotomy (clamshell incision) or thoracotomies (‘sternal
sparing’). In concordance with an overarching desire to optimize ventila-
tory function in the postoperative setting, the selection of an appropriate
analgesic strategy should be made to minimize respiratory depression
and facilitate pulmonary toilet. Multimodal analgesic strategies can help to
decrease the need for narcotic pain medication. The use of an epidural
catheter is generally well tolerated from a haemodynamic standpoint and
results in less respiratory depression than strategies that rely more on nar-
cotic pain medication. Preoperative placement of an epidural catheter is
likewise associated with decreased use of narcotic pain medication and
shorter time to extubation than the use of an epidural catheter placed
postoperatively. Erector spinae and other blocks may provide analgesia
without the small, but real risk of an epidural haematoma from the epi-
dural catheter.
386

386 Chapter 29 Postoperative care after lung transplantation

Ventilator management
Barotrauma and high airway pressures may lead to ischaemia of the bron-
chial mucosa or shearing injury to the alveoli. As such, lung-​protective
ventilatory strategies are required as soon as the allograft lungs are tran-
sitioned on to the ventilator. Adjustment of the FiO2 as low as possible
to attain a PaO2 >70 mmHg is conventional and may be associated with
decreased rates of PGD. Lower tidal volumes—​such as to 10–​12 mL/​kg—​
will reduce peak airway pressure. Remember that the donor size dictates
tidal volume ratios, not necessarily the recipient. Serial blood gas examin-
ation is employed to ensure adequate oxygenation and acid–​base status,
and to facilitate appropriate weaning of supplemental oxygen. Recipients
are extubated after a period of haemodynamic and ventilatory stability
and having tolerated conventional weaning of pressure support ventilation.
Regular bronchoscopy and endotracheal suctioning—​in addition to aggres-
sive pulmonary toilet and incentive spirometry—​reduce mucus plugging
and atelectasis. Early tracheostomy for those recipients unable to promptly
wean from the ventilator helps achieve early mobilization and adequate en-
teral nutrition.
Infection prophylaxis
Standard intraoperative antibiotic prophylaxis following lung transplant-
ation includes empiric broad Gram-​negative and Gram-​positive coverage
(cefepime for Gram-​ negative coverage, vancomycin for Gram-​ positive
coverage) and fluconazole for Candida prophylaxis. Coverage may be
tailored—​or discontinued—​as donor cultures finalize as negative. The need
for antifungal prophylaxis may be tailored to observed local rates of inva-
sive candidiasis.
Patients with known pretransplant colonization with antimicro-
bial pathogens, such as those with CF, are evaluated by transplant
infectious disease for development of a customized perioperative anti-
biotic regimen. The duration of such pathogen-​directed antimicrobials
should be determined in a multidisciplinary fashion given that complete
source control should be obtained at the time of recipient preparatory
pneumonectomy.
TMP–​SMX 80/​360 mg daily is a first-​line agent for Pneumocystis jirovecii
prophylaxis starting 7 days after the transplantation. Nystatin swish and
swallow four times daily for the first 6 months post transplantation is
standard prophylaxis for oral candidiasis.
Viral prophylaxis is dependent on donor and recipient CMV status.
Recipients who are at risk for CMV are treated initially with ganciclovir
and transitioned to valganciclovir orally daily. Recipients with prior ex-
posure to CMV are continued on CMV prophylaxis for 12 months post
transplantation. Those high risk for CMV disease due to donor CMV
IgG positivity without pre-​transplant recipient exposure are continued
on prophylaxis indefinitely as tolerated. For recipients who are both
donor and recipient CMV IgG negative, aciclovir prophylaxis is given IV
initially and then at a dose of 400 mg orally twice daily for the first 6 post-​
transplant months.
 Postoperative complications and management 387

Postoperative complications
and management
Haemorrhage
Haemorrhage is an uncommon but potentially life-​threatening complication
of lung transplantation. Extensive adhesiolysis necessitated by prior cardiac
or pulmonary surgery may increase the risk of postoperative bleeding, as
does the use of mechanical support such as CPB or ECMO. Meticulous
haemostasis and correction of undesired coagulopathy is mandatory in the
immediate postoperative period.
Airway complications
Airway stenosis, dehiscence, and necrosis are all uncommon complications
following lung transplantation. Immediate complications more frequently
reflect technical error, while late complications may manifest as a result of
tenuous anastomotic perfusion or due to infection. Early failure may require
reoperation to revise and correct technical error. Late complications may
be better managed with more conservative measures such as dilation and
stenting and less frequently require operative intervention.
Primary graft dysfunction
PGD after lung transplantation is the major contributor to both morbidity
and early mortality. Prompt diagnostic workup and a high index of suspicion
are mandatory to evaluate for alternative causes of respiratory failure, such
as vascular torsion, infection, cardiogenic oedema, or hyperacute rejection.
These reversible causes must be promptly addressed to correct ventilatory
status. The direct mechanistic sequence leading to PGD is not yet eluci-
dated, though retrospective studies suggest that the degree of ischaemic-​
reperfusion injury contributes significantly. Several factors may decrease the
risk of PGD. During the transplant procedure, several manoeuvres may be
employed to minimize the extent of reperfusion injury experienced by the
allograft. In addition to the use of extracellular preservation solutions, ad-
ministration of IV methylprednisolone (500 mg) and mannitol (25 mg) prior
to reperfusion of both allografts have been shown to reduce the degree
of reperfusion injury. Importantly, reperfusion is performed in a controlled
fashion over a period of 10–​15 minutes when done off of CPB. Similarly,
ventilation and lung recruitment should be held until the newly implanted
lung has rewarmed. Inhaled NO or other pulmonary vasodilator can used
to decrease PVR during the operation. If additional pulmonary vasodilation
is thought to be necessary, the patient can be continued on inhaled NO
or inhaled epoprostenol after initial stabilization in the ICU and prior to
extubation.
Those patients exhibiting severe PGD despite preventive and less in-
vasive rescue manoeuvres are considered for mechanical support with
ECMO. Those with peak inspiratory pressures approaching 30 cmH2O
and requiring FiO2 >0.60 after excluding other causes for failure are con-
sidered candidates for post-​transplant ECMO. VV ECMO provides short-​
term support while lung recovery is anticipated. Support can be initiated
at the bedside by way of a single dual-​lumen cannula, although many can-
nulation strategies can be employed as the situation necessitates. Once
38

388 Chapter 29 Postoperative care after lung transplantation

ECMO support is established, patients are transitioned to lung-​protective

ventilatory settings with low pressures and FiO2 of 0.21. Of those patients
requiring VV ECMO post transplantation, the vast majority are successfully
weaned from support as graft performance improves. Patients are typic-
ally weaned from ECMO within 24–​72 hours as evidence of pulmonary
recovery is observed. Though survival rates of those experiencing PGD
continue to improve with advances in ECMO technology, PGD continues
to decrease overall survival rates and leads to a decrease in overall graft
function once free from ECMO support.

Conclusion
Lung transplantation is a life-​saving therapy for candidates with end-​stage
lung disease. Improvements in donor and candidate selection and care,
surgical and anaesthetic technique, critical care management, and immuno-
suppressive strategies have made lung transplantation safer than ever be-
fore. Exceptional care of both the donor and the candidate/​recipient at all
phases of the transplantation process is critical to the overall outcome of
the recipient. In the immediate postoperative period, utmost vigilance is
required for the early detection of and rescue of the recipient from a small
number of predictable but impactful complications. The ability of the care
team to safely recover the lung transplant recipient from the transplant pro-
cedure is a critical facet of quality care delivery in thoracic transplantation.
 Chapter 30 389

Management
of rejection following
lung transplantation
Acute cellular rejection 390
Antibody-​mediated rejection 393
Induction agents in lung transplantation 394
Maintenance immunosuppression drugs 396
390

390 Chapter 30 Rejection in lung transplant

Acute cellular rejection

Despite advances in surgical, anaesthetic, surveillance, and transplant im-
munosuppression strategies, acute lung rejection remains a common post-​
lung transplant complication seen in lung transplant patients within a year
after transplantation. Rejection is often diagnosed on clinical grounds,
using a composite of symptoms, lung function, imaging, DSAs, and
immunohistological examination of transbronchial lung biopsy specimens
by experienced multidisciplinary teams. A third of lung recipients will have
an episode of treated rejection within the first year. Further ACR/​AMR
are major risk factors for early graft loss and CLAD. Registry data from the
ISHLT indicate that ACR is responsible for 3.6% of deaths within 30 days
after transplantation.
Risk factors for ACR
See Table 30.1.

Table 30.1 Risk factors for ACR

1 Degree of HLA mismatch: HLA-​DR, HLA-​B, HLA-​A, etc.

2 PGD
3 Infections: bacterial, fungal, community-​acquired respiratory virus
(CARV), CMV
4 Gastro-​oesophageal reflux disease
5 Non-​compliance with immunosuppression

Aetiology
The immunological mechanisms of lung allograft rejection are complex
and multifaceted. From a simplistic perspective, lung allograft transplant-
ation exposes the host immune system to a multitude of damage and
pathogen-​associated molecular patterns that test the discriminatory ability
of the lung innate and adaptive immune response systems. Non-​self-​major
histocompatibility complex patterns from the donor-​derived antigens and
pathogens are recognized by direct, indirect, and semi-​direct pathways of
allo-​recognition involving the T-​cell receptor (TCR)–​CD3 complex. This is
followed by allo-​activation of the T cells in the presence of co-​stimulatory
signals driving complex adaptive effector immune responses that lead to graft
rejection or tolerance. Effective immunosuppression is therefore one of the
key strategies for dampening the immune response to the lung allograft to
minimize acute rejection of the graft and improve long-​term outcomes.
Clinical manifestations and diagnosis
Symptoms of acute rejection are often unreliable and non-​specific, including
shortness of breath, cough, and low-​grade fever. CXR can demonstrate
new evolving pleural effusions or persistent effusions and infiltrates. De novo
DSAs may be detected. Symptoms and signs are more often pronounced in
severe acute rejection and help direct appropriate investigations including
imaging, bronchoscopy evaluation, and transbronchial biopsies. Routine sur-
veillance transbronchial bronchoscopies also help in identification of ACR
in asymptomatic patients particularly in the early post-​transplant period.
 Acute cellular rejection 391

Laboratory results
There are no pathognomonic markers of ACR on routine blood panels.
Their use is limited to help identify adequacy of trough CNI levels
(tacrolimus/​ciclosporin), infective, haematological, renal, and hepatic ab-
normalities post transplantation.
Pulmonary imaging
Serial CXRs are routinely obtained for all patients after lung transplantation.
The presence of bilateral pulmonary infiltrates or new evolving pleural effu-
sions or persistent pleural effusions are harbingers of acute rejection. High-​
resolution CT is not always helpful; however, the presence of interlobular
septal thickening and bilateral ground-​glass changes in the absence of fluid
overload/​consolidation or atelectasis can be complementary for diagnosis
of acute rejection.
Pulmonary function tests
Spirometry is routinely performed post transplantation on the ward, at
home, and during follow-​up visits after discharge. Our protocol for spirom-
etry is at 2 weeks post transplantation and then bi-​weekly until discharge
from hospital in a majority of uncomplicated lung transplantations. The
spirometer marker of clinical interest in the FEV1 and a decline of ≥10%
from a stable baseline in the absence of confounding factors such as pain/​
sedation/​depression or technical quality control issues should trigger add-
itional investigations. While spirometry is helpful particularly when there
are no infection/​bronchial anastomotic concerns, specificity for diagnosis
of ACR is low.
Bronchoscopy and transbronchial biopsy (TBBX)
Surveillance bronchoscopy and TBBX are critical to histological gold
standard confirmation of acute ACR and AMR. This allows examination
of the bronchial anastomosis, removal of endobronchial plugs, microbio-
logical confirmation of infections by BAL, and tissue confirmation of acute
rejection based on the ISHLT criteria. Bronchoscopy and TBBX protocols
vary across transplant centres. Our institutional practice is to perform these
at 1 month, 3 months, 6 months, and 12 months routinely with additional
bronchoscopies when clinically indicated, often by new-​onset decrease in
lung function, concerns of infection, or changes in radiology. A minimum
of five pieces of well-​expanded alveolar parenchyma along with terminal
bronchioles is required for reliably diagnosing ACR by dedicated, experi-
enced transplant pathologists after exclusion of acute infection. Acute rejec-
tion is characterized by perivascular mononuclear infiltrates with or without
endothelialitis. In practice, we provide six to nine good-​quality biopsy speci-
mens performed under conscious sedation and fluoroscopic guidance. The
ISHLT diagnostic criteria for grading and severity assessment of ACR are
provided in Table 30.2.
Emerging biomarkers
Emerging biomarkers for ACR include liquid biopsies, that is, quantification
of circulating donor-​derived cell-​free DNA from the allograft in plasma
(ddcfDNA) by next-​generation sequencing platforms, and are of immense
research interest and awaiting prospective validation. Other markers of fu-
ture interest include micro-​RNA (miRNA) signatures in peripheral blood.
392

392 Chapter 30 Rejection in lung transplant

Table 30.2 Pathological grading of ACR

Grade Meaning Appearance
A Perivascular
inflammation
0 No Normal lung parenchyma
1 Minimal Scattered, infrequent small mononuclear
perivascular infiltrates; no eosinophils
2 Mild More frequent perivascular infiltrates seen at
low magnification, eosinophils may be present
3 Moderate Dense perivascular infiltrates, eosinophils, and
neutrophils common. Pathognomonic feature is
extension into alveolar septa and air spaces
4 Severe Diffuse perivascular, interstitial, and air-​space
infiltrates with pneumocyte damage and
features of acute lung injury
B Airway-​
associated
inflammation
0 None No evidence of bronchiolar inflammation
1R Low grade Single layer of mononuclear cells in bronchiolar
submucosa
2R High grade Large infiltrates of larger and activated
lymphocytes in bronchiolar submucosa, with
potential involvement of eosinophils and
plasmacytoid cells
X Ungradable No bronchiolar tissue

Treatment of ACR
Treatment of ACR is by pulse dose steroids (usually methylprednisolone)
and the decision to treat often depends on TBBX histological grading of
rejection. Grade A2 (mild rejection) and above are definitely treated.
Asymptomatic A1 (minimal rejection) and isolated B grade may not merit
treatment and centres vary on their management of asymptomatic minimal
rejection. At the least, such patients will need to have close clinical follow-​
up with lung function and surveillance TBBX at 4–​6 weeks.
Our institutional policy to use IV pulse methyl prednisolone (10 mg/​kg
for 3 days followed by 1 mg/​kg on a tapering schedule) within the first
3 months after transplantation, with oral prednisolone pulses 1 mg/​kg being
used after 3 months. Surveillance lung function and TBBX needs to be re-
peated at 4–​6 weeks after treatment of ACR to identify persistent ACR,
which is a risk factor for CLAD progression.
The presence of persistent ACR despite augmentation is a marker of
coexistent AMR and should trigger investigations for de novo DSAs and
immune-​histological phenotyping (C4d) of TBBX tissue for AMR. In the
first instance, the pulse steroid therapy should be repeated and CNI switch
to tacrolimus can be considered if the patient has had two prior steroid
 Antibody-mediated rejection 393

augmentations. Other potential treatment considerations are a simultan-

eous switch to MMF from azathioprine. If there is recurrent/​refractory
high-​grade ACR, antibody depletion-​based treatments with rabbit anti-​
thymocyte globulin (rATG) or alemtuzumab (anti-​ CD52) can be con-
sidered. Depending on institutional practice, expert centres would consider
early total lymphoid nodal irradiation or extracorporeal photopheresis in
recurrent ACR to prevent CLAD progression.

Antibody-​mediated rejection
The role of AMR in the development of lung allograft rejection has been
rapidly growing in the last few years. However, it remains a complex patho-
logical and clinical process.
AMR is associated with different forms of rejection including:
• Hyperacute rejection, a rare but fulminant rejection occurring
following reperfusion of the allograft and due to pre-​existing anti-​HLA
antibodies.
• Candidates for transplantation may have pre-​existing anti-​HLA
antibodies following pregnancy, blood transfusion, or organ
transplantation.
• Acute AMR, believed to be consequence of the development of de
novo DSAs.
• ACR.
• CLAD.
The presence of pre-​existing anti-​HLA antibodies prior to transplant-
ation and/​or the development of de novo DSAs after transplantation is
detected with fluorescent-​based solid-​phase assay (Luminex bead tech-
nology). Generally, multiple levels of testing are performed to evaluate the
presence and the specificity of circulating anti-​HLA antibodies. The first
level uses a pool of different class I and class II antigens assessing the pres-
ence or absence of antibodies. The last level of testing will use single HLA
antigen beads to determine antibody specificity. A note of caution in the
interpretation of anti-​HLA antibodies tests: the level and function of circu-
lating anti-​HLA antibodies should not be based on the MFI of the assays,
because the MFI does not represent the strength or the titre of circulating
anti-​HLA antibodies. There is no consensus currently on how to measure
the strength of circulating DSAs. Further available options to estimate the
strength of circulating anti-​HLA antibodies includes serial dilution assay, IgG
subclasses analysis, and/​or C1q binding activity.
Differently from other SOTs, in lung transplantation the lack of specific
diagnostic features and the variable relationship between DSAs and clinical
presentation pose a challenge in defining the clinical diagnosis and treat-
ment. Clinical manifestations of acute AMR are non-​specific. Patients tend
to present with general respiratory symptoms: dyspnoea, cough, and hyp-
oxaemia. Radiologically, diffuse pulmonary opacities is the most common
finding on CXR. Histopathologically, the presence of capillaritis in lung allo-
graft tissue has been seen occasionally in cases of AMR. C4d deposition
has been an inconsistent finding in lung transplant biopsy and its role in the
diagnosis has been controversial.
394

394 Chapter 30 Rejection in lung transplant

Based on the presence of (1) allograft dysfunction, (2) histopathology,

(3) C4d staining, and (4) circulating DSA, AMR is currently divided into clin-
ical and subclinical types: clinical AMR if allograft dysfunction is present, and
subclinical if there is not allograft dysfunction. Subcategorization includes
‘definitive’, in case of all criteria being present; ‘probable’ in case of three
criteria; and ‘possible’ in case of two criteria.
Treatment of AMR
There is no consensus on the optimal treatment for AMR due to the ab-
sence of clinical trials. Based on single-​centre experiences, a combination
of drugs directed to decrease the circulating anti-​HLA antibodies and the
production of new DSAs are often used. To date, the optimal regimen for
the treatment of pulmonary AMR is unknown.
Agents that are often used include:
• IVIG: despite the exact mechanism of action being unknown, IVIG is
the foundation of most AMR treatment regimens. IVIG may neutralize
DSAs, inhibit complement activity and cytokine gene activation, and
downregulate B cells. IVIG is typically dosed at 500–​2000 mg/​kg.
• Anti-​CD20 monoclonal antibody: rituximab is able to rapidly bind to CD-​
20 expressed on pre-​B and mature B cells causing cell lysis. Generally,
375 mg/​m2 of rituximab is administered once a week for 4 weeks.
However, the optimal dose and number of cycles are also unclear.
• Proteasome inhibitors: bortezomib and carfilzomib. These two drugs are
directed against proteasome components, which are required by plasma
cells to complete production of proteins, including antibodies. Typically,
four doses of 1.3 mg/​m2 of bortezomib are given. Due to risk of
infection and other side effects associated with bortezomib, carfilzomib,
a newer proteasome inhibitor, has been recently used for the treatment
of AMR along with plasmapheresis and IVIG.
• Plasmapheresis: the goal of this therapy is to deplete circulating anti-​HLA
antibodies and mitigate graft dysfunction by separating and discarding
plasma component and replacing it with albumin or fresh frozen plasma.
There is no consensus of the number of plasmapheresis sessions
required.

Induction agents in lung transplantation

The use of strong immunosuppression in the perioperative or early
postoperative period to reduce the initial robust T-​cell response against
the lung allograft was initially motivated by evidence from other SOTs
such as kidney, heart, and liver transplantation. However, due to the lack
of significant evidence of a clear benefit and the concerns for increased
risk of infection, the use of immunosuppressive strategies for induction
varies from institution to institution. The use of induction immunosuppres-
sion in the recent era is associated with decreased incidence of BOS and
increased survival. Despite it not being considered part of the induction
therapy, just before reperfusion of the allograft, high-​dose glucocorticoids
(methylprednisolone 500 to 1000 mg IV) are administered to reduce the
risk of reperfusion injury.
 Induction agents in lung transplantation 395

IL2R antagonists
Basiliximab and daclizumab are chimeric monoclonal antibodies that act by
binding to activated T-​cells expressing CD25, blocking cell activation and
proliferation.
• Basiliximab is dosed intraoperatively or immediately after lung
transplantation at 20 mg on the first and fourth days after
transplantation and present a half-​life of 13 days with an effective IL2R
saturation of 30 days.
• Daclizumab is used at 1 mg/​kg within the first 24 hours post
transplantation and then repeated for an additional four doses every 2
weeks. Daclizumab has a half-​life of 20–​40 days with an effective IL2R
saturation of 120 days.
The use of daclizumab compared to depleting agents and/​or monoclonal
antibodies showed no differences in the rate of acute rejection, freedom
from BOS, and 2-​year survival.
Infusion of basiliximab prior to implantation led to a lower rejection score
during the first year after transplantation compared to infusion after im-
plantation or no induction. However, no change in survival or freedom from
BOS was observed.
Lymphocyte-​depleting agents
Two T-​cell-​depleting agents are polyclonal immunoglobulins: horse anti-​
thymocyte globulin (Atgam) and rATG.
• Atgam is administered IV at a dose of 5–​15 mg/​kg per day for the first
day to 14 days following lung transplantation.
• rATG is used at 1.5 mg/​kg IV over 6 hours then two or three additional
doses given 24 hours apart.
During infusion, cytokine release syndrome including fever, rigors, rash, and
myalgia, and post-​infusion leucopenia, thrombocytopenia, immune-​com-
plex glomerulonephritis, and serum sickness have been described.
Conflicting results have been observed with the use of these agents as
induction agents. The use of rATG as an induction agent showed a reduced
incidence of acute rejection and trend towards reduced BOS when com-
pared to no induction. However, these findings were not confirmed in the
follow-​up study with no change in survival at 8 years. A large randomized
controlled multicentre trial evaluating the efficacy of ATG-​Fresenius (5
and 9 mg/​kg, respectively) compared to no induction failed to show re-
duction in acute rejection, graft loss, or death in the first year after lung
transplantation.
Monoclonal antibodies
Monoclonal antibodies used in the induction phase are Muromonab–​CD3
(OKT3) and alemtuzumab. OKT3 was voluntarily withdrawn from the
United States market in 2009.
Alemtuzumab is a monoclonal antibody that targets CD52, a cell surface
receptor expressed by B and T cells, macrophages, monocytes, and nat-
ural killer cells. Alemtuzumab leads to cell depletion including B and T cells
through direct cellular and complement-​mediated cytotoxicity.
Alemtuzumab is dosed at 30 mg IV over 2 hours. It is associated with
a cytokine release syndrome that can be attenuated by pre-​ emptive
396

396 Chapter 30 Rejection in lung transplant

administration of acetaminophen 500–​1000 mg orally, diphenhydramine 50

mg orally or IV, and methylprednisolone 15 mg/​kg about 30 minutes prior
to infusion. The T-​cell depletion can last up to 3 years after infusion, while
B-​cell depletion lasts about 3 months.
Use of alemtuzumab is associated with comparable 5-​year survival but
increased freedom from acute rejection and BOS in comparison with anti-​
thymoglobulin, daclizumab, and no induction regimens.
Alemtuzumab is associated with an increased risk of opportunistic
infections.
In conclusion, all induction agents provide benefits with regard to acute
rejection but are more modest in terms of long-​term survival. Newer
strategies are being used in other SOT recipients with more tailored ap-
proaches (i.e. belatacept in living-​donor kidney transplant recipients) and
hopefully soon will start making recommendations based on more signifi-
cant evidence.

Maintenance immunosuppression drugs

Protocols
After lung transplantation, a maintenance immunosuppression regimen
including three drugs is considered the standard of care. This regimen in-
cludes one CNI, tacrolimus or ciclosporin; a corticosteroid, usually pred-
nisone; and a cell cycle inhibitor, mycophenolate or azathioprine.
According to ISHLT registry data, the most common regimen both at 1-​
and 5-​year follow-​up is tacrolimus, mycophenolate, and prednisone. mTOR
inhibitors, sirolimus and everolimus, have been used as an alternative to a
cell cycle inhibitor, or to minimize CNI dosing and more recently as an ad-
junct agent in the setting of rejection (Table 30.3).
Management of a highly sensitized host
The importance of pre-​and post-​transplant management of HLA anti-
bodies in transplant candidates remains an important limiting factor to long-​
term survival.
The development of HLA antibodies may be due to prior pregnancies,
connective tissue disease, previous transfusion, or prior organ transplant-
ation. Certain infections and ECMO may lead to cross-​reactivity antibodies
and thus detectable HLA antibodies prior to transplantation.
Acceptance of a donor with HLA to the recipient’s detected HLA anti-
bodies has been associated with episodes of hyperacute rejection, and
worse early post-​transplant outcomes.
Lung transplant recipients with detectable pre-​transplant HLA antibodies
demonstrated greater number of ventilator days, worse 30-​day mortality,
higher rate of post-​transplant detectable HLA antibodies, and an increased
risk of developing BOS, a presentation of CLAD.
• The most common type of pre-​transplant management of HLA
antibodies is avoidance of any detectable donor-​specific HLA
antibodies.
 Maintenance immunosuppression drugs 397

Table 30.3 Maintenance immunosuppression drugs

Class Target trough Side effects
level
Tacrolimus CNI 5–​15 ng/​mL Nephrotoxicity, neurotoxicity,
hyperglycaemia, hypertension,
hyperlipidaemia, hyperkalaemia,
hypomagnesaemia
Ciclosporin CNI 75–​200 ng/​ Nephrotoxicity, neurotoxicity,
mL hyperglycaemia, hypertension,
hyperlipidaemia, hyperkalaemia,
hypomagnesaemia
Sirolimus mTOR 5–​15 mg/​ Decreased wound healing,
Everolimus inhibitors mL leucopenia, thrombocytopenia,
hypertriglyceridemia,
proteinuria, and pneumonitis
Myco Cell cycle 250–​1000 Leucopenia, thrombocytopenia,
phenolate inhibitor mg twice and GI disturbances (diarrhoea,
a day abdominal pain, nausea,
vomiting)
Azathioprine Cell cycle 2 mg/​kg Leucopenia, thrombocytopenia,
inhibitor daily anaemia, hepatotoxicity,
pancreatitis
Prednisone Corticost 5–​20 mg Hypertension, weight
eroid daily gain, hyperlipidaemia,
hyperglycaemia and diabetes
mellitus, osteoporosis,
cataracts, poor wound healing,
psychiatric disturbances

• Desensitization strategies for lung transplant candidates with elevated

PRAs consist of regimens including IVIG, plasmapheresis, rituximab,
carfilzomib, or bortezomib. However, no significant improvements in
terms of survival or development of CLAD was noted. A different
approach based on clinical risk stratification, presence of DSA anti-​
HLA antibodies, and organ availability including intraoperative plasma
exchange prior to reperfusion of the first graft and followed by a total
of five sessions postoperatively, IVIG, and rATG (dosed based on
the results of the complement-​dependent cytotoxicity cross-​match),
showed less episodes of acute rejection and similar 1-​year and 5-​year
graft survival when compared to patients with detectable PRAs but
negative DSAs and unsensitized patients.
Early detection of DSA has been associated with worse outcomes. Thus,
aggressive management of these antibodies may impact longer-​term out-
comes by removing or decreasing HLA antibodies. Regimens include IVIG,
plasmapheresis, rituximab, carfilzomib, and bortezomib. However, there
is no significant evidence yet of an appropriate regimen in this scenario.
Further studies are necessary to define optimal early regimens to treat the
memory HLA response.
398
 Chapter 31 399

Late complications
following
lung transplantation
Chronic lung allograft dysfunction 400
Antibody-​mediated rejection 401
Airway complications 402
Medication-​related complications 404
Infection 405
Malignancy 406
Gastro-​oesophageal reflux disease 407
Osteoporosis 407
Neurological complications 408
Retransplantation 408
Conclusion 409
40

400 Chapter 31 Late lung transplant complications

Chronic lung allograft dysfunction

Despite improvement in surgical techniques, lung preservation, advances
in immunosuppressive regimens, and antimicrobial prophylaxis, the long-​
term survival following lung transplantation remains limited. After the first
year, the main causes of morbidity and mortality are CLAD and infections.
CLAD is defined as a persistent decline (≥3 weeks) in pulmonary func-
tion tests from baseline. There are currently three distinct phenotypes
of CLAD: (1) bronchiolitis obliterans syndrome (BOS), (2) neutrophilic
reversible allograft dysfunction (NRAD), (3) and restrictive allograft syn-
drome (RAS). Each phenotype in its pure form has supportive lung physi-
ology, histopathological, radiological, and BAL features. The incidence of
CLAD has been estimated at 48% within 5 years and 76% after 10 years.
Bronchiolitis obliterans syndrome
BOS presents with shortness of breath associated with declining obstructive
lung physiology (decrease in FEV1 and in the FEV1/​FVC ratio). It is a diag-
nosis of exclusion, characterized by progressive irreversible airflow ob-
struction due to small airway obstruction. CXRs are often normal in BOS;
however, CT scans are very sensitive and specific, with airway dilation and
gas trapping on expiration. Histological features of BOS suggest injury of
the small airways leading to excessive fibroproliferation. Transbronchial bi-
opsies have poor sensitivity for BOS (17%). BOS is classified on the basis of
lung function as detailed in Table 31.1. A number of risk factors have been
associated with the development of BOS. The single best characterized risk
factor is ACR, which is characterized by perivascular infiltration of activated
lymphocytes. One or more episodes of moderate to severe acute rejection
is a significant risk factor for BOS. Lymphocytic bronchiolitis is also a risk
factor for BOS. Although often quoted as a risk factor, evidence of an as-
sociation between HLA mismatching and BOS is unclear. Allo-​independent
factors are extremely important with viral, bacterial, and fungal infections
and compliance with medication all associated with increased risk of BOS.
Other potential risk factors for BOS include older donor age, gastro-​
oesophageal reflux with aspiration, organizing pneumonia, prolonged allo-
graft ischaemia, and persistent DSAs.
Treatment options for BOS are very limited with no established suc-
cessful regimen. Most focus on augmentation of immunosuppression to
inhibit progression; however, this treatment predisposes to pulmonary in-
fections and can increase mortality. Small, non-​controlled case series have
suggested conversion from ciclosporin to tacrolimus may result in a slowing
of BOS. Other immune-​modulating strategies that have been trialled in
stabilizing lung function include methotrexate, cyclophosphamide, extra-
corporeal photopheresis, total lymphoid irradiation, and antilymphocyte
antibodies. Montelukast has also been shown to slow the rate of decline in
FEV1 compared with a control group.
Neutrophilic reversible allograft dysfunction
NRAD is a phenotype of CLAD characterized by the presence of excess neu-
trophils ≥15% in BAL in the absence of concurrent infection. Neutrophils se-
crete chemokines, growth factors, and matrix metalloproteinase, increasing
oxidative stress damaging epithelium, and leading to an excess repair pro-
cess with fibroblast proliferation. Azithromycin is a macrolide antibiotic with
 Antibody-mediated rejection 401

Table 31.1 ISHLT consensus on BOS grading

1993 classification 2002 classification
BOS 0 FEV1 80% or more BOS 0 FEV1 >90% of baseline and
of baseline FEF25–​75 >75% of baseline
BOS 0p FEV1 81–​90% of baseline and/​
or FEF25–​75 ≤75% of baseline
BOS 1 FEV1 66–​80% of BOS 1 FEV1 66–​80% of baseline
baseline
BOS 2 FEV1 51–​65% of BOS 2 FEV1 51–​65% of baseline
baseline
BOS 3 FEV1 ≤50% of BOS 3 FEV1 ≤50% of baseline
baseline
FEF, forced expiratory flow.

both anti-​inflammatory and immune-​modulatory properties. It significantly

reduces airway neutrophilia and inflammatory cytokines. A pilot study in
2003 by Gerhardt et al. demonstrated significant improvement in FEV1, with
low-​dose azithromycin. Response to azithromycin is classified as an FEV1 in-
crease of ≥10% after 2–​3 months of treatment. There was a significant cor-
relation between the initial percentage of BAL neutrophilia and the changes
in FEV1 after 3 months of treatment with azithromycin. Patients with NRAD
typically have a good prognosis after diagnosis.
Restrictive allograft syndrome
RAS is a form of CLAD that is defined by the development of restrictive
lung physiology (persistent ≥20% decline in FEV1, concomitant ≥10% de-
cline in TLC). Features on CT scanning consist of interstitial infiltrates with
ground-​glass opacities and fibrotic changes predominantly in the upper
lobes. Histologically, RAS is characterized by extensive fibrosis in the alveoli,
with pleuroparenchymal fibroelastosis, and in the interlobular septa. RAS
develops in about 25–​35% of recipients with CLAD and has a very vari-
able course. In the primary progressive phenotype, RAS is associated with
poorer outcomes (Table 31.2). Mixed phenotypes of BOS and RAS may
develop with some patients first developing BOS with obstructive decline
in FEV1 and then developing RAS. Risk factors for RAS are similar to that of
BOS; they include frequent episodes of acute rejection, lymphocytic bron-
chiolitis, colonization with Pseudomonas, and infection. The fibrotic features
of RAS have led to a trial of pirfenidone in patients post transplantation and
a single case report suggested some improvement.

Antibody-​mediated rejection
DSAs are closely associated with AMR in most SOTs. Until recently, in the
lungs, AMR was associated with hyper-​acute rejection; however, with im-
provement in the detection of DSA it is now a recognized cause of CLAD.
The clinical manifestations of AMR are identical to other form of allograft
injury and are thus not specific. The mechanism of injury in AMR is binding
402

402 Chapter 31 Late lung transplant complications

Table 31.2 Short summary of CLAD phenotypes

Parameter BOS NRAD RAS
Lung Obstructive Obstructive Restrictive
function
BAL Limited Inflammatory Limited inflammation
inflammation neutrophilia
Raised IL-​8
Decreased IL-​10
Histology Fibroblast Inflammation Alveolar damage
proliferation with with fibroblast Parenchymal and
obliteration of proliferation pleural fibrosis
bronchioles
Radiology Gas trapping Airway wall Ground-​glass
Bronchiectasis thickening opacities,
Mosaic attenuation Mucus plugging peripheral fibrosis,
and subpleural
consolidation
Prognosis Variable Good Poor
Azithromycin Variable Good effect No effect

of DSA to the graft and subsequent inflammatory damage. In 2016, the

ISHLT developed a consensus definition for AMR based on findings in renal
and cardiac AMR and case series in lung transplantation. The diagnostic
features of AMR include (1) presence of circulating HLA or other DSAs,
(2) evidence of histological change and C4d deposition in the allograft, and
(3) clinical allograft dysfunction. The treatment of AMR remains empirical,
there are no randomized controlled trials or comparison studies of treat-
ment regimens for AMR. Treatment regimens consist of use of high-​dose
corticosteroids in combination with agents aimed at antibody depletion
such as ATG, IVIG, plasmapheresis, anti-​CD20 monoclonal antibodies, pro-
teasome inhibitors, and complement inhibition.

Airway complications
The transplanted airways are vulnerable to ischaemia as a result of the
bronchial circulation being interrupted at the time of transplantation. The
flow to the airways early postoperatively is entirely dependent on back
flow from the pulmonary circulation until such time as other blood flow can
be established. This can result in ischaemia injury occurring in the airways
and the peri-​anastomotic site is particularly vulnerable (Fig. 31.1a). The fre-
quency of this injury has been difficult to assess, in part related to the lack
of a universally accepted definition—​the published literature suggests rates
of up to 39%. Ischaemic necrosis of the airway results in scarring and fi-
brosis that can result in the loss of airway diameter. This cicatrizing process
results in airway obstruction secondary to strictures and bronchomalacia
resulting in increasing breathlessness (Fig. 31.1b). If the ischaemic strictures
are severe, airway intervention with balloon bronchoplasty, cryotherapy, or
 Airway complications 403

diathermy may improve airway narrowing (Fig. 31.2). Failure of these inter-
ventions associated with increasing symptoms may require airway stenting.
There are a variety of bronchial stents available for first-​generation airways
including self-​expanding metallic stents, silicone stents, and biodegradable
stents. Long-​term complications of bronchial stents are common and in-
clude bronchial stasis, granulation tissue, haemoptysis, stent fracture, and
airway erosion (Fig. 31.3). The use of uncovered stents is decreasing due
to the unacceptably high rate of complications. The outcomes in patients
who have airway complications are significantly worse than in patients who
do not have this complication, particularly in the presence of Aspergillus.

(a)

(b)

Fig. 31.1 (a) Right main bronchus with severe peri-​anastomotic ischaemic necrosis
of the airway. (b) Complete occlusion of right main bronchus with granulation tissue,
successfully removed by cryotherapy. See plate section.
40

404 Chapter 31 Late lung transplant complications

Fig. 31.2 Ischaemic stricture with severe airway narrowing of the right main
bronchus. Total dynamic obstruction of right main bronchus secondary to
bronchomalacia. See plate section.

Fig. 31.3 Right main bronchus stent with in-​stent bronchial granulation tissue. CT
demonstrating in-​stent stenosis. See plate section.

Medication-​related complications
Many of the long-​ term complications arise from the necessity for
antirejection medication in recipients. Globally, glucocorticoids, CNIs, and
cell cycle inhibitors, are the cornerstone of immunosuppressive regimens.
Although the intended action is immune suppression, either individually or
in concert they can cause significant unwanted side effects.
Cardiovascular
Both glucocorticoids and CNIs can increase the risk of developing hyper-
tension, diabetes mellitus, and renal disease. This is associated with a sig-
nificantly increased risk of cardiovascular disease. Approximately 90% of
 Infection 405

patients 3 years post transplantation on ciclosporin-​based therapy have at

least one new risk factor for cardiovascular disease, either hypertension,
hypercholesterolaemia, or new-​onset diabetes mellitus. Mortality from car-
diovascular disease rises from around 5% at 1 year to 8% at 10 years.
Renal
Many patients may have a degree of pre-​existing renal impairment. hyper-
tension and diabetes mellitus are risk factors for chronic kidney disease.
CNIs are directly toxic to kidneys and are a cause of renal failure post trans-
plantation. CNI-​induced chronic kidney disease arises from a combination
of vascular, glomerular, and tubular disease. Dose reduction or withdrawal
may ameliorate the effects of CNI-​induced chronic kidney disease. In pa-
tients without proteinuria, consideration of changing the CNI for an mTOR
inhibitor may offer some renal preservation.
Bone marrow suppression
While the immunosuppression targets T cells, the effects of the cell cycle
inhibitors in particular are not specific. These medications may cause
cytopenia of any of the cell lines, thus leucopenia, thrombocytopaenia,
and anaemia may arise. Bone marrow suppression may also arise from
antifungals (voriconazole) and antimicrobials (meropenem, co-​trimoxazole,
and valganciclovir).

Infection
After BOS, infection is the second highest cause of mortality in lung trans-
plant recipients. The lungs interface between the body and the external en-
vironment and they are constantly exposed to inhaled pathogens. Following
lung transplantation, multiple risk factors exist for infection. Early on, the mu-
cosal defences are compromised with reduction in ciliary function. Patients
may also be colonized with pathogenic and multidrug-​resistant microbes and
there may be donor-​derived infections. The loss of the bronchial circulation
gives rise to the potential for ischaemia of the proximal airways. Intubation
and prolonged critical care and hospital stay expose patients to further
risk. Transplant recipients are at increased risk of opportunistic and donor-​
related infections; the highest risk is in the first 6 months. This correlates with
recovery from surgery and the maximum period of immunosuppression.
Bacterial
Post-​transplant community-​ acquired bacterial infections and multidrug-​
resistant bacterial infections are frequent. Patients transplanted for CF are
at higher risk of reacquisition of their previous microbes. Sputum cultures
as well as BAL are used to guide antibiotic therapies.
Common respiratory viruses
Influenza, parainfluenza, metapneumovirus, and rhinovirus are common re-
spiratory viruses associated with pulmonary infections in the lung transplant
patient. Common respiratory viral infections are also associated with an
increased risk of developing CLAD. All patients receiving lung transplants
should receive an annual influenza vaccination.
406

406 Chapter 31 Late lung transplant complications

Herpesviruses
HSV-​ 1 (human herpesvirus (HHV)-​ 1), HSV-​ 2 (HHV-​ 2), VZV (HHV-​ 3),
HHV-​6, and Kaposi’s sarcoma-​associated herpesvirus (HHV-​8) can all com-
plicate transplantation due to immunosuppression. Patient education and
regular dermatological surveillance should be initiated to identify active in-
fections. Active cutaneous or mucocutaneous infections provide a disruption
to skin defences increasing the risk of bacterial infections. EBV (HHV-​4) and
CMV (HHV-​5) are important viral pathogens in lung transplants. Infection
with CMV is one of the most significant causes of morbidity in transplant
recipients. CMV may cause leucopenia, thrombocytopaenia, hepatitis, colitis,
nephritis, and pneumonitis. CMV may also predispose individuals to bacterial
and fungal infection. Most centres offer CMV-​positive patients pre-​emptive
prophylaxis with valganciclovir for an extended period of time.
Fungal
Aspergillus and Candida species are the most common fungal infections.
Fungal infections have an estimated incidence of 15–​35%. Aspergillus infec-
tion is the most common fungal infection after lung transplantation with an
incidence of approximately 6%. The highest risk period is the first 12 months;
however, there remains a risk throughout the life of a recipient. The diag-
nosis of Aspergillus is challenging and reliant on imaging with CT, supported
by culture and galactomannan assay. Treatment with an azole is often re-
quired for protracted periods and will require careful monitoring of the CNI.
Non-​tuberculous mycobacteria (NTM)
Active growth of fast-​dividing NTM is a relative contraindication to trans-
plantation, the most common NTMs are Mycobacterium avium intracellulare
(50%) M. abscessus (24%), M. fortuitum (12%), and M. immunogenum (12%).
Slow-​growing species such as M. gordonae have also been shown to be patho-
genic, causing 20% of NTM infections. Clinical manifestations are diverse and
include pulmonary disease, surgical site infection such as anastomotic and
sternotomy infections, lymphadenopathy, and cutaneous manifestations.

Malignancy
Post-​transplant recipients are vulnerable to solid organ, cutaneous, and
lymphomatous malignancies. PTLD is most common in the first year post
transplantation, with skin and then solid organ malignancies increasing over
time. At 1 year, the incidence of cancers is approximately 5% rising to 20%
at 5 years and 30% at 10 years. Both CNIs and cell cycle inhibitors have
been shown to reduce cellular apoptosis, suppress DNA repair, and pro-
mote metastasis. In combination, they can promote oncogenic pathways
by increasing tumour formation, growth, and spread. They may interfere
with immune surveillance responsible for identifying and destroying malig-
nancies. PTLD occurs in up to 5% of patients post transplantation. The risk
of PTLD is higher in EBV-​naïve recipients which is an oncogenic virus that
has a trophism for B cells. Immunosuppression reduces regulatory T-​cell
activity allowing infected cells to proliferate. Treatment for PTLD involves
reduction of immunosuppression burden. In patients with CD20-​positive
tumours, there should be consideration of single-​agent rituximab but for
patients with CD20-​negative tumours, chemotherapy is usually needed.
 Osteoporosis 407

Dermatological malignancies
Malignancies of the skin have been estimated to occur in 16% of patients
post lung transplantation and account for 61% of all malignancies at 1 year.
The risk of skin cancer is increased in older patients with greater sun ex-
posure and in patients with a high burden of immunosuppression. Other
risk factors that may predispose to dermatological malignancy include
human papillomavirus, underlying skin type, male sex, and duration of im-
munosuppression. There is also an increased risk in patients on long-​term
voriconazole.
Lung cancer
Lung transplant recipients are at increased risk of lung malignancy, with
an estimated tenfold increased risk compared to the general population
of developing bronchogenic carcinoma. Risk factors include the smoking
history of both donor and recipient, and environmental and occupational
exposures. The prevalence of lung cancer originating from the native lung in
single-​lung transplants has increased over time and is now estimated to be
9%. Primary bronchogenic carcinoma from transplanted lung has a signifi-
cantly lower prevalence of <0.4%.

Gastro-​oesophageal reflux disease

Gastro-​oesophageal reflux disease is very common in patients with end-​
stage lung disease with an estimated incidence of between 35% and 65%.
Post transplantation, the prevalence increases from 30% at 3 months to
50% at 12 months. Multiple studies have shown an association with gastro-​
oesophageal reflux disease chronic micro-​ aspiration and BOS. Micro-​
aspiration reduces surfactant A and D levels, and increases inflammatory
cytokines and alveolar neutrophils. Ambulatory pH monitoring is the main
investigation. Impedance testing has been shown to be useful in identifying
non-​acid reflux as well, but no reflux testing clearly demonstrates aspir-
ation. Treatment can be divided into medical and surgical, the use of proton
pump inhibitors and H2 blockers are used to reduce the acid component of
reflux. Although proton pump inhibitors reduce acid secretion they may not
reduce micro-​aspiration. Azithromycin may act as a prokinetic and improve
clearance of gastric contents and reduce reflux. Antireflux procedures such
as fundoplication have been shown to reduce gastro-​oesophageal reflux
disease-​associated lung injury and improve FEV1 in some recipients. Studies
suggest that early intervention for gastro-​oesophageal reflux disease is as-
sociated with better outcomes in the long term.

Osteoporosis
Osteoporosis as defined as a bone mineral density <–​2.5 SD. Pre-​existing
bone disease is common in lung transplant recipients with an estimated
prevalence of approximately 50%. A significant contribution to this incidence
is from prolonged corticosteroid use. Other factors that are associated with
bone demineralization include advanced age, reduced BMI, reduced vitamin
D levels, and hypogonadism. One-​year post transplantation the prevalence
408

408 Chapter 31 Late lung transplant complications

increases significantly to 73%. Glucocorticoids reduce osteoblast replica-

tion differentiation and apoptosis. Lung transplant recipients should receive
screening for osteoporosis prior to transplant with dual-​energy X-​ray ab-
sorptiometry imaging. Unless contraindicated, all patients should receive
supplementation of their daily allowance of calcium and vitamin D to en-
sure an adequate vitamin D status. Bisphosphonates should be added to
the regimen, the optimum duration of bisphosphonate treatment has yet to
be determined. Concerns over long-​term treatment with bisphosphonates
causing osteonecrosis of the jaw and atypical femoral fractures require
that treatment duration should be reviewed at 3–​5 years. If treatment is
stopped, risk assessment should be performed at 18 months to 3 years
after cessation, or after any new fracture. At this point, reinstatement of
therapy should be considered if appropriate.

Neurological complications
CNIs may cause neurological toxicity most commonly associated with
higher blood concentrations. Paraesthesia, tremor, and headache are
common manifestations and estimated to occur in 20–​39% of patients.
Posterior reversible encephalopathy syndrome is clinically similar to hyper-
tensive encephalopathy. It causes acute confusion, aphasia, cortical blind-
ness, ataxia, and seizures. The diagnosis is confirmed on MRI with findings
of vasogenic oedema. Treatment using a different CNI or withdrawal of
CNI altogether generally leads to resolution of symptoms. In the 10 years
post transplantation, it is estimated that 92% of patients will encounter
neurological problems. These are most common in the first year and often
in the perioperative period. Postoperative intracerebral haemorrhage, sub-
arachnoid haemorrhage, and ischaemic strokes can cause significant early
morbidity and mortality. Encephalopathy may also arise early post trans-
plantation secondary to drug toxicities, infection, or cerebral hypoxia.
Infections with atypical organisms may occur early post transplantation.
Neurological complications after lung transplant have an increased mortality
at 2 years. There is emerging evidence that recipients may encounter signifi-
cant neurocognitive deficit, this has been identified at 3 months post trans-
plantation. However, studies looking at this over a longer duration have yet
to be performed.

Retransplantation
A very limited number of retransplantations are performed per year
and represent <4% of all lung transplants. While survival at 1 year post
retransplantation has improved up to 71%, it remains inferior to first trans-
plantations at 86%. The median survival is significantly lower than primary
lung transplantation at 2.6 years versus 6.8 years for primary transplant-
ations. However, after allowing for PGD or acute surgical issues, the
conditional survival taken at 1 year is 6.3 years. The main indications for
retransplantation have been CLAD, airway complications, and end-​stage
BOS. Patients retransplanted for BOS had an improved survival compared
to other indications.
 Conclusion 409

Conclusion
Outcomes from lung transplantation have improved significantly over the
last 30 years. However, the burden of immunosuppression necessitated for
these outcomes results in a significant incidence of complications. Careful
monitoring of patients in specialist lung transplant centres helps to reduce
morbidity and mortality and improve both length and quality of life.
041
 Chapter 32 411

Antimicrobial prophylaxis
and treatment
after lung transplantation
Key recommendations 412
Viral infections in lung transplantation 413
Fungal infections in lung transplantation 414
Bacterial infections in lung transplantation 416
Choice of antibiotic 417
Special recipient considerations 418
241

412 Chapter 32 Antimicrobial therapy after lung transplant

Key recommendations
See Table 32.1.

Table 32.1 Key recommendations for prophylaxis and treatment after lung
transplantation

Pre-​ • Ensure appropriate respiratory cultures from

transplantation all recipients at regular intervals and consider
colonization antimicrobial treatment strategies if clinically indicated
management • Certain pathogens, namely Mycobacterium abscessus
complex and Burkholderia cenocepacia, need special
consideration. Isolation of these pathogens may impact
transplant eligibility
Management • Consider stapling the bronchus prior to explanting
during operative the lung
implantation • Take care to avoid spillage of secretions during
implantation
• Irrigate the pleural space with antibacterial solution
both after removal of the lung and at the completion
of the procedure (such as taurolidine)
• Microbiologically sampling of the donor lung: BAL
(50 mL of normal saline into a tertiary bronchus and
separate swab culture)
Post-​ • Patients at high risk, i.e. those with preoperative
trans plantation colonization, need coverage with broad-​spectrum
care antibiotics. Minimizing the duration of antimicrobials
is essential to reduce the risk of development of
antimicrobial resistance. Nebulized antimicrobials
are a useful adjunct. They help deliver appropriate
antimicrobials at the anatomic site without the risk of
patient developing antimicrobial resistance
• BAL from the donor and preoperative colonization
helps guide antimicrobial management in the
immediate perioperative period
• Avoidance of other healthcare associated infection is
also key:
1. Maintenance of good skin integrity, appropriate
management of intravascular catheters
2. HOUDINI is a useful tool to guide urinary catheter
management.
Haematuria (only requires catheter if in clot
  retention)
Obstruction/​retention
Urology surgery
Damaged skin (open sacral or perineal wound in an
  incontinent patient)
Input/​output, fluid monitoring
Nursing care end of life/​comfort care
Immobility, due to physical constraint, e.g. unstable
   fracture and unable to use bottles/​bedpans
3. Removal of drains at the earliest possible opportunity
4. Appropriate isolation facilities and adherence to
strict infection prevention and control policies
 Viral infections in lung transplantation 413

Viral infections in lung transplantation

Viral infection causes specific morbidity and mortality after lung transplant-
ation (Table 32.2). Several viral infections have been associated with long-​
term morbidity such as CLAD, previously identified as BOS.
Infection with community-​acquired respiratory viruses such as influenzae
A/​B and Paramyxoviridae (e.g. respiratory syncytial virus) occur with sea-
sonal variability and infected episodes range from easily treatable to more
persistent infection. The interventions for these infections vary depending
on the virus recovered and the currently available treatment regimens.
Prevention strategies include yearly influenza vaccination for transplant re-
cipients and their close contacts.
Herpes infection
Most transplant recipients are seropositive for HSV and VZV. Postoperative
infections are a consequence of reactivation of previous latent infection.
Assessment of serological status of both the donor and the recipient are
essential especially for seronegative recipients who are at the greatest risk
of developing symptomatic disease.
Cytomegalovirus
Belonging to the beta herpesvirus family, CMV has historically been the
number one pathogen occurring after organ transplantation. Interestingly,
the incidence of CMV disease (symptoms and active infection—​
diagnosed by NAAT which identifies CMV replication) is higher in the
lung transplant population compared to other solid organs transplanted
and is potentially related to the size of the organ and a higher state of
immunosuppression.
The symptoms are important to recognize and consist of malaise, my-
algia, low-​grade temperature, and an end-​organ disease such as colitis,
oesophagitis, or hepatitis. As well as wanting to avoid the disease for the
comfort of the recipients, it is also important to note the link between
CMV infection and future allograft dysfunction and so infection prevention
is advocated.
Epstein–​Barr virus (EBV)
EBV is a gamma herpesvirus (HHV-​4) that infects >95% of the world’s
population. EBV is responsible for a variety of clinical manifestations
ranging from an infectious mononucleosis-​ like syndrome to prolifer-
ation of monoclonal lymphocytes leading to high-​ g rade lymphoma.
Such clinical entities associated with EBV infection after transplantation
are known as PTLD and are mainly the consequence of the lack of cell-​
mediated immune surveillance of EBV-​infected lymphocytes because of
immunosuppression.
The main risk factor for the development of EBV-​associated PTLD in
lung transplant recipients is the serostatus of the donor and the recipient.
EBV D+​/​R–​patients are at the highest risk of developing primary EBV
infection which may lead to PTLD. Prevention is a major issue and is man-
aged with reduced immunosuppressive regimens, particularly regimens that
avoid the use of lymphocyte-​depleting antibodies.
41

414 Chapter 32 Antimicrobial therapy after lung transplant

Table 32.2 Viral infection after transplantation

Virus Prevention Therapy
HSV-​1 and Aciclovir prophylaxis Mucocutaneous
HSV-​2 (HHV-​1 disease: aciclovir,
and HHV 2) valaciclovir, and
famciclovir
Disseminated
disease: IV aciclovir
VZV (HHV-​3) Aciclovir prophylaxis Localized
disease: aciclovir,
valaciclovir and
famciclovir
Disseminated
disease: IV aciclovir
CMV (HHV-​5) D+​/​R–​: CMV mismatch consider End-​organ disease: IV
prophylaxis with valganciclovir ganciclovir
for 3 months
D–​/​R+​ and D+​/​R+​: no
prophylaxis, weekly surveillance
4 weeks post-​transplantation for
at least 3 months and treatment
when high viral load detected
D–​/​R–​: no prophylaxis/​
surveillance
Some centres have different
strategies including longer
duration of prophylaxis
≥200 days
HHV6, HHV7, No specific antiviral preventive End-​organ HHV6
HHV8 strategy is recommended disease: foscarnet

Other herpesvirus infections

HSV reinfection is one of the most common forms of viral infection after
organ transplantation. The clinical manifestations of HSV infection are usu-
ally like those observed in immunocompetent patients and include oral and
genital mucocutaneous vesicular rash. In highly immunocompromised pa-
tients, more extensive lesions with necrotic ulcers can be observed.

Fungal infections in lung transplantation

Invasive fungal infections are a serious complication of lung transplantation
and are associated with significant morbidity and mortality. When com-
pared to other SOT recipients, lung transplant recipients are more suscep-
tible to fungal infections. This stems from continuous and direct contact
with the environment alongside impaired clearance mechanisms caused by
allograft denervation and colonization of the airways by organisms from
the upper respiratory tract or from the native lung in cases of single-​lung
transplantation.
 Fungal infections in lung transplantation 415

Aspergillus
Aspergillus infection is the most frequent fungal infection after lung trans-
plantation; it accounts for 73% of all cases of mould infection. It generally
occurs within the first year after implantation and the most common organ-
isms are A. fumigatus, A. flavus, A. niger, A. terreus, A. versicolor, and others.
Aspergillus is a ubiquitous organism that is found in decaying organic
matter. Exposure to airborne fungal conidia can occur during construction
or renovation, gardening, composting, and farming activities. Additionally,
tobacco and marijuana can be contaminated with fungi, and recreational and
medical marijuana use has been associated with IA. Therefore, lung trans-
plant recipients should be advised to avoid such environmental exposures.
The risk factors are shown in Box 32.1.

Box 32.1 Risk factors for invasive aspergillosis in lung

transplant recipients
• Environmental and recreational exposure.
• Construction.
• Farming.
• Marijuana use (smoking).
• Net state of immunosuppression.
• High-​dose steroids.
• Anti-​lymphocytic therapy.
• Muromonab–​CD3 (OKT3).
• CMV infection.
• Hypogammaglobulinaemia.
Risk factors unique to lung transplantation
• Continuous and direct contact with the environment.
• Impaired clearance mechanisms caused by allograft denervation.
• Airway colonization before transplantation.
• Airway colonization after transplantation.
• Single-​lung transplantation.
• Prolonged ischaemia.
• Donor-​derived infection.

Four main clinical manifestations of Aspergillus infection in lung transplant

recipients are colonization, tracheobronchitis or bronchial anastomotic site
infection, pulmonary disease, and disseminated disease. Colonization of
Aspergillus can occur in 46% of lung transplant recipients. In most patients it
occurs within 3 months and patients may be asymptomatic with diagnosis
made on culture without evidence of tissue invasion. Colonization itself is
not a problem; however, it may be linked to CLAD.
Management of IA requires a twofold approach: reduction of immuno-
suppression and selection of an appropriate antifungal agent. The choice
of agent should be based on efficacy as well as toxicity with consideration
of the patient’s comorbid condition. Attention should be paid to special
circumstances. Tracheobronchitis requires bronchoscopic debridement
and inhaled antifungal therapy in addition to systemic therapy. In cases with
anastomotic dehiscence, surgical intervention with stent placement may be
required.
641

416 Chapter 32 Antimicrobial therapy after lung transplant

Prophylaxis
Given the risk for the development of IA following lung transplantation and
the significant morbidity associated with it, many lung transplant centres
implement pre-​emptive or universal prophylaxis:
• Pre-​emptive: targets therapy towards patients deemed to be at high risk
(e.g. previous history of colonization).
• Universal: targets all transplant recipients irrespective of their risk status.
A worldwide survey reported that 58.6% of lung transplant centres use
universal prophylaxis for the first 6 months. Voriconazole either alone or
in combination with inhaled amphotericin is the most commonly adminis-
tered agent.
Yeast
Candida species are responsible for 23% of all invasive fungal infections in
lung transplant recipients. Many species can cause human disease but most
infection is caused by C. albicans, followed by C. glabrata, C. parapsilosis,
C. tropicalis, and C. krusei.
Risk factors
Risk factors include broad-​ spectrum antibiotics, protracted antibiotic
therapy, presence of central venous lines, and renal replacement therapy.
Colonization with Candida is a risk factor for anastomotic infection and
candidemia. Candidiasis is usually a nosocomial infection that occurs early
after transplantation and the manifestations range from mucocutaneous in-
volvement to deep organ seeding. Bronchial anastomotic involvement is the
most common manifestation followed by bloodstream infection and dis-
seminated disease. Recovery of Candida from a respiratory tract culture
usually indicates colonization and does not require therapy. Candida pneu-
monia is rare, and its diagnosis requires demonstration of tissue invasion by
histopathological examination.
Bloodstream infections are preferably treated with fluconazole or an ap-
propriate echinocandin for at least 14 days after the first negative blood
culture. Disseminated disease requires prolonged treatment until infection
is cleared. C. glabrata demonstrates elevated minimum inhibitory con-
centrations to fluconazole hence high-​dose fluconazole or an alternative
antifungal agent should be considered. C. krusei is intrinsically resistant to
fluconazole and should be treated with another agent. Alternative agents
include echinocandin, amphotericin, and other azoles.
Despite advances in understanding of the biology and epidemiology of
fungal pathogens, improvement in diagnostic modalities, and introduction of
new more efficacious therapeutic agents, invasive fungal infection remains
an important cause of morbidity and mortality after lung transplantation.

Bacterial infections
in lung transplantation
This is the most common type of infectious complication following lung
transplantation. Most episodes occur immediately with >80% occurring
within the lung, mediastinum, and pleural space. Most infections are uncom-
plicated and respond to antibiotic therapy.
 Choice of antibiotic 417

Risk factors
Preoperative factors:
• Altered anatomy.
• Aggressive immunosuppression.
• Recipients colonizing flora.
Postoperative factors:
• Decreased cough reflex.
• Decreased mucociliary clearance.
• Loss of bronchial circulation.
• Loss of lymphatic drainage.
Lung transplantation was once described as an immune state character-
ized by an increase in infection and an alloreactive state. The process of
increasing immunosuppression to manage rejection and obliterative bron-
chiolitis coupled with markedly impaired lung function and mucus clearance
dramatically raises the predisposition to infection in such patients.
Donor-​to-​host transmission is the most common source of bacterial in-
fection post transplantation either from donor lavage or from the pres-
ervation fluid. As such, knowledge of pre-​transplant airway colonization
especially in the at-​risk population such as CF or bronchiectasis is crit-
ical. Lung transplant recipients who are colonized with pathogens such as
Pseudomonas, Staphylococcus aureus, and Aspergillus are at greater risk of
developing post-​transplantation infectious complications secondary to the
presence of these organisms.
Mycobacterial infection
Mycobacterial infection after lung transplantation is uncommon. The
relative risk in lung transplant recipients with tuberculous and NTM in-
fection is greater than that in the general population. Lung transplant
recipients have a higher incidence than other SOT recipients, NTM
infection is rare, albeit more common than Mycobacterium tuberculosis
infection. Lung transplant recipients are at greater risk for pulmonary
infections in general because the allograft is in direct contact with the
pathogen. M. abscessus is observed in lung transplant recipients at a
higher rate than in other SOTs. It has been reported to cause skin,
soft tissue, pulmonary, and disseminated disease. The most common
manifestation is cutaneous lesions. Such infection has been reported
throughout the post-​transplant period with no increase in incidence at
any specific point in that period.

Choice of antibiotic
Obtaining appropriate cultures such as blood (peripheral and line), urine,
and sputum early in the infection is critical, because they are the usual
sources. Early removal of central venous catheters, arterial catheters,
and urinary catheters is essential. In the first month, nosocomial patho-
gens predominate, and empiric therapy should have broad-​ spectrum
coverage. The emergence of multidrug-​resistant pathogens has made this
choice particularly challenging and susceptibility data should always guide
therapeutic decisions. Some of the anti-​MRSA drugs include linezolid and
841

418 Chapter 32 Antimicrobial therapy after lung transplant

tigecycline. Newer agents (e.g. tedizolid, ceftobiprole, dalbavancin) have

not been licensed for use in this population. The use of daptomycin is
restricted to bloodstream infections as it is inactivated by pulmonary sur-
factant. Glycopeptide-​resistant enterococci may be problematic in some
regions—​common infections are associated with renal tract or are central
line associated. Linezolid is the drug of choice but should be used cau-
tiously due to the risk of thrombocytopenia and peripheral neuropathy.
Other significant multidrug-​resistant pathogens include carbapenemase-​
producing Enterobacteriaceae where treatment options are very limited
(e.g. IV colistin, fosfomycin, and combination treatment) based on suscep-
tibility testing.
Burkholderia cepacia
One of the most debated areas is carriage of B. cepacia. The organism is
associated with faster decline in patients with CF, as a result of which it
is present in a disproportionately high percentage of potential transplant
recipients. The strain B. cenocepacia is associated with extremely poor out-
comes following transplantation. Many centres do not accept recipients col-
onized with B. cenocepacia, namely genomovar III.
Mycobacterium abscessus complex
M. abscessus complex is associated with rapid disease progression and
multiple reports exist of postoperative recurrences with chest wall ab-
scesses and surgical wound breakdown. Treatment needs to be well estab-
lished prior to transplant and eradication can take up to 2 years. Disease
that continues to progress pre transplantation should be regarded as a
contraindication.

Special recipient considerations

Cystic fibrosis and bronchiectasis
Affecting approximately 1 in 3000 births in northern and western Europe,
CF is a lethal disease affecting the lung via a cycle of infection and destruc-
tion of the bronchioles. As many as 98% of children show either culture
or serological evidence of Pseudomonas aeruginosa infection by 3 years
of age. Most deaths in patients with CF are from respiratory failure. In
contrast, a wide range of septic lung conditions encompass bronchiec-
tasis. The label is applied to those patients in whom dilated airways are
visible on high-​resolution CT scans with recurrent infections and sputum
production and it is much more common in the elderly. In up to half of
all patients bronchiectasis develops following a destructive infection such
as pneumonia, measles, tuberculosis, or allergenic bronchopulmonary
Aspergillus.
Both conditions can lead to inflammatory adhesions or purulent spillage
at the time of transplantation. In principle, the surgical and microbio-
logical steps taken to prevent early infection are outlined in Table 32.1
(see p. 412).
 Special recipient considerations 419

Novel emerging treatment therapies

Mycobacterium abscessus is generally considered a contraindication to lung
transplantation but, in 2019, a novel case report was presented in Nature
of a recipient aged 15 years who received a double-​lung transplant col-
onized with M. abscessus. A genetically engineered bacteriophage (phage)
was constructed involving a three-​phage cocktail which efficiently killed the
infectious M. abscessus strain. These transplants are high risk and mandate
the use of continued IV antibiotic cocktails post transplantation but evolving
gene therapy presents another string in the bow of the antimicrobial treat-
ment arm.
420
 Section 7

Combined
transplantation

33 Combined heart–​lung transplantation 423

34 Combined thoracoabdominal organ
transplantation 439
42
 Chapter 33 423

Combined
heart–​lung transplantation
Introduction 424
Indications 424
Organ allocation 426
Domino transplants 426
Donor organ procurement 427
Heart–​lung transplant recipient procedure 428
Early postoperative management 433
Allograft monitoring 433
Outcomes: patient survival 434
Causes of death in heart–​lung transplant recipients 435
Morbidity following heart–​lung transplantation 436
Retransplantation 437
42

424 Chapter 33 Combined heart–lung transplantation

Introduction
Combined heart–​lung transplantation (HLTx) is accepted as the only ef-
fective treatment for patients with end-​stage heart and lung failure. The
first HLTx was performed in 1981 in Stanford, CA, USA, on a patient with
pulmonary arterial hypertension. The practice of HLTx has evolved signifi-
cantly since led by improved medical therapies for end-​stage cardiopul-
monary failure and PH, improved outcomes following isolated heart and
lung transplantation, and also driven by limited donor organ availability.
Current activity and trends over time
See Fig. 33.1.
• Global HLTx activity peaked in the year 1990 with 225 reported cases.
• Activity steadily fell to 38 in 2015, but has risen slightly since.
• HLTx only account for <1% of all cardiothoracic transplantations. The
2016 ISHLT registry reported 58 HLTx worldwide, compared with 4554
single or bilateral lung transplants and 5832 heart transplants.
Worldwide, 89 centres have reported performing HLTx since 2009:
• Of these, 56 (63%) centres average one procedure per year.
• Only five centres have performed an average of more than three per
year.

Indications
In the early years, patients with end-​stage lung disease with concomitant
RV dysfunction tended to be considered for HLTx. However, bilateral lung
transplants are now preferred in these patients since outcomes are similar
or better, and right heart dysfunction commonly improves following suc-
cessful lung transplantation.
Currently, the commonest indications for HLTx include:
• Congenital heart disease with Eisenmenger’s syndrome—​including
patients with complex congenital heart disease who had previously
undergone palliative surgical procedures to defer the need for
transplantation.
• Idiopathic pulmonary arterial hypertension—​some evidence that
outcomes may be superior to bilateral lung transplantation in patients
ventilated on ICU pre transplantation.
• Combined end-​stage heart and lung disease—​for example, due to
systemic disease or pulmonary diseases with refractory RV or LV
dysfunction.
HLTx used to be offered to selected patients with end-​stage HF and se-
vere secondary PH (PVR >5 Wood units or a transpulmonary gradient
>15–​20 mmHg) because outcomes following isolated heart transplantation
were poor in these patients. However, many such patients are now offered
implantable LVADs instead as a bridge to heart transplantation candidacy.
The most recent ISHLT registry report summarizes the indications for the
4054 heart–​lung transplants performed worldwide (Table 33.1).
 Number of transplantations
0
50
100
150
200
250
300

1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
Fig. 33.1 Numbers of HLTx performed worldwide.

1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
425 Indications
426

426 Chapter 33 Combined heart–lung transplantation

Table 33.1 Indications for heart–​lung transplantation from 1982 to 2016

Indication Number (%)
Congenital heart disease 1227 (37.7)
Idiopathic pulmonary arterial hypertension 962 (29.5)
Cystic fibrosis 464 (14.2)
COPD 145 (4.5)
Idiopathic pulmonary fibrosis 124 (3.8)
Alpha-​1 antitrypsin deficiency 63 (1.9)
Other 272 (8.4)
Source data from ISHLT registry report.

Organ allocation
In the setting of donor organ shortages and a duty to maximize the benefits
from the donated organs, it is important to consider the number of patients
who could receive a transplant as well as their likely survival gains from
transplantation. The combined life-​years gained from carrying out one HLTx
is probably less than half that of one heart transplant and a bilateral lung
transplant. Therefore, whether a donor heart–​lung block should be used
for a single patient requiring HLTx or the organ block split for two patients,
one requiring a heart transplant and the other a bilateral lung transplant, is
an ethical dilemma.
In the USA, there is no allocation system for patients requiring HLTx.
They must be listed separately for heart and lung transplants using the
UNOS schemes. It is the heart status that usually determines organ alloca-
tion, since if donor lungs are allocated to a patient, the donor heart would
only be allocated if there is no other suitable higher heart status candidate.
The converse is not true if the patient is allocated the heart.
In Europe, national policies vary by country, but similarly tend to favour
isolated heart or lung transplant recipients due to urgency status listing for
isolated organs that is not available for potential HLTx recipients. As a re-
sult, the wait for HLTx tends to be longer.

Domino transplants
In the era before bilateral lung transplantation became fully established,
some patients with normal heart function who required lung transplantation
were offered HLTx (e.g. patients with CF or emphysema). In these cases,
the recipient could donate their heart at the time of their HLTx to a HF pa-
tient who requires a heart transplant as a ‘domino’ procedure.
The domino procedure is rarely performed in the current era when such
patients with normal cardiac function would more likely receive a bilateral
lung transplant.
However, HLTx and domino heart donation do offer some advantages
for both recipients:
 Donor organ procurement 427

• The implant time for a heart–​lung block is much shorter than that
of bilateral lung transplantation and, therefore, ischaemic time is
significantly reduced.
• Airway complications after HLTx is infrequent compared with that
following lung transplantation especially in patients with smaller stature.
• The patient scheduled to be a domino heart donor can be fully
investigated beforehand and there is even the potential to consider
donor–​recipient HLA matching.
• The domino heart donor is essentially a living donor—​the donated
heart would not have been exposed to the detrimental effects of
brainstem death.
• There is usually co-​location of the domino heart donor and the
recipient thereby minimizing donor heart ischaemic time.
Therefore, there is still a place for HLTx and domino heart donation for
selected patients with lung failure and a normal heart.

Donor organ procurement

Donor assessment
Assessment of the donor is performed upon arrival at the donor hospital:
• Heart: haemodynamics, inotrope and vasopressor requirements, ECG,
echocardiogram, right heart catheterization.
• Lung: arterial blood gas, bronchoscopy, CXR.
Size matching is important because oversized donor lungs can be challen-
ging to fit inside the chest cavity of the recipient, particularly if the recipient
has fibrotic lung disease. In general, it is reasonable to use a donor with
a predicted total lung capacity that is within 15% of that of the recipient.
Organ procurement
There are some important differences with procurement of the heart–​lung
block for HLTx.
• Median sternotomy is performed, and the pericardium and pleura are
opened widely including posteriorly, adjacent to the diaphragm.
• The heart is assessed visually and palpated for CAD. The lungs are
palpated for nodules and atelectatic lung segments are recruited.
• The PA is separated from the ascending aorta.
• The SVC is dissected and encircled with a silk ligature cranial to the
azygos vein. The azygous vein can be divided between ligatures.
• To facilitate access to the trachea at the level of the thoracic inlet,
the innominate vein and artery can be divided in turn between heavy
ligatures. The trachea is encircled with an umbilical tape in the space
between the SVC and aortic arch.
• Following systemic heparinization, cannulas are placed in the ascending
aorta and pulmonary trunk, secured with purse-​string sutures, and
connected to an infusion line for cardiac and lung preservation solutions
respectively.
• Once all retrieval teams are ready to proceed with aortic clamping, a
bolus of prostacyclin is administered into the pulmonary trunk to pre-​
dilate the pulmonary vascular bed.
428

428 Chapter 33 Combined heart–lung transplantation

• The SVC is ligated.

• The intrapericardial IVC is clamped and hemi-​transected to vent the RA.
• The tip of the LAA is widely opened to vent the LA. The heart
is allowed to beat 4–​6 times until empty before the aorta is
cross-​clamped.
• Cardiac and lung preservation fluid infusions are commenced and the
chest cavity is filled with copious cold saline for topical cooling of the
thoracic organs. Since the donor heart–​lung block is kept intact, lung
preservation fluid is only given antegrade and not retrograde.
• Following organ preservation, the SVC and IVC are divided.
• The heart–​lung block is mobilized from the posterior mediastinal
attachments with sharp dissection towards the superior mediastinum,
taking care not to breach the oesophageal lumen.
• Transecting the descending thoracic aorta at the level of the diaphragm
and taking it with the heart–​lung block helps to ensure that the
dissection plane is keep well away from the posterior surface of the
airways.
• Access to the posterior surface of the hilum can be improved by
reflecting each lung anteriorly out of the sternotomy in turn if
required.
• Superiorly, the aortic arch is mobilized by dividing the three aortic arch
branches.
• At this point, the trachea is transected at the level of the thoracic inlet
and the heart–​lung block can be lifted out of the chest cavity.
• On the back table, a sterile endotracheal tube is reinserted into the
open end of the donor trachea to inflate the lungs to a pressure of 15–​
20 cmH2O to ensure that all segments are recruited.
• The trachea is then stapled 2–​3 cm above the carina and the heart–​lung
block is packaged for transfer to the recipient centre.

Heart–​lung transplant recipient

procedure
Cardio-​pneumonectomy
• Median sternotomy is performed; the anterior pericardium and both
pleurae are opened widely.
• Throughout this procedure, great care is taken to preserve the phrenic,
left recurrent laryngeal, and vagus nerves.
• Following systemic heparinization, the patient is placed onto CPB with
aorto-​bicaval cannulation.
• Following aortic cross-​clamping, the aorta is divided above the
sinotubular junction and the pulmonary trunk just proximal to its
bifurcation; the SVC is transected as it enters the RA; the inferior RA is
incised leaving a 2 cm cuff of atrial tissue around the IVC orifice; and the
LA is divided circumferentially just anterior to the pulmonary veins and
the heart is removed (Fig. 33.2).
• Lateral pericardial windows are made longitudinally on each side
posterior to the phrenic nerves.
 Heart–lung transplant recipient procedure 429

Ao

SVC
PA

LA cuff

Fig. 33.2 Cardiectomy with the recipient posterior LA and PA bifurcation left
behind.

• The posterior LA remnant is incised down the midline to separate the

right and left pulmonary vein cuffs.
• Bilateral pneumonectomies are then performed by dissecting the lung
hila and stapling the main stem bronchi.
• The vagus can be adherent to the posterior surface of each main
bronchus and care should be taken to avoid their injury.
• A 1 cm patch of PA is preserved at the level of the ligamentum
arteriosum to reduce risk of injuring the left recurrent laryngeal nerve.
• The two transected bronchial stumps are grasped to allow traction and
dissection of the carina.
• The trachea is divided just above the take-​off of the right main
bronchus (Fig. 33.3).
• Meticulous haemostasis should be performed at this stage paying
particular attention to the posterior mediastinum.
430

430 Chapter 33 Combined heart–lung transplantation

Fig. 33.3 The bronchial stumps are retracted inferiorly, and the trachea is incised
just above the carina.

Preparation of the heart–​lung block

• The heart–​lung block is taken out of cold storage and inspected.
• Excess donor aorta and pericardium are removed.
• The paratracheal tissue should be preserved in order to maintain its
blood supply.
• The donor trachea is incised one cartilaginous ring above the take-​off of
the right main bronchus.
• Any secretions within the airways are suctioned and a specimen sent for
culture.
• The vent site on the donor LAA is now repaired.
• The donor interatrial septum is inspected through the IVC orifice and
any defect repaired.
Implantation
• The heart–​lung block is placed into the recipient chest cavity by passing
the lungs through the laterally placed pericardial windows (Fig. 33.4).
• The tracheal anastomosis is completed with a continuous polypropylene
suture; some surgeons prefer to use a continuous suture for the
membranous portion only and interrupted sutures for the cartilaginous
portion (Fig. 33.5).
 Heart–lung transplant recipient procedure 431

• The peritracheal tissues are suture approximated so that the tracheal

anastomosis is not in direct contact with a vascular structure and may
reduce the risk of fistula formation.
• The aortic anastomosis is performed with continuous 4-​0 Prolene, a de-​
airing cannula is secured, and the aortic cross-​clamp is removed.
• A fibreoptic bronchoscopy is performed to visualize the tracheal
anastomosis and for airway clearance. Ventilation of the lungs can be
initiated with half tidal volume (5 mL/​kg).
• The IVC and SVC anastomoses are performed while the donor heart is
reperfused (Fig. 33.6).
• On completion, ventilation can be increased to full tidal volume (10
mL/​kg) and the patient weaned from CPB with appropriate inotropic
support and epicardial pacing as required.
• A Swan–​Ganz catheter should be routinely placed for haemodynamic
monitoring.

Fig. 33.4 The heart–​lung block is lowered into the chest: first the left lung
into the left pleural space through the incision in the pericardium behind the left
phrenic nerve, then the right lung into the right pleural space through the incision
in the pericardium behind the right phrenic nerve, and finally the heart into the
pericardial space.
432

432 Chapter 33 Combined heart–lung transplantation

Fig. 33.5 Tracheal anastomosis.

Fig. 33.6 SVC anastomosis.

 Allograft monitoring 433

Early postoperative management

Early postoperative care requires close monitoring of the haemodynamics
and gas exchange. Allograft dysfunctions will manifest as they would for
isolated heart or lung transplant recipients. Patients should have close moni-
toring of RV function with appropriate inotropic and pulmonary vasodila-
tory therapy. Protective ventilation should be employed to minimize the
risk of barotrauma.
Early graft failure
Early graft failure resulting in death or retransplantation within the first
30 days occurs in 6.8% of HLTx. This compares unfavourable to the inci-
dence following isolated heart or lung transplantation over the same period
of 2.3%. This may be due to the combined risks of developing dysfunction
of the heart or lung allografts.
Surgical complications
HLTx recipients are at risk of developing certain surgical complications less
commonly encountered following isolated heart or lung transplantations:
• Phrenic nerve dysfunction: there is risk of injury to the phrenic nerve
when opening the pericardium posteriorly and pericardial retraction
while placing the lungs into the pleural cavities. Patients with phrenic
nerve dysfunction require ventilation for longer periods and have
prolonged ICU stays.
• Chylothorax: the extent of posterior mediastinal dissection to explant
the diseased organs can lead to unnoticed lymphatic injury. Additionally,
it is recognized that patients with congenital heart disease can have
aberrant lymphatic pathways that can be injured inadvertently. This can
prolong recovery and may require further surgical intervention.
• Gastroparesis: the vagus nerve is at risk of being injured while dissecting
and performing the tracheal anastomosis. This can lead to gastroparesis
which can significantly increase the risk of recurrent aspiration and
is associated with the development of CLAD. It can also lead to
malabsorption and malnutrition.
Immunosuppression
There is no standardized immunosuppression regimen for HLTx recipients
and protocols vary between transplant centres.
Some centres advocate the use of ATG as induction therapy while other
centres opt for more modern agents such as IL2RAs (e.g. basiliximab).
However, the clinical benefit of induction immunosuppression remains
contentious.
Maintenance immunosuppression is similar to other thoracic organ trans-
plant recipients comprising of a combination of a CNI, a cell cycle inhibitor,
and a glucocorticoid.

Allograft monitoring
Monitoring of HLTx recipients is usually lead by pulmonologists because the
surveillance is primarily bronchoscopic and the majority of complications are
related to the lung allografts. Regular monitoring of the patients includes:
43

434 Chapter 33 Combined heart–lung transplantation

Lung allograft:
• CXR.
• Pulmonary function testing.
• Bronchoscopy and surveillance transbronchial biopsy.
Heart allograft:
• TTE.
• Endomyocardial biopsy if evidence of dysfunction.
Acute rejection of the cardiac allograft following HLTx is uncommon with
some suggesting that there may be a protective benefit of the lung allo-
grafts on the donor heart. One study focusing on endomyocardial biopsy
following HLTx reported a significantly lower incidence of acute rejec-
tion compared with isolated heart transplantation. As such, most centres
choose to only perform routine surveillance transbronchial biopsies unless
there is specific evidence of cardiac dysfunction.

Outcomes: patient survival

ISHLT data demonstrate that operative mortality following HLTx has re-
duced significantly over time but remains higher than that for isolated heart
or lung transplantation. Between 1982 and 1992 operative mortality was
25.4%, compared to 16.8% for 2002.
The reduction in operative mortality has been attributed to:
• Improved operative techniques.
• Improved organ preservation techniques and solutions.
• Improved patient selection.
• Greater understanding of the risk factors for adverse outcomes.
• Improvements in immunosuppressive regimens.
Median survival following HLTx has also improved over time and currently
stands at 5.8 years, although this remains lower than that for isolated lung
(7.1 years) and heart (10.4 years) transplantations performed over the
same era. This is attributed to the fact that recipients undergoing HLTx have
more complex diagnoses and have a more severe preoperative condition,
the procedure being infrequently performed, and the risk of complications
to both the heart and lung allografts.
However, for the recipients who survive their first year after a HLTx,
median survival is significantly higher at 11.5 years for the most recent era,
which is comparable to conditional survival following bilateral lung trans-
plantation but remains shorter than that for heart transplant recipients
(Fig. 33.7).
Due to the relatively small numbers performed, robust analysis is chal-
lenging, but factors that are associated with inferior survival following HLTx
include:
• Recipient age (p =​0.03).
• Donor age (p =​0.01).
• More modern transplant era (hazard ratio (HR) 0.661; confidence
interval (CI) 0.496–​0.883; p <0.01).
• Donor CMV+​/​recipient CMV–​(HR 1.438; CI 1.019–​2.028; p =​0.04).
• Ventilator dependence preoperatively (HR 2.270; CI 1.346–​3.829; p
<0.01).
 Causes of death in heart–lung transplant recipients 435

100

75
Survival (%)

50

25

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Years

1982–1993 (N = 1600) 1994–2003 (N = 1426)

2004–6/2016 (N = 972)

Fig. 33.7 Survival rate after HLTx.

Causes of death in heart–​lung transplant

recipients
The most common causes of death within the first year following HLTx in-
clude early graft failure, technical complications, and infections (Table 33.2).
Beyond the first year, the commonest causes of death are infection and
development of BOS.

Table 33.2 Causes of death within the first year compared to beyond the first
year (ISHLT registry)
<1 year >1 year
BOS 14 (1.7%) 224 (21.9%)
Acute rejection 15 (1.8%) 10 (1.0%)
Malignancy 16 (1.9%) 96 (9.4%)
Infection 206 (24.8%) 246 (24.0%)
Graft failure 204 (24.5%) 157 (15.3%)
Cardiovascular 54 (6.5%) 96 (9.4%)
Technical 123 (14.8%) 13 (1.3%)
Multiple organ failure 104 (12.5%) 60 (5.9%)
Source data from ISHLT registry report.
436

436 Chapter 33 Combined heart–lung transplantation

Morbidity following heart–​

lung transplantation
The commonest morbidities following HLTx are associated with the long-​
term side effects of immunosuppression (Table 33.3). A high proportion of
patients develop hypertension, dyslipidaemia, and diabetes following trans-
plantation. Some degree of renal dysfunction is very common and observed
in 45.5% of patients at 5 years.
Malignancy is another important morbidity, in particular PTLD. The in-
cidence of PTLD following HLTx is 7.6% compared to 5.4% and 3.1% fol-
lowing isolated heart or lung transplantation respectively at 5 years.

Table 33.3 Cumulative post-​transplant morbidity—​ISHLT registry

Outcome Within 1 year (%) Within 5 years (%)
Severe renal dysfunction 7.1 13.7
Creatinine >2.5 mg/​dL (221 3.0 9.5
µmol/​L)
Chronic dialysis 3.8 3.4
Renal transplant 0.2 0.8
Diabetes 17.1 26.5
Hyperlipidaemia 28.0 70.4
CAV 2.5 6.9
BOS 7.1 31.1
Malignancy 5.7 10.5

Source data from ISHLT registry report.

Chronic allograft dysfunction

The development of chronic allograft dysfunction is a major cause of long-​
term morbidity and mortality following HLTx (Fig. 33.8).
Cardiac
The incidence and morbidity associated with chronic cardiac allograft dys-
function is significantly lower in HLTx recipients compared to isolated heart
transplant recipients. Incidence of CAV in surviving HLTx recipients is 2.5%
and 6.9% at 1 and 5 years respectively. For isolated heart transplant recipi-
ents these values are 7.7% and 30%. Following experimental investigation
in animal models, this phenomenon is thought to be related to lymphoid
tissue within the lung allografts localizing the recipient alloimmune response,
thereby sparing the cardiac allograft from immune-​mediated injury.
Lung
The incidence of CLAD/​BOS is significantly higher than that of the cardiac
allograft. Incidence of CLAD in surviving HLTx recipients is 7.1% and 31.1%
at 1 and 5 years respectively and this is a major cause of morbidity and
mortality in the heart–​lung transplant population. The incidence of CLAD
 Retransplantation 437

100
Freedom from CAV and BOS (%)

75

50

25

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Years

Freedom from CAV (N = 417) Freedom from BOS (N = 489)

Fig. 33.8 Chronic allograft dysfunction following HLTx.

appears to be lower than that observed following isolated lung transplant-

ation where the 1-​and 5-​year incidence is 9.3% and 41% respectively. It is
thought that the higher incidence of PGD in isolated lung transplant recipi-
ents may be the explanation for this observation.

Retransplantation
There have been a small number of retransplantations performed following
HLTx. However, outcomes are significantly inferior compared to first-​time
HLTx recipients—​with a 1-​year survival of only 43% (Fig. 33.9).
100

75
Survival (%)

50

25

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Years

Primary (N = 3912) First Retransplant (N = 46)

Fig. 33.9 Survival after retransplantation following HLTx.

438
 Chapter 34 439

Combined
thoracoabdominal
organ transplantation
Introduction 440
Indications and overview 440
Donor operation 441
Recipient preparation and team coordination 442
Outcomes relative to isolated transplantations 443
Planned mechanical circulatory support for combined heart liver
transplant 444
Conclusion 446
40

440 Chapter 34 Thoraco-abdominal transplantation

Introduction
Dual thoracoabdominal organ transplantations are the most resource
intensive cases in solid organ transplantation. They have unique chal-
lenges and are performed in selected patients at high-​volume quater-
nary care centres. There are no standard indications for combined
thoracoabdominal transplants, and individual centres evaluate the neces-
sity on a case-​by-​case basis. The most common clinical scenarios and in-
dications are as follows:
• Heart–​kidney: severe HF with advanced renal disease not expected to
recover following heart transplantation.
• Lung–​kidney: lung retransplantation with concomitant nephrotoxicity
from CNIs.
• Heart–​liver: severe amyloidosis affecting both organs.
• Lung–​liver: CF with CF-​related liver disease.
At our institution, intraoperative management is performed by car-
diothoracic anaesthesiologists and perfusionists who manage CPB
and ECMO. In heart–​k idney (HKTx) and lung–​k idney transplantations
(LKTx), the thoracic transplantation is finished in its entirety including
chest closure, followed by a period of recovery in the ICU prior to
initiating the kidney transplantation. In contrast, heart–​liver and lung–​
liver transplantations (LLTx) are completed sequentially under the same
anaesthetic.
In summary, combined thoracic/​abdominal transplantations are com-
plex endeavours. Donor and recipient selection, organ procurement, an-
aesthetic management, recipient surgery, and postoperative management
require a multidisciplinary approach and will be discussed in detail in this
chapter.

Indications and overview

Combined heart/​lung–​kidney transplantation
Renal failure following heart and lung transplantation is associated with an
increased mortality risk, thus motivating combined HKTx or LKTx in pa-
tients with concomitant advanced renal disease.
Considerations for HKTx:
• Concomitant HF and renal insufficiency warrants consideration for
HKTx when the eGFR is <60 mL/​min/​1.73 m2 without necessitating
dialysis dependence.
• Evidence for this approach includes a UNOS analysis demonstrating
higher 5-​year survival rates for HKTx recipients compared to heart
transplant recipients with renal insufficiency.
• Excellent outcomes have been achieved in older patients and in
sensitized patients.
Considerations for LKTx:
• The most common indication for LKTx is retransplantation for lung graft
failure with concomitant CNI nephrotoxicity.
 Donor operation 441

• Patient survival after LKTx at 5 years is 51.4%, similar to isolated lung

transplantation, suggesting that LKTx is a feasible therapeutic option for
lung transplant candidates with significant renal dysfunction.
• Combined transplantation is associated with a lower risk of rejection
than single-​organ transplantation.
Combined heart–​liver transplantation
Less than 200 combined transplants have been performed in the USA since
1988 for the following indications:
1. Familial amyloidosis (the most common indication).
2. Familial hypercholesterolaemia.
3. Homozygous beta-​thalassaemia.
4. Haemochromatosis.
5. Alcoholic cardiomyopathy.
6. Cryptogenic cirrhosis with underlying cardiomyopathy.
7. Glycogen storage disease.
8. Late-​stage congenital heart disease status post previous repair.
Considerations for heart–​liver transplantation:
• Heart transplantation is performed first and the chest is left open,
followed by liver transplantation and then chest closure.
• The use of MCS devices after heart transplantation can facilitate the
liver implantation phase.
• Outcomes of the combined procedure are comparable to isolated
heart and liver transplantations.
Combined lung–​liver transplantation
Between 1990 and 2016, 131 combined LLTx were performed worldwide
for the following aetiologies:
1. CF.
2. COPD.
3. Alpha-​1 antitrypsin deficiency.
4. Idiopathic pulmonary fibrosis.
5. Idiopathic PA hypertension.
6. Sarcoidosis.
7. Retransplant.
Considerations for LLTx:
• In the USA, the severity of lung disease most commonly drives organ
allocation. From a liver perspective, recipient MELD scores range from
7 to 14, much lower than for patients undergoing liver transplant alone.
• The combination of high LAS and high MELD scores increases the risk
of the procedure and diminishes outcomes.

Donor operation
Although the technique for organ procurement varies from centre to
centre, the standard approaches to the thoracic donor operations are de-
scribed in Chapter 15 (heart procurement) and Chapter 26 (lung procure-
ment). Consideration is given to coordinate donor cross-​clamp time with
the recipient team to ensure optimal cold ischaemic time.
42

442 Chapter 34 Thoraco-abdominal transplantation

Recipient preparation and

team coordination
Combined heart–​kidney and lung–​kidney
The logistics of combined thoracic–​kidney transplantation are more flexible
due to the ability of the kidney to safely tolerate prolonged cold ischaemia.
In most cases, the kidney is maintained ex situ using hypothermic machine
perfusion while the recipient is stabilized in the ICU following completion of
thoracic transplantation. Typical time between the two procedures varies
from 6 to 24 hours, during which time the kidney is undergoing machine
perfusion. General approach:
• Thoracic transplantation completed, including chest closure.
• Recipient recovered in ICU with careful management of acid–​base
status, correct coagulopathy related to CPB, and stabilize thoracic graft
function.
• Ideally, vasopressor support is weaned off or down to a single agent
with improvement in the patient’s pH and lactic acidosis, given that renal
outcomes are impaired in patients who are on multiple vasopressors at
the time of renal allotransplantation.
• Renal transplantation is then typically carried out in a standard
retroperitoneal (heterotopic) approach although, when necessary, an
intraperitoneal approach can be utilized. The renal graft is placed on the
opposite side from any femoral arterial/​venous catheters.
In all instances postoperative management is carried out in a multidiscip-
linary fashion between thoracic and abdominal transplant teams (medical
and surgical). Postoperative immunosuppression is generally managed by
the thoracic transplant team as recipients of thoracic organ transplants re-
quire higher levels of immunosuppression when compared to abdominal
organs.
Combined heart–​liver and lung–​liver
This section focuses on the management and coordination of the multiorgan
transplant recipient between surgical teams (thoracic and abdominal) as
well as the anaesthetic team. Recipients of dual-​organ transplants will typ-
ically undergo thoracic organ transplantation first, followed by abdominal
organ transplantation, although a liver-​first approach is more commonly
being utilized for LLTx. The operative approach is typically carried out as
follows:
• Once a suitable donor becomes available, procurement teams will travel
to the donor hospital for evaluation and procurement of the intended
organs.
• Following visualization of organs in the donor, the donor team will
provide the ‘go-​ahead’ call to the recipient team.
• The intended recipient will be anaesthetized and undergo placement of
necessary intraoperative monitoring lines (arterial lines, central venous
catheter, Swan–​Ganz catheter, etc.) and routine TOE is performed.
• Recipient operation gets underway.
• In order to minimize cold ischaemic time, donor cross-​clamp does not
occur until the recipient operation is underway and progressing without
complication.
 Outcomes relative to isolated transplantations 443

• Following cross-​clamp and cold preservation solution administration,

the donor heart is removed first, followed by the lungs, liver, and then
kidneys.
• For combined heart–​liver transplantation the heart will undergo
implantation first and the chest left open; the recipient will then undergo
liver transplantation via a chevron incision.
• For LLTx, we favour performing a liver-​first approach.

Outcomes relative to isolated

transplantations
Combined thoracoabdominal organ transplantations involve significant
resource utilization, and ideal allocation strategies have not been de-
termined. In the USA, combined thoracoabdominal organ recipients
receive priority over recipients on the liver-​alone and kidney-​alone
waiting lists. The most common transplant scenario is ‘thoracic-​first’
allocation in which the thoracic organ is allocated to a given recipient
through the standard heart and lung allocation systems. By UNOS
policy, the corresponding abdominal organ is then allocated to the same
recipient ahead of all potential recipients on the liver alone or kidney
alone waiting lists. Although there is considerable debate regarding the
fairness of prioritizing dual-​organ recipients in the setting of the cur-
rent organ shortages, the outcomes of combined thoracoabdominal
transplantations have justified the current system. The following points
summarize the outcomes of combined thoracoabdominal organ trans-
plantations in comparison to their thoracic-​alone and abdominal-​alone
counterparts.
Heart–​kidney
• Long-​term outcome is comparable to heart transplantation alone with
10-​year patient survival of 56.7% for heart–​kidney recipients versus
53.6% for heart-​alone recipients (p =​0.13).
• There appears to be an immunological benefit for heart–​kidney
recipients, with significantly reduced incidence of treatment for acute
rejection in the first year (8.4% for heart–​kidney recipients vs 17.4% for
heart-​alone recipients, p <0.01).
• The 5-​year kidney graft survival is comparable between heart–​kidney
recipients and kidney-​alone recipients (72% vs 73%, p =​0.71).
• There does not appear to be an increased rate of futile kidney
transplants in the combined heart–​kidney population which lends
support to the current allocation system in the USA.
Lung–​kidney
• Outcomes from LKTx are the least defined.
• There appears to be a trend towards reduced patient survival
compared to lung transplant alone at 1 year (71.0% vs 81.7%, p =​
0.06); however, by 5 years patient survival was comparable (59.9% vs
51.4%, p =​0.55).
4

444 Chapter 34 Thoraco-abdominal transplantation

Heart–​liver
• Long-​term outcome is comparable to heart transplantation alone, with
10-​year patient survival of 60.4% for combined heart–​liver transplant
compared to 53.6% for heart-​alone (p =​0.09).
• There appears to be an immunological advantage for heart–​liver
recipients, with significantly decreased rates of acute rejection during
the index hospitalization (3.1% vs 7.1%, p =​0.03) and during the first
year post transplantation (2.1% vs 17.4%, p <0.01).
• Liver graft survival is comparable between combined heart–​liver and
liver transplantation alone at 1 year, 5 years, and 10 years (83.4%,
72.8%, and 71.0% vs 79.4%, 71.0%, and 65.1% for liver transplantation
alone, p =​0.894).
Lung–​liver
• Patient survival is comparable to lung transplantation alone and appears
to be improving in the modern era.
• A UNOS analysis in the era following implementation of the LAS
demonstrated a 1-​year survival of 82.7% and 5-​year survival of 69.0%.
• Historically, liver graft survival has been impaired in combined LLTx
compared to liver transplantation alone.
• However, the favourable patient survival for LLTx in the modern
era described previously indicates that this may now more closely
approximate outcomes for liver transplantation alone.

Planned mechanical circulatory support

for combined heart liver transplant
RV dysfunction post heart transplantation is a common problem; its like-
lihood is even higher following a combined heart–​liver transplantation.
Recently, we have performed combined heart–​ liver transplants with
elective placement of an extracorporeal RVAD in order to protect the
newly transplanted heart from high-​dose inotropic support and facilitate
the liver implantation (Fig. 34.1). This temporary support has maintained
haemodynamic stability and prevented liver congestion from right heart
dysfunction. Importantly, we use this approach in all heart–​liver patients,
regardless of their predisposition to RV dysfunction. The procedure is out-
lined as follows:
• Standard bicaval heart transplantation. Once completed, adequate
de-​airing is assured using TOE and the patient is separated from CPB
decannulated and protamine is administrated to reverse heparinization.
• Once the patient is deemed to be stable and haemostasis is achieved,
5000 units of heparin are given IV and an extracorporeal RVAD is
placed to support the circulation during the liver transplant. This enables
stable haemodynamics during IVC clamping and liver reperfusion.
• A typical configuration for the RVAD is as follows:
• RA cannula (34 Fr plastic right angle tip) is placed through purse string
in the RAA.
• Femoral venous cannula (21 Fr multiple stage) is percutaneously
placed and positioned below the level of the caval anastomosis.
 MCS FOR HEART-LIVER TRANSPLANT 445

• These two cannulas are connected together thorough a Y connection

to provide inflow to a Rotaflow centrifugal pump (MAQUET
Cardiopulmonary AG, Hirrlingen, Germany) (Fig. 34.1).
• The pump outflow consists of a 19 Fr arterial cannula placed through
purse-​string sutures in the main PA.
• The pump speed is set to achieve about 3–​4 L/​min of flow. Attention
must be paid to avoid volume overload of the LV.
• Midline position of the interventricular and the interatrial septa on TOE
is reassuring for adequate filling of the LV.
• Inotropic support may still be required at this point for adequate
function of the LV.
• Once the liver transplantation is completed, the RVAD is weaned with
attention to the CVP, the cannulas are removed, and the purse strings
are tied.
• An elevated CVP may warrant more prolonged RVAD support which
can be achieved with just one of the venous cannulas.

RVAD

Fig. 34.1 A schematic illustration of the RVAD configuration during liver

transplantation. Once the heart is transplanted and the patient is stable, the inflow
to the RVAD from both the RA and the femoral vein are connected using a Y
connector. The outflow Cannula to the PA is connected and the RVAD support is
initiated and titrated to a speed that will support the heart, and ensure a bloodless
field for the liver transplant team but will not cause overflow to the lungs and
the LV. Next, the IVC is clamped below the heart and below the liver, and liver
transplantation begins.
46

446 Chapter 34 Thoraco-abdominal transplantation

• Once the patient is stable, the chest is closed and the patient is
transferred to the ICU.
Placement of an extracorporeal RVAD is safe and technically straightfor-
ward, but one should keep in mind some potential pitfalls:
• ACT should be kept around 200–​250 seconds during the procedure.
• Patients may experience bleeding complications if the ACT is not kept
in a tight range (we have used very low doses of heparin during RVAD
support).
• The cardiac surgeon must remain involved throughout the liver
transplantation to address any issues related to flow, cannulas, and
haemodynamics if needed.
In our experience, elective RVAD placement in heart–​liver transplantation
can be performed safely and effectively. Preoperative planning involving all
of the surgical and anaesthesia teams is critical to achieve a good result.

Conclusion
Thoracoabdominal transplantations are complex and rare cases that are re-
served for specific patients who suffer from multiorgan dysfunction. Team
coordination and a well-​planned operation are paramount for the success
of any combined organ transplantation.
 Section 8

Anaesthetic
considerations

35 Anaesthetic issues related to extracorporeal

membrane oxygenation and ventricular assist device
procedures 449
36 Anaesthetic issues related to heart
transplantation 459
37 Anaesthetic issues related to lung
transplantation 469
48
 Chapter 35 449

Anaesthetic issues
related to extracorporeal
membrane oxygenation
and ventricular assist
device procedures
Extracorporeal membrane oxygenation 450
Ventricular assist devices 453
450

450 Chapter 35 Anaesthesia for ECMO and VAD procedures

Extracorporeal membrane oxygenation

Indications for use: ECMO
The ELSO publishes guidelines for ECMO use (http://​www.elso.org) which
can include acute pulmonary and/​or severe cardiac failure that may be re-
versible and non-​responsive to current therapy. Clinical situations that may
prompt the initiation of ECMO include the following:
• Severe hypoxaemic or acute respiratory failure with a PaO2/​FiO2 ratio
of <100 mmHg despite optimization of ventilation (i.e. tidal volume,
PEEP, and inspiratory to expiratory (I:E) ratio).
• Hypercapnic respiratory failure with an arterial pH <7.20.
• Bridge to lung transplantation.
• Circulatory failure or refractory cardiogenic shock.
• Massive pulmonary embolism.
• Cardiac arrest.
• Inability to wean from CPB after cardiac surgery.
• Short-​term bridge to cardiac or lung transplantation or placement of a
VAD.
Absolute contraindication to ECMO is any pre-​existing state which is in-
compatible with recovery (severe neurological injury, end-​ stage malig-
nancy). Relative contraindications include uncontrollable bleeding and poor
prognosis from the primary condition.
During ECMO, blood is drained from the native vascular system, cir-
culated outside the body by a mechanical pump through an oxygenator
and heat exchanger, and returned to the circulation. The oxygenator fully
saturates haemoglobin with oxygen, while carbon dioxide is removed.
Oxygenation is determined by flow rate, where elimination of carbon di-
oxide can be controlled by adjusting the rate of countercurrent gas flow
(sweep) through the oxygenator.
There are two types of ECMO:
• VV ECMO: blood is extracted from the vena cava/​RA and returned
to the RA; provides respiratory support and requires an intact
cardiovascular system to provide stable haemodynamics.
• Venous cannulas are usually placed in one of the common femoral
veins (venous drainage) and right IJ vein (arterial return). The tip of
the femoral cannula should be maintained near the junction of the
IVC and RA, while the tip of the IJ cannula should be maintained
near the junction of the SVC and RA. Alternatively, a double-​lumen
cannula is available that is large enough to accommodate 4–​5 L/​
min of blood flow. It is available in a variety of sizes, 31 Fr being the
largest and most appropriate for adult males. 27 Fr being sufficient for
most average sized women.
• VA ECMO: blood is extracted from the SVC or RA (venous drainage)
and returned to the femoral arterial or aorta (arterial return),
completely bypassing the heart and lungs, and thereby providing both
respiratory and haemodynamic support.
• With femoral access comes an increased risk of ischaemia of
the ipsilateral lower extremity. The risks of this complication are
decreased by inserting a separate arterial cannula distal to the femoral
artery cannula and directing a portion of the infused blood for
 Extracorporeal membrane oxygenation 451

perfusion of the distal extremity. In those patients transitioning from

CPB, the cannulas used for bypass can be transferred from the heart/​
lung machine to the ECMO circuit, with blood drained from the RA
and reinfused into the ascending aorta.
Once ECMO has be initiated, and blood flow and haemodynamics are
stabilized, therapy is guided by blood gases and targets include:
• PaO2 >90% for VA ECMO; >75% for VV ECMO.
• Venous oxyhaemoglobin 20–​25% < arterial saturation.
• Evidence of adequate tissue perfusion (arterial blood gas, venous
oxygen saturation, and blood lactate levels).
Maintenance
After haemodynamic and laboratory data demonstrate stability, blood flow
rates are kept constant. Constant assessment using continuous venous ox-
imetry via PAC or direct measures of the oxyhaemoglobin saturation of the
blood in the venous limb of the ECMO circuit is performed. If venous oxy-
haemoglobin saturations fall below target, increasing one or more of the
following may be helpful: rate of flow, intravascular volume, or haemoglobin
concentration. Some practitioners will reduce patient temperature (passive
cooling) to reduce systemic oxygen uptake.
Anticoagulation
Heparin is most often used during ECMO to maintain anticoagulation and
should be titrated to an ACT of 180–​210 seconds. If bleeding develops,
the targeted ACT is decreased. The anticoagulant effect of heparin is de-
pendent on endogenous levels of anti-​thrombin (AT3). AT3 levels should
be measured if concern arises and if levels are found to be <50% of normal,
then AT3 can be administered either by giving fresh frozen plasma or re-
combinant AT3.
Platelets
Platelet transfusions are common with ECMO use as the platelets are
constantly being consumed. ECMO circuits activate platelets by constant
exposure to very large surface areas of plastic. Therefore, platelet counts
should be maintained at ≥50,000/​mL.
Red cell transfusions
Because ECMO is providing the only source of oxygen in the setting of
complete respiratory ± cardiac failure, oxygen delivery must be optimized.
This is done as a balance between flow and amount of haemoglobin, which
is maintained at >12 g/​dL in ECMO patients.
Ventilation
Ventilation should be optimized to avoid volutrauma, barotrauma, and
oxygen toxicity while preventing atelectasis. FiO2 should be kept at <0.5
and plate pressures should be kept around 20 cmH2O. Minimizing venti-
lator support will improve venous return and therefore cardiac output by
reducing intrathoracic pressure. Tracheostomy can be performed while on
ECMO with minimal risks and can improve ventilation by reducing head-
space and improving patient comfort through early extubation.
452

452 Chapter 35 Anaesthesia for ECMO and VAD procedures

Special considerations
• Blood flow should be maintained at near maximum flow rates during
VV ECMO to optimize oxygen delivery. However, in VA ECMO, the
flow rates should be just high enough to maintain adequate perfusion
pressure and venous oxyhaemoglobin saturation, but low enough to
provide sufficient preload to maintain LV output.
• Diuresis may be necessary as most patients are volume overloaded on
initiation of ECMO. Ultrafiltration can be added to the ECMO circuit if
patients experience renal failure.
• LV monitoring must be performed and must be continuous. LV failure
may occur due to insufficient offloading of the ventricle, underlying
ventricular dysfunction at baseline is now unmasked. Arterial
waveform monitoring and TOE are optimal methods of ensuring
continued ejection of the ventricle. If ventricular failure begins to
occur, the addition of inotropic support (i.e. milrinone or dobutamine)
can be started as well as mechanical support through use of IABP
counterpulsation. Immediate LV decompression is essential to avoid
pulmonary haemorrhage if LV ejection cannot be maintained despite
IABP counter-​pulsation and inotropic agents.
Weaning from ECMO
One or more trials of taking the patient off ECMO should be performed
prior to discontinuing ECMO permanently:
• VV ECMO weaning trials are performed by decreasing the sweep gas
through the oxygenator while maintaining blood flow at a constant.
Arterial blood gases are observed for several hours during which time
appropriate ventilator settings required for transition are determined.
• VA ECMO weaning trials require temporary clamping of both the
drainage and infusion lines, while allowing the ECMO circuit to circulate
through a bridge between the arterial and venous limbs to minimize
thrombus formation within the ECMO circuit or oxygenator. Due to
the heightened risk of clot formation, VA ECMO trials are shorter in
duration than VV ECMO trials.
Complications of ECMO
Bleeding
• Occurs in 30–​50% of patients on ECMO and can be life-​threatening.
• Increased risk from continuous anticoagulation and platelet dysfunction.
• Platelet counts >50,000/​microL and maintaining tight targets of ACT
reduce risk of bleeding.
• Early surgical intervention if major bleeding occurs within body cavities
(i.e. abdomen or pleural space).
Thromboembolism
• Systemic thromboembolism due to thrombus formation is a devastating
complication that occurs in 10–​20% of cases.
• VA ECMO has worse outcomes with thrombus formation than
VV ECMO.
• Hypervigilance for thrombus formation within the oxygenator
must be maintained and early detection can successfully prevent
thromboembolism in most patients.
 Ventricular assist devices 453

• Observation and inspection of all connectors and monitoring the

pressure gradient across the oxygenator is critical. Any sudden change
in the pressure gradient is highly suggestive that a thrombus is present.
• Large or mobile clots require immediate circuit or component exchange.
• Primed circuits should be kept at the bedside if the target ACT has
been reduced due to bleeding because the risk of thrombus formation
is increased.
Neurological
• Neurological injury in the ELSO registry is 10%.
• Neurological injury in cardiac failure and those in whom ECMO is
administered during CPR is 50%.
Other complications
• Vessel perforation, arterial dissection, distal ischaemia may occur due to
malpositioned cannula (drainage or arterial return).
• HIT can arise in patients on ECMO receiving heparin as anticoagulation.
• Pulmonary oedema and haemorrhage can occur during VA ECMO if LA
pressures exceed 25 mmHg.
• Intracardiac thrombosis may form if LV ejection is not maintained during
VA ECMO use.
• Cerebral or coronary hypoxia can occur from preferential perfusion of
the gut and lower extremities. If unrecognized, cerebral hypoxia can
result with catastrophic results.

Ventricular assist devices

Indications for use: VAD
VADs perform as an adjunct cardiac pump, taking blood returning to a
failing ventricle and ejecting it downstream, either into the ascending or
descending aorta in the case of a LVAD, or into the PA in the case of a
RVAD. VADS can be used in the following ways:
• Bridge to transplantation: a VAD may be implanted in a candidate for
cardiac transplantation to support a chronically failing ventricle as a
stopgap until a donor organ is available.
• Bridge to candidacy: placed in patients who are not currently eligible for
transplantation but may be able to achieve eligibility after substantial
improvement in end-​organ function.
• Destination therapy: in some patients, a long-​term VAD is implanted for
permanent ventricular support.
• Bridge to recovery: short-​term use, available for rapid implantation when
MCS is acutely needed for survival.
The devices used are continuous-​flow pumps, which produce pulseless
blood flow, that unload the failing ventricle through the action of an axial
or centrifugal impeller that rotates at very high speeds. Device components
include an impeller pump, and an external driveline connecting to an elec-
trical power source. These non-​pulsatile devices are placed in the thoracic
cavity with blood flowing through an inflow cannula in the apex of the LV to
the pump and returning it back through an outflow cannula in the ascending
or descending aorta.
45

454 Chapter 35 Anaesthesia for ECMO and VAD procedures

Pre-​anaesthetic consultation
Detailed discussion with knowledgeable clinicians having training in MCS
(e.g. a specialized VAD team) is extremely important. This facilitates
awareness of potential patient-​specific and device-​specific problems and
avoidance of pitfalls that may arise in the perioperative period, as well as
coordination of care in the postoperative period.
Preoperative checklist-​specific issues
• Understand the type of device, implantation date, pump speed, pump
flow, pump power, and pulsatility index.
• Identify the location of the externalized driveline on the abdominal wall,
looking for any evidence of infection.
• Is there a previous history of difficulties with the VAD function, any
VAD-​related complications (e.g. thromboembolism, infection, mucosal
bleeding), and is the device functioning adequately for the patient’s
current clinical status?
• Is there any evidence of end-​organ damage or impairment?
• Identify cardiovascular comorbidities (original indication for VAD
implantation, current EF, presence of right HF, aortic insufficiency,
arrhythmias, physical activity level):
• Is there evidence of right HF? Right HF may be exacerbated by
decreases in LV pressure and size after VAD implantation, leading to
interventricular septal bowing and distortion of RV geometry and
mechanics, particularly if the patient had underlying biventricular
dysfunction.
• Are there any haematological abnormalities, increased risk of
perioperative thrombosis or, conversely, coagulopathy? What is the risk
of bleeding requiring transfusion?
• What is the anticoagulation strategy for the patient? Thromboembolism
remains one of the most serious complications of VAD implantation.
All types of VADs require some type of anticoagulation, which will
be noted in the history, and the presence of anaemia or coagulopathy
identified on preoperative laboratory tests:
• Although the level of anticoagulation may be decreased towards
the lower limit of the manufacturer’s recommendation during
the perioperative period, one should not completely reverse
anticoagulation in VAD-​supported patients for any elective surgical
procedures. Exceptions include neurological procedures and
some ophthalmological procedures. The anaesthesiologist, VAD
management team, and surgeon should collaborate to plan a safe
anticoagulation plan during the perioperative period, which may
include a heparin bridge in selected patients.
• Is there a pacemaker and/​or ICD? Generally, perioperative
considerations regarding pacemakers and ICDs do not differ in VAD-​
supported patients. However, defibrillator pads must be positioned on
an area of the chest that is not directly over the device or driveline. The
best options are bilateral pad placement on opposite sides of the chest or
anterior–​posterior placement in the centre of the chest and on the back.
• Are there any pulmonary comorbidities that can be optimized (e.g. PH,
pre-​existing COPD, pneumonia, recent changes in oxygen requirements,
or decreased activity)?
 Ventricular assist devices 455

• Are there neurological comorbidities (e.g. abnormal mental status, prior

cerebrovascular accident, or embolic stroke)?
• Identify any renal comorbidity (e.g. renal insufficiency, requirement for
dialysis, urinary tract infection).
• Identify any hepatic comorbidity (e.g. liver insufficiency, congestive
hepatopathy, coagulopathy).
• Does the patient have diabetes and difficulties with glycaemic control?
• Is the patient at risk for infectious complications (e.g. related to VAD
implantation)?
Planning for postoperative care
• Planning for postoperative care should be to the most appropriate
location to deliver postoperative care.
• Planning for postoperative care should begin when the case is posted.
• Postoperative care should be in an area with appropriate monitoring
(telemetry) and staff specifically trained in emergency care of such
patients.
• Recovery from surgery and anaesthesia in the ICU may be appropriate.
Anaesthetic technique
• Patients supported with a VAD receive general anaesthesia for most
surgical interventions.
• Due to the need for anticoagulation, neuraxial anaesthetic techniques
are contraindicated in most VAD patients.
• Placement of a peripheral nerve block with ultrasound guidance is
relatively safe in anticoagulated patients.
• Sedation with monitored anaesthesia care is usually a good choice when
feasible (e.g. endoscopy procedures).
Intraoperative management
• Intraoperative management of a patient with a VAD must include
haemodynamic and VAD monitoring.
• Involvement of or consultation with a cardiac anaesthesiologist/​
cardiologist or surgeon with expertise in VAD management may
provide additional support.
• All patients with a VAD should be transported within the hospital with
the VAD under battery power, including transport to the preoperative
holding area, the operating room, post-​anaesthesia care unit, or ICU.
• After transport, the VAD should be promptly switched from battery
power and connected to the power base unit of the device, which is
connected to a wall electrical power outlet.
• Backup batteries should always be available; therefore, it is best to
connect the VAD to an external power outlet to save battery power.
Intraoperative monitoring
• VAD parameters may be monitored on a console.
• Second-​and third-​generation non-​pulsatile VADs provide continuous
displays of pump speed, flow, power, and the dimensionless pulsatility
index :
• Pump speed is measured in rpm.
• Pump flow is measured in L/​min.
456

456 Chapter 35 Anaesthesia for ECMO and VAD procedures

• Pump power is measured in watts (W).

• Pulsatility index is a dimensionless measure of the extent of LV
pulsatility.
Non-​invasive monitors
• Patients with non-​pulsatile VADs often have no palpable pulse and
minimal arterial pulse pressure; as such, automated non-​invasive blood
pressure monitors are only functional in approximately one-​half of
VAD-​supported patients:
• Hypovolaemia and/​or vasodilatation further reduce pulsatility in
patients with non-​pulsatile devices, with consequent reduction in
efficacy of non-​invasive monitors.
• Even a small pulse pressure (i.e. 10–​15 mmHg) typically allows use of
standard non-​invasive monitoring, which may be acceptable for minor
surgical procedures (e.g. colonoscopy, bronchoscopy, TOE).
• An intra-​arterial catheter provides the most accurate and continuous
blood pressure monitoring in VAD-​supported patients:
• Placement of an intra-​arterial catheter may be difficult using palpation
alone and is typically facilitated with ultrasound guidance.
Transoesophageal echocardiography
• Availability of equipment/​personnel to use TOE is useful in major
elective surgical procedures and serves as a partial substitute for PAC
placement in selected patients.
• TOE is valuable to evaluate the following conditions, particularly in
unstable patients:
• Intravascular volume status.
• RV function.
• Aortic valve opening.
• Aortic insufficiency.
• Interventricular septal shifts.
• Correct position of the VAD cannula.
• In the event of hypotension or cardiac arrest, TOE can be used to
rapidly determine contributing factors.
Central venous or pulmonary artery catheter
• A PAC may provide information to guide intraoperative fluid
management and avoid increases in PVR.
• A central venous catheter may be inserted rather than a PAC in patients
without significant right heart dysfunction, with use of the CVP to
monitor right heart function.
• A central venous catheter also provides large-​bore access for infusion
of IV fluids or blood and vasoactive infusions when these needs are
anticipated.
Positioning considerations
• Surgical positioning may alter VAD function and impact haemodynamics.
• Placement of the patient in the lateral decubitus position with one-​lung
ventilation may result in hypoxia and/​or hypercarbia increasing PVR,
limiting preload to the VAD, and leading to acute RV failure.
• Reverse Trendelenburg position decreases venous return to the heart;
Trendelenburg position increases venous return.
 Ventricular assist devices 457

Induction/​maintenance of general anaesthesia

• The decision to perform endotracheal intubation or use a supraglottic
airway in a VAD-​supported patient is based on the usual criteria for
airway management.
• Patients supported with left-​sided VADs may develop telangiectasias on
mucosal surfaces, with epistaxis being a frequent complication.
• Second-​and third-​generation non-​pulsatile VADs are relatively small
in size and typically have an intrathoracic location and do not typically
impair gastric emptying.
• Selection of induction drugs and dosing should take into consideration
the degree of dysfunction of the unassisted ventricle.
• Maintaining adequate preload, afterload, heart rate, and rhythm are
critical for haemodynamic stability.
• Reasonable targets for intraoperative MAP are within 10% of the
patient’s normal MAP, but no lower than 70–​80 mmHg.
• A VAD will only pump what is delivered to it; maintenance of
intravascular volume status is necessary to provide adequate preload for
optimal VAD function:
• Factors that decrease preload decrease pump flow and LV output
and include anaesthetic agents, dehydration, haemorrhage, and lateral
decubitus and reverse Trendelenburg positions.
• If VAD flow exceeds the available LV preload, the walls of the LV
near the inflow conduit can collapse and limit VAD inflow.
• VADs are ‘afterload-​sensitive’ devices. Increases in afterload (i.e.
systemic arterial hypertension) will decrease pump flow and reduce
VAD output:
• Appropriate depth of anaesthesia should be achieved prior to
noxious stimuli such as laryngoscopy or skin incision, and adequate
analgesia should be ensured during and after the operation has
concluded.
• Arrhythmias causing clinically significant tachycardia or bradycardia
interfere with optimal VAD function and can reduce pump flow.
Management of arrhythmias in these patients is similar to that for
patients without a VAD.
• Adequate RV function must be assured by minimizing PVR since output
from the RV determines the volume ultimately ejected by a left-​sided
VAD. Thus, increases in PVR due to hypoxaemia, hypercarbia, pain,
alpha-​agonist vasopressors, hypothermia, and/​or acidosis should be
avoided.
• If power is normal, high flow may indicate a low SVR; thus, a
vasopressor should be administered to increase vascular tone
(afterload) and maintain MAP at 70–​80 mmHg.
• If power is high in a hypotensive patient, the cause may be severe aortic
insufficiency:
• A sudden increase in power may indicate device thrombosis or
malfunction.
• These causes can be diagnosed with TOE examination.
• Cardiac arrest or severe hypotension (MAP ≤50 mmHg) is challenging
to diagnose in a patient with a non-​pulsatile VAD due to difficulty in
measuring blood pressure in a patient without an arterial line.
458

458 Chapter 35 Anaesthesia for ECMO and VAD procedures

• The first step in management is to ensure adequate airway and

breathing, similar to patients without a VAD.
• In patients with a functioning non-​pulsatile VAD, causes of severe
hypotension include:
• Hypovolaemia.
• RV failure.
• Inflow cannula obstruction.
• Outflow cannula obstruction.
• Tamponade.
• Sepsis.
• TOE can be employed to rapidly determine which of these factors is
likely.
Emergence and extubation
• Extubation criteria are the same as in patients without a VAD.
• Prolonged tracheal intubation and mechanical ventilation should be
avoided, if possible, to decrease the risk of respiratory infection.
Postoperative management
• Postoperative management should be conducted in a fully monitored
setting. Non-​invasive and invasive monitoring are continued until major
bleeding, fluid shifts, and severe pain have been managed, and implanted
devices have been checked.
• After transport from the operating room and upon arrival to the post-​
anaesthesia care unit or other monitored care setting, such as the ICU,
the VAD is promptly connected to an external power outlet and the
battery packs are simultaneously connected to allow recharging.
• Similar to the intraoperative period, haemodynamic stability in the
postoperative period is maintained by providing adequate preload,
maintaining afterload, adjusting pump speed as needed, ensuring an
adequate heart rate and rhythm, and preventing any haemodynamic
changes that would compromise function of the RV.
• Goals of pain management are to achieve adequate analgesia while
maintaining haemodynamic stability:
• Hypertension associated with pain should be avoided since this may
increase afterload and decrease pump output.
• Over-​sedation with opioids or other sedative-​analgesics must also be
avoided, since this may lead to hypercarbia, hypoxia, increased PVR,
and RV dysfunction.
• Reinstitution of prior pacemaker or ICD settings should be assured
shortly after the patient’s arrival in the post-​anaesthesia care unit, ICU,
or other monitored care setting:
• Maintenance of adequate rhythm and rate is necessary for
proper VAD function; patients with a pacemaker or ICD should
have the device interrogated by the electrophysiology team and
reprogrammed if necessary.
• Patients supported by VADs require anticoagulant and antiplatelet
therapy to reduce the risk of thrombotic complications such as
device thrombosis and embolic stroke. These agents are resumed
when feasible (depending on the type of surgery and amount of
postoperative bleeding).
 Chapter 36 459

Anaesthetic issues related

to heart transplantation
Preoperatively 460
Intraoperatively 462
Postoperatively 467
460

460 Chapter 36 Heart transplantation: anesthetic management

Preoperatively
Timings
Heart transplantation is inevitably performed as an emergency procedure.
One of the key aspects is timing. There is the requirement to coordinate
two operations at individual sites in order to minimize the ischaemic time
between cross-​clamp at the donor site to reperfusion at the recipient site.
The often-​complex logistics mean that timing estimates may be continu-
ally changing. It is important to allow ample time for drug and equipment
preparation, patient assessment, and any procedures that are required. The
timing of moving to theatre, induction of anaesthesia, and duration of sur-
gical dissection can be critical, meaning frequent and proactive communica-
tion with all members of the multidisciplinary team is essential. Utilization
of the Organ Care System (OCS) Heart (TransMedics, Andover, MA, USA)
makes it possible for the beating donor heart to be preserved throughout
transfer from donor to recipient, bringing increased flexibility for timings in
challenging patient groups (e.g. those dependent on a LVAD).
Preoperative assessment
The preoperative status of patients undergoing heart transplantation can
range from ambulatory at home on oral medications alone, to being man-
aged as an in-​patient in the ICU on MCS.
History and examination
Most of these patients will have been thoroughly investigated at the time
of transplantation. However, it is important to ensure there has been no
change in their health status since they were last assessed and that recent in-
vestigations are reviewed. The ECG, echocardiogram, left and right cardiac
catheterization results, CXR, and pulmonary function test results should be
assessed. Any concomitant organ dysfunction should be identified.
Up-​to-​date haematology and biochemistry results are necessary along
with cross-​matching of blood (4 units for transplant, 6 units if previous
sternotomy or currently on mechanical support). Identification of infec-
tion or deteriorating organ function may influence patient management
perioperatively or may affect their eligibility for transplantation. Assessment
of the level of cardiac support the patient is currently receiving is important.
Known allergies, along with previous anaesthetic and surgical his-
tory should be reviewed. Particularly relevant is prior cardiac surgery or
sternotomy, as this will influence the time required for explantation of the
native heart.
Assessment of the airway and fasting status are important to ensure safe
airway management intraoperatively. Clinical examination should be per-
formed to assess volume status and explore potential difficulties with vas-
cular access.
Medications
The use of medications for management of HF may impact management in
the perioperative period. The patient may be stable on long-​term HF man-
agement or may be deteriorating on escalating doses of inotropes.
Diuretic use may lead to biochemical abnormalities that may need to
be corrected in the perioperative period. Any vasoactive medications
should be continued until safely established on CPB and doses may need
 Preoperatively 461

to be adjusted in order to anticipate the deleterious effects of anaesthesia

and positive pressure ventilation on loading conditions and myocardial
function.
ACEis, ARBS, BBs, and the angiotensin receptor–​ neprilysin inhibitor
sacubitril/​valsartan are used to control the neurohormonal changes associ-
ated with HF and prevent the remodelling that occurs. They can have pro-
nounced effects on haemodynamics with reduced LV afterload which can
improve myocardial performance. These effects can be amplified in those
patients undergoing general anaesthesia, leading to severe hypotension.
This may necessitate the use of vasopressors to support adequate haemo-
dynamics intraoperatively.
The long-​term use of anticoagulants in selected HF patients means that
consideration needs to be given, where possible, to reversal. Ideally, pa-
tients listed for transplantation should be managed on warfarin or UFH to
simplify perioperative management. Warfarin can be reversed with pro-
thrombin complex concentrate or fresh frozen plasma along with IV vitamin
K. UFH infusions may be discontinued prior to surgery or continued until
CPB. Haematology involvement and advice may be required for more com-
plex scenarios.
Preoperative implantable cardiac devices
The use of CRT devices, ICDs, and pacemakers is common in patients with
advanced HF. These devices should be checked preoperatively to confirm
the patient’s underlying rhythm and appropriate device function. Fixed-​
mode pacing may be required intraoperatively to avoid interference from
diathermy. ICD therapy should be disabled and transcutaneous pads ap-
plied as a replacement. Device leads will usually be cut and removed at
the time of explantation and the generator box removed at the end of the
procedure.
Preoperative MCS
This may include short-​term MCS devices such as IABP, short-​term VADs,
and VA ECMO or long-​term devices such as durable VADs.
In those patients with mechanical support, it is important to review
device function, anticoagulation management, and complications or
issues while on MCS and other organ support. They will require add-
itional personnel and time to safely transfer to the operating room.
Arterial access may be challenging (patients supported by continuous-​
flow devices may not have a palpable pulse) and anticoagulation will
need to be reversed.
Patients on central MCS need additional planning for surgery as additional
time will be required to access the chest and establish CPB. They will be
at increased risk of intraoperative bleeding from adhesions and from pre-
operative use of anticoagulation.
Immunosuppression
Successful immunosuppression is key to long-​term success of the transplant
and the regimen begins in the perioperative phase. Specific regimens are
largely dependent on local protocols and the potential side effect profile
in each patient, but treatment may begin in the preoperative phase and be
continued intraoperatively.
462

462 Chapter 36 Heart transplantation: anesthetic management

Surgical brief
The surgical brief should take place once the decision has been taken to
proceed to surgery. It is an opportunity to introduce the perioperative care
team and facilitate communication between the different specialties in-
volved. It should include a summary of patient and donor details, proposed
timings for key events, an outline of planned perioperative management
including surgical approach, cardiovascular support, and immunosuppres-
sion therapy.

Intraoperatively
The time of arrival into the operating room is key. There needs to be
enough time to establish vascular access, safely anaesthetize the recipient,
and initiate CPB in time for the donor heart arriving in the operating room.
The intraoperative time can be defined by three phases: pre CPB, CPB,
and post CPB.
Pre cardiopulmonary bypass
This is the period in preparation for donor heart arrival and needs to be
coordinated to ensure the recipient is established on CPB at the time the
donor heart arrives. During this period, priorities are safe induction and
maintenance of anaesthesia, haemodynamic stability, and end-​organ perfu-
sion and preparation for CPB.
Monitoring
Routine monitoring including continuous ECG, invasive blood pressure
(non-​invasive if invasive not yet established), pulse oximetry, capnography,
nasopharyngeal temperature, urinary catheter, and depth of anaesthesia
monitor should all be used. Additional monitoring such as CVP and PAC
measurements (including PCWP and cardiac output) can be instituted as
appropriate vascular devices are placed.
Induction and maintenance
No specific anaesthetic agents are mandated or preferred for induction
or maintenance of anaesthesia with choice of agent dependent on experi-
ence and preference of anaesthetists involved in patient care. The goal is to
achieve balanced anaesthesia without large changes in preload, afterload,
contractility, rate, or rhythm. Inotropic and vasoconstrictive support is
often required in the pre-​CPB period. Depending on fasting status, rapid
sequence induction of anaesthesia may be necessary.
In those patients established on MCS, there can be relative stability for
induction of anaesthesia where the pump is performing the work of the
heart. Volume replacement and vasopressor infusion may be required to
maintain perfusion pressure during this time. Although there is no evidence
specific to transplantation, there is no benefit to either total IV anaesthesia
or volatile anaesthesia in patients undergoing cardiac surgery.
Vascular access
Vascular access can be challenging as a result of multiple previous cannu-
lation attempts, low intravascular volume, and current implantable cardiac
devices. Traditionally, the right IJ vein has been avoided in order to ensure
 Intraoperatively 463

patency of this vessel for subsequent long-​term cardiac biopsies to be per-

formed from this site. However, this is not borne out in clinical practice and
any site can be used. In those with an ICD in situ, the left subclavian vessels
may not be an option.
Given the requirement for long-​term immunosuppression, a strict ap-
proach to asepsis is essential to prevent infection. Ultrasound use should be
routine and a multi-​lumen central venous catheter and sheath introducer,
to allow insertion of a PAC, should be placed. If the PAC is used prior to
heart excision, the catheter needs to be pulled back before explantation
of the recipient heart and can be re-​advanced later for assessment of the
transplanted heart.
The decision to place vascular access before or after induction of gen-
eral anaesthesia will depend on multiple factors. The perceived difficulty,
patient cooperation, staff experience, cardiovascular stability, logistics, and
timing will all need to be taken into account. Placing vascular access before
induction of general anaesthesia may minimize the time from induction and
intubation to initiation of CPB and may assist in delivery of vasopressor and
inotropic support during induction. Significant delays to induction of an-
aesthesia should be avoided and alternative strategies such as surgical cut-​
down for vascular access should be discussed if difficulties are encountered.
Immunosuppression
High doses of corticosteroids may be administered at induction of anaes-
thesia and/​or prior to reperfusion of the donor heart based on institutional
protocol.
Microbiology
Antimicrobials are essential and should be guided by local protocols and
patient microbiological history. Prophylactic antibiotics should be admin-
istered prior to commencing surgery and re-​administered during the case
as required.
Surgical pause
This should take place prior to surgical incision and ensure that the donor
heart has been retrieved at the donor site and that it is appropriate for the
transplant operation to proceed. Any additional concerns of the implanting
team should be identified and addressed.
Repeat sternotomy
In those patients who are having a repeat surgery or who are on MCS,
planning needs to take place as accessing the chest and establishing the pa-
tient on CPB will be more challenging. Prolonged time to access the chest
should be accounted for, cross-​matched blood should be available, and ex-
ternal defibrillation pads should be placed. It is not uncommon for CPB
to be established through femoral venous, and femoral or axillary arterial
cannulation.
Blood and coagulation
Cross-​matched blood should be available for commencement of surgery.
Antifibrinolytics such as tranexamic acid or epsilon-​aminocaproic acid are
used to minimize perioperative bleeding. When ready for CPB, heparin is
administered and ACT is monitored. Further dosing is targeted to ACT
during bypass.
46

464 Chapter 36 Heart transplantation: anesthetic management

HIT is a major challenge for cardiac transplantation and, where appro-

priate, transplantation may need to be delayed until >3 months since pre-
vious heparin exposure. In cases where this is not possible, alternative
agents such as the direct thrombin inhibitor bivalirudin can be used and
exposure to heparin in the perioperative period avoided. Other options for
patients with current or remote HIT and positive HIT antibodies for whom
surgery cannot be postponed include (1) preoperative plasmapheresis
and intraoperative administration of heparin, (2) preoperative administra-
tion of IVIG and intraoperative anticoagulation using heparin, and (3) co-​
administration of epoprostenol intraoperatively with heparin.
Cardiopulmonary bypass
This period involves recipient heart explantation and donor heart implant-
ation. Priorities are to ensure immunosuppression is administered according
to the agreed perioperative plan, coagulation monitoring is continued, and
that preparations are made for coming off CPB.
Haemodynamic priorities include identifying and managing low SVR and
preparing for the post-​CPB period. Vasoplegia is common following cardiac
transplantation, may manifest during CPB through low MAP with poor end-​
organ perfusion, and requires the administration of vasopressors.
It is common practice to continue low tidal volume ventilation with or
without inhaled NO during the CPB period in a bid to minimize bypass-​
induced lung ischaemia and mitigate the elevation in RV afterload following
transplantation.
Post cardiopulmonary bypass
This period includes weaning from CPB and establishing the new heart in the
recipient circulation. This involves continually assessing heart function, and
instituting and adjusting vasopressor and inotropic support as appropriate.
Weaning from CPB
Reperfusion time is required prior to weaning from CPB. The longer the
ischaemic period, the longer this reperfusion should be. Low-​dose infusions
of inotrope and vasoconstrictor can be commenced at this time and titrated
according to cardiovascular response. Temporary epicardial atrial and ven-
tricular pacing wires are placed and a heart rate of 90–​110 bpm targeted.
A PAC can be floated with transient reduction in flows to monitor pul-
monary pressures and cardiac output during separation from CPB.
Weaning should not commence until this reperfusion period is complete,
the patient is normothermic, there is no acidaemia, blood glucose is under
control (<10 mmol/​L), haemoglobin concentration >7.5 g/​dL, and potas-
sium normalized.
The weaning process should be gradual and performed under continuous
assessment of haemodynamic parameters, TOE, and visual inspection. This
information should be integrated to assess overall graft function and car-
diovascular status. The well-​functioning heart should be able to support
the circulation with a MAP >60 mmHg, a CVP <16 cmH2O, and a car-
diac index >2.2 L/​min/​m2 with low or moderate doses of vasopressors
and inotropes. If initial evaluation reveals inadequate graft function, the
graft should be further rested on CPB and changes made to cardiovascular
management.
 Intraoperatively 465

TOE
Intraoperative TOE is an invaluable diagnostic and monitoring tool
in patients undergoing heart transplantation. TOE is employed for
(1) intraoperative monitoring in the pre-​transplantation period, (2) evalu-
ation of cardiac allograft function and surgical anastomoses in the imme-
diate post-​transplantation period, and (3) diagnosis and management of
haemodynamic abnormalities.
In the pre-​CPB phase, the future anastomotic sites, ascending aorta, PA,
and caval veins should be evaluated for the presence of atherosclerosis
or thrombus, respectively. Also, the presence of thrombus at sites of low
blood flow and stasis should be evaluated, as clot can be dislodged during
surgical manipulation.
Following CPB, a comprehensive evaluation of the newly transplanted
heart should be performed, including LV and RV function, as well as com-
plete evaluation of valve function by two-​dimensional and colour flow
Doppler imaging. In addition, the anastomotic sites of the ascending aorta,
main PA, IVC, and SVC should show no discrete areas of narrowing by
two-​dimensional and have laminar flow by colour flow Doppler. Moreover,
throughout the post-​CPB and in the immediate postoperative period, TOE
is paramount in haemodynamic management, aiding in the decision to es-
calate haemodynamic support or to further deploy MCS.
Ongoing cardiovascular management
The main cardiovascular issues following heart transplantation are related
to PGD, often predominantly RV dysfunction, and vasoplegia. These fea-
tures rarely exist in isolation and the failing graft following transplantation
can result from varying degrees of each. Monitoring of clinical and echocar-
diographic parameters will allow diagnosis and targeted treatment.
Primary graft dysfunction
PGD is likely responsible for the majority of early transplant deaths. It is
characterized by the transplanted heart being unable to support recipient
circulation due to predominantly LV or RV dysfunction or a combination
of both. Risk factors which include donor, recipient, and surgical factors
should be identified by the perioperative team to facilitate early diagnosis
and intervention. Management comprises a spectrum of cardiovascular
support ranging from low-​to high-​dose inotropes and the use of MCS.
Devices including IABP, VA ECMO, or short-​term VAD (LVAD, RVAD, or
BiVAD) should be considered early in the context of persistent graft dys-
function despite high-​dose inotropic support. PGD is covered in more detail
in Chapter 17.
RV dysfunction
RV dysfunction is common following heart transplantation and is a cause of
major morbidity and mortality. It results from a combination of factors; the
recipient often has a degree of PH resulting from end-​stage HF. In addition,
the donor heart has undergone a series of insults from donor brain death,
cardioplegia, ischaemic-​ reperfusion injury, and CPB. In the early post-​
transplant setting it may be classified as PGD or secondary graft dysfunction
where there is a discernible cause (e.g. PH).
The diagnosis is challenging and a combination of clinical and echocar-
diographic parameters, along with clinical judgement is used. The RV may
46

466 Chapter 36 Heart transplantation: anesthetic management

appear dilated or poorly functioning on visual inspection. Invasive moni-

toring may demonstrate an elevated CVP, an elevated CVP:PA wedge pres-
sure ratio, and a low cardiac index. Echocardiography findings may include
increased RV volume and reduced RV contractility by reduced longitudinal
contractility (tricuspid annular plane systolic excursion) or reduced frac-
tional area change with or without a reduced LV end-​diastolic volume.
Ventricular interdependence means that any approach to management of
RV failure needs to target both ventricles. Goals include optimizing RV pre-
load, minimizing RV afterload, optimizing inotropic support, and maintaining
systemic blood pressure. A faster heart rate (90–​110 bpm) may help pre-
vent RV distension. Sinus rhythm or AV sequential pacing maintains AV syn-
chrony and consistent RV filling.
Although the RV is usually preload tolerant, in the dilated and failing RV,
fluid administration can have deleterious effects causing a deterioration in
global cardiac function by ventricular interdependence. In a response to
a small fluid challenge, if there are increased filling pressures along with a
minimal or even negative response to cardiac output, volume overload of
the RV should be considered. Volume reduction can be targeted by diuretic
bolus and/​or infusion and management should focus on the other factors
that can impact RV performance.
Minimizing RV afterload comes from the use of selective pulmonary
vasodilators (such as inhaled NO, often started pre-​emptively as described
previously) and avoidance of factors that increase RV afterload such as hyp-
oxia, acidaemia, hypercarbia, and high airway pressures.
The ideal inotropic agent would improve RV contractility, reduce RV
afterload, and have minimal effects on the systemic vasculature. This can
improve the coupling of the RV to the pulmonary circulation without
lowering systemic blood pressure, which is essential to maintain RV per-
fusion. Similarly, any vasopressors used should increase systemic pressure
while having minimal effect on the pulmonary vasculature. Combinations of
inotropes and vasopressors are often used with all of these factors in mind.
IABP should be considered particularly in the context of LV systolic dys-
function with the goal of reducing PCWP, RV afterload, and improving cor-
onary perfusion. If these measures are insufficient to support the newly
transplanted graft, short-​term RVAD (RV support only) or VA ECMO
(biventricular support) should be considered early.
Vasoplegia
Vasoplegia in this setting lacks a universal diagnosis but is clinically evident
when a patient has low systemic blood pressure, low SVR, and normal car-
diac output. It is often refractory to high doses of vasopressors.
It is associated with postoperative morbidity and mortality with risk fac-
tors including the pre-​transplant use of mechanical circulatory and pro-
longed CPB times. Vasoplegia can be managed with vasopressors such as
vasopressin and/​or norepinephrine (noradrenaline) along with methylene
blue in resistant cases. Agents such as inodilators and anaesthetic agents
need to be titrated as clinically indicated.
Bleeding/​coagulation
Bleeding can be problematic following cardiac transplantation and can be as
a result of prolonged CPB times, hypothermia, haemodilution, preopera-
tive antiplatelet/​anticoagulant use, and inflammatory activation. Bleeding
 Postoperatively 467

risk is increased in those who have been on preoperative mechanical sup-

port. Requirement for blood products can be determined by laboratory
tests including full blood count, laboratory coagulation screen, and point-​of-​
care testing such as TEG.
Once cardiovascular parameters are stable and bleeding is controlled, the
chest is closed and the patient is transferred back to the ICU.

Postoperatively
Once the patient is stable, they should be transferred to the ICU for ongoing
management. Depending on the support instituted in the operating room,
this may require additional personnel to ensure it occurs safely. All car-
diovascular monitoring should be continued during this period. In selected
patients, the TOE probe may be left in situ for continued monitoring, but
should be removed as soon as cardiovascular stability is confirmed and the
risk of tamponade has reduced.
It is important that a detailed structured handover is given to the ICU
team who will continue ongoing care. Anaesthetic, surgical, and perfusion
representatives should be present. Checklists ensure that salient points are
covered including postoperative immunosuppression.
The early problems in the ICU will largely be similar to those in the post-​
CPB period. Cardiovascular issues remain such as PGD, RV failure, and
vasoplegia. Haemodynamic instability should prompt careful clinical assess-
ment and institution of support as appropriate. Patients are at ongoing risk
of bleeding and any clinical deterioration may require prompt echocardiog-
raphy to identify potential cardiac tamponade. In addition, recipients are at
risk of hypothermia, acute kidney injury, and acid–​base disturbance, all of
which should be actively corrected.
In those patients on minimal cardiovascular and respiratory support who
are warm, not bleeding, and neurologically appropriate, tracheal extubation
should be considered as soon as possible.
468
 Chapter 37 469

Anaesthetic issues related

to lung transplantation
Introduction 470
Surgical strategies and anaesthetic considerations 472
Intraoperative management 473
Conclusion 480
470

470 Chapter 37 Anaesthesia for lung transplantation

Introduction
Preoperative evaluation for lung transplantation should include a compre-
hensive review of medical and surgical history, medications, and available
testing. While testing is often comprehensive and available at the time of
listing, optimization remains challenging in the setting of end-​stage cardio-
pulmonary disease given the narrow time frame of the ‘transplant window’.
Review of history
Indications for lung transplantation are organized by four primary diagnoses
of end-​stage pulmonary disease, including (1) obstructive lung disease, (2) re-
strictive lung disease, (3) CF or immunodeficiency disorders, and (4) pulmonary
vascular disease. Comorbid conditions that exist by indication may influence
anaesthetic choice and surgical planning (Table 37.1).
Both medical and surgical complexity continue to expand through in-
creased acceptance of donation after circulatory death and high risk donors,
wider adoption of bridging with mechanical cardiopulmonary support.
Although the LAS reflects urgency of transplantation, there has not yet
been a surgical scoring system for this population that predicts outcomes,
particularly that includes functional status and frailty. These are associated
with worsened outcomes in lung transplantation, but are difficult to rou-
tinely quantify and optimize prior to transplantation.

Table 37.1 Common indications for lung transplantation with associated

comorbidities
Indication for Common clinical presentation
transplantation
CF Younger age, suppurative secretions, empyema, antibiotic-​
resistant infections, pleural adhesions, indwelling ports
Malnutrition, malabsorption, and vitamin K deficiency;
pancreatic insufficiency
Diabetes, hepatic dysfunction, bowel dysfunction
(obstruction)
Fibrotic disease Older age, small chest cavity, concomitant cardiovascular
disease (CAD, pulmonary vascular disease, carotid disease)
Steroid dependency, chronic immunosuppression,
association with systemic immune disease (Sjögren’s
syndrome, rheumatoid arthritis), inflammatory myositis,
hepatic dysfunction
Obstructive Older age, concomitant cardiovascular disease (CAD,
disease pulmonary vascular disease, carotid disease), steroid
dependency, bullous disease
Pulmonary Vascularized adhesions, RV dysfunction, venous congestion
vascular disease of the liver and kidneys
Impingement of the recurrent laryngeal nerve by
enlarged PAs
Clotting or bleeding disorders, indwelling ports or lines for
ongoing pulmonary vasodilators
 Introduction 471

Review of medications
Preoperative medications such as bronchodilators, antibiotics, pulmonary
vasodilators, and immunosuppressants should be continued until trans-
plantation. Inhaled or IV pulmonary vasodilators should be replaced by IV
or inhaled therapies, or discontinued after initiation of VA ECMO or CPB.
Anxiety, depression, and pre-​existing pain conditions are predictive of
poor pain outcomes following lung transplantation, and should be opti-
mized preoperatively if possible. Medications with respiratory depressant
or sedative effects should be minimized.
Some lung transplant patients require systemic anticoagulation until trans-
plantation for thromboprophylaxis of atrial fibrillation or thromboembolic
conditions, which may require reversal once in the operating room. UFH
and low-​molecular-​weight heparin, warfarin, and more recently direct-​
acting oral anticoagulants have been described in the perioperative period
for lung transplantation. Timing and dosing of reversal agents should be
considered in the context of MCS.
Steroid-​dependent and immunosuppressed patients are at increased risk
for steroid-​dependent diabetes. This warrants evaluation of glycaemic con-
trol (HbA1c), as well as reviewing markers of any comorbid known endo-
crine disorders.
Fasting status and previous anaesthetic records should be reviewed. The
consent process includes risks of the anaesthetic, death, intraoperative re-
call, invasive line and TOE placement, and review of the pain management
plan and postoperative recovery. Use of anxiolytics and narcotics such as
benzodiazepines or opiates in the preoperative area is not recommended
unless absolutely necessary.
Physical examination
Physical examination should focus on the airway, heart, lungs, neurological
system and spine, and extremities for evidence of cardiopulmonary failure,
and evaluating ease of intravascular access.
Laboratory testing and imaging
Pulmonary function tests, blood gases, and V/​Q scans help predict ability
to tolerate one-​lung ventilation and inform the sequence of recipient native
pneumonectomy. If the non-​operative lung has little perfusion, or the room
air PaO2 is <45 mmHg, intraoperative mechanical circulatory support, or
MCS, may be necessary.
Cardiac testing assists in identifying candidacy, urgency of transplant-
ation, and planning intraoperative pulmonary support (e.g. VV ECMO), or
cardiopulmonary MCS. ECG, resting TTE, and right heart catheterization;
CI; CVP; and pulmonary pressures are key diagnostic modalities to identify
patients at risk for intraoperative haemodynamic instability. Older patients
may have CAD established by left heart catheterization. Cardiac MRI is
helpful to identify irrecoverable cardiac scarring and to quantify RV function
for patients with long-​standing congenital heart disease or PH.
Oesophageal imaging and function tests may demonstrate evidence of
stricture, dysfunction, or aspiration. If aspiration is present, a rapid se-
quence intubation must be considered. If oesophageal stricture is found,
the risk of placing a TOE probe should be weighed carefully. The presence
of oesophageal dysfunction may alter the route of enteral feeding, nutri-
tion, and recovery.
472

472 Chapter 37 Anaesthesia for lung transplantation

Certain disease states raise suspicion for liver or renal dysfunction (e.g.
CF) or RV dysfunction (e.g. idiopathic pulmonary fibrosis, Table 37.1).
Complete blood counts, chemistry panels, liver and renal function tests, as
well as coagulation studies should be reviewed. Preoperatively, avoidance
of transfusion is desirable to mitigate risks of alloimmunization. However,
blood products are commonly needed for lung transplantation, and blood
type, screen, and cross-​match should be performed in advance.
Intraoperative medication management
The anaesthesia team will often administer induction immunosuppression.
Awareness of drug mechanism, dosing, metabolism, and drug compatibility
is necessary. Expensive monoclonal antibody therapies should be dosed
after assurance that MCS and/​or plasmapheresis are not needed, so the
drug reaches therapeutic blood concentrations.
Gram-​positive coverage with both a cephalosporin and vancomycin in the
setting of immunosuppression is warranted. For recipients with significant,
long-​standing resistant microbial burden, a specialized antibiotic regimen
will be designed. Dosing and re-​dosing of both immunosuppression and
antibiotics should be adjusted accordingly if MCS is utilized.

Surgical strategies and

anaesthetic considerations
Anaesthesia for particular surgical strategies: MCS
Cardiopulmonary MCS may be present preoperatively, or planned for pa-
tients with (1) severe PH or anticipated RV failure as a result of PA clamping
and recipient native pneumonectomy; (2) concomitant cardiac procedures
(e.g. intracardiac shunts, coronary bypass or valve repair); (3) complex
airway resulting in inability to safely and adequately achieve lung isolation;
(4) small donor lung size relative to the recipient, to avoid full cardiac output
through a small vascular bed; and (5) fragile LA tissue in the recipient or
inadequate donor LA cuff that prohibits clamping of the atria and comple-
tion of the pulmonary venous anastomoses. Recently, some centers have
demonstrated a perioperative outcome benefit with an elective VA strategy
for intraoperative support.
Intraoperative teams must communicate prior to induction to confirm the
surgical plan. This includes review of the surgical approach, single or bilateral
procedure, positioning, cannulation sites and type of MCS. VV ECMO, VA
ECMO, or CPB may be cannulated centrally or peripherally. If the patient
is on preoperative VV ECMO, it may be continued into the postoperative
phase. Patients requiring peripheral VA ECMO should have a right upper
extremity arterial line to monitor oxygenation to the head and upper body
to monitor for potential Harlequen (North-South) syndrome, where de-
oxygenated, inadequately mixed blood is pumped from the left ventricle to
the upper body. Urgent MCS may be needed due to unanticipated haemor-
rhage, RV failure, or poor function of the newly implanted allograft.
The surgical team and perfusionist should be present in the room on in-
duction of anaesthesia, especially for high-​risk patients such as those with
severe pulmonary hypertension. The need for MCS is possible at any stage,
 Intraoperative management 473

and teams, equipment (appropriately sized cannulas, circuits, and pumps),

and systemic heparin doses should be prepared.
Anaesthetic planning
Intraoperative monitoring
Standard monitoring includes a five-​lead ECG, blood pressure, central and
peripheral temperature, end-​tidal carbon dioxide monitoring, pulse oxim-
etry, and urine output. Pulse oximeters placed on the extremities will often
lose signal during the procedure due to positioning, vasoconstriction, or
poor distal perfusion related to surgical manipulation or loss of pulsatile
flow. Fluid warmers and warming devices should be utilized, as it is difficult
to maintain body temperature with a large amount of exposed body sur-
face and extracorporeal MCS.
Invasive haemodynamic monitoring is necessary for lung transplantation.
Cannula position for MCS may influence the site of arterial line placement.
Radial arterial lines may become dampened during the case and a femoral
arterial line may be needed. Monitoring CVP, pulmonary pressures, cardiac
output, and mixed venous oxygenation is extremely useful in guiding volume
resuscitation and haemodynamic management. The choice of central line
site may be influenced by VV ECMO choice of cannulation site.
Equipment
Single-​and double-​lumen tubes of various sizes should be available for lung
isolation. A single-​lumen tube may be placed for pre-​transplant bronchos-
copy for management of secretions, and may remain in place if the pro-
cedure will be performed completely on VA ECMO or CPB. Fibreoptic
bronchoscopy confirms lung isolation and allows for bronchopulmonary
toilet. During bronchial anastomosis, the anaesthesia team should avoid
inserting suction catheters or bronchoscopes in the operative lung, while
the surgical team carefully avoids balloon puncture of the bronchial cuff
during dissection and native lung pneumonectomy. Insertion of airway ex-
change catheters for exchange of double-​to single-​lumen tubes at the com-
pletion of the procedure should be carefully performed in light of the fresh
bronchial anastomoses.

Intraoperative management
Anaesthesia induction and maintenance
Induction is a key time of potential haemodynamic instability, particularly in
the setting of PH or underlying RV dysfunction, and MCS should be imme-
diately available. Although prophylactic IABP placement has been described,
there is little evidence to support its use in this setting. Usually, ECMO or
CPB are prepared. A rapid sequence intubation with use of rapid-​acting
neuromuscular blockers may be chosen if the patient is not appropriately
NPO prior to induction. Induction agents should include those that avoid
myocardial depression and reduction of SVR. Drugs or manoeuvres that in-
crease pulmonary vascular tone (i.e. PVR) should also be avoided, as this
may precipitate right heart decompensation in patients with borderline RV
function. Importantly, surgical stimulation, poor pain control, or shivering
may all precipitate increases in PVR that may be poorly tolerated in a patient
47

474 Chapter 37 Anaesthesia for lung transplantation

with severe PH or borderline RV function. MCS circuits can absorb signifi-

cant amounts of medication used for pain control and sedation (e.g. lipophilic
or highly protein-​bound drugs), due to an increased volume of distribution
and decreased drug clearance. Similarly, absorption binding of fentanyl to
pulmonary endothelium results in significant reductions in plasma concen-
trations following initial native pneumonectomy and may require redosing.
Inotropic and vasopressor support should be available for support of the
RV in the setting of PA clamping. Inhaled pulmonary vasodilators (e.g. inhaled
NO and prostaglandins) are thought to reduce mean pulmonary pressures,
improve V/​Q matching and thereby the PaO2/​FiO2 ratio, and RV function in
lung transplant patients. Randomized controlled studies have failed to show
benefit for inhaled NO for prevention of ischaemia-​reperfusion injury or
demonstrate any clear perioperative or long-​term benefits. A recent ran-
domized controlled trial recently demonstrated non-inferiority when com-
paring iNO to inhaled epoprostenol in the lung transplant population.
Positioning
For a SOLT, a posterior thoracotomy incision is typically performed, and
the patient will be positioned laterally with the operative side up. For a mid-
line sternotomy approach, the patient is typically supine with arms tucked.
In the bilateral thoracotomy or thoracosternotomy (‘clamshell’) approach,
the patient will be positioned supine with arms tucked, or arms lifted an-
teriorly, flexed slightly at the level of the elbow, and abducted carefully,
with the forearms resting on cushioned support. Avoidance of direct nerve
compression or over-​abduction is critical to avoid peripheral nerve injury.
Careful padding of the axillary and ulnar nerves is recommended.
Ventilation strategies
Supporting oxygenation, ventilation, and avoiding acidosis during one-​lung
ventilation is a challenge in patients with end-​stage disease. In patients with
suppurative disease, it is helpful to achieve early lung isolation and avoid
cross-​contamination from the chronically infected native lung into the sur-
gical field, chest cavity, and newly transplanted lungs.
Positive pressure ventilation, high inspiratory pressures, or high PEEP ex-
acerbates haemodynamic instability by intrathoracic venous compression.
Patients with increased lung compliance and expiratory obstruction are at
risk for dynamic lung hyperinflation, or ‘breath stacking’, and permissive
hypercapnia may be necessary. Recipient pneumonectomy typically begins
with to the lung with less perfusion to minimize hypoxaemia due to shunting
in the deflated lung. If hypoxaemia due to shunting occurs, this improves
dramatically with PA clamping and recipient pneumonectomy. Avoidance
of drugs that interfere with hypoxic pulmonary vasoconstriction is im-
portant during this stage. Protective strategies should always be employed
for any native lung not being explanted, and the newly transplanted lungs.
This includes a low FiO2 <40%, driving pressures not to exceed 15 cmH2O,
5–​10 cmH2O PEEP with low tidal volumes. Most centres use 6 cc/​kg of
predicted body weight as a starting strategy to reduce barotrauma. There
is significant institutional variability on how these goals are achieved, with
little information about the donor lungs being provided to the care team in
order to tailor the ventilation strategy. After a SOLT, significant differences
in compliance between the native and transplanted lungs may exist, which
rarely may require differential ventilation.
 Intraoperative management 475

Haemodynamic instability is encountered with intravascular volume loss,

surgical manipulation of the heart, PA clamping and unclamping, as well as
reperfusion of the pulmonary allograft (Table 37.2). Communication be-
tween teams during these steps is critical.
Fluid management
The newly transplanted lung is susceptible to developing pulmonary oe-
dema due to disrupted lymphatic drainage, which can lead to increased
extravascular lung water. Additional injury is incurred due to ischaemia-​
reperfusion, endothelial disruption, and subsequent primary graft dysfunc-
tion, which makes fluid management particularly challenging.

Table 37.2 Anticipating anaesthetic interventions by surgical step

Surgical step Anaesthetic steps
Exposure and Evaluate oxygenation, ventilation, and
dissection haemodynamic performance on one-​lung ventilation
Bleeding during dissection may require early
resuscitation and inotrope/​pressor support
PA clamp Assess RV function frequently via TOE, CI, mixed
venous saturation
If RV failure ensues, consider MCS
Recipient When organ arrives in room, second check for
pneumonectomy organ compatability is performed
Heparin is administered to achieve and maintain a
suitable ACT for planned level of MCS
Implantation of Avoid airway suctioning. Surgical team prompts the
donor lung: bronchial anaesthesia team to test for bronchial leak/​integrity
anastomosis via slow manual inflation with room air.
Implantation of donor Maintain anticoagulation goals, prepare for
lung: PA anastomosis reperfusion
Implantation of donor Potential for volume loss. Volume, inotropic, and
lung: LA anastomosis pressor support may be required
Steroids and mannitol are given just after posterior
anastomosis complete
De-​airing of the graft Partial release of PA clamp and full removal of LA
clamp prior to finishing anterior LA anastomosis
Assess for air on TOE
Aortic venting with needle if left sided air observed
Notify any available team members of reperfusion
(next step)
Reperfusion Gradual release of PA clamp by surgical team
Potential for haemodynamic instability, volume loss.
May require inotropic or pressor support
Teams coordinate manual ventilation post
reperfusion with low FiO2, and initiate mechanical
ventilation with inhaled pulmonary vasodilators, and
lung protective settings
476

476 Chapter 37 Anaesthesia for lung transplantation

Lung transplant patients are volume resuscitated with balanced crystalloids

or colloid if they do not require transfusion. In an off pump population, each
litre of intraoperative fluid increases grade 3 PGD by 13% after adjusting for
patient weight. With each litre of colloid administered, a decreased PaO2/​
FiO2 ratio at 12 hours and delay in extubation was observed. An elevated
CVP has been associated with delayed extubation due to early graft dys-
function, with the proposed mechanism being increased hydrostatic pres-
sure through the pulmonary vasculature, leading to increased alveolar fluid.
On the other hand, avoidance of volume resuscitation in lieu of high-​
dose vasopressors or inotropes to support the haemodynamics risks
ensuing end-​organ malperfusion and injury. Lung transplant patients carry
a significant risk of acute kidney injury, amplified by perioperative immuno-
suppressants, antibiotics, diuretics, and high-​dose vasoactive drug use. We
recommend a goal-​directed approach to volume replacement using invasive
monitoring, for example, cardiac output, mixed venous, invasive pressures,
and TOE to guide volume resuscitation to optimize perfusion, oxygen de-
livery, and lactate clearance.
Bleeding and transfusion
Transfusion is commonly required during lung transplantation, particularly
after the blood loss associated with graft de-​airing and reperfusion. Bleeding
risks include a bilateral procedure, previous thoracic surgery, known pleural
adhesions, and CPB. Literature describing transfusion-​related outcomes in
lung transplantation is derived from small, retrospective, observational trials
from single centres, and confounded by risk factors for bleeding and the use
of MCS. A few studies suggest a transfusion-​related immunomodulatory
effect in lung transplantation, but more recent studies indicate that peri-
operative transfusion is associated with worsened PGD, CLAD, and overall
graft and patient survival in lung transplantation.
Transfusion has been implicated in allograft rejection and poor out-
comes. Some centers ensure all lung transplant patients receive EBV and
CMV negative products, particularly when the recipient and the donor are
also EBV and CMV negative, although this is challenging if massive trans-
fusion is needed or blood shortages are encountered. Leucoreduction of
products for SOT is widely accepted. However, if patients have had a prior
bone marrow transplant, or are receiving rATG, irradiated products are re-
quired. High-​volume red blood cell transfusion is associated with morbidity
and mortality in lung transplantation, while platelet transfusion has been
implicated in the development of post-​lung transplant anti-​HLA antibodies,
with worsened graft and patient survival. Preoperative transfusion is linked
to the development of new HLA antibodies, challenges in organ matching,
and worsened lung transplantation outcomes, it remains unclear how or if
intraoperative allogeneic red blood cell or platelet transfusion affects HLA
alloimmunization and allograft function.
Antifibrinolytics have been described for lung transplantation. Aprotinin
reduces bleeding and transfusion requirements, but is associated with acute
kidney injury and death in cardiac surgical patients and no longer available
in the US. Tranexamic acid and aminocaproic acid reduce bleeding and
transfusion requirements in cardiac and non-​cardiac surgery, but these
have not been well studied in lung transplantation.
 Intraoperative management 477

Prothrombin complex concentrates allow for rapid replacement of

clotting factors without transfusion of fresh frozen plasma, albeit with
the problems of limited availability, significant cost, and associated risk
of thromboembolism. In non-​surgical populations, the risk of thrombo-
embolism associated with prothrombin complex concentrate administra-
tion for warfarin reversal is similar to those receiving fresh frozen plasma,
approximating 7%. Off-​label use of prothrombin complex concentrates as
well as activated factor VII has been described for postoperative haem-
orrhage in lung transplantation. Extreme caution should be taken in ad-
ministering these medications in the setting of MCS, as catastrophic
thromboembolism may occur.
Transoesophageal echocardiography
TOE allows for evaluation of intracardiac shunting, monitoring of LV and RV
function, and preload optimization, and procedural guidance for cannulae
placement. During the dynamic phases of lung transplantation, guide thor-
ough deairing during reperfusion, and evaluation of both PA and pulmonary
vein anastomoses post lung transplantation. Despite the utility of real-​time
evaluation, there remains no consensus about routine use of TOE during
lung transplantation, which carries a class 2b indication according to the
American College of Cardiology/​American Heart Association/​American
Society of Echocardiography guidelines. Fig. 37.1a shows a normal pul-
monary vein tracing with pulsed wave Doppler, with peak systolic pul-
monary vein velocities well below 100 cm/​s, while Fig. 37.1b shows an
isolated elevated peak systolic pulmonary vein velocity exceeding 100 cm/​
s suggestive of pulmonary vein stenosis or kinking. Fig. 37.2 shows RV dila-
tion, hypertrophy, and dysfunction during lung transplantation in a mid-​
oesophageal four-​chamber view (Fig. 37.2a) and transgastric short-​axis
view (Fig. 37.2b), with a ‘D-​shaped’ LV indicating both volume and pressure
overload.
Acute pain management
Good perioperative pain management provides physiological benefits and
improved graft function, extubation times, patient satisfaction, and reduced
ICU length of stay during hospitalization. The evidence in lung transplant-
ation is limited to small, single-​centre studies that highlight both individual
and institutional practice variability.
Thoracic epidural catheters are identified as a gold standard for pain
control. Timing of catheter placement, choice of local anaesthetic solu-
tion, and adjunct use makes comparing outcomes difficult. An ideal strategy
employs catheter placement prior to weaning the ventilator, and such
that it covers both the incision and chest tubes. In the setting of systemic
anticoagulation or coagulopathy, epidural catheter placement or removal
should be avoided to avoid the risk of epidural haematoma. Regional cath-
eters, e.g. paravertebral, thoracic fascial plane blocks, and catheters and
local wound infiltration with and without intermittent bolus techniques have
been described as effective alternatives. Adjuncts such as acetaminophen,
gabapentin, lidocaine, ketamine and dexmedetomidine may be employed as
non-​narcotic adjuncts, while non-​steroidal anti-​inflammatory drugs are typ-
ically avoided due to renal toxicity issues. It remains unclear how or if these
strategies translate into improved long-​term pain outcomes.
478

478 Chapter 37 Anaesthesia for lung transplantation

(a)

(b)

Fig. 37.1 (a) Pulsed wave Doppler of the left superior pulmonary vein in the
mid-​oesophageal, two-​chamber view. Systolic dominance of a normal pulmonary
vein tracing, with velocities <100 cm/​s and without turbulence is noted (colour
flow Doppler not shown). (b) Abnormal continuous-​wave Doppler of the left
superior pulmonary vein in the mid-​oesophageal, two-​chamber view. Here velocities
throughout systole and diastole exceed 200 cm/​s, and pulsed wave Doppler is not
possible because the velocities are too high. See plate section.

Chronic pain management

Chronic pain after lung transplantation is estimated at approximately 10–​
49%. Post-​ thoracotomy pain syndrome and post-​ sternotomy pain are
well-​described, significant problems in the general thoracic and cardiac sur-
gical population with 25–​80% attributed to extensive nerve damage. Risk
 Intraoperative management 479

(a)

(b)

Fig. 37.2 (a) TOE demonstrating RV dilation and dysfunction in the mid-​
oesophageal, four-​chamber view. This patient required VAV ECMO for support during
lung transplantation. (b) TOE demonstrating RV dilation, hypertrophy, and dysfunction
in the transgastric, short-​axis view of the RV. This patient has the classically described
‘D-shaped left ventricle’ suggestive of both pressure and volume overload, and
required VAV ECMO for support during lung transplantation. See plate section.

factors in thoracic surgery include youth, female sex, and the presence of
preoperative pain. In patients undergoing clamshell incisions, the incidence
is not described. In lung transplant patients undergoing various surgical ap-
proaches, about half of survey respondents describe their pain as signifi-
cantly compromising their quality of life. The source of pain is multifactorial
and related to pre-​existing pain conditions, CNIs, and pain at incision and
thoracostomy tube sites.
480

480 Chapter 37 Anaesthesia for lung transplantation

Management of chronic pain after lung transplantation is exceedingly

challenging due to potential drug interactions, interference with immuno-
suppression regimens, and the potentially catastrophic risk of respiratory
depression associated with opioids. Multimodal pain management strat-
egies, cognitive behavioural therapy, non-​narcotic adjuncts, and therapeutic
nerve blocks or ablations should be considered whenever possible.

Conclusion
Preparing a comprehensive anaesthetic plan for lung transplantation re-
quires an in-​depth evaluation of pre-​existing conditions, medications, avail-
able testing, and understanding of the surgical plan. Anaesthetic challenges
include intraoperative management of anaesthesia haemodynamics, volume
resuscitation, bleeding, and pain.
 481

Index
For the benefit of digital users, indexed terms that span two pages (e.g., 52–​53) may, on occasion,
appear on only one of those pages.
Tables, figures, and boxes are indicated by t, f, and b following the page number
A extracorporeal membrane asphyxiated lung donors 340
oxygenation 450–​53 Atgam 395
acanthamoeba 253t heart atrial fibrillation 202–​3
ACE inhibitors 9 transplantation 459–​67 atrial flutter 202–​3
acute cellular rejection 242–​ lung transplantation 372–​ ATTR-​ACT trial 35–​36
43, 390–​93 73, 469–​80 Avalon cannula 103
Acute Decompensated Heart ventricular assist azathioprine 213, 237–​38,
Failure National Registry devices 453–​58 239t, 397t
(ADHERE) score 13–​14 angiotensin receptor–​ azithromycin 400–​1
acute kidney injury 203 neprilysin inhibitor 9
acute lung injury 368 angiotensin II receptor
adenovirus 257 blockers 9
B
adherence to therapy anomalous pulmonary bacterial infections
223, 231 veins 272 405, 416–​17
ADHERE score 13–​14 anthrax vaccine 262t barrier contraception 222
adult congenital heart antibiotics see antimicrobial basiliximab 235t, 235, 395
disease 303–​16 prophylaxis and BCG vaccine 262t
assessment 307–​16 treatment beta blockers 8–​9
contraindications for antibody-​mediated rejection biomarkers, lung allograft
transplantation 308 243–​44, 393–​94, 401–​2 rejection 391
history and anti-​CD20 monoclonal bisoprolol 8–​9
examination 312–​13 antibodies 256, 394 bisphosphonates 407–​8
indications for anticoagulation bleeding
transplantation 308 ECMO 93, 283–​85, 451 ECMO 95, 452
investigations 314–​15 total artificial heart 113–​14 GI tract with ventricular
outcome of antimicrobial prophylaxis and assist device 71–​72
transplantation 304–​5 treatment 199–​200, heart transplantation 201–​
perioperative issues 315–​16 249–​50, 367, 411–​ 2, 466–​67
pulmonary artery 19, 463 Impella devices 76–​77
reconstruction 311–​12 antiproliferative agents 213, lung transplantation 476–​77
pulmonary 237–​38, 239t surgical ventricular assist
hypertension 308–​9 anti-​thymocyte globulin 234–​ devices 71
sensitization 309 35, 235t, 395 BODE score 321–​22
single ventricle 309–​10 aortic coarctation/​ body surface area
waiting list criteria 305 hypoplasia 272 estimation 88
ADVANCE trial 66 aortic valve bone marrow
AG-​10 35–​36 repair during LVAD suppression 405
A-​HeFt trial 10 surgery 57–​58 bone mineral density 219–​20
AL amyloid 31–​36 replacement for heart bortezomib 394
albendazole 261 failure 13 brain death donors 172–​
alemtuzumab 235t, 235–​ aortopulmonary 77, 348–​51
36, 395–​96 collaterals 311–​12 brainstem death 140
AlloMap testing 240–​41 aprotinin 476–​77 bronchiectasis 418
aminocaproic acid 476–​77 arrhythmias bronchiolitis obliterans
amphotericin, heart failure 11–​12 syndrome 392t, 400,
lipid 258–​59 post-​heart 401t, 436–​37
amphotericin B 259–​60 transplantation 202–​3 bronchoscopy
ampicillin 254 ventricular assist lung donors 342–​43
amyloid cardiomyopathy devices 72–​73 post-​lung
31–​36, 37–​38 arterial blood gases 328–​29 transplantation 391
anaemia 114 Aspergillus spp. 258–​59, bumetanide 10
anaesthesia 406, 415–​16 Burkholderia cepacia 418
482

482 Index

C COAPT trial 12–​13 donation after brain death

coarctation of the aorta 272 172–​77, 348–​51
calcineurin inhibitors 212, collapse test 342 donation after circulatory
236–​37, 239t complete atrioventricular death 180–​83, 300,
calculated PRA 245–​46 septal defect 307 344, 352–​53
cancer computed donor-​specific antibodies
post-​transplantation risk tomography see CT 243–​44, 245–​46
214–​15, 406–​7 congenital heart disease 132–​ dopamine 197t
screening 331 35, 190, 270–​73, 291, drowning, lung donors 340
Candida infections 406, 416 see also adult congenital
carbapenemase-​producing heart disease
Enterobacteriaceae congenitally corrected
E
417–​18 transposition of the Ebstein's anomaly 306–​7
carbon monoxide great arteries 306–​7 ECG
poisoning 340 contraception 222 amyloid
cardiac allograft vasculopathy coronary angiography, heart cardiomyopathy 32–​34
210–​11, 240–​41, 436 donors 145–​46 heart donors 143
cardiac catheterization 7 coronary artery myocarditis 18–​20
cardiac donors see disease 4–​7 pre-​transplantation 314–​15
heart donors corticosteroids 213, 238, echinocandin 258–​59
cardiac MRI 18–​20, 239t, 392–​93 echocardiography
20b, 32–​34 cross-​clamp 174–​77, amyloid
cardiac resynchronization 186–​87 cardiomyopathy 32–​34
therapy 12 cross-​matching 245–​46 heart donors 143–​44
cardiac transplantation see cryptococcosis 259–​60 myocarditis 18–​20
heart transplantation Crystal City algorithm 151f pre-​heart
cardiomyopathy CT transplantation 314–​15
dilated 16, 290 lung allograft rejection 391 pre-​lung
Hypertrophic 290 lung donors 343 transplantation 329
non-​ischaemic 4–​7 pre-​heart restrictive
see also restrictive transplantation 314–​15 cardiomyopathies
cardiomyopathies pre-​lung 28–​31, 30t
cardiopulmonary bypass transplantation 328–​29 Eisenmenger
heart transplantation 270–​ cystic fibrosis 323–​24, 327, syndrome 306–​7
71, 464 418, 470t electrocardiogram see ECG
left ventricular assist cytomegalovirus (CMV) 216, elexacaftor–​tezacaftor–​
device 52 250t, 255–​56, 386, 406, ivacaftor 323–​24
cardiopulmonary exercise 413, 414t endocrine
testing 7–​8 management 149–​52
carfilzomib 394 endomyocardial biopsy
carvedilol 8–​9 D acute cellular rejection
CFTR modulators 323–​24 daclizumab 395 242–​43, 243t
Chagas' disease 253t Dallas criteria 18–​20, 20b amyloid
chest trauma 340 daptomycin 417–​18 cardiomyopathy 32–​34
chest X-​ray depression 223 myocarditis 18–​20
lung allograft rejection 391 desensitization 309, 333 rejection
lung donors 343 Destination Therapy Risk surveillance 240–​41
myocarditis 18–​20 Score (DTRS) 45–​46 restrictive
cholera vaccine 262t DEVELOP-​UK trial 360t cardiomyopathies 28–​
chronic kidney dextrocardia 272 31
disease 220–​21 dextro-​transposition of the ENDURANCE trial 66
chronic lung allograft great arteries 306–​7 eosinophilic
dysfunction 400–​1, diabetes myocarditis 23–​24
402t, 436–​37 heart failure risk 4–​7 epinephrine 197t
chronic obstructive new onset after eplerenone 9
pulmonary disease 321–​ transplantation 218–​19 Epstein-​Barr virus 250t, 256,
22, 326, 470t digoxin 10–​11 406, 413
chylothorax 433 DIG trial 10–​11 Eurotransplant Heart Donor
ciclosporin 212, 236–​37, dilated cardiomyopathy 16, Score 154, 155t
239t, 397t 290 everolimus 238, 239t, 397t
Clostridium difficile-​associated dobutamine 10–​11, 197t EXPAND Heart Pivotal
diarrhoea 255 domino transplants 426–​27 Trial 168
 Index 483

extracorporeal intensive care 87 ex vivo heart perfusion 159,

cardiopulmonary intraoperative, lung see also Organ Care
resuscitation transplantation 375 System (OCS) Heart
(E-​CPR) 85–​88 left atrial venting 278 ex vivo lung perfusion 338–​
extracorporeal left left ventricular assist 39, 355–​69
ventricular assist device device 52 acute lung injury 368
43–​44, 70 left ventricular distension antimicrobials 367
extracorporeal membrane and venting 94–​95 cellular or acellular
oxygenation (ECMO) low mixed venous oxygen perfusate 365
ambulation strategy 227 saturation 95 criteria for assessment and
anaesthetic issues 450–​53 maintenance 451 reconditioning 362–​63
anticoagulation 93, 283–​ monitoring 282–​83 decision to transplant
85, 451 neck cannulation 280–​81 lungs 363
aortic position 89 neurological injury 453 gene therapy 368
arterial cannula 89–​90 nutrition 229 HCV-​infected
axillary artery cannula 89 operating room 87 donors 367–​68
bleeding 95, 452 outcomes 286 history of 356–​59
blood product paediatrics see paediatric Lund protocol 358t
transfusion 94 extracorporeal OCS protocol 358t, 359
bridging support membrane portable 359, 366
for transplant oxygenation specialized
candidates 137–​38 peripheral arterial cannula centres 365–​66
cannulation flow capacity 89–​90 static 366
sequence 91–​93 peripheral cannulation Steen Solution 356–​58
cannulation strategy 88, 280–​82 stem cell therapy 368
88, 279–​82 platelet transfusion 451 surgical technique 363–​64
cannula types 88–​91 post-​cardiotomy Toronto protocol 358t, 359
cardiac cath lab 87 cardiogenic translational
central shock 120–​21 research 367–​69
cannulation 88, 280 post-​lung transplantation trials 360t
children see paediatric 381, 387–​88 urokinase use 367
extracorporeal pre-​lung ventilation 364–​65
membrane transplantation 332
oxygenation pressure alarms 95
clinical applications 80–​93 recirculation 94
F
coming off urgently 286 red cell transfusion 451 fibrotic lung disease 470t
complications 285, 452–​53 single ventricle 286 fidaxomicin 255
contraindications 450–​51 stepdown unit 87–​88 fluid management 198
conversion of subclavian artery 5-​fluocytosine 259–​60
configuration 279 cannula 89 follow-​up clinics 232
decannulation 93 TandemLife system 104–​8 Fontan circulation 297–98.
distal perfusion cannula 90 technical considerations 307, 309–​10
dual-​lumen cannula 91, 103–​4 88, 277–​79 functional testing, heart
emergency department 87 thromboembolism 452–​53 failure 7–​8
extracorporeal venoarterial (VA ECMO) fungal infections 258–​60,
cardiopulmonary 82–​83, 277, 280–​82, 406, 414–​16
resuscitation (E-​ 283, 450–​51, 452 furosemide 10
CPR) 85–​88 venoarteriovenous (VAV
fem-​fem VV ECMO 80–​81 ECMO) 84–​85, 278
fem-​IJ VV ECMO 80–​81 venous cannula 90–​91
G
femoral cannula 90–​ venous cannula siphonage ganciclovir 255–​56
91, 281–​82 capacity 90–​91 gastrointestinal tract
flow rates 283 venovenous (VV ECMO) bleeding with ventricular
haemodynamic 80–​81, 94, 277, assist devices 71–​72
monitoring 94 279–​80, 283, pre-​lung transplantation
hypercarbia/​ 450–​51, 452 evaluation 329–​30
hypoxaemia 94 ventilation 451 gastro-​oesophageal reflex
impaired gas exchange 94 weaning 95–​96, 285–​ disease 407
indications 80–​93, 86, 452 gastroparesis 433
276–​77, 450–​51 extracorporeal ventricular gene therapy 368
innominate artery assist device 42–​44 giant cell myocarditis 16–​
cannula 89 complications 70–​75 18, 23–​24
48

484 Index

glycopeptide-​resistant optimization 148–​49 heart–​lung transplantation

enterococci 417–​18 paediatrics 269, 291, 292–93 132–​35, 334–​
post-​procurement 35, 423–​37
assessment 146 allograft monitoring 433–​34
H predicted heart cardio-​
haemodynamics mass 147b pneumonectomy 428–​
ECMO 94 principles of selection 141 29
post-​heart pulmonary artery causes of death in
transplantation 196–​98 catheterization 145 recipients 435
restrictive cardiomyopathies respiratory chronic allograft
28–​31, 30t management 152–​53 dysfunction 436–​37
haemolysis risk scoring systems 154 chylothorax 433
Impella devices 76–​77 'rule of 100' 149b domino transplants 426–​27
surgical ventricular assist size matching 147, 269 donor assessment 427
devices 74 temperature early graft failure 433
Haemophilus influenzae B management 153–​54 early postoperative
vaccine 262t thromboembolic management 433
haemorrhage, post-​lung prevention 153 gastroparesis 433
transplantation 387 heart failure immunosuppression 433
Harlequin syndrome 84–​85 ACC/​AHA staging 4, 6t implantation 430–​31
heart donors adult congenital heart indications 424, 426t
age 142 disease 307 organ allocation 426
assessment of quality 141 aetiology 4–​7, 6b organ procurement 427–​28
blood tests 142–​43 cardiac catheterization 7 outcomes 434
cause of death 142 cardiac resynchronization phrenic nerve
circulatory therapy 12 dysfunction 433
management 149–​52 clinical assessment 4 post-​transplantation
comorbidity 142 epidemiology 4 morbidity 436–​37
coronary functional testing 7–​8 recipient
angiography 145–​46 history 5b procedure 428–​31
cross-​clamp 174–​ ICD therapy 11–​12 retransplantation 437
77, 186–​87 medical evaluation 7–​8 statistics 424
Crystal City algorithm 151f NYHA functional HeartMate IP 62–​63
desirable and undesirable classification 4, 6t HeartMate II 58–​59, 66–​67
characteristics 141t pharmacotherapy 8–​11 BTT trial 64–​65
donation after brain physical examination 5b DT trial 65
death 172–​77 prognostication 13–​14 HeartMate 3 (HM3) 58–​
donation after circulatory recurrence after heart 59, 65–​67
death 180–​83, 300 transplantation 223 HeartMate XVE 63
donor–​recipient referral to advanced heart heart transplantation
compatibility 146–​48 failure team 14 acute cellular
ECG 143 risk scores 13–​14 rejection 242–​43
echocardiogram 143–​44 signs and symptoms 5b acute kidney injury 203
endocrine sudden cardiac death adherence to therapy
management 149–​52 prevention 11–​12 223, 231
ex vivo organ perfusion 159 transcatheter aortic valve adults with congenital
general critical care 153–​54 replacement 13 heart disease see
haematological transcatheter mitral valve adult congenital heart
management 153 repair 12–​13 disease
history and transthoracic allograft
examination 142 echocardiography 7 implantation 190–​93
infections derived ventricular anaesthetic issues 459–​67
from 251–​52 arrhythmias 11–​12 antibiotic prophylaxis 199–​
intraoperative Heart Failure Survival Score 200, 249–​50, 463
assessment 146 (HFSS) 44–​45, 132–​35 antibody-​mediated
legal issues 291–​92 heart–​kidney transplantation rejection 243–​44
marginal donors 158–​59, 293 132–​35, 440–​41, aortopulmonary
mortality risk scores 154 442, 443 collaterals 311–​12
nutrition 153 heart–​liver transplantation arrhythmias 202–​3
OCS Heart 159, 132–​35, 440, 441, 442–​ back-​table preparation 190
160–​68, 177–​80 43, 444–​46 bleeding 201–​2, 466–​67
 Index 485

blood and medical assessment of timings 460

coagulation 463–​64 candidates 132–​35 TOE 465
bone mineral medication vascular access 462–​63
density 219–​20 management 460–​61 vasoplegia 206–​7, 466
bridging support 137–​38 multidisciplinary team weaning 198, 271, 464
cardiac allograft working 201 wound management 230
vasculopathy 210–​ myocarditis 22–​23 HeartWare Ventricular Assist
11, 240–​41 nutrition 200 Device (HVAD) 58–​59,
cardiac explantation 187–​89 osteoporosis 219–​20 65–​67, 70
cardiopulmonary bypass paediatrics see paediatric HELP trial 359, 360t
270–​71, 464 heart transplantation heparin-​induced
cardiovascular pain management 230 thrombocytopenia
management 196–​98 patient education 231 93, 463–​64
children see paediatric heart physical hepatitis A vaccine 262t
transplantation rehabilitation 230–​31 hepatitis B 251–​52
chronic post-​transplant monitoring hepatitis B vaccine 262t
complications 209–​23 and investigations 196 hepatitis C 251–​52, 367–​68
chronic kidney pre-​cardiopulmonary herpesviruses 406, 413–​
disease 220–​21 bypass 462–​64 14, 414t
congenital heart disease pregnancy 221–​22 histocompatibility 146
132–​35, 270–​73, see preoperative HIV 251–​52
also adult congenital assessment 460–​61 hormonal contraception 222
heart disease primary graft dysfunction HOUDINI 412t
contraindications 132–​35, 193, 204, 465 HSV-​1 and HSV-​2 406,
134t, 268–​69, 308–​12 psychological 413, 414t
critical care 196–​201, 467 morbidity 152–​53 human herpesviruses 406,
depression 223 psychological 413–​14, 414t
dextrocardia 272 rehabilitation 230 human leucocyte antgen
diabetes mellitus 218–​19 psychosocial (HLA) 295
donors see heart donors assessment 135–​37 human metapneumovirus 257
early complications 201–​3 psychosocial support 200 human papillomavirus
first successful pulmonary artery vaccine 262t
transplant 172 reconstruction 311–​12 hydralazine 10
fluid management 198 pulmonary hyperacute rejection
following a congenital hypertension 308–​9 201, 393–​94
operation 190 recipient operation 186–​87 hypertension
follow-​up clinics 232 recipient selection 132–​37, heart failure risk 4–​7
history and examination 268–​69, 312–​16 post-​heart
312–​13, 460 recurrence of heart transplantation 221
hyperacute rejection 201 failure 223 hydralazine 290
hypertension 221 rehabilitation 229–​32
immunizations, post-​ rejection
transplantation surveillance 240–​41
I
261, 262t renal replacement ideal cardiac index 88
immunosuppression 199, therapy 203 immunizations, post-​
212–​14, 222, 234–​38, repeat sternotomy transplantation 261, 262t
461, 463 189, 463 immunosuppression 199,
implantable cardiac reproductive health 221–​22 212–​14, 433, 461, 463
devices 461 respiratory induction 234–​36, 394–​96
indications 132–​35, 133b, management 198 maintenance 236–​38, 396–​
268, 308 restrictive 97, 397t
induction and maintenance cardiomyopathies 37–​ myocarditis 23–​24
of anaesthesia 462 38 post-​lung transplantation
infection management 199–​ retransplantation 223 complications 404–​5
200, 215–​16, 231, right ventricular dysfunction pregnancy 222
247–​61, 463 203, 465–​66 Impella devices 76–​77, 99–​102
investigations 314–​15 sensitization 245–​46, 309 bridging support
LVAD explantation 189–​90 single ventricle 272, 309–​10 for transplant
malignancy 214–​15 surgical brief 462 candidates 137–​38
mechanical circulatory surgical pause 463 post-​cardiotomy cardiogenic
support 197, 461 tamponade 202 shock 121–​22
486

486 Index

implantable cardioverter left heart catheterization 329 sew-​then-​core

defibrillators left superior caval vein 272 technique 52–​53
(ICDs) 11–​12 left ventricular assist device sternotomy approach 50
infection management (LVAD) see also stroke 72
post-​transplantation 199–​ ventricular assist devices TandemHeart 106–​7
200, 215–​16, 231, air entrainment 75 third-​generation
247–​61, 386, 405–​6, alternative to heart devices 65–​67
411–​19, 463 transplantation 46–​47 thoracotomy
pre-​transplantation ambulation 227, 228 approach 55–​56
assessment 251–​ arrythmias 72–​73 thrombosis 74
52, 331 axillary artery outflow 57 timing of
VAD driveline 72 bridging support implantation 45–​46
VAD insertion site 75 for transplant VA ECMO cannulas 52
infective endocarditis candidates 137–​38 valvular disease
prophylaxis 216 cannula site bleeding 71 management 57–​58
influenza 257 cardiopulmonary bypass 52 weaning from bypass 54–​55
influenza vaccine 262t clinical trials with durable levosimendan 10–​11, 197t
inotersen 35–​36 devices 61–​67 linezolid 417–​18
inotropes 10–​11, 197t complications 69–​77 Listeria monocytogenes 254
INSPIRE trial 359, 360t, 366 continuous flow devices 64 liver assessment
INTERMACS registry 64 contraindications 44–​45 pre-​heart
International EXPAND Lung core-​then-​sew transplantation 314–​15
Pivotal trial 360t technique 52–​53 pre-​lung
interstitial lung disease de-​airing 54–​55 transplantation 330
325, 327–​28 descending aorta 57 live vaccines 261
intra-​aortic balloon pump driveline infection 72 loop diuretics 10
(IABP) 98–​99 driveline tunnelling 50–​51 lung cancer, post-​lung
ambulation strategy 226–​27 durable systems 44–​47, 70 transplantation 407
bridging support EACTS indications 44–​45 lung donors 337–​45
for transplant explantation 189–​90 acceptance criteria 339–​44
candidates 137–​38 extracorporeal 43–​44, 70 age 340
rehabilitation 226–​29 first-​generation asphyxiated donors 340
restrictive devices 62–​63 blood gases 341–​42
cardiomyopathies 36–​ GI tract bleeding 71–​72 bronchoscopy 342–​43
37 haemolysis 74 carbon monoxide
intravascular ventricular assist hypoxaemia 73 poisoning 340
system (iVAS) 99 inflow 52–​53 care of donor 338
intravenous immunoglobulin left atrial appendage chest trauma 340
(IVIG) 394 occlusion 57–​58 collapse test 342
INTrEPID trial 63–​64 mediastinal bleeding 71 declining lungs 338–​39
isavuconazole 258–​59 mortality rate 70–​71 donation after brain
ischaemic heart disease 145b myocardial recovery 46 death 348–​51
isosorbide dinitrate 10 myocarditis 22–​23 donation after circulatory
ivabradine 9–​10 nutrition 229 death 344, 352–​53
ivermectin 261 off-​pump implantation 52 drowning 340
operative planning 50 exclusion criteria 338–​39
outflow 53 imaging 343
J outflow graft ischaemic time 343–​44
Japanese encephalitis obstruction 75 preparation of lungs 377
vaccine 262t physiological criteria for race matching 341
candidacy 44–​45 reconditioned lungs see
pump exchange 58–​59 ex vivo lung perfusion
K pump pocket 50 sex matching 341
kidney restrictive size of donor 341
acute injury 203 cardiomyopathies 36–​ smoking history 342
chronic disease 220–​21 37 suicidal hanging 340
right ventricular surgical assessment 339
failure 73–​74 ventilation 341–​42
L risk stratification 45–​46 lung–​kidney transplantation
left atrial appendage second-​generation 335, 440–​41, 442,
occlusion 57–​58 devices 64–​65 443
 Index 487

lung–​liver transplantation indications for postoperative physiology

335–​36, 440, 441, transplantation/​ 384–​85
442–​43, 444 activation on waiting post-​transplant
lung transplantation list 326–​28, 470t lymphoproliferative
acute cellular induction and maintenance disorder 406–​7
rejection 390–​93 of anaesthesia 473–​74 preoperative ECMO 332
adherence to therapy 231 induction preoperative evaluation
airway complications immunosuppression 328–​31, 470–​72
387, 402–​3 394–​96 primary graft dysfunction
anaesthetic issues 372–​ infection management 387–​88
73, 469–​80 231, 331, 386, 405–​6, psychological rehabilitation
analgesia 385 411–​19 230
antibody-​mediated interstitial lung disease 325, pulmonary evaluation
rejection 393–​ 327–​28 328–​29
94, 401–​2 intraoperative pulmonary vascular disease
antimicrobial prophylaxis cardiopulmonary (hypertension) 322–​
and treatment 411–​19 support 374–​75 23, 326–​27, 372–​73,
bilateral anterior intraoperative ECMO 375 470t
thoracosternotomy intraoperative medications recipient pneumonectomy
(clamshell) 373 472 376–​77
bleeding and intraoperative monitoring rehabilitation 229–​32
transfusion 476–​77 473 rejection management
bronchiectasis 418 laboratory tests 471–​72 389–​97
bronchiolitis obliterans late complications renal evaluation 330
syndrome 392t, 399–​409 restrictive allograft
400, 401t maintenance syndrome 401, 402t
cancer screening 331 immunosuppression restrictive lung disease 325
cardiac evaluation 329 396–​97, 397t retransplantation 334,
chest closure 380 malignancy risk 406–​7 381, 408
chronic lung allograft mechanical circulatory right heart dysfunction
dysfunction 400–​ support 472–​73 372–​73
1, 402t medications review 471 sedation 385
clinical management after multiorgan transplant sensitization 333
listing 331–​32 334–​36 single versus double
contraindications 321b neurological complications 332–​33
critical care 384–​86 408 statistics 320
cystic fibrosis 323–​24, 327, neutrophilic reversible sternal-​sparing
418, 470t allograft dysfunction thoracotomy 373
de-​listing 336 400–​1, 402t sternotomy 373–​74
donor lung preparation 377 nutrition 330 TOE 477
donors see lung donors obstructive lung disease ventilation 386, 474–​75
EVLP lungs 363 321–​22, 326, 470t wound management 230
fibrotic lung disease 470t operative technique lymphocytic choriomeningitis
fluids 475–​76 372–​80 virus 253t
follow-​up clinics 232 osteoporosis 407–​8 lymphocytic myocarditis
gastrointestinal pain management 230, 23–​24
evaluation 329–​30 477–​80
gastro-​oesophageal reflex patient education 231
disease 407 physical examination 471
M
haemorrhage 387 physical rehabilitation malaria 253t
hepatic evaluation 330 230–​31 malignancy
history-​taking 470 pleural adhesions 380–​81 post-​transplant risk 214–​15,
imaging 471–​72 positioning 474 406–​7
immunosuppression 394–​ posterolateral thoracotomy screening 331
97, 397t 374 marginal donors 158–​59, 293
immunosuppression-​related postoperative mean fluorescent intensity
complications 404–​5 complications 387–​88 245–​46
implantation 378 postoperative measles, mumps, rubella
incision choice 373–​74 considerations 385–​86 vaccine 262t
indications for postoperative ECMO 381, mechanical circulatory
referral 320–​25 387–​88 support
48

488 Index

ambulation strategy 227 echocardiography 18–​20 O

heart–​liver transplantation endomyocardial biopsy
444–​46 18–​20 obstructive lung disease 321–​
lung transplantation 472–​73 eosinophilic 23–​24 22, 326, 470t
percutaneous 97–​108 epidemiology 16 OCS Heart 159, 160–​68,
post-​cardiotomy fulminant 24–​25 177–​80
cardiogenic shock giant cell 16–​18, 23–​24 OCS Lung INSPIRE trial
122–​25, 125t heart 360t, 366
post-​heart transplantation transplantation 22–​23 OKT3 235t, 236, 395–​96
197 immunosuppression 23–​24 oral anti-​hyperglycaemics
preoperative in heart infectious causes 16–​18 219
transplantation 461 laboratory findings 18–​20 Organ Care System (OCS)
rehabilitation 226–​29 left ventricular assist Heart 159, 160–​68,
restrictive devices 22–​23 177–​80, 300
cardiomyopathies lymphocytic 23–​24 oseltamivir 257
36–​37 non-​infectious causes osteoporosis 219–​20, 407–​8
MELD score 45–​46 16–​18
methylprednisolone 239t, outcomes 24–​25 P
392–​93 recovery-​related factors
metoprolol succinate 8–​9 23b paediatric extracorporeal
metronidazole 255 therapy 22–​24 membrane oxygenation
milrinone 10–​11, 197t three phase model 16–​18 275–​86
mineralocorticoid receptor Myocarditis Treatment Trial anticoagulation 283–​85
antagonists 9 23–​25 cannulation strategies 279–​82
minimally invasive ventricular coming off urgently 286
assist devices 76–​77 complications 285
MitraClip 12–​13
N conversion of configuration
MITRA-​FR trial 12–​13 Neisseria meningitides vaccine 279
mitral valve repair 262t exclusion criteria 276
for heart failure 12–​13 neurological complications flow rates 283
LVAD surgery 57–​58 ECMO 453 inclusion criteria 276
MMR vaccine 262t post-​lung transplantation indications 276–​77
Model for End-​Stage Liver 408 monitoring 282–​83
Disease (MELD) score neutrophilic reversible outcomes 286
45–​46 allograft dysfunction patient selection 276–​77
MOMENTUM 3 trial 52–​53, 400–​1, 402t single ventricle 286
66–​67 new onset diabetes after venoarterial (VA ECMO)
mTOR inhibitors 214, 238, transplantation 277, 280–​82, 283
239t (NODAT) 218–​19 venoarterial with left heart
multidrug-​resistant pathogens Nocardia spp. 254 drainage 278
417–​18 non-​HLA antibodies 243–​44 venoarteriovenous (VAV
muromonab (OKT3) 235t, non-​ischaemic ECMO) 278
236, 395–​96 cardiomyopathies 4–​7 venovenous (VV ECMO)
Mustard procedure 306–​7 non-​tuberculous 277, 279–​80, 283
mycobacteria, non-​ mycobacteria 406, 417 weaning 285–​86
tuberculous 406, 417 norepinephrine 197t paediatric heart
Mycobacteria abscessus North–​South syndrome transplantation 267–​73
complex 418, 419 84–​85 ABO incompatibility 293–94
mycophenolate mofetil 213, Novacor LVAS trial 62–​64 anomalous pulmonary
237–​38, 239t, 397t NOVEL Lung trial 360t veins 272
mycophenolate sodium 239t NT-​proBNP 34 aorta 272
myocarditis 15–​25 nuclear imaging, amyloid bicaval anastomosis 271
aetiology 16–​18, 17b cardiomyopathy 32–​34 bicaval versus biatrial 271–​72
antiviral therapy 23–​24 nutrition bypass 270–​71
cardiac MRI 18–​20, 20b ECMO 229 cardiac tumours 291
chest X-​ray 18–​20 heart donors 153 cardiectomy 270
clinical presentation and LVAD 229 cardiomyopathy 290
diagnosis 18–​20 post-​heart transplantation challenges 289
Dallas criteria 18–​20, 20b 200 congenital heart disease 291
definition 16 pre-​lung transplantation contraindications 268–​69
ECG 18–​20 330 dextrocardia 272
 Index 489

donation after circulatory early 118 RADIAL score 204, 206t

death 300 ECMO 120–​21 RADIANCE trial 10–​11
donor evaluation 269 latent 118 rATG 395
donor heart preparation 271 mechanical circulatory recruitment manoeuvres
ex vivo cardiac perfusion 300 support 122–​25, 125t 348–​51
Fontan patients 297 outcomes 124–​25 rehabilitation 225–​32
history of 288 step-​wise approach to high-​ rejection
indications 268 risk patients 119–​22 acute cellular 242–​43,
left superior caval vein 272 ventricular assist devices 390–​93
marginal donor 293 121–​22 antibody-​mediated 243–​44,
operative steps 270 posterior reversible 393–​94, 401–​2
outcome 272 encephalopathy hyperacute 201, 393–​94
oversize donors 293 syndrome 408 surveillance 240–​41
plastic bronchitis 300 post-​transplant REMATCH trial 63
primary transplantation lymphoproliferative renal function
289–90, 291 disorder 214–​15, 406–​7, acute kidney injury 203
protein-losing enteropathy 436–​37 chronic kidney disease
299 Potential Donor-​Derived 220–​21
pulmonary arteries 272 Disease Transmission immunosuppression-​related
pulmonary hypertension Event 251–​52 complications 405
296-97 PREDICTA score 204, 207t pre-​lung transplantation
recipient selection 268–​69 predicted heart mass 147b 330
rescue therapy 291 prednisone 239t, 397t total artificial heart 114
sensitization 294 pregnancy 221–​22 renal replacement therapy
single ventricle 272, 297–98 primary graft dysfunction 203
size matching 269 193, 204, 387–​88, 465 renin–​angiotensin–​
unique characteristics 289 PROCEED II trial 168 aldosterone system
waiting list 291 proteasome inhibitors 394 8–​11
weaning 271 protein-losing enteropathy renin–​angiotensin–​
pain management 230, 385, 299 aldosterone system
477–​80 ProtekDuo cannula 76–​77, antagonists 9
panel reactive antibody 107–​8, 121–​22 reproductive health 221–​22
(PRA) testing 245–​46 prothrombin complex respiratory management
PARADIGM-​HF trial 9 concentrates 476–​77 heart donors 152–​53
parainfluenza 257 PROVED trial 10–​11 heart transplantation 198
Park's stitch 57–​58 psychological morbidity respiratory syncytial virus
patient education 152–​53 257
programme 231 psychological rehabilitation restrictive allograft syndrome
patisiran 35–​36 230 401, 402t
pentamidine 257–​58 psychosocial assessment restrictive cardiomyopathies
percutaneous ventricular 135–​37 27–​38
assist devices 76–​77 psychosocial support 200 aetiology 28t
Perfadex 348–​51 pulmonary artery amyloid cardiomyopathy
peripheral arterial cannula catheterization 145, 456 31–​36, 37–​38
flow capacity 89–​90 pulmonary artery clinical presentation and
pertussis vaccine 262t reconstruction 311–​12 diagnosis 28–​31
phrenic nerve dysfunction 433 pulmonary function tests echocardiography 28–​31,
physical rehabilitation 230–​31 lung allograft rejection 391 30t
Pierce–​Donachy VAD 62–​63 pre-​lung transplantation endomyocardial biopsy
plasmapheresis 394 328–​29 28–​31
plastic bronchitis 300 pulmonary vascular disease haemodynamics 28–​31, 30t
pleural effusions 343 (hypertension) 296–97, heart transplantation
Pneumocystis jirovecii 216, 308–​9, 322–​23, 326–​27, 37–​38
250t, 257–​58, 386 372–​73, 470t mechanical circulatory
polio vaccine 262t pyrimethamine/​sulfadiazine/​ support 36–​37
posaconazole 258–​59 leucovorin 260 medical therapy 28–​31
post-​cardiotomy cardiogenic restrictive lung disease 325
shock 117–​25 retransplantation 223, 334,
adverse prognostic
R 381, 408, 437
indicators 124–​25 rabies 253t rhinovirus 257
definitions 118 rabies vaccine 262t right atrial pressure 7
490

490 Index

right heart catheterization SynCardia see total artificial transthoracic

7, 329 heart echocardiography
right heart dysfunction (TTE), heart failure 7
372–​73 traveller's diarrhoea vaccine
right ventricular assist device
T 262t
42–​43, 444–​46 tacrolimus 212, 236–​37, tricuspid valve repair, LVAD
right ventricular dysfunction 239t, 397t surgery 57–​58
203, 465–​66 tafamidis 35–​36 trimethoprim–​
right ventricular failure, tamponade 202 sulfamethoxazole
ventricular assist devices TandemHeart 106–​7, 137–​38 (TMP–​SMX) 254, 257–​
73–​74 TandemLife system 104–​8 58, 386
rituximab 256, 394 TandemLung 107–​8 Trypanosoma cruzi 253t
rotavirus vaccine 262t TAVR UNLOAD trial 13 TTR 31–​36
'rule of 100' 149b tetanus vaccine 262t
thiazolidinediones 219
thoracoabdominal organ
U
S transplantation 439–​46 unfractionated heparin
sacubitril–​valsartan 9 donor operation 441 283–​85
Salmonella typhi vaccine 262t indications 440–​41 univentricular heart see single
Seattle Heart Failure Model outcomes 443–​44 ventricle
(SHFM) 13–​14, 44–​45, recipient preparation UNOS Improving Patient
132–​35 442–​43 Safety 251–​52
Senning procedure 306–​7 thrombosis/​ urokinase 367
sensitization 245–​46, 294, thromboembolism
309, 333 ECMO 452–​53
sepsis 253 Impella devices 76–​77
V
SHIFT trial 9–​10 surgical ventricular assist vaccines, post-​transplantation
single ventricle devices 74 261, 262t
ECMO 286 TMP–​SMX 254, 257–​58, 386 valganciclovir 255–​56
heart transplantation 272, torsemide 10 vancomycin 255
297–98, 309–​10 total artificial heart varicella vaccine 262t
sirolimus 214, 238, 239t, 397t (SynCardia) 109–​15 varicella zoster virus 256,
six minute walk tests 7–​8, adverse events 115, 116t 406, 413, 414t
328–​29 anaemia 114 vasoplegia 206–​7, 466
skin cancer 214, 407 anticoagulation 113–​14 vasopressin 197t
smallpox vaccine 262t assessing for fit 111 vasopressors 197t
smoking, lung donors 342 indications 110–​11 ventilation
solid tumours 215 intraoperative management ECMO 451
spirometry 391 111–​12 EVLP lungs 364–​65
spironolactone 9 outcomes 114–​15 lung donors 341–​42
Steen Solution 356–​58 postoperative lung transplantation
stem cell therapy 368 complications 114 474–​75
steroids 213, 238, 239t, postoperative management post-​lung transplantation
392–​93 113–​14 386
STITCHES trial 7 preoperative management ventricular arrhythmias
STITCH trial 7 111–​12 11–​12
Streptococcus pneumoniae renal dysfunction 114 ventricular assist devices
vaccine 262t Toxoplasma gondii 250t, (VADs) see also left
stroke 253t, 260 ventricular assist device
Impella devices 76–​77 tranexamic acid 476–​77 air entrainment 75
surgical ventricular assist transbronchial biopsy 391 anaesthetic issues 453–​58
devices 72 transcatheter aortic valve arrythmias 72–​73
Strongyloides stercoralis replacement 13 bleeding 71, 76–​77
253t, 261 transcatheter mitral valve bridge to transplantation
sudden cardiac death repair 12–​13 453
prevention 11–​12 TRANSFORM-​HF trial 10 cannula displacement 73
suicidal hanging, lung donors transoesophageal cannula site bleeding 71
340 echocardiography catheter migration 76–​77
supraventricular tachycardia (TOE) 456, 465, 477 central venous catheter 456
202–​3 transposition of the great complications 69–​77
surgical site infection 249–​50 arteries 306–​7 driveline infection 72
 Index 491

emergence and extubation mortality rate 70–​71, ventricular tachycardia 203

458 76–​77 VIENNA trial 360t
extracorporeal 42–​44, 70 non-​invasive monitors 456 viral infections 406, 413–​14
GI tract bleeding 71–​72 outflow graft obstruction viral myocarditis 16–​18
haemolysis 74, 76–​77 75 virtual cross-​matching 245–​46
heart–​liver transplantation percutaneous 76–​77 VO2max 7–​8
444–​46 positioning 456 voriconazole 258–​59
hypoxaemia 73 post-​cardiotomy cardiogenic V/​Q scan 328–​29
indications 41–​47, 453 shock 121–​22
induction and maintenance postoperative care 455, 458
of anaesthesia 457–​58 pre-​anaesthetic
W
insertion site infection 75 consultation 454 West Nile virus 253t
intraoperative management preoperative checklist wound management 230
and monitoring 454–​55
455–​56 pulmonary artery catheter
intravascular ventricular 456
Y
assist system short term therapy 42–​44 yellow fever vaccine 262t
(iVAS) 99 stroke 72, 76–​77
limb ischaemia 76–​77 thromboembolism 74, 76–​77
mediastinal bleeding 71 TOE 456
Z
minimally invasive 76–​77 vascular injury 76–​77 zanamivir 257